sign in sign up
<<
Home , Epiglottitis

applyparastyle "fig//caption/p[1]" parastyle "FigCapt"

Journal of Crohn’s and Colitis, 2019, 1–10 doi:10.1093/ecco-jcc/jjz165

Advance Access publication October 11, 2019 Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 Original Article

Original Article Interstitial and Granulomatous Lung Disease in Inflammatory Bowel Disease Patients Elena Eliadou,a Joana Moleiro,b Davide Giuseppe Ribaldone,c Marco Astegiano,c Katja Rothfuss,d Carlos Taxonera,e Fahd Ghalim,f Franck Carbonnel,f Bram Verstockt,g, Stefano Festa,h Luís Maia,i Ana Berrozpe,j Edyta Zagorowicz,k Edoardo Savarino,l Pierre Ellul,m Stephan R. Vavricka,n Marta Calvo,o Ioannis Koutroubakis,p Frank Hoentjen,q Luis Fernández Salazar,r Francesca Callela,s Fiorella Cañete Pizarro,t Konstantinos Soufleris,u Elena Sonnenberg,v Maryan Cavicchi,w Joanna Wypych,x Christophe Hommel,y Alessandro Ghiani,z and Gionata Fiorinoaa,bb, ; for the ECCO CONFER COMMITTEEcc

aGastroenterology Department, Manchester Royal Infirmary, Manchester UK bInstituto Portugues de Oncologia de Lisboa Francisco Gentil, Lisboa, Portugal cGastroenterologia-U, Città della Salute e della Scienza di Torino, Turin, Italy dRobert-Bosch Hospital, Department of Gastroenterology, Hepatology and Endocrinology, Stuttgart, Germany eDepartment of Gastroenterology, Hospital Clínico San Carlos and Instituto de Investigación del Hospital Clínico San Carlos [IdISSC], Madrid, Spain fGastroenterology Department, Kremlin Bicêtre Hospital, University Paris Sud, Paris, France gDepartment of Gastroenterology and Hepatology, University Hospitals Leuven, and Department of CHROMETA, KU Leuven, Leuven, Belgium hOspedale San Filippo Neri, UOS Malattie Infiammatorie Croniche Intestinali iGastroenterology Department, Centro Hospitalar do Porto, Porto, Portugal jIBD Unit, Bellvitge’s Hospital, Barcelona, Spain kMaria Sklodowska Curie Memorial Cancer Centre and Institute of Oncology, Department of Gastroenterology,Warsaw, Poland lDepartment of Surgery, Oncology and Gastroenterology,University of Padua, Padua, Italy mDivision of Gastroenterology, Mater Dei Hospital, Valleta, Malta nDepartment of Gastroenterology and Hepatology, Center for Gastroenterology and Hepatology, Zurich, Switzerland oGastroenterology Department, Hospital Universitario Puerta de Hierro Majadahonda, Madrid, Spain pDepartment of Gastroenterology, University of Crete, Heraklion, Greece qDepartment of Gastroenterology, Radboud University Medical Center, Nijmegen, The Netherlands rDepartment of Gastroenterology, Hospital Clínico Universitario. Valladolid, Spain sUOC Gastroenterologia, Ospedale San Giuseppe, Empoli, Firenze, Italy tHospital Universitari Germans Trias i Pujol in Badalona, Barcelona, Spain uDepartment of Gastroenterology, Theagenion Cancer Hospital of Thessaloniki, Thessaloniki, Greece vDepartment of Gastroenterology, Charité, Berlin, Germany wDepartment of Gatroenterology, Clinique de Bercy, Creteil, France xSurgery & Gastroenterology Department, Copernicus Hospital, Gdansk, Poland yDepartment of Gastroenterology and Hepatology, CHU UCL Namur, Yvoir, Belgium,and Catholic University of Louvain, Brussels, Belgium zSchillerhoehe Lung Clinic [Robert-Bosch-Hospital], Department of Pneumology and Respiratory Medicine, Gerlingen, Germany aaHumanitas Clinical and Research Center, Gastroenterology Department, Rozzano, Milan, Italy bbHumanitas University, Department of Biomedical Sciences, Rozzano, Milan, Italy ccECCO CONFER Steering Committee, European Crohn’s and Colitis Organisation, Vienna, Austria

Abstract Background: Interstitial lung [ILD] disease and granulomatous lung disease [GLD] are rare respiratory disorders that have been associated with inflammatory bowel disease [IBD]. Clinical presentation is polymorphic and aetiology is unclear.

© Crown copyright 2019. 1 2 E. Eliadou et al.

Methods: This was an ECCO-CONFER project. Cases of concomitant ILD or GLD and IBD, or drug- induced ILD/GLD, were collected. The criteria for diagnosing ILD and GLD were based on definitions from the American Thoracic Society and the European Respiratory Society and on the discretion Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 of reporting clinician. Results: We identified 31 patients with ILD. The majority had ulcerative colitis [UC] n[ = 22]. Drug- related ILD was found in 64% of these patients, 25 patients [80.6%] required hospitalisation, and one required non-invasive ventilation. The causative drug was stopped in all drug-related ILD, and 87% of patients received systemic steroids. At follow-up, 16% of patients had no respiratory symptoms, 16% had partial improvement, 55% had ongoing symptoms, and there were no data in 13%. One patient was referred for lung transplantation, and one death from lung fibrosis was reported. We also identified 22 GLD patients: most had Crohn’s disease [CD] n[ = 17]. Drug-related GLD was found in 36% of patients and 10 patients [45.4%] required hospitalisation. The causative drug was stopped in all drug-related GLD, and 81% of patients received systemic steroids. Remission of both conditions was achieved in almost all patients. Conclusions: ILD and GLD, although rare, can cause significant morbidity. In our series, over half of cases were drug-related and therefore focused pharmacovigilance is needed to identify and manage these cases.

1. Introduction Box 1. Classification of pulmonary abnormalities in Up to 50% of IBD patients experience at least one extra-intestinal association with inflammatory bowel disease [non manifestation [EIM], such as pyoderma gangrenosum, uve- drug-related]. itis, episcleritis, polyarthritis, or thromboembolic disease.1–3 Bronchopulmonary manifestations, despite being considered rare 1. Upper airways with an unknown prevalence, are increasingly recognised.4 ◦ Epiglottitis The presentation of bronchopulmonary manifestations in IBD ◦ Tracheobronchitis patients is polymorphic, as all segments of the can 2. Large airways be affected. Generally, pulmonary involvement in IBD may be associ- ◦ ated with IBD medication or an EIM of the disease itself.4 ◦ Acute or chronic IBD-related lung disease can be subclassified into airway diseases, 3. Small airways autoimmune disorder, interstitial lung disease, granulomatous disease, ◦ and fistulas [Box 1].5 Particularly interstitial lung disease [ILD] and ◦ Bronchiolitis obliterans granulomatous lung disease [GLD] are rare respiratory conditions. 4. Interstitial disease: GLD, mimicking parenchymal , may be observed in CD ◦ Non-specific interstitial patients. In fact, patients with CD and concomitant sarcoidosis have ◦ Acute interstitial pneumonia been reported in the literature, suggesting a link between the two dis- ◦ Cryptogenic organizing pneumonia eases which share susceptibility genes.4 ILD is a heterogeneous group 5. Autoimmune disease: of disorders characterised by varying degrees of fibrosis and inflam- ◦ Wegener’s granulomatosis mation of lung parenchyma. There are estimates of more than 200 ◦ Pulmonary vasculitis known causes of ILD leading to symptoms and radiological changes. ◦ Churg Strauss syndrome These diseases can be classified based on the definitions from the 6. Vascular disease American Thoracic society and the European Respiratory Society.5–7 ◦ Drug-related lung disease can present with either interstitial lung 7. Other pulmonary manifestations disease or granulomatous disease. The diagnosis of drug-related dis- ◦ Necrobiotic nodules ease can be based on several criteria8: ◦ Pleuritis ◦ Fistulae 1. a history of drug exposure with correct identification of the drugs, its duration and administration; 2. clinical imaging and histopathological patterns which are con- 2. Methods sistent with earlier observations of the same drug; 2.1. Study design 3. exclusion of other lung disease; 4. improvement after discontinuation of drug suspected; This observational multicentre study retrospectively collected cases 5. recurrence of symptoms on rechallenge. across the world through the CONFER [COllaborative Network For Exceptionally Rare case reports] project9 and supported by the In our case series, the diagnosis was made at the discretion of different European Crohn’s and Colitis Organisation [ECCO]. The CONFER clinicians and once other causes such as were excluded. project was initiated by ECCO in order to specifically identify and We aimed to describe a series of IBD patients with a diagnosis of report rare IBD disease associations, which are otherwise seldom re- ILD or GLD and try to elicit the impact of the on ported due to their exceptional rarity. Briefly, the CONFER method- IBD and outcome. ology comprises selecting a topic worthy of investigation out of case ILD and GLD in IBD Patients 3 proposals submitted by ECCO members. The steering committee of 3. Results CONFER chooses the topic, and ECCO launches a call to identify

