Analysis of 1060 Cases of Drug-Induced Acute Pancreatitis

Abstract

Acute pancreatitis (AP) has a mortality of approximately 3%.1Parniczky A. et al.PLoS One. 2016; 11e0165309Crossref PubMed Scopus (154) Google Scholar Its reported incidence is variable across countries (10 to 100/100,000 inhabitants), and in the United States, AP is a significant cause of acute hospitalization for gastrointestinal disorders.2Peery A.F. et al.Gastroenterology. 2019; 156: 254-272.e11Abstract Full Text Full Text PDF PubMed Scopus (904) Google Scholar Drug-induced AP (DIAP) is regarded as a rare and mild entity; yet, it is estimated to account for approximately 2% to 5% of AP episodes worldwide.3Nitsche C. et al.Curr Gastroenterol Rep. 2012; 14: 131-138Crossref PubMed Scopus (75) Google Scholar,4Badalov N. et al.Clin Gastroenterol Hepatol. 2007; 5: 648-661Abstract Full Text Full Text PDF PubMed Scopus (300) Google Scholar Because DIAP has no unique features, rechallenge with the offending drug would be the only way to provide the most robust evidence to confirm the etiology. However, giving back the drug only for understanding the etiology of index AP is ethically unacceptable.3Nitsche C. et al.Curr Gastroenterol Rep. 2012; 14: 131-138Crossref PubMed Scopus (75) Google Scholar Therefore, unsurprisingly, DIAP often remains a speculative diagnosis. A more detailed introduction of DIAP can be found in Supplementary Document 1. We aimed to systematically search the medical literature, analyze the outcomes of all reported cases of DIAP, and compare them against a general cohort of AP. We comprehensively searched the literature for reported cases of proven DIAP and extracted detailed data of each case on both first episodes and rechallenges. We compared DIAP to the large cohort of AP cases collected by the Hungarian Pancreatic Study Group. Details of the methods are in Supplementary Document 1. As a result of the comprehensive search and selection, we identified and analyzed 1060 eligible patients in 856 reports (details of the articles can be found in Supplementary Document 2). In a large proportion of the 1060 patients, antineoplastic (179 [16.89%]), antibiotic (128 [12.08%]), and anticonvulsant (103 [9.72%]) drugs caused DIAP. A combination of drugs caused 78 (7.36%) of the DIAP episodes (Figure 1A). In approximately half of the 1060 patients, drugs were given to manage the diseases of the gastrointestinal tract (214 [20.19%]) or neurologic (158 [14.91%]) or hematologic (155 [14.62%]) conditions (Figure 1B). The 10 most common drugs resulting in AP are shown in Figure 1C. Male patients comprised 536 of 1054 patients (50.85%) affected by DIAP (Figure 1D). Interestingly, 228 of 1054 cases (21.63%) were reported in children (aged <18 years), and cases were seen at a younger age than in AP of other common etiologies (Figure 1E). Among the 1060 patients, DIAP was severe in 213 (20.09%), moderately severe in 118 (11.13%), and mild in 729 (68.77%), if the first episodes were analyzed (Figure 1F). DIAP had a mortality of 90 of 1033 (8.71%) for all severities (Figure 1G). There was a significant difference in the median (interquartile range [IQR]) length of hospitalization (LOH) between mild and moderately severe (7 days [IQR, 4–11.5 days] vs 16 days [IQR, 7–25 days], P .001) and between mild and severe DIAP (7 days [IQR, 4 –11.5 days] vs 18 days [IQR, 6.5–42 days], P < .001). There was no difference in the LOH between moderately severe and severe DIAP (Figure 1H). We found information on rechallenge in 960 of the 1060 patients in our analysis. Epidemiology (indication, sex, and age) and outcome parameters (severity, mortality, and LOH) of the rechallenge episodes of DIAP can be found in Figure 2A–F. The association of the drug categories and primary conditions with the severity and mortality rates of DIAP can be found in Supplementary Document 1 and Supplementary Tables 1 and 2. In 147 of 241 patients (70.00%), no data were available on the dose for rechallenge. Rechallenge was performed in 49 of 241 patients (20.33%) with the same dose as given in the first DIAP episode. The dose was decreased in 33 patients (13.69%) and was increased in 12 (4.98%) compared with the drug dose given in the first episode. If the same dose was given that provoked first episode, DIAP was severe