Prognostic Value of Skin Manifestations of Infective Endocarditis

Research

Original Investigation Prognostic Value of Skin Manifestations of Infective Endocarditis

Amandine Servy, MD; Laurence Valeyrie-Allanore, MD; François Alla, MD, PhD; Catherine Lechiche, MD; Pierre Nazeyrollas, MD, PhD; Christian Chidiac, MD, PhD; Bruno Hoen, MD, PhD; Olivier Chosidow, MD, PhD; Xavier Duval, MD, PhD; for the Association Pour l’Etude et la Prévention de l’Endocardite Infectieuse Study Group

Supplemental content at IMPORTANCE Infective endocarditis (IE) is a rare disease with poor prognosis. When IE is jamadermatology.com suspected, skin examination is mandatory to look for a portal of entry and classic skin lesions to help diagnose and manage the condition.

OBJECTIVES To describe the prevalence of and factors associated with dermatological manifestations in patients with definite IE.

DESIGN Observational, prospective, population-based epidemiological study between January 1 and December 31, 2008. Subsequently, collected dermatological data were subjected to post hoc analysis.

SETTING AND PARTICIPANTS Patients (n = 497) diagnosed in 7 French regions and hospitalized in France for definite IE satisfying modified Duke criteria.

MAIN OUTCOMES AND MEASURES Patient and disease epidemiological information was collected, focusing on the most classic dermatological manifestations of IE (Osler nodes, Janeway lesions, purpura, and conjunctival hemorrhages). Disease outcome was also recorded.

RESULTS Among 497 definite IE cases, 487 had known dermatological status. Of 487 cases, 58 (11.9%) had skin manifestations, including 39 (8.0%) with purpura, 13 (2.7%) with Osler nodes, 8 (1.6%) with Janeway lesions, and 3 (0.6%) with conjunctival hemorrhages (5 patients had 2 skin manifestations). Patients with skin manifestations had a higher rate of IE-related extracardiac complications than patients without skin manifestations, particularly cerebral emboli (32.8% vs 18.4%, P = .01), without increased mortality. Patients with purpura had larger cardiac vegetations (18.1 vs 13.7 mm, P = .01), and Janeway lesions were associated with more extracerebral emboli (75.0% vs 31.8%, P = .02).

CONCLUSIONS AND RELEVANCE Specific skin manifestations of IE are associated with a higher risk of complications and should alert physicians to examine for extracardiac complications, notably with cerebral imaging.

Author Affiliations: Author affiliations are listed at the end of this article. Group Information: The Association Pour l’Etude et la Prévention de l’Endocardite Infectieuse Study Group investigators are listed at the end of this article. Corresponding Author: Amandine Servy, MD, Department of Dermatology, Centre Hospitalier Universitaire Henri-Mondor Hospital, Assistance Publique–Hôpitaux de Paris, 51 Avenue du Maréchal-de-Lattre-de-Tassigny, JAMA Dermatol. 2014;150(5):494-500. doi:10.1001/jamadermatol.2013.8727 94010 Créteil CEDEX, France Published online February 5, 2014. ([email protected]).

494 jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Skin Manifestations of Infective Endocarditis Original Investigation Research