3.1. Interstitial lung disease patients Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 similar cases encountered by IBD physicians worldwide. The call to physicians is made through announcements in the A total of 31 patients with ILD were identified from 14 medical ECCO annual congress and in national IBD meetings across Europe centres. Patients’ characteristics are shown in Table 1. All patients and during several international IBD meetings. In addition, the call had a diagnosis of IBD prior to ILD, with a median time of 10.3 years for similar cases is disseminated by direct emails to all ECCO mem- [range 0.3–51 years]. Eight [25.8%] patients had IBD disease ac- bers and affiliated physicians, on the ECCO website, and in the tivity at the time of ILD diagnosis; 11 [35.5%] patients had non ECCO eNews. Physicians are then prompted to report their cases to drug-related ILD; and 20 [64.5%] had drug-related ILD [mesalazine the CONFER database using a pre-determined standardised case re- n = 9, methotrexate n = 1, golimumab n = 1, vedolizumab n = 1, port form. The call for the present case series was entitled ‘Interstitial and infliximab [IFX] n = 8]. One patient had two different drug- and granulomatous lung disease in IBD’. induced ILD, related to vedolizumab and golimumab, available as Supplementary data at ECCO-JCC online and each event was ana- lysed separately. Main characteristics and outcomes of reported ILD 2.2. Patients and procedures are shown in Figure 1. All adult IBD patients [age >16 years] with a GLD or ILD diagnosis, according to established classifications6,7 throughout the course 3.1.1. Non drug-related data of IBD or prior to its diagnosis, were eligible for inclusion in this Eleven patients were identified, 10 [90.9%] male, with mean age study. GLD and ILD diagnoses were made based on clinical presen- at event of 51.6 years [range 20–67 years]. Mean duration of IBD tation according to previous case reports and reviews. Relationship was 13.3 years [range 1.3–51 years] at the time of the event. Eight between IBD treatment and pulmonary disease was established by [72.7%] patients had previous lung disease ( n = 1, previous the physician, based on the patients’ medical history and relevant Pneumocystis jirovecii [PJ] pneumonia n = 1, pulmonary embolus tests, when applicable. Data were collected using a case report form, n = 1, previous perinuclear anti-neutrophil cytoplasmic antibodies which was divided into five domains. Section 1 included patient [epi- [p ANCA] n = 1, not specified n = 4) [Table 1]. The main reported demiological data, past medical history, smoking habits, previous symptoms were cough [n = 10, 90.9%], shortness of breath [SOB] lung disease] and disease characteristics. Section 2 was focused on [n = 9, 81.8%], [n = 2, 18.1%], and lethargy [n = 3, 27.2%] event description, Section 3 on event presentation and investigations, with duration of symptoms varying from 12 days to 1 year. Section 4 on treatment, and Section 5 on respiratory and IBD out- Investigations showed that eight patients [72.2%] had ab- comes. All the participating centres reviewed patient charts and en- normal chest X-ray [CXR], two had positive autoantibodies (one rolled all the patients eligible. The data were collected and analysed antinuclear antibodies [ANA] positive and one ANCA positive). anonymously and handled according to local regulations. All the ad- All had abnormal chest high resolution computerised tomography verse events discussed in the text and related to any drug have been [HRCT]. Bronchoscopy was done in seven patients [63.6%] [ab- reported to pharmacovigilance according to the local regulations. normal n = 3, normal n = 4]. Only six patients [54.5%] underwent lung biopsy. Diagnoses of ILD included cryptogenic organising pneu- 2.3. Statistics monia [COP] in five patients [45.4%], idiopathic For the statistical analysis, IBM Software SPSS Statistics Version [IPF] in three [27.3%], and three [27.3%] unclassified. Nine patients 22.0.0 [2013 SPSS Science, Inc., Chicago, IL] was used. Depending [81.8%] were hospitalised and seven [63.6%] received steroids, four on the number and the distribution of the data, appropriate tests of those long term. One of the patients required non-invasive ven- were used to analyse the data. tilation with bilevel positive airway pressure [BIPAP] for 1 month.

Table 1. Clinical characteristics of inflammatory bowel disease patients.

Patients with GLD [n = 22] Patients with ILD [n = 31]

Median age [years] 46.4 [range 18–86] 47 [range17-84] Sex 17 [77.3%] males: 5 [22.7%] females 19 [61.3%] males: 12 [38.7%] females Median age at diagnosis [years] 32.2 [range 10–74] 49.65 [range 17–84] CD/UC/indeterminate colitis 17 [77.3%]/5 [22.7%]/0 8/22/1 Montreal classification A1/A2/A3 1/10/6 2/5/3 L1/L2/L3 4/8/5 0/7/3 B2/B2/B3 11/5/0 7/2/1 E1/E2/E3 0/ 2/3 0/6/17 Perianal disease 5 [22.7%] 4 [12.9%] Extra-intestinal manifestations 8 [36.4%] 11 [35.5%] Positive family history of IBD 3 [13.6%] 4 [12.9%] Previous lung disease 3 [13.6%] 11 [35.5%] Smoking: current/past/never 3 [13.6%]/7 [31.8%]/12 [54.5%] 0 /10 [32.3%]/21 [67.7%] IBD activity at event: active/quiescent 4 [18.2%]/18 [81.8%] 11 [35.5%]/20 [64.5%] Drug-related lung disease 8 [36.4%] 20 [64.5%] 5-ASA/MTX/AZA/anti-TNF/VEDO 0/ 0/ 1/6/1 9/1/0/ 9/1

CD, Crohn’s disease; UC, ulcerative colitis; IBD, inflammatory bowel disease; 5-ASA. 5-aminosalicylic acid; MTX, methotrexate; AZA, azathioprine; anti-TNF, anti-tumour necrosis factor; VEDO, vedolizumab. 4 E. Eliadou et al.

