Impact of the Decentralization and Integration of Control Activities in Basic Health Services

Faculty of Medicine and Health Department of Public Health Sciences

Contribution to the improvement of the control of Buruli ulcer in the Territory of Songololo: Impact of the decentralization and integration of control activities in basic health services

Dissertation submitted for the degree of Doctor in Medical Sciences at the University of Antwerp to be defended by

Delphin Phanzu Mavinga

PROMOTERS

Prof. Dr. Marleen Boelaert, ITM Prof. Dr. Greet Ieven, UA Prof. Dr. Pascal Lutumba Tshindele, UNIKIN

MENTOR

Em. Prof. Dr. Françoise Portaels

Antwerp, 2015

ISBN: 9789057284809 Legal deposit number: D/2015/12.293/9

Contribution à l’amélioration du contrôle de l’Ulcère de Buruli dans le Territoire de Songololo : Impact de la décentralisation et de l’intégration des activités de lutte dans les services de santé de base

*** Bijdrage aan het verbeteren van de bestrijding van Buruli ulcus in het gebied Songololo. Impact van de decentralisatie en integratie van de ziektebestrijding in de basisgezondheidsdiensten.

Delphin Phanzu Mavinga

Doctoral committee

Chair

Prof. Dr. Jean-Pierre Van geertruyden University of Antwerp, Belgium

Promoters

Prof. Dr. Marleen Boelaert Institute of Tropical Medicine, Belgium

Prof. Dr. Greet Ieven University of Antwerp, Belgium

Prof. Dr. Pascal Lutumba University of Kinshasa, DR Congo

Institut National de Recherche Biomédicale, DR Congo

Mentor

Em. Prof. Dr Françoise Portaels Institute of Tropical Medicine, Belgium

Member

Prof. Dr. Lut Lynen Institute of Tropical Medicine, Belgium

External jury members

Prof. Dr. Tjip S. van der Werf University of Groningen, Netherlands

Em. Prof. Dr. Jacques Aubry University of Nantes, France

This study was supported by a PhD grant of the Institute of Tropical Medicine (ITM) supported by the Belgian Development Cooperation (DGD) in the frame of institutional cooperation.

Dédicace

A ma chère Patrie,

Ab imo pectore

Don béni de Dieu nous légué par nos aïeux Languis tout en nous n’est plus que ténébreux Retroussons nos manches afin que tous à l’ouvrage Nous nacrions l’image ternie de cet indivis héritage

Que vers le progrès tous nous tendions Bien en quête de nouveaux horizons Possible serait-il par des hommes tout autres Refaisant parure de ce paradis terrestre

Extrait du recueil "Délices cachées" Delphin Phanzu Mavinga 19 mai 1992

A toi, Godelive, A vous, Albi-Priscilla, Jonathan-Raphaël, Arnaud-Goddel, Don-Parfait & Roselyne la grâce Phanzu, Trouvez ici l’expression de mon profond amour.

Table des matières Table des matières ...... 9 Liste des abréviations ...... 11 Résumé ...... 13 Samenvatting ...... 17 CHAPITRE I. INTRODUCTION GENERALE ...... 21 I.1. DÉFINITION ...... 23 I.2. DESCRIPTION CLINIQUE DE L’UB ...... 23 I.3. EPIDÉMIOLOGIE ...... 25 I.4. DIAGNOSTIC ...... 30 I.5. TRAITEMENT ...... 34 I.6 CONTRÔLE DE LA MALADIE ...... 35 I.7. BREF APERÇU HISTORIQUE DE L’ UB EN RDC...... 40 I.8. RÉFÉRENCES ...... 41 CHAPITRE II. OBJECTIFS ET ORGANISATION DE LA THESE ...... 59 II.1. ENONCÉ DU PROBLÈME DE RECHERCHE ...... 61 II.2. JUSTIFICATION DE LA THÈSE ...... 62 II.3. OBJECTIF GÉNÉRAL ...... 63 II.4. OBJECTIFS SPÉCIFIQUES ...... 63 CHAPITRE III. METHODOLOGIE ...... 65 III.1. LA RÉPUBLIQUE DÉMOCRATIQUE DU CONGO ...... 67 III.2. CADRE DU TRAVAIL ...... 70 III.3. TYPES D’ÉTUDES ...... 71 III.4. ECHANTILLONS ÉTUDIÉS ...... 71 III.5. DÉFINITIONS DE CAS ...... 72 III.6. PROCÉDURES DE CONFIRMATION PAR LE LABORATOIRE ...... 73 III.7. COLLECTE DES DONNÉES ...... 74 III. 8. ANALYSE DES DONNÉES ...... 74 III. 9. ASPECTS ÉTHIQUES ...... 75 CHAPITRE IV. RESULTATS ...... 77 IV.1. PARTIE CLINIQUE ...... 79 IV.1.1. Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. 81 IV.1.2. Short report: edematous Mycobacterium ulcerans infection (Buruli ulcer) on the face: a case report 97 IV.1.3. Short Report: Under treated necrotizing fasciitis masquerading as ulcerated edematous Mycobacterium ulcerans infection (Buruli ulcer). 109 IV.1.4. Mycobacterium ulcerans infection (Buruli ulcer) on the face: a comparative analysis of 13 clinically suspected cases from the Democratic Republic of Congo 123 IV.2. PARTIE SANTÉ PUBLIQUE 145 IV.2.1. Effect of a control project on clinical profiles and outcomes in Buruli ulcer: a before/after study in Bas-Congo, Democratic Republic of Congo. 147 IV.2.2. Burden of Mycobacterium ulcerans disease (Buruli ulcer) and underreporting ratio in the Territory of Songololo, Democratic Republic of Congo 167 IV.2.3. Effectiveness of routine BCG vaccination on Buruli Ulcer Disease: a case-control study in the DR Congo, Ghana and Togo 195 IV.2.4. Management of Mycobacterium ulcerans disease (Buruli ulcer) in the Territory of Songololo, Democratic Republic of Congo: Outcomes, Challenges and Prospects. 231 CHAPITRE V. DISCUSSION GENERALE ...... 253 CHAPITRE VI. CONCLUSION ET PERSPECTIVES D’AVENIR ...... 275 REMERCIEMENTS ...... 281 CURRICULUM VITAE ...... 285

Liste des abréviations

AAF Aspiration à l’Aiguille Fine AFIP Armed Forces Institute of Pathology/Washington DC ALM American Leprosy Missions AS Aire de Santé BAAR Bacilles Acido-Alcoolo Résistants CHU Centre Hospitalier Universitaire CP-LT Coordination Provinciale Lèpre-Tuberculose CS Centre de Santé CSDT Centre de Santé de Dépistage et de Traitement CSR Centre de Santé de Référence CUL Culture sur milieu de Löwenstein CZN Coloration de Ziehl-Neelsen HGR Hôpital Général de Référence HIS Examen Histopathologique IEC Information, Education, Communication IME Institut Médical Evangélique de Kimpese IMT Institut de Médecine Tropicale d’Anvers IMUB Initiative Mondiale de lutte contre l’Ulcère de Buruli INRB Institut National de Recherche Biomédicale/Kinshasa IT Infirmier titulaire LT Lèpre-Tuberculose MTN Maladies Tropicales Négligées OMS Organisation Mondiale de la Santé PCR Polymerase chain reaction PNLUB Programme National de Lutte contre l’Ulcère de Buruli POD Prevention of Disability PS Poste de santé RDC République Démocratique du Congo RECO Relais Communautaire UB Ulcère de Buruli VIH Virus de l’immunodéficience humaine ZN Ziehl-Neelsen ZS(R) Zone de Santé (Rurale)

Résumé

L’ulcère de Buruli (UB), une des dix-sept maladies tropicales négligées, constitue à l’heure actuelle un réel problème de santé publique dans plusieurs pays en Afrique comme le Bénin, le Cameroun, la Côte d'Ivoire, la République Démocratique du Congo (RDC) et le Ghana. L’UB se manifeste sous une variété des formes, dont les plus spectaculaires sont des ulcérations étendues qui aboutissent à des séquelles invalidantes. Dans les foyers endémiques, la maladie est bien connue dans les communautés touchées.

En RDC , l’UB est surtout endémique dans la province du Bas-Congo et en particulier dans le territoire de Songololo, l’un des foyers les plus importants dans le pays. L’hôpital IME/Kimpese héberge en son sein un programme spécialisé de prise en charge de l’UB depuis 2004. Après plus de trois ans de mise en œuvre des activités de lutte dans les deux zones de santé du territoire, la morbidité et les incapacités dues à l’UB restaient élevées, et l’hospitalisation restait une très lourde charge financière pour les patients et leurs familles, malgré la gratuité des soins offerte. Pour faire face à cette situation, un système décentralisé et intégré de prise en charge de l’UB dans le territoire de Songololo fut introduit à partir de 2007 et notre projet de thèse de doctorat avait pour objectif d’en évaluer l’impact. Nous avons étudié le programme de lutte en termes d’efficacité opérationnelle, d’efficience, de qualité d’accès, de mise à l’échelle et de pérennisation, avec comme but final de pouvoir (ou non) généraliser cette expérience dans d’autres foyers endémiques de la province et de la RDC.

Notre projet d’étude doctorale a évolué de la recherche clinique (Partie I) vers la recherche en santé publique, avec l’évaluation dudit programme de contrôle de la maladie (Partie II) et couvre une période d’une douzaine d’années. Nous avons étudié deux séries de malades : premièrement, tous les cas d’UB observés de 2002 à 2007 à l’Hôpital Général de Référence (HGR) de l’Institut Médical Evangélique (IME) de Kimpese et en deuxième lieu, tous les cas de UB enregistrés entre 2008 et 2014 dans les deux

Zones de Santé Rurales (ZSR) du Territoire de Songololo : la ZSR de Kimpese et la ZSR de Nsona-Mpangu.

La première publication dans la partie clinique porte sur l’expérience de l’HGR IME/Kimpese dans la prise en charge des patients souffrant de l’UB entre 2002 et 2004. Cette étude nous a montré que 70,5% des cas probables ont été confirmés par le laboratoire et que 38,9% des cas étaient des rechutes. Nonante quatre pourcent des cas confirmés étaient porteurs de lésions ulcérées et 25% présentaient une limitation fonctionnelle au moment du diagnostic. Dix-sept pourcent des patients présentaient des atteintes osseuses confirmées par le laboratoire. Le délai médian depuis le début des symptômes jusqu’au diagnostic était de 60 jours et la durée médiane d’hospitalisation était de 102 jours. Trente-sept pourcent des patients ont été déclarés guéris avec une limitation des mouvements articulaires et une mortalité de 19,4% a été observée.

La deuxième publication porte sur la prise en charge de la forme œdémateuse de l’UB à localisation faciale. Les difficultés diagnostiques et thérapeutiques rencontrées et la nécessité d’un traitement conservateur ont été mises en évidence. L’antibiothérapie spécifique a un effet bénéfique dans la réduction de l’œdème et la limitation de l’étendue des excisions chirurgicales.

La troisième publication porte sur le diagnostic différentiel de l’UB. Elle a montré les difficultés qui peuvent être rencontrées, même dans une zone endémique connue, à diagnostiquer l’UB purement sur une base clinique et souligne l’importance d’examens complémentaires de laboratoire, y compris les analyses histopathologiques.

La quatrième publication traite de la prise en charge des lésions faciales de l’UB durant une période de cinq années allant de 2003 à 2007. La fréquence de ces lésions faciales était de 7% parmi les patients confirmés. La réponse clinique à l’antibiothérapie spécifique était meilleure pour les lésions les plus précoces. Cette étude souligne l'importance de la formation des professionnels de santé en zones endémique pour le

14 diagnostic précoce et une référence rapide des cas d’UB, y compris les localisations atypiques.

Pour la deuxième partie, celle de santé publique, la première publication est une étude comparative avant-après qui montre l’effet du projet de lutte contre l’UB trois années après sa mise en œuvre. Cette période a été marquée par un accroissement important du nombre des cas détectés et un changement substantiel du profil clinique des patients UB traités à l’HGR IME/Kimpese.

La deuxième publication relate l’enquête exhaustive réalisée dans le territoire de Songololo en juillet-août 2008 qui a permis de déterminer l’ampleur de la maladie ainsi que sa distribution géographique. L’enquête a démontré une variation importante de la prévalence entre les aires de santé de deux ZSR du territoire, Kimpese et Nsona-Mpangu, ainsi qu’une sous-notification importante des cas d’UB dans le système de rapportage basé à l’hôpital.

La troisième étude discute le rôle potentiel de la vaccination BCG contre l’UB.

Enfin, la quatrième et dernière étude se propose d’évaluer sur le plan clinique et épidémiologique, l’impact de la décentralisation du dépistage et du traitement de l’UB dans le territoire de Songololo, en comparaison au modèle précédent qui limitait la prise en charge des patients UB aux centres de référence.

En conclusion, notre travail montre que, compte tenu du caractère focalisé et d’une distribution éparse de la maladie, la détection précoce des cas d’UB doit aller de pair avec une décentralisation de la prise en charge. Sans quoi cette détection n’apportera que très peu dans la mise en œuvre d’une lutte efficace contre l’UB. Cette décentralisation de la prise en charge est faisable en travaillant à travers le réseau des services de santé de premier échelon, même dans une zone difficile d’accès comme Songololo.

Samenvatting

Buruli ulcus (BU), een van zeventien verwaarloosde tropische ziekten, is momenteel een groot probleem voor de volksgezondheid in verschillende afrikaanse landen, zoals Benin, Kameroen, Ivoorkust, de Democratische Republiek Congo en Ghana. BU manifesteert zich onder verschillende vormen , de meest spectaculaire ervan zijn grote necrotiserende zweren met zwaar invaliderende sekwellen. In endemische foci is de ziekte zeer goed gekend in de getroffen gemeenschappen.

In de Democratische Republiek Congo (DRC) is met name het grondgebied van Songololo, in de provincie Bas-Congo, één van de belangrijkste haarden van BU in het land. Sinds 2004 coordineert het ziekenhuis “Institut Médical Evangélique” (IME) in Kimpese een gespecialiseerd programma ter bestrijding van BU. Drie jaar na de lancering van deze bestrijdings-activiteiten in twee gezondheidszones bleven de morbiditeit en arbeidsongeschiktheid als gevolg van UB echter hoog. De lange hospitalisatie vomde ook een ernstig financieel obstakel voor de patiënten en hun families, ondanks het feit het ziekenhuis zorgen voor UB gratis verstrekte. Er werd daarom vanaf 2007 werk gemaakt van een gedecentraliseerd en geïntegreerd systeem van BU bestrijding op het grondgebied van Songololo en de bedoeling van ons doctoraatsonderzoek was om de impact hiervan te bestuderen. We evalueerden het bestrijdingsprogramma in termen van operationele effectiviteit, efficientie, kwaliteit, schaalvergroting en duurzaamheid, met als uiteindelijke bedoeling de betekenis hiervan in te schatten voor de andere haarden van BU in de Democratische Republiek Congo (RDC).

Ons doctoraats-onderzoek startte met klinisch onderzoek over BU (deel I) om te eindigen met het evalueren van een bestrijdingsprogramma (deel II) en behelst een periode van twaalf jaar (2002-2014). In de eerste plaats concentreerden we ons op de BU gevallen die in de periode 2002- 2007 werden opgenomen in het Algemene Referentie Ziekenhuis (Hôpital

Général de Référence (HGR)) van het Medische Evangelische Instituut (IME) in Kimpese. Daarna bestudeerden we alle gevallen die geregistreerd werden tussen 2008 en 2014 in beide Rurale Gezondheids Zones (Zone de Santé Rurale (ZSR)) van het gebied Songololo: de ZSR Kimpese en de ZSR Nsona-Mpangu.

In het klinische deel beschrijft de eerste publicatie de ervaringen van het ziekenhuis IME/Kimpese met de behandeling van BU patiënten tussen 2002 en 2004. Deze studie toonde aan dat 70,5% van de waarschijnlijke gevallen werden bevestigd door het laboratorium en in 38,9% van de gevallen ging het om een recidief. In de groep bevestigde gevallen bleek dat 94,4% patienten waren met zwerende letsels en 25% had een functionele beperking op het moment van de diagnose. 16,7% van de patiënten vertoonden botletsels- bevestigd door het laboratorium. De mediane tijd tussen optreden van symptomen en diagnose was 60 dagen. De mediane duur van de ziekenhuisopname was 102 dagen, 37% van de patiënten genazen met een bewegingsbeperking en de mortaliteit was 19,4%. De tweede publicatie gaat over het beleid bij de oedemateuze vorm van BU, een lokalisatie in het aangezicht. De diagnostische en therapeutische problemen en de noodzaak van een conservatieve behandeling worden toegelicht. Bovendien toonden we het gunstige effect van specifieke antibiotica aan: ze verminderen zwelling en beperken de omvang van de chirurgische excisie. Het derde artikel behandelt de differentiële diagnose van BU. Het toont de moeilijkheden die zich kunnen voordoen, zelfs in een bekend endemisch gebied, wanneer men BU wil diagnosticeren op een puur klinische basis. Het benadrukt het belang van aanvullende laboratoriumtests, waaronder histopathologische analyse. De vierde publicatie bespreekt de behandeling van BU letsels in het aangezicht, aan de hand van ervaringen over een periode van vijf jaar, van 2003 tot 2007. De frequentie van deze letsels in het gezicht was 7% in de groep van bevestigde gevallen. De klinische respons op specifieke antibiotische therapie is beter voor de vroegtijdige letsels. Deze studie benadrukt het belang van de opleiding van gezondheidswerkers in BU

18 endemische gebieden voor een vroege diagnose en snelle verwijzing van de patiënt, die soms zeer atypische lokalisaties van het letsel vertoont.

Het tweede deel van de thesis bekijkt BU vanuit een volksgezondheid perspectief. De eerste publicatie is een voor-na vergelijking die het effect van het BU bestrijdingsproject onderzoekt, drie jaar na de start ervan. We stelden een aanzienlijke toename van het aantal gedetecteerde gevallen vast en een drastische verandering van het klinische profiel van de BU patiënten in het IME/Kimpese ziekenhuis. De tweede publicatie is gebaseerd op data van een exhaustieve census in het grondgebied Songololo in juli-augustus 2008, die de omvang van het BU probleem en de geografische spreiding ervan laten zien. Uit de enquête blijkt een grote variatie in prevalentie tussen de twee gezondheidszones in het gebied, Kimpese en Nsona-Mpangu, en tevens een aanzienlijke onderrapportage van het werkelijke aantal BU gevallen in de statistieken van het ziekenhuis. De derde studie is een bijdrage aan de literatuur over de ontwikkeling van de BCG-vaccinatie en zijn potentiëel tegen BU. Ten slotte beschrijft de vierde en laatste studie de impact van de decentralisatie van de screening en behandeling van BU op het grondgebied van Songololo, in vergelijking met de aanpak voordien die BU behandeling enkel aanbood in enkele referentiecentra. Samenvattend toont ons doctoraatswerk aan dat, gezien het zeer gelokaliseerd voorkomen van de ziekte, de vroegtijdige opsporing van gevallen van BU hand in hand moet gaan met decentralisatie van de zorg. Zoniet vormt ze een struikelblok voor een doeltreffende bestrijding. Deze decentralisatie is haalbaar, ook in moeilijk toegankelijke gebieden als Songololo.

CHAPITRE I. INTRODUCTION GENERALE

I.1. Définition

L’ulcère de Buruli (UB) est une maladie nécrosante de la peau, du tissu sous-cutané et de l’os, d’origine infectieuse. L’agent causal est une mycobactérie de l’environnement appelée Mycobacterium ulcerans (Portaels F et al., 2009). Le début de la maladie est habituellement caractérisé par un nodule indolore qui évolue en l’absence de traitement vers une ulcération étendue, suivie de séquelles très invalidantes telles des contractures déformantes, des amputations des membres et des pertes d’organes comme l’œil, les seins et les organes génitaux (Asiedu K et al., 2000). Mycobacterium ulcerans est une mycobactérie de l’environnement à croissance lente. Pour se développer, M. ulcerans a besoin d’une température située entre 29 et 33°C et d’une concentration en oxygène (2,5%) (Portaels F et al., 2001). La pathogénicité de M. ulcerans est liée à la production d’une exotoxine lipidique appelée Mycolactone, laquelle a des propriétés cytotoxiques et immunosuppressives, responsables de la nécrose tissulaire. C’est le premier macrolide connu produit par un agent pathogène pour l’homme et le seul identifié dans le genre Mycobacterium (Portaels F et al., 2001).

I.2. Description clinique de l’UB

Signes et symptômes

Initialement, l’UB se caractérise souvent par une tuméfaction indolore (papule, rare en Afrique ou un nodule). Il peut aussi se présenter au début sous forme d’une induration indolore (plaque) ou d’un œdème diffus indolore. C’est le pouvoir immunosuppresseur local de la mycolactone permet à la maladie d’évoluer sans douleur ni fièvre (Asiedu K et al., 2000 ; WHO, 2013). Sans traitement ou parfois sous traitement antibiotique, ces lésions évoluent vers un ulcère indolore à bords creusés aboutissant souvent à des séquelles invalidantes. Il arrive parfois que les os soient détruits (Asiedu K et al., 2000 ; WHO, 2013).

La localisation des lésions

L’UB peut se développer sur toutes les parties du corps humain. En général, les lésions siègent préférentiellement sur les membres, 55% sur les membres inférieurs, 35% sur les membres supérieurs et 10% sur d’autres parties du corps (Asiedu K et al., 2000 ; WHO, 2013 ; Portaels F et al., 2001). La variabilité dans la distribution anatomique des lésions peut être due à des différences dans la fréquence des contacts entre les différentes parties du corps et l’agent causal dans l’environnement (Smith JH, 1970).

Manifestations cliniques

Les formes cliniques de l'UB peuvent être classées en 3 grands groupes: les formes pré-ulcératives (papule, nodule, plaque et œdème), l'ulcère et la cicatrice (Asiedu K et al., 2000 ; Portaels F et al., 2001 ; WHO, 2005). Dans de nombreux pays où l’UB est endémique, 70 à 100 % des patients se présentent avec des lésions ulcéreuses et 0 à 30 % avec des lésions non ulcéreuses (WHO, 2013). Une papule est une tuméfaction cutanée indolore de moins de 1 cm de diamètre (Figure 3). Un nodule est une lésion ferme, indolore, palpable, de 1 à 2 cm de diamètre, adhérente à la peau et mobile sur les tissus sous-jacents (Figure 4). Une plaque est caractérisée par une lésion ferme et indurée, indolore, bien démarquée et surélevée, dépassant 2 cm de diamètre, avec des bords irréguliers (Figure 5). Un œdème est une tuméfaction diffuse, ferme, à bords mal définis, généralement non douloureux et ne prenant pas le godet (Figures 6, 7a et 7b). Un ulcère est une plaie indolore, ayant un centre nécrosé, des bords décollés et une peau foncée et épaissie autour (Figure 8 et 9). La cicatrice caractéristique est fibreuse, rétractile, souvent affaissée en étoile (Figures 10 et 11), pouvant occasionner la limitation de

24 mouvements des articulations avoisinantes (Asiedu K et al., 2000 ; Portaels F et al., 2001). Les atteintes osseuses peuvent survenir soit de proche en proche (ostéomyélites de contiguïté), soit à distance des lésions évolutives (ostéomyélites hématogènes ou métastatiques) (Lagarrigue V et al., 2000 ; Portaels F et al., 2003 ; Portaels F et al., 2008). Elles concernent surtout les extrémités (Portaels F et al., 2009) quoique les autres localisations peuvent être également intéressées (Figures 12a et 12b).

I.3. Epidémiologie

Ampleur du problème

L’UB est l’une des 17 maladies tropicales négligées (MTN) (WHO, 2013). L’UB sévit selon le mode endémique en zones intertropicales et touche surtout les populations démunies vivant en milieu rural près des rivières et des marécages (Asiedu K et al., 2000). Sur le plan global, un intérêt sans cesse croissant a été accordé à la maladie ces dix dernières années sous l’égide de l’Initiative Mondiale de l’Organisation Mondiale de la Santé contre l’UB (IMUB) (WHO, 2008). Cependant, la maladie reste encore largement ignorée par de nombreux programmes nationaux de santé publique dans le monde, faisant ainsi partie des MTN les plus méconnues et sa prévalence globale n’est pas bien connue (WHO, 2013 ; Aujoulat I et al., 1996 ; Hotez PJ et al., 2007). L’IMUB a rapporté que seulement 15 sur une trentaine de pays connus endémique, ont transmis des données en 2012 (WHO, 2013).

L’UB a été rapporté dans au moins 33 pays des zones de climat tropical, subtropical et tempéré, mais seuls 25 d’entre eux ont eu à ce jour des cas confirmés par le laboratoire (WHO, 2013 ; WHO, 2008 ; Portaels F et al. 2001 ; Janssens P et al. 2005). Les foyers endémiques sont répartis en Afrique, en Amérique latine, en Asie et en Océanie. L’Afrique subsaharienne est la région du monde la plus touchée. Durant les deux dernières décennies, l’UB y a émergé dramatiquement, en particulier en

Afrique de l’ouest (Asiedu K et al., 2000 ; WHO, 2013 ; WHO, 2008). L’UB touche essentiellement les populations pauvres vivant en milieu rural et ayant un faible accès aux services de santé. Il peut survenir à tout âge et près de la moitié des personnes affectées en Afrique sont des sujets de moins de 15 ans (WHO, 2013 ; WHO, 2008). Le pic de détection se situe habituellement entre 10 -15 ans et 75-79 ans (Debacker M et al., 2004). N’entraînant que peu de décès, elle est à l’origine d’une morbidité sérieuse et notamment d'incapacités fonctionnelles provoquant des problèmes socio-économiques et de développement permanents (WHO, 2008 ; Asiedu K et al., 1998). La proportion des patients développant des incapacités permanentes peut atteindre jusqu’à 20-25% du total du nombre des patients soignés dans les services de santé en Afrique (Portaels F et al., 2001). L’UB, comme de nombreuses autres MTN, crée souvent un problème économique majeur pour les familles touchées et pousse les familles davantage dans la pauvreté. Le coût moyen du traitement a été estimé à US $ 783 parmi les malades suivis dans un hôpital de district du Ghana entre 1994 et 1996 (Asiedu K et al.1998). Parmi 102 patients d’UB traités chirurgicalement entre 2000 et 2005, Kibadi rapporte un coût direct approximatif d’US $ 520 pour ceux ayant des formes ulcérées (Kibadi K, 2005). Bafende et al. rapportent un coût direct moyen de US $ 466 parmi les malades traités à l’hôpital IME/Kimpese, dans le territoire de Songololo, en RDC, où la majorité de personnes vivent avec moins de 1 US $ par jour (Bafende AE et al., 2004). Dans les hôpitaux en Afrique, la durée d’hospitalisation moyenne est d’environ trois mois et impose à la même période la co-hospitalisation d’un parent non malade pour s’occuper du patient (Asiedu K et al., 2000). Une étude récente, réalisée au Cameroun dans deux hôpitaux avec programme spécialisé contre l’UB, a montré que malgré la gratuité des soins offerts, le coût direct non- médical, la perte de productivité pour les patients et pour les ménages étaient énormes. Les auteurs rapportent que 62% des familles ont interrompu leur soutien social et financier au patient, conduisant à l’isolement social de ce dernier à l’hôpital (Peeters Grietens K et al.,

2008). D’autres conséquences sociales sont : l’interruption de la scolarité pour les enfants malades, la privation de l’attention d’un des parents pour les enfants restés à domicile, l’opprobre, l’isolement social, les perspectives amenuisées de mariage et les divorces pour les femmes malades (Asiedu K et al., 2000). Kibadi et al. ont répertorié près de vingt appellations de l’UB rattachées à l’incurabilité de la maladie ou à la sorcellerie dans neuf pays endémiques africains (Kibadi K et al., 2007). Ainsi, le recours au traitement traditionnel en premier ressort est très important dans la plupart des communautés affectées (Muelder AA et al., 2008 ; Stienstra Y et al., 2002 ; Renzaho AM et al., 2007). Par conséquent, les patients se présentent souvent aux structures de santé aux stades avancés de la maladie avec comme corollaires, l’indication fréquente du traitement chirurgical, une longue durée d’hospitalisation, le développement des séquelles invalidantes et un coût exorbitant (WHO, 2008 ; Asiedu K et al., 1998 ; Sizaire V et al., 2006). Les rechutes sont une autre cause de souffrance supplémentaire pour les patients. Des taux variables de rechutes après traitement chirurgical uniquement sont décrits dans la littérature, allant de 6.1% au Bénin (Debacker M et al., 2005), 16% au Ghana (Amofah G et al., 1998) à 22% en RDC (Kibadi K, 2005).

Fréquence et facteurs de risque

L’UB est à l’heure actuelle chez l’homme, la troisième maladie mycobactérienne après la tuberculose et la lèpre, et la moins comprise des trois (Portaels F, 2009 ; Sizaire V et al., 2006). Dans certains pays d’Afrique, notamment le Bénin, l’UB est pour le moment la deuxième mycobactériose après la tuberculose (Sopoh GE et al., 2007). L’IMUB rapporte que chaque année, 15 des 33 pays considérés endémiques dans le monde notifient entre 5000 et 6000 cas (WHO, 2013). En Afrique, environ 48% des malades sont des sujets de moins de 15 ans, alors qu’en Australie ils ne représentent que 10% des cas et au Japon 19% (WHO, 2013. Ceci semble dire que l’age n’est pas un facteur de risque particulier, puisque la distribution d’âge des cas d’UB semble suivre celle de la

27 population générale. La répartition de l’UB par sexe varie entre les différents foyers endémiques : en Afrique, le rapport entre le sexe masculin et féminin est de 52% sur 48%, et le sexe n’y représenterait donc pas un facteur de risque particulier, tandis qu’au Japon, les femmes semblent plus exposées, avec un rapport hommes/femmes de 34%/66% (WHO, 2013). A ce jour, ni la prévalence, ni la population à risque ne sont bien déterminées sur le plan mondial (Hotez PJ et al., 2007). Les taux d’incidence varient énormément d’un continent à l’autre, d’un pays à l’autre et entre différents foyers endémiques dans un même pays. Des estimations précises de l’incidence par région ou pour tout le pays sont rarement disponibles (WHO, 2004a). Une étude réalisée en 2004 au Bénin, dans la commune de Lalo, avait confirmé le caractère très focalisé de la distribution l’UB et a pu ainsi montrer que l’unité géographique pertinente d’analyse de l’UB était une entité plus réduite que le niveau national ou régional (Johnson RC et al., 2005).

Figure 1. Répartition géographique de l’UB en 2012 dans le monde (www.who.int).

Les principaux facteurs de risque rapportés dans des études cas-témoins sont un âge inférieur à 15 ans (Debacker M et al., 2006), l’exposition à

28 l’eau non potable (Debacker M et al., 2006 ; Nackers F et al., 2007 ; Pouillot R et al., 2007), le manque d’hygiène personnelle (Nackers F et al., 2007), le port d’habits courts durant les travaux de champs (Pouillot R et al., 2007) et récemment l’infection à VIH (Johnson RC et al., 2008). Dans leur étude sur les facteurs de risque pour l’UB au Bénin, Debacker et al. rapportent qu’après ajustement pour la cicatrice de vaccination BCG, le lieu de résidence et les sources d’eau, les enfants de 5 à 14 ans avaient un risque élevé pour l’UB (Debacker M et al., 2006). Le risque pour l’UB était également associé à l’exposition à l’eau non potable (Debacker M et al., 2006). Au Cameroun, Pouillot et al. ont montré aussi que le contact avec les marécages du fleuve Nyong était un facteur de risque pour UB (Pouillot R et al., 2007). Dans la même étude, ils ont trouvé que le port des habits courts durant les activités de champs était un facteur de risque pour l’UB (Pouillot R et al., 2007). Une autre étude cas-témoins réalisée au Bénin avait suggéré que des simples mesures d’hygiène pouvaient être importantes dans la prévention de l’UB. En effet, se laver avec du savon, et soigner les plaies et écorchures avec du savon ou une poudre antibiotique semblaient protecteurs (Nackers F et al., 2007). Johnson et al ont récemment rapporté l’association entre l’infection HIV et UB (Johnson RC et al., 2008).

Transmission

Il n’y a pas d’évidences qui permettraient d’évoquer une contagion interhumaine. En même temps, le mécanisme exact de la transmission de Mycobacterium ulcerans de l’environnement à l’homme n’est pas encore élucidé (Portaels F et al., 2009). Les deux hypothèses prépondérantes proposées dans la littérature sont l’introduction du germe dans la peau soit par traumatisme, depuis une surface cutanée contaminée, soit par piqûres d’insectes aquatiques (Meyers WM et al., 1974a ; Portaels F et al., 1999 ; Portaels F et al., 2001 ; Marsollier L et al., 2002 ; Debacker M et al., 2003 ; Silva MT et al., 2007).

I.4. Diagnostic

Clinique

Bien qu’en zone d’endémie un personnel de santé expérimenté soit capable de poser le diagnostic de présomption sur base de la clinique, la confirmation par le laboratoire revêt un caractère crucial, en particulier depuis l’introduction de l’antibiothérapie en 2004 (WHO, 2004b). Les éléments cliniques et épidémiologiques révélateurs de l’UB sont les suivants : (i) la présence chez un patient d’une ou plusieurs formes cliniques citées ci-haut ; (ii) les lésions sont habituellement indolores ; (iii) les lésions sont le plus souvent localisées sur les membres ; (iv) la notion de résidence ou de séjour dans une zone endémique connue ; (v) la plupart des patients ont moins de 15 ans (Asiedu K et al., 2000 ; Portaels F et al., 2001). Après le diagnostic clinique, il convient de recueillir les échantillons avant le traitement du patient. Trois techniques de prélèvement sont utilisées: écouvillonnage, aspiration à l’aiguille fine et biopsie (à l'emporte-pièce ou chirurgicale). Comme les mycobactéries ne sont pas réparties de façon homogène dans les lésions, il convient de recueillir au moins deux échantillons cliniques de chaque lésion. En routine, les techniques moins invasives, telles les écouvillonnages et l’aspiration à l’aiguille ﬁne, suffisent pour la confirmation de l’UB. Les prélèvements d'écouvillons et les biopsies à l'aiguille fine sont des procédures simples, pouvant être faites à n'importe quel niveau des soins, au cours de la prise en charge en routine ou pendant la recherche des cas dans les communautés (Portaels F, 2014). Pour les analyses bactériologiques différées ou loin des sites de prélèvement, les différents types d’échantillons recueillis ont été transportés avec succès dans un bouillon de base de Dubos avec de l’albumine pour bouillon de Dubos additionnée de PANTA ou un bouillon de Dubos avec 10 % d’OADC et du PANTA (Eddyani M et al., 2008 ; Eddyani M et al., 2009 ; Herbinger KH et al., 2009 ; Siegmund V et al., 2007 ;

Yeboah-Manu D et al., 2004). Ces milieux permettent aussi l’analyse des échantillons par PCR. Les biopsies destinées à l’analyse histopathologique doivent être placées dans du formol à 10 % (Portaels F et al., 2001).

Ecouvillons

L’écouvillonnage est la technique la plus utilisée face à des ulcères qui ont des bords creusés. Les échantillons obtenus par écouvillonnage doivent être prélevés sur les bords creusés de l’UB.

Aspiration à l’aiguille fine

Elle est principalement utilisée pour prélever après le diagnostic clinique des échantillons à partir de lésions non ulcéreuses telles nodule, plaque ou œdème. Seuls des médecins ou des agents de santé expérimentés devraient l’exécuter ; les agents de santé devront bénéficier de formation continue et d’une supervision régulière pour améliorer leurs compétences.

Biopsie (à l’emporte-pièce ou chirurgicale)

Les biopsies ne constituent plus les prélèvements de choix et ne sont utilisées que dans un nombre limité de circonstances. On préfèrera la biopsie chirurgicale lorsque l’on doit prélever de grands spécimens pour l’analyse histopathologique.

Méthodes d’analyse au laboratoire

Il n’existe pas encore de méthodes diagnostiques rapides de l’UB (tests dermiques ou sérologiques) qui puissent permettre de détecter l’infection chez les sujets asymptomatiques, avant qu’ils ne développent la première lésion de la peau (Portaels F et al., 2009). Les quatre tests de laboratoire de confirmation de l’UB actuellement disponibles sont l’examen de frottis avec la méthode de Ziehl-Neelsen (ZN), la culture in vitro (CUL), l’examen histopathologique (HIS) et l’amplification génique (PCR) (Asiedu K et al., 2000 ; Portaels F et al., 2001). L’OMS préconise au moins deux tests positifs sur les quatre pour confirmer formellement le diagnostic de l’UB. Cependant, en zones

31 endémiques, c’est l’examen de frottis avec la méthode de ZN qui est souvent la seule disponible, les trois autres techniques ne peuvent être réalisées que dans des laboratoires bien équipés éloignés des foyers (Portaels F et al., 2009). Selon les formes cliniques et les types de prélèvement, la sensibilité du ZN varie de 20 à 80% et pour CUL de 20 à 60% (WHO, 2008 ; Portaels F et al., 2001 ; Eddyani M et al., 2008 ; Eddyani M et al., 2009 ; Phillips R et al., 2005 ; Phillips R et al., 2009). Indépendamment de la forme clinique, HIS a une sensibilité d’environ 90% et la PCR de plus de 90% (WHO, 2008 ; Portaels F et al., 2001).

L’examen direct L’examen direct des frottis à la recherche des bacilles acido-alcoolo résistants (BAAR) est le premier examen disponible pour la confirmation de l’UB en zones endémiques (Portaels F, 2014). Il peut être pratiqué à partir des écouvillons, des biopsies ou des aspirations à l’aiguille fine. Après coloration de Ziehl-Neelsen, les BAAR apparaissent en rouge sur fond bleu, sous-forme des bâtonnets droits ou incurvés, isolés ou regroupés en corde ou en globis.

La culture in vitro M. ulcerans est une bactérie à croissance lente qui peut être cultivée in vitro à 32°C sur un milieu classique pour les mycobactéries (MacCallum P, 1948). Les premiers résultats sont obtenus au bout de 6 à 8 semaines à l’aide des protocoles classiques d’identification (Lévy-Frébault VV et al., 1992). A cause de ce délai, les cultures ne sont pas indispensables pour le diagnostic de l’infection et la prise en charge clinique immédiate des patients. Elles sont néanmoins utiles pour l’identification des cas d’échecs thérapeutiques et des rechutes de l’infection (Portaels F, 2014).

L’amplification génique (PCR) Elle peut être réalisée à partir d’écouvillons, d’aspirations à l’aiguille fine ou de biopsies.

Cette méthode permet d’amplifier de petites quantités d’ADN afin d’en obtenir des quantités facilement décelables au laboratoire. Elle cible la séquence d’insertion IS2404, présente à raison de plus de 200 copies dans le génome de M. ulcerans (Ross BC et al., 1997 ; Stinear T et al., 1999). L’OMS recommande actuellement qu’au moins 70 % des cas cliniques suspects d’UB soient conﬁrmés par un test positif par PCR. Un réseau de laboratoires effectuant la PCR a été mis en place dans les pays d’endémie(Portaels F, 2014).

L’histopathologie Cette méthode requiert des biopsies. Elle est utile pour faire le diagnostic différentiel et surveiller les réactions inattendues au traitement. Lorsque ces services sont disponibles, l’obtention des résultats est assez rapide. Sa Sensibilité de 90 % environ. Le nouveau manuel de l’OMS sur le diagnostic de l’UB au laboratoire décrit en détails les images histopathologiques (Portaels F, 2014).

Méthodes de détection de la mycolactone La toxine de M. ulcerans se répartit plus uniformément dans une lésion que les bacilles, ce qui semble indiquer que la mise au point d’un anticorps (ou d’un autre réactif) anti-mycolactone pourrait aboutir à la production d'un test diagnostique rapide pour le terrain (WHO, 2008). Les nouvelles méthodes de détection de la mycolactone dans les tissus donnent des résultats encourageants (George KM et al., 2000 ; Hong H et al., 2003 ; Hong H et al., 2005 ; Hong H et al., 2008 ; Spangenberg T et al., 2010 ; Kishi Y, 2011 ; Converse PJ et al., 2014). Les échantillons prélevés sur les parties nécrosées des lésions ulcéreuses peuvent fournir de la mycolactone pour les analyses. Pour les lésions non-ulcéreuses en particulier, de nouvelles études sont nécessaires pour déterminer la faisabilité de la détection de la mycolactone à partir d’échantillons obtenus par les techniques simples de prélèvement, utilisées en routine (WHO, 2013).

I.5. Traitement

Jusqu’il y a tout récemment, le protocole thérapeutique recommandé par l’OMS pour le traitement de l’UB était le suivant : Pour la catégorie I (une lésion unique de moins de 5 cm de diamètre) : association antibiotique (rifampicine et streptomycine) uniquement pendant un délai minimum de 4 semaines. Si une excision est indiquée, elle aura lieu au moins 24 heures après début des antibiotiques. Pour la catégorie II (une lésion unique de 5 à 15 cm) : association antibiotique pendant au moins 4 semaines, chirurgie si nécessaire, puis poursuivre l’antibiothérapie pendant 4 semaines supplémentaires. Enfin, pour la catégorie III [lésion unique dépassant 15 cm de diamètre, des lésions multiples, une ou plusieurs lésions sur des localisations délicates (œil, sein, organes génitaux) et une ostéomyélite] : antibiothérapie pendant 1 semaine, suivi de la chirurgie, puis poursuite de l’antibiothérapie pendant 7 semaines (WHO, 2004b). Le suivi pour la surveillance des rechutes est de 12 mois. Selon un rapport de l’OMS, les rechutes signalées dans le passé, allant de 16 à 30% des cas traités uniquement par chirurgie, ont été ramenées à moins de 2% après l’introduction des antibiotiques (WHO, 2008). Le traitement avec la rifampicine et la streptomycine, ou avec la rifampicine et d’autres schémas thérapeutiques par voie orale, a permis une décentralisation du traitement par rapport à l’époque où le traitement chirurgical à l’hôpital était la seule option (Johnson PD et al., 2005 ; Etuaful S et al., 2005). Avec le protocole thérapeutique actuel (WHO, 2012), l’antibiothérapie spécifique est davantage mise en avant plan. La chirurgie est désormais principalement utilisée pour le débridement et pour effectuer des greffes sur les ulcérations importantes afin d’accélérer la cicatrisation. Les interventions pour réduire le plus possible ou prévenir les incapacités sont nécessaires selon le stade de la maladie. Aujourd’hui, 30% à 50% des cas sont traités uniquement par antibiotiques, ce qui facilite l'accès au traitement.

I.6 Contrôle de la maladie

Le diagnostic et le traitement précoces sont à ce jour les seuls moyens pour réduire au maximum la morbidité et éviter les incapacités (Asiedu K et al., 2000). Nous listons les stratégies d’interventions qui sont théoriquement possibles et compatibles avec les connaissances actuelles contre la transmission. La figure 2 est une représentation schématique simplifiée d’une réalité bien plus complexe dont divers aspects restent à clarifier (Sizaire V et al., 2006).

L’espoir d’un vaccin Il n’existe pas de vaccin spécifique contre l’UB à l’heure actuelle (WHO, 2008). Des études observationnelles, réalisées au Bénin et au Cameroun, ont suggéré que la vaccination BCG protégeait les enfants, aussi à un certain degré les adultes, contre les formes sévères de l’UB, en particulier les ostéomyélites et les lésions disséminées (Portaels F et al., 2002 ; Portaels F et al., 2004 ; Noeske J et al. 2004), tandis qu’il n’y a aucune évidence de réduction du risque d’infection par la vaccination BCG de routine (Nackers F et al., 2006). Dans une perspective de santé publique, la vaccination BCG de routine a donc peu de valeur pour le contrôle de l’UB dans la population. L’identification et le développement des candidats-vaccins spécifiques contre l’UB sont devenus une priorité pour un meilleur contrôle de l’UB (Huygen K et al., 2009). Les perspectives courantes pour une vaccination contre l’UB impliquent l’amélioration de vaccins à base de BCG, l’atténuation rationnelle d’un isolat vivant de Mycobacterium ulcerans, la production de vaccins sous-unités protéiques (à partir des protéines de surface) et le développement de l’immunogénicité de la mycolactone (Portaels F et al., 2009).

Lutte contre le réservoir et diminution de contact homme- réservoir Présentement, nous disposons d’évidences suffisantes pour considérer que la maladie est liée à l’eau, notamment les eaux stagnantes des zones

35 marécageuses (Debacker M et al., 2006 ; Silva MT et al., 2007 ; Portaels F, 1989 ; Portaels F et al., 2008). Néanmoins, les modes de transmission de l’agent causal de l’environnement à l’homme n’ont pas encore été clairement établis (Portaels F et al., 2009, Meyers WM et al., 1974a ; Duker AA et al., 2006). En outre, l’éradication des niches aquatiques de Mycobacterium ulcerans paraît impossible, particulièrement dans les pays endémiques africains (Sizaire V et al., 2006).

Lors d’une étude de prévalence, réalisée dans un arrondissement de la commune de Lalo au Bénin en 2002, Johnson et al. ont observé une relation inverse entre la prévalence de l’UB et la distance à la rivière Couffo. En effet, le nombre de patients pour 1000 habitants augmentait graduellement de 0.58 à 32.62 avec la diminution de la distance à la rivière de 10 Km à 1 Km. Au-delà de cette observation, ils ont également constaté que l’UB était plus prévalent dans un village appelé Tandji où les familles utilisaient directement l’eau de la rivière à des fins domestiques contrairement à d’autres villages qui disposaient des sources d’eau protégées. Cette étude suggère que l’utilisation des sources d’eau non protégées pourrait jouer un rôle dans la prévalence élevée de l’UB dans le village susmentionné. Ils ont suggéré que la fréquence de la maladie pourrait diminuer suite au développement des programmes de santé publique basés sur les stratégies d’approvisionnement en eau potable (Johnson RC et al., 2005). Une préoccupation demeure cependant, certains groupes professionnels comme les agriculteurs, les pêcheurs et d’autres personnes qui, pour des raisons de survie, s’exposent fréquemment à ces zones humides. En effet, le problème que pose l’UB n’est pas seulement d’ordre médical ; il dépend également du développement socio-économique général des régions endémiques (Asiedu K et al., 1998).

Détection précoce et prise en charge de l’infection On estime que la durée d’incubation est égale à environ 1 à 3 mois (Andersen FO, 1965 ; Barker DJP, 1973). Elle peut cependant varier de 2

36 semaines à plusieurs années (Meyers WM et al., 1974a). En Australie, une étude récente réalisée dans le Victoria a rapporté une période d’incubation moyenne de 19,2 semaines (minumum 4,5 semaines, maximum 37,7 semaines) (Trubiano JA et al., 2013). A ce jour, il n’existe pas encore des méthodes diagnostiques rapides de l’UB (tests dermiques ou sérologiques) utilisables sur le terrain qui puissent permettre de détecter l’infection chez les sujets asymptomatiques, avant qu’ils ne développent la première lésion de la peau (Asiedu K et al., 2000 ; WHO, 2013). Des recherches sont en cours pour en mettre un au point. A l’heure actuelle, les efforts de contrôle sont concentrés sur la promotion de la détection précoce des cas, à l’aide du dépistage actif des cas, de l’information et l’éducation sanitaire, et un traitement rapide adéquat (WHO, 2008). Des études ont montré que la détection et le traitement précoces permettaient d’éviter à un coût abordable, les conséquences sérieuses de l’UB (Amofah G et al., 1998 ; Kanga JM et al., 2006 ; Kanga JM et al., 2007). Au Ghana, Amofah et al. ont mené une étude prospective entre 1993 et 1995 dans un foyer endémique du district d’Amansie west et ont démontré l’efficacité opérationnelle, dans la prévention des invalidités, de l’excision de lésions pré-ulcératives d’UB suivie de suture primaire sur le terrain. La durée médiane pour la cicatrisation des nodules excisés était de 31,5 jours et aucune invalidité n’a été notée après traitement (Amofah G et al., 1998). Entre 1998 et 1999, Evans et al. ont confirmé les observations de la précédente étude. Ils ont remarqué que la détection précoce des nodules et le traitement de proximité avaient modifié la présentation clinique de l’UB à l’hôpital par une augmentation des formes débutantes diagnostiquées (Evans MR et al., 2003). Des programmes similaires développés depuis 1998 en Côte-d’Ivoire ont eu un impact immédiat et à long terme. Le taux de cicatrisation en un mois était supérieur à 96% et une inversion du rapport entre formes non ulcérées et ulcères a été constatée les années suivantes (Kanga JM et al., 2006 ; Kanga JM et al., 2007 ; Ecra E et al., 2005). Il ressort de toutes ces expériences que les efforts de santé publique, focalisés sur le dépistage

37 précoce et le traitement, réduisent fortement la morbidité de l’UB et le coût (Asiedu K et al., 1998 ; Amofah G et al., 1998 ; Johnson PD et al.,2005). Cependant une bonne surveillance, la détection et le traitement précoces ne seront pas probablement suffisants pour réduire à long terme l’incidence et la prévalence de l’UB (Sizaire V et al., 2006 ; Huygen K et al., 2009).