in 2 of 41 patients (4.88%), moderate in 3 (7.32%), and mild in 36 (87.80%). If decreased doses were given, we found no moderately severe cases. Rechallenge in 28 patients caused 1 (3.57%) severe case and 27 (96.43%) mild cases of DIAP (Figure 2G). The descriptive statistics of the general AP cohort are available in Supplementary Document 1 and Supplementary Figure 1. Our data showed that severity and mortality were increased in all DIAP cases compared with AP of other etiologies by 18.41% vs 5.63% (P < .001) and 7.30% vs 2.20% (P < .001), respectively. DIAP had the second highest mortality rate of all etiologies (8.49%) (Supplementary Document 1 and Supplementary Figure 1). One of the most critical findings of our study is that compared with AP of other etiologies, reported patients with DIAP have a more severe disease course. Most medications causing severe DIAP are given to treat significant preexisting pathologies and primary diseases such as cancers and autoimmune disorders. These patients will have a higher risk of organ failure. In some patients, organ failure is present at the introduction of the offending drug, before the DIAP event. We hypothesize that this accounts for the increased proportion of moderately severe and severe cases of AP in the DIAP cohort. A primary disease itself is a comorbidity and often has other comorbidities. We believe that the more severe the primary disease was, the higher doses of the offending drugs were used, leading to more severe courses of the DIAP patients. The offending drugs likely cause the DIAP in a dose-dependent way. In our recent meta-analysis, older age led to a more severe disease course,5Marta K. et al.Front Physiol. 2019; 10: 328Crossref PubMed Scopus (24) Google Scholar and our recent cohort analysis proved that comorbidities are more critical in AP than age.6Szakacs Z. et al.Front Physiol. 2018; 9: 1776Crossref PubMed Scopus (0) Google Scholar These conclusions are in line with the findings of the present study and support our above-detailed hypothesis. Besides the negative effect of comorbidities on the outcome of pancreatitis, culprit drugs have direct toxic effects on acinar cells as well. Asparaginase, for example, was shown to cause cellular necrosis.7Peng S. et al.Biol Sci. 2016; 37: 1700Google Scholar Importantly, here we report for the first time that when rechallenge was done with a decreased dose of the offending drug, it resulted in less severe outcomes. The main strength and limitations of this study are in Supplementary Document 1. Here we conclude that reported cases of DIAP have worse outcomes than AP of other etiologies and seem to be dose-dependent. If rechallenge is necessary, we recommend that patients are closely monitored and receive a reduced drug dose. Evidence-based guidelines on DIAP and rechallenge should be developed. Ágnes Meczker MSC (Conceptualization: Equal; Data curation: Equal; Investigation: Equal; Writing – original draft: Lead). Lilla Hanák MSC (Conceptualization: Equal; Data curation: Equal; Formal analysis: Lead; Writing – original draft: Equal). Andrea Párniczky, MD PhD (Formal analysis: Equal; Writing – review & editing: Equal). Andrea Szentesi, PhD (Formal analysis: Equal; Writing – review & editing: Equal). Bálint Erőss, MD, PhD (Conceptualization: Equal; Investigation: Equal; Writing – original draft: Lead). Péter Hegyi, MD, PhD, DSc, MAE (Conceptualization: Equal; Data curation: Equal; Writing – review & editing: Lead). Hungarian Pancreatic Study Group: Erdősi D, Mikó A, Szakács Zs, Dobszai D, and Szapáry L: Institute for Translational Medicine, Medical School, Szentágothai Research Centre, University of Pécs, Pécs, Hungary; Bajor J, Mikó A, and Vincze Á: Department of Gastroenterology, First Department of Medicine, Medical School, University of Pécs, Pécs, Hungary; Matuz M and Csupor D: Department of Clinical Pharmacy, Faculty of Pharmacy, University of Szeged, Hungary; Gábor Pethő: Department of Pharmacology and Pharmacotherapy, Medical School, University of Pécs, Pécs, Hungary; and Heim Pál: Children’s Hospital, Budapest. Drug-induced acute pancreatitis (DIAP) is regarded as a rare entity; yet, it is estimated to account for approximately 2% to 5% of acute pancreatitis (AP) episodes worldwide.1Eland I.A. Van Puijenbroek E.P. Sturkenboom M.J.C.M. Wilson J.H.P. Stricker B.H.C. Drug-associated acute pancreatitis: twenty-one years of spontaneous reporting in the Netherlands.Am J Gastroenterol. 