nfective endocarditis (IE) affects 2 to 6 individuals per which remains unclear. Therefore, using data from a large French 100 000 person-years in industrialized countries. Inpa- general population–based epidemiological study, we analyzed I tient mortality approaches 20%, particularly due to septic the prevalence of the 4 main dermatological manifestations of shock, heart failure, or stroke.1 The epidemiological profile of IE (Osler nodes, Janeway lesions, purpura, and conjunctival IE has evolved progressively, with increasing percentages of hemorrhages) and their possible associated factors. health care–associated IE, Staphylococcus aureus involve- ment, and prosthetic and degenerative native valve IE.2-4 A complete dermatological physical examination is essen- Methods tial for the management of IE. Indeed, a cutaneous portal of entry is found for more than 20% of the cases,5 and dermato- Study Design logical lesions suggestive of septicemia (Osler nodes, Janeway The study was approved by an institutional review board lesions, purpura, and splinter and conjunctival hemor- (Comité de Protection des Personnes, Besançon, France, De- rhages) may help physicians reach a clinical diagnosis. Nev- cember 2007). In accord with French law at that time, pa- ertheless, it can be difficult to discriminate among these clas- tients were informed of the study orally but did not have to sic dermatological lesions (Figure) and nonspecific provide written individual consent. descriptions. Distinguishing between Janeway lesions and Os- Between January 1 and December 31, 2008, the Associa- ler nodes is not that obvious.6 Osler nodes are described as tion Pour l’Etude et la Prévention de l’Endocardite Infec- purple painful nodes, mainly localized on fingertips, pulp of tieuse Study Group17,18 conducted an exhaustive observa- the toes, palms, soles, or sometimes on the ears.6 Lesions dis- tional and prospective analysis of all IE cases occurring in 7 appear within a few days without sequelae, occasionally in a French regions (Franche-Comté, Ille-et-Vilaine, Languedoc- few hours. Janeway lesions are often small, nontender, ery- Roussillon, Lorraine, Marne, Île-de-France, and Rhône-Alpes).17 thematous, and painless macules or papules on the palms or The 15 million inhabitants of these regions account for 32% of soles, sometimes purpuric or hemorrhagic.5,7 They tend to last the French population (http://www.insee.fr/fr/themes days to weeks before healing completely. Histological find- /document.asp?ref_id=ip1220). Only definite IE cases satisfy- ings for both lesions, described in case reports,8-15 include sep- ing Duke criteria revised by Li et al19 (eTable in the Supple- tic emboli with inflammatory reactions, and culture of le- ment) among patients 18 years or older and living in one of the sions can grow pathologic microorganisms. Vascular purpura, study regions were considered for this analysis. usually necrotic, is more frequent and typical. Purpura is most All hospital physicians usually managing IE (ie, cardiolo- often localized on lower parts of the body (legs and back) and gists [P.N. and X.D.], cardiac surgeons, infectious disease spe- sometimes on mucosa (conjunctivae and mouth) or near the cialists [C.L., C.C., and B.H.], intensivists, and internists) were clavicles. The findings on histological examination of pur- aware of this epidemiological study and notified all their pa- pura have been shown to be primarily septic emboli or leuko- tients. Then, Association Pour l’Etude et la Prévention de cytoclastic vasculitis.16 Splinter (longitudinal striations along l’Endocardite Infectieuse investigators transferred collected data the major nail axis) and conjunctival hemorrhages are also ob- from medical records to standardized case report forms. The fol- served in IE but are not pathognomonic of it. lowing information was collected for each patient: sex, age, Despite the importance of the skin for positive and etio- medical history, procedures and at-risk factors for IE, clinical logical IE diagnosis, few nondermatological investigations have signs and symptoms, portal of entry, laboratory and microbi- analyzed the manifestations of cutaneous IE, the prevalence of ology test results, imaging findings, medical and surgical treat-

Figure. Classic Lesions Associated With Infective Endocarditis

A B C

A, Osler nodes on the right thumb characterized by a painful distal the sole corresponding to a Janeway lesion. From the collection of the erythematous and hemorrhagic bullous lesion. B, Extensive distal infiltrated Department of Dermatology, Centre Hospitalier Universitaire Henri-Mondor purpura evolving to necrosis of the legs. C, Erythematous purpuric macules of Hospital, Assistance Publique–Hôpitaux de Paris, Créteil, France.

jamadermatology.com JAMA Dermatology May 2014 Volume 150, Number 5 495

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Research Original Investigation Skin Manifestations of Infective Endocarditis