Prévention Iaire Sujet non infecté

Sujet infecté Non malade (Infection primaire)

Sujet infecté Prévention IIaire Infection latente malade

Formes Formes localisées de la disséminées de maladie la maladie

Guérison Nodule Plaque Guérison spontanée Papule œdème spontanée

Guérison Ulcération Ulcération spontanée mineure étendue Cicatrices Ulcérations Guérison atrophiques multiples ou spontanée rétractiles/ Cicatrices étendues Contractures atrophiques Guérison rétractiles/ spontanée Contractures

Prévention IIIaire Prévention IIIaire

Séquelles Guérison Ostéomyélite Guérison Séquelles Ostéomyélite par contiguité invalidantes spontanée hématogène spontanée invalidantes

Décès (complication rare)

Prévention primaire Prévention secondaire Prévention tertiaire

· Education sanitaire · Détection précoce · Chirurgie réparatrice · Accès à l’eau potable et traitement · Réhabilitation physique · BCG? précoce · Réintégration sociale et réadaptation

Figure 2: Modèle épidémiologique de l’UB et interventions possibles Le présent modèle épidémiologique de l’UB (voir figure 1) a été adapté de celui récemment publié par Portaels et al.(1).

I.7. Bref aperçu historique de l’ UB en RDC

En République Démocratique du Congo (RDC), des foyers endémiques ont été décrits dans la quasi majorité des provinces (Meyers WM et al., 1974b). La province du Bas-Congo contient l’un des foyers les plus importants du pays, à savoir le territoire de Songololo, dans le district de Cataractes (Meyers WM et al., 1974 ; Andersen FO, 1965 ; Smith JH, 1970 ; Portaels F, 1989 ; Kibadi K et al., 2003 ; Bafende AE et al., 2003 ; Phanzu DM et al., 2006). Meyers et al. ont rapporté, sur base des interviews réalisés auprès des anciens patients, que l’UB existait dans cette région bien avant 1935 (Meyers WM et al., 1974). Plusieurs noms y sont employés pour décrire la maladie, comme « mputa mpangu », à cause de nombreux cas de plaies chroniques compatibles à l’UB rapportés dans la localité de Mpangu au nord de la ZSR de Kimpese ; « mputa matadi », c’est-à-dire plaie dure comme la pierre ; « mputa mbuaki » ou plaie rouge et « mbasu », qui évoque une origine mystique ou un mauvais sort (Meyers WM et al., 1974 ; Kibadi AK, 2004 ; Kibadi K et al., 2007). Les premiers cas publiés de la région remontent dans les années 60 et 70 (Meyers WM et al., 1974 ; Andersen FO, 1965 ; Smith JH, 1970) suivies par une longue période de près d’un quart de siècle sans cas déclarés. Cependant, un regain d’intérêt sur la maladie a été observé en début des années 2000 (Kibadi K et al., 2003 ; Bafende AE et al., 2004). Après une période de silence épidémiologique d’environ 25 ans, des cas d’UB sont à nouveau rapportés dans les anciens foyers à partir de 2000 (Janssens P et al., 2005). En décembre 2001, un programme national de lutte contre l’UB (PNLUB) a été créé en RDC avec comme objectif global, la réduction de la morbidité et des incapacités liées à l’UB (Plan Directeur, 2005). Depuis sa création, le PNLUB a été largement sous-financé et est donc peu fonctionnel. Une enquête nationale organisée en 2004 a permis d’identifier 487 cas probables d’UB repartis dans les provinces du Bas-Congo, du Bandundu, du Katanga, de Kinshasa, du Maniema et du Sud-Kivu (Kibadi K et al., 2006).

La province du Bas-Congo contient à l’heure actuelle toujours l’un des foyers les plus importants du pays, à savoir le même territoire de Songololo, dans le district de Cataractes (Kibadi K et al., 2003). Depuis 1999, l’hôpital général de référence (HGR) de l’Institut Médical Evangélique (IME) de Kimpese, situé dans ce territoire, à 220 Km au sud- ouest de Kinshasa, reçoit régulièrement des cas d’UB (Bafende AE et al., 2004). Il héberge en son sein depuis fin 2004 le projet de lutte contre l’UB.

I.8. Références

Amofah G, Asamoah S, Fram-Gyening C. Effectiveness of excision of preulcerative Buruli lesions in field situations in a rural district in Ghana. Trop Doct 1998; 28: 81-3.

Andersen FO. Mycobacterial skin ulcers-Clinical experience. Cent Afr J Med. 1965; 11: 131-35.

Asiedu K, Etuaful S. Socioeconomic implications of Buruli ulcer in Ghana: a three-year review. Am J Trop Med Hyg 1998; 59: 1015-22.

Asiedu K, Scherpbier R, Raviglione M. Buruli ulcer-Mycobacterium ulcerans infection. World Health Organization. 2000.WHO/CDS/CPE/GUI/2001.1.

Aujoulat I, Huguet-Ribas MP, Koïta Y. L'ulcère de Buruli: un problème de santé publique méconnu, appelant une mobilisation internationale. Dév Santé 1996; 125: 22-30.

Aujoulat I, Johnson C, Zinsou C, Guedenon A, Portaels F. Psychosocial aspects of health seeking behaviours of patients with Buruli ulcer in southern Benin. Trop Med Int Health 2003; 8: 750-9.

Bafende AE, Phanzu MD, Imposo BB. Buruli ulcer in the Democratic Republic of Congo: epidemiology, presentation and outcome. Trop Doct 2004; 34: 82-4.

Barker DJP. Epidemiology of Mycobacterium ulcerans infection. Trans R Soc Trop Med Hyg 1973; 67: 43-50.

Converse PJ, Xing Y, Kim KH, Tyagi S, Li S-Y, Almeida DV, Nuermberger EL, Grosset JH, Kishi Y. Accelerated Detection of Mycolactone Production and Response to Antibiotic Treatment in a Mouse Model of Mycobacterium ulcerans Disease. PLoS Negl Trop Dis 2014; 8(1): e2618.

Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Portaels F. Buruli ulcer recurrence, Benin. Emerg Infect Dis 2005; 11: 584-9.

Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Scott JT, Dramaix M, Portaels F. Mycobacterium ulcerans disease: role of age and gender in incidence and morbidity. Trop Med Int Health 2004; 9: 1297- 304.

Debacker M, Portaels F, Aguiar J, Steunou C, Zinsou C, Meyers WM, Dramaix M. Risk factors for Buruli ulcer, Benin. Emerg Infect Dis 2006; 12: 1325-31.

Debacker M, Zinsou C, Aguiar J, Meyers WM, Portaels F. First case of Mycobacterium ulcerans disease (Buruli ulcer) following a human bite. Clin Infect Dis 2003; 36: e67-e68.

Duker AA, Portaels F, Hale M. Pathways of Mycobacterium ulcerans infection: a review. Environ Int 2006; 32: 567-73.

Ecra E, Kanga JM, Gbery ID, Sangare A, Kacou E, Kouame K, Kaloga M, Kamelan CO, Yao J. [Detection and treatment of early forms of Mycobacterium ulcerans infection in Ivory Coast]. Med Trop 2005; 65: 334-8.

Eddyani M, Debacker M, Martin A, Aguiar J, Johnson CR, Uwizeye C, Fissette K, Portaels F. Primary culture of Mycobacterium ulcerans from human tissue specimens after storage in semi-solid transport medium. J Clin Microbiol 2008; 46: 69-72.

Eddyani M, Fraga AG, Schmitt F, Uwizeye C, Fissette K, Johnson C, Aguiar J, Sopoh G, Barogui Y, Meyers WM, Pedrosa J, Portaels F. Fine needle aspiration, an efficient sampling technique for the bacteriological diagnosis of nonulcerative Buruli ulcer. J Clin Microbiol 2009; 47: 1700-4.

Etuaful S, Carbonnelle B, Grosset J, Lucas S, Horsfield C, Phillips R, Evans M, Ofori-Adjei D, Klustse E, Owusu-Boateng J, Amedofu GK, Awuah P, Ampadu E, Amofah G, Asiedu K, Wansbrough-Jones M. Efficacy of the combination rifampin-streptomycin in preventing growth of Mycobacterium ulcerans in early lesions of Buruli ulcer in humans. Antimicrob Agents Chemother. 2005; 49: 3182-6.

Evans MR, Phillips R, Etuaful SN, Amofah G, Adomako J, Adjei O, Dennis- Antwi J, Lucas SB, Wansbrough-Jones MH. An outreach education and treatment project in Ghana for the early stage of Mycobacterium ulcerans disease. Trans R Soc Trop Med Hyg. 2003; 97: 159-60.

George KM, Pascopella L, Welty DM, Small PL. A Mycobacterium ulcerans toxin, mycolactone, causes apoptosis in guinea pig ulcers and tissue culture cells. Infect Immun 2000 68: 877–83.

Herbinger KH, Adjei O, Awua-Boateng NY, Nienhuis WA, Kunaa L, Siegmund V, Nitschke J, Thompson W, Klutse E, Agbenorku P, Schipf A, Reu S, Racz P, Fleischer B, Beissner M, Fleischmann E, Helfrich K, van der Werf TS, Löscher T, Bretzel G. Comparative study of the sensitivity of different diagnostic methods for the laboratory diagnosis of Buruli ulcer disease. Clin Infect Dis 2009; 48 :1055–64.

Hong H, Demangel C, Pidot SJ, Leadlay PF, Stinear T. Mycolactones: immunosuppressive and cytotoxic polyketides produced by aquatic mycobacteria. Nat Prod Rep 2008; 25: 447–454.

Hong H, Gates PJ, Staunton J, Stinear T, Cole ST, et al. Identification using LC-MSn of co-metabolites in the biosynthesis of the polyketide toxin mycolactone by a clinical isolate of Mycobacterium ulcerans. Chem Commun 2003: 2822–23.

Hong H, Spencer JB, Porter JL, Leadlay PF, Stinear T. A novel mycolactone from a clinical isolate of Mycobacterium ulcerans provides evidence for additional toxin heterogeneity as a result of specific changes in the modular polyketide synthase. Chem biochem 2005; 6: 643–48.

Hotez PJ, Molyneux DH, Fenwick A, Kumaresan J, Sachs SE, Sachs JD, Savioli L. Control of Neglected Tropical Diseases. N Engl J Med 2007; 357: 1018-27.

Huygen K, Adjei O, Affolabi D, Bretzel G, Demangel C, Fleischer B, Johnson RC, Pedrosa J, Phanzu DM, Phillips RO, Pluschke G, Siegmund V, Singh M, van der Werf TS, Wansbrough-Jones M, Portaels F. Buruli ulcer disease: prospects for a vaccine. Med Microbiol Immunol. 2009; 198: 69- 77.

Janssens P, Pattyn S, Meyers WM, Portaels F. Buruli Ulcer: An Historical Overview with Updating to 2005. Bull Séanc Acad R Sci Outre-Mer 2005; 51: 265-99.

Johnson PD, Stinear T, Small PL, Pluschke G, Merritt RW, Portaels F, Huygen K, Hayman JA, Asiedu K. Buruli ulcer (M. ulcerans infection): new insights, new hope for disease control. PLoS Med 2005; 2: e108.

Johnson RC, Makoutode M, Hougnihin R, Guedenon A, Ifebe D, Boko M, Portaels F. [Traditional treatment for Buruli ulcer in Benin]. Med Trop 2004; 64: 145-50.

Johnson RC, Makoutode M, Sopoh GE, Elsen P, Gbovi J, Pouteau LH, Meyers MW, Boko M, Portaels F. Buruli ulcer distribution in Benin. Emerg Infect Dis 2005; 11: 500-1.

Johnson RC, Nackers F, Glynn JR, DE Biurrun BE, Zinsou C, Aguiar J, Tonglet R, Portaels F. Association of HIV infection and Mycobacterium ulcerans disease in Benin. AIDS 2008; 22: 901-3.

Johnson RC, Sopoh GE, Boko M, Zinsou C, Gbovi J, Makoutode M, Portaels F. [Distribution of Mycobacterium ulcerans (Buruli ulcer) in the district of Lalo in Benin]. Trop Med Int Health 2005; 10: 863-71.

Kanga JM, Kacou ED, Kouame K, Kassi K, Kaloga M, Yao JK, Dion-Lainé M, Avoaka LE, Yoboué-Yao P, Sangaré A, Ecra JE, Ahogo C, Djédjé MS, Kadiri AJ, Ayé C. [Fighting against Buruli ulcer: the Cote-d'Ivoire experience]. Bull Soc Pathol Exot 2006; 99: 34-8.

Kanga JM, Kacou ED, Yao JK, Kassi K, Kouame K, Kaloga M, Kamelan CO. [Impact of Buruli ulcer secondary prevention program in an endemic area in Cote d'Ivoire]. Bull Soc Pathol Exot 2007; 100: 277-81.

Kibadi AK, Singa JN, Wembanyama H, Portaels F. Résultats de l'enquête nationale préliminaire sur l'ulcère de Buruli en République démocratique du Congo. Bull de l'ALLF 2006; 18: 24-26.

Kibadi K, Aujoulat I, Meyers WM, Mokassa L, Muyembe T, Portaels F. [Study of names and folklore associated with Mycobacterium ulcerans infection in various endemic countries in Africa]. Med Trop 2007; 67: 241- 8.

Kibadi K, Tsakala M, Mputu-Yamba JB, Muyembe T, Kashongwe M, Imposo B, Nsiala A. [Buruli ulcer in Angolese refugees in the Kimpese area, Lower Congo, D.R. Congo]. Santé. 2003; 13: 39-41.

Kibadi K. [Mycobacterium ulcerans disease (Buruli ulcer): surgical treatment of 102 cases in the Democratic Republic of Congo]. Med Trop 2005; 65: 444-8.

Kibadi KA. [Knowledge-attitudes-practices survey of the Songololo population (D.R. Congo) about Buruli ulcers]. Bull Soc Pathol Exot 2004; 97: 302-5.

Kibadi K, Boelaert M, Kayinua M, Minuku JB, Muyembe-Tamfum JJ, Portaels F, Lefèvre P. Therapeutic itineraries of patients with ulcerated forms of Mycobacterium ulcerans (Buruli ulcer) disease in a rural health

45 zone in the Democratic Republic of Congo. Trop Med Int Health 2009; 14: 1110-6.

Kishi Y. Chemistry of mycolactones, the causative toxins of Buruli ulcer. Proc Natl Acad Sci USA 2011; 108: 6703–8.

Lagarrigue V, Portaels F, Meyers WM, Aguiar J. L'ulcère de Buruli: attention aux atteintes osseuses! A propos de 33 cas observés au Bénin. Méd Trop 2000; 60: 262-66.

Lévy-Frébault VV, Portaels F. Proposed minimal standards for the genus Mycobacterium and for description of new slowly growing Mycobacterium species. Int J System Bact 1992; 42: 315-23.

MacCallum P, Tolhurst JC, Buckle G, Sissons HA. A new mycobacterial infection in man. J Path Bacteriol 1948; 60: 93-122.

Marsollier L, Robert R, Aubry J, Saint André J-P, Kouakou H, Legras P, Manceau A-L, Mahaza C, Carbonelle B. Aquatic insects as a vector for Mycobacterium ulcerans. Appl Environ Microbiol 2002; 68: 4623-28.

Meyers WM, Connor DH, McCullough B, Bourland J, Moris R, Proos L. Distribution of Mycobacterium ulcerans infection in Zaïre, including the report of new foci. Ann Soc belge Méd Trop 1974b; 54: 147-57.

Meyers WM, Shelly WM, Connor DH, Meyers EK. Human Mycobacterium ulcerans infections developing at sites of trauma to skin. Am J Trop Med Hyg 1974a; 23: 919-23.

Mulder AA, Boerma RP, Barogui Y, Zinsou C, Johnson RC, Gbovi J, van der Werf TS, Stienstra Y. Healthcare seeking behaviour for Buruli ulcer in Benin: a model to capture therapy choice of patients and healthy community members. Trans R Soc Trop Med Hyg 2008; 102: 912-20.

Nackers F, Dramaix M, Johnson RC, Zinsou C, Robert A, DE Biurrun Bakedano E, Glynn JR, Portaels F, Tonglet R. BCG vaccine effectiveness against Buruli ulcer: a case-control study in Benin. Am J Trop Med Hyg 2006; 75: 768-74.

Nackers F, Johnson RC, Glynn JR, Zinsou C, Tonglet R, Portaels F. Environmental and health-related risk factors for Mycobacterium ulcerans disease (Buruli ulcer) in Benin. Am J Trop Med Hyg 2007; 77: 834-6.

Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, Mbuagbaw J, Portaels F, Pluschke G. Buruli ulcer disease in Cameroon rediscovered. Am J Trop Med Hyg 2004; 70: 520-6.

Peeters Grietens K, Boock AU, Peeters H, Hausmann-Muela S, Toomer E, Ribera JM. "It is me who endures but my family that suffers": social isolation as a consequence of the household cost burden of Buruli ulcer free of charge hospital treatment. PLoS Negl Trop Dis 2008; 2:e321.

Phillips R, Horsfield C, Kuijper S, Lartey A, Tetteh I, Etuaful S, Nyamekye b, Awuah P, Nyarko KM, Osei-Sarpong F, Lucas s, Kolk AH, wansbrough- Jones M. Sensitivity of PCR targeting the IS2404 insertion sequence of Mycobacterium ulcerans in an Assay using punch biopsy specimens for diagnosis of Buruli ulcer. J clin Microbiol 2005; 43: 3650-56.

Phillips R, sarfo FS, Osei-Sarpong F, Boateng A, Tetteh I, Lartey A, Adentwe E, Opare W, Asiedu K, Wansbrough-Jones M. Sensitivity of PCR targeting Mycobacterium ulcerans by use of fine-needle aspirates for diagnosis of Buruli ulcer. J Clin Microbiol 2009 ; 47 : 924-26.

Plan directeur de développement du Programme National de Lutte contre l’Ulcère de Buruli (PNLUB-RDC) 2005-2008, Janvier 2005.

Portaels F, Aguiar J, Debacker M, Guedenon A, Steunou C, Zinsou C, Meyers WM. Mycobacterium bovis BCG vaccination as prophylaxis against Mycobacterium ulcerans osteomyelitis in Buruli ulcer disease. Infect Immun 2004; 72: 62-5.

Portaels F, Aguiar J, Debacker M, Steunou C, Zinsou C, Guedenon A, Meyers WM. Prophylactic effect of Mycobacterium bovis BCG vaccination against osteomyelitis in children with Mycobacterium ulcerans disease (Buruli Ulcer). Clin Diagn Lab Immunol 2002; 9: 1389-91.

Portaels F, Chemlal K, Elsen P, Johnson PD, Hayman JA, Hibble J, Kirkwood R, Meyers WM. Mycobacterium ulcerans in wild animals. Rev Sci Tech 2001; 20: 252-64.

Portaels F, Johnson P, Meyers WM. Ulcère de Buruli-Diagnostic de l'infection à Mycobacterium ulcerans.World Health Organisation. WHO/CDS/CPE/GBUI/2001.4.

Portaels F, Meyers WM, Ablordey A, Castro AG, Chemlal K, de Rijk P, Elsen P, Fissette K, Fraga AG, Lee R, Mahrous E, Small PL, Stragier P, Torrado E, Van Aerde A, Silva MT, Pedrosa J. First Cultivation and Characterization of Mycobacterium ulcerans from the Environment. PLoS Negl Trop Dis 2008; 2: e178.

Portaels F, Silva MT, Meyers WM. Buruli ulcer. Clin Dermatol 2009; 27: 291-305.

Portaels F. Epidémiologie des ulcères à Mycobacterium ulcerans. Ann Soc belge Méd Trop 1989; 69: 91-103.

Portaels F. Laboratory diagnosis of Buruli ulcer-A manual for health care providers. World Health Organization. 2014. WHO/HTM/NTD/IDM/2014.1.

Portaels F, Elsen P, Guimaraes-Peres A, Fonteyne P-A, Meyers WM. Insects in the transmission of Mycobacterium ulcerans infection. Lancet 1999; 353: 986.

Portaels F, Johnson RC, Aguiar J, Meyers WM, Debacker M. Etude de 106 cas d'ulcères de Buruli avec atteintes osseuses traités à Zagnanado, Bénin. Bull de l'ALLF. 2008; 23: 48-50.

Portaels F, Zinsou C, Aguiar J, Debacker M, de Biurrun E, Guédénon A, Josse R, Lagarrigue V, Silva MT, Stenou C, Meyers WM. Les atteintes osseuses dans l'ulcère de Buruli: à propos de 73 cas. Bull Séanc Acad R Sci Outre-Mer 2003; 49: 161-90.

Pouillot R, Matias G, Wondje CM, Portaels F, Valin N, Ngos F, Njikap A, Marsollier L, Fontanet A, Eyangoh S. Risk factors for Buruli ulcer: a case control study in Cameroon. PLoS Negl Trop Dis 2007; 1: e101.

Renzaho AM, Woods PV, Ackumey MM, Harvey SK, Kotin J. Community- based study on knowledge, attitude and practice on the mode of transmission, prevention and treatment of the Buruli ulcer in Ga West District, Ghana. Trop Med Int Health 2007; 12: 445-58.

Ross BC, Marino L, Oppedisano F, Ewards R, Robins-Browne RM, Johnson PD. Development of a PCR assay for rapid diagnosis of Mycobacterium ulcerans infection. J Clin Microb 1997; 35: 1696-1700.

Siegmund V, Adjei O, Nitschke J, Thompson W, Klutse E, Herbinger KH, Thompson R, van Vloten F, Racz P, Fleischer B, Loescher T, Bretzel G. Dry reagent-based polymerase chain reaction compared with other laboratory methods available for the diagnosis of Buruli ulcer disease. Clin Infect Dis 2007; 45: 68-75.

Silva MT, Portaels F, Pedrosa J. Aquatic insects and Mycobacterium ulcerans: an association relevant to Buruli ulcer control? PLoS Med 2007; 4: e63.

Sizaire V, Nackers F, Comte E, Portaels F. Mycobacterium ulcerans infection: control, diagnosis, and treatment. Lancet Infect Dis 2006; 6: 288-96.

Smith JH. Epidemiologic observations on cases of Buruli ulcer seen in a hospital in the lower Congo. Am J Trop Med Hyg 1970; 19: 657-63.

Sopoh GE, Johnson RC, Chauty A, Dossou AD, Aguiar J, Salmon O, Portaels F, Asiedu K. Buruli ulcer surveillance, Benin, 2003-2005. Emerg Infect Dis 2007; 13: 1374-6.

Spangenberg T, Kishi Y. Highly sensitive, operationally simple, cost/time effective detection of the mycolactones from the human pathogen Mycobacterium ulcerans. Chem Comm 2010; 46: 1410–12.

Stienstra Y, van der Graaf WT, Asamoa K, van der Werf TS. Beliefs and attitudes toward Buruli ulcer in Ghana. Am J Trop Med Hyg.2002;67:207- 13.

Stinear T, Ross BC, Davies JK, Marino L, Robins-Browne RM, Oppedisano

F, Sievers A, Johnson PD. Identification and characterization of IS2404 and IS 2606: two distinct repeated sequences for detection of Mycobacterium ulcerans by PCR. J Clin Microb 1999; 37: 1018-23.

Trubiano JA, Lavender CJ, Fyfe JAM, Bittmann S, Johnson PDR. The Incubation Period of Buruli Ulcer (Mycobacterium ulcerans Infection). PLoS Negl Trop Dis 2013; 7: e2463.

World Health Organisation, 2013. Buruli ulcer- Mycobacterium ulcerans Infection. Fact sheet n°199, June 2013 ( available at http://www.who.int/mediacentre/factsheets/fs199/fr/; accessed on June 29th, 2014).

World Health Organisation. Buruli ulcer disease. Wkly Epidemiol Rec. 2004a; 79: 194-9.

World Health Organization, 2004b. Provisional guidance on the role of specific antibiotics in the management of Mycobacterium ulcerans (Buruli ulcer). WHO/CDS/CPE/GBUI/2004.10.

World Health Organization, 2005. Modules de formation en organisation et gestion de programme de lutte contre l’ulcère de Buruli au niveau du district. OMS Afro, Harare, Zimbabwe, décembre 2005.

World Health Organization, 2012. Treatment of Mycobacterium ulcerans disease (Buruli ulcer)-Guidance for health workers. WHO/HTM/NTD/IDM/2012.1.

World Health Organization. Buruli ulcer: progress report, 2004-2008. Wkly Epidemiol Rec 2008; 83: 145-54.

Yeboah-Manu D, Bodmer T, Mensah-Quainoo E, Owusu S, Ofori-Adjei D, Pluschke G. Evaluation of decontamination methods and growth media for

50 primary isolation of Mycobacterium ulcerans from surgical specimens. J Clin Microbiol 2004 Dec; 42: 5875-6.

Figure 3. Une papule à la tempe, associée à un œdème diffus de la face (Photo : Dr Phanzu).

Figure 4. Un nodule à la face latérale interne de la cuisse gauche (Photo : Dr Phanzu).

Figure 5. Une plaque typique à localisation abdominale (Photo : Dr Phanzu).

Figure 6. Un œdème diffus du membre supérieur gauche (Photo : Dr Phanzu).

Figure 7a. Un œdème diffus de la face avec fermeture permanente des yeux (photo : Dr Phanzu).

Figure 7b. Résolution complète de l’œdème facial après traitement antibiotique seul (même patient que la figure 7a, Photo : Dr Phanzu).

Figure 8. Une lésion ulcérée précoce de catégorie I (Photo : Dr Phanzu).

Figure 9. Une lésion ulcérée étendue de catégorie III (Photo : Dr Phanzu).

Figure 10. Une lésion cicatricielle retractile du bas du dos après greffe cutanée (Photo: Dr Phanzu).

Figure 11. Une cicatrice retractile du coude droit avec limitation des mouvements articulaires (Photo: Dr Phanzu).

Figure 12a. Forme mixte disséminée de l’UB avec atteinte osseuse des os du gril costal (Photo : Dr Phanzu).

Figure 12b. Résultat du traitement après antiothérapie spécifique combinée à la chirurgie (Photo : Dr Phanzu).

CHAPITRE II. OBJECTIFS ET ORGANISATION DE LA THESE

II.1. Enoncé du problème de recherche

L’UB touche essentiellement les populations pauvres vivant en milieu rural et ayant un faible accès aux services de santé. La mortalité liée à l’UB est faible mais la morbidité et les incapacités sont élevées (Portaels F et al., 2001). Le recours à l’auto-prise en charge ou au traitement traditionnel en premier ressort est très important dans la plupart des communautés affectées d’Afrique (Mulder AA et al., 2008 ; Stienstra Y et al., 2002 ; Renzaho AM et al., 2007 ; Sizaire V et al., 2006 ; Noeske J et al., 2004 ; Aujoulat I et al., 2003 ; Johnson RC et al., 2004), y compris dans le territoire de Songololo en RDC (Kibadi K, 2004 ; Kibadi K et al., 2009). Par conséquent, les patients se présentent souvent aux structures de santé aux stades avancés de la maladie avec comme corollaires, l’indication fréquente du traitement chirurgical, une longue durée d’hospitalisation, le développement des séquelles invalidantes et un coût économique pour la famille exorbitant (Asiedu K et al. 1998 ; Kibadi K, 2005 ; Bafende AE et al., 2004 ; Peeters-Grietens K et al., 2008 ; Sizaire V et al., 2006). La proportion des patients guéris avec incapacités peut atteindre jusqu’à 20-25% du total du nombre des patients soignés dans les services de santé en Afrique (Portaels F et al., 2001).

Entre 2002 et 2004, parmi les 64 patients admis pour UB à l’HGR IME/Kimpese, 95% ont été admis au stade le plus avancé, c'est-à-dire, le stade ulcéré, 36% présentaient déjà une limitation fonctionnelle au moment du diagnostic, et 23% ont développé des incapacités permanentes. La durée médiane d’hospitalisation était de 89 jours. En outre, 90% des patients n’étaient pas capables de payer les frais d’hospitalisation et un taux de létalité anormalement élevée de 19% a été observé (Phanzu DM et al., 2011). Afin de faire face à ces problèmes causés par l’UB dans le territoire de Songololo, un projet de lutte contre l'UB fut initié par American Leprosy Missions (ALM) et IME en fin 2004. Les deux composantes principales dudit projet étaient l’amélioration de la prise en charge des malades UB admis à l’HGR IME/Kimpese et la promotion de la détection précoce à base

61 communautaire. Cependant, après plus de trois ans de mise en œuvre des activités de lutte par le projet, la morbidité et les incapacités dues à l’UB restaient élevées, sur 190 patients admis à l’hôpital, 86% ont été dépistés au stade ulcéré et 20% étaient guéris avec incapacités permanentes (Phanzu DM et al., 2011).

II.2. Justification de la thèse

Eu égard à ce qui vient d’être décrit ci-haut, l’UB constitue un réel problème de santé publique dans le Territoire de Songololo, un des foyers endémiques les plus importants en RDC.

La morbidité et les incapacités occasionnées par l’UB y sont élevées. Plusieurs facteurs concourent à cet état des choses, d’une part, la méconnaissance des formes précoces de l’UB dans les communautés, le recours important à l’automédication et au traitement traditionnel, une faible utilisation des formations sanitaires par la population, et d’autre part, le diagnostic tardif et une prise en charge inadéquate dans les structures de santé.

En outre, la sous-notification de l’UB par la plupart des structures de santé limite une meilleure connaissance de sa fréquence ainsi que de sa distribution.

Notre étude, au-delà des aspects cliniques, épidémiologiques et microbiologiques de la problématique, adresse la question centrale de l’organisation du dépistage et des soins des patients atteints d’UB. Quel est l’apport d’une décentralisation du dépistage et du traitement de l’UB dans un district comparé au modèle existant qui limitait la prise en charge des malades au centre de référence?

II.3. Objectif général

L’objectif général de notre étude est de contribuer à l’amélioration du contrôle de l’UB dans le territoire de Songololo par une meilleure connaissance et une prise en charge adéquate de la maladie, ainsi qu’une meilleure organisation des activités de lutte.

II.4. Objectifs spécifiques

Pour atteindre cet objectif général, nous nous sommes fixés comme objectifs spécifiques :

Dans la Partie I, de recherche clinique :

1. Réviser l’expérience clinique de l’hôpital IME dans la prise en charge des patients UB sur base des données disponibles avant la mise en place du projet de lutte de façon générale et en particulier par rapport aux lésions faciales.

Dans la Partie II, de recherche en santé publique : 2. Déterminer l’ampleur de l’UB comme problème de santé publique dans le foyer de Songololo en termes de prévalence et de distribution géographiques, ainsi que de ses caractéristiques cliniques et épidémiologiques. 3. Evaluer l’effet du vaccin BCG sur l’UB. 4. Evaluer l’impact de la décentralisation et de l’intégration des activités de lutte dans les services de santé de base sur l’évolution du profil clinique et épidémiologique des malades UB dans le foyer de Songololo.

CHAPITRE III. METHODOLOGIE

La méthodologie générale du travail est décrite dans ce chapitre III. Toutefois, la méthodologie propre à chaque étude est à consulter dans la publication correspondante. La source des données sur la présentation générale de la RDC est tiré du Rapport sur le Développement Humain 2013 disponible sur : http://www.cd.undp.org/content/rdc/fr/home/countryinfo/. Province du Bas-Congo : http://www.cd.undp.org/content/dam/dem_rep_congo/docs/povred/UNDP -CD-Profil-PROVINCE-Bas-Congo.pdf

III.1. La République Démocratique du Congo

La RDC est un vaste pays situé en Afrique centrale et à cheval sur l’Équateur. Elle a une superficie de 2.345.000 Km2.soit quatre fois plus grand que la France et 80 fois plus que la Belgique. Elle compte 9 pays limitrophes avec lesquels elle partage 9165 Km de frontières. Elle est limitée au nord par la République Centrafricaine et le Sud-Soudan, à l’est par l’Ouganda, le Rwanda, le Burundi et la Tanzanie, au sud par la Zambie et l’Angola, et à l’ouest par la République du Congo. La RDC est actuellement subdivisée en 11 provinces (Bandundu, Bas-Congo, Equateur, Kasaï Occidental, Kasaï Oriental, Katanga, Maniema, Nord-Kivu, Province Orientale, Sud-Kivu et La ville-province Kinshasa), 45 districts et 225 territoires administratifs et communes. Il existe une langue officielle (le français), quatre langues nationales (kikongo, lingala, swahili et tshiluba) et environ 450 dialectes.

Situation démographique La population de la RDC en 2012 est estimée à 70.000.000 d’habitants. C’est une population jeune dont la tranche jeune représente environ 70% et à majorité rurale (60%). Quelques indicateurs démographiques : - Taux de croissance démographique : 2,54% - Taux de natalité : 36,34 naissances/1000 habitants par an - Taux de mortalité infantile (pour 100 naissances) : 17% - Espérance de vie à la naissance : 49 ans - Taux brut d’enrôlement scolaire (filles/garçons) : 54% - Taux d’alphabétisation (15 ans et +) : 66,8%.

Situation socio-économique Bien que le pays dispose d’énormes potentialités minières, énergétiques et agricoles, l’instabilité politique et les conflits armés à répétition ont eu ces dernières années un impact négatif sur la qualité de vie des populations. Les indicateurs socio-économiques en 2012 : - Revenu par tête : 319$

- Taux de croissance du Produit intérieur brut : 7,2% - Indice de développement humain : 0,304 (186ème position sur 186 pays classés) - Taux de pauvreté : 71%.

Système de santé Le système de santé est de type pyramidal. Il comprend 3 niveaux : - Le niveau central : c’est le Ministère de la Santé Publique avec ses différentes directions. Il a essentiellement un rôle normatif, stratégique et régulateur. La structure de prise en charge de niveau tertiaire est représentée par les cliniques universitaires. - Le niveau intermédiaire est constitué de 11 divisions provinciales. Il joue le rôle d’appui technique, d’accompagnement et d’encadrement aux zones de santé. La structure de prise en charge des malades est représentée par les hôpitaux régionaux. - Le niveau périphérique : La Zone de Santé (ZS) est le niveau opérationnel du système de santé. Les structures de santé obligatoires pour une ZS sont l’ hôpital général de référence (HGR) et centre de santé (CS), tandis que les centres de santé de référence (CSR) et les postes de santé (PS) sont des structures facultatives en fonction de la superficie ou de la densité de la population. Chaque ZS est constitué d’une équipe cadre de ZS, d’un HGR et un nombre variable de centres de santé (CS), en général 15 à 20, qui desservent chacun une aire de santé (AS). La RDC compte 515 ZS, qui desservent chacune en moyenne 100.000 habitants en milieu rural et 150.000 en milieu urbain.

La Province du Bas-Congo Elle est située au sud-ouest de la RDC. Sa superficie est de 53 920 Km² et sa population était estimée à 3 734 594 habitants en 2006, avec une densité de 69.26 habitants au Km². Elle est limitée au nord par l’enclave angolaise de Cabinda et la République du Congo, au sud par l’Angola, à

68 l’est par la Ville de Kinshasa et la province de Bandundu et à l’ouest par l’Océan Atlantique.

Figure 1. La situation géographique de la Province du Bas-Congo.

Le climat est de type tropical dont la saison sèche s’étend sur un peu plus de 4 mois (du 15 mai au 25 septembre). La longue saison des pluies est souvent interrompue par une petite saison sèche au mois de février. La température moyenne annuelle oscille autour de 25° C. La végétation comprend les mangroves des terrains marécageux de l’embouchure du fleuve Congo, les steppes des plateaux de la côte de Moanda, la forêt dans le district du Bas-Fleuve et une région de savane entrecoupée par des lambeaux de forêt dans les districts des Cataractes et de la Lukaya. La province est divisée en entités urbaines et rurales, dont 2 villes ayant le statut de district (Matadi et Boma), 3 districts (Bas-Fleuve, Cataractes et

Lukaya), 6 communes, 10 territoires, 55 secteurs, 17 cités, 113 quartiers, 27 postes d’encadrement administratif, 367 groupements, 6.783 Villages. Environ 70% de la population vit en milieu rural et 70 % de la population vit avec un revenu annuel estimé à 138,6 USD (soit 11,55 USD par mois et 0,39 USD par jour) 1. La province du Bas-Congo compte 5 districts médicaux, 31 zones de santé, 381 aires de santé, 30 hôpitaux généraux de référence, 76 centres de santé de référence, 389 centres de santé, 747 postes de santé, 539 autres établissements de santé. Elle dispose d’un médecin pour 17.356 habitants, d’un pharmacien pour 131.069 habitants, d’un lit pour 514 habitants, d’un HGR pour 126.700 habitants, d’un centre de santé de référence pour 50.013 habitants.

III.2. Cadre du travail

Le cadre de notre étude est le territoire de Songololo. Il est l’un des 10 territoires de la province du Bas-Congo et est situé dans le district des Cataractes. Sa superficie est de 8190 Km2 et sa population est estimée à 237 418 habitants en 2008, avec une densité de 28.98 habitants au Km2. Il est limité au nord par le fleuve Congo, à l’ouest par le territoire de Sekebanza, à l’est par le territoire de Mbanza-Ngungu et au sud par la frontière nord de la République d’Angola. Le bassin hydrographique comprend des grandes rivières qui se déversent dans le fleuve Congo. La population vit essentiellement de l’agriculture de subsistance. Il contient deux zones de santé rurales (ZSR), la ZSR de Kimpese et la ZSR de Nsona-Mpangu dont chacune est subdivisée en 20 aires de santé. Le système sanitaire est de type pyramidal constitué au niveau primaire par les PS, les CS et les centres de santé de référence (CSR) et au niveau secondaire par l’hôpital général de référence (HGR). Selon la classification des ZS en application en RDC2, du fait de leur situation stable sur le plan sécuritaire et de l’existence d’une activité socio-économique, les deux sont

1 Document provincial de stratégie de réduction de la pauvreté, Province du Bas-Congo, 2007. 2 Recueil des normes de la zone de santé. Secrétariat général du Ministère de la Santé, RDC, Août 2006.

70 des ZS en développement. La ZS de Kimpese est une ZS à potentiel élevé au développement tandis que celle de Nsona-Mpangu est à faible potentiel au développement.

III.3. Types d’études

Nos études épidémiologiques sont observationnelles à visée descriptive et analytique. En premier lieu, des études descriptives sous forme de séries de cas (clinical case-report) ont été réalisées. Ensuite, nous avons mené une étude comparative avant/après décrivant le profil des malades UB à l’hôpital IME/Kimpese, suivie d’une étude descriptive sur la prévalence et la distribution de l’UB dans la région de Songololo. Enfin, une étude cas- témoins et une étude d’intervention sur l’impact de la décentralisation des activités de lutte ont été réalisées. Cette dernière se propose d’évaluer sur le plan clinique et épidémiologique, l’impact de la décentralisation du dépistage et du traitement de l’UB dans le territoire de Songololo en comparaison au modèle précédent qui limitait la prise en charge des patients UB aux centres de référence.

III.4. Echantillons étudiés

Avant le lancement de la décentralisation de la prise en charge de l’UB en 2009 dans le territoire de Songololo, nos premières études se sont basées sur les 254 patients consécutifs cliniquement diagnostiqués comme UB et admis dans le service de chirurgie générale de l’HGR IME/Kimpese de janvier 2002 à décembre 2007, dans le territoire de Songololo, province du Bas-Congo, en RDC. En 2008, l’échantillon étudié à l’enquête de prévalence était constitué de 259 patients porteurs des lésions évolutives d’UB et de 516 sujets porteurs des lésions inactives (cicatrices, séquelles). Concernant l’étude cas-témoin menée de janvier 2010 à décembre 2012, l’échantillon pour la RDC est constitué de 128 patients confirmés par au moins un des quatre tests recommandés par l’OMS et 494 contrôles.

Pour l’étude d’intervention sur l’impact de la décentralisation, l’échantillon étudié est constitué par les malades souffrant de l’UB cliniquement diagnostiqués dans les structures de santé du territoire de Songololo de janvier 2009 à septembre 2014, y compris l’HGR IME/Kimpese. Au total, nous avons enregistrés 901 nouveaux cas d’UB durant cette période.

III.5. Définitions de cas

Cas probable Toute personne présentant une tuméfaction indolore (sous forme d’un nodule, d’un placard induré ou d’un œdème) ou un ulcère, indolore, aux bords décollés, qui a commencé comme un gonflement indolore (formes actives), ou une cicatrice (forme inactive) consécutive à un tel ulcère, qui habite dans une zone endémique ou qui a visité une zone endémique, est un cas probable d'UB (Asiedu K et al., 2000 ; WHO, 2005).

Cas confirmé d’UB Un cas confirmé d'UB est un cas probable qui a été confirmé par un ou plusieurs des quatre examens de laboratoire recommandés par l’OMS(Asiedu K et al., 2000 ; WHO, 2005).

Cas compliqué d’UB Si l'une des formes cliniques susmentionnées de l'UB est associée à l'une des situations suivantes, elle est considérée comme un cas d'UB compliqué : Infection bactérienne secondaire, formation d'abcès, tétanos, saignement modéré ou grave, lésion étendue, ostéomyélite, formation de sinus, infirmités par contraction des articulations, cicatrisation excessive, amputation, destruction d'un organe, malignité (cancer). Tous les cas compliqués d'UB doivent être référés pour des soins spécialisés. Les autres cas à référer pour soins spécialisés comprennent les lésions d’UB sur les sites ci-dessous ou dans les situations suivantes : la tête et le cou, le périnée, les organes génitaux, les hanches, la main, les seins, les

72 lésions qui recouvrent des articulations, l'UB pendant la grossesse et toute autre maladie systémique (Asiedu K et al., 2000 ; WHO, 2005).

Cas de rechute Un cas de rechute d’UB est un cas d’UB traité médicalement et/ou chirurgicalement et déclaré guéri (cicatrisation des lésions ulcérées) et qui présente à nouveau une forme active d’UB confirmée par le laboratoire moins d’une année après la guérison de la précédente lésion. La rechute doit être considérée comme un échec du traitement précédent. Tous les autres cas de reprise évolutive de la maladie plus de 12 mois après la guérison doivent être considérés comme des nouveaux cas d’UB (Asiedu K et al., 2000 ; WHO, 2005).

UB à lésions multiples Un malade peut avoir en même temps deux lésions ou davantage. Les lésions peuvent être à la même phase d'évolution ou à différentes étapes (Asiedu K et al., 2000 ; WHO, 2005).

III.6. Procédures de confirmation par le laboratoire

Le diagnostic clinique des cas est basé sur les éléments clinico- épidémiologiques selon l’OMS détaillés aux sections I.2, I.3 et I.4. du chapitre I. Ensuite, les prélèvements pour la confirmation des cas suspectés sont réalisés. Selon la forme clinique en présence et le plateau technique de la structure de santé, les écouvillonnages pour les ulcères, les aspirations à l’aiguille fine pour les formes non-ulcérées ou biopsie pour toutes les formes, ulcérées ou non, sont réalisés.

Généralement, 3 échantillons sont prélevés par lésion, dont un étalement direct à analyser localement (CSDT), un autre à placer dans un milieu de transport pour le laboratoire de l’IME et un dernier à placer également dans un milieu de transport pour l’INRB. Le formulaire UB 03 de l’OMS est

73 utilisé pour la demande d’analyses au CSDT, au laboratoire de l’IME ou à l’INRB. Entre ces laboratoires, un réseau de confirmation des cas d’UB était mis en place avec quatre niveaux de compétence, local et régional, pour la collecte des échantillons et la réalisation des premiers examens (ZN surtout), national, pour la confirmation par PCR et supranational, pour le contrôle de qualité, assuré par l’unité de Mycobactérie de l’IMT/Anvers. Depuis 2007, la culture de M. ulcerans a été lancée à l’HGR IME. Avant, seuls la coloration de ZN et l’histopathologie étaient disponibles.

III.7. Collecte des données

La notification des cas est standardisée dans les deux ZS du territoire de Songololo. Les données des patients UB sont collectées par l’usage de la fiche clinique et thérapeutique UB01 et de la fiche d’enregistrement de cas UB02 de l’OMS, lesquelles ont été introduites à l’HGR IME/Kimpese depuis 2002. Ensuite, elles ont été saisies dans une base de données Excel (Microsoft Corporation, Redmond, WA). Pour l’étude cas-témoins multicentrique (DRC, Ghana and Togo) sur l’efficacité de la vaccination BCG sur l’UB, une fiche supplémentaire spécifique du projet de recherche "Burulivac" (FP7/2010-2013 ; grant agreement N°241500) a été utilisée. Puis, toutes les données sociodémographiques, cliniques et de laboratoire ont été saisies dans une base de données sur le web, conçu spécifiquement pour le projet.

III. 8. Analyse des données

Les données ont été analysées à l’aide des logiciels Epi Info version 3.3.2 (Centers for Diseases Control and Prevention, Atlanta, GA) et SPSS version 9.0 pour Windows (SPSS Inc., Chicago, IL). Pour l’analyse des tables de contingences, le test de Chi-carré de Pearson a été utilisé pour comparer deux proportions d’échantillons, et lorsque ce dernier n’était pas applicable, le test exact de Fisher. Pour l’étude cas- témoins, une régression logistique par le logiciel Stata, version 9.0.

(Stata Corporation, College Station, TX) a été utilisée. Le seuil de signification était de 5% pour toutes les analyses. Les principaux paramètres d’intérêt sont : le nombre de cas enregistrés pour chaque période, l’âge, le sexe, le nombre de nouveaux cas et de rechutes, la catégorie de lésions, la proportion des cas avec limitation de mouvements articulaires au moment du diagnostic, la proportion de cas confirmés par au moins un examen de laboratoire, la proportion des formes ulcérées au moment du diagnostic, le type de traitement appliqué, la proportion de cas guéris avec limitation de mouvements articulaires, la durée d’hospitalisation ou du traitement, et le taux de létalité.

III. 9. Aspects éthiques

Les données des participants aux différentes études ont été traitées de façon confidentielle. La participation est volontaire et un consentement éclairé est obtenu auprès de personnes concernées ou de leurs représentants pour les sujets de moins de 18 ans. L’approbation éthique a été accordée par le Comité d’Ethique de l’Ecole de Santé Publique de l’Université de Kinshasa (Réf. N°ESP/CE/057/2010).

CHAPITRE IV. RESULTATS

IV.1. Partie clinique

"Les rapports de cas et les séries de cas ont leur propre rôle dans le progrès de la science médicale. Ils permettent la découverte de nouvelles maladies et des effets inattendus (négatif ou positif) ainsi que l'étude de mécanismes, et ils jouent un rôle important dans la formation médicale. Les rapports de cas et les séries de cas ont une grande sensibilité pour la détection des nouveautés et restent donc un des piliers du progrès de la médecine; ils fournissent beaucoup d'idées nouvelles en médecine. En même temps, de bons rapports de cas exigent une orientation claire afin de rendre explicite au public pourquoi une observation particulière est importante dans le contexte des connaissances existantes."

Jan P. Vandenbroucke. In Defense of Case Reports and Case Series. Ann Intern Med 2001; 134:330-334.

IV.1.1. Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004.

Delphin M. Phanzu1, Eric A. Bafende1, Barthelemy K. Dunda1, Désiré B. Imposo1, Anatole K. Kibadi2, Samuel Z. Nsiangana1, Jackie N. Singa2, Wayne M. Meyers3, Patrick Suykerbuyk4 , Françoise Portaels4

1 Institut Médical Evangélique, Kimpese Hospital, Kimpese, Bas-Congo, Democratic Republic of Congo 2 Programme National de Lutte contre l’Ulcère de Buruli, Kinshasa, Democratic Republic of Congo 3 Armed Forces Institute of Pathology, Washington, District of Columbia 4 Mycobacteriology Unit, Department of Microbiology, Institute of Tropical Medicine, Antwerp, Belgium

Personal contribution: Conceived and designed the experiments: DMP; capacity strengthening: DMP; performed the experiments: DMP; analyzed the data: DMP; wrote the paper: DMP.

Am J Trop Med Hyg. 2006 Aug;75(2):311-4.

Abstract

Buruli ulcer (BU), which is caused by Mycobacterium ulcerans, is an important disabling skin disease. Its prevalence is highest in west and central Africa. We report an apparent resurgence of BU in the Bas-Congo Province, Democratic Republic of Congo. During a 28-month period in 2002–2004, the rural hospital of the Institut Médical Evangélique at Kimpese admitted 51 patients suspected of having BU. Bacteriologic, molecular biologic, and histopathologic studies confirmed BU in 36 of these patients. Extensive clinical data, treatment outcomes, and socioeconomic correlations are summarized. Osteomyelitis was an important complication. A multidisciplinary approach to BU control in the Bas-Congo is proposed, aimed primarily at active case detection.