Interstitial Lung Disease (ILD) (n=31) Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020

Non-drug Drug-induced induced ILD ILD (n=20) (n=11)

Golimumab (n=1)/ Cryptogenic Organizing Idiopathic brosis Unclassied 5-ASA Methotrexate In†iximab Vedolizumab Pneumonia (COP) (n=3) (n=3) (n=9) (n=1) (n=8) (n=5) (n=1)

COP (n=2) Idiopathic brosis (n=1) COP (n=3) Steroids (n=3) Steroids (n=2) Steroids (n=2) Unclassied (n=5) Unclassied (n=1) idiopathic brosis (n=2) COP under golimumab ILD unclassied after None (n=2) None (n=1) None (n=1) Eosinophilic Unspecied (n=3) vedolizumab pneumonia (n=1)

Ongoing symptoms (n=2) Stop 5-ASA (n=9) Stop methotrexate Stop in†iximab (n=8) Stop therapy (n=1) Stop therapy (n=1) Recovery (n=5) Ongoing symptoms (n=3) No date (n=1) Steroids (n=8) Steroids Steroids(n=6) Steroids (n=1) Steroids (n=1)

Symptom improvement (n=2) Symptom Ongoing symptoms (n=5) improvement (n=1) Symptom No data (n=1) Recovery (n=1) No data (n=2) improvement (n=3) Death (n=1) Ongoing symptoms (n=5)

Figure 1. Characteristics and respiratory outcomes of ILD related to IBD [non drug-related and drug-related] in our cohort. ILD, interstitial lung disease; IBD, inflammatory bowel disease.

At mean follow-up of 10 months [ranging from 4 months to At mean follow-up of 4.6 months [ranging from 1–14 months], 9 years] five [45.4%] patients had ongoing symptoms which included five [62.5%] patients had ongoing symptoms; five [62.5%] had one patient with ongoing symptoms despite long-term steroids, who normal CXR; and one [12.5%] still had ongoing abnormalities. was then given methotrexate and referred for consideration of lung CT was normal in one patient [12%], but the majority [six] had transplantation. Five [45.4%] patients had complete resolution of ongoing abnormalities [75%]. Three of these patients showed symptoms and respiratory data were missing for one patient. On re- improvement. With regards to IBD, four had active disease at view, two [18.2%] had abnormal CXR, five [45.4%], three [27.3%] follow-up [50%] and three had quiescent disease [37.5%]. Two had abnormal HRCT. Eight patients [72.7%] had active IBD, two required steroids to control their disease and five had new medica- were [18.2%] quiescent and three [27.3%] had a flare. Out of those tions initiated (5-aminosalylates [5-ASA] n = 1, azathioprine [AZA] who had a flare, one was started on adalimumab, one on IFX, and n = 1, adalimumab n = 1, vedolizumab n = 1, ustekinumab n = 1). one on tacrolimus. One of the patients underwent proctocolectomy There was only one case of a female patient [5% of drug-related for colonic dysplasia. ILD characteristics and outcome are also cases] with UC E3 receiving golimumab and vedolizumab. Clinical summarised in Supplementary Table 1, available as Supplementary characteristics are included in Supplementary Table 2. In this case data at ECCO-JCC online. there was previous use of IFX and vedolizumab and previous lung disease due to vedolizumab with ILD unclassified. She presented 3.1.2. ILD: drug related data at age 24 after two doses of golimumab and active IBD at event. A total of 20 cases were identified [mesalazinen = 9, methotrexate Symptoms included cough, SOB, fever, and lethargy. She had ab- n = 1, golimumab n = 1, vedolizumab n = 1, and IFX n = 8]. Patients’ normal CXR and HRCT diagnosed as cryptogenic organising pneu- epidemiological and IBD characteristics are summarised in Table monia, and was admitted to hospital and treated with steroids for 1 and Supplementary Table 2, available as Supplementary data at 1 month. Golimumab was stopped and, despite steroid treatment, ECCO-JCC online. she had ongoing active IBD that required admission and subtotal colectomy and ileostomy. 3.1.2.2. Biologics The same patient [as above] presented with ILD [unclassi- There were eight cases [40% of drug-related cases] with IFX- fied]: disease duration at event was 12 years and age at pres- induced interstitial lung disease. Patients were on IFX [5 mg/ entation was 24 years. She had previous use of infliximab and kg] with a mean duration of 5.4 doses [range 2–8 doses]. Main was switched to vedolizumab due to ongoing active IBD. Lung symptoms were: cough [n = 6, 75%], SOB [n = 8, 100%], fever disease presented with SOB after two doses of vedolizumab, with [n = 3, 37.5%], and lethargy [n = 4, 50%] with a mean dur- cough, SOB, fever, and lethargy. She had abnormal CXR, HRCT, ation of symptoms of 6.3 weeks [1 day–24 weeks]. All patients and bronchoscopy. She was diagnosed with ILD [unclassified], had negative virology and bacterial screen. Five [62.5%] had ab- treated with steroids, and admitted to hospital. Vedolizumab was normal CXR and six [75%] of them had abnormal HRCT; only stopped, but she had ongoing active IBD at review 4 months later one had abnormal bronchoscopy, and three [37.5%] underwent and was admitted with a flare that required steroids and initi- lung biopsy. ation of golimumab. Diagnosis included three with COP [37.5%], two with IPF [25%], and three [37.5%] with unclassified interstitial lung disease 3.1.2.3. 5-ASA [ILD unclassified]. Five patients [62.5%] were admitted, and IFX There were nine patients [45% of drug-related cases], six of them was stopped in all patients. Once infection was excluded, six patients male [66.7%]. Two [22.2%] patients had previous lung disease received steroids [75%] for duration between 1–6 months. [asthma n = 1, Sjogren’s syndrome n = 1]. Patients’ ILD characteristics ILD and GLD in IBD Patients 5 and outcomes are summarised in Supplementary Table 3, available 1 month later, she had lethargy and normal CXR. Subsequently she as Supplementary data at ECCO-JCC online. Mean age at time of had a flare of her CD, and adalimumab was started 3 years later. event was 52 years [ranging 19–80 years] with mean duration of IBD Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 6.6 years [ranging 0.8–14 years]. At the time of event, two [22.2%] 3.2. Granulomatous lung disease patients patients had active disease and seven [77.8%] had quiescent disease. We identified 22 GLD patients from 18 university hospitals, most of 5-ASA mean dose at event was 2.3 g [range 0.8–3 g] and them with CD [77.3%]. Patients’ characteristics are shown in Table mean duration of 5-ASA use was 5.11 years [range 0.8–14 years]. 1. In 17 patients, IBD diagnosis preceded lung disease diagnosis after Presenting symptoms were: cough [n = 7], SOB [n = 8], fever [n = 6], a median time of 10.6 years [range 0.5–27 years]; only four patients and lethargy [n = 1], with mean duration of 7.5 weeks [ranging had active disease at the time of GLD diagnosis. In four patients, 1.5–12 weeks]. All patients had abnormal CXR and CT. One had lung disease was diagnosed at a median time of 14 months [range abnormal bronchoscopy [11.1%] and only two had lung biopsy 8–24 months] before IBD and, in one patient, IBD and lung disease [22.2%]. Diagnosis included two patients with COP [22.2%]; five diagnoses were established simultaneously. had ILD unclassified [55.5%], one with idiopathic interstitial dis- A total of 14 cases [63.6%] were considered non drug-related ease [3.3%] and one with [3.3%]. The drug GL [primary sarcoidosis n = 7, fungal infection n = 2, and unspeci- was stopped in all patients. Eight [88.9%] patients were admitted to fiedn = 5]. Eight cases [36.4%] were considered drug-related GLD hospital and treated with steroids [88.9%] with mean duration of [IFX n = 4, certolizumab n = 1, adalimumab n = 1, vedolizumab 6.47 months [0.3–24 months]: two needed long-term use. Any infec- n = 1, and azathioprine [AZA] n = 1], among whom five had tion was excluded before commencing steroids. sarcoidosis. Characteristics of GLD and outcomes are shown in At mean follow-up of 11.6 months, [ranging 3–24], five patients Figure 2. [55.5%] had ongoing symptoms, four had abnormal CXR [44.4%], and two [50%] experienced improvement. Three [33.3%] had ab- 3.2.1. Non drug-related GLD normal HRCT. One [11%] became steroid dependent and died We identified 14 patients [63.6%] with non drug-related GLD 5 years later, due to progressive lung disease. After treatment, one pa- and IBD, 11 male [78.6%], with mean age at event of 44.5 years tient [11%] had a flare that required steroids, three [33.3%] started [range 18–85 years]. At event only four patients had active IBD dis- new medications [AZA, steroids, one restarted 5-ASA]. 5-ASA in- ease. Nine patients presented with cough [64.3%], four with SOB duced ILD characteristics and outcome are also summarised in [28.6%], two with fever [14.3%], and two with weight loss [14.3%]. Supplementary Table 3. Investigations showed that 12 patients had abnormal CT scan [85.7%], 11 [78.6%] had abnormal CXR, and four had abnormal 3.1.2.4. Methotrexate spirometry [28.6%, restrictive pattern n = 2, obstructive pattern Only one patient [5%] was identified, a female with CD L2. Age n = 2]. Three had abnormal bronchoscopy, and 12 patients [85.7%] at event was 44 years, and at presentation she had received metho- had a lung biopsy compatible with granulomatous non-necrotising trexate [MTX] for 2 years at 15 mg every week, and her IBD was in disease. remission. She presented with cough, SOB, fever, and lethargy, that Diagnosis was primary sarcoidosis in seven patients [50%] lasted a few days. Abnormal CXR and spirometry [mixed restrictive [six males; four patients with CD and three with UC]. One pa- and obstructive pattern] and also abnormal HRCT were reported. tient was receiving anti-tumour necrosis factor [TNF] therapy She was treated with steroids for 1 month after being admitted to [adalimumab], one immunosuppressive therapy [AZA], and two hospital. A diagnosis of ILD [unclassified] was made. At review 5-ASA therapy; three were without therapy. The first patient had