ment, and complications. A dermatological examination was not though transthoracic echocardiography diagnosed IE in al- systematically performed. Among clinical data, the following most half of the patients, transesophageal echocardiography 4 classic dermatological manifestations of IE were diagnosed was almost always performed to confirm the diagnosis and to according to each physician’s personal experience with IE and identify heart complications. recorded: Osler nodes, Janeway lesions, purpura, and conjunc- Time to IE diagnosis was similar regardless of dermato- tival hemorrhages. All case report forms were checked by an ad- logical manifestation status (P = .90), usually less than 1 month. judication committee composed of an infectious disease spe- However, compared with patients having IE without skin le- cialist, a cardiologist, a microbiologist, and a cardiac surgeon sions, those with dermatological manifestations had signifi- and were classified according to modified Duke criteria. cantly more extracardiac complications (75.9% vs 54.8%, Then, one of us (A.S.) examined case report forms to ex- P = .002), and their manifestations were associated with a tract information for a post hoc analysis of collected derma- higher rate of cerebral complications (41.4% vs 25.4%, P = .01), tological data. Concordance between data collection and medi- mainly cerebral emboli (32.8% vs 18.4%, P = .01), while only a cal records was checked for 10.0% of patients with IE having trend was found for a subgroup with symptomatic cerebral skin lesions. complications (Table 2). Compared with patients having IE without cerebral emboli, those with this complication had sig- Statistical Analysis nificantly higher percentages of IE-associated skin signs All statistical analyses were performed using commercially (Table 3). Despite significantly higher complication rates for available software (SAS, version 8.02; SAS Institute Inc). P <.05 patients with dermatological manifestations of IE, mortality defined significance. The prevalences of dermatological mani- and hospital length of stay were comparable to those for pa- festations of definite IE cases were calculated. After elimina- tients without dermatological manifestations (Table 2). tion of 10 patients with unknown dermatological status, fac- Considering patients with each type of cutaneous lesion tors associated with cutaneous lesions were subjected to individually vs those without, purpura was significantly as- univariate analysis. First, we compared all definite IE cases ac- sociated with larger initial cardiac vegetations (18.1 vs 13.7 mm, cording to the presence or absence of at least 1 of 4 cutaneous P = .01). Janeway lesion was significantly associated with ex- IE lesions considered. Then, we performed another analysis tracerebral emboli (75.0% vs 31.8%, P = .02) and often with with each dermatological manifestation individually vs all symptomatic emboli in lung, spleen, heart, kidney, liver, or pe- other patients having IE in the cohort without it. Quantitative ripheral localizations. Bicuspid aortic valve was significantly variables, expressed as means (SDs) with 95% CIs, were ana- associated with Osler node (23.0% vs 4.0%, P = .01). No other lyzed using t test or Wilcoxon rank sum test. Qualitative vari- characteristics were associated with the presence of each type ables, expressed as numbers (percentages), were compared of cutaneous lesion (data not shown). using χ2 test or Fisher exact test for all comparisons. Concordance was good (80.0%) between case report forms and medical records. Although some symptomatological or dif- ferential diagnoses can be discussed, no patient was ex- Results cluded because of incomplete or imprecise medical records. Prevalence of Dermatological Manifestations of IE Among 497 definite IE cases, 487 had known dermatological Discussion status (presence or absence of a classic IE-associated cutaneous lesion). Four hundred twenty-nine patients (88.1%) had no Based on the Association Pour l’Etude et la Prévention de skin manifestation. Fifty-eight patients (11.9%) had at least 1 l’Endocardite Infectieuse prospective population study of pa- of 4 skin manifestations of IE considered, including 39 (8.0%) tients with definite IE, 11.9% had dermatological manifesta- with purpura, 13 (2.7%) with Osler nodes, 8 (1.6%) with Janeway tions. According to the literature,20 the frequency of cutane- lesions, and 3 (0.6%) with conjunctival hemorrhages. Five pa- ous lesions varies widely (5%-25% of IE cases) across tients had 2 skin manifestations. investigations. In a 1973 review article21 covering 1927 to 1967, the frequencies were 10% to 23% of patients with IE having Os- Factors Associated With Dermatological Manifestations of IE ler nodes22,23 and 19% to 40% of patients with IE having The main characteristics of patients with IE according to the purpura.24,25 These manifestations seem to be reported less fre- presence or absence of dermatological manifestations are sum- quently (approximately 10% of cases) in more recent studies.5,7 marized in Table 1. Patients with cutaneous lesions were The prevalence rates may be underestimated in the absence younger, while no differences were observed for the frequen- of systematic dermatological examination and photographs for cies of intravenous drug use or associated comorbidities ex- retrospective assessment. We found the IE dermatological cept for a lower percentage of patients with type 1 diabetes manifestations to be significantly associated with sympto- mellitus among patients with skin manifestations. In particu- matic and asymptomatic cerebral embolic events. Conse- lar, the frequency of previously known underlying heart dis- quently, the presence of these cutaneous lesions might attest ease, including a prosthetic valve, did not differ significantly to an active embolic process responsible for systemic compli- between patients with vs without skin manifestations. Com- cations and thus a sign of IE severity, which is in agreement parable numbers of patients with or without skin signs had with reported septic emboli in Osler nodes, Janeway lesions, heart murmurs (new or previously known) at IE diagnosis. Al- and purpura.8-15 Moreover, we observed the largest cardiac

496 JAMA Dermatology May 2014 Volume 150, Number 5 jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Skin Manifestations of Infective Endocarditis Original Investigation Research

Table 1. Characteristics of 487 Patients Having Definite Infective Endocarditis (IE) With or Without Dermatological Manifestationsa