1999; 94: 2417-2422Crossref PubMed Scopus (115) Google Scholar, 2Trivedi C.D. Pitchumoni C.S. Drug-induced pancreatitis: an update.J Clin Gastroenterol. 2005; 39: 709-716Crossref PubMed Scopus (314) Google Scholar, 3Nitsche C. Maertin S. Scheiber J. Ritter C.A. Lerch M.M. Mayerle J. Drug-induced pancreatitis.Curr Gastroenterol Rep. 2012; 14: 131-138Crossref PubMed Scopus (110) Google Scholar, 4Balani A.R. Grendell J.H. Drug-induced pancreatitis: incidence, management and prevention.Drug Saf. 2008; 31: 823-837Crossref PubMed Scopus (154) Google Scholar, 5Vinklerová I. Procházka M. Procházka V. Urbánek K. Incidence, severity, and etiology of drug-induced acute pancreatitis.Dig Dis Sci. 2010; 55: 2977-2981Crossref PubMed Scopus (87) Google Scholar However, estimates vary due to the challenging diagnosis and the difficulties of causality assessment.6Badalov N. Baradarian R. Iswara K. Li J. Steinberg W. Tenner S. Drug-induced acute pancreatitis: an evidence-based review.Clin Gastroenterol Hepatol. 2007; 5 (quiz 664): 648-661Abstract Full Text Full Text PDF PubMed Scopus (403) Google Scholar, 7Naranjo C.A. Busto U. Sellers E.M. et al.A method for estimating the probability of adverse drug reactions.Clin Pharmacol Ther. 1981; 30: 239-245Crossref PubMed Scopus (9017) Google Scholar, 8Karch F.E. Lasagna L. Adverse drug reactions. A critical review.JAMA. 1975; 234: 1236-1241Crossref PubMed Scopus (461) Google Scholar Because DIAP has no unique features that would help in distinguishing a case of DIAP from other etiologies, a rechallenge with the offending drug resulting a relapse of DIAP still means the most reliable evidence in confirming the etiology.3Nitsche C. Maertin S. Scheiber J. Ritter C.A. Lerch M.M. Mayerle J. Drug-induced pancreatitis.Curr Gastroenterol Rep. 2012; 14: 131-138Crossref PubMed Scopus (110) Google Scholar,9Jones M.R. Hall O.M. Kaye A.M. Kaye A.D. Drug-induced acute pancreatitis: a review.Ochsner J. 2015; 15: 45-51PubMed Google Scholar In most cases, intentional rechallenge is considered unethical due to the potentially life-threatening complications of AP; therefore, DIAP remains a speculative diagnosis of exclusion. The subject of past reviews is usually the categorization of the drugs based on their reported frequency of provoking DIAP4Balani A.R. Grendell J.H. Drug-induced pancreatitis: incidence, management and prevention.Drug Saf. 2008; 31: 823-837Crossref PubMed Scopus (154) Google Scholar,10Spanier B.W. Tuynman H.A. van der Hulst R.W. Dijkgraaf M.G. Bruno M.J. Acute pancreatitis and concomitant use of pancreatitis-associated drugs.Am J Gastroenterol. 2011; 106: 2183-2188Crossref PubMed Scopus (46) Google Scholar and the analysis of the strength of the causal relationship between the drug intake and the AP episode.11Mallory A. Kern Jr., F. Drug-induced pancreatitis: a critical review.Gastroenterology. 1980; 78: 813-820Abstract Full Text PDF PubMed Scopus (354) Google Scholar, 12McArthur K.E. Review article: drug-induced pancreatitis.Aliment Pharmacol Ther. 1996; 10: 23-38Crossref PubMed Scopus (95) Google Scholar, 13Wilmink T. Frick T.W. Drug-induced pancreatitis.Drug Saf. 1996; 14: 406-423Crossref PubMed Scopus (139) Google Scholar According to the literature, most cases of DIAP are mild, self-limited, and dose-independent, with a rapid resolution upon discontinuation of the offending drug.14Lankisch P.G. Dröge M. Gottesleben F. Drug induced acute pancreatitis: incidence and severity.Gut. 1995; 37: 565-567Crossref PubMed Scopus (222) Google Scholar However, in our previous study on 5-aminosalicylic acid–induced DIAP, we found that DIAP might not be dose-independent, and we saw more moderately severe cases than expected.15Meczker A. Miko A. Hegyi P. 5-ASA induces mild acute pancreatitis. Case report and review of the literature.J Gastrointestin Liver Dis. 2018; 27: 189-194Crossref PubMed Scopus (11) Google Scholar We performed a systematic literature search according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline.16Moher D. Shamseer L. Clarke M. et al.Preferred Reporting Items for Systematic Review and Meta-analysis protocols (PRISMA-P) 2015 statement.Syst Rev. 2015; 