INTRODUCTION

Buruli ulcer (BU) is a severe disfiguring and disabling infectious disease caused by Mycobacterium ulcerans, a bacterium related to those causing leprosy and tuberculosis. The disease has emerged dramatically over the past decade particularly in west Africa, and has been detected or suspected in at least 31 countries (WHO, 2000, 2004). The first case, which was reported in 1950 in the Democratic Republic of Congo (DRC), was an American boy living in the Kwilu region (Van Oye et al., 1950). Information on BU in the DRC was summarized by Janssens in 1972 (Janssens et al., 1972) and by Meyers and others in 1974 (Meyers et al., 1974). The District of Cataractes in Bas-Congo (Lower Congo) Province is one of the main regions of the DRC where BU has been reported (Meyers et al., 1974; Andersen et al., 1965; Smith et al., 1970; Bafende et al., 2002, 2004; Kibadi et al., 2002). Bas-Congo is one of the 11 provinces of the DRC, is situated in the southwestern part of the country, and has a population of 2,835,000. This province has 27 health zones and five districts (Figure 7): Matadi (the main town), Boma (the second largest town), Bas-Fleuve, Cataractes, and Lukaya. The population of the DRC was approximately 50 million in 2002 and

82 reached 60 million in 2005 (National Institute of Statistics). Seventy percent of the people live in rural areas, 57% are ≤ 19 years old, and 39% are 20–50 years old. Between 1972 and 1990 the national health budget decreased from 2.4% to 0.7% with one physician per 22,637 inhabitants and one nurse per 1,714. The Institut Médical Evangélique (IME)/Kimpese serves an area with approximately 150,000 inhabitants, has 400 beds, and receives most of the hospitalized BU patients in the province. This hospital is 220 km southwest of Kinshasa in Songololo Territory, the main known focus of BU in the DRC (Meyers et al., 1974; Andersen et al., 1965; Smith et al., 1970; Bafende et al., 2002, 2004; Kibadi et al., 2002, 2003; Portaels et al., 1989). Although many observations have been reported from IME/Kimpese (Meyers et al., 1974; Andersen et al., 1965; Smith et al., 1970; Bafende et al., 2002, 2004; Kibadi et al., 2002), no recent data have been published on patients with a laboratory-confirmed diagnosis of BU according to the World Health Organization (WHO) criteria (WHO, 2001). This study reviews the experience of IME/Kimpese in the management of hospitalized laboratory-confirmed BU patients from May 2002 to August 2004.

MATERIALS AND METHODS

We studied 51 consecutive suspected BU patients admitted to the surgical service of IME/Kimpese in the Province of Bas-Congo from May 2002 to August 2004. Age, sex, address, date of onset, date of diagnosis, clinical features, hospital stay, type of management, and evolution of the disease were recorded. Recognized clinico-epidemiologic features were used to suspect lesions as BU (WHO, 2000, 2001). As required by the Medical Committee of IME/Kimpese all patients, or their responsible relative or guardian, gave informed consent for all diagnostic and treatment procedures. Specimens of exudates and tissues from all 51 patients were obtained and evaluated according to WHO recommendations (WHO, 2001). The initial

83 direct smear examinations for acid-fast bacilli and histopathologic analyses were made at the IME/ Kimpese laboratory. Other specimens from the same patient were sent in a transport medium (WHO, 2001), to the Mycobacteriology Unit of the Institute of Tropical Medicine (ITM) in Antwerp, Belgium, where Ziehl-Neelsen (ZN) staining, in vitro culture, and an IS2404 polymerase chain reaction (PCR) were performed (WHO, 2001). Formalin-fixed tissues were sent to the Department of Infectious and Parasitic Diseases Pathology of the Armed Forces Institute of Pathology in Washington DC, for confirmation of the histopathologic diagnosis. Data were recorded using Excel (Microsoft Corporation, Redmond, WA) and analyzed with Epi-Info (Centers for Diseases Control and Prevention, Atlanta, GA) and SPSS version 9.0 for Windows (SPSS Inc., Chicago, IL).

RESULTS AND DISCUSSION

Among 51 patients clinically suspected to have BU, 36 (70.5%) had laboratory-proven M. ulcerans disease. They had positive results for at least one of the laboratory tests used, i.e.: ZN staining, culture, PCR or histopathologic diagnosis. Twenty-six of 36 had at least two positive tests results and 10 of 36 had only one positive test result. The distribution of the positive test results for diagnosis of BU was ZN staining (61.1%), PCR (75%), histopathologic diagnosis (75%), and culture (2.8%). Except for culture results, this distribution compares favorably with previously published data (WHO, 2001). Several explanations for the low percentage of positive primary cultures have been discussed by several investigators (WHO, 2001; Pettit et al., 1966; Portaels et al., 1997). Secondary infections of skin ulcers in 52.8% of our patients, difficulties with sampling, and transportation delay (> 2 months) to the ITM in Belgium were the most likely contributing factors. Ideally, specimens should be processed on the day of collection to obtain maximum positivity rates for primary cultures (WHO, 2000). Some investigators have obtained > 60% positivity rates for primary cultures

84 after storage of tissue specimens in a semisolid transport medium (WHO, 2001) kept at 4°C for one week before processing (Yeboah-Manu et al., 2004). Fifteen patients (29.4%) had negative results. Among them, three had negative results in the four tests and 12 were partially tested. Five patients had another diagnosis (one with malignant non-Hodgkin’s lymphoma, one with Kaposi’s sarcoma, and three with non-specific chronic inflammation and necrosis). The age distribution of the BU patients during this period showed that 16 (44.4%) of 36 patients were less than 15 years of age, 36.1% were 16– 45 years of age, and 19.4% were more than 45 years of age. The median age was 17 years (age range = 3 months to 67 years). This is similar to previous studies that reported rates of 48.0% to 64.0% of patients less than 15 years of age in the same area (Bafende et al, 2004; Debacker et al., 2004a). A study of 151 BU patients treated and followed-up between 1998 and 2000 by the health center of Lalo (Couffo Department, Benin) showed similar results: 46.4% less than 15 years of age, 40.4% between 15 and 49 years of age, and 13.3% more than 49 years of age (Debacker et al., 2004a). However, the proportion of 70% of patients less than 15 years of age found in other disease-endemic countries is higher than in the present study (WHO, 2000). Debacker and others in a study of 1630 BU patients in Benin reported a median age of 15 years and an age range of 2 months to 83 years (Debacker et al., 2004b). This is also similar to our observations. From May 2002 to August 2004, three times more males (n = 27) than females (n = 9) were diagnosed (75% versus 25%), which is consistent with a previous report from the same geographic area (Kibadi et al., 2003). However, other reports show an equal proportion of males and females, both in the area of this study (Bafende et al., 2004) and other areas (WHO, 2000; Portaels et al., 1989; Debacker et al., 2004b). It is likely that socioeconomic factors in our region result in males seeking hospital care more often than females. Among our BU patients, 91.7% came from the Cataractes District, 5.6% from the Bas-Fleuve District, and 2.8% from the Matadi District (Table 1).

No patients were diagnosed in the Boma and Lukaya Districts. Most (58.3%) BU patients were Congolese and 41.7% (15 of 36) were Angolan (Table 1). Bas-Congo Province borders northern Angola. The possibility of infection in Angola is considered low, although cases of BU have been reported in Angola (Bär et al., 1998; Pszolla et al., 2003). Meyers and others reported BU in the Lower Congo in newly arrived patients from Angola beginning in 1965 (Meyers et al. 1974). In our report, however, most of the Angolan BU patients had lived in the DRC for several years, and some young Angolan patients born in Bas-Congo had never visited Angola. Among Angolans living in refugee camps, we believe most patients were either infected while living near Songololo, the usual point of entrance for Angolan refugees, or in refugee camps located in this region that is highly endemic for BU. Despite the availability of a safe water supply, the refugees frequent surrounding rivers and wetlands for domestic water (laundry, bathing and agricultural activities). Recently, an environmental study showed that the Kimu River near the Angolan refugee camp of Nkondo in Kimpese was positive for M. ulcerans by IS2404 PCR (Suykerbuyk et al., 2004). Songololo Territory, which is where our hospital is located, has remained the primary source of BU patients for more than 40 years (Meyers et al., 1974; Andersen et al., 1965; Smith et al., 1970; Portaels et al., 1989). Most (94.4%) patients had ulcerative skin lesions and two (5.6%) had non-ulcerative M. ulcerans disease (one edema and one plaque). There was superinfection of skin ulcers in 52.8% of the BU patients. Disability was noted in 25% on admission. Clinical forms were distributed as follows: mixed forms in 61.1%, single ulcerative lesions in 33.3%, plaque in 2.8%, and edema in 2.8%. Ulcers were present in all mixed forms. Bone involvement was suspected in 36.1% of the patients. Osteomyelitis caused by M. ulcerans was confirmed in 6 (16.70%) of these patients. Two patients had gram-positive and gram-negative bacterial infections and one had a fungal infection of the bone. These findings are consistent with the rate reported in BU patients with osteomyelitis (14%) in Benin (Debacker et al., 2004a; Portaels et al.,

2003). Table 2 shows the location of the lesions and the total distribution of lesions including those that had multiple areas involved. The lower limbs were the most frequently involved (63.9%), followed by the upper limbs (44.4%), thorax (5.5%), back (5.5%), and head (5.5%). This is consistent with another report on the distribution of lesions (Debacker et al., 2004a). During the 28 months from May 2002 to August 2004, the admission rate at the IME was 15.4 patients/year. Between 1960 and 1970, 130 patients were treated at IME (13/year). A study program for M. ulcerans infection that included a special treatment ward was established in late 1970 at the same hospital, and several rural dispensaries and outlying hospitals were contacted. This resulted in the treatment of 39 patients in 1971 (Meyers et al., 1974). The admission rate was 4.5 patients/year from 1989 to 1999 and 31 patients/year from 2000 to 2001 (Bafende et al., 2004). The lowest rates of admission were recorded from 1989 to 1999. This is probably due to a loss of specialized personnel in the hospital during this period. The highest rate of admission, which was recorded from 2000 to 2001, was most likely attributable to the high proportion of Angolan refugees referred by new health centers in refugee camps, which represented 50% of all BU patients (Kibadi et al., 2003). Since most reports are based on passive presentation of patients to the hospital, these numbers represent gross underreporting (WHO, 2000). In 1974, Meyers and others reported that the patients presenting at IME/Kimpese with BU represented only a small percentage of the total number afflicted by this disease (Meyers et al, 1974). In July 2002, a joint research team of IME/ Kimpese and the ITM/Antwerp reported 34 suspected cases during two half days of active case finding in Songololo Territory (Polman K and Phanzu MD, unpublished data). In October 2003, 21 suspected BU patients were reported after one day of active survey within 5 km of Nsona-Mpangu Hospital where only one patient was hospitalized (Phanzu DM and others, unpublished data). The median delay in seeking medical care was 60 days with a minimum of 7 days and a maximum of 840 days (Phanzu DM and others, unpublished

87 data). Overall median delay in Benin from 1989 to 2001 decreased from approximately four months to one month (Debacker et al., 2004a). In this country, the introduction of a BU program of the Directorate General for Development Cooperation (DGDC, Belgium) was an important factor in the marked reductions in patient delay. Moreover, in 2000, after promotional sessions on BU organized by the DGDC and the National BU Program in the Zou, Oueme, and Atlantique Departments, patients reported earlier than in 1999 to the health center at Zagnanado in the Zou Region (Debacker et al., 2004a). We expect the admission rate to increase at the IME/ Kimpese Hospital with the inauguration of the Program National de Lutte contre l’Ulcère de Buruli (PNLUB), which was started by the Ministry of Health of the DRC in 2002. In May 2003, the first conference on the control of BU was organized in Bas-Congo, the main disease-endemic region nearest the national capital of Kinshasa. The first national workshop took place in September 2004. The PNLUB seeks collaboration with national or international organizations to strengthen control measures and treatment programs. In this study, 5 patients presented spontaneously and 31 (86.1%) were referred by government health centers or other health professionals. Only 45.2% (14 of 31) were referred with a suspicion of BU, among whom 12 were referred by a health center where a well-trained doctor well- informed about BU was working. At the health center in Zagnanado, Benin, 68.3% of the patients were referred by a previously treated patient, 22.1% by a family member acquainted with the health center, 5.9% by village outreach activities of the health center, and only 3.7% by health professionals (Debacker et al., 2004a). The strikingly different referral method compared with those at Kimpese is most likely related to the high number (> 4,000) of successfully treated patients in Benin since 1989 (Debacker et al., 2004a). In Zagnanado, Benin, referral of patients to a BU treatment center depends largely on word-of-mouth information from former BU patients. We are hopeful that similar results will develop at IME/Kimpese.

Among our patients 61.1% were new cases (n=22) and 38.9% were relapses (n=14). Among relapsed cases, eight were previously treated by surgery at IME/Kimpese, three at health centers, and three by traditional practitioners. Rates of recurrence for any of these treatment situations are unknown. A study at the health center of Lalo (Couffo Department, Benin) reported a relapse rate of 9.3%, whereas the recurrence rate in Zagnanado (Zou department, Benin) of patients treated after up to 7 years follow-up was 6.1% (Ahodehou Gandji FU, 2000. Contribution à l’Etude des Récidives d’Ulcère de Buruli à Lalo dans le Département du Couffo. Thèse de Doctorat en Médecine. Faculté des Sciences de la Santé, Cotonou, Bénin. No 872) (Debacker et al., 2005). The low recurrence rate in Zagnanado is partly due to the early presentation of patients, easy access to treatment, educational sessions conducted by the health center in village, and the quality of surgical treatment (Debacker et al., 2005). Our patients must travel long distances to reach the BU treatment center of IME/Kimpese. Physical impairments were present in 37% of our patients with obvious long-term social and economic impacts. This is high compared with other reports, indicating that patients with healed lesions had inappropriate management of their physical impairments (WHO, 2001). The observations in this study demonstrate the importance of appropriate management of the disease. Our hope is that treatment can be further improved with better resources that are now available. The fact that 94.4% of our patients had an ulcerative form sometimes associated with other clinical forms suggests that our BU patients arrived at an advanced stage of illness, usually with physical impairments or other serious complications such as sepsis, dissemination, and bone involvement. Consequently, these patients had long hospital stays (median = 102 days) and an unusually high mortality rate of 19.4%. Meyers and others noted that the reasons for many BU patients delaying to seek medical assistance were obviously complex, but cultural, economic, and transportation factors were especially important (Meyers et al., 1974). Thus, this complexity requires a multidisciplinary approach that involves education of the

89 population, training healthcare workers, and managing cases adequately to shorten hospitalization time and stimulate patient initiative (WHO, 2000; Debacker et al., 2004a). Such a multidisciplinary approach will be possible in Kimpese through a Specific Targeted Research Project sponsored by the European Commission entitled “Buruli ulcer: multidisciplinary research for improvement of control in Africa.” In conclusion, the Province of Bas-Congo (Lower Congo) is a region with endemic BU, particularly in the Cataractes District. Prevalence and incidence are high but surveys are urgently needed. The economic and social impacts of BU are believed to be serious but have never been evaluated. One goal of this report was to establish the need of a Program for BU at IME/Kimpese to serve as a pilot project for the management of this disease. A multidisciplinary approach is being implemented to improve BU control. This includes early detection by active case finding and early treatment to prevent severe handicaps, relapses, unusually high death rates, and exorbitant costs.

ACKNOWLEDGMENTS

We thank the staff of the IME/Kimpese as well as the staff of the Mycobacteriology Unit of the ITM/Antwerp for patient care and microbiologic analyses.

FINANCIAL SUPPORT

This study was supported by the Damien Foundation (Brussels, Belgium) and the European Commission (International Science and Technology Cooperation Development Program), project no. INCO-CT-2005-051476- BURULICO. Delphin M. Phanzu received a grant from the Directorate General for Development and Cooperation (Brussels, Belgium).

REFERENCES

Andersen FO. Mycobacterial skin ulcers-Clinical experience. Cent Afr J Med 1965; 11: 131–135.

Bafende AE, Lukanu NP, Numbi AN. Buruli ulcer in an AIDS patient: case report. S Afr Med J 2002; 92: 437. Bafende AE, Phanzu MD, Imposo DB. Buruli ulcer in the Democratic Republic of Congo: epidemiology, presentation and outcome. Trop Doct 2004; 34: 82-4. Bär W, Rüsh-Gerdes S, Richter E, Marquéz de Bär G, Dittmer C, Papsdorf H, et al. M. ulcerans infection in a child from Angola: diagnosis by direct detection and culture. Trop Med Int Health 1998; 3:189-96. DOI: 10.1046/j.1365-3156.1998.00225.x. Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Guédénon A, Scott JT, Dramaix M, Portaels F. Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, Southern Benin, 1997-2001. Emerg Infect Dis 2004a; 10: 1391-8. Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Scott JT, Dramaix M, Portaels F. Mycobacterium ulcerans disease: role of age and gender in incidence and morbidity. Trop Med Int Health 2004b; 9: 1297-304. Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Portaels F. Buruli ulcer recurrence, Benin. Emerg Infect Dis 2005; 11: 584-9. Janssens PG, 1972. Skin ulcers caused by acid-fast bacilli. Marshall J, ed. Essays in Dermatology.Volume 2. Amsterdam: Excerpta Medica: 264–95. Kibadi K, Tsakala TM, Mputu-Yamba JB, Kashongwe M, Muyembe T, Imposo BB. Essai thérapeutique de l’association chloramine- métronidazole-nitrofurantoine dans l’ulcère de Buruli surinfecté. Med Afr Noire 2002; 49: 239–43. Kibadi K, Tsakala TM, Mputu-Yamba JB, Muyembe T, Kashongwe M, Imposo B, Nsiala A. L’ulcère de Buruli chez les réfugiés Angolais des sites de Kimpese, Bas-Congo. Cahiers Santé 2003; 13: 39-41. Meyers WM, Connor DH, McCullough B, Bourland J, Moris R, Proos L. Distribution of Mycobacterium ulcerans infections in Zaire, including the report of new foci. Ann Soc Belg Med Trop 1974; 54: 147-57. Pettit JHS, Marchette NJ, Rees RJW. Mycobacterium ulcerans infection.

Clinical and bacteriological study of the first cases recognized in south east Asia. Br J Dermatol 1966; 78: 187–97. Portaels F, Aguiar J, Fissette K, Fonteyne PA, De Beenhouwer H, De Rijk P, Guédénon A, Lemans R, Steunou C, Zinsou C, Dumonceau JM, Meyers WM. Direct detection and identification of Mycobacterium ulcerans in clinical specimens by PCR and oligonucleotide-specific capture plate hybridization. J Clin Microbiol 1997; 35: 1097–100. Portaels F, Zinsou C, Aguiar J, Debacker M, de Biurrun E, Guédénon A, Josse R, Lagarrigue V, Silva MT, Steunou C, Meyers WM. Les atteintes osseuses dans l’ulcère de Buruli: à propos de 73 cas. Bull Seanc Acad R Sci Outre Mer 2003; 49: 161–90. Portaels F. Epidémiologie des ulcères à Mycobacterium ulcerans. Ann Soc Belg Med Trop 1989; 69: 91-103. Pszolla N, Sarkar MR, Strecker W, Kern P, Kinzl L, Meyers WM, Portaels F. Buruli ulcer: A systemic disease. Clin Infect Dis 2003; 37: e78–82. Smith JH Epidemiologic observations on cases of Buruli ulcer seen in a hospital in the Lower Congo. Am J Trop Med Hyg 1970; 19: 657– 63. Suykerbuyk P, Phanzu MD, Polman K, Bafende E, Meirte D, Leirs H, Trefois P, Portaels F, 2004b. The potential role of freshwater snails, aquatic insects, fish, rodents, lizards and amphibians as reservoirs for Mycobacterium ulcerans. Eighth Maastricht Medical Students Research Conference. Maastricht, The Netherlands: University of Maastricht. Van Oye E, Ballion M. Faudra-t-il tenir compte d’une nouvelle affection à bacilles acido-résistants en Afrique? (Note préliminaire). Ann Soc Belg Med Trop 1950; 30: 619–21. World Health Organization. Buruli Ulcer. Diagnosis of Mycobacterium ulcerans Disease. Portaels F, Johnson P, Meyers WM, eds. Geneva: World Health Organization. 2001. WHO/CDS/GBUI/2001.4. World Health Organization. Buruli ulcer. Mycobacterium ulcerans infection. WHO/CDS/CPE/GBUI/2000/1. Asiedu K, Scherpbier R, Raviglione M, editors. Geneva: The Organization; 2000. p.118.

World Health Organization, 2004. Buruli ulcer disease. Mycobacterium ulcerans infection: an overview of reported cases globally. Wkly Epidemiol Rec 79: 194–200. Yeboah-Manu D, Bodmer T, Mensah-Quainoo E, Owusu S, Ofori-Adjei D, Pluschke G. Evaluation of decontamination methods and growth media for primary isolation of Mycobacterium ulcerans from surgical specimens. J Clin Microbiol 2004; 42: 5875–6.

Table 1. Origin of Buruli ulcer patients, 2002–2004, Democratic Republic of Congo (Bas-Congo)

Table 2. Site of lesions in Buruli ulcer patients

Figure 1. Map of Bas-Congo province with 27 Health Zones among which are three main regions endemic for Buruli ulcer: the zones of Kimpese, Nsona-Mpangu, and Kwilu- Ngongo.

IV.1.2. Short report: edematous Mycobacterium ulcerans infection (Buruli ulcer) on the face: a case report

Delphin M. Phanzu1, Anthony Ablordey2, Desire B. Imposo1, Lucien Lefevre3, Roger L. Mahema1, Patrick Suykerbuyk4, Wayne M. Meyers5, and Françoise Portaels4

1. Institut Médical Evangélique (IME), Kimpese, Bas-Congo, Democratic Republic of Congo (DRC) 2. Noguchi Memorial Institute for Medical Research, Legon, Ghana 3. Onze Lieve Vrouw Hospital, Aalst, Belgium 4. Mycobacteriology Unit, Department of Microbiology, Institute of Tropical Medicine (ITM), Antwerpen, Belgium 5. Armed Forces Institute of Pathology, Washington, D.C.

Personal contribution: Conceived and designed the experiments: DMP; performed the experiments: DMP; analyzed the data: DMP; wrote the paper: DMP.

Am J Trop Med Hyg. 2007 Dec;77(6):1099-102.

Abstract We report a case of a four-year-old Angolan boy with the edematous form of Buruli ulcer on the face and scalp, who was treated at a rural hospital in the Bas-Congo Province, Democratic Republic of Congo. Treatment consisted of a series of surgical interventions and antimycobacterial chemotherapy (rifampin and ciprofloxacin) for 2 months. This case demonstrates the diagnostic and management difficulties of an edematous lesion of BU on the face and suggests an enhancement of healing and limitation of extent of excision by specific antibiotherapy. The outcome in this patient also underscores the importance of prompt referral of suspected cases and training of health professionals in the early diagnosis of BU.

INTRODUCTION

Buruli ulcer (BU) is a severe disabling and disfiguring disease caused by Mycobacterium ulcerans. Its affects primarily children less than 15 years of age in many tropical and subtropical countries.1--3

In early or preulcerative lesions, M. ulcerans produces a lipid toxin, mycolactone, which is responsible for necrosis of the dermis, panniculus, and fascia, culminating in extensive ulcers.4 Preulcerative lesions and small ulcers may be surgically excised and closed. However, antimicrobial therapy is often effective for treatment of early lesions.5 Large ulcers or extensive edematous lesions generally require excision followed by skin grafting.6

The Bas-Congo Province of the Democratic Republic of Congo is endemic for BU and the disease is generally underdiagnosed.7--9 In the surgical ward of the Institut Médical Evangélique Hospital at Kimpese, BU patients are regularly admitted, more than 90% are in an advanced stage of the disease. As required by the Medical Committee of Institut Médical Evangélique, all patients or their responsible relative or guardian must provide informed consent for all diagnostic and treatment procedures and

98 publication of any or all images derived from the management of the patient, including clinical photographs that might reveal patient identity.

We report the diagnostic and management difficulties in a patient with an edematous form of BU localized to the face and scalp.

CASE REPORT A four-year-old Angolan boy from a refugee camp in Mpete, Democratic Republic of Congo, was admitted to the Institut Médical Evangélique Hospital on September 3, 2002, with facial edema that began two weeks earlier with painless swelling of the eyelids (Figure 1). He had no history of trauma or insect bite and his temperature was normal. Various clinical features and types of treatment given over the observation period are shown in Table 1. Several antibiotic regimens were administered: oral ampicillin, and topical chloramphenicol ointment, followed by chloramphenicol and gentamycin, and chloramphenicol, gentamycin, indomethacin, and dexamethasone. Three weeks after admission, a diagnosis of M. ulcerans infection was suspected, but the results of a superficial skin biopsy proved nonspecific. On October 11, 2002 the skin on his forehead ulcerated and a diagnosis of BU was confirmed by the detection of numerous acid-fast bacilli in the necrotic base of the ulcer. On October 22, 2002 necrotic tissue was excised under general anesthesia. Results of a polymerase chain reaction and culture were positive for M. ulcerans, and histopathologic examination showed extensive coagulation necrosis of the deep tissue, vasculitis with thrombosis, numerous foci of mineralization, and areas of acute inflammation. There were clusters of acid-fast bacilli deep in the panniculus. Thus, active BU was confirmed by all four tests recommended by the World Health Organization.1 Ulceration that extended to the right eyelid and base of the nose (Figure 2) was treated by four surgical interventions, daily wound dressing, and skin grafts. In December 2002, lagophthalmos, upper eyelid ectropion, dacryocystitis, and right periorbital active BU lesion developed after the first skin graft.

Beginning on November 11, 2002, rifampin was given for two months in combination with ciprofloxacin to limit extension of the disease to the eyeball (Figure 3). The edema resolved and ulcerative lesions healed rapidly during this period. On January 12, 2003 a fourth excision was performed with limited debridement of residual periorbital lesions after skin grafting. On February 28, 2003, a central tarsorrhaphy, followed by plastic surgery of the superior eyelid, was performed. Results of the procedures were acceptable results with minimal scar formation. In March 2003, the patient was discharged from the hospital after 195 days and there was no relapse at a one year follow-up (Figure 4).

DISCUSSION Foci endemic for BU are most frequent in remote rural areas where there is limited access to adequate health care facilities for treatment of this disease. The indolent course of BU commonly results in patients seeking appropriate medical care only as a last resort, by which time the disease has often reached an advanced edematous or ulcerative stage.

Our patient was referred from an Angolan refugee clinic for edema of the face that developed spontaneously without a known history of insect bite, trauma, fever or pain.10 The differential diagnosis of edematous BU includes cutaneous cellulitis, edema from underlying osteomyelitis, systemic diseases (e.g., renal or cardiac failure), malnutrition, traumatic edema, Kaposi’s sarcoma or lymphatic obstruction. Our patient had been living in Bas-Congo for three years. He frequently accompanied his mother to a field near Kasi village, a well-known BU endemic area.9,11 His parents had a fatalistic attitude, believing that the illness had a mystical origin and was the result of a bewitching, currently called “Mbasu”in Kikongo.12 Belief in a supernatural origin of BU has been reported by several investigators in other African countries.13,14 We

100 frequently organize health education sessions to help BU patients and their families to better understand the nature of the disease.7 Although our patient was initially treated with antibiotics and anti- inflammatory agents, his lesion progressed. Exacerbation of BU under anti-inflammatory treatment has been observed in an Australian patient who was treated with corticosteroids.15 Diagnosis of BU in our patient was initially tentative because of the rarity of patients with the edematous nonulcerative form of BU, particularly diffuse edema of the face, who came to our hospital. Approximately one month after admission, a weeping erythematous preulcerative lesion appeared in the center of his forehead, from which a punch biopsy specimen (3mm) was obtained. It is recommended that specimens for laboratory confirmation from nonulcerative forms (papules, nodules, plaques and edematous forms) should be taken from the center of the surgically excised tissue and should include the entire thickness of infected tissue.1 Fine needle aspiration was recently proposed as an alternative to punch biopsies because it does not require local anesthesia (Eddyani M and others, unpublished data). Until recently, the only effective treatment of BU was surgical excision of the affected tissue.1,16 As early as around 1950, an edematous form of BU on the eyelid of a four-year-old child was diagnosed in eastern Democratic Republic of Congo.3 The difficulties encountered in treating such cases and the need to be surgically conservative were emphasized.3 In the 1960s, some investigators noted healing of early lesions of BU with antibiotics only.17 A number of well-documented cases of early lesions at our hospital in 1970 were treated successfully with rifampin (Meyers WM, unpublished data). The Global Buruli Ulcer Initiative (GBUI) has recently recommended the initial use of conservative treatment, particularly when lesions are located on the face, breast, and genitalia. The suggested approach is to treat the patient with antibiotics (rifampin and streptomycin) for eight weeks, with only debridement and dressings before surgery.16 Our patient had two surgical excisions before antibiotic treatment was initiated. Additional excisions were performed after three and eight weeks

101 of antibiotherapy. However, in addition to surgery, the adjunctive medical treatment may have facilitated healing of our patient; the edema resolved and the ulcerative lesions healed rapidly during medical treatment and only limited debridements were required. For lesions involving the eye, clinicians may consider initial specific antibiotherapy for preulcerative lesions to promote early healing and to avoid ocular complications. Edematous forms of BU and lesions on the face have been reported in different countries. However, there is little information on thei frequency. During a five-year study (1997-2001) of 1,700 patients in Benin, edematous forms were found in only 0.7% of the patients, and lesions were found on the face and neck in 4.2% of children (0-14 years of age).18,19 In Bas-Congo during a six-year period (2002-2007), 16 (7.3%) of 219 patients had lesions on the face and 10 (62.5%) were edematous lesions (Phanzu D, unpublished data). Thus, edematous forms on the face may be more frequent in Bas-Congo than in Benin. These differences deserve special attention. In conclusion, this case report demonstrates the diagnostic and management difficulties of an edematous form of BU on the face and suggests the possible benefit of specific antibiotherapy to enhance healing and limit the extent of surgical excisions. Our observation highlight the importance of this clinical presentation of BU and the need for health workers managing BU lesions of the head and neck to be aware of the special challenges these lesions present. Ophthalmologists, ear, nose and throat specialists, and plastic surgeons should take special note. Prompt referral of suspected BU cases and the training of the staff at both the peripheral level and health professionals at regional hospitals in the early diagnosis of BU are of utmost importance. Finally, all clinicians treating BU patients should stay well informed of developments in the antimicrobial treatment of BU, an apparently fast-moving area of research.

Acknowledgments: We thank the Institut Médical Evangélique in Kimpese staff as well as the staff of the Mycobacteriology Unit of the Institute of Tropical Medicine in Antwerp for patient care and microbiologic

102 analyses. We also thank Gaëlle Antoine for an outstanding work in the preparation of the manuscript.

Financial support: This study was supported by the European Commission (International Science and Technology Cooperation Development Program), Project No. INCO-CT-2005-05-051476- BURULICO.

REFERENCES 1. World Health Organization, 2001. Buruli ulcer. Diagnosis of Mycobacterium ulcerans disease. Portaels F, Johnson P, Meyers WM, eds. Geneva: World Health Organization. WHO/CDS/CPE/GBUI/2001.4. 2. World Health Organization, 2004. Buruli ulcer disease. Mycobacterium ulcerans infection: an overview of reported cases globally. Wkly Epidemiol Rec 79: 194-200. 3. Janssens P, Pattyn S, Meyers WM, Portaels F, 2005. Buruli ulcer: An historical overview with updating to 2005. Bull Seances Acad R Sci Outre Mer 51: 265-299. 4. World Health Organization, 2000. Buruli ulcer--Mycobacterium ulcerans infection. Asiedu K, Sherpbier R, Raviglione M, eds. Geneva : World Health Organization. WHO/CDS/CPE/GBUI/2001.1. 5. Etuaful S, Carbonnelle B, Grosset J, Lucas S, Horsfield C, Phillips R, Evans M, Ofori-Adjei D, Klustse E, Owusu-Boateng J, Amedofu GK, Awuah P, Ampadu E, Amofah G, Asiedu K, Wansbrough-Jones M, 2005. Efficacy of the combination rifampin-streptomycin in preventing growth of Mycobacterium ulcerans in early lesions of Buruli ulcer in humans. Antimicrob Agents Chemother 49: 3182-3186. 6. Sizaire V, Nackers F, Comte E, Portaels F, 2006. Mycobacterium ulcerans infection: control, diagnosis, and treatment. Lancet Infect Dis 6: 288-296. 7. Phanzu DM, Bafende AE, Dunda BK, Imposo DB, Kibadi KA, Nsiangana ZS, Singa NJ, Meyers WM, Suykerbuyk P, Portaels F, 2006.

103

Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg 75: 311-314. 8. Bafende AE, Phanzu MD, Imposo BB, 2004. Buruli ulcer in the Democratic Republic of Congo: epidemiology, presentation and outcome. Trop Doct 34: 82-84. 9. Meyers WM, Connor DH, McCullough B, Bourland J, Moris R, Proos L, 1974. Distribution of Mycobacterium ulcerans infection in Zaïre, including the report of new foci. Ann Soc Belg Med Trop 54: 147-157. 10. Kibadi K, Tsakala TM, Mputu-Yamba JB, Muyembe T, Kashongwe M, Imposo B, Nsiala A, 2003. L’ulcère de Buruli chez les réfugiés Angolais des sites de Kimpese, Bas-Congo. Cahiers Santé 13: 39-41. 11. Portaels F, 1989. Epidémiologie des ulcères à Mycobacterium ulcerans. Ann Soc Belg Med Trop 69: 91-103. 12. Kibadi K, 2004. Enquête connaissances-attitudes-pratiques (CAP) de la population de Songololo (RD Congo) sur l’ulcère de Buruli. Bull Soc Pathol Exot 97: 302-305. 13. Johnson RC, Makoudote M, Hougnihin R, Guédénon A, Ifebe D, Boko M, Portaels F, 2004. Le traitement traditionnel de l’ulcère de Buruli au Bénin. Med Trop 64: 145-150. 14. Kibadi K, Aujoulat I, Meyers WM, Mokassa L, Muyembe T, Portaels F, 2007. Etude des appellations et des représentations attachées à l’infection à Mycobacterium ulcerans dans les différents pays endémiques d’Afrique. Med Trop 67: 241-248. 15. Portaels F, Silva MT, Meyers WM. Buruli ulcer. Clin Dermatol. Accepted. 16. World Health Organization, 2004. Provisional guidance on the role of specific antibiotics in the management of Mycobacterium ulcerans disease (Buruli ulcer). Geneva: World Health Organization. WHO/CDS/CPE/GBUI/2004.10. 17. Meyers WM, 1995. Mycobacterial infections of the skin. Doerr W, Seifert G, eds. Tropical Pathology. Second edition. Berlin: Springer- Verlag, 291-377.

104

18. Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Guédénon A, Scott JT, Dramaix M, Portaels F, 2004. Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital, southern Benin, 1997-2001. Emerg Infect Dis 10: 1391-1398. 19. Debacker M, Aguiar J, Steunou C, Zinsou MD, Meyers WM, Scott JT, Dramaix M, Portaels F, 2004. Mycobacterium ulcerans disease: role of age and gender in incidence and morbidity. Trop Med Int Health 9: 1297-1304.

105

Table 1

Chronology of the clinical features and treatment of the patient ______September 3, 2002 (Figure 1) Admission to hospital Painless edema on the face with complete closure of the eyes Treatment with ampicillin, indomethacin, and Chloramphenicol ointment September 28, 2002 Extension of edema reaching the scalp Clinical diagnosis of Buruli ulcer suggested October 11, 2002 Ulceration in center of forehead Ziehl-Neelsen stain of smear from lesion revealed acid-fast bacilli Skin biopsy performed October 22, 2002 (Figure 2) Wide ulcer on forehead reaching the right eyelid and bridge of nose First excision November 9, 2002 Second excision November 11, 2002 Treatment with rifampin and ciprofloxacin until January 11, 2003 November 21, 2002 Third excision: limited debridement of necrotic tissue December 18, 2002 Skin grafting December 2002 (Figure 3) Lagophthalmos, ectropion, dacryocystitis of the right eye Right periorbital active necrotic ulcerative disease January 12, 2003 Fourth excision: (limited debridement of residual periorbital lesions developing after skin grafting February 28, 2003 Central tarsorrhaphy and blepharoplasty March 2003 Discharge from hospital after 195 days. Follow-up: One year, no relapse observed (Figure 4) ______

106

Figure 1: Edematous form of Mycobacterium ulcerans disease over forehead and upper face.

Figure 2: Yellowish-white necrosis of skin, subcutaneous tissue and fascia, typical of Buruli ulcer.

107

Figur 3: Patient two months after admission, surgery and skin grafting. Active disease remains around patient’s right eye.

Figure 4: Patient one year after discharge. There is no evidence of relapse.

108

IV.1.3. Short Report: Under treated necrotizing fasciitis masquerading as ulcerated edematous Mycobacterium ulcerans infection (Buruli ulcer).

Mavinga D. Phanzu1, Aombe E. Bafende1, Bofunga B.D. Imposo1, Wayne M. Meyers2, and Françoise Portaels3 1. Institut Médical Evangélique (IME), Kimpese, Bas-Congo, Democratic Republic of Congo; 2. Armed Forces Institute of Pathology (AFIP), Washington, DC; 3. Institute of Tropical Medicine (ITM), Antwerp, Belgium.

Personal contribution: Conceived and designed the experiments: DMP; performed the experiments: DMP; analyzed the data: DMP; wrote the paper: DMP.

Am J Trop Med Hyg 2010; 82:478-81.

109

Abstract We report here a case of under-treated necrotizing fasciitis (NF) in a 65- year-old woman with diabetes misdiagnosed as M. ulcerans infection. She presented to the Institut Médical Evangélique (IME) with an extensive painful edematous ulcerated lesion on the dorsum of the right foot and ankle. The diagnosis of Buruli ulcer (BU) was based initially on clinical findings and place of residence (Songololo Territory, the largest known focus of BU in Bas-Congo province). Tissue specimens gave negative results for acid-fast bacilli (AFB), culture and PCR for M. ulcerans. Histopathologic analysis revealed marked necrosis of the lower dermis and subcutaneous tissue. No AFB were found. Later, scattered foci of intracellular gram-positive cocci typical of streptococci were seen. Clinicopathologic correlation of these findings strongly supported the diagnosis of NF. This patient demonstrates the difficulties that may be encountered even in known endemic areas in recognizing BU cases purely on clinical findings.

Mycobacterium ulcerans infection, popularly known as Buruli ulcer (BU), has been identified by the WHO as a rapidly re-emerging disease.1 The disease often afflicts people who live or work in riverine areas, wetlands and stagnant bodies of water.2-3 Penetrating trauma to skin is believed by many authorities to introduce the etiologic agent from M. ulcerans- contaminated skin surface, or the penetrating object itself is similarly contaminated.4-5 The trauma may be as slight as a hypodermic injection or as severe as a gunshot or landmine wound. Following inoculation into the skin, M. ulcerans proliferates and elaborates a toxin that causes necrosis of the dermis, panniculus, and fascia.6-7 The usual incubation period of BU ranges from 2 months to several years.5 As required by the Medical Committee of IME, all patients or their responsible relative or guardian must provide informed consent for all diagnostic and treatment procedures and publication of any or all images derived from the management of patient, including clinical photographs that might reveal patient identity. We report here a case of under-treated necrotizing

110 fasciitis (NF) misdiagnosed on admission to hospital as an ulcerated edematous lesion of M. ulcerans infection. Onset of disease followed well- documented local trauma to the skin of a resident of Songololo Territory, the largest known focus of BU in the Democratic Republic of Congo (DRC).8

CASE REPORT A 65-year-old woman with diabetes from the village of Luvuvamu (Songololo Territory, Bas-Congo Province, DRC) presented on September 9, 2002 to the hospital of the Institut Médical Evangélique (IME) with an extensive painful edematous ulcerated lesion on the dorsum of the right foot and ankle. Seven days prior to admission (September 2, 2002), while tilling the soil, a thorn had pierced the skin deeply at this site. The day following this trauma the site of penetration was surrounded by warm and tender swelling. Her daughter, a nurse, immediately took her to the nearby peripheral health center, where she received analgesic and ampicillin therapy for 6 days. On admission to hospital laboratory tests showed neutrophilia, increased erythrocyte sedimentation rate (ESR) and hyperglycemia. Cefoxitin, metronidazole and ciprofloxacin were administrated for 10 days, with surgical exploration and debridement of devitalized tissue on September 19, 2002. Response to therapy was favourable (Figure 1). A split skin graft was performed 4 weeks later and the patient was discharged from hospital 58 days after admission (Figure 2). The diagnosis of Buruli ulcer (BU) was based initially on clinical findings, place of residence, and of the report of rare acid-fast bacilli (AFB) in the Ziehl-Neelsen (ZN) stained smears. Culture and PCR tests for M. ulcerans were not available locally. Cultivation for bacteria was not performed because the patient had already received antibiotics. Tissue specimens, however, were submitted to the Mycobacteriology laboratory of the Institute of Tropical Medicine (ITM), in Antwerp. Smears, culture and PCR for M. ulcerans were tested and gave negative results for AFB and M.

111 ulcerans9. Histopathologic analysis of formalin-fixed tissue performed at the Armed Forces Institute of Pathology (AFIP) revealed marked necrosis of the lower dermis and subcutaneous tissue, with numerous acute and a few chronic inflammatory cells (Figure 3). No AFB were found. Initially, the histopathologic findings were interpreted as nonspecific acute and chronic inflammation with necrosis (Figure 4). BU was not ruled-out. Later the specimen was studied exhaustively and scattered foci of intracellular gram-positive cocci were seen in Brown-Brenn stained sections (Figure 5). Silver staining showed numerous carcasses of coccal forms (Figure 6). In the lower dermis and panniculus there was vasculitis of small vessels, with dense neutrophilic infiltrates, a few lymphocytes and thrombi (Figure 7). Clinicopathologic correlation of these findings strongly supported the diagnosis of NF. The morphology of the bacteria was interpreted as typical of streptococci.

DISCUSSION The initial BU diagnosis was suspected on clinico-epidemiological features: the patient lived in an endemic region, the lesion was extensive, located on a lower limb, minimally painful, edematous and ulcerated. Whereas NF classically presents acutely with pain, fever and systemic toxicity, often with minimal clinical cutaneous findings. NF can extend to cutaneous structures, causing vasculitis and thrombosis, skin discoloration, crepitus, anesthesia and bulla formation.10 Erythema and edema frequently appear early. Pain disproportionate to local findings may be present and should be a clue to consider NF.11--14 In our patient, ampicillin therapy prescribed in the village probably significantly modified the initial clinical picture. Moreover, we presume that antibiotic therapy had rendered many gram- positive organisms nonstainable by the Brown-Brenn method. Common features of M. ulcerans infection and NF are compared in Table 1. We also have added features of tropical phagedenic ulcer (TPU), an acute or chronic skin disease frequently seen in BU endemic areas15-16 and misdiagnosed as BU17. TPU, mostly solitary18 is characterized by necrosis of the epidermis and underlying superficial tissue16,19.

112

On admission to hospital, the BU diagnosis was also based on the ZN staining done at IME which showed rare AFB. These results, however, could not be confirmed on tissue specimens analyzed at the ITM. Other studies on smear microscopy in tuberculosis concluded that a smear result of ≤3 AFB is unreliable20 and that single smear-positive results require confirmation by another examination. We therefore believe the IME report of AFB was a false-positive result. A study done on BU patients hospitalized at IME from May 2002 to August 2004 showed that the median delay in seeking medical attention was 60 days for all clinical forms21. Often BU patients do not remember the time of onset of the disease. However, those who have had specific trauma at the site where the lesion developed are often able to recall the circumstances more precisely. Published data from such patients indicate that incubation periods range from 2 weeks to 3 years with a mean of 2-3 months.4-5,22-23 A study of six patients diagnosed with NF and treated in the Department of Surgery, University College Hospital, in Ibadan, Nigeria, between January 1998 and December 2001 showed that the duration of symptoms ranged between 3 and 14 days.24 The seven day delay for our patient between the skin trauma and presentation to IME medical center is therefore more compatible with NF than with BU, as well as the reported edema on the day following this trauma.Diabetes, trauma and old age frequently play a role in the rapid spread of NF. NF is commonly caused by a group A streptococcus, (Streptococcus pyogenes), but a variety of aerobic and anaerobic organisms have been reported.10-14,25 Abnormal inflammatory tests are frequent in cases of NF. Laboratory tests in our case showed a neutrophilia and increased ESR. In Bas-Congo, superinfection of skin ulcers by other bacterial species was found in 52.8% of BU patients, with similar abnormal inflammatory tests.21 It is generally believed that in endemic areas, the experienced health worker usually can make an accurate presumptive diagnosis of BU on clinical grounds alone, or by using a combination of clinical features and a positive ZN staining.9 Laboratory confirmation of clinically suspected BU cases, however, is

113 highly recommended. Indeed several studies on the microbiological confirmation of clinically suspected cases of BU have demonstrated that up to 70 % of clinically suspected cases can not be confirmed by microbiological tests.9,26 Moreover, clinicians must remember that misdiagnosis of BU can require the patient to receive 8 weeks of antibiotic therapy with a combination of oral rifampicin and an intramuscular aminoglycoside (usually streptomycin) under direct observation.27 Our patient healed well with a combination of antibiotics, surgery and daily wound dressings followed by skin graft (Figure 2). Our case illustrates the importance of early detection and adequate management of NF, and any other significant tropical skin disease such as TPU or BU. Effective collaboration between peripheral health centers and reference hospitals is of utmost importance. In conclusion, this patient demonstrates the difficulties that may be encountered, even in known endemic areas, in recognizing BU cases purely on clinical findings.

Acknowledgments We thank the staff of the Institut Médical Evangélique in Kimpese and the staff of the Mycobacteriology Unit of the Institute of Tropical Medicine in Antwerp for patient care and microbiologic analyses. We also thank Esther K Meyers for her outstanding work in preparation of the manuscript.

Financial support This study was supported by the European Commission (International Science and Technology Cooperation Development Program), Project No. INCO-CT-2005-05-051476-BURULICO. Author’s addresses: Mavinga D. Phanzu, Aombe E. Bafende, Bofunga B.D. Imposo, Wayne M. Meyers, and Françoise Portaels*. Institut Médical Evangélique (IME), Kimpese, Bas-Congo, Democratic Republic of Congo; Armed Forces Institute of Pathology (AFIP), Washington, DC; *Institute of Tropical Medicine (ITM), Antwerp, Belgium, Tel: 00-32-3-2476317, Fax: 00-32-3-2476333, e-mail: [email protected]

114

REFERENCES 1. World Health Organization, 2000. Buruli ulcer-Mycobacterium ulcerans infection. Asiedu K, Scherpbier R, Raviglione M, eds. Geneva: World Health Organization. 2. Walsh DS, Portaels F, Meyers WM, 2008. Review: Buruli ulcer (Mycobacterium ulcerans infection). Trans R Soc Trop Med Hyg 102: 969-978. 3. Aiga H, Amano T, Cairncross S, Domako JA, Nanas OK, Coleman S, 2004. Assessing water-related risk factors for Buruli ulcer: A case- control study in Ghana. Am J Trop Med Hyg 71: 387-392. 4. Debacker M, Zinsou C, Aguiar J, Meyers WM, Portaels F, 2003. First case of Mycobacterium ulcerans disease (Buruli ulcer) following a human bite. Clin Infect Dis 36: e67-e68. 5. Meyers WM, Shelly WM, Connor DH, Meyers EK, 1974. Mycobacterium ulcerans infections developing at sites of trauma to skin. Am J Trop Med Hyg 23: 919-923. 6. Abalos FMV, Aguiar J, Guédénon A, Portaels F, Meyers WM, 2000. Mycobacterium ulcerans Infection (Buruli ulcer): A case report of the disseminated nonulcerative form. Ann Diagn Pathol 4: 386-390. 7. Portaels F, Silva MT, Meyers WM, 2009. Buruli ulcer. Clin Dermatol 27: 291–305. 8. Meyers WM, Connor DH, McCullough B, Bourland J, Moris R, Proos L, 1974. Distribution of Mycobacterium ulcerans infection in Zaïre, including the report of new foci. Ann Soc belge Méd Trop 54: 47-157. 9. World Health Organization, 2001. Buruli ulcer. Diagnosis of Mycobacterium ulcerans disease. Portaels F, Johnson P, Meyers WM, eds. Geneva: World Health Organization. WHO/CDS/CPE/BBUI/2001.4. 10. Braunwald E, Fauci AS, Kasper DL, Hauser SL, Longo DL, Jameson JL, 2001. 15th Edition Harrison’s Manual of Medicine, McGraw-Hill 176,180,378.