Granulomatous lung disease (GLD) (n=22)

Non-drug related Drug-related (n=14) (n=8)

Anti-TNF (n=6) Fungal In iximab (n=3) Primary sarcoidosis infection-induced Unspeci‚ed (n=5) Vedolizumab (n=1) Azathioprine (n=1) (n=7) Adalimumab (n=1) GLD (n=2) Certolizumab (n=1)

Steroids (n=7) Steroids (n=5) Stop anti-TNF (n=6) Stop vedolizumab Stop azathioprine (n=1) Methotrexate (n=1) Recovered (n=2) In iximab (n=4) Steroids (n=5) Steroids Steroids (n=1) Azathioprine (n=2)

Recovery (n=2) Recovered (n=3) Recovered (n=6) Recovered (n=1) Recovery (n=1) under steroids+ after in iximab azathioprine

Figure 2. Characteristics and respiratory outcomes of GLD related to IBD [non- and drug-related] in our cohort. GLD, granulomatous lung disease; IBD, inflammatory bowel disease. 6 E. Eliadou et al. received adalimumab therapy for just only 1 month when the IBD symptoms. Three patients were started on vedolizumab therapy, diagnosis of sarcoidosis was established. Due to progressive de- and one patient treated with IFX was successfully switched to terioration with weight loss, fatigue, and arthralgia, further in- adalimumab [data not available for one patient]. One male patient Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 vestigation was made which revealed systemic granulomatosis with CD, who was on IFX for 72 months, had a diagnosis of GLD with skin, lung, spleen, bone, heart, bone marrow, joint [right [non-sarcoidosis] related to anti-TNF. The anti-TNF was stopped and knee], and gastro-intestinal involvement. Due to severe lung de- he was given steroids, with resolution of respiratory symptoms. When terioration and cardiac involvement, a high dose of steroids was reviewed at 3 months, AZA was started for maintenance of his IBD. initiated. One male patient with CD [12.5% of drug-related cases] had the In the remaining six patients, diagnoses of both IBD [CD n = 3, diagnosis of CD established 7 years before and had already been treated UC n = 3] and sarcoidosis were made. Initially all patients re- with AZA and golimumab. After 4 months of vedolizumab therapy ceived steroids, and during the follow-up two patients started on [300 mg q4w], he developed cough, SOB, and lethargy. He had a diag- immunosuppression with AZA, achieving remission of both con- nosis of GLD secondary to vedolizumab. Vedolizumab was stopped and ditions, whereas one UC patient, who had already been submitted he was treated with steroids, with resolution of respiratory symptoms. to proctocolectomy, was started on immunosuppression with MTX Due to active IBD, 15 months later he was started on anti-TNF therapy. due to ongoing respiratory symptoms. Five patients [36%] were considered to have an unspecified GLD 3.2.2.2. Azathioprine [AZA] [three males; all of the five with CD]: two patients had no IBD medi- There was only one such case, of male patient with CD [12.5% of cation [lung disease was established before or at the same time as drug-related cases]. This patient had the diagnosis of CD established IBD], two patients were under 5-ASA, and one under MTX. All pa- 7 years before and was in remission under AZA [2.5 mg/kg]. After tients initially received steroids. Additionally, during the follow-up, 24 months of AZA he developed cough, SOB, and lethargy, and the four patients started anti-TNF therapy [adalimumab], of whom diagnosis of unspecified GLD was established. The patient was ad- three achieved steroid-free remission of both conditions. mitted, AZA was stopped, and he was treated with steroids. Four We identified two patients [14%] [two males; one patient with years later the patient still needed steroids [10 mg/day] to control CD and one with UC] with a GLD caused by a fungal infection: respiratory symptoms and still had abnormal CT chest and spirom- one patient was under systemic steroid therapy and the other pa- etry. There was no need for new IBD therapy. tient had the diagnosis of pulmonary histoplasmosis made before Among patients with drug-related GLD, hospital admission was IBD diagnosis. At follow-up, neither patient had ongoing pulmonary needed in two patients [25%] but none needed invasive ventilation. symptoms. With regards to IBD, the first patient had pancolitis and Remission of both conditions was achieved in all patients, with only during the follow-up was submitted to surgery [ileal pouch-anal one patient still needing steroid to control lung disease. anastamosis: IPAA] to control IBD, and the other still had disease activity despite the introduction of budesonide therapy. Among patients with non drug-related GLD, hospital admission 4. Discussion was needed ineight8 patients [57.1%], but none needed invasive ven- Even though pulmonary manifestations in IBD are rare they are in- tilation. Remission of both conditions was achieved in almost all creasingly recognised.10,11 The true prevalence in IBD remains un- patients. known.12,13 Our retrospective study describes a series of IBD patients affected by ILD [n = 31] and GL [n = 22]. Most cases were reported 3.2.2. Drug-related GLD to be drug-related [n = 28] and 25 were non drug-related cases. Most We identified eight patients [36.4%] with drug-related GLD [IFX of the current data in literature comes from rheumatological patients n = 4, certolizumab n = 1, adalimumab n = 1, vedolizumab n = 1, and also small case series; our study provides insight into the IBD and AZA n = 1] and IBD, six males [75.0%], with mean age at population. event of 48.1 years [35–76 years]. At event presentation, only one Establishing an accurate diagnosis of ILD can be challenging for patient had active IBD disease. Six patients presented with cough clinicians, as there are more than 200 different subtypes and over the [75.0%], five with SOB [62.5%], five with lethargy [62.5%], past decade ILD has been reclassified in comprehensive international three with fever [37.5%], and two with weight loss [25.0%]. consensus statements.6,7 Determination of disease subtypes requires Investigations showed that all patients had abnormal thorax CT consideration of patients’ clinical features, radiological pattern, ser- scan and three had abnormal spirometry [37.5%, restrictive pat- ology, and in some cases lung biopsy.14–21 tern n = 2, mixed pattern n = 1]. Bronchoscopy was done in seven Making a diagnosis of granulomatous lung disease diagnosis can patients [abnormal n = 2, normal n = 5]. Most patients [n = 5, also be challenging. Even with histology, it can be difficult to distin- 62.5%] had lung biopsy, all of them compatible with granuloma- guish between sarcoidosis and Crohn’s disease. In fact, both entities tous non-necrotising disease. are barrier disorders of unknown aetiology, with several aspects in common [non-necrotising granulomas, an increased intestinal per- 3.2.2.1. Biologics meability, local and peripheral T cell activation, and an increased There was a total of six cases [75.0% of drug-related cases] with proinflammatory cytokine release].22–27 Moreover, the patterns of anti-TNF induced GLD. Diagnosis included five patients with sar- possible organ involvement outside the primary manifestations in coidosis [83.3%] and one with GLD unspecified [16.7%]. In the the lung and gut are similar and include eye, skin, joints, and liver. sarcoidosis group [n = 5; three males, four with CD; mean age of Sarcoidosis manifestations in the gut, CD manifestations in the lung, 41.4 years] all were under anti-TNF therapy [IFX n = 3, adalimumab and manifestations of both diseases in the same patient, have been n = 1, certolizumab n = 1] for a mean time of 55 months [range reported in rare cases, which may impose considerable diagnostic 10–108 months]. The causative drug was stopped in all patients and difficulties.28–30 two also received steroids. At mean follow-up of 33 months [range In our case series, 11 patients had non drug-induced ILD with male 12–84 months] there were no patients with ongoing respiratory or predominance, 73% with pre-existing lung disease and concomitant ILD and GLD in IBD Patients 7

