Pancreatitis as a novel complication of aspirin therapy in patients with aspirin-exacerbated respiratory disease

1684 LETTERS TO THE EDITOR

J ALLERGY CLIN IMMUNOL JUNE 2012

P 5 .001. However, both groups had identical median skin prick test wheal size (7 mm) (Fig 1). The sensitivity, specificity, positive predictive values, and negative predictive values for various egg sIgE cutoffs are provided in Table II. The majority of reactions (71%) occurred before 50% of the total dose had been ingested. Five of the 31 (16%) reactions occurred during the 2-hour observation period after the patient had ingested the entire serving size. Multisystem symptoms occurred in 23% of the OFCs, whereas isolated symptoms affected the gastrointestinal tract (39%), skin (19%), oropharynx (16%), and respiratory tract (3%). The majority of reactions (80.6%) were mild and treated with antihistamine (liquid diphenhydramine or cetirizine) only. In addition to an oral antihistamine, 1 patient received a nebulized bronchodilator (albuterol), 3 patients received oral steroids, and 4 patients received intramuscular epinephrine. No patient received more than 1 injection of epinephrine or required observation beyond 2 hours. There were no reported delayed reactions. We describe results of the first 100 consecutive baked-egg challenges done in our outpatient allergy clinic, confirming that the majority of egg-allergic children tolerate baked egg. Neither clinical history (ever having reaction or anaphylaxis to egg) nor skin prick test provided helpful information to identify the patients who tolerated baked egg. Serum egg white sIgE levels, however, were statistically different between the 2 groups. Varying predictive values are given in Table II. For example, an egg white sIgE cutoff of 2.5 kUA/L had a negative predictive value of 89%, a positive predictive value of 44%, a sensitivity of 87%, and a specificity of 48%. In addition, the symptoms in patients with a serum egg white sIgE of less than 2.5 kUA/L who reacted were mild and treated with oral antihistamines only. On the other hand, if one were to select an egg sIgE of less than 10 in this cohort, the negative predictive value would decrease to 71%, a number that may still be acceptable to some clinicians; however, the specificity would increase to 94%, thus capturing a great deal more children who were tolerant to baked egg. While creating cutoffs from a retrospective cohort may be biased by patient selection, it can provide guidance to the everyday clinician seeking patients who are likely to pass a baked-egg challenge in order to minimize in-office reactions. In addition, given the range of egg white skin prick test wheal sizes (0-19 mm) and sIgE levels (undetectable [<0.35]75.4 kUA/L) in our cohort, we feel that the study population is representative of egg-allergic patients seen in a typical allergy office. Finally, reactions to baked egg were comparable in severity to the reactions reported in our large series of open OFCs; 12.9% of the reactions to baked egg were treated with epinephrine, whereas 9% of the reactions to all OFCs were treated with epinephrine.9 In summary, we report data from 100 consecutive open OFCs to baked egg performed in an allergy office. Given the nutritional, social, and immunologic benefits that baked-egg–containing diets may provide, our data support the conclusion that the benefits of performing outpatient OFCs to baked egg far outweigh the associated risk of the challenge for select patient populations. Jay A. Lieberman, MD Faith R. Huang, MD Hugh A. Sampson, MD Anna Nowak-We˛grzyn, MD From the Division of Allergy and Immunology, Department of Pediatrics, Elliot and Roslyn Jaffe Food Allergy Institute, Mount Sinai School of Medicine, New York, NY. E-mail: [email protected].