115

11. Dahl PR, Perniciaro C, Holmkvist KA, O’Connor MI, Gibson LE, 2002. Fulminant group A streptococcal necrotizing fasciitis: clinical and pathological findings in 7 patients. J Am Acad Dermatol 47: 489-492. 12. Hasham S, Matteucci P, Stanley PR, Hart NB. Necrotising fasciitis, 2005. Br Med J 330: 830-833. 13. Smeets L, Bous A, Heymans O, 2007. Necrotizing fasciitis: Case report and review of literature. Acta Chir Belg 107: 29-36. 14. Angoules AG, Kontakis G, Drakoulakis E, Vrentzos G, Granick MS, Giannoudis PV, 2007. Necrotising fasciitis of upper and lower limb: a systematic review. Injury 38S: S18-S25. 15. Fleming RA, 1962. The Causes, Pathologic Aspects and Treatment of Phagedenic Ulcer in West Africa. J Int Col Surg 38:120-128. 16. Robinson DC, Adriaans B, Hay RJ, Yesudian P, 1988. The Clinical and Epidemiologic Features of Tropical Ulcer (Tropical Phagedenic Ulcer). Int J Derm 27:49-53. 17. Janssens P, Pattyn S, Meyers WM, Portaels F, 2005. Buruli ulcer: An Historical Overview with Updating to 2005. Bull Séanc Acad R Sci Outre-Mer 51:265-299. 18. Bulto T, Maskel FH, Fisseha G, 1993. Skin lesions in resettled and indigenous populations in Gambela, wiyh special emphasis on the epidemiology of tropical ulcer. Ethiop Med J 31:75-82. 19. Tumwine JK, Dungare PS, Tswana SA, Maoneke WR, 1989. Tropical ulcers in a remote area in Zimbabwe. Centr Afr J Med 35:413-416. 20. Van Deun A, Hamid Salim A, Cooreman E, Daru P, Das APK, Aung KJM, Rieder HL, 2004. Scanty AFB smears: what’s in a name? Int J Tuberc Lung Dis 8: 816-823. 21. Phanzu MD, Bafende AE, Imposo BBD, Dunda KB, Nsiangana ZS, Kibadi KA, Singa NJ, Meyers WM, Portaels F, 2006. Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg 75: 311-314. 22. Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Guédénon A, Scott JT, Dramaix M, Portaels F, 2004. Mycobacterium ulcerans

116

Disease (Buruli ulcer) in a Rural Hospital, Southern Benin, 1997- 2001. Emerg Infect Dis 10: 1391-1398. 23. Van der Werf TS, van der Graaf WTA, Tappero JW, Asiedu K, 1999. Mycobacterium ulcerans infection. Lancet 354: 1013-1018. 24. Ogundiran TO, Akute OO, Oluwatosin OM, 2004. Necrotising fasciitis. Trop Doct 34: 175-178. 25. Cox NH, 1999. Streptococcal necrotizing fasciitis and the dermatologist. Br J Dermatol 141: 613-614. 26. Siegmund V, Adjei O, Nitschke J, Thompson W, Klutse E, Herbinger KH, Thompson R, van Vloten F, Racz P, Fleischer B, Loescher T, BretzelG, 2007. Dry Reagent–Based Polymerase Chain Reaction Compared with Other Laboratory Methods Available for the Diagnosis of Buruli Ulcer Disease. Clin Infect Dis 45:68-75. 27. Provisional guidance on the role of specific antibiotics in the management of Mycobacterium ulcerans disease (Buruli ulcer). WHO/CDS/CPE/GBUI/2004.10 Geneva: World Health Organization; 2004.

117

Table 1. Comparison of common features of Buruli ulcer, necrotizing fasciitis and tropical phagedenic ulcer (TPU)

BURULI ULCER NECROTIZING TROPICAL FASCIITIS PHAGEDENIC ULCER Epidemiology focal in the tropics cosmopolitan mostly tropical Most frequent host age More frequent in all ages children and young and children 15 or diabetes? HIV adults comorbidities younger immunosuppression malnutrition ? and peripheral anemia ? vascular diseases poor hygiene ? Clinical features Body distribution more frequent non-selective most commonly on on limbs lower limbs (mostly lateral malleolus) Trauma related frequently frequently frequently Odor no malodorous strongly malodorous Onset slow (2-3 months) rapid (1-2 days) relatively rapid (2 weeks to 1-2 months) Lesion type nodule, or ulcer, red plaque, or violaceous in color edema swelling ulcer ulcer Pain non-tender tender tender relatively painless painful painful Skin temperature normal warm normal Erythema no yes no Systemic signs none systemic toxicity none fever → shock Etiology M. ulcerans polybacterial polybacterial

118

Figure 1. Wound after surgical excision and daily wound care several days after admission.

Figure 2. Patient in her village several weeks after discharge from hospital. Note good healing without cicatricial retraction.

119

Figure 3. Full thickness section of skin from near the edge of the ulcer, showing inflammation and necrosis, most marked in the deep areas of the panniculus.

Figure 4. Parallel section of panniculus of specimen in Figure 3. Note heavy infiltration of neutrophilis mixed with a few lymphocytes, and necrosis. H&E, x75.

120

Figure 5. Brown-Brenn stained parallel section of area in Figure 4. Showing gram-positive cocci in a small cluster (arrow). There were similar cocci, singly and in clusters, widely scattered in the panniculus (see Figure 6), x330.

Figure 6. Gomori’s methenamine-silver stained parallel section of area in Figure 6 showing many silvered bodies interpreted as carcasses of cocci (arrows). GMS, x330.

121

Figure 7. Vasculitis of small vessels in a parallel section of Figure 3. Note dense neutrophilic infiltrates, a few lymphocytes and thrombus. H&E, x75.

122

IV.1.4. Mycobacterium ulcerans infection (Buruli ulcer) on the face: a comparative analysis of 13 clinically suspected cases from the Democratic Republic of Congo

Delphin M. Phanzu1,4, Roger L. Mahema1, Patrick Suykerbuyk5, Désiré- Hubert B. Imposo1, Linda F. Lehman2, Elie Nduwamahoro5, Wayne M. Meyers3, Marleen Boelaert4 and Françoise Portaels5

1. Institut Médical Evangélique, Kimpese Hospital, Kimpese, Bas-Congo, Democratic Republic of Congo 2. American Leprosy Missions, Greenville, South Carolina 3. Armed Forces Institute of Pathology, Washington, DC 4. Epidemiology and Disease Control Unit, Department of Public Health, Institute of Tropical Medicine, Antwerp, Belgium 5. Mycobacteriology Unit, Department of Microbiology, Institute of Tropical Medicine, Antwerp, Belgium

Personal contribution: Conceived and designed the experiments: DMP; performed the experiments: DMP; analyzed the data: DMP; wrote the paper: DMP.

Am J Trop Med Hyg. 2011 Dec;85(6):1100-5.

123

Abstract

We report our experience in managing 13 consecutive clinically suspected cases of Buruli ulcer on the face treated at the hospital of the Institut Médical Evangélique at Kimpese, diagnosed between 2003 and 2007. During specific antibiotherapy, facial edema diminished, thus minimizing the subsequent extent of surgery and severe disfigurations. The following complications were observed: (1) lagophthalmos in four patients from scarring, with associated ectropion in three of them; (2) blindness in one eye in one patient; (3) disfiguring exposure of teeth and gums resulting from excision of the left labial commissure affected speech, drinking and eating in one patient, and (4) dissemination of M. ulcerans infection in three patients. Our study highlights the importance of this clinical presentation of BU, and the need for health workers in endemic areas to be aware of the special challenges management of BU of the face presents.

INTRODUCTION

Cutaneous infection by Mycobacterium ulcerans, popularly known as “Buruli ulcer,” represents the third most common mycobacterial disease in the world after tuberculosis and leprosy.1 Buruli ulcer (BU) has emerged dramatically over the past decade particularly in West Africa, and has been reported or suspected in 312,3 countries, but confirmed by laboratory tests in only 26 countries.4,5 BU is a severely disfiguring and disabling disease (Figure 1) which affects primarily children under 15 years of age in many tropical and subtropical countries.3-5 Although BU lesions on the face have been reported in different countries there is little information on their frequency.2,5,6-10 The difficulties encountered in treating such cases and the importance of a conservative surgical approach to their management have been emphasized.5,7-10 The Global Buruli Ulcer Initiative (GBUI) of the World Health Organization (WHO) has recently recommended the initial use of conservative

124 treatment, particularly when lesions are located on the face, breast, and genitalia. The suggested approach for treatment of such patients is a combination of specific antibiotics, with or without surgery, and to follow an appropriate prevention of disability (POD) program assiduously.11,12 Our first experience was with a four-year-old Angolan boy (recently published and not included in the present study) who was treated for edematous M. ulcerans infection on the face and drew our attention to the special features and therapeutic requirements of such cases.10 We now report our experience in managing 13 consecutive patients with lesions on the face suspected to be BU at the Institut Médical Evangélique (IME) hospital at Kimpese over a five-year period, and compare their clinical presentations and outcomes.

PATIENTS AND METHODS

The IME hospital at Kimpese serves an area with approximately 150.000 inhabitants, the Kimpese Health Zone in the Songololo Territory, and manages the treatment of most of the BU patients in the Bas-Congo Province of the Democratic Republic of Congo.

In this prospective descriptive study we included all consecutive patients with facial lesions suspect for BU admitted to IME Kimpese hospital, from January 1, 2003 to December 31, 2007. All lesions were first suspected as BU on the basis of clinico-epidemiologic features.2,4 Age, sex, address, date of onset, date of diagnosis, clinical features, hospital stay, type of management, and disease progression were recorded.2

- Specimens of exudates and tissues were analyzed by laboratory tests according to WHO recommendations.2,4 Initial direct smear examination by Ziehl-Neelsen (ZN) stain and histopathologic analysis were performed at the IME/Kimpese laboratory. Other specimens from the same patient were sent in a transport medium13 to the Mycobacteriology Unit of the Institute of Tropical Medicine (ITM) in Antwerp, Belgium, where ZN stain, in vitro culture, and an IS2404 polymerase chain reaction (PCR)

125 were performed with appropriate controls.4,14 Data were entered in an Excel database (Microsoft Corporation, Redmond, WA) and analyzed with Epi-Info version 3.3.2 (Centers for Disease Control and Prevention, Atlanta GA, USA).

As required by the Medical Committee of IME, all patients, or their responsible relative or guardian, provided informed consent for all diagnostic and treatment procedures and publication of any or all images derived from the management of the patient, including clinical photographs that might reveal patient identity.

RESULTS AND DISCUSSION

During the five-year period of our study, 238 suspected BU patients were admitted to the IME hospital, of whom 147 (62%) were confirmed by at least one laboratory test.4 Facial lesions were observed among 13 suspected BU patients of whom 10 (77%) had M. ulcerans disease confirmed by PCR. Thus, the frequency of facial BU was 5% (13 of 238) among suspected BU patients and 7% (10 of 147) among confirmed BU patients. These results are in agreement with the frequency of facial BU lesions reported in the literature (between 0.8 and 16.7%) as shown in Table 1.

The characteristics of the 13 patients are presented in Table 2.

Nine out of 13 patients (69,2%) presented with ulcerated lesions and swabs were first taken for ZN staining of smears, followed by swabs for PCR and culture. For patients with preulcerative lesions, neither punch nor incisional skin biopsies were employed to minimize initiating of the ulceration process. In such patients fine needle aspiration (FNA) was used for rapid diagnosis of facial BU and if an ulcer developed later, swabs were taken from the early ulcerative lesion. Specimens of tissue were also obtained during debridements. Based on our experience, we recommend

126

ZN on swabs from ulcers or FNA from preulcerative forms for rapid diagnosis of facial BU.15 Punch or surgical biopsies from preulcerative forms should be avoided: in our experience biopsies from preulcerative forms became the starting point of extensive ulceration.10 Laboratory tests were negative for two cases (Table 2). One of these patients had ulcerated lesions on the neck and cheek. Conditions other than BU (eg. cervical mycobacterial lymphadenitis) should not be ruled out. In this case, histopathology was not performed. The case emphasizes the difficulties in recognizing some BU cases on a clinical basis only, and stresses the importance of additional laboratory tests including histopathologic analysis for the differential diagnosis of BU.

The median age was 11 years (age range = 4 to 68 years) in our series. Consistent with other studies, the age distribution showed that 9 (69.2%) of 13 patients were less than 15 years of age 7,8,12,16, 3 (23.1%) were 15- 45 years of age, and one (7.7%) was 68 years old. There were 8 males and 5 females, yielding a sex ratio of 1.6:1, although this difference may not be significant, as in several other studies, case numbers of BU were similar for males and females.2,9,12,17,18 Origins of patients were: 12 (92.3%) from Cataractes District (11 from Songololo Territory, and one from the Mbanza-Ngungu Territory), and one patient came from the Bas-Fleuve District (Sekebanza Territory). This geographic distribution is consistent with other studies in the same area.14,17,18 Median delay in seeking medical attention was 12 weeks with a range of 3 weeks to 52 weeks. This delay seems prolonged compared to a previous report from the same area.14 Despite the extent of necrosis and tissue damage, BU lesions are usually painless. This absence of pain often deters patients from seeking early treatment, resulting in severe sequelae, including amputations of limbs.19,20 The different clinical forms were distributed as follows: mixed forms in 61.5%, single ulcerative lesions in 15.4%, and single edematous forms in 23.1%. There was bone involvement in two (15.4%) of these patients.

127

Three patients had disseminated lesions affecting the head and lower limbs. Edema was present in 9 (69.2%) patients with tendency to regional infiltration, particularly among children (Figure 2). A similar observation was reported recently in Ivory Coast.7 The question arises whether or not children are more likely to develop extensive facial edema than older individuals. Extensive edematous forms are considered to be among the category of severe disseminated lesions.1 Why children are prone to develop edematous forms remains unknown but the cause is probably multifactorial, including immunity. Children may have a less effective protective immune response against M. ulcerans than adults.1 The evidence that mycolactone plays a major role in the pathogenicity of BU has been established.2,19,21 However, factors determining the extent of the lesions are unknown.21 Evans and Wansbrough-Jones, and Meyers and Hayman suggest that the host’s cell-mediated immune response is likely to play an important role in BU.2 In Benin, Debacker et al. noted that children tended to develop the disease more often on the trunk, head, neck and upper limbs. Because of their shorter stature, the entire body of children is nearer to the soil which may explain why children are more apt than adults to develop lesions on the head.9 No routine HIV-testing was done in our series, so we cannot exclude that immune suppression could play a role in facial presentations; however, HIV-AIDS is infrequent in children in this area of DR Congo. The periorbital region was the site of predilection among our patients. Similar observations have been reported recently in the study mentioned above.7 Table 3 shows the location of the lesions on the face and on other parts of the body for those with disseminated disease. Among our patients, 12 were new cases (92.3%) and 1 a relapse (7.7%). Physical impairments were present in 5 patients (38.5%) initially when admitted in hospital and at the end of treatment. Initial impairments consisted essentially of significant swelling with eye closure (Figure 3) in four patients and limited eyelid closure for one patient.

128

Nine patients (69.2%) were treated by a combination of antibiotics (rifampicin and streptomycin) and surgery. Three patients (23.1%) were treated only with surgery, and one (7.7%) was treated only with antibiotics (Figure 4). Among patients treated with antibiotics, 8 received 60 days of antibiotic treatment, and it was prolonged to 90 days for 2 patients (Table 2). Surgery was performed at different times (during or after antibiotic treatment) when considered necessary. Seven patients (53.8%) received skin grafts. In total, twelve patients were surgically treated (92.3%). The degree of facial edema, in all patients where it was present, was reduced during antibiotherapy, minimizing the extent of surgery, thus helping to prevent severe disfigurement. We noted in our series that clinical response to specific antibiotic therapy seemed to be related to the stage of the lesion; the earlier the diagnosis of a BU the facial lesion, the better the response to antibiotic therapy. In the 1960s, Meyers, who also worked at IME Kimpese hospital, noted healing of early lesions of BU after treatment with rifampicin alone.22 A recent report suggests the possible benefit of specific antibiotherapy to enhance healing and limit the extent of surgical excisions in edematous forms of BU on the face.10 Complications observed in our study were as follows: lagophthalmos in four patients from scarring, with associated ectropion in three of them, loss of sight in one patient (Figure 5), disfigurement with exposure of teeth and gums, excision of left labial commissure in one patient, and BU dissemination despite specific antibiotherapy in two patients. Preservation of vision is of upmost importance and special attention must be given to minimize the risk of corneal dryness, ulcers, and infection. If the eyelids cannot close fully, the cornea needs to be lubricated with artificial tears and with ointment at night. Protection of the eyes during the day with hats and glasses and at night with an eye shield can further reduce dryness from exposure and protect the eye from dust and other foreign objects. Frequent exercise to close the eye helps lubricate the cornea, strengthen weakened muscles and maintain full movement for eye closure. Gentle massage over the scar area softens it, stretches it and

129 lessens the adhesions to adjacent structures that can limit movement. Taping can also be used to facilitate good eyelid positioning. Sopoh et al. reported difficulties in managing the technical and cosmetic aspects of lesions around the head, neck, and trunk in Benin, where plastic surgeons were not available.16 Our study highlights the importance of this clinical presentation of BU, and the need for health workers in endemic areas managing BU lesions of the face to be aware of the special challenges such lesions present. Scars can cause the lower eyelid to evert (ectropion) and adhesions to adjacent structures can make it difficult to complete full eye closure. Edema can cause the eyelids to invert (entropion) creating problems with eyelashes turning in and touching the cornea (trichiasis). Special attention to care for the eye is important to preserving vision even when there is an early diagnosis. Complete eye closure is important; the affected eye should open and close the same as the unaffected side. If not, exercises should be given. The cornea should be lubricated when there is incomplete eye closure with corneal exposure by using artificial tears during the day and ointment at night. An eye shield at night will provide extra protection and prevent foreign matter from falling into the eye that does not completely close. Surgery may be required to correct eyelid and eye lash positions. Excessive edema can cause entropion inverting the eyelash position causing them to touch the cornea during blink. Ectropion involving the lower eyelid may occur as ulcers heal and the strong contracting forces scar tissue limits the eye from closing completely. Scar management and exercises may improve eyelid closure but if the eyelid correction is not obtained with these simple measures, surgical correction of the eyelid is required. If the eye must be removed, special surgical procedures should be given during the initial excision to assure that a prosthesis can be fit in the future. The faces of children who are not fitted with a prosthesis will not develop normally and their faces will be asymmetrical. The involvement of the maxillar and mandibular areas can create functional problems for speaking, eating and drinking. Stigma can also be a major problem for BU lesions on the face.23

130

Thus, community education for early case detection, chemotherapy, POD and rehabilitation, and reconstructive surgery, all become key components of the current BU control strategy.12

The median hospital stay in this series of patients was 125 days (range 32-301 days). This range is similar to that of a previous study in the same geographic area.14

Follow-up at one year detected three relapsed patients (23%), two at the original site (facial) and one at a remote site (lower limb). Among the relapsed patients, two had been previously treated by surgery alone and one by antibiotics alone.

Concerning all patients in our study who received specific antibiotics (n=10), we observed only one relapse (10%). As no samples were taken for culture, we cannot exclude a possible paradoxical reaction to dead bacilli in this patient. The GBUI reported recently that recurrences after surgical treatment alone have fallen to < 2% following the introduction of antibiotics. The development of antibiotic treatment is one of many major advances in the 10 years since the GBUI began.12

In conclusion, effective care and support methods to improve the management of BU include community education to improve hygiene and awareness to the early detection by a community-based health surveillance system, early antibiotic treatment, with or without surgery, and POD measures. For rapid diagnosis of facial BU, ZN stain on swabs from ulcers or fine needle aspiration (FNA) from preulcerative forms is recommended. Our study underscores the importance of prompt referral of suspected BU cases, and training of health professionals in BU endemic areas in the early diagnosis of BU, including atypical localizations.

Acknowledgements We thank the staff of the IME/Kimpese as well as the staff of the Mycobacteriology Unit of the ITM/Antwerp for patient care and microbiologic analyses. We thank Bouke de Jong for critical reading of this

131 manuscript. We also thank Karin Janssens for outstanding work in preparation of the manuscript.

Financial support This study was supported by the American Leprosy Missions (Greenville, SC, USA) and the European Commission (International Science and Technology Cooperation Development Program), Project No. INCO-CT- 2005-05-051476-BURULICO. Address correspondence: Phanzu M. Delphin, Institut Médical Evangélique, Kimpese Hospital, Kimpese, Bas-Congo, Democratic Republic of Congo. E- mail: [email protected]

REFERENCES

1. Portaels F, Silva MT, Meyers WM, 2009. Buruli ulcer. Clin Dermatol 27: 291-305. 2. World Health Organization, 2000. Buruli ulcer-Mycobacterium ulcerans infection. Asiedu K, Sherpbier R, Raviglione M, eds. Geneva : World Health Organization. WHO/CDS/CPE/GBUI/2001.1. 3. World Health Organization, 2004. Buruli ulcer disease. Mycobacterium ulcerans infection: an overview of reported cases globally.Wkly Epidemiol Rec 79: 194-200. 4. World Health Organization, 2001. Buruli ulcer. Diagnosis of Mycobacterium ulcerans disease. Portaels F, Johnson P, Meyers WM, eds. Geneva: World Health Organization WHO/CDS/CPE/GBUI/2001.4. 5. Janssens P, Pattyn SR, Meyers WM, Portaels F, 2005. Buruli ulcer: An historical overview with updating to 2005. Bull Séances Acad R Sci Outre Mer 51: 265-299. 6. Ouattara D, Aka GK, Meningaud JP, Sica A, Kaba L, Gadegbeku S, 2003. Les localisations faciales de l’ulcère de Buruli : à propos de deux cas. Rev Stomatol Chir Maxillo Fac 104: 231-234.

132

7. Kouame K, Ecra E, Cisse M, Gbery I, Kacou E, Kassi K, Kouassi A, Ahogo C, Yoboue P, Kanga JM, 2008. Ulcère de Buruli céphalique : huit observations au CHU d’Abidjan, Côte d’Ivoire. Méd Trop 68: 643- 644. 8. Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, Mbuagbaw J, et al. 2004. Buruli ulcer disease in Cameroon rediscovered. Am J Trop Med Hyg 70:520-526.

9. Debacker M, Aguiar J, Stenou C, Zinsou C, Meyers WM, Scott JT, Dramaix M, Portaels F, 2004. Mycobacterium ulcerans disease: role of age and gender in incidence and morbidity. Trop Med Int Health 9: 1297-1304. 10. Phanzu MD, Ablordey A, Imposo BD, Lefevre L, Mahema RL, and Suykerbuyk P, Meyers WM, Portaels F, 2007. Short report: Edematous Mycobacterium ulcerans infection (Buruli ulcer) on the face: A case report. Am J Trop Med Hyg 77: 1099-1102. 11. World Health Organization, 2004. Provisional guidance on the role of specific antibiotics in the management of Mycobacterium ulcerans disease (Buruli ulcer), WHO/CDS/CPE/GBUI/2004.10 Geneva: The Organization; 2004. 12. World Health Organization, 2008. Buruli ulcer: progress report, 2004- 2008. Wkly Epidemiol Rec 83: 145-156. 13. Eddyani M, Debacker M, Martin A, Aguiar J, Johnson RC, Uwizeye C, Fissette K, Portaels F, 2008. Primary culture of Mycobacterium ulcerans from human tissue specimens after storage in a semi-solid transport medium. J Clin Microbiol 46: 69-72. 14. Phanzu DM, Bafende AE, Dunda BK, Imposo DB, Kibadi KA, Nsiangana ZS, Singa NJ, Meyers WM, Suykerbuyk P, Portaels F, 2006. Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg 75: 311-314. 15. Eddyani M, Fraga AG, Schmitt F, Uwizeye C, Fissette K, Johnson C, Aguiar J, Sopoh G, Barogui Y, Meyers WM, Pedrosa J, Portaels F, 2009. Fine-Needle Aspiration, an efficient sampling technique for

133

bacteriological diagnosis of nonulcerative Buruli ulcer. J Clin Microbiol 47: 1700-1704. 16. Sopoh GE, Johnson RC, Chauty A, Dossou AD, Aguiar J, Salmon O, Portaels F, Asiedu K, 2007. Buruli ulcer surveillance, Benin, 2003- 2005. Emerg Infect Dis 13: 1374-1376. 17. Bafende AE, Phanzu MD ,Imposo BB, 2004. Buruli ulcer in the Democratic Republic of Congo: epidemiology, presentation and outcome. Trop Doct 34: 82-84. 18. Portaels F, 1989. Epidémiologie des ulcères à Mycobacterium ulcerans, Ann Soc Belge Méd Trop 69: 91-103. 19. En J, Goto M, Nakanaga K, Higashi M, Ishii N, Saito H, Yonezawa S, Hamada H, Small PL, 2008. Mycolactone is responsible for the painlessness of Mycobacterium ulcerans Infection (Buruli Ulcer) in a murine study. Infect Immun 76: 2002-2007. 20. Kibadi K, Boelaert M, Kayinua M, Minuku JB, Muyembe-Tamfum JJ, Portaels F, Lefèvre P, 2009. Therapeutic itineraries of patients with ulcerated forms of Mycobacterium ulcerans (Buruli ulcer) disease in a rural health zone in the Democratic Republic of Congo. Trop Med Int Health 14: 1110-1116. 21. Silva MT, Portaels F, Pedrosa JR, 2009. Pathogenetic mechanisms of the intracellular parasite Mycobacterium ulcerans leading to Buruli ulcer. Lancet Infect Dis 9: 699-710. 22. Meyers WM, 1995. Mycobacterial infections of the skin. Doerr W, Seifert G, eds. Tropical Pathology . Second edition. Berlin: Springer- Verlag, 291-377. 23. World Health Organization, 2006. Buruli ulcer-Prevention of disability. Lehman L, Simonet V, Saunderson P, Agbenorku P, eds. Geneva: World health Organization. 24. Connor DH, Lunn HF, 1966. Buruli ulceration. Arch Pathol 81:183- 199. 25. Reid IS, 1967. Mycobacterium ulcerans infection: A report of 13 cases at the Port Moresby General Hospital, Papua. Med J Aust 1:427-431.

134

26. Smith JH, 1970. Epidemiologic observations on cases of Buruli ulcer seen in a hospital in the Lower Congo. Am J Trop Med Hyg 19:657- 663. 27. Barker DJP, 1971. Buruli disease in a district of Uganda. J Trop Med Hyg 74:260-264. 28. The Uganda Buruli Group, 1971. Epidemiology of Mycobacterium ulcerans infection (Buruli ulcer) at Kinyara, Uganda. Trans R Soc Trop Med Hyg 65:763-775. 29. Barker DJP, 1972. The distribution of Buruli disease in Uganda. Trans R Soc Trop Med Hyg 66:867-874. 30. Meyers WM, Connor DH, McCullough B, Bourland J, Moris R, Proos L, 1974. Distribution of Mycobacterium ulcerans infection in Zaïre, including the report of new foci. Ann Soc belge Méd Trop 54:147-157. 31. Oluwasanmi JO, Solankee TF, Olurin EO, Itayemi SO, Alabi GO, Lucas AO, 1976. Mycobacterium ulcerans (Buruli) skin ulceration in Nigeria. Am J Trop Med Hyg 25:122-128. 32. Ravisse P, 1977. L’ulcère cutané à Mycobacterium ulcerans au Cameroun. I. Etude clinique, épidémiologique et histologique. Bull Soc Pathol Exot 70 :109-124. 33. Van der Werf TS, Van der Graaf WTA, Groothuis DG, Knell AJ, 1989. Mycobacterium ulcerans infection in Ashanti region, Ghana. Trans R Soc Trop Med Hyg 83:410-413. 34. Darie H, Le Guyadec T, Touze JE, 1993. Aspects épidémiologiques et cliniques de l’ulcère de Buruli en Côte d’Ivoire. A propos de 124 observations récentes. Bull Soc Pathol Exot 86 :272-276. 35. Josse R, Guédénon A, Aguiar J, Anagonou S, Zinsou C, Prost C, Foundohou J, Touze JE, 1994. L’ulcère de Buruli, une pathologie peu connue au Bénin. A propos de 227 cas. Bull Soc Pathol Exot 87 :170- 175. 36. Addo HA, 1995. Mycobacterium ulcerans infection (Buruli ulcer) in Ga District of Greater Accra Region. Ghana Med J 29:595-602.

135

37. Aujoulat I, Huguet-Ribas MP, Koïta Y, 1996. L'ulcère de Buruli: un problème de santé publique méconnu, appelant une mobilisation internationale. Dév Santé 125:22-30. 38. Aguiar J, Domingo MC, Guédénon A, Meyers WM, Stenou C, Portaels F, 1997. L’ulcère de Buruli, une maladie mycobactérienne importante et en recrudescence au Bénin. Bull Séanc Acad R Sci Outre-Mer 3 :325-356. 39. Asiedu K, Etuaful S, 1998. Socioeconomic implications of Buruli ulcer in Ghana: a three-year review. Am J Trop Med Hyg 59:1015-1022. 40. Ouoba K, Sano D, Traoré A, Ouédraogo R, Sakandé B, Sanou A, 1998. Les ulcères cutanés à Mycobacterium ulcerans au Burkina Faso: à propos de six observations et revue de la littérature. Nouv Dermatol 17 :358-362. 41. Amofah G, Bonsu F, Tetteh C, Okrah J, Asamoa K, Asiedu K, Addy J, 2002. Buruli ulcer in Ghana : results of a national case search. Emerg Infect Dis 8 :167-170. 42. Schierle HP, Lemperle G, Erdmann D, 2002. The Buruli type ulcer. Plast Reconstr Surg 109:2608. 43. Kibadi K, Tsakala M, Mputu-Yamba JB, Muyembe T, Kashongwe M, Imposo B, Nsiala A, 2003. L’ulcère de Buruli chez les réfugiés angolais des sites de Kimpese, Bas-Congo, RD Congo. Santé 13 :39- 41. 44. James K, Attipou KK, James YE, Blakime M, Tignokpa N, Abete B, 2003. L’ulcère de Buruli au Togo: à propos d’une enquête hospitalière. Cah Santé 13 :43-47. 45. Hospers IC, Wiersma IC, Dijkstra PU, Stienstra Y, Etuaful S, Ampadu EO, van der Graaf WT, van der Werf TS, 2005. Buruli ulcer in Ghana: results of a national case search. Distribution of Buruli ulcer lesions over body surface area in a large case series in Ghana: Uncovering clues for mode of transmission. Trans R Soc Trop Med Hyg 3:196- 201.

136

46. Raghunathan PL, Whitney EA, Asamoa K, Stienstra Y, Taylor TH Jr, Amofah GK, Ofori-Adjei D, Dobos K, Guarner J, Martin S, Pathak S, Klutse E, Etuaful S, van der Graaf WT, van der Werf TS, King CH, Tappero JW, Ashford DA, 2005. Risk factors for Buruli ulcer disease (Mycobacterium ulcerans infection): results from a case-control study in Ghana. Clin Infect Dis 40:1445-1453. 47. Johnson PD, Azuolas J, Lavender CJ, Wishart E, Stinear TP, Hayman JA, Brown L, Jenkin GA, Fyfe JA, 2007. Mycobacterium ulcerans in mosquitoes captured during outbreak of Buruli ulcer, southeastern Australia. Emerg Infect Dis 11:1653-1660. 48. Pouillot R, Matias G, Wondje CM, Portaels F, Valin N, Ngos F, et al. 2007. Risk factors for Buruli ulcer: a case control study in Cameroon. PLoS Negl Trop Dis 1:e101.

137

Table 1. Reports of facial Buruli ulcer disease, 1966-2008.

Year of Country Study design Number of % Reference report patients 1966 Uganda Case series 1/38 2.6 Connor and Lunn24 1967 Papua Case report 2/13 15.4 Reid25 1970 DRC Case series 19/93 20.4** Smith26 1971 Uganda Case series 30/164 18.3** Barker27 1971 Uganda Case series 39/220 17.7** Uganda Buruli Group28 1972 Uganda Case series 27/539 5.0 Barker29 1974 DRC Review 1/14 7.1 Meyers et al.30

1976 Nigeria Case report 1/22 4.5 Oluwasanmi et al.31 1977 Cameroon Case report 4/37 10.8** Ravisse et al.32 1989 Ghana Case series 12/96 12.5 Van der Werf et al.33 1993 Ivory Coast Case series 1/124 0.8 Darie et al.34 1994 Benin Case series 7/227 3.1 Josse et al.35 1995 Ghana Case series 1/26 3.8 Addo36 1996 Ivory Coast Review 1 NA Aujoulat et al.37 1997 Benin Case series 23/867 2.7 Aguiar et al.38

1998 Ghana Case series 1/102 1.0 Asiedu and Etuaful39 1998 Burkina Case report 1/6 16.7 Ouoba et al.40 Faso 2002 Ghana Case series 219/5770 3.8 Amofah et al.41 2002 Togo Case report 1* NA Schierle et al.42

2003 DRC Case series 1/27 3.7 Kibadi et al.43 (Angola) 2003 Ivory Coast Case series 2* NA Ouatttara et al.6 2003 Togo Case series 7/160 4.4 James et al.44

2004 Benin Case series 58/1700 3.4 Debacker et al.9 2004 Cameroon Cross- 4/202 2.0 Noeske et al.8 sectional 2005 DRC Review 1 NA Janssens et al.5 2005 Ghana Case series 32/748 4.3 Hospers et al.45 2005 Ghana Case-control 5/121 4.1 Raghunathan et al.46

138

2007 Australia Case series 1/79 1.3 Johnson et al.47 2007 DRC Case report 1* NA Phanzu et al.10

2007 Benin Case series 19/2598 0.7 Sopoh et al.16 2007 Cameroon Case-control 3/163 1.8 Pouillot et al.48 2008 Ivory Coast Case series 8* NA Kouame et al.7

Note: DRC = Democratic Republic of Congo; NA = not applicable. *Number in specific or dedicated report of facial Buruli ulcer disease **Head, neck and trunk combined

139

Table 2. Characteristics of patients with facial BU admitted at IME Kimpese hospital, 2003-2007

Age Gender Classification Clinical forms Sites Lesions Patient delay Hospital Treatment Laboratory Physical impairment Follow-up (1st Year) Confirmation (Years) New/Relapse (Number) (Weeks) stay R+S Surgery by Initial At the end Relapse (Days) (Days) Yes/No 18 Male N U,O* LL,Head 3 52 80 No Yes PCR Lagoph Lagoph Yes

68 Male N E,U Head 1 3 68 No Yes Neg No Lagoph No

5 Male N E Head 1 12 213 Yes (90) Yes ZN,PCR PEC Lagoph,LOS No

11 Male N E,U Head 1 8 125 Yes (90) Yes ZN,PCR No No No

17 Male N U Head 1 12 106 Yes (60) Yes ZN,PCR No No No

17 Male R Q,U,N LL,Head 2 12 84 Yes (60) Yes PCR No No No

12 Male N E Head 2 4 127 Yes (60) Yes PCR No No No

9 Female N E,U,O* Head 1 16 301 Yes (60) Yes ZN,PCR,HIS No Lagoph,ELC No

5 Female N U Head, neck 1 48 32 No Yes Neg No No Yes

4 Female N E,P Head 2 8 66 Yes (60) No NT PEC No Yes

4 Female N E,U LL,Head 3 24 297 Yes (60) Yes ZN,PCR PEC No No

5 Female N E Head 3 6 133 Yes (60) Yes ZN,PCR PEC No No

13 Male N E,U Head 1 16 151 Yes (60) Yes ZN,PCR No No No

N, New case; R, Relapse (new diagnosis of BU less than one year after being declared cured with previous antibiotic and/or surgical treatment); U, Ulcer; O, Osteomyelitis (*Not laboratory confirmed); E, Edema; P, Papule; Q, Plaque; N, Nodule; LL, Lower Limb; R+S, Rifampicin & Streptomycin; HIS, Histopathologic analysis; Neg, Negative; NT, not tested ; Lagoph, Lagophthalmos; PEC, Permanent eye closure; LOS, Loss of sight (right eye); ELC, Excision of labial commissure.

Table 3. Site of lesions among patients with facial lesions (n=13) admitted at IME Kimpese hospital from 2003 to 2007

Site of lesions Children % Adults % Total %

N=9 N=4 N=13

Periorbital 6* 66.7 1* 25 7 53.8

Periorbital & 1 11.1 1 25 2 15.4 cheek

Cheek 2§ 22.2 2* 50 4 30.8

* One patient with another lesion on the lower limb (disseminated BU case) § One patient with lesion reaching the neck

141

Figure 1. Illustration of late diagnosis: A large disfiguring BU in a 9 years old female patient from Songololo.

Figure 2. Periorbital advanced edematous lesions on right eye with early extension on left side

142

Figure 3: Papula on the temple and painless swelling of the right eyelids (early detection).

Figure 4: Patient (See Figure 3) one year after specific antibiotic treatment without surgery. No complications.

143

Figure 5: Loss of sight in the right eye, no complications on left side (same patient in Figure 2).

144

IV.2. Partie santé publique

"L’intégration d’activités de programmes dans des services de santé polyvalents est plus qu’une simple décentralisation opérationnelle des activités dans l’espace, d’une structure centralisée spécialisée vers des structures décentralisées. Elle implique un véritable transfert de responsabilités, des droits et des devoirs vers les ‘horizontalistes’, c’est-à- dire le personnel responsable de la gestion des structures de santé polyvalentes. Elle ne signifie pas que le programme vertical disparaisse, ni que le personnel spécialisé n’a plus de rôle à jouer, bien au contraire."

Criel B & De Brouwere V. Conditions, limites et potentiel de l’intégration In Intégrations et Recherche, Van Lerberghe W & de Béthune X eds. Studies in Health Services Organisation & Policy 1998; 8:83-109, Antwerpen:ITGPress.

145

146

IV.2.1. Effect of a control project on clinical profiles and outcomes in Buruli ulcer: a before/after study in Bas-Congo, Democratic Republic of Congo.

Delphin Mavinga Phanzu1,7, Patrick Suykerbuyk2, Désiré Bofunga B. Imposo1, Philippe Ngwala Lukanu3, Jean-Bedel Masamba Minuku4, Linda F. Lehman5, Paul Saunderson5, Bouke C. de Jong2, Pascal Tshindele Lutumba6, Francoise Portaels2, Marleen Boelaert7

1 General Reference Hospital of Kimpese, Institut Medical Evangelique, Kimpese, Bas- Congo, Democratic Republic of Congo 2 Department of Microbiology, Mycobacteriology Unit, Institute of Tropical Medicine, Antwerp, Belgium 3 Central Office of the Rural Health Zone of Kimpese, Bas-Congo, Democratic Republic of Congo 4 Central Office of the Rural Health Zone of Nsona Mpangu, Bas-Congo, Democratic Republic of Congo 5 American Leprosy Missions, Greenville, South Carolina, United States of America 6 Institut National de Recherche Biomédicale, Kinshasa, Democratic Republic of Congo 7 Department of Public Health, Unit of Epidemiology and Disease Control, Institute of Tropical Medicine, Antwerp, Belgium

Personal contribution: Conceived and designed the experiments: DMP; capacity strengthening: DMP; performed the experiments: DMP; analyzed the data: DMP; wrote the paper: DMP.

PLoS Negl Trop Dis. 2011 Dec;5(12):e1402

147

Abstract

Background: Buruli ulcer (BU) is a necrotizing bacterial infection of skin, subcutaneous tissue and bone caused by Mycobacterium ulcerans. Although the functional impairment caused by BU results in severe suffering and in socio-economic problems, the disease remains largely neglected in Africa. The province of Bas-Congo in Democratic Republic of Congo contains one of the most important BU foci of the country, i.e. the Songololo Territory in the District of Cataractes. This study aims to assess the impact of a BU control project launched in 2004 in the Songololo Territory. Methods: We used a comparative non-randomized study design, comparing clinical profiles and outcomes of the group of patients admitted at the General Reference Hospital (GRH) of the ''Institut Médical Evangélique'' (IME) of Kimpese 3 years before the start of the project (2002-2004) with those admitted during the 3 years after the start of the project (2005-2007). Results: The BU control project was associated with a strong increase in the number of admitted BU cases at the GRH of IME/Kimpese and a fundamental change in the profile of those patients; more female patients presented with BU, the proportion of relapse cases amongst all admissions reduced, the proportion of early lesions and simple ulcerative forms increased, more patients healed without complications and the case fatality rate decreased substantially. The median duration since the onset of first symptoms however remained high, as well as the proportion of patients with osteomyelitis or limitations of joint movement, suggesting that the diagnostic delay remains substantial. Conclusion: Implementing a specialized program for BU may be effective in improving clinical profiles and outcomes in BU. Despite these encouraging results, our study highlights the need of considering new strategies to better improve BU control in a low resources setting.

148

INTRODUCTION

Buruli ulcer (BU) is a necrotizing bacterial infection of skin, subcutaneous tissue and bone, caused by an environmental pathogen, Mycobacterium ulcerans (Portaels et al., 2009). Although the functional impairment caused by BU results in severe suffering and in socio-economic problems (Asiedu et al., 1998), the disease remains largely neglected by health authorities in Africa (Aujoulat et al., 1996). BU is considered as one of the Neglected Tropical Diseases with a poorly known global prevalence (Hotez et al., 2007). The province of Bas-Congo (Lower Congo) in the Democratic Republic of Congo (DRC) contains one of the most important BU foci of the country, i.e. the Songololo Territory in the District of Cataractes (Andersen et al., 1965; Smith et al., 1970; Meyers et al., 1974; Portaels et al., 1989; Kibadi et al., 2003; Phanzu et al., 2006). Meyers et al. reported that BU existed in that region before 1935 on the basis of interviews of former patients (Meyers et al., 1974). The first BU case reports were published in the sixties (Andersen et al., 1965; Smith et al., 1970; Meyers et al., 1974) followed by a long period without reported cases. Since 1999, the general reference hospital (GRH) of the Institut Médical Evangélique (IME)/Kimpese, located in the Songololo Territory, 220 km southwest of Kinshasa, regularly admits BU cases. Between 2002 and 2004 this hospital admitted 64 patients, 95% of them in the ulcerative stage. During this period, 48 patients out of 64 (75%) were referred by government health centers or other health professionals, 9 (14.1%) by family members, and 7 (10.9%) presented spontaneously. Surgery was the main method of treatment applied amongst these patients (93.7%). An abnormally high case fatality rate (18.7%) was observed among these 64 patients, and whereas 36% presented already a functional limitation at the time of diagnosis, 23% were discharged with permanent disability. The median length of hospitalization was 89 days and, -noteworthy- 90% of the patients were not able to pay their hospitalization costs. To address these poor clinical outcomes, the

149

American Leprosy Mission and the IME hospital launched a BU control project in Songololo Territory in 2004. The principal aims of this project were (i) the improvement of the patient care of BU patients admitted at the GRH IME/Kimpese and (ii) the promotion of early community-based detection of suspected BU cases. The aim of this study is to evaluate the impact of this specialized BU control program on clinical profiles and outcomes.

MATERIALS AND METHODS

Ethics Statement

Ethical clearance for this study was obtained from the Institutional Review Board of IME. All patients, or their guardian in the case of minors, provided informed consent for all diagnostic and treatment procedures and publication of any or all images derived from the management of the patient, including clinical photographs that might reveal patient identity. The BU control project started at the end of 2004 and introduced free patient care for BU patients during their admission at GRH IME/Kimpese, whereas this was hitherto to be paid on a fee-forservice basis. Furthermore, the patients benefited from a free daily nutritional supplement, and specific antibiotherapy was introduced in accordance with WHO recommendations (WHO, 2004), as well as a physiotherapy program for prevention of disabilities. Simultaneously the project organized awareness raising campaigns in the endemic communities, based on a mass-media approach targeting the general public, followed by active case-finding and referral of suspected cases to the specialized BU care centre. The project was based on the following five components: Improving facilities’ management and treatment skills; Prevention of disabilities and physical rehabilitation; Feeding patients and psychological and social support for those affected; Stepping up Information, Education and Communication for the general public and community-based surveillance, and Training and research. To evaluate the effect of this control project,

150 we used a comparative non-randomized study design, comparing patient demographic profiles and clinical outcomes of the group of patients admitted at the GRH IME/Kimpese in the 3 years before the start of the project (2002–2004) with those admitted during the 3 years after the start of the project (2005–2007). We have included all consecutive patients clinically diagnosed as BU and admitted to the Surgical Department of GRH IME/ Kimpese from January 2002 to December 2007. The clinical case definition elaborated by the World Health Organization (WHO) was used to diagnose BU (WHO, 2001). Additionally for the second period, as recommended by the WHO (WHO, 2004), we introduced patients’ categorization as follows: A single lesion <5 cm (Category I); A single lesion 5–15 cm (Category II); A single lesion >15 cm, multiple lesions, lesions at critical sites (face, breast and genitalia) or osteomyelitis (Category III). For all patients included in this study, the diagnostic confirmation process consisted of swabs from ulcerative lesions and biopsies for the laboratory confirmation (bacteriology and/or histopathology) of suspected cases according to WHO recommendations (WHO, 2001). The initial direct smear examinations for acid-fast bacilli and histopathologic analyses were made at the IME/Kimpese laboratory. Other specimens from the same patient were sent in a transport medium to the Mycobacteriology Unit of the Institute of Tropical Medicine (ITM) in Antwerp, Belgium (Eddyani et al., 2008), where Ziehl-Neelsen (ZN) staining, in vitro culture on Löwenstein-Jensen medium, and PCR for the detection of M. ulcerans DNA were performed according to WHO recommendations (WHO, 2001). Formalin-fixed tissues were sent to the Department of Infectious and Parasitic Diseases Pathology of the Armed Forces Institute of Pathology in Washington DC, for the histopathological confirmation of diagnosis (Phanzu et al., 2006). Throughout the whole study period, clinical data of BU patients were recorded on standardized Case Report Forms elaborated by WHO (known as form BU01) and the data were entered in a standardized case registry form (BU02) (WHO, 2000). Next, these data were entered into an Excel database (Microsoft Corporation, Redmond, WA) and analyzed with Epi-

151

Info version 3.3.2 (Centers for Diseases Control and Prevention, Atlanta, GA). The Pearson chi-square test was used to compare proportions with a significance level set at 5%, as well the Fisher’s exact test when an expected cell value was less than 5. To evaluate the relevance and the effect of the BU control project, we used the conceptual framework to evaluate public health programs proposed by Habicht et al. (Habicht et al., 1999). The principal indicators considered for the data analysis are the number of recorded cases for each period, the number of new cases and relapses, the proportion of cases with functional limitation of joints at diagnosis, the proportion of cases confirmed by at least one laboratory test, the proportion of ulcerative forms at diagnosis, the type of treatment applied, the proportion of discharged cases with functional limitation of joints, the median duration of hospitalization, and the case-fatality rate. Relapse was defined in both study periods as a new confirmed diagnosis of BU less than one year after being declared cured from BU after treatment (surgical only in the first period, antibiotic and/or surgical in the second period). Functional limitation was defined as any reduction in the range of motion of one or more joints, and was assessed based on clinical observation. Lesions were considered as mixed forms when simultaneous presence of different forms of disease including bone and joint involvement in the same patient was noticed. Besides, we defined as simple ulcerative forms (SUF) the ulcerative lesions not associated with other clinical lesions such as papule, nodule, plaque, edema or osteomyelitis at the same site.

RESULTS

The number of suspected BU cases admitted at GRH IME/ Kimpese strongly increased after the start of the BU control project. The average number of annual admissions for BU tripled, from 21 cases per year for the period 2002–2004, to 63 cases per year for 2005–2007 (Figure 8). The clinico-epidemiological features and the results of patient management are shown in Tables 3 and 4. The origin of patients remains

152 mainly the Songololo Territory, Cataractes District, where the GRH IME/ Kimpese is located (Figure 9). The median age of patients (20 years) was similar for both periods. The proportion of female patients increased significantly from 30% before to 49% after the project was initiated (p =0.005). In both periods, the majority of BU patients were new cases, yet the proportion of relapse cases amongst all admissions reduced from 32.8% to 11.6% (p<0.001) after 2004. The proportion of ulcerative forms at admission decreased from 95.3% to 85.8% after 2004 (p = 0.041), and the proportion of SUF increased from 32.8% to 60.7% amongst the ulcers (p<0.001) (Figure 10). There was no change in the proportion of confirmed osteomyelitis nor in the proportion of patients presenting with joint movement limitations. The reported median duration of the disease since the appearance of first symptoms increased from 6 to 8 weeks. Globally, the proportion of patients who healed with complications did not change significantly from 23.4% to 19.5% (p = 0.496), even amongst patients declared cured only, from 31.3% to 21.0% (p= 0.136). However, the number of cases that healed without complications increased significantly from 51.6 to 73.2% (Figure 11) (p = 0.001). The proportion of cases confirmed by at least one laboratory test positive for M.ulcerans remained the same (70% in 2002–2004 versus 61% in 2005–2007, p= 0.183). Antibiotic therapy was introduced as part of the control project, and was prescribed to 56.3% of patients, although most patients continued to receive surgery (93.7% previously compared to 84.2% after 2004, p = 0.052). Ninety patients (47.4%) were treated by a combination of antibiotics (rifampicin and streptomycin) and surgery. Seventy patients (36.8%) were treated with surgery alone, seventeen patients (8.9%) only with antibiotics, and thirteen (6.8%) were treated with daily wound dressing. The median duration of hospitalization, around 90 days, was approximately similar during both periods (Table 4) and varied by disease category during the second period, respectively 60 days for category I (Figure 12 and 13), 81 days for category II, and 118 days for category III.