UC. None of the pre-existing lung disease included previous ILD. Only IBD patients.37,38 The most important side effects include increased six patients had lung biopsy, making classification difficult, with three risk of opportunistic .39 There has been some evidence cases unclassified. Three cases were diagnosed as IPF which is a pro- that anti-TNF therapy can cause acute respiratory distress syn- Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 gressive lung disease; two of these had ongoing respiratory symptoms drome, diffuse alveolar haemorrhage,40 and ILD.41–44 A large-scale and radiological changes, but for one no respiratory data exist. One of post-marketing surveillance study in Japan, on safety of IFX in the ILD unclassified cases had progressive deterioration of his respira- rheumatology patients, revealed that the development of ILD was tory symptoms requiring referral for consideration of lung transplant- a rare event at 0.5% [25/5000]. Mean age of those patients was ation, despite ongoing steroid treatment. A high proportion of the cases 62.9 years and occurred after a mean dose of 2.8 in- were admitted and treated with steroids, with resolution of symptoms fusions of IFX.45 Another case-control study from Nakashita in- in 45% of patients, which possibly is related to the ILD subtype of vestigated the potential risk of TNF inhibitors on progression of COP which has a good prognosis and responds to steroid treatment. ILD in patients with rheumatoid arthritis [RA] revealed increased We also identified 14 patients with non drug-related GLD, and risk of ILD events and also progression of pre-existing ILD dis- 12 patients had lung biopsy. Seven patients were diagnosed with sar- ease.46 Patients presented mostly with dyspnoea and cough with coidosis and five had GLD unclassified, which again demonstrates IFX use of 6–14 weeks preceding presentation. Diagnosis included: the difficulty in diagnosis; two cases were due to fungal infection. non-specific interstitial pneumonitis, drug-related alveolitis, and Almost all cases recovered with the use of steroids and AZA [sar- drug-induced interstitial lymphocytic pneumonitis; all had dis- coidosis] or IFX [unspecified GLD]. continuation of IFX with complete resolution of symptoms with The most common pulmonary manifestations of IBD are drug- steroids.47–49 There has been one case report of fatal acute intersti- induced lung disease. Most IBD drugs have been implicated, and tial pneumonia after an accelerated dose of IFX for acute severe definitive diagnosis of drug-induced lung disease can be difficult. UC.50 In our case series, we identified eight IFX-induced ILD cases Clinical presentation and exclusion of other causes play important with male predominance. Only two patients had lung biopsy. One roles in making the diagnosis, but sometimes invasive procedures young male was diagnosed with idiopathic pulmonary fibrosis, but such as bronchoscopy and lung biopsy may be necessary.22–27 he responded to steroids and it was therefore unclear whether that Patients can present with virtually all histopathological patterns diagnosis was the correct one. Most of our patients had ongoing of either ILD or GLD. Moreover, some drugs can produce more symptoms and radiological changes despite steroid treatment and than one pattern of respiratory involvement. There have been iden- discontinuation of the drug. The different outcome might be due tified some risk factors for developing drug-related lung disease, to the diversity of the subtype of ILD, as different subtypes have including: extremes of age, sex, ethnicity, dose of medication, con- different outcomes. comitant use of oxygen, drug interaction, and underlying lung dis- Another uncommon pulmonary adverse event associated with ease.31 Diagnostic criteria exist as mentioned earlier, but despite that anti-TNF therapy is the development of sarcoid-like lesions. The it can still be a difficult diagnosis.8 estimated frequency is 1:28 000 [0.04%] which comes from case Side effects of sulphasalazine and mesalazine are rare and can reports published from rheumatological data.51–53 Symptoms are be dose-related or idiosyncratic. Mesalazine-induced lung injury is non-specific, including dyspnoea, non-productive cough, erythema rare and usually presents as interstitial and eosinophilic pneumon- nodosum, parotid enlargement, and neuro-ocular manifestations. itis, bronchiolitis obliterans, and eosinophilic .32 Radiologically, the most common findings are mediastinal and hilar Case report data suggest that mesalazine-induced lung toxicity fol- adenopathy, and less frequently, upper lobe opacities. The diag- lows a milder course; however, there have been case reports with nosis is confirmed histologically by the presence of well-formed more severe presentation requiring intubation.33 In a recent review non-necrotising granulomas. Ultimately, sarcoidosis is a diagnosis of 50 cases of sulphasalazine lung toxicity, most patients presented of exclusion, once all infectious causes have been ruled out.54 In a with cough, breathlessness, fever, and weight loss. Most cases dem- recent review of the available literature, the authors evaluated the onstrated chronic interstitial pneumonia, desquamative interstitial published cases of sarcoidosis associated with anti-TNF therapy and pneumonitis, eosinophilic pneumonitis, or acute interstitial pneu- found a total of 90 patients with median age of 43 years [7–72] and monia. Clinical improvement occurred in majority of cases, with a female predominance [61%]. The underlying anti-TNF therapy resolution occurring on average 6.5 weeks after initiation of therapy. was etanercept [n = 53, 59%], adalimumab [n = 21, 23%], and IFX There were five deaths due to pulmonary pathology, and all cases [n = 16, 18%], and the median duration between initiation of anti- had UC as underlying condition.34 In another review of case re- TNF therapy and diagnosis was 22.5 months [1–84]. Regarding ports of 5-ASA-induced lung toxicity, mean age of presentation was treatment options, in 43 cases, the anti-TNF therapy was discon- 46 years, with most cases being female and having UC as underlying tinued and steroids were started, with an improvement or resolution disease. The time for appearance of symptoms ranged from 2 days to in 41 cases [95%]; in 37 cases, anti-TNF discontinuation was the 8 years.35 5-ASA lung toxicity usually responds to discontinuation of only intervention, giving improvement in 32 cases [86%]; in five drug, and some may require a course of steroids.36 In our case series cases, addition of steroid therapy alone gave improvement in four we identified nine cases and, unlike the literature, most of them were cases [80%]; and in one out of three cases, improvement was ob- male [67%] and 89% had ulcerative colitis. The mean duration use served without any intervention [33%]. In 20 cases, the anti-TNF of 5-ASA was 5.1 years [range 0.8–14 years] which is similar to the therapy was restarted: in six cases the same drug was used, which previous review. Only two patients had lung biopsy and most cases resulted in recurrence in four; and in 14 cases, another anti-TNF was were ILD unclassified. At follow-up, 55% of them had ongoing re- initiated, with relapse in three.55 In our series we found a male pre- spiratory symptoms. There was one death due to progressive fibrosis. dominance, but the other characteristics and outcomes are similar to This was an 81 year-old male patient who was diagnosed with idio- those previously reported. All patients fully recovered by stopping pathic pulmonary fibrosis and became steroid-dependent. anti-TNF therapy and receiving steroids. The use of anti-TNF drugs [IFX, adalimumab, golimumab, No data exist in the IBD population regarding golimumab- certolizumab] in IBD has increased and improved the outcomes of induced ILD. According to the Europeanmedicines Agency [EMA] 8 E. Eliadou et al.