Disclosure of potential conflict of interest: J. A. Lieberman has received a grant for salary from the AAAAI/Elliot and Rosalyn Jaffe 3rd year Fellowship. H. A. Sampson is a consultant for Allertein Therapeutics, LLC; is on the advisory board for ImmusanT and Novartis; has received research support from the Food Allergy Initiative (FAI) and the National Institute of Allergy and Infectious Disease/National Institutes of Health (grant no. AI-044236); is a consultant/scientific advisor for the FAI; is a medical advisor for the Food Allergy & Anaphylaxis Network; is a scientific advisor for the University of Nebraska FARRP and the Asthma & Allergy Foundation of America; and is 45% owner of Herbal Springs, LLC. The rest of the authors declare that they have no relevant conflicts of interest. REFERENCES 1. Osborne NJ, Koplin JJ, Martin PE, Gurrin LC, Lowe AJ, Matheson MC, et al. Prevalence of challenge-proven IgE-mediated food allergy using population-based sampling and predetermined challenge criteria in infants. J Allergy Clin Immunol 2011; 127:668-76. 2. Lemon-Mule H, Sampson HA, Sicherer SH, Shreffler WG, Noone S, Nowak-Wegrzyn A. Immunologic changes in children with egg allergy ingesting extensively heated egg. J Allergy Clin Immunol 2008;122:977-83. 3. Nowak-Wegrzyn AH, Leonard SA, Bencharitiwond R, Noone S, Lemon-Mule H, Sampson HA. Follow-up of the Heated Egg (HE) Diet Trial. J Allergy Clin Immunol 2011;127:AB25. 4. Urisu A, Ando H, Morita Y, Wada E, Yasaki T, Yamada K, et al. Allergenic activity of heated and ovomucoid-depleted egg white. J Allergy Clin Immunol 1997;100:171-6. 5. Ando H, Moverare R, Kondo Y, Tsuge I, Tanaka A, Borres MP, et al. Utility of ovomucoid-specific IgE concentrations in predicting symptomatic egg allergy. J Allergy Clin Immunol 2008;122:583-8. 6. Konstantinou GN, Giavi S, Kalobatsou A, Vassilopoulou E, Douladiris N, Saxoni-Papageorgiou P, et al. Consumption of heat-treated egg by children allergic or sensitized to egg can affect the natural course of egg allergy: hypothesis-generating observations. J Allergy Clin Immunol 2008;122:414-5. 7. Knight AK, Shreffler WG, Sampson HA, Sicherer SH, Noone S, Mofidi S, et al. Skin prick test to egg white provides additional diagnostic utility to serum egg whitespecific IgE antibody concentration in children. JAllergy Clin Immunol 2006;117:842-7. 8. Nowak-Wegrzyn A, Assa’ad AH, Bahna SL, et al. Work group report: oral food challenge testing. J Allergy Clin Immunol 2009;123:S365-83. 9. Lieberman JA, Cox AL, Vitale M, Sampson HA. Outcomes of office-based, open food challenges in the management of food allergy. J Allergy Clin Immunol 2011;128:1120-2. doi:10.1016/j.jaci.2012.04.007

Pancreatitis as a novel complication of aspirin therapy in patients with aspirin-exacerbated respiratory disease To the Editor: Aspirin-exacerbated respiratory disease (AERD), previously termed Samter triad, is an adult-onset disease characterized by asthma, hyperplastic chronic sinusitis, nasal polyposis, and hypersensitivity to aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs). Although the upper and lower airway inflammation seen in patients with AERD is exacerbated by the ingestion of aspirin and other NSAIDs, the disease progresses independently of exposure to these medications. Since 1922, when this syndrome was originally identified by Widal,1 and 1967, when it was further described by Samter and Beers,2 the clinical efficacy of aspirin desensitization and ongoing aspirin therapy has been well established.3 Aspirin desensitization is a procedure whereby patients receive increasing doses of aspirin, usually to a dose of 650 mg, over 2 to 3 days. If these patients continue to take aspirin on a daily basis after the desensitization procedure, many of them demonstrate improved control of their asthma and sinus disease. Chronic aspirin therapy can result in adverse drug reactions in a minority of patients with AERD, with gastrointestinal involvement being the most common. Berges-Gimeno et al3 evaluated 172 aspirin-desensitized subjects, 24 (14%) of whom had to discontinue aspirin therapy because of side effects: 14 had epigastric