153

The case fatality rate was significantly decreased from 18.7% during the previous period (12 out of 64 patients) to 3.2% (6 out of 190 patients) during the second period (p<0.001). Conditions associated with mortality among BU patients in the previous period were as follows: sepsis in four patients out of twelve (33%), malnutrition and anaemia in nine patients (75%), edematous disseminated disease in two patients (16.6%), postsurgical shock in one patient (8%), and cancerization in two patients (16.6%).

DISCUSSION

The BU control project was associated with a strong increase in the number of admitted BU cases at GRH IME/Kimpese and a fundamental change in the profile of those BU patients. Since the implementation of the control project we observed equal numbers of men and women presenting with BU, significant decrease in the proportion of relapse cases and significant increases in the proportion of early lesions and simple ulcerative forms, and in the proportion of patients healed without complications. Importantly, the case fatality rate decreased significantly from 18.7% to 3.2%. While those parameters indicate a positive impact of the project, we are aware of the limitations of our study. For our evaluation, we used a historical control group: BU patients admitted at the hospital before the project (2002–2004) were compared to those who benefited from the implementation of the control project (2005–2007). Although such before/after evaluation design does not provide conclusive evidence that the observed changes are attributable to the control project itself, it is usually considered sufficient by policy makers to conclude to a beneficial effect (Habicht et al., 1999). The threefold increase in the number of BU cases admitted annually can to a large extent be explained by the active casefinding and the reduction of the financial barrier, as patient care was free after 2004, but is probably also due to the improvement of patient management and the quality of

154 clinical results. While both aspects are likely partially involved in the observed results, the observational study design will not allow us to distinguish between the two. The capacity strengthening of medical staff on the surgical management of BU patients through local and international training, the introduction of specific antibiotherapy (rifampicin and streptomycin), and implementation of a program for the prevention of disabilities have contributed to improvement of clinical outcomes (the increased proportion of patients healed without complications, the reduction of the proportion of relapses, and the reduction of the case fatality rate). Furthermore, we assume that the improved access to adequate and prompt BU treatment in the second period through the free patient care, and the free daily nutritional supplement offered played a major role in the improvement of clinical outcomes during the second period. Indeed, in Africa, the challenge for health care professionals working with BU patients is to break up the cycle of poor clinical outcomes leading to loss of confidence of the affected communities in the hospital (Renzaho et al., 2007). Debacker et al. reported that in the Centre Sanitaire et Nutritionnel Gbemoten (CSNG), Zagnanado, Benin, 68.3% of patients were referred to the hospital by a former BU patient. The improved quality of care at CSNG resulted in a reduction of the median duration of hospitalization from 9 months in 1989 to 1 month in 2001, and the median delay in seeking medical care dropped from 4 months in 1989 to 1 in 2001 (Debacker et al., 2004). The introduction of a BU program was an important factor in the marked reduction in patient delay. Furthermore, after promotional sessions on BU organized in 2000 by the National BU program in the Zou, Oueme, and Atlantique Departments, patients reported earlier than in 1999 (Debacker et al., 2004). We are hopeful that similar results will develop at the Territory of Songololo in Bas-Congo. Awareness raising campaigns followed by active case-finding have contributed to the dissemination of information on BU among the communities in Songololo during the intervention period. We assume that the active case-finding activities have contributed to the change of the

155

Male/Female ratio from 2.4/1 before the project to 1.02/1 during the project period, and thus, the project seems to have contributed to equilibrate the gender balance. During the first period, male BU patients were more frequent probably due to sociocultural barriers for women to seek care, whereas during the second period the active case-finding activities helped the female patients to overcome these barriers. Progressively, more early lesions and more SUF were diagnosed at the hospital. However, rather surprisingly, the median duration since the onset of first symptoms remained high after the project was launched. Reasons why the median delay in seeking medical attention was higher during the second period compared to the first one remain unclear, and need to be assessed. This may explain the fact that the number of confirmed osteomyelitis cases, limitations of joint movement, both at diagnosis as well as at healing, and patients needing surgery, remained similar. This is problematic, as the huge clinical impact of BU is mainly due to the late detection of cases (Sizaire et al., 2006). Indeed, an extended delay before presentation to the hospital has been identified as one of the most important risk factors for bone involvement. Between 1996 and 2007, out of 930 confirmed and treated BU patients at Zagnanado, Benin, 106 (11.4%) presented an osteomyelitis caused by M. ulcerans. The median delay between onset of symptoms and consultation was 167 days for patients with bone involvement and 61 days for those with cutaneous lesions (p<0.001) (Portaels et al., 2008). In most endemic regions, consulting the hospital seems to be the last resort when other attempts were unsuccessful and when the disease has reached an advanced stage with large cutaneous ulcerations or other complications, such as joint contractures or osteomyelitis (Portaels et al., 2008). Stienstra et al. reported in their study on the beliefs and attitudes towards BU in Ghana that in 59% of cases, witchcraft was mentioned as cause of the disease. Among the interviewed patients, 52% applied herbs on their lesions and consulted a hospital as last resort. The reasons evoked were (i) financial difficulties [30% of patients], (ii) the fear of treatment at the hospital and in particular amputation, and (iii) expectations of a spontaneous healing

156

(Stienstra et al., 2002). Recently, a study conducted by Renzaho et al. in Ga West district in Ghana demonstrated that 71.8% of BU patients consulted a traditional practitioner first and that the hospital was consulted as last resort (Renzaho et al., 2007). Meyers and others noted that in the Songololo Territory, DRC, the reasons for which many BU patients delayed seeking medical assistance were obviously complex, but cultural, economic, and transportation factors were especially important (Meyers et al., 1974). Recently, a study conducted in the same area showed that all interviewed patients first adopted a ‘‘wait and see’’ attitude which lasted on average 2 months (Kibadi et al., 2009). Similar observations were reported in other African countries as Cameroun (Noeske et al., 2004), and Benin (Aujoulat et al., 2003; Johnson et al., 2004; Mulder et al., 2008). Those studies were realized when surgery was still the treatment of choice; the recent introduction of specific antibiotherapy as first line treatment may alter this behavior (Johnson et al., 2005; Etuaful et al., 2005; Nienhuis et al., 2010). These social, economical, geographical and cultural reasons, that limit the access to health care in endemic regions, suggest that the number of admitted patients at GRH IME/Kimpese may represent only the emerged part of the iceberg. The free of charge policy offered to patients does not resolve completely the problem of financial barriers related to the patient management of BU. The study conducted by Grietens et al. in two hospitals with a specialized program for BU in Cameroun, similar to ours, has shown that in spite of the reduction of the treatment costs, the hospitalization for BU remains financially and socially untenable for patients and their households, leading to the abandonment of biomedical treatment or a complete refusal (Grietens et al., 2008). Therefore, there is a need to consider new control strategies which are both socially and financially acceptable and appropriate for the concerned communities.

157

CONCLUSION

Overall, the results after 3 years of implementation of BU control activities in Songololo Territory are encouraging. However, the morbidity and disabilities due to BU remain high among our patients. The burden of BU in terms of human suffering, long duration of hospitalization, the development of disabling sequelae, and socio-economic repercussions, is mainly attributable to the late detection of cases. For this reason, secondary prevention through earlier case detection and treatment remains one of the key measures in the control of BU (WHO, 2008). To reduce the burden and to increase the coverage of the population at risk, we consider that a dedicated BU control program at central and provincial level, that operates in close collaboration with the existing polyvalent health services, would be the most efficient way to organize the control of BU in Songololo Territory. The aforesaid program should involve education of the population in the endemic areas, training of healthcare workers, early detection by active case-finding and adequate case management provided free of charge. Further decentralization and integration of BU control activities may improve access to diagnosis and care at the most peripheral level of the health system. A close collaboration between the BU control project and the health zones is essential for the implementation of a simple, functional, and efficient active surveillance system in a resource limited context.

REFERENCES

Andersen FO. Mycobacterial skin ulcers-Clinical experience. Cent Afr J Med 1965; 11: 131–135. Asiedu K, Etuaful S Socioeconomic implications of Buruli ulcer in Ghana: a three-year review. Am J Trop Med Hyg 1998; 59: 1015–22. Aujoulat I, Huguet-Ribas MP, Koïta Y. L’ulcère de Buruli: un problème de santé publique méconnu, appelant une mobilisation internationale. Dév Santé 1996; 125: 22–30. Aujoulat I, Johnson C, Zinsou C, Guedenon A, Portaels F. Psychosocial

158

aspects of health seeking behaviours of patients with Buruli ulcer in southern Benin. Trop Med Int Health 2003; 8: 750–9. Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Guédénon A, Scott JT, Dramaix M, Portaels F. Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, Southern Benin, 1997-2001. Emerg Infect Dis 2004; 10: 1391-8. Eddyani M, Debacker M, Martin A, Aguiar J, Johnson CR, Uwizeye C, Fissette K, Portaels F. Primary culture of Mycobacterium ulcerans from human tissue specimens after storage in semisolid transport medium. J Clin Microbiol 2008; 46: 69-72. Etuaful S, Carbonnelle B, Grosset J, Lucas S, Horsfield C, et al. Efficacy of the combination rifampin-streptomycin in preventing growth of Mycobacterium ulcerans in early lesions of Buruli ulcer in humans. Antimicrob Agents Chemother 2005; 49: 3182–86. Grietens KP, Boock AU, Peeters H, Hausmann-Muela S, Toomer E, et al. ‘‘It is me who endures but my family that suffers’’: social isolation as a consequence of the household cost burden of Buruli ulcer free of charge hospital treatment. PloS Negl Trop Dis 2008; 2: e321. Habicht JP, Victoria CG, Vaughan JP. Evaluation designs for adequacy, plausibility and probability of public health programme performance and impact. Int J Epid 1999; 28: 10–8. Hotez PJ, Molyneux DH, Fenwick A, Kumaresan J, Sachs SE, et al. Control of Neglected Tropical Diseases. N Engl J Med 2007; 357: 1018–27. Johnson PD, Stinear T, Small PL, Pluschke G, Merritt RW, Portaels F, Huygen K, Hayman JA, Asiedu K. Buruli ulcer (Mycobacterium ulcerans infection): new insights, new hope for disease control. PLoS Med 2005a; 2(4):e108. Johnson RC, Makoudoté M, Hougnihin R, Guédénon A, Ifebe D, Boko M, Portaels F. Le traitement traditionnel de l'ulcère de Buruli au Bénin. Med Trop (Mars) 2004; 64: 145-50. Kibadi K, Boelaert M, Kayinua M, Minuku JB, Muyembe-Tamfum JJ, et al. Therapeutic itineraries of patients with ulcerated forms of Mycobacterium ulcerans (Buruli ulcer) disease in a rural health

159

zone in the Democratic Republic of Congo. Trop Med Int Health 2009; 14: 1110–6. Kibadi K, Tsakala TM, Mputu-Yamba JB, Muyembe T, Kashongwe M, Imposo B, Nsiala A. L’ulcère de Buruli chez les réfugiés Angolais des sites de Kimpese, Bas-Congo. Cahiers Santé 2003; 13: 39-41. Meyers WM, Connor DH, McCullough B, Bourland J, Moris R, Proos L. Distribution of Mycobacterium ulcerans infections in Zaire, including the report of new foci. Ann Soc Belg Med Trop 1974; 54: 147-57. Mulder AA, Boerma RP, Barogui Y, Zinsou C, Johnson RC, et al. Healthcare seeking behaviour for Buruli ulcer in Benin: a model to capture therapy choice of patients and healthy community members. Trans R Soc Trop Med Hyg 2008; 102: 912–20. Nienhuis WA, Stienstra Y, Thompson WA, Awuah PC, Abass KM, et al. Antimicrobial treatment for early, limited Mycobacterium ulcerans infection: a randomised controlled trial. Lancet 2010; 375: 664–72. Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, et al. Buruli ulcer disease in Cameroon rediscovered. Am J Trop Med Hyg 2004; 70: 520–6. Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, Nsiangana SZ, Singa JN, Meyers WM, Suykerbuyk P, Portaels F. Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg 2006; 75: 311-4. Portaels F, Johnson RC, Aguiar J, Meyers WM, Debacker M. Etude de 106 cas d’ulcères de Buruli avec atteintes osseuses traité à Zagnanado, Bénin. Bull ALLF 2008; 23: 48–50. Portaels F, Silva MT, Meyers WM. Buruli ulcer. Clin Dermatol 2008; 27: 291-305. Portaels F. Epidémiologie des ulcères à Mycobacterium ulcerans. Ann Soc Belg Med Trop 1989; 69: 91-103. Renzaho AM, Woods PV, Ackumey MM, Harvey SK, Kotin J. Community- based study on knowledge, attitude and practice on the mode of transmission, prevention and treatment of the Buruli ulcer in Ga West District, Ghana. Trop Med Int Health 2007; 12: 445–58.

160

Sizaire V, Nackers F, Comte E, Portaels F. Mycobacterium ulcerans infection: control, diagnosis, and treatment. Lancet Infect Dis 2006; 6: 288-96. Smith JH. Epidemiologic observations on cases of Buruli ulcer seen in a hospital in the Lower Congo. Am J Trop Med Hyg 1970; 19: 657– 63. Stienstra Y, van der Graaf WT, Asamoa K, van der Werf TS. Beliefs and attitudes toward Buruli ulcer in Ghana. Am J Trop Med Hyg 2002; 67: 207-13. World Health Organization. Buruli Ulcer. Diagnosis of Mycobacterium ulcerans Disease. Portaels F, Johnson P, Meyers WM, eds. Geneva: World Health Organization. 2001. WHO/CDS/GBUI/2001.4. World Health Organization. Buruli ulcer. Mycobacterium ulcerans infection. WHO/CDS/CPE/GBUI/2000/1. Asiedu K, Scherpbier R, Raviglione M, editors. Geneva: The Organization; 2000. p.118. World Health Organization. Buruli ulcer: progress report, 2004-2008. Wkly Epidemiol Rec 2008; 17: 145-54. World Health Organization. Provisional Guidance on the Role of Specific Antibiotics in the Management of Mycobacterium ulcerans Disease (Buruli ulcer). Geneva: World Health Organization 2004; WHO/CDS/CPE/GBUI/2004.10.

161

Table 1. The clinic-epidemiological features of BU patients at admission in IME/Kimpese Hospital.

162

Table 2. Results of the management of BU patients in IME/Kimpese Hospital

163

Figure 1. Evolution of number of annual admissions of BU cases to the GRH IME/Kimpese from 2002 to 2007. doi:10.1371/journal.pntd.0001402.g001

Figure 2. Origin of BU patients admitted in IME/Kimpese Hospital, 2002- 2007.doi:10.1371/journal.pntd.0001402.g002

164

Figure 3. A simple ulcerated form of disease on the right arm.doi:10.1371/journal.pntd.0001402.g003

Figure 4. Healing without complications after antibiotherapy combined with surgery.doi:10.1371/journal.pntd.0001402.g004

165

Figure 5. Single ulcerative lesion ,5 cm diameter (confirmed by IS2404-PCR). doi:10.1371/journal.pntd.0001402.g005

Figure 6. Healed lesion without complication after antibiotherapy alone without surgery.doi:10.1371/journal.pntd.0001402.g006

166

IV.2.2. Burden of Mycobacterium ulcerans disease (Buruli ulcer) and underreporting ratio in the Territory of Songololo, Democratic Republic of Congo

Delphin Mavinga Phanzu 1, 7Ω, Patrick Suykerbuyk2Ω, Paul Saunderson3, Philippe Ngwala Lukanu4, Jean-Bedel Masamba Minuku5, Désiré Bofunga B. Imposo1, Blanchard Mbadu Diengidi1, Makanzu Kayinua5, Jean-Jacques Tamfum Muyembe6, Pascal Tshindele Lutumba6, Bouke C. de Jong2, Françoise Portaels2, Marleen Boelaert7

1 General Reference Hospital of Kimpese, Institut Médical Evangélique, Kimpese, Bas- Congo, Democratic Republic of Congo 2 Department of Biomedical Sciences, Unit of Mycobacteriology, Institute of Tropical Medicine, Antwerp, Belgium 3 American Leprosy Missions, Greenville, South Carolina, United States of America 4 Central Office of the Rural Health Zone of Kimpese, Bas-Congo, Democratic Republic of Congo 5 Central Office of the Rural Health Zone of Nsona Mpangu, Bas-Congo, Democratic Republic of Congo 6 Institut National de Recherche Biomédicale, Kinshasa, Democratic Republic of Congo 7 Department of Public Health, Unit of Epidemiology and Disease Control, Institute of Tropical Medicine, Antwerp, Belgium Ω These authors contributed equally to this work

Personal contribution: Conceived and designed the experiments: DMP; capacity strengthening: DMP; performed the experiments: DMP; analyzed the data: DMP; wrote the paper: DMP.

PLoS Negl Trop Dis. 2013 Dec; 7(12): e2563

167

Abstract

Background: Cutaneous infection by Mycobacterium ulcerans, also known as Buruli ulcer (BU), represents the third most common mycobacterial disease in the world after tuberculosis and leprosy. Data on the burden of BU disease in the Democratic Republic of Congo are scanty. This study aimed to estimate the prevalence rate and the distribution of BU in the Songololo Territory, and to assess the coverage of the existing hospital-based reporting system. Methods: We conducted a cross-sectional survey (July-August 2008) using the door-to-door method simultaneously in the two rural health zones (RHZ) of the Songololo Territory (RHZ of Kimpese and Nsona- Mpangu), each containing twenty health areas. Cases were defined clinically as active BU and inactive BU in accordance with WHO-case definitions. Results: We detected 775 BU patients (259 active and 516 inactive) in a total population of 237,418 inhabitants. The overall prevalence of BU in Songololo Territory was 3.3/1000 inhabitants, varying from 0 to 27.5/1000 between health areas. Of the 259 patients with active BU, 18 (7%) had been reported in the hospital-based reporting system at Kimpese in the 6-8 months prior to the survey. Conclusion: The survey demonstrated a huge variation of prevalence between health areas in Songololo Territory and gross underreporting of BU cases in the hospital-based reporting system. Data obtained may contribute to better targeted and improved BU control interventions, and serve as a baseline for future assessments of the control program. INTRODUCTION Cutaneous infection by Mycobacterium ulcerans, also known as Buruli ulcer (BU), represents the third most common mycobacterial disease in the world after tuberculosis and leprosy (1). In Africa, children under 15 years old have the highest incidence, but healthy persons of all ages, races, and socioeconomic classes are susceptible (2,3). Rates of infection among males and females are equal (3). BU most affects the extremities

168

(2,4), and is diagnosed in the majority of patients at the ulcerative stage (5). The disease has a scattered focal distribution within endemic regions, which impedes accurate estimation of disease burden (5,6). BU is considered as one of the Neglected Tropical Diseases (NTDs) with a poorly known global prevalence (7), and mainly affects remote rural African communities (8). A recent review on prevalence (9) reported that, of the estimated 7,000 cases of BU reported annually worldwide, more than 4,000 cases occur in Sub-Saharan Africa. The largest numbers of reported BU cases are from the West African countries of Côte d’Ivoire (about 2,000 cases annually), Benin and Ghana, each reporting about 1,000 cases a year (3). Various prevalence rates (Table 1) have been reported from different endemic regions in Sub-Saharan Africa (6,10-13). In the Democratic Republic of Congo (DRC), more than 500 BU cases had been reported before 1980 (14). The first BU case reports in the Province of Bas-Congo were published in the 1960s and 1970s (15-17). However, in-depth interviews of former patients conducted in the Bas-Congo by Meyers et al. strongly supported the concept that BU was an ancient disease in that region (14). After 1980, there was a silent period of 20 years without any cases reported in the scientific literature (14). A national hospital-based survey conducted in 2004 identified 487 clinically suspected cases of BU from six provinces (18). Between 2002-2004, an apparent resurgence of BU was reported in Songololo Territory (4), known to be the main focus of BU in the country (17). Since the end of 2004, the General Reference Hospital (GRH) of the Institut Médical Evangélique (IME) of Kimpese launched a specialized BU program sponsored by American Leprosy Missions, offering in-patient treatment free-of-charge and supplementary aid. A recent study has shown a strong increase in the number of admitted BU cases at the IME Hospital after the start of the BU Control Project (19). Although the number of BU cases admitted in the hospital was rising, data on the exact prevalence and the extent of the disease in the region was lacking. We set up a study to obtain relevant information for planning subsequent control activities, and to provide baseline data for future control program assessments. This study aimed

169

(i) to assess the prevalence and the geographic distribution of BU, (ii) to determine the epidemiologic characteristics of BU, and (iii) to determine the project coverage in Songololo Territory, the target endemic region of the project.

METHODS Ethics Statement The Congolese Ministry of Health granted approval to conduct the survey. We obtained ethical clearance for this study from the Institutional Review Board of IME (N° IME/CS/01/2008). All patients, or their guardian in the case of minors, provided written informed consent for all diagnostic and treatment procedures and publication of any or all images derived from the management of the patient, including clinical photographs that might reveal patient identity. After informed consent had been given, data were recorded on a Community BU Form recommended by WHO. Patient care was free of charge.

Survey zone The case search covered two rural health zones (RHZ), Kimpese and Nsona-Mpangu, both located in Songololo Territory (Figure 1), one of ten territories of Bas-Congo Province. It is situated in the District of Cataractes and covers an area of 8,190 Km2, approximately 15.2% of the total surface of the province, with a population of 237,418 inhabitants in 2008 (enumeration conducted on December 2007 by the Central Offices of the 2 RHZ). An average of 6 persons per household was used as a regional estimate, giving a total of 39,569 households to be visited by 80 community health workers (CHW). Songololo Territory is limited in the north by the Congo River, in the west by Sekebanza Territory, in the east by Mbanza-Ngungu Territory and in the south by the northern border of Angola. Each RHZ is subdivided into 20 health areas (Table S1 & Table S2). The primary level of health care facilities includes the Rural Health Posts (HP), Health Centres (HC) and

170

Reference Health Centres (RHC), and the secondary level is represented by the GRH.

Patients, data collection and analysis We conducted a cross-sectional survey (July-August 2008) using the door- to-door method simultaneously in the two RHZ of the Songololo Territory (i.e., Kimpese and Nsona-Mpangu), each containing twenty health areas. Cases were defined clinically as active BU and inactive (healed) BU in accordance with WHO-case definitions (20). We defined functional limitation as any reduction in the range of motion of one or more joints, and assessed it by clinical observation. Lesions were considered as mixed forms when the simultaneous presence of different forms of disease, including bone and joint involvement, in the same patient was noted. In addition, we defined as simple ulcerative forms (SUF) the ulcerative lesions not associated with other clinical lesions such as papule, nodule, plaque, edema or osteomyelitis at the same site. Lesions were categorized as follows: A single lesion < 5 cm (Category I); a single lesion 5-15 cm (Category II); a single lesion > 15 cm, multiple lesions, and lesions at critical sites (face, breast and genitalia) or osteomyelitis (Category III). The status of relapse was assessed by questioning the patients, or their guardian in the case of minors, on the history of the lesion, and defined as the reappearance of an ulcer or another form of the disease at the original site of the lesion or elsewhere during the 12 months that followed the end of the previous treatment (antibiotics and/or surgery). This study was conducted in two phases: a preparatory phase and an investigation phase. During the preparatory four-week phase (June 2008), the purpose of the study was explained to the local political and health authorities, and their approval was obtained. Then, 80 CHW, i.e., 40 per RHZ, were trained in the use of the survey tools (BU community form, pictorial document to recognize BU) and in the identification of suspected BU cases in their communities. We also trained six physicians (working in the RHC of both RHZ), two nurse-supervisors of the leprosy and tuberculosis program (LT), and 40 head nurses (in charge of peripheral

171 health areas), in active case-finding of BU cases and in the use of the survey tools. For the survey, each RHZ was provided with 1 motor bike, 1 Global Position System device, 4 digital photo cameras, 30 bicycles (at least 1 for each health area), 25 megaphones (at least 1 for each health area), drugs and required medical and laboratory consumables. The investigation phase was divided in two periods. The first period (two to three weeks depending on health area) consisted of making an inventory of all BU-like cases by the CHW, using the door-to-door approach in all villages and in each section of two cities in Songololo Territory (Songololo city and Kimpese city). The recommendation to CHW was to visit 40 households per day. A pictorial document, showing different clinical manifestations of BU, was presented to the head of the household or his/her representative asking if any household members presented similar lesions. If the head of the household was not present, the household was revisited once. The second period (6 weeks) included the clinical validation of suspected BU cases by trained health professionals. The eight validation teams were each composed of two people: firstly, a team member of the BU Project (physician or nurse), or another physician, or a LT supervisor, and secondly one of the head nurses. The diagnostic confirmation process of suspected cases involved the collection of swabs from ulcerative lesions and fine needle aspirates from non-ulcerative lesions, followed by laboratory analyses (bacteriology and/or molecular biology) according to WHO recommendations (20). The initial direct smear examinations for acid-fast bacilli were made at the IME/Kimpese laboratory, followed by in vitro culture for M. ulcerans. Samples were sent in tubes to the “Institut National de Recherche Biomédicale” in Kinshasa, DRC, where PCR for the detection of M. ulcerans DNA was performed, according to WHO recommendations (20). The external quality control was conducted by the Unit of Mycobacteriology of the Institute of Tropical Medicine in Antwerp, Belgium.

172

The study was carried out simultaneously in the different health areas of both RHZ. Data were recorded on a standardized Case Registry Form elaborated by WHO (BU02), entered into an Excel database (Microsoft Corporation, Redmond, WA) and analyzed with Epi-Info version 3.3.2 (Centers for Diseases Control and Prevention, Atlanta, GA). The Pearson chi-square test was used to compare proportions with a significance level set at 5%, and the Fisher’s exact test when an expected cell value was less than 5. Coverage was calculated as the number of active cases detected who had visited the BU reference center in IME Hospital. We produced the distribution maps of BU in Songololo Territory using ArcGIS 9.2 (ESRI, Redlands, CA, USA).

RESULTS The CHW visited a total of 39,044 households distributed across 9 sections of two cities (Kimpese and Songololo), 46 hamlets and camps, and 547 villages of the Songololo Territory. The estimated coverage of the study was 98.6%. During the household visits, the CHW inventoried 2,516 persons with BU-like lesions, among which 775 (30.8%) were validated in a second step as probable cases of BU, all forms included (i.e., 259 with active and 516 with inactive lesions). A total of 72 out of 241 (30%) patients with active lesions in whom a sample could be taken were confirmed by at least one positive laboratory test for M. ulcerans. The overall prevalence of BU (active and inactive) in Songololo Territory was 3.3/1000 inhabitants, varying from 0 to 27.5/1000 between health areas, while the prevalence of active BU was 1.1/1000 inhabitants with the minimum of 0.3/1000 when only active, laboratory confirmed BU, was considered. Table 2 shows the prevalence of different BU forms in both RHZ of Songololo Territory, and the distribution per health area is presented in Figures 2, 3, S1 and S2. The overall prevalence for the RHZ of Kimpese was 2.6 per 1000 inhabitants and could vary between health areas from 0.1 (Kimbanguiste) to 24.4 (Mukimbungu). The prevalence of BU active forms was 1 per 1000 inhabitants, varying between health areas from 0.1 (Kimbanguiste) to 5.7 (Mukimbungu). The health areas of

173

Mukimbungu and Kasi, located in the North of the RHZ of Kimpese, are the most endemic, representing together 60% of the identified patients during the survey (Table S1). Sixty percent of the identified patients in the RHZ of Nsona-Mpangu were from 3 health areas, Kisonga, Nkamuna, and Songololo (Table S2). The overall prevalence in this RHZ was 4.4 per 1000 inhabitants, varying from 0 (health areas Nduizi, Nkenge and Pala Bala) to 27.5 (Kisonga). The prevalence of active forms of BU was 1.3 per 1000 inhabitants, varying between health areas from 0 (Nduizi, Nkenge and Pala Bala) to 3.8 (Kisonga).

The age distribution of all cases ranged from 2 to 94 years (Median 27, Interquartile range (IQR) 14-44) with no significant differences between active and inactive cases. The supplementary tables provide the detailed information. We observed a predominance of female gender (60%) among the recorded cases. Among the 259 patients with active lesions, no sex difference was observed, as 130 (50.2%) were female. The proportion of new cases was far higher (94%) than the relapses. The ages ranged from 2 to 94 years (Median 27 years; IQR 11-47 years), and the distributions in the two RHZ were similar. Among these 259 patients, 192 (74%) had ulcerative lesions and 62 (23.9%) were diagnosed with functional joint limitations. Lesions on the limbs were predominant, representing 90% of the sites of lesions. Regarding the patients’ categorization, 48.8% were in category I, 31.5% category II, and 19.7% category III. The proportion of patients with ulcerative lesions was higher (p<0.001) in the RHZ of Kimpese (83%) compared to the RHZ Nsona-Mpangu (63.6%). Less than half of the patients of the RHZ of Kimpese (41.2%) and more than half (57.6%) in the RHZ of Nsona-Mpangu were in category I (p=0.031) (Table S3). Female patients predominated amongst active confirmed cases compared to unconfirmed cases; on the other hand, male patients were more frequent in active unconfirmed patients (p=0.029). No differences in the

174 age distribution were observed between active confirmed and unconfirmed patients. The lower limb locations were significantly more frequent amongst active unconfirmed patients (p<0.001). Upper limb sites predominated (p<0.001) amongst active confirmed patients (Table 3). Features of active cases in the two RHZ were quite similar, with a few exceptions. The ulcerated forms (p<0.001) and functional limitations on diagnosis (p<0.001) predominated in the RHZ of Kimpese. Features of inactive cases in the two RHZ were similar but functional limitations were more often observed in the RHZ of Kimpese (p=0.005) (Table S4).

Only 25 BU patients were admitted and notified at the General Hospital IME/Kimpese between January and August 2008, amongst which 18 were still under treatment for active BU during the survey. Thus, 93% of all active BU patients at the time of the community survey were not captured by the hospital-based reporting system, corresponding to a ratio of 1 reported case for approximately 13 unreported cases.

DISCUSSION The present study is the first exhaustive population-based survey in DRC aiming to assess the prevalence and distribution of BU in a well- circumscribed endemic region. The survey demonstrated a huge variation in prevalence between health areas and gross underreporting of BU cases in Songololo Territory, compared with the ongoing hospital-based reporting system. Case-definition during the survey was essentially clinical. Case validation was performed by physicians from the BU project and physicians working in the area, well-trained in BU diagnosis, assisted by either a nurse from the BU project or a LT-supervisor, with the nurse responsible for the health area. We are aware of the limitations of clinical diagnosis, which is dependent on the range of experience of health professionals. This may account for certain non-BU cases included in this study. In endemic regions, depending on the clinical stage of the disease, BU may be confused with many other conditions such as nodular onchocerciasis, cyst,

175 lipoma, lymphadenitis, phagedenic tropical ulcer , pyomyositis , necrotizing fasciitis (20,21), to name a few. Our study showed that 72 out of the 241 (30%) patients who were tested, were confirmed in the laboratory. The low confirmation rate is mostly due to the relatively high number (almost half) of the ulcers being in an advanced stage of healing. Likewise, the technical problems encountered by peripheral health professionals when sampling non-ulcerated lesions and wounds, where mixtures of traditional herbs had been applied, may have played a role. Nevertheless, lesions due to another etiology misclassified as BU cannot be excluded, as lower limb locations were significantly more frequent among active unconfirmed patients. Indeed, among 92 clinically suspected patients recruited from the RHZ of Nsona Mpangu, Kibadi et al. found 31 (33.7%) PCR negative patients and among them, 25 with histopathological features not compatible with BU (chronic inflammation and bacterial infections due to gram positive cocci) (22). Despite these limitations, we suggest that our results reflect the endemicity of BU in Songololo Territory reasonably well. In fact, the areas previously established as most endemic were corroborated through this survey, as were the non- or hypoendemic areas (15-17,4). When considering only active lesions, no sex difference was observed, similar to findings in other studies (2,11,12, ,23,24), although our study showed a predominance of females among all cases detected (active and inactive), because among inactive cases, 64.9% (335/516) were females and only 35.1% (181/516) were males. Females predominated also among active confirmed BU cases. This preponderance may be due to time itself, or the fact that the population was predominantly female. When referring to the national census figures (July 2008 estimates), for a total population of 66,514,504 inhabitants, 50.3% were female and 49.7% male. Among the 259 patients with active lesions, the majority (66%) were over age 15, similar to previous findings in the same area (19). Ages observed in this survey were higher than found in other disease-endemic countries (2,10,12,25). The median ages for both RHZ were similar with the median

176 age of 25 years found in Ghana (11), and relatively high when compared to the 15.5 years observed in Cameroon (13). The predominant clinical presentation was an ulcerative lesion in 192 cases (74%). This is consistent with studies in Côte d’Ivoire (10) and Cameroon (12,13), while the percentage of ulcerative lesions was lower in some other studies, for example, 48.5% in Ghana (11), approximately 50% in Benin from 1997 to 2001 (26) and 57.5% in 2004 in the same country (6). Of the 259 active cases, 62 (23.9%) were diagnosed with joint functional limitations, similar to previous findings in the same area (19), and in other African endemic regions (6,12). The general finding of limbs being most affected was confirmed in this study (2,11-13,23,24). The results presented in Table 3 shows that nearly 50% of the BU patients had category I lesions. A similar observation was made in the District of Akonolinga, Cameroon (13). Ambulatory treatment, based on antibiotic therapy in the primary health care facility, is indicated for this category of patients. Indeed, most category I and some category II lesions may heal completely with antibiotic treatment alone (3,27). The introduction of antibiotic therapy (28) has shifted the balance between surgical treatment, mainly limited to reference centers, and antibiotics administered at the most peripheral level of the health system (3). The clinical presentation of BU was different in the two health zones (Table S3). The degree of functional limitation was significantly higher in patients in Kimpese and they had more often ulcerated lesions. We speculate that this difference is most likely due to differences in health seeking behavior, with higher patient delays in Kimpese, notwithstanding the fact that they were living at shorter distance from the IME hospital. In recent years, an influential religious sect has been a factor in the reluctance to seek medical care in the Kimpese area. Although the number of BU patients admitted at the hospital has increased in recent years, the survey results have demonstrated that the coverage of the population at risk was still insufficient. Of the 259 patients with active BU, 18 (7%) had been reported in the hospital-based reporting system. Porten et al. reported a coverage of 16%, limited to the area

177 close to the Akonolinga hospital in Cameroon, where Médecins Sans Frontières (MSF) opened a BU programme in 2002. The need for improved access to care in the high prevalence areas was emphasized (13). In the same area, Grietens et al. found that despite the significant reduction in costs for medical care, hospital treatment for BU often remained financially and socially unaffordable for patients and their households, leading to the abandonment of biomedical treatment, or avoiding it altogether. They concluded in their study that from a socio-economic perspective, a decentralized treatment system may limit the impoverishment of households caused by a long hospitalization period (29). We agree with this opinion because bringing treatment as close as possible to the communities will have a significant mitigating impact on the socio-economic repercussions of BU. The survey demonstrated large variations in prevalence between health areas within an endemic health zone consistent with previous studies in other African BU-endemic regions (6,12,13). Tables S1 and S2 show that in both RHZ, 60% of patients were respectively identified from 2 out of 20 health areas (Mukimbungu, Kasi) in the RHZ of Kimpese and 3 out of 20 health areas (Kisonga, Nkamuna, Songololo) in the RHZ of Nsona-Mpangu. Therefore, priority in case detection should be given to the most endemic health areas. A close collaboration with the provincial Leprosy & Tuberculosis control officers may facilitate the integration of BU activities at the primary health care centers. In fact, the use of the same case- confirmation network or the organization of integrated supervisions would help to reduce the BU intervention costs.

Data obtained in this survey may contribute to better targeted and improved BU control interventions, and serve as a baseline for future assessments of the control program.

178

ACKNOWLEDGEMENTS We are grateful to all participants in this study, the staff of the IME/Kimpese Hospital, community health workers and health professionals in the health zones of Kimpese and Nsona-Mpangu, as well as the staff of the Mycobacteriology Unit and the Unit of Epidemiology and Disease Control of the ITM/Antwerp. We particularly thank Dr WM Meyers for his unconditional support for all projects on Buruli ulcer in the Lower Congo for more than 40 years. Address for correspondence: Marleen Boelaert, Department of Public Health, Unit of Epidemiology and Disease Control, Institute of Tropical Medicine, Antwerp, Belgium.

REFERENCES 1. Portaels F, Silva MT, Meyers WM (2009) Buruli ulcer. Clin Dermatol 27: 291-305.

2. Debacker M, Aguiar J, Stenou C, Zinsou C, Meyers WM, et al. (2004) Mycobacterium ulcerans disease: role of age and gender in incidence and morbidity. Trop Med Int Health 9: 1297–1304.

3. WHO (2008) Buruli ulcer: progress report, 2004-2008. Wkly Epidemiol Rec 83: 145-154. 4. Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, et al. (2006) Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg 75: 311-314.

5. Van der Werf TS, Stienstra Y, Johnson RC, Phillips R, Adjei O, et al. (2005) Mycobacterium ulcerans disease. Bull of the WHO 83: 785- 791.

6. Johnson RC, Sopoh GE, Boko M, Zinsou C, Gbovi J, et al (2005) Distribution of Mycobacterium ulcerans Infection (Buruli Ulcer) in the District of Lalo in Benin. Trop Med Int Health 10:863-871.

179

7. Hotez PJ, Molyneux DH, Fenwick A, Kumaresan J, Sachs SE, et al. (2007) Control of Neglected Tropical Diseases. N Engl J Med 357: 1018-1027.

8. Sizaire V, Nackers F, Compte E, Portaels F (2006) Mycobacterium ulcerans infection: Control, diagnosis, and treatment. Lancet Infect Dis 6:288-296.

9. Hotez PJ, Kamath A (2009) Neglected Tropical Diseases in Sub- Saharan Africa: Review of Their Prevalence, Distribution, and Disease Burden. PLoS Negl Trop Dis 3(8): e412. Doi:10.1371/journal.pntd.0000412.

10. Kanga JM, Kacou ED (2001) Epidemiological aspects of Buruli ulcer in Côte d’Ivoire: results of a national survey. Bull Soc Pathol Exot 94:46-51.

11. Amofah G, Bonsu F, Tetteh C, Okrah J, Asamoa K, et al. (2002) Buruli ulcer in Ghana: Results of a National Case Search. Emerg Infect Dis 8:167-170.

12. Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, et al. (2004) Buruli ulcer disease in Cameroon rediscovered. Am J Trop Med Hyg 70: 520- 526.

13. Porten K, Sailor K, Comte E, Njikap A, Sobry A, et al. (2009) Prevalence of Buruli Ulcer in Akonolinga Health District, Cameroon: Results of a Cross Sectional Survey. PLoS Negl Trop Dis 3(6): e466. doi:10.1371/journal.pntd.0000466.

14. Janssens P, Pattyn S, Meyers WM, Portaels F (2005) Buruli ulcer: An Historical Overview with Updating to 2005. Bull Séanc Acad R Sci Outre-Mer 51:265-299.

15. Andersen FO (1965) Mycobacterial skin ulcers-Clinical experience. Cent Afr J Med 11: 131-135.

180

16. Smith JH (1970) Epidemiologic observations on cases of Buruli ulcer seen in a hospital in the Lower Congo. Am J Trop Med Hyg 19: 657- 663.

17. Meyers WM, Connor DH, McCullough B, Bourland J, Moris R, et al. (1974) Distribution of Mycobacterium ulcerans infection in Zaïre, including the report of new foci. Ann Soc Belge Méd Trop 54: 147- 157.

18. Kibadi AK, Singa JN, Wembanyama H, Portaels F (2006) Résultats de l’enquête nationale préliminaire sur l’ulcère de Buruli en République Démocratique du Congo. Bull de l’ALLF 18 :24-26.

19. Phanzu DM, Suykerbuyk P, Imposo DBB, Lukanu PN, Minuku J-BM, et al. (2011) Effect of a Control Project on Clinical Profiles and Outcomes in Buruli Ulcer: A Before/After Study in Bas-Congo, Democratic Republic of Congo. PLoS Negl Trop Dis 5(12): e1402. doi:10.1371/journal.pntd.0001402.

20. World Health Organization, 2001 (2001) Buruli ulcer-Diagnosis of Mycobacterium ulcerans disease. Portaels F, Johnson P, Meyers WM, eds. WHO/CDS/CPE/GBUI/2001.4, World Health Organization, Geneva.

21. Phanzu MD, Bafende AE, Imposo BBD, Meyers WM, Portaels F (2010) Short Report: Undertreated Necrotizing Fasciitis Masquerading as Ulcerated Edematous Mycobacterium ulcerans Infection (Buruli ulcer). Am J Trop Med Hyg 82:478-481.

22. Kibadi K, Boelaert M, Fraga AG, Kayinua M, Longatto-Filho A, et al. (2010) Response to Treatment in a Prospective Cohort of Patients with Large Ulcerated Lesions Suspected to Be Buruli Ulcer (Mycobacterium ulcerans Disease). PLoS Negl Trop Dis 4(7): e736. doi:10.1371/journal.pntd.0000736.

181

23. Landier J, Boisier P, Fotso Piam F, Noumen-Djeunga B, Sime´ J, et al. (2011) Adequate Wound Care and Use of Bed Nets as Protective Factors against Buruli Ulcer: Results from a Case Control Study in Cameroon. PLoS Negl Trop Dis 5(11): e1392. doi:10.1371/journal.pntd.0001392.

24. Adu EJK, Ampadu E, Acheampong D. (2011) Surgical management of Buruli ulcer disease: a four-year experience from four endemic districts in Ghana. Ghana Med J 45:4-9.

25. Sopoh GE, Barogui YT, Johnson RC, Dossou AD, Makoutode´ M, et al. (2010) Family Relationship, Water Contact and Occurrence of Buruli Ulcer in Benin. PLoS Negl Trop Dis 4(7): e746. doi:10.1371/journal.pntd.0000746.

26. Debacker M, Aguiar J, Stenou C, Zinsou C, Meyers WM, et al. (2004) Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, southern Benin, 1997–2001. Emerg Infect Dis 10: 1391–1398.

27. Nienhuis WA, Stienstra Y, Thompson WA, Awuah PC, Abass KM, et al. (2010) Antimicrobial treatment for early, limited Mycobacterium ulcerans infection: a randomised controlled trial. Lancet 375: 664-72.

28. World Health Organization, 2004 (2004) Provisional guidance on the role of specific antibiotics in the management of Mycobacterium ulcerans disease (Buruli ulcer), WHO/CDS/CPE/GBUI/2004.10, World Health Organization, Geneva.

29. Peeters Grietens K, Um Boock A, Peeters H, Hausmann-Muela S, Toomer E, et al. (2008) “It is Me Who Endures but My family That Suffers”: Social Isolation as a Consequence of the Household Cost Burden of Buruli Ulcer Free of Charge Hospital Treatment. PloS Negl Trop Dis 2(10): e321. doi:10.1371/journal.pntd.0000321.

182

Table 1. Prevalence of Buruli ulcer disease in Africa.

183

Table 2. Prevalence of Buruli ulcer in the Territory of Songololo, July-August 2008

184

Table 3. Clinico-epidemiological features of active BU cases in the Territory of Songololo, July-August 2008.

185

Table S1 Distribution of active and inactive BU cases in the Rural Health Zone of Kimpese (July-August 2008).

Active cases Inactive cases Total Pop. Prev. Prev. Prev. Health areas 2008 Number (x1000) Number (x1000) Number (x1000) BEMBA 3220 2 0,6 2 0,6 4 1,2 CBCO 13046 3 0,2 0 0,0 3 0,2 CECO 9948 3 0,3 0 0,0 3 0,3 KASI 11413 25 2,2 77 6,7 102 8,9 KIASUNGUA 6349 1 0,2 1 0,2 2 0,3 KIKEBA 4872 5 1,0 1 0,2 6 1,2 KILUEKA 5382 2 0,4 8 1,5 10 1,9 KIMBALA 11478 18 1,6 1 0,1 19 1,7 KIMBANGUISTE 16180 2 0,1 0 0,0 2 0,1 LOVO 4893 7 1,4 8 1,6 15 3,1 LUKUNGA 5201 6 1,2 3 0,6 9 1,7 MALANGA 4170 3 0,7 0 0,0 3 0,7 MBANZA NSANDA 4381 5 1,1 1 0,2 6 1,4 MUKIMBUNGU 5070 29 5,7 95 18,7 124 24,5 NKUANZA 9628 7 0,7 31 3,2 38 3,9 SONGA LUMUENO 6919 2 0,3 0 0,0 2 0,3 VIAZA 3520 7 2,0 5 1,4 12 3,4 VILA 5309 4 0,8 1 0,2 5 0,9 VUNDA NSOLE 3336 2 0,6 1 0,3 3 0,9 YANGA DIA SONGA 11793 8 0,7 0 0,0 8 0,7 Total 146108 141 1,0 235 1,6 376 2,6

186

Table S2 Distribution of active and inactive BU cases in the Rural Health Zone of Nsona Mpangu (July-August 2008).

Active cases Inactive cases Total Pop. Prev. Prev. Prev. Health areas 2008 Number (x1000) Number (x1000) Number (x1000) KINGANGA 4714 1 0,2 3 0,6 4 0,8 KINZOLANI 1991 1 0,5 9 4,5 10 5,0 KISONGA 3380 13 3,8 80 23,7 93 27,5 KIZULU 5425 2 0,4 1 0,2 3 0,6 LOMBE 3829 6 1,6 3 0,8 9 2,4 LUANIKA 3062 2 0,7 1 0,3 3 1,0 LUFU 4728 14 3,0 17 3,6 31 6,6 MANTEKE 3355 1 0,3 2 0,6 3 0,9 MAYANGA 5667 4 0,7 9 1,6 13 2,3 MBANZA NGOMBE 2206 3 1,4 0 0,0 3 1,4 MBANZA NKAZI 3474 1 0,3 2 0,6 3 0,9 MINKELO 5428 9 1,7 21 3,9 30 5,5 NDUIZI 4162 0 0,0 0 0,0 0 0,0 NGOMBE 8582 1 0,1 0 0,0 1 0,1 NKAMUNA 4689 17 3,6 59 12,6 76 16,2 NKENGE 7759 0 0,0 0 0,0 0 0,0 NSONA MPANGU 3774 4 1,1 24 6,4 28 7,4 NTADI A NZADI 2527 6 2,4 12 4,7 18 7,1 PALABALA 2451 0 0,0 0 0,0 0 0,0 SONGOLOLO 10107 33 3,3 38 3,8 71 7,0 Total 91310 118 1,3 281 3,1 399 4,4

187

Table S3 Comparison of active case features in the two Rural Health Zones of Songololo Territory, July-August 2008.

188

Table S4 Comparison of inactive case features in the two Rural Health Zones of Songololo Territory, July-August 2008.

189

Figure 1: Location of the survey zone. A. Map of Africa showing the location of the Democratic Republic of Congo. B. Map of the Democratic Republic of Congo showing the location of the province of Bas-Congo. C. Map of the Province of Bas-Congo showing the location of the two health zones surveyed in 2008.

190

Figure 2: Distribution of total number of BU cases (active and inactive) in the Songololo Territory, July-August 2008.

191

Figure 3: Distribution of active BU cases in the Songololo Territory, July-August 2008.

192

Figure S1: Distribution of confirmed active BU cases in the Songololo Territory, July- August 2008.

193

Supporting Figure S2: Distribution of inactive BU cases in the Songololo Territory, July- August 2008.