4: 1Crossref PubMed Scopus (14963) Google Scholar The review was registered on PROSPERO under the ID number CRD42017079196. The following PECO items were used: P = patients with AP; E = DIAP; C = AP caused by other etiologies; and O = severity, mortality, length of hospitalization (LOH), imaging alterations, symptoms, and time to resolution of AP. The search was performed in May 2019 on PubMed, EMBASE, and Cochrane Library with the search terms “acute pancreatitis” and “drug” and was limited to English-language and human studies (if applicable), regardless of the date of publication. Study selection was performed in parallel by 2 independent researchers. Studies that contained pooled statistical data of DIAP were excluded because they did not provide relevant data for our analysis. Records that contained relevant data on patients with DIAP were eligible for our study irrespective of study design (case reports included as well). Cases reported as DIAP in which alcoholic or gallstone or different obvious etiology could be suspected were not included. We developed an assessment tool for the reporting quality of the identified articles to exclude poorly reported cases, which would threaten data quality and the analysis. We identified 3 categories of reporting quality, based on the reported symptoms and signs of AP (abdominal pain, pancreatic enzyme elevation, imaging changes) and their causality with the offending drug. Strong evidence: The report contained data sufficient to reevaluate the event as DIAP. Moderate evidence: The report described the event as DIAP, but data only partially confirmed it (could not be reevaluated as AP). Weak evidence: The report described the event as DIAP, but there was no detailed data for reevaluation. In our analysis, we included in the statistical analysis only cases with strong evidence levels. We reevaluated all events documented by the authors as AP. Each was considered as AP if it met the criteria detailed in the evidence-based guidelines for the management of AP.17Working Group IAP/APA Acute Pancreatitis GuidelinesIAP/APA evidence-based guidelines for the management of acute pancreatitis.Pancreatology. 2013; 13: e1-e15PubMed Google Scholar,18Hritz I. Czakó L. Dubravcsik Z. et al.Acute pancreatitis. Evidence-based practice guidelines, prepared by the Hungarian Pancreatic Study Group.Orv Hetil. 2015; 156 ([in Hungarian]): 244-261Crossref PubMed Scopus (44) Google Scholar To assess severity, any organ failure reported by the authors was accepted, even if there were no supporting data. Persistent organ failure was defined that lasted longer than 48 hours or was described as persistent by the original authors themselves, transient organ failure was defined that resolved within 48 hours, or described as transient by the authors. We accepted the pancreatic enzyme level elevation as higher than triple the upper limit of normal if (1) the exact enzyme level and the upper limit of normal were described and the enzyme level exceeded more than 3 the of elevation compared with the upper limit of normal was and was more than and the exact pancreatic enzyme were given their but were higher than in the case of and in the case of We considered the result of rechallenge if a with the suspected offending drug resulted in the of the pancreatic enzyme with or pain, or To the severity of DIAP, we performed an the data by the authors. We each case for the of and complications and organ failure. If the detailed data were the severity of DIAP was by the irrespective of the by the original authors. If the of data did not to the severity of DIAP, we the severity reported by the authors. The offending drugs were given to manage disorders. We defined as the primary the statistical analysis of the DIAP cases to analyze them against AP of other etiologies, we the detailed data of the AP cohort of the Hungarian Pancreatic Study as described in our previous Z. N. D. et and comorbidities in acute pancreatitis a of collected Physiol. 2018; 9: 1776Crossref PubMed Scopus Google Scholar, N. Hanák L. Mikó A. et al.A cohort analysis of cases to support design in acute Physiol. 2019; 10: PubMed Scopus Google Scholar, A. T. E.M. et in acute pancreatitis: from to evidence based 2019; Full Text Full Text PDF PubMed Scopus Google Scholar, A. A. Vincze et in acute pancreatitis: of each Physiol. 