J ALLERGY CLIN IMMUNOL VOLUME 129, NUMBER 6

pain, 2 had gastrointestinal bleeding, 2 had bleeding from the nose and ear, and 6 had aspirin-induced urticaria. The majority of the gastrointestinal reactions in patients with AERD receiving chronic aspirin are presumed to be either gastritis or peptic ulcer disease, with the former being considered the most common cause. Stevenson and colleagues reported the incidence of gastritis as 18.5% of 65 such patients in their 1990 study4 and 13.6% of 75 different patients involved in their 1996 report.5 However, the patients with gastrointestinal side effects were not extensively studied in regard to the cause of their gastrointestinal distress, and the diagnosis of gastritis in these studies seems to have been made clinically. Although pancreatitis is listed as a rare complication of chronic NSAID use,6 there is no known association between pancreatitis and aspirin therapy in patients with AERD. We describe 3 cases of previously unreported pancreatitis as the cause of abdominal distress during or after aspirin desensitization. The first of these patients is a 55-year-old man who presented with a several-year history of chronic sinusitis and nasal polyposis requiring 2 sinus surgeries, adult-onset asthma, and sensitivity to aspirin and other NSAIDs. On several occasions, NSAID ingestion had caused marked worsening of nasal congestion, exacerbation of his asthma, and severe epigastric pain, which had been attributed to gastroesophageal reflux disease. At presentation, the patient’s systemic medications included omeprazole, glucosamine, montelukast, and simvastatin. During the first day of the patient’s aspirin challenge/desensitization, he had his typical reaction after the 325-mg aspirin dose. Although his respiratory symptoms responded to inhaled bronchodilators and anticholinergics, intravenous corticosteroids, and intravenous H1/H2 antihistamines, his severe epigastric pain progressed and was accompanied by intense cramping and nausea. He had no diarrhea, vomiting, hives, or hypotension. He was admitted to the hospital, where he was found to have an increased lipase level of 425 U/L (normal, 13-63 U/L) and radiographic evidence of pancreatic inflammation but no gallstones or gallbladder sludge on abdominal computed tomography (CT). He was given a diagnosis of pancreatitis and was ordered to take nothing by mouth, with improvement over the next few days. Other common causes of pancreatitis were excluded, including drugs, alcohol, hypercalcemia, hypertriglyceridemia, infection, and biliary tract disease. He did not complete the desensitization procedure, and his lipase level normalized to 55 U/L on repeat testing several days later when asymptomatic. Our second patient is a 59-year-old man with a history of AERD who came to our clinic for a repeat aspirin desensitization procedure. A year prior, he had undergone aspirin desensitization at an outside clinic. On the first day of the previous desensitization, he had a respiratory reaction typical of his respiratory symptoms after NSAID exposure, including increasing upper and lower airway symptoms with a documented decrease in FEV1. He underwent the second day of this initial desensitization without difficulty. Although he did not have any gastrointestinal symptoms on either day of the procedure, he began complaining of severe epigastric pain and nausea the next day. These symptoms persisted for 3 straight days, at which point he was given a clinical diagnosis of gastritis, instructed to discontinue his aspirin therapy, and prescribed a daily proton-pump inhibitor. He improved gradually with this regimen but was not fully asymptomatic from a gastrointestinal standpoint for several weeks. After a year of proton-pump inhibitor therapy with no recurrence of his

LETTERS TO THE EDITOR 1685

gastrointestinal symptoms, the patient decided to attempt aspirin desensitization again because both his asthma and sinus disease had become difficult to control and he was needing frequent oral steroids. At this point, systemic medications included only the following over-the-counter supplements: DHEA, omega-3, and a multivitamin. During day 1 of his repeat aspirin challenge, the patient again exhibited a respiratory reaction with a documented decrease in FEV1 but no gastrointestinal complaints. Shortly after discharge, however, he began to complain of severe epigastric pain, nausea, and vomiting in addition to further worsening of his asthma. He eventually went to the emergency department, where he was given albuterol nebulizer treatments and intravenous corticosteroids for his asthma, as well as antiemetics and pain medications for his gastrointestinal complaints. In addition, he was given a diagnosis of pancreatitis based on an increased lipase level of 789 U/L (normal, 13-63 U/L), hospitalized, and ordered to take nothing by mouth. After an overnight stay, his symptoms improved, and he was discharged on a bland diet. His lipase level the morning after his reaction was 128 U/L, and both amylase and lipase levels were normal at 75 U/L (normal, 30-110 U/L) and 37 U/L, respectively, by later that afternoon. He had an abdominal ultrasound that showed gallbladder sludge but no gallstones. Other causes of pancreatitis, such as drugs, alcohol, and hypercalcemia, were excluded. Our third patient, a 74-year-old woman, also had pancreatic inflammation after her aspirin desensitization procedure but with a somewhat more subacute presentation than in the first 2 cases. She also had a clinical history classic for AERD and tolerated her challenge without gastrointestinal complaints. Her systemic medications at the time included cetirizine, clopidogrel, montelukast, vitamin D3, rosuvastatin, and alendronate. Four days after starting high-dose aspirin therapy at a dose of 650 mg twice daily, she experienced abdominal pain. However, the patient did not mention the discomfort until her follow-up visit 3 weeks later, at which point amylase and lipase measurements were taken and found to be increased at 152 U/L (normal, 30-110 U/L) and 207 U/L (normal, 16-63 U/L), respectively. Her aspirin dose was decreased to 81 mg daily, and she was hospitalized for several days to undergo extensive testing, which included laboratory work, an abdominal CT scan, and both an upper and lower endoscopy, the results of which were within normal limits except for her increased pancreatic enzyme levels. Of particular note, the patient did not have evidence of gallbladder sludge or gallstones on her CT scan. Her abdominal pain improved with the lower aspirin dose, and follow-up amylase and lipase measurements taken 6 weeks later had normalized at 56 and 27 U/L, respectively. Several case reports have been published describing pancreatitis as a result of aspirin use in conjunction with statins,7 but there are no previous reports of pancreatitis complicating aspirin challenges or desensitization procedures in patients with AERD. Although the first and third patients were receiving chronic statin therapy at the time of their procedures, which could have contributed to their pancreatic inflammation, the second patient was not. The development of pancreatic inflammation in all 3 patients only after aspirin challenge/desensitization suggests that pancreatitis might be a complication of this procedure in patients with AERD. More importantly, pancreatitis might be the actual cause of abdominal discomfort in some of the patients with AERD who complain of gastrointestinal side effects rather than gastritis or peptic ulcer disease, which are commonly implicated because they are recognized side effects of aspirin therapy. For our 3