Dermatological Manifestations of IE None ≥1 P IE-Associated Characteristic (n = 429) (n = 58) Valueb Characteristics of Patients Age, mean (SD), y 63.1 (15.5) 57.0 (17.3) .01 Male-female ratio 320:109 42:16 .73 Intravenous drug users, No. (%) 22 (5.1) 6 (10.3) .13 Comorbidities, No. (%) ≥1 210 (49.0) 23 (39.7) .18 Hypertension 207 (48.3) 21 (36.2) .09 Diabetes mellitus 103 (24.0) 8 (13.8) .08 Type 1 diabetes mellitus 53 (12.4) 2 (3.4) .04 Heart disease, No. (%) Underlying, previously known 201 (46.9) 29 (50.0) .65 Prosthetic valve 94 (21.9) 10 (17.2) .65 Congenital heart disease 18 (4.2) 3 (5.2) .65 Native valve disease 84 (19.6) 16 (27.6) .16 Previous IE 27 (6.3) 5 (8.6) .57 Intracardiac devices 60 (14.0) 4 (6.9) .13 Microbiological Characteristics Source of acquisition, No. (%) .14 Community 307 (71.6) 47 (81.0) Nosocomial 99 (23.1) 9 (15.5) Health care related 14 (3.3) 0 At-risk procedures, No. (%) Dental 34 (7.9) 2 (3.4) .29 Indwelling central catheter 30 (7.0) 5 (8.6) .60 Cardiac catheter 12 (2.8) 0 .38 Digestive 19 (4.4) 4 (6.9) .51 Urinary 6 (1.4) 1 (1.7) .60 Gynecological 0 1 (1.7) .12 Portal of entry, No. (%) Cutaneous 74 (17.2) 10 (17.2) .97 Active intravenous drug use 22 (5.1) 6 (10.3) .13 Buprenorphine hydrochloride injection 15 (3.5) 5 (8.6) .08 Microorganisms, No. (%) .56 Streptococcaceae 206 (48.0) 31 (53.4) .44 Streptococci 154 (35.9) 23 (39.7) .58 Staphylococcaceae 154 (35.9) 21 (36.2) .96 Staphylococcus aureus 108 (25.2) 20 (34.5) .13 Other microorganism 37 (8.6) 3 (5.2) .61 >1 Microorganism 7 (1.6) 2 (3.4) .29 No pathogen identified 25 (5.8) 1 (1.7) .36 Time to IE diagnosis, No. (%) First symptom to hospitalization interval .90 Appeared after admission 26 (6.1) 3 (5.2) <1 mo 288 (67.1) 38 (65.5) 1-3 mo 74 (17.2) 10 (17.2) >3 mo 33 (7.7) 6 (10.3) Heart murmur at diagnosis, No./total No. (%) Yes 238/375 (63.5) 36/54 (66.7) .65 New murmur 180/375 (48.0) 30/54 (55.6) .30 Previous murmur unchanged 63/375 (16.8) 8/54 (14.8) .71 Previous murmur modified 41/375 (10.9) 5/54 (9.3) .71

(continued)

jamadermatology.com JAMA Dermatology May 2014 Volume 150, Number 5 497

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Research Original Investigation Skin Manifestations of Infective Endocarditis

Table 1. Characteristics of 487 Patients Having Definite Infective Endocarditis (IE) With or Without Abbreviations: TEE, transesophageal Dermatological Manifestationsa (continued) echocardiography; TTE, transthoracic echocardiography. Dermatological Manifestations of IE a Osler node, Janeway lesion, None ≥1 P purpura, or conjunctival IE-Associated Characteristic (n = 429) (n = 58) Valueb hemorrhage. Echocardiographic IE findings, No./total No. (%) b χ2 Test or Fisher exact test for TTE 392 (91.4) 54 (93.1) .80 qualitative variables; t test or TEE 376 (87.6) 51 (87.9) .95 Wilcoxon rank sum test for TTE alone diagnosed IE 174/393 (44.3) 24/53 (45.3) .89 quantitative variables. c Echocardiographic abnormality 396 (92.3) 55 (94.8) .79 The mean (SD) vegetation sizes were 14.4 (11.0) mm for patients without Vegetationc 371 (86.5) 55 (94.8) .08 dermatological manifestations Perforation 80 (18.6) 8 (13.8) .37 (n = 297) and 15.8 (9.1) mm for Cardiac abscess 70 (16.3) 10 (17.2) .85 patients with at least 1 dermatological manifestation (n = 44) (P = .19). Dehiscenced 16/94 (17.0) 3/10 (30.0) .39 d Exclusively for IE developing on Severe regurgitation 161 (37.5) 28 (48.3) .13 prosthetic valves.

Table 2. Complications and Outcomes of 487 Patients Having Definite Infective Endocarditis (IE) With or Without Dermatological Manifestationsa

Dermatological Manifestations of IE None ≥1 Complication or Outcome of IE (n = 429) (n = 58) P Valuea Heart failure, No. (%) 143 (33.3) 20 (34.5) .86 Extracardiac complications, No. (%) 235 (54.8) 44 (75.9) .002 Septic shock 22 (5.1) 6 (10.3) .13 Cerebral complications 109 (25.4) 24 (41.4) .01 Symptomatic cerebral complication 81 (18.9) 17 (29.3) .06 Occurred before antibiotics 43 (10.0) 7 (12.1) .63 Cerebral embolism 79 (18.4) 19 (32.8) .01 Cerebral hemorrhage 24 (5.6) 3 (5.2) >.99 Extracerebral embolism 129 (30.1) 30 (51.7) .001 Symptomatic extracerebral embolism 61 (14.2) 13 (22.4) .07 Occurred before antibiotics 31 (7.2) 7 (12.1) .20 Vascular phenomena 188 (43.8) 39 (67.2) .01 Immunological phenomena 34 (7.9) 24 (41.4) <.001 Serum creatinine level ≥2.0 mg/dL 118 (27.5) 21 (36.2) .20 SI conversion factor: To convert Cardiac surgery 190 (44.3) 27 (46.6) .75 creatinine level to micromoles per liter, multiply by 88.4. In-hospital mortality, No. (%) 98 (22.8) 13 (22.4) .94 a Osler node, Janeway lesion, Hospital length of stay, mean (SD), d 54.3 (48.5) 57.1 (50.5) .73 purpura, or conjunctival (n = 57) hemorrhage.