2009 Assessment report, serious non-infectious pulmonary adverse and whether a drug was the caveat depended on the clinician sub- events were observed in five Phase 3 trials of golimumab for the mitting the relevant information and, as cases were collected from treatment of rheumatoid arthritis, ankylosing spondylitis, or psori- 2000–17, the classification might have not been the correct one—as Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 atic arthritis: one case of interstitial disease, two of pneumonitis, and classification of ILD has been revised over 2013–15. one of fibrosing alveolitis.56–61 Most of patients were also on metho- Due to the low sample size, the correlation between IBD ther- trexate, and it was therefore not possible to make a definite associ- apies and lung disease is challenging to assess. Also no risk factors or ation between ILD and golimumab. We reported a case of a young predictors can be investigated, due to the very low sample size and woman with previous use of IFX and previous lung disease due to the heterogeneity of cases. vedolizumab, who had golimumab-induced COP and required ster- In conclusion, ILD and GLD although rare can cause significant oids. The golimumab was stopped and she was given steroids and, morbidity. About half of our cases were drug-related, and therefore due to ongoing active UC, she had subtotal colectomy and ileostomy. clinicians should be more vigilant regarding IBD medications, espe- Vedolizumab, unlike other drugs used for IBD, has not been asso- cially when infectious causes have been excluded. Early recognition ciated with the development of serious opportunistic infections and and treatment is important, as most respond to withdrawal of medi- malignancies.62 However, clinical observations indicate an increased cation and use of steroids. risk for extra-intestinal autoimmune events under this therapy.63 There have been two case reports of ILD in the literature.64,65 The only case of vedolizumab-related ILD [unclassified subtype] reported Funding in our cohort responded to steroids, but had further admissions due No funding has been received for this project. to flare of her disease and required golimumab therapy. We also reported a case of a CD male patient who devel- Conflict of Interest oped cough, shortness of breath, and lethargy 4 months after starting vedolizumab. He was diagnosed as GLD associated with None declared. vedolizumab therapy but, as the similar case recently published,66 this patient can be reclassified as a patient with a newly developed Author contributions pulmonary manifestation of CD under vedolizumab. Methotrexate [MTX] may be useful in IBD treatment but can EE and JM initiated the study: EE analysed and interpreted ILD data and cause adverse events in lungs which can be lethal.67 The mechanism drafted and revised the manuscript. JM was responsible for analysis and inter- pretation of data and drafting the GLD part of manuscript. GF participated in is unclear and thought to be an idiosyncratic reaction that can cause study design and critically revised the manuscript. All the authors contributed interstitial pneumonitis and bronchiolitis.68 MTX pneumonitis is a the cases and were responsible for revision of the manuscript as well as for the rare complication, with only 123 case reports in the literature. Most treatment of the patients. All other collaborating authors contributed with at cases have a subacute onset and it is most common within the first least one case and critically revised the manuscript. year of treatment, with a reported incidence that varies from 0.8% to 6.9%.69,70 Treatment of MTX pneumonitis includes withdrawal of MTX, high-dose , and supportive therapy.70 We iden- Supplementary Data tified one case that presented with ILD unclassified after 2 years of Supplementary data are available at ECCO-JCC online. weekly MTX use due to CD; this patient was treated with 1 month of steroids, with resolution of symptoms. Pulmonary toxicity due to AZA/6-MP [6-mercaptopurine] has References been reported infrequently in the literature, although interstitial 1. Bernstein CN, Wajda A, Blanchard JF. The clustering of other chronic in- pneumonitis, , Goodpasture-like syndrome, flammatory diseases in inflammatory bowel disease: a population-based and pulmonary haemorrhage have been described after use of these study. Gastroenterology 2005;129:827–36. drugs.12 The largest series of lung toxicity related to AZA, with de- 2. Storch I, Sachar D, Katz S. Pulmonary manifestations of inflammatory scription of seven cases on AZA therapy, was reported in patients bowel disease. Inflamm Bowel Dis 2003;9:104–15. undergoing renal allograft transplantation immunosuppression. 3. Rothfuss KS, Stange EF, Herrlinger KR. Extraintestinal manifestations Lung biopsies revealed interstitial pneumonitis in five patients and and complications in inflammatory bowel diseases. World J Gastroenterol 2006;12:4819–31. in two patients: three patients died. The 4. Kraft SC, Earle RH, Roesler M, Esterly JR. Unexplained other four improved after stopping AZA, and in two of these pa- bronchopulmonary disease with inflammatory bowel disease. Arch Intern tients cyclophosphamide therapy to resolve completely this side ef- Med 1976;136:454–9. 71 fect was needed. There are no cases of GLD due to AZA/6-MP 5. Black H, Mendoza M, Murin S. Thoracic manifestations of inflammatory in the literature. Here we described a case of a CD male patient bowel disease. Chest 2007;131:524–32. in remission who developed cough, shortness of breath, and leth- 6. Sverzellati N, Lynch DA, Hansell DM, et al. American Thoracic Society- argy 24 months after AZA initiation. He was diagnosed with AZA- European Respiratory Society classification of the idiopathic interstitial pneu- related GLD and was admitted to hospital. He responded to steroids monias: advances in knowledge since 2002. Radiographics 2015;35:1849–71. and became steroid-dependent to control his respiratory symptoms, 7. Travis WD, Costabel U, Hansell DM, et al.; ATS/ERS Committee on with ongoing radiological and spirometry changes 4 years later. Idiopathic Interstitial . An official American Thoracic Society/ European Respiratory Society statement: update of the international Our study has several limitations: it was a retrospective case re- multidisciplinary classification of the idiopathic interstitial pneumonias. port data collection, and it relied on voluntary submission of cases Am J Respir Crit Care Med 2013;188:733–48. by physicians who responded to the ECCO calls and therefore might 8. Bradley B, Branley HM, Egan JJ, et al.; British Thoracic Society Interstitial be subject to geographical and selection biases. Only 12 ILD cases Lung Disease Guideline Group, British Thoracic Society Standards of [38.7%] had lung biopsy and 17 GLD cases [77%], which resulted Care Committee; Thoracic Society of Australia; New Zealand Thoracic in a lot of cases having ILD unclassified. The classification of ILD Society; Irish Thoracic Society. Interstitial lung disease guideline: the ILD and GLD in IBD Patients 9