2019; 10: PubMed Scopus Google Scholar We descriptive statistical to the and frequency and median and range were To analyze the between the severity for the LOH and the time that the enzyme level and symptoms we the with the as a between drug and disease categories and for DIAP against other etiologies of AP were the We regarded a P of as The analysis was for data. were performed for We compared the severity and mortality rates of DIAP to AP caused by the more common etiologies and the of The detailed descriptive statistics of the AP cohort are shown in Supplementary Figure 1. We found that DIAP showed the most severe episodes if only the first episodes were analyzed of 1060 Supplementary Figure 1). If the severe cases of first and events were the rate of severe cases was but not of were seen between the rate of severe and mortality rates of DIAP cases if the first episodes are compared with at vs (P < .001) and vs (P < .001), respectively. and mortality were increased in all DIAP compared with cases with all other etiologies, at 18.41% vs 5.63% (P < .001) and 7.30% 2.20% (P < .001), respectively. DIAP had the second highest mortality rate of all etiologies AP of alcoholic and etiology had a mortality rate Supplementary Figure 1). and were for the most severe cases of DIAP in 12 of patients of and of respectively. and had the highest mortality at 12 of patients of and 5 of respectively. were P < P < P < and P < .001) had a higher of severe disease than patients other In to patients drugs for disease P < P < .001), or other drugs than the P < had a of a severe than patients other had a higher of moderately severe DIAP P < .001) than patients other P < .001), P < P < P < and patients on medications P < had a higher of mortality than other However, patients on had a of mortality than patients other drugs P < Supplementary 1). The severity of DIAP was analyzed for of primary which showed that patients with hematologic and failure had the highest rates of severe DIAP, at of of and of respectively. gastrointestinal tract disease had a of severe DIAP P < .001) than other had a higher rate of a severe DIAP P < .001) than in other conditions had a higher for a moderately severe DIAP P < .001) than in other was patients with gastrointestinal tract disease P < and higher in diseases P < and diseases P < .001), and diseases P < than in other diseases (Supplementary 2). To our this is the only study that comprehensively searched and identified all DIAP cases in the We a data and quality analysis of each to an of reported cases of DIAP. We this to analyze the of DIAP. Our systematic and comprehensive search identified and resulted in a detailed data of 1060 cases of DIAP. To this is the and most comprehensive analysis of all reported cases of DIAP. Because we collected data on the first episodes of DIAP and on we could compare the 2 study is based on data extracted from case reports and case which all of the limitations of the most and T. R. The case a review of and PubMed Scopus Google Scholar The was increased by the which we had to use due to a large number of identified by the Case reports and which are in which is a The between drug and the of the pancreatitis was not and this is a of our to the of could not be T. R. The case a review of and PubMed Scopus Google Scholar which is significant of the of the The of DIAP rechallenge The of pancreatic enzyme and resolution of an of acute pancreatitis the assessment of pancreatitis following the rechallenge with the suspected drug. Preferred Reporting Items for Systematic Review and Meta-analysis Details of of the Drug the and of Acute based on severity of < < < < < < < < < drugs for < than < < < < P < P < P < in a Supplementary of the the and of Acute based on severity affected by the primary < and < < < < < < P < P < P < in a drug. Drug-induced Acute vs have with the analysis by Meczker et reporting on 1060 patients with drug-induced acute pancreatitis The authors collected data from case reports and case of DIAP and compared them with a Hungarian cohort of patients with acute pancreatitis of any Because case reports and case are in we were the authors did not data from cohort studies for their analysis, because this was not excluded by the study PDF

Topics & Concepts