Table 3. Characteristics of 487 Patients Having Definite Infective Endocarditis (IE) With or Without Cerebral Emboli

Cerebral Emboli, No. (%) Symptomatic Cerebral Emboli, No. (%) No Yes No Yes IE-Associated Characteristic (n = 385) (n = 102) P Valuea (n = 414) (n = 73) P Valuea Cutaneous IE lesions ≥1 Lesion 39 (10.1) 19 (18.6) .02 45 (10.9) 13 (17.8) .09 Conjunctival hemorrhage 2 (0.5) 1 (1.0) .50 3 (0.7) 0 >.99 Purpura 27 (7.0) 12 (11.8) .11 31 (7.5) 8 (11.0) .30 Osler node 8 (2.1) 5 (4.9) .16 9 (2.2) 4 (5.5) .11 Janeway lesion 6 (1.6) 2 (2.0) .67 6 (1.4) 2 (2.7) .34 Microorganisms .16 .05 Streptococcaceae 187 (48.6) 50 (49.0) .94 204 (49.3) 33 (45.2) .52 Staphylococcaceae 138 (35.8) 37 (36.3) .94 143 (34.5) 32 (43.8) .13

a χ2 Test or Fisher exact test.

498 JAMA Dermatology May 2014 Volume 150, Number 5 jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Skin Manifestations of Infective Endocarditis Original Investigation Research

vebetations in patients with purpura, and other authors have be performed to assist in therapeutic management and to im- reported vegetation size to be associated with cerebral prove our knowledge. emboli.20,26,27 Janeway lesions were directly associated with This study has some limitations. Because of the exploratory embolic events other than cerebral emboli, and Osler nodes nature of the associated factors sought, collected dermatologi- were associated with bicuspid aortic valves. cal data were subjected only to post hoc analysis, and system- Embolic events are one of the most frequent (20%-50% of atic dermatological physical and histological examinations were cases20) and severe complications of IE. Their incidence is the not conducted, thereby limiting the certainty of the diagnosis of highest during the first 2 weeks after beginning an antibiotic dermatological lesions and their characterization. Consequently, regimen.20,27,28 Approximately 30% of emboli are silent,29 but the prevalence of dermatological manifestations of IE could dif- they may affect prognosis. The identification of patients with fer slightly.Nevertheless, the absence of systematic dermatologi- the highest frequency of cerebral emboli may justify cerebral cal examination reflects real clinical practice and enhances the magnetic resonance imaging and affect patient care.29 Physi- pertinence of our findings concerning factors associated with cu- cians should be aware of this association, which might iden- taneous manifestations of IE. Moreover, differentiating between tify new predictive factors. Early systematic dermatological ex- dermatological lesions is often complex, and it was important amination, preferably by a dermatologist or an experienced to simplify the main message and clinical practice that any clas- physician alert to these skin manifestations and their appear- sic manifestation of IE is a sign of disease severity. In addition, ance, could shorten the time to reach a specific diagnosis, even Osler nodes and Janeway lesions are so similar clinically and his- for small and transient lesions. tologically that their distinction is debatable. No clear relationship has been established between the on- set of specific IE dermatological lesions and a responsible mi- 30 croorganism. As did Arber et al, we observed fewer coagulase- Conclusions negative staphylococci infections in patients having IE with skin signs. Although Martínez-Marcos et al31 recently re- Classic skin manifestations of IE are signs of infection sever- ported that dermatological manifestations occurred fre- ity, associated with a higher risk of complications that should quently in patients with enterococcal IE, few studies8,9,11,15,32 alert physicians to look for secondary complications, notably have reported the presence of S aureus in needle aspirate or with cerebral imaging. In addition to their diagnostic contri- biopsy cultures of Osler nodes or Janeway lesions and blood. bution, dermatological manifestations may have a prognos- Microbiological cultures of classic dermatological lesions can tic role in IE and influence therapeutic decisions.