British Thoracic Society in collaboration with the Thoracic Society of 33. Abraham A, Karakurum A. Acute secondary Australia and New Zealand and the Irish Thoracic Society. Thorax to mesalamine-induced interstitial pneumonitis. BMJ Case Rep

2008;63[Suppl 5]:v1–58. 2013;2013:Aug 20. doi: 10.1136/bcr-2013-009834. Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 9. Katsanos KH, Domènech E, Rahier JF, et al. Making a case for case re- 34. Parry SD, Barbatzas C, Peel ET, Barton JR. Sulphasalazine and lung tox- ports: the ECCO-CONFER viewpoint on investigating rare events in a icity. Eur Respir J 2002;19:756–64. medical world reigned by group-comparative statistics. J Crohns Colitis 35. Sossai P, Cappellato MG, Stefani S. Can a drug-induced pulmonary hyper- 2017;11:256–7. sensitivity reaction be dose-dependent? A case with mesalamine. Mt Sinai 10. Hoffmann RM, Kruis W. Rare extraintestinal manifestations of inflamma- J Med 2001;68:389–95. tory bowel disease. Inflamm Bowel Dis 2004;10:140–7. 36. Foster RA, Zander DS, Mergo PJ, Valentine JF. Mesalamine-related lung 11. Loftus EV Jr. Clinical epidemiology of inflammatory bowel disease: disease: clinical, radiographic, and pathologic manifestations. Inflamm Incidence, prevalence, and environmental influences. Gastroenterology Bowel Dis 2003;9:308–15. 2004;126:1504–17. 37. Cheifetz AS. Management of active Crohn disease. JAMA 12. Casella G, Villanacci V, Di Bella C, Antonelli E, Baldini V, Bassotti G. 2013;309:2150–8. Pulmonary diseases associated with inflammatory bowel diseases.J 38. Ghosh S. Anti-TNF therapy in Crohn’s disease. Novartis Found Symp Crohns Colitis 2010;4:384–9. 2004;263:193–205. 13. Lu DG, Ji XQ, Liu X, Li HJ, Zhang CQ. Pulmonary manifestations of 39. Colombel JF, Loftus EV Jr, Tremaine WJ, et al. The safety profile of Crohn’s disease. World J Gastroenterol 2014;20:133–41. infliximab in patients with Crohn’s disease: The Mayo Clinic experience in 14. Disayabutr S, Calfee CS, Collard HR, Wolters PJ. Interstitial lung diseases 500 patients. Gastroenterology 2004;126:19–31. in the hospitalized patient. BMC Med 2015;13:245. 40. Panagi S, Palka W, Korelitz BI, Taskin M, Lessnau KD. Diffuse alveolar 15. Schlesinger C, Koss MN. The organizing pneumonias: an update and re- hemorrhage after infliximab treatment of Crohn’s disease. Inflamm Bowel view. Curr Opin Pulm Med 2005;11:422–30. Dis 2004;10:274–7. 16. Cordier JF. Cryptogenic organising pneumonia. Eur Respir J 41. Ostor AJ, Chilvers ER, Somerville MF, et al. Pulmonary complications 2006;28:422–46. of infliximab therapy in patients with rheumatoid arthritis.J Rheumatol 17. Polverosi R, Maffesanti M, Dalpiaz G. Organizing pneumonia: typical and 2006;33:622–8. atypical HRCT patterns. Radiol Med 2006;111:202–12. 42. Ramos-Casals M, Brito-Zeron P, Munoz S, et al. Autoimmune diseases 18. Yoo JW, Song JW, Jang SJ, et al. Comparison between cryptogenic organ- induced by TNF-targeted therapies: Analysis of 233 cases. Medicine izing pneumonia and connective tissue disease-related organizing pneu- [Baltimore] 2007;86:242–51. monia. Rheumatology [Oxford] 2011;50:932–8. 43. Villeneuve E, St-Pierre A, Haraoui B. Interstitial pneumonitis associated 19. Drakopanagiotakis F, Paschalaki K, Abu-Hijleh M, et al. Cryptogenic and with infliximab therapy. J Rheumatol 2006;33:1189–93. secondary organizing pneumonia: clinical presentation, radiographic find- 44. Weatherhead M, Masson S, Bourke SJ, Gunn MC, GP. Interstitial ings, treatment response, and prognosis. Chest 2011;139:893–900. pneumonitis after infliximab therapy for Crohn’s disease. Inflamm Bowel 20. Cohen AJ, King TE Jr, Downey GP. Rapidly progressive bronchio- Dis 2006;12:427–8. litis obliterans with organizing pneumonia. Am J Respir Crit Care Med 45. Takeuchi T, Tatsuki Y, Nogami Y, et al. Postmarketing surveillance of the 1994;149:1670–5. safety profile of infliximab in 5000 Japanese patients with rheumatoid 21. Johkoh T, Muller NL, Taniguchi H, et al. Acute interstitial pneumonia: arthritis. Ann Rheum Dis 2008;67:189–94. thin-section CT findings in 36 patients. Radiology 1999;211:859–63. 46. Nakashita T, Ando K, Kaneko N, Takahashi K, Motojima S. Potential risk 22. Freeman HJ, Davis JE, Prest ME, Lawson EJ. Granulomatous bron- of TNF inhibitors on the progression of interstitial lung disease in patients chiolitis with necrobiotic pulmonary nodules in Crohn’s disease. Can J with rheumatoid arthritis. BMJ Open 2014;4:e005615. Gastroenterol 2004;18:687–90. 47. Sen S, Peltz C, Jordan K, Boes TJ. Infliximab-induced nonspecific intersti- 23. Shulimzon T, Rozenman J, Perelman M, Bardan E, Ben-Dov I. Necrotizing tial pneumonia. Am J Med Sci 2012;344:75–8. granulomata in the lung preceding colonic involvement in 2 patients with 48. Veerappan SG, O’Morain CA. Infliximab-associated alveolitis after treat- Crohn’s disease. Respiration 2007;74:698–702. ment for severe left-sided ulcerative colitis. Eur J Gastroenterol Hepatol 24. Ikehara N, Fujimoto K, Sadohara J, et al. Interstitial pneumonia associated 2009;21:830–2. with ulcerative colitis: High-resolution computed tomography and patho- 49. Wiener CM, Muse VV, Mark EJ. Case records of the Massachusetts logic findings. J Thorac Imaging 2013;28:W21–3. General Hospital. Case 33–2008. A 63-year-old woman with dyspnea on 25. Marten K, Fend F, Hautmann H, et al. Case report: fatal acute exacer- exertion. N Engl J Med 2008;359:1823–32. bation of usual interstitial pneumonia in ulcerative colitis. Br J Radiol 50. Rofaiel R, Kohli S, Mura M, Hosseini-Moghaddam SM. A 53-year-old 2005;78:762–6. man with dyspnoea, respiratory failure, consistent with infliximab- 26. Camus P, Piard F, Ashcroft T, Gal AA, Colby TV. The lung in inflammatory induced acute interstitial pneumonitis after an accelerated induction bowel disease. Medicine [Baltimore] 1993;72:151–83. dosing schedule. BMJ Case Rep 2017;2017:May 12. doi: 10.1136/ 27. Teague WG, Sutphen JL, Fechner RE. Desquamative interstitial pneu- bcr-2017-219956. monitis complicating inflammatory bowel disease of childhood. J Pediatr 51. Daien CI, Monnier A, Claudepierre P, et al. Sarcoid-like granuloma- Gastroenterol Nutr 1985;4:663–7. tosis in patients treated with tumor necrosis factor blockers: 10 cases. 28. Fellermann K, Stahl M, Dahlhoff K, et al. Crohn’s disease and sarcoidosis: Rheumatology [Oxford] 2009;48:883–6. systemic granulomatosis? Eur J Gastroenterol Hepatol 1997;9:1121–4. 52. Dhaille F, Viseux V, Caudron A, et al. Cutaneous sarcoidosis occurring during 29. Fries W, Grassi SA, Leone L, et al. Association between inflammatory anti-TNF-alpha treatment: report of two cases. Dermatology 2010;220:234–7. bowel disease and sarcoidosis. Report of two cases and review of the lit- 53. Massara A, Cavazzini L, La Corte R, Trotta F. Sarcoidosis appearing erature. Scand J Gastroenterol 1995;30:1221–3. during anti-tumor necrosis factor alpha therapy: a new “class effect” 30. Takashima M, Hanai H, Sato Y, et al. Association of Crohn’s ileitis with paradoxical phenomenon. Two case reports and literature review. Semin pulmonary and cutaneous sarcoidosis in a Japanese patient. Gastrointest Arthritis Rheum 2010;39:313–9. Endosc 1998;48:637–9. 54. Wollina U, Hansel G, Koch A, Schönlebe J, Köstler E, Haroske G. 31. Schwaiblmair M, Behr W, Haeckel T, et al. Drug induced interstitial lung Tumor necrosis factor-alpha inhibitor-induced psoriasis or psoriasiform disease. Open Respir Med J 2012;6:63–74. exanthemata: first 120 cases from the literature including a series of six 32. Cilloniz R, Chesrown SE, Gonzalez-Peralta RP. Asymptomatic presenta- new patients. Am J Clin Dermatol 2008;9:1–14. tion of mesalamine-induced lung injury in an adolescent with Crohn dis- 55. Decock A, Van Assche G, Vermeire S, Wuyts W, Ferrante M. Sarcoidosis- ease. BMJ Case Rep 2009;2009. doi: 10.1136/bcr.09.2008.0908. [Epub like lesions: another paradoxical reaction to anti-TNF therapy? J Crohns 20 Mar 2009.] Colitis 2017;11:378–83. 10 E. Eliadou et al.