194

IV.2.3. Effectiveness of routine BCG vaccination on Buruli Ulcer Disease: a case-control study in the DR Congo, Ghana and Togo

Richard Odame Phillips1, Delphin Mavinga Phanzu2, Marcus Beissner3, Kossi Badziklou4, Elysée Kalundieko Luzolo2, Fred Stephen Sarfo1, Wemboo Afiwa Halatoko4, Yaw Amoako1, Michael Frimpong5, Abass Mohammed Kabiru6, Ebekalisai Piten7, Issaka Maman4, Bawimodom Bidjada4, Adjaho Koba4, Koffi Somenou Awoussi4, Basile Kobara8, Jörg Nitschke3/9, Franz Xaver Wiedemann9, Abiba Banla Kere4, Ohene Adjei1, Thomas Löscher3, Bernhard Fleischer10, Gisela Bretzel3, and Karl-Heinz Herbinger3. These authors contributed equally to this work.

1 Department of Medicine, School of Medical Sciences, Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana 2 Institut Médical Evangélique (IME) de Kimpese, Projet Ulcère de Buruli, Kimpese, Democratic Republic of the Congo 3 Department of Infectious Diseases and Tropical Medicine (DITM), University Hospital, Ludwig-Maximilians-University, Munich, Germany 4 Institut National d´Hygiène (INH), Ministry of Health, Lomé, Togo 5 Kumasi Centre for Collaborative Research in Tropical Medicine (KCCR), Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana 6 Agogo Presbyterian Hospital, Agogo, Ghana 7 Centre Hospitalier Régional Maritime (CHR Maritime), Tsévié, Togo 8 Programme National de Lutte contre l’Ulcère de Buruli - Lèpre et Pian (PNLUB-LP), Ministry of Health, Lomé, Togo 9 German Leprosy and Tuberculosis Relief Association (DAHW), Würzburg, Germany 10 Bernhard Nocht Institute for Tropical Medicine (BNITM), Hamburg, Germany

Personal contribution: Conceived and designed the experiments: DMP; contributed reagents/materials/analysis tools: DMP; performed the experiments: DMP; Reviewed the paper: DMP.

PLoS Negl Trop Dis. 2014 Jan; 9(1):e3457

195

Abstract

Background

After tuberculosis and leprosy, Buruli Ulcer Disease (BUD) is the third most common mycobacterial disease. The only available vaccine that potentially could be beneficial against these diseases contains live attenuated bovine tuberculosis bacillus (Mycobacterium bovis) also called Bacillus Calmette-Guérin (BCG), named according to its inventors. Even though the BCG vaccine is still widely used since the first vaccination of humans in 1921, results on its effectiveness in preventing mycobacterial diseases are partially contradictory, especially regarding BUD. One of the main reasons for that might be that BCG vaccination is administered with different BCG strains with variable immunogenicity produced by more than 40 manufacturers worldwide. The aim of this case-control study was to evaluate possible protective effect of BCG vaccination on BUD.

Methodology

The present retrospective case-control study was performed in three different countries and sites where BUD is endemic: in the Democratic Republic of the Congo, Ghana, and Togo between February 2010 and April 2013. After excluding clinically suspected patients without laboratory confirmation and individuals with unknown BCG status, the large study population was comprised of 401 cases and 826 controls. The cases were defined as patients affected by BUD, whose diagnosis was laboratory confirmed with microscopy, polymerase chain reaction, and culture. Most controls had a close relationship with the cases, like family members (27.2%) and neighbors (62.7%). The status of BCG vaccination of every case and control was assumed by detection of a typical scar on one of the shoulders or anterior side of the forearm. Since 1980, three different BCG strains are used for immunization, delivered mostly by UNICEF: BCG- Russia, BCG-Japan, and BCG-Danish.

196

Principal Findings

The proportion of males was 44.4% (among 401 cases) / 46.1% (among 826 controls). Most individuals were younger than 20 years: 65.1% / 55.4%. As controls were significantly younger (p = 0.01), all of the subsequent analyses were performed by stratification by age groups or by multivariate analysis considering age. Among the 401 cases, 95.5% had single, whilst 4.5% had multiple lesions. Ulcerative lesions (58.4%) were more common than non-ulcerative (41.7%) ones. The most common locations of lesions were lower (44.3%) and upper (41.4%) limbs, followed by trunk (11.5%) and face (2.9%).

After stratification by the three countries, two sexes and four age groups, no significant correlation was found between the presence of BCG scar (yes or no) and BUD status of individual (cases or controls). After stratification by the three countries and four age groups, the correlation was only significant (p = 0.03) in Ghana in age group 10-19 years. Multivariate analysis has shown that the independent variables country (p = 0.31), sex (p = 0.24), age (p = 0.96), and presence of a BCG scar (p = 0.07) did not significantly influence the dependent variable, if an individual develops BUD category I or category II/III. Furthermore, the status of BCG vaccination was also not significantly related to duration of BUD or time to healing of lesions.

Conclusions

In our study, we did not observe any evidence of a protective effect of routine BCG vaccination with Mycobacterium bovis on the risk to develop BUD or to develop severe forms of BUD after stratifying by country, age and sex. As no exact data on BCG strains used in these three countries were available, no final conclusion can be taken if any BCG strain could be protective against BUD or not. As for tuberculosis and leprosy, also for BUD, well-designed studies on different already existing BCG vaccine

197 strains are needed and further research on safe and specific vaccines against BUD should be supported.

Introduction

Buruli Ulcer Disease (BUD), caused by Mycobacterium (M.) ulcerans, is an infectious disease affecting skin, subcutanous adipose tissue, and in rare cases, bones. It is one of the 17 neglected tropical diseases defined by the World Health Organization (WHO). BUD has been reported in 33 countries, with a major endemic focus in West and Central Africa. The exact mode of transmission of M. ulcerans is still unknown. However, recent studies suggest that the pathogen is acquired from the environment with different modes of transmission in different geographic areas and epidemiological settings. Consequently, except for early case detection, confirmation and treatment, primary measures to prevent BUD are currently lacking. Furthermore, no effective vaccine against BUD is available so far [1.WHO, 2013].

After tuberculosis and leprosy, BUD is the third most common mycobacterial disease among immunocompetent human hosts. The only available vaccine against these diseases contain live attenuated bovine tuberculosis bacillus (M. bovis), also called Bacillus Calmette-Guérin (BCG), named after its inventors [2.Calmette I, 1927]. Calmette and Guérin began their research for an antituberculosis vaccine at the Pasteur Institute in Lille, France, in 1900. The first use in humans dates from 1921, when Turpin and Weill-Hallé vaccinated infants at the Charité Hospital in Paris by oral, later also by subcutaneous and intracutaneous routes [3.Calmette, 1924; 4.Wünsch-Filho, 1993; 5.Luca, 2013]. From 1924 to 1928, 114,000 infants were vaccinated without serious complications, however with limited effectiveness on preventing tuberculosis [6.Calmette II, 1927].

198

From the late 1940s onward, many studies had appeared providing evidence for the effectiveness of BCG, with widely varying results, ranging from 0% to 80% effectiveness for tuberculosis [5.Luca, 2013; 7.Moise, 2009]. For these disparate results, two principal hypotheses were discussed. The first one stated that exposure to various environmental mycobacteria could itself provide some protection against tuberculosis and affect the immune system in various ways, implying that BCG could not improve greatly upon that background [8.Palmer, 1966; 5.Luca, 2013]. The second hypothesis attributed the differences to variation between strains of BCG [9.Osborn, 1983; 5.Luca, 2013]. It was recognized that strains produced by diverse manufacturers differed in microbiological properties [10.Milstien, 1990] and hence it was not unreasonable to suggest that these might be reflected in differences in immunogenicity [11.Hart, 1967; 5.Luca, 2013]. However, in children, the effectiveness of BCG was estimated to be 50%, or even approximately 80% in preventing tuberculous meningitis and miliary tuberculosis [12.Colditz, 1994; 13.WHO, 1999; 14.Debacker, 2006].

Worldwide, over 90% of children are immunized with BCG, making it the most commonly administered vaccine, with more than 12 million doses used each year [15.Ritz, 2008]. Although BCG has been administerd to more people than any other vaccine, its history has been clouded by variable efficacy and reports of strain variability [16.Behr, 1999]. BCG has never been cloned, and there are now several different BCG seed strains in use, produced by more than 40 manufacturers [17.Clements, 2003]. Nineteen major vaccine strains are described in the literature, whereas the original vaccine used from 1921 was lost: BCG-Moreau (“Brazilian strain”: 1924), BCG-Russia (BCG-Moscow or “Russian strain”: 1924; genetically identical to BCG-Bulgaria or BCG-Sophia: 1950s), BCG-Japan (“Tokyo strain 172”: 1925), BCG-Romania (1925), BCG-Sweden (“Goethenburg strain”: 1926), BCG-Birkhaug (1927), BCG-Danish (BCG- Denmark or BCG-Copenhagen or “Danish strain 1331”: 1931), BCG-Tice (BCG-Chicago or “Tice strain”: 1934), BCG-Frappier (BCG-Montreal:

199

1937), BCG-Phipps (BCG-New York, BCG-Park, BCG-Philadelphia: 1938), BCG-Prague (“Czechoslovakian” strain: 1947), BCG-China (BCG-Beijing: 1947 or 1948), BCG-Shanghai (1948), BCG-Lanzhou (1948), BCG- Connaught (BCG-Toronto or “Theracys strain”: 1948), BCG-Polish (1950s), BCG-Glaxo (“BCG-London F10” or “Glaxo strain 1077”: 1954), BCG-Pasteur (“Pasteur strain 1173P2”: obtained in 1961), BCG-Mexico (1970), BCG-Mérieux (1989).

The following eight strains are the most common BCG strains in present use: Moreau, Russia, Japan, Danish, Tice, Connaught, Glaxo, and Pasteur. These five BCG strains represent more than 90% of the global BCG production: Russia, Japan, Danish, Glaxo, and Pasteur [16.Behr, 1999; 18.Smith, 2012]. According to Ritz et al., for some BCG strains (Russia, Japan, Danish, Prague, Glaxo, and Pasteur) results from at least nine studies were published from each strain, whereas for others, very little or no study results were found in the literature [15.Ritz, 2008]. Studies and observations have shown that BCG-Pasteur and BCG-Danish are “strong” vaccines with higher immunogenicity and with greater complications rates than BCG-Japan or “weak” vaccines as BCG-Russia or BCG-Glaxo [18.Smith, 2012, 19.Ritz, 2012].

Each of these BCG vaccines is produced in a different manner, and they are recognized to differ in various qualities, such as the proportion of viable cells per dose [10.Milstien, 1990; 5.Luca, 2013]. However, the majority of the world's population is supplied with BCG vaccine procured by UNICEF (The United Nations Children's Fund) on behalf of the GAVI Alliance (formerly “Global Alliance for Vaccines and Immunization”). UNICEF uses only four BCG vaccine suppliers who produce only three different BCG vaccine strains: BCG-Russia, BCG-Japan, and BCG-Danish [5.Luca, 2013].

BCG is also recognized to cause cross-protection against leprosy [20.Merle, 2010; 21.Setia, 2006]. A meta-analysis found that

200 experimental studies demonstrated an overall protective effect of 26% (95% CI 14-37%) and that observational studies overestimated the protective effect. Over the years, several vaccine trials using BCG have been performed to establish its limited protective effect against leprosy, often in combination with M. leprae or related mycobacterium vaccines. BCG was as good as, or superior to the other mycobacterium vaccines [22.Richardus, 2013; 23.Richardus, in press].

Additionally, cross-protection of BCG against BUD was also shown in several studies, but their results are partially contradictory. An earlier study in Uganda showed an immune protection by BCG vaccination lasting six months [24.Uganda Buruli group, 1969]. The findings are consistent with another study in Uganda concluding that BCG vaccination provides only short-term protection against M. ulcerans disease [25.Smith, 1976]. In two studies in Benin, BCG was shown to be protective against more severe M. ulcerans diseases, notably osteomyelitis [26.Portaels, 2002; 27.Portaels, 2004]. A study performed in Cameroon concluded that BCG appeared to protect children against more severe forms of BUD with multiple lesions [28.Noeske, 2004]. However, none of these studies described the BCG strain used for vaccination.

In a mouse model experiments, the potential mechanisms for cross- protection were studied. A study identified and characterized the M. ulcerans homologue of the important protective mycobacterial antigen 85 (Ag85A) from BCG. This antigen was sufficiently conserved to allow cross- reactive protection, as demonstrated by the ability of M. ulcerans-infected mice to exhibit strong cellular immune responses to both BCG and its purified Ag85 complex [29.Tanghe, 2001]. It was also shown, that BCG vaccine offered short-term protection against experimental footpad infections of mice with M. ulcerans, and that duration of this protection could not be prolonged by a booster vaccination [30.Tanghe, 2007]. Another study using a mouse model observed that BCG vaccination significantly delayed the onset of M. ulcerans growth and footpad swelling

201 through the induction of an earlier and sustained IFN-γ triggered T cell response in the draining lymph node. BCG vaccination also resulted in cell-mediated immunity in M. ulcerans-infected footpads [31.Fraga, 2012].

Two epidemiologic studies, performed in Benin, could not find any evidence of a protective effect of routine BCG vaccination against BUD. In the second study, in persons aged >5 years, a BCG scar even resulted in a risk factor of 2.5 for BUD compared with those without a BCG scar [14.Debacker, 2006; 32.Nackers, 2006]. The first two studies on the effectiveness of BCG vaccines performed in Ghana did not show any significant difference between cases and controls regarding their BCG vaccination status [33.Amofah, 1993, 34.Raghunathan, 2005]. None of these studies described the BCG strain used for vaccination.

Although many studies of BCG vaccine were performed, the results regarding the vaccine’s effectiveness against mycobacterial diseases including BUD differ immensely. Based on this unclear situation, the present case-control study was conducted with a large study population in the Democratic Republic of the Congo (DR Congo), Ghana, and Togo. In these three countries, only three different BCG strains were used since BCG was introduced between 1978 and 1984: BCG-Russia, BCG-Japan, and BCG-Danish. In the context of the EC-funded research project “BuruliVac” (FP7/2010-2013; grant agreement N° 241500), the aim of the present study was to evaluate possible protective effectiveness of routine BCG vaccination containing live attenuated bovine tuberculosis bacillus M. bovis on BUD in the DR Congo, Ghana, and Togo.

Materials and Methods

BuruliVac

BuruliVac was founded in 2009 as consortium of 16 European and African partners. As there is currently no existing vaccine lead candidate

202 available, BuruliVac aimed to identify and develop new vaccine candidates of three different types: (1) Mycolactone-directed vaccines, (2) attenduated live vaccines, and (3) subunit protein vaccines. Furthermore, BuruliVac evaluated resulting vaccine candidates using bioinformatics, applied genomics and proteomics, and subjected them to consecutive test systems. BuruliVac was funded by the European Commission under the 7th Framework Programme of the European Union [35.CORDIS, 2013].

BCG in study countries

The present study was performed in the DR Congo, Ghana, and Togo. These three countries follow the WHO recommendations for routine immunization, which are part of their national immunization programs. This includes the advice to administer the one-time BCG vaccine intracutaneously, as soon as possible, either at birth or directly after, but not later than twelve months after birth, because at that age the vaccination is usually of limited benefit, although it is not harmful or contraindicated. Booster shots are not recommended [36.WHO I, 2013]. The WHO estimates the BCG coverage rates in these three African countries as follows: 78% in the DR Congo, 98% in Ghana, and 97% in Togo [37.WHO II, 2013].

Study sites

This study consists of data collected at the three following different sites, which are members of BuruliVac. The Project Ulcère de Buruli is performed by the Institut Médical Evangélique (IME) de Kimpese, in the DR Congo. Since 1999, the general reference hospital (GRH) of the IME, located in the Songololo Territory, 220 km southwest of Kinshasa, regularly admits BUD cases. In 2004, the GRH launched a specialized BUD program offering in-patient treatment free-of-charge and supplementary aid. The principal aims of this project are the improvement of the patient care of BU patients admitted at the IME and the promotion of early community-

203 based detection of suspected BU cases. Patients and controls were recruited from Kimpese and Nsona-Mpangu health zones, both located in the Songololo Territory, Province of Bas-Congo [38.Phanzu, 2006; 39.Phanzu, 2011].

The Department of Medicine and the Kumasi Centre for Collaborative Research (KCCR) of the School of Medical Sciences at Kwame Nkrumah University of Science and Technology (KNUST) is based in Kumasi, Ghana. They have been involved with BUD in the area of training, diagnostic confirmation, provision of specialist care for BUD patients in disease endemic districts, recruited patients and controls from the Ahafo Ano North, Asante Akim North, Atwima-Nwabiagya, and the Upper Denkvira districts that are all within 70 km of the Ashanti Regional capital Kumasi [40.Phillips, 2005; 41.Schunk, 2009].

The Centre Hospitalier Régional Maritime (CHR Maritime) in Tsévié, Togo, collaborates since 2007 with the German Leprosy and Tuberculosis Relief Organization (DAHW). This collaboration is supported by the Togolese National Buruli Ulcer Control Program (“Programme National de Lutte contre L´Ulcère de Buruli – Lèpre et Pian” [PNLUB-LP]), in the area of training, laboratory confirmation and treatment of BUD. In 2007, the CHR Maritime was appointed National Reference Centre for BUD in Togo [42.Bretzel, 2011; 43.Beissner, 2013].

BCG strains

In the DR Congo, BCG vaccination was introduced in 1984. The following BCG strains were used for vaccinations: 1984 – 2003: BCG-Russia (equivalent to “BCG-Bulgaria”; produced by Bulbio [BB-NCIPD], Sofia, Bulgaria, and by Serum Institute of India); 2004: BCG-Japan (produced by Japan BCG Laboratory); 2005 – 2009: BCG-Japan (produced by Japan BCG Laboratory) and BCG-Russia (produced by Serum Institute of India); 2010 – 2011: BCG-Japan (produced by Japan BCG Laboratory); 2012:

204

BCG-Japan (produced by Japan BCG Laboratory) and BCG-Russia (produced by Serum Institute of India); January to July 2013: BCG-Russia (equivalent to “BCG-Bulgaria”; produced by Bulbio [BB-NCIPD], Sofia, Bulgaria); August and September 2013: BCG-Russia (produced by Serum Institute of India); October 2013 to date: BCG-Bulgaria = BCG-Russia (produced by Bulbio [BB-NCIPD], Sofia, Bulgaria).

In Ghana, BCG vaccination was introduced in 1978. The following BCG strains were used for vaccinations: 2007: BCG-Danish (produced by Danish Statens Serum Institute); 2008 – 2009: BCG-Bulgaria = BCG- Russia (produced by Bulbio [BB-NCIPD]; 2010 to date: BCG-Japan (produced by Japan BCG Laboratory, Tokyo, Japan). Exact data on BCG strains used in Ghana between 1978 and 2006 were not available.

In Togo, BCG vaccination was introduced in 1980. The following BCG strains were used for vaccinations: 2004: BCG-Japan (produced by Japan BCG Laboratory, Tokyo, Japan); 2004 – 2009: BCG-Russia (produced by Serum Institute of India); 2010 to date: BCG-Russia (equivalent to “BCG- Bulgaria”; produced by Bulbio [BB-NCIPD], Sofia, Bulgaria, and by Serum Institute of India). Exact data on BCG strains used in Togo between 1980 and 2003 were not available.

Study design and definitions

In these three study sites, the recruitment of both BUD cases (drawn from patients “clinically suspected” of having BUD) and healthy controls was conducted. The present retrospective case-control study defined cases (CA) as patients affected by BUD, whose diagnosis was laboratory confirmed by testing with microscopy, polymerase chain reaction (PCR), and culture. Any CA had a least one positive test result. Patients who were “clinically suspected” (CS) for BUD, but without laboratory confirmation (none of the tests results was positive) were not considered in the study

205 population. The controls (CO) were defined as healthy persons who were in close relationship with the CA.

Study population

In the time period February 2010 to April 2013, data of 1,335 individuals were collected. Out of them, 406 (30.41%) were CA, 103 (7.72%) were CS, and 826 (61.87%) were CO. From these data, 622 participants (128 CA: 20.58%; no CS; 494 CO: 79.42%) were from the DR Congo, 504 participants (196 CA: 38.89%; 65 CS: 12.90%; 243 CO: 48.21%) were from Ghana, and 209 participants (82 CA: 39.23%; 38 CS: 18.18%; 89 CO: 42.58%) were from Togo. Four CA from Ghana and one CA from Togo had unknown BCG status and were excluded out of the study.

Consequently, the study population was comprised of 1,227 participants (401 CA: 32.68%; 826 CO: 67.32%), including 622 from the DR Congo (128 CA: 20.58%; 494 CO: 79.42%), 435 from Ghana (192 CA: 44.14%; 243 CO: 55.86%), and 170 from Togo (81 CA: 47.65%; 89 CO: 52.35%). The 826 CO were in the following relationship with the CA: 225 (27.24%) were family members, 518 (62.71%) neighbors, 32 (3.87%) friends or classmates, and 51 (6.17%) were others or those with unspecified relationship.

Data collection

Data collection was conducted by means of the WHO “BU01” form, and standardized project specific “BuruliVac” laboratory data entry forms (Form S1). All socio-demographic, clinical, and laboratory data were entered in a web-based database specifically designed for the “BuruliVac” project [43.Beissner, 2013]. Following WHO guidance, the categories of BUD were defined as follows: Category I were single lesions <5 cm in diameter; Category II were single lesions between 5 and 15 cm in

206 diameter; Category III were single lesions >15 cm in diameter, multiple lesions, or osteomyelitis [44.WHO, 2008].

The status after BCG vaccination was assessed from all CA and CO of the study population by examining both sides of the arms or shoulders, if they presented a scar typical for vaccination with BCG or not, but not by documents such as vaccination certificates or hospital registers. Former studies that evaluated the presence or absence of BCG scars to determine vaccination status reported that scars develop in most vaccinated persons: Scarring rates >80% [45.Fine, 1989; 46.Floyd, 2000; 14.Debacker, 2006, 47.Roth, 2005].

Sample collection

In the DR Congo, fine needle aspirates were only collected from non- ulcerative lesions. Routinely, a direct smear was conducted at peripheral health centers from the first fine needle aspiration (FNA) and then the sample was stored in transport media (7H9 + PANTA liquid) and forwarded to IME for microscopy and culture. The second FNA (if possible) or a suspension was forwarded to the Institut National de Recherche Biomédicale (INRB) in Kinshasa via IME, where microscopy and PCR was performed. Similar procedures were applied for swabs and tissue biopsies, stored in semi-liquid transport medium (Dubos + PANTA semi-liquid).

In Ghana and Togo, diagnostic samples were collected according to standardized procedures [43.Beissner, 2013]. Briefly, swabs were collected by circling the entire undermined edges of ulcerative lesions. Fine needle aspirates were collected from the center of non-ulcerative lesions or from undermined edges of ulcerative lesions. Punch biopsy samples were only collected from advanced ulcers with scarred edges if fine needle aspirates were tested negative according to recent WHO recommendation [48.WHO, 2013].

207

Standardized specimen collection bags including swabs, biopsy punches, syringes and needles, slides, containers with transport media (700 µl [swab and punch biopsy samples], 300 µl [FNA samples] CLS [cell lysis solution, Qiagen, Hilden, Germany] for PCR samples) and data entry forms were provided to the study sites in Ghana and Togo [49.Bretzel, 2006; 50.Siegmund, 2007; 51.Bretzel, 2007; 52.Herbinger I, 2009; 53.Herbinger II, 2009; 54.Eddyani, 2009; 55.Phillips, 2009; 56.Herbinger, 2010, 57.WHO, 2010].

Samples for PCR analysis were transported in CLS at ambient temperature in an upright position in custom-made specimen collection bags from the field to the laboratories from the two study sites in Ghana and one study site in Togo within a maximum of 48 hours and stored at 4-8°C until further processing. Slides for microscopy were transported in slide boxes at ambient temperature to the laboratory.

Laboratory diagnostics

Direct smears for microscopy were prepared from swab and fine needle aspirates at the laboratory and were subjected to Ziehl-Neelsen staining. Slides were analyzed according to the WHO recommended grading system [58.WHO, 2001, 56.Beissner, 2010]. For PCR analysis, DNA was prepared using the Gentra Puregene DNA extraction kit (Qiagen, Hilden, Germany) with minor modifications of the manufacturer’s protocol [60.Siegmund, 2005].

In the study sites in Ghana and Togo, three IS2404 PCR formats (dry- reagent-based [DRB] IS2404 PCR [INH], standard gel-based IS2404 PCR and IS2404 qPCR [DITM]) were applied in this study. Briefly, for DRB-PCR the oligonucleotides MU5 and MU6 were lyophilized in reaction tubes. Illustra PuReTaq Ready-To-Go PCR beads (GE Healthcare, Munich, Germany) were added and dissolved in water before adding template DNA [50.Siegmund, 2007; 51.Bretzel, 2007; 60.Siegmund, 2005;]. Standard

208

IS2404 PCR was performed according to the protocol described by Stinear et al. [61.Ross, 1997; 62.Stinear, 1999]. IS2404 qPCR was performed as recently described using a BioRad CFX96 real-time PCR detection system [63.Fyfe, 2007; 64.Beissner, 2012]. All PCR assays included negative extraction controls, positive, negative and inhibition controls.

In the study site in the DR Congo, the Maxwell® 16 DNA extraction procedure was carried out with the Maxwell® 16 Tissue DNA Purification Kit and the Maxwell® 16 Instrument, according to manufacturer's instructions: 200 μl of specimen was added to 200 μl of lysis buffer (10 mM Tris-HCl pH 7.5, 10 mM NaCl, 10 mM EDTA, 50 ml 10% SDS solution) and 10 μl proteinase K (20 mg/ml) and incubated overnight at 60 °C in a shaker incubator. Only IS2404 qPCR was performed on an Applied Biosystems 7500 Fast Real-Time PCR System using the method of [63.Fyfe, 2007].

Ethics statements

In Kimpese, in the DR Congo, the ethical clearance was obtained through the Comite d’Éthique of the Ecole de Santé Publique of the University of Kinshasa (Ref. No. ESP/CE/057/2010). In Kumasi, in Ghana, the ethical clearance was obtained through the Committee on Human Research Publication and Ethics of the College of Health Sciences of the Kwame Nkrumah University of Science and Technology (Ref. No. CHRPE/91/10). In Tsévié, in Togo, the ethical clearance was obtained through the national Togolese ethics committee (“Comité de Bioéthique pour la Recherche en Santé”) at the University of Lomé (14/2010/CBRS) and the study was approved by the “Ministère de la Santé de la République Togolaise” Lomé, Togo (Ref. No. 0009/2011/MS/DGS/DPLET). All samples analyzed in this study were collected for diagnostic purposes within the EC funded research project “BuruliVac”. Written informed consent was obtained from all study participants, their guardians if aged <18 years according the recommendations of the respective ethical committees. In case of

209 illiterates, informed consents were countersigned by means of thumb prints.

Statistical analysis

All data assessed at these three study sites were entered into the web- based database of BuruliVac and descriptively analyzed with Excel 2007 (Microsoft, Redmond, WA). The hypothesis of the present study was to evaluate associations between the presence of BCG scars (independent variable), which are caused by BCG vaccinations, and risk for BUD (dependent variable). Bivariate approximative tests (χ2-tests) and exact test (Fisher’s tests) were conducted using EpiInfo, version 3.3.2. (Centers for Disease Control and Prevention, Atlanta, GA) and multiple logistic regression by Stata software, version 9.0. (Stata Corporation, College Station, TX) and. Significant differences were defined as p-values below 0.05.

Results

Baseline data of cases and controls

Among the study population of 1,227 individuals (401 CA and 826 CO) males comprised 45.56% (559), which was not significantly (p = 0.57) different between CA (44.39%: 178) and CO (46.13%: 381). Stratification by the three countries found no significant differences in the proportion of males among CA and CO. Among the 401 CA, the range of age was 1 to 78 years (y) and the median of age was 13 y. Among the 826 CO, the range of age was 1 to 90 y and the median of age was 16 y. Age distribution in CA and CO was significantly (p = 0.01) different, as the CA were younger than the CO: Age group (AG) 0–9 y (30.42% in CA vs. 26.63% in CO), AG 10–19 y (34.66% vs. 28.81%), AG 20–39 y (21.95% vs. 29.78%), and AG 40–90 y (12.95% vs. 14.77%). Stratified by the three countries, significant differences (p <0.01 each) of the proportions

210 of these four AG among CA and CO were found in Ghana and Togo, but not in the DR Congo (p = 0.97) (Table 1).

Lesions of cases

Among the 401 CA, 383 (95.50%) CA were detected with a single lesion, 15 (3.74%) with two lesions each, two (0.50%) with three lesions each, and one (0.25%) with four lesions. Out of them, 167 (41.65%) CA had non-ulcerative and 234 (58.35%) ulcerative lesions. The proportion of detected non-ulcerative lesions was as follows: nodules (74: 18.45%), plaques, (58: 14.46%), edema only (27: 6.73%), papules (7: 1.75%), and osteomyelitis (1: 0.25%).

Among the 401 CA, microscopy was performed for 399 (99.50%) CA, PCR for 384 (95.76%) CA, and culture for 159 (39.65%) CA. The sensitivity of the three tests was as follows: PCR 97.14% (373/384), microscopy 69.42% (277/399), and culture 35.22% (56/159). Of 384 (95.76%) CA, the location of the lesions was known: 2.86% (11/384) were diagnosed with BUD with lesion in the face, 41.41% (159) on the upper limbs, 11.46% (44) on the trunk, and 44.27% (170) on lower limbs. The right lower limb (26.30%: 101) was significantly (p <0.01) more frequently affected than the left lower limb (17.97%: 69), whereas no significant differences where found between presence of lesions on the right and left side of the body for the face, upper limbs, or trunk.

BCG scars of cases and controls

Among 401 CA, 175 (43.64%) had no BCG scar (CAA), whilst 226

(56.36%) had BCG scar (CAB). Among 826 CO, 277 (33.54%) had no BCG scar (COA), whilst 549 (66.46%) had BCG scar (COB). The proportion of those with BCG scar was significantly (p <0.01) higher among the CO than among CA. When stratified by the three countries, a significant difference of the proportion of individuals with a BCG scar among CA and

211

CO was only found in Ghana (p = 0.03), but not in the DR Congo (p = 0.22) and in Togo (p = 0.67) (Table 1).

Stratified by four age groups, a significant higher proportion of those with BCG scar among CO was only found in AG 10–19 y and AG 40–90 y (p <0.01 each). Stratified by the three countries and four age groups, a significant higher proportion of those with BCG scar among CO was only found in Ghana in AG 10–19 (p = 0.03) (Table 1). Multivariate analysis (multiple logistic regression model) confirmed, that the independent variables country (p <0.01), age (p <0.01), and status of BCG vaccination (p = 0.02) did significantly influence the dependent variable, if an individual develops BUD (CA) or not (CO).

Stratified by sex, a significant higher proportion of those with BCG scar among CO was only found among females (p <0.01), but not males (p = 0.09). When stratified by sex and by country, no significant difference of that proportion was found. After stratification by three countries, two sexes and four age groups, no significant correlation was found between the presence of BCG scare (yes or no) and BUD status of individual (CA or CO).

BCG scars and categories of cases

Among the 175 CAA and 226 CAB, from 85.14% (149/175) and 77.43% (175 /226) respectively, the BUD category was clearly known. The proportions of CA with category I, II and III among CAA were 48.99 (73), 41.61% (62), and 9.40% (14), whereas these proportions were 60.57%

(106/175), 27.43% (48), and 12.00% (21) among CAB. Consequently, among the CAA, the proportion of those with categories II and III was

51.01% (76), which was significantly (p = 0.04) higher than among CAA (39.43%: 69). Stratified by the three countries, no significant correlation was found between presence of BCG scar (yes or no) and categories (I or

II/III) was found. Among the 175 CAA, the proportion of CA detected with

212 multiple lesions was 4.57% (8), whereas it was not significantly (p =

0.94) higher than those of 4.42% (10) among the 226 CAB (Table 2).

Among the 324 individuals with known BUD category (149 CAA and 175

CAB), the proportion of males was 44.75% (145), which was not significantly (p = 0.55) different between CAA (42.95%: 64) and CAB (46.29%: 81). Stratified by sex, no significant correlation between presence of BCG scar (yes or no) and categories (I or II/III) was found.

Among the 149 CAA, the range of age was 1 to 78 y and the median of age was 18 y. Among the 175 CAB, the range of age was 2 to 70 y and the median of age was 12 y. Age distribution in CAA and CAB was significantly

(p <0.01) different, as the CAA were younger than the CAB: AG 0–9 y

(20.81% in CAA vs. 37.71% in CAB), AG 10–19 y (34.23% vs. 34.29%), AG 20–39 y (24.83% vs. 19.43%), and AG 40–90 y (20.13% vs. 8.57%) (Table 2).

After stratification by the three countries, two sexes and four age groups, no significant correlation was found between presence of BCG scar (yes or no) and categories (I or II/III). Multivariate analysis (multiple logistic regression model) confirmed, that the independent variables country (p = 0.31), sex (p = 0.24), age (p = 0.96), and presence of BCG scar (p = 0.07) did not significantly influence the dependent variable, if an individual develops BUD category I or category II/III.

BCG scars and duration of BUD

Among the 175 CAA, the proportions of duration of BUD of 0–30 days (d), 31–60 d, 61–90 d, 91–180 d, and >180 d were 46.29% (81), 21.71% (38), 12.57% (22), 13.14% (23), and 6.29% (11), whereas these proportions were 46.02% (104), 22.12% (50), 13.72% (31), 11.06%

(25), and 7.08% (16) among the 226 CAB. The difference was not significant (p = 0.97), neither after stratification by the BUD categories.

213

BCG scars and time to healing

Among the 401 CA, 305 (76.06%) CA were treated adequately by only antibiotics, 87 (21.70%) CA by antibiotics and surgery, seven (1.75%) by only surgery, and from two (0.50%) CA, no data on treatment were available. Among the 175 CAA and 226 CAB, from 82.29% (144/175) and 80.97% (183 /226) respectively, the time to healing was known (time difference between onset of treatment up to the point of time of macroscopic healing of BUD lesion). Among the 144 CAA, the proportions of time to healing of 7–90 d, 91–180 d, and >180 d were 27.08% (39), 45.83% (66), and 27.08% (39), whereas these proportions were 32.79%

(60), 33.33% (61), and 33.88% (62) among the 226 CAB. The difference was not significant (p = 0.07), and neither after stratification by the BUD categories.

Discussion

This is one of the largest observational studies on the effectiveness of Bacillus Calmette-Guérin (BCG) vaccines on Buruli Ulcer Disease (BUD). The aim of the present retrospective case-control study was to evaluate possible protection of routine BCG vaccination with live attenuated bovine tuberculosis bacillus Mycobacterium bovis against BUD in the DR Congo, Ghana, and Togo. Since the first human vaccination with BCG occurred in 1921, many studies of BCG vaccines have been performed to estimate their effectiveness, but their results differed immensely. These discrepancies are explained by three main factors: the BCG strain used for vaccination, the population vaccinated, and the mycobacterial disease or its manifestation.

The past and continued use of both strong and weak vaccine strains makes interpretation and comparison of clinical trials extremely difficult, thus no conclusions can be made that one BCG strain is clearly superior to

214 another in the protection of humans against tuberculosis or other mycobacterial diseases [17.Smith, 2012; 65.Comstock, 1988]. More than 20 different BCG seed strains are in use for vaccination, which are produced by more than 40 manufacturers. African countries like the DR Congo, Ghana and Togo, were mainly supplied with BCG vaccine procured by UNICEF as BCG-Russia, BCG-Japan, and BCG-Danish. As explained above, the BCG vaccines used in these three countries changed very often, so if was not possible to figure out retrospectively with which BCG strain a certain study participant was vaccinated if that person has shown a typical BCG scar. As no documentation in hospital files or on vaccination cards was performed, no data on exact time of vaccination could have been assessed neither. Consequently, the present study could not consider the BCG strain used for vaccination even it is known that strong strains as BCG-Danish, less strong strains as BCG-Japan and weak strains as BCG-Russia were in use in these three countries. This classification refers only to tuberculosis and it is totally unknown if this might be also conferrable on BUD [17.Smith, 2012, 19.Ritz, 2012].

This study assessed the effectiveness of BCG vaccination on BUD only. Tuberculosis, leprosy or any other disease which might influence the data, were not considered. The study population included 401 laboratory confirmed BUD cases and 826 adequate controls. To minimize confounding, the association between presence of BCG scar (yes or no) and BUD status (case or control) were calculated after stratification by the three countries, two sexes, and four age groups, and by multiple analysis.

Several studies have shown that the effectiveness of BCG is dependent on the population in which the vaccination is used. Age plays a role, as effectiveness among children is much higher in preventing tuberculous meningitis and miliary tuberculosis [12.Colditz, 1994; 13.WHO, 1999; 14.Debacker, 2006]. On the other side, BCG vaccines seem to be more effective against leprosy among adults [20.Merle, 2010; 21.Setia, 2006]. To avoid influence of age, all analyses were performed after stratification

215 by four age groups. It is totally unknown if there is any age-depending vaccine effectiveness against BUD if found against tuberculosis and leprosy. After stratification into three countries and four age groups, the present study found only a significant higher proportion of those with BCG scar among CO in Ghana in AG 10–19 (p = 0.03), but confounded by sex. After stratification by three countries, two sexes and four age groups, no significant correlation was found between the presence of BCG scare (yes or no) and BUD status of individual (CA or CO).

Furthermore, that vaccine effectiveness was calculated to be different in populations with high or low exposure to environmental mycobacteria. High exposure to mycobacteria affects the immune system in various ways and thus, BCG might not improve greatly upon that background [8.Palmer, 1966; 5.Luca, 2013]. In the three study sites of the present study, we assumed an equal or at least comparable exposure to mycobacteria among the populations. To avoid influence of country specific populations in general, all analyses were performed after stratification by the three countries.

In the present study, multivariate analysis has shown that country, sex, age, and presence of BCG scar did not significantly influence the dependent variable, whether an individual develops BUD category I or category II/III. Furthermore, the status of BCG vaccination was also not significantly related to duration of BUD nor to time of healing. These results underline those of four studies performed in Benin [14.Debacker, 2006; 32.Nackers, 2006] and in Ghana [33.Amofah, 1993, 34.Raghunathan, 2005], which did not show any significant difference between cases and controls regarding their BCG vaccination status. These results contradict those of two other studies performed in Benin which generated the hypothesis that BCG vaccination might protect children against more severe M. ulcerans diseases, notably osteomyelitis [26.Portaels, 2002; 27.Portaels, 2004], and another study performed in Cameroon which concluded that BCG appeared to protect children against

216 more severe forms of BUD with multiple lesions [28.Noeske, 2004]. Again, none of the studies considered the BCG strain used for vaccination and they could not answer the question if certain BCG strains might protect better than others against BUD.

The present study had the same limitation. Exact data on BCG vaccination among the study participants could not be assessed by documents, such as vaccination certificates or hospital registers. Thus, the status of BCG vaccination of every case and control was just assumed by detection of a typical scar on one shoulder or anterior side of the forearm, based on the fact that scars develop in most vaccinated persons as described before [45.Fine, 1989; 46.Floyd, 2000; 14.Debacker, 2006; 47.Roth, 2005]. Probably a certain proportion of individuals were defined as “vaccinated”, even though the scar was caused by something other than a BCG vaccination (“false positive”). On the other side, also a certain proportion might have been defined as “not vaccinated”, if no scare was found on the shoulder or anterior side of the forearm, because BCG vaccination did not lead to a “typical scar” (“false negative”). The number of such “false positive” and “false negative” cases and controls is not known and could not be estimated in the present study. Furthermore, no other data on the BCG vaccination (e.g. way of application, booster vaccination, and side effects) could be assessed. This inaccuracy cannot be estimated neither, but might be equally distributed among cases and controls. To minimize this bias, we have chosen a case-control-design.

From the time since the first studies were conducted on the effectiveness of BCG vaccine, the results are varying and will continue to vary as long as retrospective studies with little precise data are performed. As a consequence of this, we demand to conduct prospective studies, with an exact documentation as to which vaccine was administered. Given the fact that some BCG strains might have a short-time protection against BUD in certain populations as shown in some studies [24.Uganda Buruli group, 1969; 25.Smith, 1976], this effect would hardly decrease the incidence of

217

BUD. A safe and effective specific vaccine with long-time protection against BUD which could be used in several populations of the most BUD endemic countries would be an adequate preventive tool to reduce the risk for this disease.

Given the fact that some BCG strains might have certain protection to avoid more severe forms of BUD, notably osteomyelitis [26.Portaels, 2002; 27.Portaels, 2004] and multiple lesions [28.Noeske, 2004], this effect would also not decrease the incidence of BUD, because only a small proportion of BUD cases are diagnosed with osteomyelitis (in the present study <1%) and only a very small proportion of BUD cases are diagnosed with multiple lesions (in the present study <5%).

Even though only a limited number of studies on BCG effectiveness for the prevention BUD have been conducted, the probability of finding an effective BCG strain against BUD is low, thus efforts to research specific vaccines against BUD should be accelerated. This objective is exactly that of the consortium BuruliVac which was founded in 2009 aiming to identify and develop new vaccine candidates against BUD.

In our study, we did not observe any evidence of a protective effect of routine BCG vaccination with Mycobacterium bovis on the risk to develop BUD or to develop severe forms of BUD. As no exact data on BCG strains used were available, no final conclusion can be taken if any BCG strain could be protective against BUD or not. As for tuberculosis and leprosy, also for BUD, well-designed prospective studies on different already existing BCG vaccine strains are needed and further research on new vaccines, which are safe and specific against BUD, should be supported.

218

Author Contributions

Conceived and designed the projects: ROP, DMP, MB, KB, TL, BF, GB. Performed the field survey: ROP, DMP, KB, EKL, FSS, WAH, YA, MF, AMK, EP, IM, BB, AK, KSA, BK, JN, FXW, ABK, OA. Contributed reagents and materials: ROP, DMP, KB, EP, FXW, GB. Analyzed the data: KHH. Wrote the paper: KHH. Reviewed the paper: ROP, DMP, MB, TL, GB.

Acknowledgement

The authors thank all study participants in Kimpese (DR Congo), in Kumasi (Ghana), and in Tsévié (Togo) for their great cooperation, and all district and regional controllers of the existing networks for their continuous highly motivated assistance. Furthermore, the authors appreciate the financial support for BuruliVac from the fund of the European Commission under the 7th Framework Programme of the European Union. There is no conflict of interest among any author.

References

1. I World Health Organization (2013) Buruli Ulcer (Mycobacterium ulcerans infection). Fact sheet N°199. Geneva: World Health Organization. Homepage: http://www.who.int/mediacentre/factsheets/fs199/en/ 2. I Calmette A, Guérin C, Boquet A, Nègre L (1927) La vaccination préventive contre la tuberculose par le "BCG". Masson, Paris. 3. Calmette A, Guérin C, Weill-Halle B (1924) Bull Acad Med. Essai d'immunisation contre l'infection tuberculeuse. 91:787–96. 4. Wünsch-Filho V, Moncau JE, Nakao N (1993) Methodological considerations in case-control studies to evaluate BCG vaccine effectiveness. Int J Epidemiol. 22(1):149–55.

219

5. Luca S, Mihaescu T (2013) History of BCG vaccine. Maedica. 8(1):53– 8. 6. II Calmette A, Guérin C, Nègre L, Boquet A (1927) Sur la vaccination préventive des enfants nouveau-nés contre la tuberculose par le BCG. Ann Inst Pasteur. 41:201–32. 7. Moise L, McMurry J, de Groot AS (2009) Mycobacterium tuberculosis. Barrett ADT, Stanberry LR (eds.): Vaccines for Biodefense and Emerging and Neglected Diseases. London, Elsevier: 1219-1241. 8. Palmer CE, Long MW (1966) Effects of Infection with Atypical Mycobacteria on BCG Vaccination and Tuberculosis. Am Rev Respir Dis. 94(4):553–68. 9. Osborn TW (1983) Changes in BCG strains. Tubercle. 64(1):1–13. 10. Milstien JB, Gibson JJ (1990) Quality Control of BCG Vaccine by WHO: a Review of Factors that May Influence Vaccine Effectiveness and Safety. Bull World Health Organ. 68(1):93–108. 11. Hart PD (1967) Efficacy and Applicability of Mass B.C.G. vaccination in tuberculosis control. BMJ. 1(5540):587–92. 12. Colditz GA, Brewer TF, Berkey CS, Wilson ME, Burdick E, Fineberg HV, Mosteller F. (1994) Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. JAMA. 271(9):698–702. 13. World Health Organization (1999) Issues relating to the use of BCG in immunization programmes: a discussion document. Publication No. WHO/V + 13/99.23. Geneva: World Health Organization. 14. Debacker M, Portaels F, Aguiar J, Steunou C, Zinsou C, Meyers W, Dramaix M (2006) Risk factors for Buruli ulcer, Benin. Emerg Infect Dis. 12(9):1325–31. 15. Ritz N, Hanekom WA, Robins-Browne R, Britton WJ, Curtis N (2008) Influence of BCG vaccine strain on the immune response and protection against tuberculosis. FEMS Microbiol Rev. 32(5):821–41. 16. Behr MA, Small PM (1999) A historical and molecular phylogeny of BCG strains. Vaccine. 17(7–8):915–22.

220

17. Clements CJ (2003) Vaccination – The Current Status of BCG. Kaufmann SHE, Hahn H (eds.): Mycobacteria and TB. Issues Infect Dis. Basel, Karger. 2:46–66. 18. Smith KC, Starke JR (2012) Bacille Calmette-Guérin Vaccine. Plotkin SA, Orenstein W, Offit PA (eds.): Vaccines. Saunders. 6:188–209. 19. Ritz N, Dutta B, Donath S, Casalaz D, Connell TG, Tebruegge M, Robins-Browne R, Hanekom WA, Britton WJ, Curtis N (2012) The influence of bacilli Calmette-Guerin vaccine strain on the immune response against tuberculosis: a randomized trail. Am J Respir Crit Care Med. 185(2):213–22. 20. Merle CS, Cunha SS, Rodrigues LC (2010) BCG vaccination and leprosy protection: review of current evidence and status of BCG in leprosy control. Expert Rev Vaccines. 13(2):209–22. 21. Setia MS, Steinmaus C, Ho CS, Rutherford GW (2006) The role of BCG in prevention of leprosy: a meta-analysis. Lancet Infect. 13(3):162–70. 22. Richardus RA, Alam K, Pahan D, Feenstra SG, Geluk A, Richardus JH (2013) The combined effect of chemoprophylaxis with single dose rifampicin and immunoprophylaxis with BCG to prevent leprosy in contacts of newly diagnosed leprosy cases: a cluster randomized controlled trial (MALTALEP study). BMC Infect Dis. 13:456. 23. Richardus JH, Oskam L (probably 2014) Protecting people against leprosy: Chemoprophylaxis and Immunoprophylaxis (vaccination). Clin Dermatol. In press. 24. Uganda Buruli group (1969) BCG vaccination against Mycobacterium ulcerans infection (Buruli ulcer). First results of a trial in Uganda. Lancet. 1(7586):111–5. 25. Smith PG, Revill WD, Lukwago E, Rykushin YF (1976) The protective effect of BCG against Mycobacterium ulcerans disease: a controlled trial in an endemic area of Uganda. Trans R Soc Trop Med Hyg. 70(5– 6):449–57. 26. Portaels F, Aguiar J, Debacker M, Steunou C, Zinsou C, et al. (2002) Prophylactic effect of Mycobacterium bovis BCG vaccination against

221

osteomyelitis in children with Mycobacterium ulcerans disease (Buruli ulcer). Clin Diagn Lab Immunol. 9(6):1389–91. 27. Portaels F, Aguiar J, Debacker M, Guédénon A, Steunou C, et al. (2004) Mycobacterium bovis BCG vaccination as prophylaxis against Mycobacterium ulcerans osteomyelitis in Buruli ulcer disease. Infect Immun. 72(1):62–5. 28. Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, et al. (2004) Buruli ulcer disease in Cameroon rediscovered. Am J Trop Med Hyg. 70(5):520–6. 29. Tanghe A, Content J, Van Vooren JP, Portaels F, Huygen K (2001) Protective efficacy of a DNA vaccine encoding antigen 85A from Mycobacterium bovis BCG against Buruli ulcer. Infect Immun. 69(9):5403–11. 30. Tanghe A, Adnet PY, Gartner T, Huygen K (2007) A booster vaccination with Mycobacterium bovis BCG does not increase the protective effect of the vaccine against experimental Mycobacterium ulcerans infection in mice. Infect Immun. 75(5):2642–4. 31. Fraga AG, Martins TG, Torrado E, Huygen K, Portaels F, et al. (2012) Cellular immunity confers transient protection in experimental Buruli ulcer following BCG or mycolactone-negative Mycobacterium ulcerans vaccination. PLoS One. 7(3):e33406. 32. Nackers F, Dramaix M, Johnson RC, Zinsou C, Robert A, et al. (2006) BCG vaccine effectiveness against Buruli ulcer: a case-control study in Benin. Am J Trop Med Hyg. 75(4):768–74. 33. Amofah GK, Sagoe-Moses C, Adjei-Acquah C, Frimpong EH (1993) Epidemiology of Buruli ulcer in Amansie West district, Ghana. Trans R Soc Trop Med Hyg. 87(6):644–5. 34. Raghunathan PL, Whitney EA, Asamoa K, Stienstra Y, Taylor TH jr, et al. (2005) Risk factors for Buruli ulcer disease (Mycobacterium ulcerans Infection): results from a case-control study in Ghana. Clin Infect Dis. 40(10):1445–53.