ARTICLE INFORMATION and Alla contributed equally to this work, and Drs investigators: B. Hoen and X. Duval. Other Accepted for Publication: September 14, 2013. Chosidow and Duval contributed equally to this members: F. Alla, A. Bouvet, S. Briançon, E. work. Cambau, M. Celard, C. Chirouze, N. Danchin, T. Published Online: February 5, 2014. Study concept and design: Lechiche, Hoen, Alla, Doco-Lecompte, F. Delahaye, J. Etienne, B. Iung, V. doi:10.1001/jamadermatol.2013.8727. Chosidow, Duval. Le Moing, J. F. Obadia, C. Leport, C. Poyart, M. Author Affiliations: Department of Dermatology, Analysis and interpretation of data: Servy, Revest, C. Selton-Suty, C. Strady, P. Tattevin, and F. Centre Hospitalier Universitaire (CHU) Valeyrie-Allanore, Alla, Chosidow, Duval. Vandenesch. Region study coordinating Henri-Mondor Hospital, Assistance Drafting of the manuscript: Servy, Valeyrie-Allanore. investigators: Y. Bernard, S. Chocron, C. Chirouze, B. Publique–Hôpitaux de Paris (AP-HP), Créteil, France Critical revision of the manuscript for important Hoen, P. Plesiat, I. Abouliatim, C. De Place, P. (Servy, Valeyrie-Allanore, Chosidow); Centre intellectual content: Alla, Nazeyrollas, Chidiac, Tattevin, M. Revest, P. Y Donnio, F. Alla, J. P. d’Investigation Clinique–Epidémiologie Clinique, Hoen, Chosidow, Duval. Carteaux, T. Doco-Lecompte, C. Lion, N. Aissa, C. CHU de Nancy, Nancy, France (Alla); Department of Statistical analysis: Alla. Selton-Suty, B. Baehrel, R. Jaussaud, P. Nazeyrollas, Infectious and Tropical Diseases, CHU de Administrative, technical, or material support: Servy. C. Strady, V. Vernet, E. Cambau, X. Duval, B. Iung, P. Caremeau, Nîmes, France (Lechiche); Departments Study supervision: Chosidow, Duval. Nataf, C. Chidiac, M. Celard, F. Delahaye, J. F. of Cardiology and Therapeutics, CHU de Reims, Conflict of Interest Disclosures: None reported. Obadia, F. Vandenesch, H. Aumaître, J. M. Frappier, Faculté de Médecine, Reims, France (Nazeyrollas); V. Le Moing, E. Oziol, A. Sotto, and C. Sportouch. Department of Infectious and Tropical Diseases, Funding/Support: This study was supported by a Centre National de Référence des Streptocoques: C. CHU Hôpital de la Croix Rousse, Lyon, France research grant from the French Ministry of Health Poyart and A. Bouvet. Centre National de Référence (Chidiac); Department of Infectious and Tropical (Programme Hospitalier de Recherche Clinique des Staphylocoques: F. Vandenesch, M. Celard, and Diseases, CHU de Besançon, Unité Mixte de 2007), by grants from the Société Française de M. Bes. Investigators: P. Abassade, E. Abrial, C. Acar, Recherche Centre Nationale de Recherche Cardiologie, by the European Society of Clinical N. Aissa, J. F. Alexandra, N. Amireche, D. Amrein, P. Scientifique 6249 Chrono-Environnement, Microbiology and Infectious Diseases, and by André, M. Appriou, M. A. Arnould, P. Assayag, A. Université de Franche-Comté, Besançon, France Novartis France. The sponsor was the Délégation à Atoui, F. Aziza, N. Baille, N. Bajolle, P. Battistella, S. (Hoen); Université Paris–Est Créteil Val-de-Marne, la Recherche Clinique et au Développement, Centre Baumard, A. Ben Ali, J. Bertrand, S. Bialek, M. Bois Institut National de la Santé et de la Récherche Hospitalier Universitaire de Besançon. Grosse, M. Boixados, F. Borlot, A. Bouchachi, O. Médicale (INSERM), Centre d’Investigation Clinique Role of the Sponsor: The funding source had no Bouche, S. Bouchemal, J. L. Bourdon, A. Bouvet, L. (CIC) 006, Créteil, France (Chosidow); INSERM CIC role in the design and conduct of the study; Brasme, F. Bricaire, E. Brochet, J. F. Bruntz, A. Cady, 007, AP-HP, CHU Bichat, INSERM Unité 738, collection, management, analysis, or interpretation J. Cailhol, M. P. Caplan, B. Carette, J. P. Carteaux, O. Université Paris Diderot, Unité de Formation et de of the data; preparation, review, or approval of the Cartry, C. Cazorla, M. Celard, H. Chamagne, H. Recherche de Médecine, Site Bichat, Paris, France manuscript; and decision to submit the manuscript Champagne, G. Chanques, J. Chastre, B. Chevalier, (Duval). for publication. C. Chirouze, F. Chometon, C. Christophe, A. Cohen, Author Contributions: Drs Alla and Duval had full Group Information: The Association Pour l’Etude N. Colin de Verdiere, N. Danchin, V. Daneluzzi, L. access to all the data in the study and take et la Prévention de l’Endocardite Infectieuse Study David, P. De Lentdecker, F. Delahaye, V. Delcey, P. responsibility for the integrity of the data and the Group (AEPEI) investigators were as follows: AEPEI Deleuze, E. Donal, X. Duval, B. Deroure, V. accuracy of the data analysis. Drs Valeyrie-Allanore Study Group on Infective Endocarditis: Principal Descotes-Genon, K. Didier Petit, A. Dinh, V. Doat, F. Duchene, F. Duhoux, M. Dupont, S. Ederhy, O.