56. Emery P, Fleischmann RM, Moreland LW, et al. Golimumab, a human multicentre, randomised, double-blind, placebo-controlled, phase III trial. anti-tumor necrosis factor alpha monoclonal antibody, injected sub- Lancet 2009;374:210–21.

cutaneously every four weeks in methotrexate-naive patients with active 62. Colombel JF, Sands BE, Rutgeerts P, et al. The safety of vedolizumab for Downloaded from https://academic.oup.com/ecco-jcc/advance-article-abstract/doi/10.1093/ecco-jcc/jjz165/5585570 by universite catholique de louvain user on 09 January 2020 rheumatoid arthritis: twenty-four-week results of a phase III, multicenter, ulcerative colitis and Crohn’s disease. Gut 2017;66:839–51. randomized, double-blind, placebo-controlled study of golimumab before 63. Baumgart DC, Bokemeyer B, Drabik A, Stallmach A, Schreiber S; methotrexate as first-line therapy for early-onset rheumatoid arthritis. Vedolizumab Germany Consortium. Vedolizumab induction therapy for Arthritis Rheum 2009;60:2272–83. inflammatory bowel disease in clinical practice – a nationwide consecutive 57. Hadjinicolaou AV, Nisar MK, Bhagat S, Parfrey H, Chilvers ER, Ostör AJ. German cohort study. Aliment Pharmacol Ther 2016;43:1090–102. Non-infectious pulmonary complications of newer biological agents 64. Nambiar S, Karippot A, Oliver T. Vedolizumab-associated acute respira- for rheumatic diseases – a systematic literature review. Rheumatology tory distress syndrome. Am J Ther 2018;25:e592–3. [Oxford] 2011;50:2297–305. 65. Collins HW, Frye JW. Interstitial lung disease in a 70-year-old man with 58. Inman RD, Davis JC Jr, Heijde Dv, et al. Efficacy and safety of ulcerative colitis. ACG Case Rep J 2018;5:e28. golimumab in patients with ankylosing spondylitis: results of a random- 66. Lissner D, Glauben R, Allers K, et al. Pulmonary manifestation of Crohn’s ized, double-blind, placebo-controlled, phase III trial. Arthritis Rheum disease developed under treatment with vedolizumab. Am J Gastroenterol 2008;58:3402–12. 2018;113:146–8. 59. Kavanaugh A, McInnes I, Mease P, et al. Golimumab, a new human tumor ne- 67. Imokawa S, Colby TV, Leslie KO, Helmers RA. Methotrexate pneumon- crosis factor alpha antibody, administered every four weeks as a subcutaneous itis: review of the literature and histopathological findings in nine patients. injection in psoriatic arthritis: twenty-four-week efficacy and safety results of Eur Respir J 2000;15:373–81. a randomized, placebo-controlled study. Arthritis Rheum 2009;60:976–86. 68. Sostman HD, Matthay RA, Putman CE, Smith GJ. Methotrexate-induced 60. Keystone EC, Genovese MC, Klareskog L, et al.; GO-FORWARD Study. pneumonitis. Medicine [Baltimore] 1976;55:371–88. Golimumab, a human antibody to tumour necrosis factor {alpha} given 69. Grove ML, Hassell AB, Hay EM, Shadforth MF. Adverse reactions by monthly subcutaneous injections, in active rheumatoid arthritis des- to disease-modifying anti-rheumatic drugs in clinical practice. QJM pite methotrexate therapy: the GO-FORWARD Study. Ann Rheum Dis 2001;94:309–19. 2009;68:789–96. 70. Saravanan V, Kelly CA. Reducing the risk of methotrexate pneumonitis in 61. Smolen JS, Kay J, Doyle MK, et al.; GO-AFTER study investigators. rheumatoid arthritis. Rheumatology [Oxford] 2004;43:143–7. Golimumab in patients with active rheumatoid arthritis after treatment 71. Bedrossian CW, Sussman J, Conklin RH, Kahan B. Azathioprine-associated with tumour necrosis factor alpha inhibitors [GO-AFTER study]: a interstitial pneumonitis. Am J Clin Pathol 1984;82:148–54.