222

35. CORDIS, Community Research and Development Information Service of the European Commission. Seventh Framework Programme (FP7). Homepage: http://cordis.europa.eu/fp7/home_en.html 36. II World Health Organization (2013) Recommendations for Routine Immunization – Summary of WHO Position Papers. Homepage: http://www.who.int/immunization/policy/Immunization_routine_table 3.pdf 37. III World Health Organization (2013) WHO-UNICEF estimates of BCG coverage WHO. Homepage: http://apps.who.int/immunization_monitoring/globalsummary/timese ries/tswucoveragebcg.html 38. Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, et al. (2006) Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg. 75(2):311–4. 39. Phanzu DM, Suykerbuyk P, Imposo DB, Lukanu PN, Minuku JB, et al. (2011) Effect of a control project on clinical profiles and outcomes in buruli ulcer: a before/after study in Bas-Congo, Democratic Republic of Congo. PLoS Negl Trop Dis. 5(12):e1402. 40. Phillips R, Horsfield C, Kuijper S, Lartey A, Tetteh I, et al. (2005) Sensitivity of PCR targeting the IS2404 insertion sequence of Mycobacterium ulcerans in an Assay using punch biopsy specimens for diagnosis of Buruli ulcer. J Clin Microbiol. 43(8):3650–6. 41. Schunk M, Thompson W, Klutse E, Nitschke J, Opare-Asamoah K, et al. (2009) Outcome of patients with buruli ulcer after surgical treatment with or without antimycobacterial treatment in Ghana. Am J Trop Med Hyg. 81(1):75–81. 42. Bretzel G, Huber KL, Kobara B, Beissner M, Piten E, et al. (2011) Laboratory confirmation of Buruli ulcer disease in Togo, 2007-2010. PLoS Negl Trop Dis. 7:e1228. 43. Beissner M, Huber KL, Badziklou K, Halatako WA, Maman I, et al. (2013) Implementation of a national reference laboratory for Buruli ulcer disease in Togo. PLoS Negl Trop Dis. 7(1):e2011.

223

44. World Health Organization (2008) Buruli ulcer: progress report, 2004–2008. Wkly Epidemiol Rec. 83(17):145–54. 45. Fine PE, Ponnighaus JM, Maine N (1989) The distribution and implications of BCG scars in northern Malawi. Bull World Health Organ. 67(1):35–42. 46. Floyd S, Ponnighaus JM, Bliss L, Warndorff DK, Kasunga A, Mogha P, et al. (2000) BCG scar in northern Malawi: sensitivity and repeatability of scar reading, and factors affecting scar size. Int J Tuberc Lung Dis. 4(12):1133–42. 47. Roth A, Sodemann M, Jensen H, Poulsen A, Gustafson P, et al. (2005) Vaccination . Vaccination technique, PPD reaction and BCG scarring in a cohort of children born in Guinea-Bissau 2000-2002. Vaccine. 23(30):3991–8. 48. IV World Health Organization (2013) Guidance on sampling techniques for laboratory-confirmation of Mycobacterium ulcerans infection. Homepage: http://www.who.int/buruli/Guidance_sampling_techniques_MU_infect ion.pdf 49. Bretzel G, Siegmund V, Nitschke J, Herbinger KH, Thompson R, et al. (2006) External quality assurance for the laboratory diagnosis of Buruli ulcer disease in Ghana. Trop Med Int Health. 11(11):1688–93. 50. Siegmund V, Adjei O, Nitschke J, Thompson W, Klutse E, et al. (2007) Dry reagent-based polymerase chain reaction compared with other laboratory methods available for the diagnosis of Buruli ulcer disease. Clin Infect Dis. 45(1):68–75. 51. Bretzel G, Siegmund V, Nitschke J, Herbinger KH, Thompson W, et al. (2007) A stepwise approach to the laboratory diagnosis of Buruli ulcer disease. Trop Med Int Health. 12(1):89–96. 52. I Herbinger KH, Brieske D, Nitschke J, Siegmund V, Thompson W, et al. (2009) Excision of pre-ulcerative forms of Buruli Ulcer Disease: A curative treatment? Infection. 37(1):20–5. 53. II Herbinger KH, Adjei O, Awua-Boateng NY, Nienhuis WA, Kunaa L, et al. (2009) Comparative study of the sensitivity of different

224

diagnostic methods for the laboratory diagnosis of Buruli ulcer disease. Clin Infect Dis. 48(8):1055–64. 54. Eddyani M, Fraga AG, Schmitt F, Uwizeye C, Fissette K, et al. (2009) Fine needle aspiration, an efficient sampling technique for the bacteriological diagnosis of nonulcerative Buruli ulcer lesions. J Clin Microbiol. 47(6):1700–4. 55. Phillips RO, Sarfo FS, Osei-Sarpong F, Boateng A, Tetteh I, et al. (2009) Sensitivity of PCR targeting Mycobacterium ulcerans by use of fine-needle aspirates for diagnosis of Buruli ulcer. J Clin Microbiol. 47(4):924–6. 56. Herbinger KH, Beissner M, Huber K, Awua-Boateng NY, Nitschke J, et al. (2010) Efficiency of fine-needle aspiration compared with other sampling techniques for laboratory diagnosis of Buruli ulcer disease. J Clin Microbiol. 48(10):3732–4. 57. World Health Organization (2010) Guidance on sampling techniques for laboratory-confirmation of Mycobacterium ulcerans infection (Buruli ulcer disease). Geneva: World Health Organization. 58. World Health Organization (2001) Diagnosis of Mycobacterium ulcerans disease. Geneva: World Health Organization. 59. Beissner M, Herbinger KH, Bretzel G (2010) Laboratory diagnosis of Buruli ulcer disease. Future Microbiol. 5(3): 363–70. 60. Siegmund V, Adjei O, Racz P, Berberich C, Klutse E, et al. (2005) Dry-reagent-based PCR as a novel tool for laboratory confirmation of clinically diagnosed Mycobacterium ulcerans-associated disease in areas in the tropics where M. ulcerans is endemic. J Clin Microbiol. 43(1):271–6. 61. Ross BC, Marino L, Oppendisano F, Edwards R, Robins-Browne RM, et al. (1997) Development of a PCR Assay for rapid Diagnosis of Mycobacterium ulcerans Infection. J Microbiol Clin. 35(7): 1696–700. 62. Stinear T, Ross BC, Davies JK, Marino L, Robins-Browne RM, et al. (1999) Identification and characterization of IS2404 and IS2606: two distinct repeated sequences for detection of Mycobacterium ulcerans by PCR. J Clin Microbiol. 37(4):1018–23.

225

63. Fyfe JA, Lavender CJ, Johnson PD, Globan M, Sievers A, et al. (2007) Development and application of two multiplex real-time PCR assays for the detection of Mycobacterium ulcerans in clinical and environmental samples. Appl Environ Microbiol. 73(15):4733–40. 64. Beissner M, Symank D, Phillips RO, Amoako YA, Awua-Boateng NY, et al. (2012) Detection of Viable Mycobacterium ulcerans in Clinical Samples by a Novel Combined 16S rRNA Reverse Transcriptase/IS2404 Real-Time qPCR Assay. PLoS Negl Trop Dis. 6(8):e1756. 65. Comstock GW (1988) Identification of an effective vaccine against tuberculosis. Am Rev Respir Dis. 138(2):479–80.

226

Table 1. Baseline data of casesa and controlsb from the Democratic Republic of the Congo (DR Congo), Ghana, and Togo, collected between February 2010 and April 2013.

227

228

Table 2. BCGa vaccination and casesb of BUD categoryc I and categoryc II/III from the Democratic Republic of the Congo (DR Congo), Ghana, and Togo, collected between February 2010 and April 2013.

229

230

IV.2.4. Management of Mycobacterium ulcerans disease (Buruli ulcer) in the Territory of Songololo, Democratic Republic of Congo: Outcomes, Challenges and Prospects.

Delphin Mavinga Phanzu 1,8, Marcus Beissner2, Koen Vandelannoote3, Philippe Ngwala Lukanu4, Dierto Kembo Mputu5, Désiré Bofunga B. Imposo1, Elysée Kalundieko Luzolo1, Blanchard M. Diengidi1, Oscar Nlemvo Kiabanzawoko1, Bibiche Mundabi6, Nadine Mi Makuth Mintsey6, Pascal Tshindele Lutumba6, Miriam Eddyani3, Bouke de Jong3, Gisela Bretzel2, Françoise Portaels3, Greet Ieven7, Marleen Boelaert8

1 General Reference Hospital of Kimpese, Institut Médical Evangélique, Kimpese, Bas- Congo, Democratic Republic of Congo 2 Department of Infectious Diseases and Tropical Medicine (DITM), University Hospital, Ludwig-Maximilians University, Munich, Germany 3 Mycobacteriology Unit, Microbiology Group, Department of Biomedical Sciences, Institute of Tropical Medicine, Antwerp, Belgium 4 Central Office of the Rural Health Zone of Kimpese, Bas-Congo, Democratic Republic of Congo 5 Central Office of the Rural Health Zone of Nsona Mpangu, Bas-Congo, Democratic Republic of Congo 6 Institut National de Recherche Biomédicale, Kinshasa, Democratic Republic of Congo 7 Laboratory of Medical Microbiology, University Hospital of Antwerp, Antwerp, Belgium 8 Department of Public Health, Institute of Tropical Medicine, Antwerp, Belgium.

Personal contribution: Conceived and designed the experiments: DMP; capacity strengthening: DMP; performed the experiments: DMP; analyzed the data: DMP; wrote the paper: DMP.

Paper in preparation

231

Abstract Background: Mycobacterium ulcerans disease (Buruli ulcer), a necrotizing infection mainly of the subcutaneous tissue, is the third most prevalent human mycobacteriosis after tuberculosis and leprosy. Without treatment, the infection leads to destruction of skin and soft tissue with large ulcers usually on limbs, resulting in long-term functional disability. The General Reference Hospital of the “Institut Médical Evangélique” of Kimpese, located in the Territory of Songololo, known to be one of the main Buruli ulcer (BU) foci in the Democratic Republic of Congo, in 2004 launched a specialized BU program, offering in-patient treatment free-of- charge and supplementary aid. Although the number of BU cases admitted in the hospital was rising, a survey conducted in July-August 2008, in the Territory of Songololo, showed that only one of thirteen active BU cases was notified at the hospital in Kimpese in the 8 months prior to the survey. Methods: This is a retrospective, observational study, from January 2009 through September 2014, to assess the impact of decentralization and integration of BU control activities at the primary health care level in Songololo. Results: The impact evaluation of the decentralization showed: (i) strong increase in case detection from 211 new cases in the period 2002-2007, before the Songololo survey conducted in 2008, to 901 new cases in the period 2009-2014; (ii) increase in number of health areas reporting cases each year, from an average of 12.5 out of 40 health areas before to 26 out of 40 after; (iii) Moreover, in the last six years most patients were diagnosed at an earlier stage (category I); (iv) three-quarters of reported cases have been treated in peripheral health centres; (v) and half of reported cases were confirmed by at least one laboratory test. Conclusion: Decentralization and integration of BU control activities in primary health care in Songololo are effective, and results obtained so far suggest these to be key ways of improved diagnosis and treatment access for BU patients at the most peripheral level of the health system. A positive interrelation between a BU control program and the rest of the

232 health system is of utmost importance to sustain an efficient surveillance system in a resource-limited context, exportable to other BU endemic regions of the country. Meanwhile, sustaining the project’s impact remains a big challenge in the region.

INTRODUCTION Buruli ulcer is caused by infection with Mycobacterium ulcerans, an organism which belongs to the family of bacteria that causes tuberculosis and leprosy. It occurs mainly in West and Central Africa, and it often affects people in remote rural areas with limited access to health care. Without treatment, the infection leads to destruction of skin and soft tissue with large ulcers usually on limbs, resulting in long-term functional disability. Early diagnosis and treatment are the only ways to minimize morbidity and prevent disability (1). Until recently, surgery was the only available treatment. Thus, surgery was recommended for all BU patients, but availability was limited to major BU treatment centers, mainly built or sponsored by international NGOs in Africa (2-6). The introduction of antibiotic therapy in 2004 (7) has shifted the balance between surgical treatment, mainly limited to reference centers, and antibiotics administered at the most peripheral level of the health system. Globally, between 30% and 50% of patients are now treated solely with antibiotics, facilitating greater access to treatment (5). In the Territory of Songololo, despite encouraging results in improving clinical profiles and outcomes in BU obtained by a specialized program for BU implemented in the hospital of the “Institut Médical Evangélique” (IME) (8), a survey conducted in July-August 2008 showed that only one of thirteen active BU cases had been reported in the hospital-based reporting system at Kimpese in the 6-8 months prior to the survey (9), and brought out the need to decentralize and integrate surveillance and control activities at primary health-care level to increase the therapeutic coverage and control effectiveness. The aim of study is to evaluate the impact of BU

233 control activities decentralization in Songololo from January 2009 to September 2014. We present the outcomes, challenges and prospects.

PATIENTS AND METHODS Study design This retrospective observational study reports results of BU surveillance in the Territory of Songololo in the context of patient’s management decentralization (Table 1), using the World Health Organization (WHO) standardized BU02 forms, from January 2009 through September 2014. In order to improve early case detection and management, at least 260 community volunteers (around 1 per 900 inhabitants) and 160 health professionals (1 per 1500 inhabitants) were trained. Study site The study was conducted in two rural health zones (RHZ), Kimpese and Nsona-Mpangu, both located in Songololo Territory (Phanzu et al., 2013), one of ten territories of Bas-Congo Province. It is situated in the District of Cataractes and covers an area of 8,190 Km2, approximately 15.2% of the total surface of the province, with a population of 237,418 inhabitants in 2009. Songololo Territory is limited in the north by the Congo River, in the west by Sekebanza Territory, in the east by Mbanza-Ngungu Territory and in the south by the northern border of Angola. Each RHZ is subdivided into 20 health areas. The primary level of health care facilities includes the Rural Health Posts (HP), Health Centres (HC) and Reference Health Centres (RHC), and the secondary level is represented by the GRH (Table 1). Case definition BU was defined as painless or minimally painful preulcerative lesions or ulcers with subcutaneous tissue necrosis, undermined edges and peripheral induration. Preulcerative lesions include papule, nodule, plaque and edema, as described by the WHO Global Buruli Ulcer Initiative (10- 12). A new case was defined as a patient with no previous history of or treatment of BU, and a recurrent case was defined as a patient presenting

234 within one year from the end of the last treatment with a new lesion at the same or a different site. We defined functional limitation as any reduction in the range of motion of one or more joints, and assessed it by clinical observation. We categorized lesions as follows: A single lesion < 5 cm (Category I); a single lesion 5-15 cm (Category II); a single lesion > 15 cm, multiple lesions, and lesions at critical sites (face, breast and genitalia) or osteomyelitis (Category III). Sample collection In order to detect the maximum number of BU cases, active case finding was implemented by the health zone teams of Kimpese and Nsona- Mpangu, in close collaboration with the IME BU project team. First level health workers were trained to detect BU cases during routine work. Samples were routinely divided for different laboratory tests: one specimen processed at the peripheral laboratory within the TB network, one specimen sent to the BU regional laboratory (IME, Kimpese) and one specimen sent to the BU national reference laboratory (INRB, Kinshasa). Fine Needle aspirates (FNA) were collected from non-ulcerative lesions (13). A direct smear staining was done at peripheral level (RHC) from the first FNA and then the sample was stored in transport media (7H9 +PANTA liquid) and forwarded to IME and further processed for microscopy and culture. The second FNA (if possible) was forwarded to INRB via IME, where PCR was performed. Similar procedures were applied for swabs and tissue biopsies, stored in semi-solid transport medium (Dubos + PANTA semi-liquid) (14). Furthermore, the corresponding laboratory data entry forms BU03 and BU01 forms were filled in on-site during sample collection by nurses or physicians of the health centres. Laboratory diagnosis Direct smears for microscopy were prepared from swab and fine needle aspirates at the laboratory and were subjected to Ziehl-Neelsen staining. Slides were analyzed according to the WHO recommended grading system (11,15). For PCR analysis, the Maxwell® 16 DNA extraction procedure was carried out with the Maxwell® 16 Tissue DNA Purification Kit and the Maxwell® 16 Instrument, according to manufacturer's instructions: 200 μl

235 of specimen was added to 200 μl of lysis buffer (10 mM Tris-HCl pH 7.5, 10 mM NaCl, 10 mM EDTA, 50 ml 10% SDS solution) and 10 μl proteinase K (20 mg/ml) and incubated overnight at 60 °C in a shaker incubator. qPCR for M.ulcerans specific target IS2404 was performed on an Applied Biosystems 7500 Fast Real-Time PCR System (16). Data collection and statistical analysis Patient information was collected from the WHO BU02 forms of all health areas of Kimpese and Nsona-Mpangu health zones in Songololo Territory from January 1st, 2009 to September 30th, 2014. BU02 registers are available in 3 sheets of different colours, white, yellow and pink (4). Data from BU02 registers were transmitted on a monthly basis to the health zone office and the IME BU project coordination. Each month, the nurses in charge of health areas brought the white and yellow filled sheets of the previous month to the health zone review, and the pink sheet was kept and stored in health centres. Demographic data, clinical form, location of lesions, category of lesions, presence of functional limitations on diagnosis, results of laboratory tests and information on treatment were collected. Data were entered in a web-based database specifically designed for the “Burulivac” project (17), exported to Excel 2007 (Microsoft Corporation, Redmond, WA) and analyzed with Epi-Info version 3.3.2 (Centers for Disease Control and Prevention, Atlanta, GA). Ethics Statement The study protocol was approved by the Ethical Committee of the School of Public Health, University of Kinshasa (Ref. N° ESP/CE/057/2010). All participants, or their guardian in the case of minors, provided written informed consent for all diagnostic and treatment procedures, and were treated according to the WHO recommendations (7). Management of patients was free of charge.

RESULTS Case detection From January 1, 2009, through September 30, 2014, a total of 901 new cases were reported and treated in Songololo. After a peak of detection in

236

2012 with 188 cases, a downward trend was observed in recent years (Figure 1). The average annual rate of detection increased from 35.2 to 150.2 cases when comparing with 211 new cases reported during the six years period (2002-2007) before the Songololo survey conducted in 2008 (Figure 1). Most patients lived in the Songololo Territory, where the Coordination of the BU project is based. The BU patients living out of Songololo represented 5% (41 out of 901) of patients (Table 2). Their major origin locations were the Territory of Mbanza-Ngungu (46%), followed by the Territory of Luozi (37%), located both with the Territory of Songololo in Cataractes District (Table 3). We observed an increasing number of health areas reporting cases each year, from an average of 12.5 out of 40 health areas before, to 26 out of 40 during the study period. The age distribution of BU patients showed that 40% of the 901 cases were <15 years of age, 45% were 16-49 years, and 16% more than 49 years. During the study period, 53% of patients were female and 47% male. Clinical presentations The past six years, 76% of patients were detected with ulcerated lesions, ranging from 72 to 82% (Figure 2). Limitation of joint movement was observed in 33% of patients on diagnosis. Osteomyelitis represented 3.2% of cases. Out of 923 listed lesions, 572 (62%) were located in the lower limbs, 246 (27%) to the upper limbs and 105 (11%) on other sites. Among other locations, the most frequent were, thorax (29%), head and neck (27%), and buttock and perineum (22%), followed by abdomen (11%) and the back (11%). The preponderance of category I lesions these past six years is remarkable (Figure 3). Of the total number of patients, 46% presented lesions of category I, 26% category II and 28% category III. Thus, category I and II represented 72% of all lesions. We observed an increased proportion of category III lesions in 2010, followed by a plateau the following three years, and a downward trend in 2014 (Figure 3).

237

Laboratory confirmation From 2009 to 2014, the laboratory confirmation rate has grown exponentially from 36 to 64%. In total, half of reported cases has been confirmed by at least one laboratory test. During the years 2002 to 2007, laboratory confirmation rate ranged from 50 to 80% (Figure). The trends of lab confirmation rate since 2002 is presented in Figure 4. Treatment During the study period, three-quarters of reported cases have been treated in basic health facilities (Figure 5), and only 18 cases (2%) of relapse have been observed. In these peripheral treatment centers, standard antibiotic regimens recommended by WHO, wound care, and the prevention of disability constituted an important part of the treatment offered to patients. The remaining quarter was referred to the hospital, mainly for the indication of surgery.

DISCUSSION Buruli ulcer is a public health problem in most countries in West and Central Africa, and it predominantly affects the rural poor. In Africa, the first control actions were carried out by specialized institutions, mainly private (2,4,6,8). This period was characterized by poor geographical and therapeutic coverage of the population at risk and lack of accurate estimation of the disease burden (3,18). Care for BU was largely hospital- based (19,20). Moreover, patients often reported to specialized units at a late stage of the disease and suffered long-term functional disability (1). In the Province of Bas-Congo, a recent survey, conducted in the Territory of Songololo, demonstrated a large variation in prevalence between health areas and gross underreporting of BU cases in the hospital-based notification system (9). The recent advances in diagnosis and treatment of BU by specific antibiotic therapy, combined with early detection, have the potential to shift the focus to community-based care (7,20-21).

238

Since the launching of decentralization of BU control activities in Songololo, we observed a strong increase in case detection. Indeed, 4.3 times more cases were detected the last six years compared to the period before. The number of health areas that reported cases doubled. We noticed a remarkable change in the lesion category’s profile with the preponderance of Category I lesions. Importantly, three-quarters of reported patients were treated in basic health facilities, and the laboratory confirmation rate was notably improved. Several factors, such as a better coordination of BU control activities (planning, monitoring and evaluation), extensive collaboration across all levels, capacity strengthening of health professionals in various aspects of BU management, training of community-based surveillance volunteers, intensification of health education and community-based surveillance, contributed to obtain current results. Songololo Territory has remained the primary source of BU patients. Out of 11 known BU endemic health zones in the Province of Bas-Congo, since more than a decade, BU control activities remained focused in Kimpese and Nsona-Mpangu, two health zones of Songololo Territory in the District of Cataractes. For patients living out of Songololo, the main origin locations were Mbanza-ngungu and Luozi Territories. This is consistent with previous studies in the same region (8,22). Despite active surveillance efforts, the predominant clinical presentation was an ulcerative lesion in 76% of cases. This is consistent with previous studies, in the same areas (9) or in other endemic areas in Africa (1,4). As in most BU endemic regions, the real challenge in Songololo is how to further reduce the percentage of ulcerative lesions and sustain such surveillance efforts (4). According to the WHO, early diagnosis and treatment are the only ways to minimize morbidity and prevent disability (1). Intensifying health education at the community level and surveillance, training of health workers and community-based surveillance volunteers would ultimately increase early reporting and treatment for all cases.

239

A recent study in a highly endemic district in Ghana, where intensified public health efforts, mainly revolving around training and motivating community-based surveillance volunteers (CBSVs) were implemented, has shown that CBSVs are an important resource in the early detection of BU (21). Previous studies in other BU endemic countries reported similar observations (23-25). Besides CBSVs, teachers, family members, and former patients also proved to be crucial in the referral system (25). Thus, it is incumbent upon us to make certain the public is well-informed and cases adequately treated. While most of our BU patients were successfully treated in primary level (Table 1), peripheral health workers experienced great difficulty relating to immediate work conditions. Indeed, one of the major obstacles to the control of BU in Africa, is inadequate and poorly-equipped health facilities in the affected areas (23). To improve early BU case management, strengthening existing clinics to increase access to antibiotic therapy and wound care is critical (20). There is a growing interest in improving the relationship between disease control programmes and the rest of the health system in low-and middle-income countries (26). Recently, the WHO Stop TB Programme highlighted the need for mutual strengthening of TB programmes and ‘general health services’ (27). The effectiveness of such vision for BU is of utmost importance. Another major difficulty in sustaining the programme is the almost total dependence on external financing and the absence of government budgetary funding. This is common in African endemic countries including DRC (20). Ackumey et al. suggested that for effective BU control, particularly for poorly resourced rural health systems, documented cases could serve as indices for classifying communities into three categories: priority-endemic areas, requiring the most interventions, endemic and non-endemic, requiring further research to enhance understanding of the disease (20). A similar approach was applied in Songololo since 2010. A top-ten list of priority-endemic health areas in the two health zones Kimpese and Nsona-Mpangu was established, helping for an efficient allocation of limited available resources.

240

CONCLUSION Decentralization and integration of BU control activities in primary health care in Songololo are effective, and results obtained so far suggest these to be key ways of improved diagnosis and treatment access for BU patients at the most peripheral level of the health system. A positive interrelation between a BU control program and the rest of the health system is of utmost importance to sustain an efficient surveillance system in a resource-limited context, exportable to other BU endemic regions of the country. Meanwhile, sustaining the project’s impact remains a big challenge in the region.

ACKNOWLEDGEMENTS We are grateful to all participants in this study, the staff of the IME/Kimpese Hospital and health professionals in the health zones of Kimpese and Nsona-Mpangu, as well as the staff of the Mycobacteriology Unit and the Unit of Epidemiology and Disease Control of the ITM/Antwerp. This study was supported by the American Leprosy Missions (Greenville, South Carolina, United States of America), European Community’s Seventh Framework Programme (FP7/2010–2013) under grant agreement N° 241500 and the Directorate General for Development and Cooperation (Brussels, Belgium). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Address for correspondence: Marleen Boelaert, Department of Public Health, Unit of Epidemiology and Disease Control, Institute of Tropical Medicine, Antwerp, Belgium.

241

REFERENCES 1. World Health Organization (2013) Buruli Ulcer (Mycobacterium ulcerans infection). Fact sheet N°199. Geneva: World Health Organization. Homepage:http://www.who.int/mediacentre/factsheets/fs199/en/

2. Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Guédénon A, Scott JT, Dramaix M, Portaels F. Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, Southern Benin, 1997-2001. Emerg Infect Dis 2004; 10: 1391-8.

3. Johnson RC, Sopoh GE, Boko M, Zinsou C, Gbovi J, Makoutode M, Portaels F. [Distribution of Mycobacterium ulcerans (Buruli ulcer) in the district of Lalo in Benin]. Trop Med Int Health 2005; 10: 863-71.

4. Sopoh GE, Johnson RC, Chauty A, Dossou AD, Aguiar J, Salmon O, Portaels F, Asiedu K. Buruli ulcer surveillance, Benin, 2003-2005. Emerg Infect Dis 2007; 13: 1374-6.

5. World Health Organization. Buruli ulcer: progress report, 2004-2008. Wkly Epidemiol Rec 2008; 83: 145-154.

6. Peeters Grietens K, Boock AU, Peeters H, Hausmann-Muela S, Toomer E, Ribera JM. "It is me who endures but my family that suffers": social isolation as a consequence of the household cost burden of Buruli ulcer free of charge hospital treatment. PLoS Negl Trop Dis 2008; 2:e321.

7. World Health Organization (2004) Provisional guidance on the role of specific antibiotics in the management of Mycobacterium ulcerans disease (Buruli ulcer), WHO/CDS/CPE/GBUI/2004.10, World Health Organization, Geneva.

8. Phanzu DM, Suykerbuyk P, Imposo DB, Lukanu PN, Minuku JB, Lehman LF, Saunderson P, de Jong BC, Lutumba PT, Portaels F, Boelaert M. Effect of a control project on clinical profiles and

242

outcomes in Buruli ulcer: a before/after study in Bas-Congo, Democratic Republic of Congo. PLoS Negl Trop Dis 2011; 5: e1402.

9. Phanzu DM, Suykerbuyk P, Saunderson P, Lukanu PN, Minuku JBM, Imposo DBB, Diengidi BM, Kayinua M, Muyembe JJT, Lutumba PT, de Jong BC, Portaels F, Boelaert M. Burden of Mycobacterium ulcerans disease (Buruli ulcer) and underreporting ratio in the Territory of Songololo, Democratic Republic of Congo. PLoS Negl Trop Dis 2013; 7: e2563.

10. Asiedu K, Scherpbier R, Raviglione M. Buruli ulcer-Mycobacterium ulcerans infection. World Health Organization. 2000.WHO/CDS/CPE/GUI/2001.1.

11. Portaels F, Johnson P, Meyers WM. Ulcère de Buruli-Diagnostic de l'infection à Mycobacterium ulcerans. World Health Organisation. WHO/CDS/CPE/GBUI/2001.4.

12. World Health Organization, 2005. Modules de formation en organisation et gestion de programme de lutte contre l’ulcère de Buruli au niveau du district. OMS Afro, Harare, Zimbabwe, décembre 2005.

13. Eddyani M, Fraga AG, Schmitt F, Uwizeye C, Fissette K, Johnson C, Aguiar J, Sopoh G, Barogui Y, Meyers WM, Pedrosa J, Portaels F. Fine needle aspiration, an efficient sampling technique for the bacteriological diagnosis of nonulcerative Buruli ulcer. J Clin Microbiol 2009; 47: 1700-4.

14.Eddyani M, Debacker M, Martin A, Aguiar J, Johnson CR, Uwizeye C, Fissette K, Portaels F. Primary culture of Mycobacterium ulcerans from human tissue specimens after storage in semi-solid transport medium. J Clin Microbiol 2008; 46: 69-72.

15. Beissner M, Herbinger KH, Bretzel G. Laboratory diagnosis of Buruli ulcer disease. Future Microbiol 2010; 5: 363–70.

243

16. Fyfe JA, Lavender CJ, Johnson PD, Globan M, Sievers A, Azuolas J, Stinear TP. Development and application of two multiplex real-time PCR assays for the detection of Mycobacterium ulcerans in clinical and environmental samples. Appl Environ Microbiol 2007; 73: 4733–40.

17. Beissner M, Huber KL, Badziklou K, Halatako WA, Maman I, Vogel F, Bidjada B, Awoussi KS, Piten E, Helfrich K, Mengele C, Nitschke J, Amekuse K, Wiedemann FX, Diefenhardt A, Kobara B, Herbinger KH, Kere AB, Prince-David M, Löscher T, Bretzel G. Implementation of a national reference laboratory for Buruli ulcer disease in Togo. PLoS Negl Trop Dis 2013; 7: e2011.

18. Van der Werf TS, Stienstra Y, Johnson RC, Phillips R, Adjei O, Fleischer B, Wansbrough-Jones MH, Johnson PDR, Portaels F, van der Graaft WTA, Asiedu K. Mycobacterium ulcerans disease. Bull World Health Organ 2005; 83:785-791.

19. Asiedu K, Etuaful S. Socioeconomic implications of Buruli ulcer in Ghana: a three-year review. Am J Trop Med Hyg 1998; 59: 1015- 1022.

20. Ackumey MM, Kwakye-Maclean C, Ampadu EO, de Savigny D, Weiss MG. Health Services for Buruli Ulcer Control: Lessons from a Field Study in Ghana. PLoS Negl Trop Dis 2011; 5: e1187.

21. Abass KM, van der Werf TS, Phillips RO, Sarfo FS, Abotsi J, Mireku SO, Thompson WN, Asiedu K, Stienstra Y, Klis SA. Short report: Buruli Ulcer Control in a Highly Endemic District in Ghana: Role of Community-Based Surveillance Volunteers. Am J Trop Med Hyg 2014; 14-0405.

22. Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, Nsiangana SZ, Singa JN, Meyers WM, Suykerbuyk P, Portaels F. Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002–2004. Am J Trop Med Hyg 2006; 75: 311–314.

244

23. Vouking MZ, Tamo VC, Mbuagbaw L. The impact of community health workers (CHWs) on Buruli ulcer in Sub-Saharan Africa: a systematic review. Pan Afr Med J 2013; 15: 19.

24. Vouking MZ, Takougang I, Mbam LM, Mbuagbaw L, Tadenfok CN, Tamo CV. The contribution of community health workers to the control of Buruli ulcer in the Ngoantet area, Cameroon. Pan Afr Med J 2013; 16: 63.

25. Barogui YT, Sopoh GE, Johnson RC, de Zeeuw J, Dossou AD, et al. Contribution of the Community Health Volunteers in the Control of Buruli Ulcer in Benin. PLoS Negl Trop Dis 2014; 8: e3200.

26. Collins C, Gonzalez Block MA, Tang S. Disease control and health systems in low- and middle-income countries: enhancing positive interrelation. Trop Med Int Health 2012; 17: 646-651.

27. Uplekar M & Raviglione MC . The ‘‘vertical–horizontal’’ debates: time for the pendulum to rest (in peace)? Bull World Health Organ 2007; 85: 413–414.

245

Table 1. Overview of BU patient management at different health care levels.

TERTIARY PRIMARY SECONDARY CENTRES MANAGEMENT HP HC RHC GRH UH/INRB

Category of I I & II I & II II & III III and sequelae lesion

Clinical + + + + + diagnosis + + + + + Swab

Fine Needle + + + + + Aspiration - - + + + Biopsy

ZN staining - - + + + (Microscopy) - - - - + PCR

- - - + + Histopathology

- - - + + Culture

+ + + + + Wound dressing

Antibiotics + + + + + administration - - + + + Surgery

+ + + + + POD

+ + + + + Documentation

+ : Could be performed or available at this level. - : Not allowed or unavailable at this level. HP: Health Post; HC: Health Centre; RHC: Reference Health Centre; GRH: General Reference Hospital; UH : University Hospital; INRB : Institut National de Recherche Bio- Médicale; FNA : Fine Needle Aspiration; ZN : Ziehl-Neelsen; PCR : Polymerase Chain Reaction; POD : Prevention of Disability.

246

Table 2. Yearly trends of BU new cases reported in Songololo Teritory, 2009-2014

Origin of 2009 2010 2011 2012 2013 2014 Total BU n (%) n (%) n (%) n (%) n (%) n (%) n (%) patients

Kimpese Health 69 (40) 58 (47) 99 (57) 84 (45) 58 (42) 60 (56) 428 (47) Zone*

Nsona- Mpangu 99 (58) 56 (46) 66 (38) 98 (52) 71 (51) 42 (39) 432 (48) Health Zone*

Outside of 4 (2) 8 (7) 8 (5) 6 (3) 10 (7) 5 (5) 41 (5) Songololo

Total 172(100) 122(100) 173(100) 188(100) 139(100) 107(100) 901(100)

*Located in Songololo Territory, District of Cataractes.

Table 3. Origin of Buruli ulcer patients reported in Songololo Territory, 2009-2014.

Health Total % per Country Province District Territory Zone n Territory Bas- Sekebanza Sekebanza 1 0.1 Fleuve

Gombe- 3 Matadi Mbanza- Kimpangu 1 2.1 Ngungu Bas- Kwilu- 15 DRC Congo Ngongo Cataractes Kibunzi 3 Luozi 1.7 Luozi 12

Kimpese 428 Songololo Nsona- 95.4 432 Mpangu Kinshasa 4 0.4

Angola 2 0.2

Total 901 100

247

Figure 1. Number of BU cases reported in Songololo, 2002-2014.

Figure 2. Ulcerative and non-ulcerative lesions at diagnosis in Songololo, 2002-2014.

248

Figure 3. Evolution of categories of lesions in Songololo, 2005-2014.

Figure 4. Evolution of laboratory confirmation of BU cases in Songololo, 2002-2014.

249

Figure 5. Evolution of BU treatment decentralization in Songololo, 2009-2014.

250

Figure 6. Single ulcerative lesion <15 cm diameter diagnosed and treated in primary health care level.

Figure 7. Healed lesion without complication after antibiotic therapy alone without surgery (same patient in Figure 6).

251

252

CHAPITRE V. DISCUSSION GENERALE

253

254

En 2001, lorsque nous découvrons l’UB à l’hôpital IME/Kimpese, nous disposions d’une connaissance très limitée du tableau clinique de la maladie, de son ampleur et de sa distribution dans le territoire de Songololo. Dans la région, la maladie est restée longtemps méconnue par le personnel de santé et par conséquent sous-diagnostiquée. Ainsi, notre recherche de doctorat s’est donné comme objectif général de contribuer à l’amélioration du contrôle de l’UB dans le territoire de Songololo par une meilleure connaissance et une prise en charge adéquate de la maladie, ainsi qu’une meilleure organisation des activités de lutte. Le déroulement de notre projet de doctorat a été caractérisé par l’évolution de l’observation clinique à la mise en œuvre d’un programme de contrôle de la maladie. Concernant la partie clinique, la première publication porte sur l’expérience de l’HGR IME/Kimpese dans la prise en charge des patients souffrant de l’UB entre 2002 et 2004 (Phanzu DM et al., 2006). Cette étude nous a montré que le fardeau de l’UB en termes de souffrances humaines, de longues durées d’hospitalisation, de développement des séquelles invalidantes et de répercussions socio-économiques, est principalement dû à la détection tardive des cas, comme d’autres auteurs ont démontré (Sizaire V et al., 2006). Le délai médian de deux mois entre le début des symptômes et le diagnostic observé dans notre étude explique la prédominance marquée des formes ulcérées au moment du diagnostic. En effet, la forme de la maladie est liée au délai à la consultation. Debacker et al. ont pu montrer, dans une étude se rapportant sur 1700 patients traités dans le centre de santé de Zagnanado au Bénin entre 1997 et 2001, que le délai médian à la consultation était de 30 à 46 jours pour les formes non-ulcérées (nodule, œdème et plaque) et de 60 jours pour les formes ulcérées (Debacker M et al., 2004). En plus, un délai fortement prolongé à se présenter à l’hôpital a été identifié comme l’un des facteurs de risque les plus importants pour l’atteinte osseuse. Ce délai était de 91 jours pour les atteintes osseuses (Debacker M et al., 2004). Dans le même centre, pour la période allant de 1996 à 2007, parmi les 930 patients d’UB confirmés et traités, 106 (11.4%)

255 présentaient une ostéomyélite causée par M.ulcerans. Le délai médian à la consultation était de 167 jours pour les patients avec atteintes osseuses et de 61 jours pour ceux qui présentaient uniquement des lésions cutanées (Portaels F et al., 2008). Dans la plupart des zones endémiques, la consultation à l’hôpital apparaît comme dernier recours des malades lorsque d’autres tentatives de recherche de soins ont échoué. La maladie a alors atteint un stade évolué avec des ulcérations cutanées étendues ou d’autres complications telles les contractures articulaires ou des ostéomyélites. Les raisons de cette consultation tardive sont complexes, à la fois sociales, culturelles et économiques (Meyers WM et al., 1974; Sizaire V et al., 2006). Stiendra et al. rapportent dans leur étude au Ghana sur les croyances et attitudes sur l’UB, que dans 59 % de cas la sorcellerie était mentionnée comme cause de la maladie. L’absence d’informations précises sur le mode de transmission, la présentation clinique de la maladie et l’absence de traitement médical efficaces renforçaient la croyance à la sorcellerie. Parmi ces patients interrogés 52% ont appliqué des herbes sur leurs lésions et ont consulté à l’hôpital en dernier ressort. Les raisons évoquées pour justifier cette situation étaient les difficultés financières (30% des patients), une autre raison était la peur du traitement à l’hôpital et en particulier l’amputation, enfin d’autres espéraient une cicatrisation spontanée (Stienstra Y et al., 2002). Le recours à l’auto-prise en charge ou au traitement traditionnel en premier ressort est très important dans la plupart des communautés affectées (Stienstra Y et al., 2002 ; Renzaho AM et al., 2007). Renzaho et al. ont montré que 71,8% de patients atteints d’UB consultaient premièrement un tradipraticien et allaient à l’hôpital en dernier ressort dans le district de Ga west au Ghana (Renzaho AM et al., 2007). Des observations similaires ont été également rapportées dans d’autres pays d’Afrique comme la République Démocratique du Congo (Kibadi K et al., 2004), le Cameroun (Noeske J et al., 2004 ; Peeters Grietens K et al., 2012) et le Bénin (Aujoulat I et al., 2003 ; Johnson RC et al., 2004 ; Mulder AA et al., 2008). Il est également possible que la nature indolore, la progression lente de la maladie et la mauvaise interprétation des

256 symptômes jouent également un rôle important pour ne pas consulter plus tôt les structures sanitaires (Asiedu K et al., 2000 ; Quek TYJ et al., 2007). Par conséquent, les patients se présentent souvent aux structures de santé aux stades avancés de la maladie avec comme corollaires, l’indication fréquente du traitement chirurgical, une longue durée d’hospitalisation, le développement des séquelles invalidantes et un coût exorbitant (Sizaire V et al., 2006).

Les rechutes après traitement sont une cause de souffrances supplémentaires pour les patients et leurs familles. En outre, ces rechutes gonflent encore les coûts médicaux et sapent la confiance des patients dans ce type de traitement. Le taux observé dans notre étude était très élevé par rapport à ce qui était antérieurement décrit dans la littérature. En effet, des taux variables de rechutes après traitement chirurgical exclusif ont été rapportés, allant de 6,1% au Bénin (Debacker M et al., 2005), 16% au Ghana (Amofah G et al., 1998) à 22% en RDC (Kibadi K, 2005). Le taux très bas décrit au Bénin pouvait s’expliquer par la détection précoce des patients suite aux séances de sensibilisation organisées par l’équipe du centre de santé de Zagnanado dans les villages, l’accès aisé au traitement et à la qualité du traitement chirurgical y pratiqué. Quoique la morbidité soit élevée, la mortalité liée à l’UB quant à elle est faible (Portaels F et al., 2000). Cependant, durant cette période une mortalité anormalement élevée a été observée dans notre étude. Les facteurs associés à cette mortalité étaient le sepsis, la malnutrition et l’anémie, la forme disséminée de la maladie, la cancérisation et le choc post-chirurgical. Plus du tiers de nos patients ont développé des limitations des mouvements articulaires. Dans la plupart des pays endémiques en Afrique, la proportion généralement décrite peut atteindre jusqu’à 20-25% du total du nombre des patients soignés dans les services de santé (Portaels F et al., 2000). Toutefois, une étude au Ghana avait montré que 58% des sujets traités présentaient un handicap à un degré plus ou moins fort (Ellen DE et al., 2003). Quant à la durée médiane d’hospitalisation de 102 jours observé dans notre étude, elle correspond à ce qui est décrit dans la littérature. En effet, dans les hôpitaux en Afrique,

257 la durée d’hospitalisation moyenne est d’environ trois mois et impose la même période la co-hospitalisation d’un parent non malade pour s’occuper du patient (Asiedu K and Etuaful S, 1998 ; Asiedu K et al., 2000). Il n’est pas étonnant que l’insolvabilité des patients atteints d’UB ait dépassé 90% dans notre hôpital durant cette période. Les conséquences économiques et sociales de l’UB sont donc énormes, mais insuffisamment étudiées (Asiedu K et al., 2000 ; Asiedu K and Etuaful S, 1998).

La forme œdémateuse étendue est considérée comme une forme sévère de l’UB (Portaels F et al., 2009). Sa prise en charge s’avère difficile, en particulier sur des localisations atypiques comme la face. Dans leur revue historique sur l’UB, Janssen et al. ont mentionné les difficultés diagnostiques et thérapeutiques rencontrées dans la prise en charge d’un œdème de paupière dû à l’UB chez un enfant de quatre ans diagnostiqué en 1950 à l’est de la RDC. La nécessité d’un traitement chirurgical conservateur a été mise en exergue (Janssen P et al., 2005). Cependant, peu d’études spécifiques ont été consacrées à ce sujet (Schierle HP et al., 2002 ; Ouattara D et al., 2003). Notre première expérience avec une forme œdémateuse à localisation faciale concerne également un enfant de 4 ans, d’origine Angolaise, diagnostiqué en 2002. Notre étude a montré qu’à cause de la rareté de cette localisation, le diagnostic ainsi que la mise en œuvre d’un traitement adéquat peut être substantiellement retardé. Elle a également suggéré l’emploi d’une méthode alternative moins invasive de prélèvement pour les formes pré-ulcératives. En effet, d’après notre expérience clinique, l’endroit où les biopsies ont été prélevées était devenu le point de départ d’une ulcération étendue. En outre, notre étude a pu montrer l’apport bénéfique de l’antibiothérapie spécifique dans la réduction de l’œdème et la limitation de l’étendue des excisions chirurgicales. Cette étude souligne l'importance de la formation des professionnels de santé en zones endémique pour le diagnostic précoce et une référence rapide des cas d’UB, y compris les localisations atypiques. Après notre première expérience sur la prise en charge des lésions faciales de l’UB (Phanzu DM et al., 2007), la quatrième étude a traité du même sujet, mais cette fois-ci sur une série des cas admis à l’HGR IME/Kimpese

258 durant une période de cinq années allant de 2003 à 2007. La fréquence de ces lésions faciales était de 5% (13 sur 238) parmi tous les cas rapportés et 7% (10/147) parmi les patients confirmés. Ces chiffres sont compatibles avec ce qui est rapporté dans la littérature. En effet, la revue de la littérature que nous avons réalisé, allant de 1966 à 2008, a montré que la fréquence de l’UB facial selon les pays variait de 0,8% à 16,7% (Phanzu DM et al., 2011). La réponse clinique à l’antibiothérapie spécifique était meilleure pour les lésions les plus précoces. L’étendue de l’œdème facial chez tous les patients où il était présent (9 sur 13) a été réduit durant l’antibiothérapie, minimisant ainsi l’étendue des excisions chirurgicales. Cela a permis d’éviter des graves défigurations. La lagophtalmie était la complication la plus fréquente, avec l’ectropion associé. Les autres complications observées étaient la perte de la vision, la défiguration avec exposition des dents et de la gencive, l’excision de la commissure labiale et la dissémination de la maladie. La prévention, le traitement et le suivi de ces complications impose une prise en charge multidisciplinaire (WHO, 2006). Il ressort de ce qui vient d’être développé ci-haut que l’éducation de la communauté pour une détection précoce de la maladie, l’antibiothérapie spécifique, la prévention des incapacités et la chirurgie reconstructive sont des composantes clés pour le contrôle de l’UB.

Pour l’instant, la définition de cas de l’UB reste essentiellement clinique, avec ou sans confirmation par le laboratoire (WHO, 2012). Il est généralement admis qu’en zone endémique le personnel médical expérimenté peut poser aisément le diagnostic de l’UB sur base clinique (Asiedu K et al., 2000 ; WHO 2004). Cependant, nous reconnaissons les limites du diagnostic clinique, lequel est fortement influencé par la disparité d’expérience des professionnels de santé sur terrain. En zones endémiques, suivant le stade clinique, l’UB peut être confondu avec plusieurs conditions pathologiques, dont les plus fréquentes sont le nodule onchocerquien, le kyste sébacé, le lipome, la lymphadénite ou la lymphadénopathie, la cellulite cutanée infectieuse, l’infection mycotique profonde, l’ulcère phagédénique tropical, la pyomyosite, la fasciite

259 nécrosante, le carcinome épidermoïde ulcéré (Andersen FO, 1965 ; Guédénon A and Portaels F, 2002 ; Sizaire V et al., 2006 ; Portaels F et al., 2001 ; van der Werf TS et al., 2005, Phanzu DM et al., 2010). Notre troisième publication porte sur le diagnostic différentiel de l’UB, dont l’importance est devenue cruciale depuis l’avènement de l’antibiothérapie spécifique. Elle a montré les difficultés qui peuvent être rencontrées, même dans une zone endémique connue, à diagnostiquer l’UB purement sur une base clinique et souligne l’importance d’examens complémentaires de laboratoire, y compris les analyses histopathologiques. En effet, plusieurs études sur la confirmation microbiologique des cas suspectés cliniquement ont pu montrer la difficulté d’en confirmer au-delà de 70% par les tests microbiologiques (Portaels F et al., 2001 ; Siegmund V et al., 2007). En Australie, les formes œdémateuses de l'infection à M. ulcerans sont souvent initialement diagnostiquées et traitées comme cellulite bactérienne. Cela conduit à des retards dans le diagnostic, à la progression de maladie, à une morbidité et une complexité accrues, et un coût exorbitant du traitement (Jenkin GA et al., 2002 ; Pak J et al., 2012 ; Trevillyan JM and Johnson PD, 2013 ; O’Brien DP et al., 2014). Dans toutes nos études cliniques, une place de choix a été réservée à la discussion du diagnostic différentiel. Parmi les patients non confirmés de notre première étude, les résultats histopathologiques ont révélé le lymphome malin non-Hodgkinien, le sarcome de Kaposi et des inflammations chroniques non-spécifiques (Phanzu DM et al., 2006). Discutant du diagnostic différentiel des formes œdémateuses de l’UB dans notre deuxième étude, les pathologies comme la cellulite cutanée infectieuse (actuellement dermo-hypodermite bactérienne), œdème couvrant une ostéomyélite, œdème traumatique, les maladies systémiques (insuffisance cardiaque, insuffisance rénale), la malnutrition, le sarcome de Kaposi et l’obstruction lymphatique ont été citées. Enfin, dans la quatrième étude, une adénite tuberculeuse a été évoquée face un patient ayant présenté un nodule au cou. Ainsi, nos études ont permis d’aider les professionnels de santé dans la région à prendre en charge de manière optimale la maladie sur base des connaissances exactes et de

260 l’expérience accumulée au fil des années. Un diagnostic correct est gage d’une lutte efficace contre l’UB. En effet, il permet d’éviter un gaspillage des ressources déjà limitées dans la plupart des zones endémiques.