jamadermatology.com JAMA Dermatology May 2014 Volume 150, Number 5 499

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Research Original Investigation Skin Manifestations of Infective Endocarditis

Epaulard, M. Evest, J. F. Faucher, B. Fantin, E. 6. Farrior JB, Silverman ME. A consideration of the Infectious Diseases (ESCMID) and the International Fauveau, T. Ferry, M. Fillod, T. Floch, T. Fraisse, J. M. differences between a Janeway’s lesion and an Society of Chemotherapy (ISC) for Infection and Frapier, L. Freysz, B. Fumery, B. Gachot, S. Gallien, I. Osler’s node in infectious endocarditis. Chest. Cancer. Guidelines on the prevention, diagnosis, Gandjbach, P. Garcon, A. Gaubert, J.L. Genoud, S. 1976;70(2):239-243. and treatment of infective endocarditis (new Ghiglione, C. Godreuil, A. Grentzinger, L. Groben, D. 7. Murdoch DR, Corey GR, Hoen B, et al; version 2009. Eur Heart J. 2009;30(19):2369- Gherissi, P. Guéret, A. Hagege, N. Hammoudi, F. International Collaboration on Endocarditis– 2413. Heliot, P. Henry, S. Herson, B. Hoen, P. Houriez, L. Prospective Cohort Study (ICE-PCS) Investigators. 21. Weinstein L, Rubin RH. Infective endocarditis: Hustache-Mathieu, O. Huttin, S. Imbert, B. Iung, S. Clinical presentation, etiology, and outcome of 1973. Prog Cardiovasc Dis. 1973;16(3):239-274. Jaureguiberry, M. Kaaki, A. Konate, J. M. Kuhn, S. infective endocarditis in the 21st century. Arch 22. Geraci JE, Martin WJ. Antibiotic therapy of Kural Menasche, A. Lafitte, B. Lafon, F. Lanternier, Intern Med. 2009;169(5):463-473. V. Le Chenault, V. Le Moing, C. Lechiche, S. Lefevre, bacterial endocarditis, VI: subacute enterococcal Thibaut, A. Lefort, A. Leguerrier, J. Lemoine, L. 8. Espinosa Parra FJ, Ramos Rincón JM, Herrero endocarditis: clinical, pathologic and therapeutic Lepage, C. Leport, C. Lepousé, J. Leroy, P. Lesprit, L. Huerta F, Pretel Serrano L, Lorenzo AA. Diagnostic consideration of 33 cases. Circulation. Letranchant, D. Loisance, G. Loncar, C. Lorentz, P. utility of Osler’s nodules in infectious endocarditis 1954;10(2):173-194. Mabo, I. Magnin-Poull, T. May, A. Makinson, H. Man, among parenteral drug users [in Spanish]. An Med 23. Newman W, Torres JH, Guck JK. Bacterial M. Mansouri, O. Marçon, J. P. Maroni, V. Masse, F. Interna. 2002;19(6):299-301. endocarditis; an analysis of fifty-two cases. Am J Maurier, M. C. Meyohas, P. L. Michel, C. Michelet, F. 9. Alpert JS, Krous HF, Dalen JE, O’Rourke RA, Med. 1954;16(4):535-542. Mechaï, O. Merceron, D. Messika-Zeitoun, Z. Bloor CM. Pathogenesis of Osler’s nodes. Ann Intern 24. Kaipainen WJ, Seppala K. Endocarditis lenta; a Metref, V. Meyssonnier, C. Mezher, S. Micheli, M. Med. 1976;85(4):471-473. review of 118 patients treated during the ten-year Monsigny, S. Mouly, B. Mourvillier, O. Nallet, P. 10. Lin CS, Lee RL, Kuo CM, Lee LC. Positive culture period 1945 to 1954. Acta Med Scand. Nataf, P. Nazeyrollas, V. Noel, J. F. Obadia, E. Oziol, of the Janeway lesion in acute bacterial 1956;155(1):71-81. T. Papo, B. Payet, A. Pelletier, P. Perez, J. S. Petit, F. endocarditis. Taiwan Yi Xue Hui Za Zhi. 25. Vogler WR, Dorney ER, Bridges HA. Bacterial Philippart, E. Piet, C. Plainvert, B. Popovic, J. M. 1980;79(1):99-102. Porte, P. Pradier, R. Ramadan, M. Revest, J. endocarditis. Am J Med. 1962;32:910-921. Richemond, M. Rodermann, M. Roncato, I. Roigt, O. 11. Vinson RP, Chung A, Elston DM, Keller RA. 26. Di Salvo G, Habib G, Pergola V, et al. Ruyer, M. Saada, J. Schwartz, C. Selton-Suty, M. Septic microemboli in a Janeway lesion of bacterial Echocardiography predicts embolic events in Simon, B. Simorre, S. Skalli, F. Spatz, C. Strady, J. endocarditis. J Am Acad Dermatol. infective endocarditis. J Am Coll Cardiol. Sudrial, L. Tartiere, A. Terrier De La Chaise, M. C. 1996;35(6):984-985. 