Pour la partie santé publique, la première publication est une étude comparative avant-après qui décrit l’effet du projet de lutte contre l’UB trois années après sa mise en œuvre. Cette période a été marquée par un accroissement important du nombre des cas détectés et un changement substantiel du profil clinique des patients UB traités à l’HGR IME/Kimpese (Phanzu DM et al., 2011). La triple augmentation du nombre moyen des cas admis par an s’explique non seulement par la réduction partielle de la barrière financière, du fait de la gratuité des soins offerts aux malades, mais aussi par l’amélioration de la prise en charge et la qualité des résultats obtenus. Notre étude a en plus montré une augmentation substantielle au fil du temps de la proportion des patients guéris sans complications, une réduction marquée de la proportion des rechutes et une réduction notable de la mortalité. En Afrique, le défi à relever pour les professionnels de santé est de briser le cycle de mauvais résultats de traitement conduisant à une perte de confiance au traitement hospitalier dans les communautés touchées (Renzaho AM et al., 2007). Debacker et al. ont rapporté qu’au Centre sanitaire et nutritionnel Gbemoten, à Zagnanado, au Bénin, 68.3% de patients étaient référés par un ancien patient d’UB. L’amélioration de la qualité du traitement a conduit à une réduction de la durée médiane d’hospitalisation de 9 mois en 1989 à 1 mois en 2001. Le délai à la consultation avait chuté de 4 mois en 1989 à 1 mois en 2001 (Debacker M et al., 2004). Dans notre étude, la durée médiane d’hospitalisation est restée stationnaire autour de 90 jours. Il est cependant intéressant de noter que durant la deuxième période, elle était respectivement de 60 jours pour les patients de catégorie I, 81 jours pour ceux de la catégorie II et enfin 118 jours pour la catégorie III. Sur un total de 190 patients, 20% étaient de la catégorie I, 25% de la catégorie II et 55% de la catégorie III. Ces données montrent que pour des lésions simples, le traitement est plus simple et la guérison rapide.

261

Les raisons sociales, économiques et culturelles ci-haut mentionnées, limitant l’accès aux formations hospitalières en zones endémiques, suggéraient que le nombre des patients admis à l’hôpital IME/Kimpese pourrait représenter uniquement la partie émergée de l’iceberg. La gratuité des soins offerte aux patients par l’avènement du projet ne résout pas totalement le problème de barrières financières liées à la prise en charge de l’UB. En effet une étude menée au Cameroun dans deux hôpitaux avec un programme spécialisé centralisé contre l’UB, similaire au nôtre, a montré qu’en dépit des subventions du traitement, l’hospitalisation pour l’UB demeurait financièrement et socialement insoutenable pour les patients et leurs ménages, conduisant fréquemment à l’abandon de traitement ou complètement à un refus (Peeters Grietens K et al., 2008). La conclusion était de penser d’ores et déjà à des nouvelles stratégies de lutte qui soient à la fois socialement et financièrement acceptables et appropriées pour les communautés touchées. Nous avons utilisé pour notre évaluation un groupe contrôle historique, les patients ayant bénéficié de la mise en œuvre du projet de 2005 à 2007 ont été comparés à ceux admis au sein de la même institution de 2002 à 2004 avant le projet. Bien qu’une telle évaluation n’apporte pas une preuve scientifique définitive, à cause des difficultés de contrôler tous les confondants, elle est souvent considérée comme suffisante pour les décideurs. Elle nous donne un faisceau d’arguments qui à suffisance indiquent que les changements observés sont dus au programme (Habicht JP et al., 1999). La deuxième publication porte sur la première enquête exhaustive en RDC visant à déterminer la prévalence et la distribution de l’UB dans une région endémique bien circonscrite (Phanzu DM et al., 2013). Elle s’est réalisée dans le territoire de Songololo en juillet-août 2008 et a permis de déterminer l’ampleur de la maladie ainsi que sa distribution géographique. Elle a confirmé une distribution éparse et un caractère focalisé de la maladie tel que décrits dans d’autres études (Johnson RC et al., 2005 ; Porten K et al., 2009). Une étude réalisée en 2004 au Bénin, dans la commune de Lalo, avait confirmé le caractère très focalisé de la

262 distribution l’UB et a pu ainsi montrer que l’unité géographique pertinente d’analyse de l’UB était une entité plus réduite que le niveau national ou régional (Johnson RC et al., 2005). En outre, notre enquête a démontré une variation importante de la prévalence entre les aires de santé de deux ZSR du territoire, Kimpese et Nsona-Mpangu, ainsi qu’une sous- notification importante des cas d’UB dans le système de rapportage basé à l’hôpital. Malgré le fait que le nombre de cas admis à l’hôpital ne cessait d’augmenter les années précédentes, les résultats de l’enquête ont montré que la couverture de la population à risque était encore insuffisante. En Afrique, les premières actions de contrôle de l’UB ont été effectuées par des établissements spécialisés, essentiellement privés (Debacker M et al., 2004 ; Sopoh G et al. 2007 ; Peeters Grietens K et al., 2008 ; Phanzu DM et al., 2011). Le traitement de l’UB se déroulait en grande partie en milieu hospitalier (Asiedu K et al., 1998 ; Ackumey MM et al. 2011). Cette période a été caractérisée par la faible couverture géographique et thérapeutique de la population à risque et l'absence d'une estimation précise de la charge de morbidité (Johnson et al., 2005 ; van der Werf et al. 2005). En outre, les patients se présentaient à ces unités spécialisées à un stade avancé de la maladie et souffraient d’une incapacité fonctionnelle à long terme (WHO, 2014). Le taux de confirmation des patients détectés pendant l’enquête était faible. Nous pensons, toutefois, que nos résultats sont une traduction fiable de l’endémicité de l’UB dans le territoire de Songololo. En effet, les foyers connus hyperendémiques antérieurement, ont été confirmés par l’enquête, de même ceux connus moins ou non endémiques les ont été également.

Les données obtenues dans cette enquête ont contribué ces dernières années à mettre en œuvre des interventions de lutte mieux ciblées et servent de référence pour les évaluations actuelles du programme de contrôle. Dans une étude récente au Ghana, sur les services de santé pour le contrôle de l’UB, Ackumey et al. ont suggéré que pour un contrôle efficace de l’UB, particulièrement pour les systèmes de santé ruraux à ressources limitées, les cas documentés pourraient servir d'indices pour

263 classer les communautés dans trois catégories : zones d'endémie prioritaires, exigeant la plupart des interventions, endémiques et non endémiques, exigeant davantage de recherche pour améliorer la compréhension de la maladie (Ackumey et al., 2011). Une approche similaire est d’application à Songololo depuis 2010. Une liste de dix aires de santé endémiques prioritaires par zone de santé couverte par le projet a été créé et contribue à une affectation efficace des ressources.

A l’heure actuelle, il n’existe pas de vaccin spécifique pour la prévention primaire de l’UB (WHO, 2008 ; WHO, 2004). Bien que plusieurs études sur le vaccin BCG aient été effectuées, les résultats concernant l'efficacité du vaccin contre les maladies mycobactériennes, y compris l’UB diffèrent énormément. Les raisons évoquées pour expliquer la variabilité des résultats concernant l’efficacité du BCG sont entre autres les différences entre les vaccins utilisés, l’interaction des vaccins avec différentes mycobactéries de l’environnement, les différences régionales de souches de Mycobacterium tuberculosis, les différences nutritionnelles ou génétiques entre les différentes populations d’étude ou encore les différences dans la méthodologie des études elles-mêmes (Fine PE, 2001). Notre troisième étude avait pour objectif d’évaluer l’effet protecteur potentiel du vaccin BCG contre l’UB (Phanzu DM et al., 2014). Il s’agit d’une étude cas-témoins, réalisée dans trois pays endémiques, le Ghana, la RDC et le Togo, avec une grande population d’étude comprenant 401 cas d’UB confirmés et 826 témoins. C'est une des plus importantes études d'observation sur l'efficacité des vaccins BCG sur l'UB. Dans notre étude, nous n'avons pas mis en évidence un effet protecteur du vaccin BCG sur le risque de développer l’UB ou de développer des formes sévères de l’UB. Une limitation de notre étude était que les données exactes sur la vaccination par le BCG chez les participants à l'étude ne pouvaient être vérifiées par des documents, tels que des certificats de vaccination ou les registres de l'hôpital. Probablement une certaine proportion d'individus ont été définis comme « vaccinés », même si la cicatrice a été causée par autre chose qu'une vaccination par le BCG ("faux positif"). De l'autre côté, une certaine proportion aurait pu être également défini comme « pas

264 vaccinés », si aucune cicatrice n’ait été trouvée sur l'épaule ou la partie antérieure de l'avant-bras, parce que la vaccination par le BCG n’a pas conduit à une « cicatrice typique » ("faux négatif"). Un grand nombre de malclassés ("faux positif" et "faux négatif") ferait baisser le rapport de cotes observé, et nous ferait sousestimer la vraie force de l’association, un phénomène connu comme « non-differential misclassification ». Nous n’avons pas de moyen de vérifier dans quel mesure ce phénomène a joué. En outre, comme aucune donnée exacte sur les souches de BCG utilisées n'était disponibles, aucune conclusion définitive ne peut être tirée par rapport à un effet spécifique selon la souche. Aussi bien pour la tuberculose et la lèpre, que pour l’UB, des études prospectives ultérieures sur les différentes souches de vaccins BCG existantes sont nécessaires. En plus, la poursuite des recherches sur des nouveaux vaccins, sûrs et spécifiques contre l’UB est nécessaire.

Enfin, la quatrième et dernière étude a permis d’évaluer sur le plan clinique et épidémiologique, l’impact de la décentralisation du dépistage et du traitement de l’UB dans le territoire de Songololo, en comparaison au modèle précédent qui limitait la prise en charge des patients UB aux centres de référence. Depuis le lancement de la décentralisation des activités de lutte contre l’UB à Songololo, nous avons observé une forte augmentation dans la détection des cas. Le nombre d’aires de santé qui rapportent des cas a doublé. Nous avons remarqué un changement remarquable du profil de la catégorie de lésions avec la prépondérance de la catégorie I. En outre, les trois-quarts des patients détectés ont été traités dans les centres de santé de premier échelon et le taux de confirmation de laboratoire a été considérablement amélioré. Nous constatons effectivement que les progrès récents dans le diagnostic et le traitement de l'UB par un traitement antibiotique spécifique, combiné avec la détection précoce, sont susceptibles de changer l'orientation de soins principalement dans les services de santé polyvalents de premier échelon (WHO, 2004 ; Ackumey MM et al., 2011 ; Abass KM et al., 2014). Cependant, pour renforcer la détection précoce, l’information aux communautés touchées doit obligatoirement être couplée à une prise en

265 charge efficace des malades (Klis S et al., 2014). Notre étude a également montré que, compte tenu du caractère focalisé et d’une distribution éparse de la maladie, la détection précoce des cas d’UB sans une décentralisation de la prise en charge est un exercice futile dans la mise en œuvre d’une lutte efficace contre l’UB. Dans une récente étude sur le rôle joué par des perceptions traditionnelles sur les itinéraires thérapeutiques et le retard de consultation au Caméroun, Peeters Grietens et al. rapportent que l'attribution persistente de retard et le choix de traitement à des croyances constitue une simplification excessive de la situation et place le poids de la responsabilité directement sur les épaules des personnes souffrant de l’UB. Cela néglige potentiellement les autres facteurs déterminants qui sont l’efficacité du traitement, la place du traitement et la qualité de la relation personnel soignant-malade. Ils concluent qu’un traitement plus efficace dans le secteur biomédical est susceptible de réduire l’implication mystique perçue dans la maladie, la décentralisation supplémentaire pourrait constituer un élément clé pour réduire les retards et d'accroître l’observance du traitement biomédical (Peeters Grietens K et al., 2012). Le renforcement des structures sanitaires de première ligne revêt alors un caractère crucial, afin d’accroître l’accès à l’antibiothérapie et aux soins des plaies aux personnes touchées (Ackumey MM et al., 2011).

Comme dans la plupart des régions endémiques de l’UB, le véritable défi dans le territoire de Songololo est comment réduire davantage les séquelles incapacitantes, et donc la proportion des lésions ulcéreuses et de catégorie III au diagnostic, et maintenir ces efforts de surveillance (Sopoh GE et al., 2005 ; WHO, 2014). Il est également crucial d’accroître la proportion des patients confirmés par le laboratoire, afin d’éviter au maximum un traitement inadéquat d’autres pathologies. Pour ce faire, la formation et la remise à niveau régulières du personnel soignant est d’une grande importance.

Tout compte fait, les efforts de contrôle déployés ces dernières années dans le territoire de Songololo ont porté leurs fruits. L’intégration et la

266 décentralisation des activités de lutte dans les services de santé polyvalents est bel et bien une réalité. Le projet UB de l’IME/Kimpese est un projet pionnier en RDC, dont l’expérience devrait être partagée avec d’autres foyers endémiques du pays. A ce jour, en terme de détection de cas et de mise sous traitement, la RDC figure parmi les cinq pays qui rapportent la majorité des cas d’UB dans le monde (WHO, 2014). Nous espérons diminuer le fardeau d’UB en nombre de cas rapportés dans ce pays, tout en assurant une couverture maximale de la population à risque.

REFERENCES

Abass KM, van der Werf TS, Phillips RO, Sarfo FS, Abotsi J, Mireku SO, Thompson WN, Asiedu K, Stienstra Y, Klis SA. Short report: Buruli Ulcer Control in a Highly Endemic District in Ghana: Role of Community-Based Surveillance Volunteers. Am J Trop Med Hyg 2014; 14-0405.

Ackumey MM, Kwakye-Maclean C, Ampadu EO, de Savigny D, Weiss MG. Health Services for Buruli Ulcer Control: Lessons from a Field Study in Ghana. PLoS Negl Trop Dis 2011; 5: e1187.

Amofah G, Asamoah S, fram-Gyening C. Effectiveness of excision of preulcerative Buruli lesions in field situations in a rural district in Ghana. Trop Doct 1998; 28:81-3.

Andersen FO. Mycobacterial skin ulcers-Clinical experience. Cent Afr J Med 1965; 11:131-135.

Asiedu K, Etuaful S. Socioeconomic implications of Buruli ulcer in Ghana: a three-year review. Am J Trop Med Hyg 1998; 59:1015-22.

Asiedu K, Scherpbier R, Raviglione M. Buruli ulcer-Mycobacterium ulcerans infection. World Health Organization. 2000. WHO/CDS/CPE/GUI/2001.1.

Aujoulat I, Johnson C, Zinsou C, Guedenon A, Portaels F. Psychosocial aspects of health seeking behaviours of patients with Buruli ulcer in southern Benin. Trop Med Int Health 2003; 8:750-9.

267

Bafende AE, Phanzu MD, Imposo BB. Buruli ulcer in the Democratic Republic of Congo: epidemiology, presentation and outcome. Trop Doct 2004; 34:82-4.

Clancey JK, Dodge OG, Lunn HF, Oduori ML. Mycobacterial skin ulcers in Uganda. Lancet 1961; 2:951.

Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Guedenon A, et al. Mycobacterium ulcerans disease (Buruli ulcer) in rural hospital, Southern Benin, 1997-2001. Emerg Infect Dis 2004; 10:1391-8.

Debacker M, Aguiar J, Steunou C, Zinsou C, Meyers WM, Portaels F. Buruli ulcer recurrence, Benin. Emerg Infect Dis 2005; 11:584-9.

Ellen DE, Stienstra Y, Teelken MA, Dijkstra PU, van der Graaf WT, van der Werf TS. Assessment of functional limitations caused by Mycobacterium ulcerans infection: towards a Buruli ulcer functional limitation score. Trop Med Int Health 2003; 8:90–6.

Fine PE. BCG: the challenge continues. Scand J Infect Dis 2001; 33: 243- 245.

Habicht J.P., Victoria C.G., Vaughan J.P. Evaluation designs for adequacy, plausibility and probability of public health programme performance and impact. Int J Epid 1999; 28:10-18.

Janssens P, Pattyn S, Meyers WM, Portaels F. Buruli Ulcer: An Historical Overview with Updating to 2005. Bull Séanc Acad R Sci Outre-Mer 2005; 51:265-299.

Jenkin GA, Smith M, Fairley M, Johnson PD. Acute, oedematous Mycobacterium ulcerans infection in a farmer from far north Queensland. Med J Aust 2002; 176: 180–181.

Johnson RC, Makoutode M, Hougnihin R, Guedenon A, Ifebe D, Boko M, et al. [Traditional treatment for Buruli ulcer in Benin]. Med Trop 2004; 64:145-50.

268

Johnson RC, Sopoh GE, Boko M, Zinsou C, Gbovi J, Makoutode M, et al. [Distribution of Mycobacterium ulcerans (Buruli ulcer) in the district of Lalo in Benin.]. Trop Med Int Health 2005; 10:863-71.

Kibadi K [Mycobacterium ulcerans disease (Buruli ulcer): surgical treatment of 102 cases in the Democratic Republic of Congo]. Med Trop 2005; 65:444-8.

Kibadi K, Tsakala M, Mputu-Yamba JB, Muyembe T, Kashongwe M, Imposso B, Nsiala A. [Buruli ulcer in Angolese refugees in the Kimpese area, Lower Congo, D.R. Congo]. Santé 2003; 13:39-41.

Kibadi Kapay A. Knowledge-attitudes-practices survey of the Songololo population (D.R. Congo) about Buruli ulcers. Bull Soc Pathol Exot 2004; 97:302-5.

Klis S, Ranchor A, Phillips RO, Abass KM, Tuah W, Loth S, Velding K, van der Werf TS, Stienstra Y. Good Quality of Life in Former Buruli Ulcer Patients with Small Lesions: Long-Term Follow-up of the BURULICO Trial. PLoS Negl Trop Dis 2014; 8: e2964.

Meyers WM, Connor D.H., McCullough B., Bourland J., Moris R., Proos L. Distribution of Mycobacterium ulcerans infection in Zaïre, including the report of new foci. Ann Soc belge Med Trop 1974; 54:147-157.

Mulder AA, Boerma RP, Barogui Y, Zinsou C, Johnson RC, Gbovi J, et al. Healthcare seeking behaviour for Buruli ulcer in Benin: a model to capture therapy choice of patients and healthy community members. Trans R Soc Trop Med Hyg 2008; 102:912-20.

269

Noeske J, Kuaban C, Rondini S, Sorlin P, Ciaffi L, Mbuagbaw J, et al. Buruli ulcer disease in Cameroon rediscovered. Am J Trop Med Hyg 2004; 70:520-6.

O’Brien DP, Friedman ND, McDonald A, Callan P, Hughes A, Athan E. Clinical Features and Risk Factors of Oedematous Mycobacterium ulcerans lesions in an Australian Population: Beware Cellulitis in an Endemic Area. PLoS Negl Trop Dis 2014; 8: e2612.

Ouattara D, Aka GK, Meningaud JP, Sica A, Kaba L, Gadegbeku S. Les localisations facials de l’ulcère de Buruli céphalique: à propos de deux cas. Rev Stomatol Chir Maxillofac 2003 ; 104 :231-234.

Pak J, O’Brien DP, Quek TY, Athan E. Treatment costs of Mycobacterium ulcerans in the antibiotic era. International Health 2012; 4: 123–127.

Peeters Grietens K, Toomer E, Um Boock A, Hausmann-Muela S, Peeters H, Kanobana K, Gryseels C, Ribera JM. What Role Do Traditional Beliefs Play in Treatment Seeking and Delay for Buruli Ulcer Disease?–Insights from a Mixed Methods Study in Cameroon. PLoS ONE 2012; 7: e36954.

Perquis P, Muret G, Ravisse P, Maydat L. Ulcères tropicaux à mycobactéries. A propos de 8 observations. Med Trop 1968 ; 26 : 642- 648.

Phanzu DM, Ablordey A, Imposo DB, Lefevre L, Mahema RL, Suykerbuyk P, et al. Short report: edematous Mycobacterium ulcerans infection (Buruli ulcer) on the face: a case report. Am J Trop Med Hyg 2007;7 7:1099-102.

Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, Nsiangana SZ, Singa JN, Meyers WM, Suykerbuyk P, Portaels F. Mycobacterium ulcerans

270 disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg 2006; 75:311-4.

Phanzu DM, Suykerbuyk P, Imposo DB, Lukanu PN, Minuku JB, Lehman LF, Saunderson P, de Jong BC, Lutumba PT, Portaels F, Boelaert M. Effect of a control project on clinical profiles and outcomes in Buruli ulcer: a before/after study in Bas-Congo, Democratic Republic of Congo. PLoS Negl Trop Dis 2011; 5:e1402.

Phanzu DM, Suykerbuyk P, Saunderson P, Lukanu PN, Minuku JBM, Imposo DBB, Diengidi BM, Kayinua M, Muyembe JJT, Lutumba PT, de Jong BC, Portaels F, Boelaert M. Burden of Mycobacterium ulcerans disease (Buruli ulcer) and underreporting ratio in the Territory of Songololo, Democratic Republic of Congo. PLoS Negl Trop Dis 2013; 7: e2563.

Phanzu MD, Bafende AE, Imposo BBD, Meyers W.M., Portaels F. Undertreated necrotizing fasciitis masquerading as ulcerated edemetous Mycobacterium ulcerans infection (Buruli ulcer). Am J Trop Med Hyg 2010; 82:478-481.

Phanzu MD, Mahema RL, Suykerbuyk P, Imposo BD, Lehman LF, Nduwamahoro E, Meyers WM, Boelaert M, Portaels F. Mycobacterium ulcerans infection (Buruli ulcer) on the face : A comparative analysis of 13 clinically suspected cases from the Democratic Republic of Congo. Am J Trop Med Hyg 2011; 85:1100-105.

Phillips RO, Phanzu DM, Beissner M, Badziklou K, Luzolo EK, Sarfo FS, Halatoko WA, Amoako Y, Frimpong M, Kabiru AM, Piten E, Maman I, Bidjada B, Koba A, Awoussi KS, Kobara B, Nitschke J, Wiedemann FX, Kere AB, Adjei O, Löscher T, Fleischer B, Bretzel G, Herbinger KH. Effectiveness of routine BCG vaccination on Buruli Ulcer Disease: a case- control study in the DR Congo, Ghana and Togo. PLoS Negl Trop Dis 2014 SUBMISSION PNTD-D-14-01436.

Portaels F, Guédénon A. Buruli ulcer (Mycobacterium ulcerans infection) Diagnostic guide for health professionals, 2002.

271

Portaels F, Johnson P, Meyers WM. Ulcère de Buruli-Diagnostic de l'infection à Mycobacterium ulcerans. World Health Organisation. WHO/CDS/CPE/GBUI/2001.4.

Portaels F, Johnson RC, Aguiar J, Meyers WM, Debacker M. Etude de 106 cas d'ulcères de Buruli avec atteintes osseuses traités à Zagnanado, Bénin. Bull de l'ALLF 2008; 23:48-50.

Portaels F, Silva MT, Meyers WM. Buruli ulcer. Clin Dermatol 2009; 27:291-305.

Portaels F. Epidémiologie des ulcères à Mycobacterium ulcerans. Ann Soc belge Med Trop 1989; 69:91-103.

Porten K, Sailor K, Comte E, Njikap A, Sobry A, et al. Prevalence of Buruli Ulcer in Akonolinga Health District, Cameroon: Results of a Cross Sectional Survey. PLoS Negl Trop Dis 2009 ; 3: e466.

Queck TYJ, Henry MJ, Pasco JA, O’Brien DP, Johnson PDR, Hughes A, Cheng AC, Redden-Hoare J, Athan E. Mycobacterium ulcerans infection: factors influencing diagnostic delay. Med J Aust 2007; 187: 561-63.

Ravisse P. Ulcère cutané à Mycobacterium ulcerans au Cameroun. Etude clinique, épidémiologique et histopathologique. Bull Soc Pathol Exot 1977 ; 70 : 109-124.

Schierle HP, Lemperle G, Erdmann D. The Buruly type ulcer. Plast Reconstr Surg 2002 109: 2608.

272 laboratory methods available for the diagnosis of Buruli ulcer disease . Clin Infect Dis 2007; 45: 68 – 75 .

Sizaire V, Nackers F, Comte E, Portaels F. Mycobacterium ulcerans infection: control, diagnosis, and treatment. Lancet Infect Dis 2006; 6:288-96.

Smith J.H. Epidemiologic observations on cases of Buruli ulcer seen in a hospital in the lower Congo. Am J Trop Med Hyg 1970; 19:657-663.

Sopoh GE, Johnson RC, Chauty A, Dossou AD, Aguiar J, Salmon O, et al. Buruli ulcer surveillance, Benin, 2003-2005. Emerg Infect Dis 2007; 13:1374-6.

Stienstra Y, van der Graaf WT, Asamoa K, van der Werf TS. Beliefs and attitudes toward Buruli ulcer in Ghana. Am J Trop Med Hyg 2002; 67:207- 13.

Trevillyan JM, Johnson PD. Steroids control paradoxical worsening of Mycobacterium ulcerans infection following initiation of antibiotic therapy. Med J Aust 2013; 198: 443–444.

Van der Werf TS, Stienstra Y, Johnson RC, Phillips R, Adjei O, Fleischer B, Wansbrough-Jones MH, Johnson PDR, Portaels F, van der Graaft WTA, Asiedu K. Mycobacterium ulcerans disease. Bull World Health Organ 2005; 83:785-791.

Van der Werf TS, Van der Graaf NTS, Groothuis DG, Knell AJ. Mycobacterium ulcerans infection in Ashanti Region, Ghana. Trans R Soc Trop Med Hyg 1989; 83:410-441.

World Health Organization. Provisional guidance on the role of specific antibiotics in the management of Mycobacterium ulcerans infection (Buruli ulcer). 2004. WHO/CDS/CPE/GBUI/2004.10.

World Health Organization. Treatment of Mycobacterium ulcerans disease (Buruli ulcer)-Guidance for healthworkers. 2012. WHO/HTM/NTD/IDM/2012.1.

273

World Health Organization. Buruli ulcer- Mycobacterium ulcerans Infection. Fact sheet n°199, July 2014 ( available at http://www.who.int/mediacentre/factsheets/fs199/fr/; accessed on November 17th, 2014).

World Health Organization. Buruli ulcer: Prevention of Disability. Lehman L, Simonet V, Saunderson P, Agbenorku P, eds. 2006 Geneva: World health Organization.

World Health Organization. Buruli ulcer: progress report, 2004-2008. Wkly Epidemiol Rec 2008; 83:145-54.

274

CHAPITRE VI. CONCLUSION ET PERSPECTIVES D’AVENIR

275

276

L’UB présente un polymorphisme qui ne rend pas toujours le diagnostic clinique aisé, même pour des professionnels de santé expérimenté. Son diagnostic différentiel est très large et se discute avec toutes les différentes formes de la maladie. Le présent travail a contribué à une meilleure connaissance de la maladie et en même temps a attiré plus d’attention sur les difficultés diagnostiques et thérapeutiques qui peuvent être rencontrées lors de sa prise en charge. Le personnel de santé oeuvrant en zones endémiques, et même non endémiques, devrait en être davantage informé pour une prise en charge optimale des personnes touchées. Le traitement de l’UB peut être simple en cas de détection rapide ou compliqué et coûteux en présence d’un patient à un stade avancé. A ce jour, le diagnostic et le traitement précoces sont les seuls moyens pour réduire au maximum la morbidité et éviter les incapacités.

La mise en œuvre d’un programme spécialisé de prise en charge de l’UB offrant la gratuité des soins aux patients peut être efficace dans l’amélioration du profil clinique des malades et des résultats du traitement offert. En dépit de la réduction du coût pour le traitement et des aides supplémentaires accordées aux patients durant leur hospitalisation, la couverture géographique et thérapeutique est restée limitée.

L’enquête de prévalence a montré une large variation entre les aires de santé et une sous-notification importante du système antérieur qui limitait les soins aux centres de référence. Compte tenu du caractère focalisé et d’une distribution éparse de la maladie, la détection précoce des cas d’UB sans une décentralisation de la prise en charge est une pierre d’achoppement pour une mise en œuvre efficace de la lutte contre l’UB.

Le rationnel et la motivation pour l’intégration et la décentralisation des activités de lutte dans les soins de santé primaires à Songololo étaient d’améliorer le dépistage précoce, la prise en charge des cas, l’observance du traitement et l’accessibilité aux services.

277

Et intégration ne signifie pas pour nous que le programme vertical disparaisse, ni que le personnel spécialisé n’a plus de rôle à jouer, bien au contraire. En effet, Les compétences multidisciplinaires acquises par l’équipe du projet UB au cours des dernières années en matière du diagnostic clinique, du traitement médical et chirurgical de l’UB, de la confirmation bactériologique, des soins de plaies et de la POD, devront continuer à servir pour l’organisation des programmes d’actions périphériques dans les communautés, la supervision et le renforcement des capacités des professionnels de santé œuvrant en périphérie. Cela nécessite, néanmoins, une étroite collaboration avec les zones de santé endémiques.

Recommandations et perspectives - Poursuivre avec l’intégration et la décentralisation des activités de lutte contre l’UB dans les services de santé polyvalents de base - Intensifier l’éducation, l’information et la communication destinée au grand public, en particulier dans les communautés affectées, afin de promouvoir la détection précoce de la maladie - Renforcer davantage les services de santé polyvalents dans les zones endémiques, afin d’améliorer l’intégration des activités de lutte. - Poursuivre avec les enquêtes de prévalence dans les zones de santé endémiques couvertes par le programme, afin de vérifier la tendance à la baisse de la détection de cas constatée ces dernières années. - Partager l’expérience accumulée sur la prise en charge de l’UB et la gestion du programme de lutte avec d’autres foyers endémiques connus de la province du Bas-Congo et du reste du pays. - Mettre sur pied une base de données disponible avec géoréférences de tous les villages du territoire de Songololo. Cela permettra aux responsables locaux et aux chercheurs de suivre la dynamique de l’évolution de l’UB dans la région par la production des cartes précises de distribution au niveau «village ».

278

- Renforcer nos participations à différentes recherches sur l’UB: la mise au point d’un traitement antibiotique par voie orale, la mise au point d’un test de diagnostic rapide à faire sur le lieu de soins, le mode de transmission.

279

280

REMERCIEMENTS

Au moment où je finis la rédaction de cette thèse, je réalise plus que jamais que ce travail est le fruit d’une collaboration enrichissante entre plusieurs institutions à travers le monde.

Je voudrais en premier lieu rendre des actions de grâce à Dieu Tout- Puissant, l’auteur de toute vie, qui crée en nous le vouloir et le faire.

Mes remerciements les plus sincères vont ensuite aux promoteurs de cette thèse.

Je tiens à remercier chaleureusement le Professeur Marleen Boelaert, qui a cru en moi, qui m’a ouvert les portes de la famille de l’unité d’épidémiologie et de contrôle de maladies et qui m’a permis d’arriver au bout de cette thèse. Je ne cesserai de la remercier pour son encadrement, sa rigueur scientifique tout au long de ma formation doctorale et sa marque de générosité à mon égard.

Je dis grand merci au Professeur Françoise Portaels, un des précurseurs dans la lutte contre l’ulcère de Buruli, qui a guidé mes premiers pas dans la recherche. Je n’oublierai jamais son attention, son souci permanent du travail bien accompli et la bienveillance dont elle a toujours fait preuve à ma personne. Voilà onze ans maintenant depuis qu’elle m’accueillit dans la famille de l’unité de Mycobactériologie. Je garde encore le petit mot qu’elle m’écrivit à la fin de mon stage : « A Delphin, ce bateau pour lui souhaiter un très bon voyage, un merveilleux avenir. Nous avons tous apprécié ta présence parmi nous. Tu es maintenant mon fils et tu es le frère de tous les membres du laboratoire. Merci pour ta gentillesse, ton sens de l’humour et de ta compétence scientifique ! Je te souhaite tout le bonheur que tu mérites. Que Dieu te garde ainsi que tous les tiens ! Et n’oublie jamais…le chocolat dans la voiture… ». C’est bien vrai, les paroles s’envolent, mais les écrits restent. Je n’ai jamais oublié, merci encore !

281

J’adresse toute ma gratitude au Professeur Greet Ieven et au Professeur Pascal Lutumba pour tous vos conseils et votre encadrement tout au long de mon parcours de recherche doctorale.

Je remercie vivement tous les distingués membres du jury pour avoir accepté sans hésitation d’examiner notre travail.

Je tiens à exprimer mes plus vifs remerciements au Docteur Wayne M. Meyers, un des pionniers, pour sa précieuse contribution à l’organisation de la lutte contre l’ulcère de Buruli dans la Province du Bas-Congo et à la réalisation de ce travail. En effet, ton travail a été une grande source d’inspiration pour moi. J’ose espérer que nous avons apporté quelque chose sur ce que vous avez réalisé il y a plus de quarante ans maintenant.

Notre profonde gratitude va aux organismes ayant financé ce travail, dont la Direction Générale de Coopération au Développement du Royaume de Belgique, American Leprosy Missions et la Commission Européenne.

J’exprime toute ma reconnaissance à la Direction de l’IMT Anvers pour avoir accepté notre candidature et nous avoir permis de travailler dans les conditions les meilleurs.

Je tiens également à remercier les différents partenaires des consortiums de recherche BURULICO (FP6-Project No. INCO-CT-2005-05-051476) et BURULIVAC (FP7-Project No. 241500) et l’Initiative Mondiale de lutte contre l’Ulcère de Buruli (IMUB) pour la collaboration fructueuse.

Au Professeur Jean-Jacques Muyembe Tamfum, Directeur de l’Institut National de Recherche Biomédicale (INRB), je dis grand merci, pour avoir soutenu ma candidature, pour votre grand esprit d’ouverture et pour vos conseils précieux. Merci au staff de l’INRB, en particulier du laboratoire d’ulcère de Buruli.

Je remercie vivement la Direction de l’Institut Médical Evangélique de Kimpese pour m’avoir permis de me consacrer à cette juste cause. Un vibrant merci au Directeur Général Docteur Théophile Luzaïsu Zioka et au

282

Médecin-Directeur Docteur Roger Mahema Lutabu, pour votre soutien et votre encadrement.

J’exprime tous mes remerciements à l’ensemble du staff de l’Unité de Mycobactériologie et de l’Unité d’épidémiologie et contrôle de maladies. Merci pour vos encouragements, vos conseils et le climat favorable d’apprentissage lors de nos nombreux séjours à l’IMT-Anvers. Un grand merci fleuri pour Bouke de Jong, Miriam Eddyani, Krista Fissette, Karin Janssens, Ciska Maeckelbergh, Arabella Huys, Anne Marie Trooskens, Greet Verhulst et Evelien Paessens, pour tout ce que vous avez fait pour moi.

Je tiens à remercier vivement Helga Bodges et tout son staff du service des étudiants, pour votre serviabilité combien remarquable et votre disponibilité.

Je remercie le département de la reprographie de l’Université d’Anvers et de l’IMT-Anvers, en particulier Jean-Pierre Wenseleers, pour votre aide et pour le travail de qualité réalisé.

Mes remerciements s’adressent au Docteur Eric Bafende Aombe, pour vos conseils et votre soutien indéfectible.

Je dis merci à toute l’équipe du projet UB de l’IME/Kimpese, aux médecins chef de zone, aux superviseurs et à tous les acteurs de terrain dans les zones de santé de Kimpese et Nsona-Mpangu pour la collaboration, le travail abattu dans les conditions difficiles toutes ces années. Votre contribution est tout simplement inestimable.

Je remercie sincèrement la famille Pierre Mbombolo pour son encadrement et son accompagnement durant toutes ces années.

Je remercie particulièrement Patrick Suykerbuyk, mon ami et frère, mon compagnon de lutte, pour toutes ces années de dur labeur sur terrain et pour ton soutien sans faille. Je n’oublierai pas ta chère épouse Ann, pour son hospitalité et son soutien.

283

Je tiens à remercier tous mes collègues et amis pour votre soutien et vos encouragements. Je citerai en particulier Symphorien Dimfumu Muanamundele, Adrien Nsiala Kumbi, Patrick Mitashi, Michel Kaswa, Alain Mpanya, Christian Johnson, Ghislain Sopoh, Dissou Affolabi, Koen Vandelannoote, Peter Ndjadi Yela, Hélène Bonnet, Cécile Menard et Jules Sabwa.

J’adresse toute ma gratitude à la famille Jean-louis Dikondja Dingondo, pour son hospitalité et son soutien. Toutes ces années, vous m’avez toujours chaleureusement accueilli et offert un cadre serein chez vous.

Grand merci à vous, Nanou Mbayi Ngangasa, Oliveira Benico Masiala, pour votre hospitalité, votre serviabilité et votre amitié.

Je remercie toute ma famille et toute ma belle-famille. A mes frères et sœurs "les Phanzu" que j’aime de tout mon cœur, trouvez ici l’expression de toute ma reconnaissance. Mes sincères remerciements au Pasteur Innocent Phanzu et toute sa famille en Belgique, à toute la famille Vincent Kimpioka en France, à la famille Lady Ndompetelo en Suisse, pour tout ce que vous avez fait pour moi durant toutes ces années.

A vous tous, que je n’ai pas pu citer nommément ici, qui avez contribué à la réalisation de ce travail, par votre soutien tant moral, spirituel, matériel que financier, trouvez ici l’expression de toute notre gratitude.

Mes remerciements vont enfin à mon épouse, l’amour de ma vie, Godelive Phanzu, pour son soutien indéfectible, son amour, sa tendresse et la compréhension dont elle a toujours fait preuve à mon égard.

284

CURRICULUM VITAE

PHANZU, Delphin Mavinga

Personal information

Date and place of birth: June 21, 1971; Kinshasa, DR Congo

Nationality: Congolese

Marital status: married to Godelive Masolokele Mambi, five children

Home address: 0048B, Camp Missionnaire, IME/Kimpese, City of Kimpese, Province of Bas-Congo, DR Congo Tel.: +243 819481410 +243 898534350

Professional address: Institut Medicale Evangélique (IME), Matadi Road, Kimpese, Province of Bas-Congo, DR Congo PO Box: 68 Kimpese E-mail:[email protected] /[email protected]

Languages

- Lingala, Kikongo: native languages - French: official language, fluent - English

285

Education and postgraduate training

1.October 1989- December 1999: Study of Medicine, University of Kinshasa (Medical Doctor).

2. June 2001- March 2004: 3 years Resident Medical Doctors Programme, IME/Kimpese, DR Congo / “Family medicine and primary health care”, MEDUNSA, Southern Africa (External student). Research topic: “what factors contribute towards the late presentation of Buruli ulcer patients at IME/Kimpese hospital?”

3. September 2008- July 2009: Master in Public Health, Institute of Tropical Medicine, Antwerp, Belgium.

4. Since 2009, doctoral study, Institute of Tropical Medicine/Antwerp University, Antwerp, Belgium.

Positions

1. June 2001–March 2004: Resident Medical Doctor, IME/Kimpese Hospital (General medicine).

2. April 2004- November 2005 : Orthopedics & traumatology, IME/Kimpese Hospital.

3. December 2005- March 2009: Coordinator of BURULICO/DRC Research Project (Buruli ulcer : multidisciplinary research for improvement of control in Africa) funded by the European Commission (FP6/ Project No. INCO-CT-2005-05-051476-BURULICO).

4. March 2010- May 2013: Coordinator of BURULIVAC/DRC Research Project (Identification and development of vaccine candidates for Buruli ulcer disease) funded by the European Commission (FP7/ Project No. 241500).

286

5. Since July 2013: Principal Investigator of SOFI/DRC Research Project (Global spread and transmission dynamics of Mycobacterium ulcerans) funded by the Department of Economy, sciences and innovation of the Flamish Government, Belgium.

6. Since August 2013: Coordinator of IME/Kimpese Buruli ulcer Project/Community, mainly funded by American Leprosy Missions (ALM).

Participation in international meetings, training, research

1. September- December 2004: Training in Mycobacteriology, ITM/Antwerp, Belgium.

2. June 2005 : Co-investigator in the research project « Epidemiology and control of Shistosomiasis in DR Congo of Today » (phase I) funded by TDR-WHO.

3. December 2005 : International Colloquium on TB and Buruli ulcer organised by ITM/Antwerp, Cotonou, Benin.

4. December 2005- March 2009: Research Project (Buruli ulcer : multidisciplinary research for improvement of control in Africa, FP6/ Project No. INCO-CT-2005-05-051476-BURULICO). 5. March 2006 : WHO Annual meeting on Buruli ulcer, Geneva, Switzerland.

6. December 2006 : Co-investigator in the research project « Epidemiology and control of Shistosomiasis in DR Congo of Today » (phase II) funded by TDR-WHO.

7. April 2007 : WHO Annual meeting on Buruli ulcer, Geneva, Switzerland.

8. September 2007 : International course on M.ulcerans Microbiology (M2U), Centre Pasteur, Yaoundé, Cameroon.

287

9. Mars 31- April 2, 2008 : WHO Annual meeting on Buruli ulcer, Geneva, Switzerland.

10. March 22-24, 2010 : WHO Annual meeting on Buruli ulcer, Geneva, Switzerland.

11. March 2010- May 2013: Research Project (Identification and development of vaccine candidates for Buruli ulcer disease, FP7/ Project No. 241500-BURULIVAC). 12. March 28-30, 2011 : WHO Annual meeting on Buruli ulcer, Geneva, Switzerland.

13. February 20-26, 2011 : Training in « Laboratoire de Référence des Mycobactérie », Cotonou, Benin.

14. July 8-10, 2011: ISID-NTD Congress on Infectious Diseases, Boston, Massachusetts, USA.

15. November 1st-4, 2011 : Colloquium on Zoonoses and Neglected Infectious Diseases of Africa, Johannesburg, South Africa.

16. June 13-16, 2012 : 15th International Congress on Infectious Diseases , Bangkok, Thaïlande.

17. March 25-27, 2013 : WHO Annual meeting on Buruli ulcer, Geneva, Switzerland.

18. Since July 2013: Research Project (Global spread and transmission dynamics of Mycobacterium ulcerans) funded by the Department of Economy, sciences and innovation of the Flamish Government, Belgium. 19. January 19-21, 2014: 18th National Conference of Wounds and Wound healing, Paris, France.

20. April 2-5, 2014: 16th International Congress on Infectious Diseases, Cape Town, South Africa.

21. June 16-20, 2014: Conference on the laboratory diagnosis of Buruli ulcer via the detection of Mycolactone A/B, the causative agent of the

288

disease, Kishi Laboratory, Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA, USA.

22. July 28- August 2, 2014: “Stop Buruli consortium” workshop on diagnostics and wound management in Buruli ulcer, Limbé, Cameroon.

23. November 24-27, 2014: ITM Colloquium “The Human Factor-Social sciences in global health research”, ITM Antwerp, Belgium.

24. March 23-25, 2015 : WHO meeting on Buruli ulcer, Geneva, Switzerland.

Memberships of scientific societies

- Member of International Society of Infectious Diseases since 2011. - Member of French and French-speaking Society of Wounds and Wound Healing since January 2014 (SFFPC in French). - Member of the WHO working group on Epidemiology & Surveillance of Buruli ulcer.

Awards

- September 17th, 2009: Provincial Prize for Development Cooperation 2009 granted by the Executive Board of the Province Council of Antwerp.

Publications

1. Bafende AE, Phanzu MD ,Imposo BB. (2004) Buruli ulcer in the Democratic Republic of Congo: epidemiology, presentation and outcome. Trop Doc 34: 82-84.

2. Kibadi K, Muyembe T, Phanzu MD, Mbala L, Meyers WM, Portaels F. (2004) Association large ulcère de Buruli thoraco-abdominal et

289

pleurésie massive homolatérale : aspects diagnostiques et difficultés de prise en charge. A propos d’un cas. Med Afr Noire 51 : 643-648.

3. Phanzu DM, Bafende EA, Dunda BK, Imposo DB, Kibadi AK, Nsiangana SZ, Singa JN, Meyers WM, Suykerbuyk P, Portaels F. (2006) Mycobacterium ulcerans disease (Buruli ulcer) in a rural hospital in Bas-Congo, Democratic Republic of Congo, 2002-2004. Am J Trop Med Hyg 75 : 311-314.

4. Phanzu MD, Ablordey A, Imposo BD, Lefevre L, Mahema RL, and Suykerbuyk P, Meyers WM, Portaels F. (2007) Short report: Edematous Mycobacterium ulcerans infection (Buruli ulcer) on the Face: A case report. Amer J Trop Med Hyg 77: 1099-1102.

5. Suykerbuyk P, Wambacq J, Phanzu DM, Haruna H, Nakazawa Y, Ooms K, Kamango K, Stragier P, Singa JN, Ekwanzala F, De Herdt E, De Maeyer P, Kestens L, Portaels F. (2009) Persistence of Mycobacterium ulcerans Disease (Buruli Ulcer) in the Historical Focus of Kasongo Territory, the Democratic Republic of Congo. Am J Trop Med Hyg 81: 888-894.

6. Huygen K, Adjei O, Affolabi D, Bretzel G, Demangel C, Fleischer B, Johnson RC, Pedrosa J, Phanzu DM, Phillips RO, Pluschke G, Siegmund V, Singh M, van der Werf TS, Wansbrough-Jones M, Portaels F. (2009) Buruli ulcer disease: prospects for a vaccine. Med Microbiol Immunol 198:68-77.

7. Phanzu MD, Bafende AE, Imposo BBD, Meyers WM, Portaels F. (2010) Under-treated necrotizing fasciitis masquerading as ulcerated edemetous Mycobacterium ulcerans infection (Buruli ulcer): A case report. Am J Trop Med Hyg 82: 478-481.

8. Sopoh GE, Johnson RC, Anagonou SY, Barogui YT, Dossou AD, Houézo JD, Phanzu DM, Tente BH, Meyers WM, Portaels F. (2011) Buruli Ulcer Prevalence and Altitude, Benin. Emerg Inf Dis 17: 153- 154.

290

9. Phanzu MD, Mahema RL, Suykerbuyk P, Imposo BD, Lehman LF, Nduwamahoro E, Meyers WM, Boelaert M, Portaels F. (2011) Mycobacterium ulcerans infection (Buruli ulcer) on the face : A comparative analysis of 13 clinically suspected cases from the Democratic Republic of Congo. Am J Trop Med Hyg 85: 1100-1105.

10. Phanzu DM, Suykerbuyk P, Imposo DBB, Lukanu PN, Minuku J-BM, et al. (2011) Effect of a Control Project on Clinical Profiles and Outcomes in Buruli Ulcer: a Before/After Study in Bas-Congo, Democratic Republic of Congo. PLoS Negl Trop Dis 5(12): e1402. doi: 10.1371/journal.pntd.0001402.

11. Phanzu DM, Suykerbuyk P, Saunderson P, Lukanu PN, Minuku J-BM, et al. (2013) Burden of Mycobacterium ulcerans (Buruli Ulcer) and underreporting ratio in the Territory of Songololo, Democratic Republic of Congo. PLoS Negl Trop Dis 7(12): e2563. doi:10.1371/journal.pntd.0002563.

12. Vandelannoote K, Jordaens K, Bomans P, Leirs H, Durnez L, Affolabi D, Sopoh G, Aguiar J, Phanzu DM, Kibadi K, Eyangoh S, Manou LB, Phillips RO, Adjei O,Ablordey A, Rigouts L, Portaels F, Eddyani M, de Jong BC. (2014) Insertion sequence element single nucleotide polymorphism typing provides insights into the population structure and evolution of Mycobacterium ulcerans across Africa. Appl Environ Microbiol. 80(3):1197-209. doi: 10.1128/AEM.02774-13. Epub 2013 Dec 2.

13. Phillips RO, Phanzu DM, Beissner M, Badziklou K, Luzolo EK, et al. (2015) Effectiveness of routine BCG vaccination on Buruli Ulcer Disease: A case-control study in the Democratic Republic of Congo, Ghana and Togo. PLoS Negl Trop Dis 9(1): e3457. doi:10.1371/journal.pntd.0003457.

Antwerp, March 28th, 2015 Dr. Delphin Mavinga PHANZU

291