2001;37(4):1069-1076. Thiercelin, D. Thomas, M. Thomas, L. Toko, F. 12. Mylonakis E, Calderwood SB. Infective 27. Thuny F, Di Salvo G, Belliard O, et al. Risk of Tournoux, A. Tristan, J. L. Trouillet, L. Tual, A. endocarditis in adults. N Engl J Med. embolism and death in infective endocarditis. Vahanian, F. Verdier, V. Vernet Garnier, V. Vidal, P. 2001;345(18):1318-1330. Circulation. 2005;112(1):69-75. Weyne, M. Wolff, A. Wynckel, N. Zannad, and P. Y. 13. Matsui Y, Okada N, Nishizawa M, et al. An Zinzius. 28. Dickerman SA, Abrutyn E, Barsic B, et al; ICE Osler’s node and a Janeway lesion. Intern Med. Investigators. The relationship between the Additional Contributions: Janet Jacobson 2009;48(16):1487-1488. initiation of antimicrobial therapy and the incidence provided editorial assistance. We thank the 14. Kerr A Jr, Tan JS. Biopsies of the Janeway lesion of stroke in infective endocarditis. Am Heart J. Association Pour l’Etude et la Prévention de of infective endocarditis. J Cutan Pathol. 2007;154(6):1086-1094. l’Endocardite Infectieuse Study Group for sharing 1979;6(2):124-129. their data. These individuals did not receive 29. Duval X, Iung B, Klein I, et al. IMAGE compensation for their help. 15. Cardullo AC, Silvers DN, Grossman ME. Janeway (Resonance Magnetic Imaging at the Acute Phase of lesions and Osler’s nodes. J Am Acad Dermatol. Endocarditis) Study Group. Effect of early cerebral REFERENCES 1990;22(6, pt 1):1088-1090. magnetic resonance imaging on clinical decisions in 16. Lévesque H, Marie I. Infection and vascular infective endocarditis. Ann Intern Med. 1. Hoen B, Duval X. Clinical practice: infective 2010;152(8):497-504, W175. endocarditis. N Engl J Med. 2013;368(15):1425- purpura [in French]. J Mal Vasc. 1999;24(3):177-182. 1433. 17. Selton-Suty C, Célard M, Le Moing V, et al; 30. Arber N, Militianu A, Ben-Yehuda A, Krivoy N, AEPEI Study Group. Preeminence of Pinkhas J, Sidi Y. Native valve Staphylococcus 2. Moreillon P, Que YA. Infective endocarditis. epidermidis endocarditis. Am J Med. Lancet. 2004;363(9403):139-149. Staphylococcus aureus in infective endocarditis. Clin Infect Dis. 2012;54(9):1230-1239. 1991;90(6):758-762. 3. Que YA, Moreillon P. Infective endocarditis. Nat 31. Martínez-Marcos FJ, Lomas-Cabezas JM, Rev Cardiol. 2011;8(6):322-336. 18. Duval X, Delahaye F, Alla F, et al; AEPEI Study Group. Temporal trends in infective endocarditis in Hidalgo-Tenorio C, et al; Grupo Para el Estudio de 4. Miro JM, Anguera I, Cabell CH, et al; the context of prophylaxis guideline modifications. las Infecciones Cardiovasculares de la Sociedad International Collaboration on Endocarditis Merged J Am Coll Cardiol. 2012;59(22):1968-1976. Andaluza de Enfermedades Infecciosas. Database Study Group. Staphylococcus aureus Enterococcal endocarditis. Enferm Infecc Microbiol native valve infective endocarditis. Clin Infect Dis. 19. Li JS, Sexton DJ, Mick N, et al. Proposed Clin. 2009;27(10):571-579. 2005;41(4):507-514. modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 32. Thadepalli H, Francis CK. Diagnostic clues in 5. Hoen B, Alla F, Selton-Suty C, et al; Association 2000;30(4):633-638. metastatic lesions of endocarditis in addicts. West J Pour l’Etude et la Prévention de l’Endocardite Med. 1978;128(1):1-5. Infectieuse (AEPEI) Study Group. Changing profile 20. Habib G, Hoen B, Tornos P, et al; ESC of infective endocarditis. JAMA. 2002;288(1):75-81. Committee for Practice Guidelines; Endorsed by the European Society of Clinical Microbiology and

500 JAMA Dermatology May 2014 Volume 150, Number 5 jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/29/2021