1686 LETTERS TO THE EDITOR

patients, the development of pancreatitis might relate to the marked increase in leukotriene levels during aspirin desensitization from an already high predesensitization level, as is typical of patients with AERD.5 The connection between acute pancreatitis and leukotrienes has been suggested by several animal models. One murine model using sodium taurocholate to induce pancreatitis led to increased levels of several leukotrienes.8 When cerulein is used to induce pancreatitis, knockout mice lacking the 5-lipoxygenase enzyme crucial for leukotriene synthesis have less pancreatic inflammation and tissue injury, neutrophil infiltration, and adhesion molecule expression compared with mice that do not lack 5-lipoxygenase.9 Using this same model in rats, blocking the effects of leukotrienes at the level of their receptors has been shown to decrease pancreatic inflammation.10 Interestingly, all 3 patients received a dose of montelukast before aspirin challenge, and the first and third patients were taking montelukast on a chronic basis at the time of the challenge. This suggests that the leukotriene receptor antagonism provided by montelukast is incomplete and might be insufficient to block the effects of leukotrienes in this subset of patients with levels of leukotrienes and leukotriene receptors that are high at baseline and increase further during the aspirin challenge. Ultimately, however, the mechanism of pancreatitis in these patients is not yet understood, and the role of leukotrienes is just a theoretical construct at this point. Although our observations relate to just 3 cases, the temporal association between aspirin desensitization and the onset of gastrointestinal distress with normal hepatic panels but evidence for pancreatic inflammation suggests that pancreatitis, although possibly rare, might be a previously unrecognized complication of aspirin desensitization. In addition, the third patient described here raises the possibility that pancreatitis might be a dosedependent complication of aspirin challenge and long-term aspirin therapy because decreasing her aspirin dose ultimately led to resolution of her abdominal discomfort and normalization of her pancreatic enzyme levels. As in all cases of pancreatitis, secondary causes, such as gallbladder disease, must be ruled out, and the possibility of idiopathic pancreatitis always exists. These 3 cases exhibit a strong temporal association that would be difficult to explain based on gallbladder pathology or chance alone. Until now, cases of pancreatitis during or shortly after aspirin desensitization have likely been classified erroneously as gastritis, given the relatively high prevalence of gastritis in chronic users of high-dose aspirin or other NSAIDs.11 To better elucidate the association between pancreatic inflammation and

J ALLERGY CLIN IMMUNOL JUNE 2012

aspirin desensitization in patients with AERD, we will prospectively examine the effect of high-dose aspirin on the pancreas in patients with AERD undergoing desensitization. On the basis of these 3 cases, the authors recommend that patients with AERD who experience gastrointestinal symptoms either during their aspirin desensitization procedure or days after and those patients with AERD with underlying pancreatic disease should have their pancreatic enzyme levels monitored. Flavia C. L. Hoyte, MD Richard W. Weber, MD Rohit K. Katial, MD From the Allergy and Immunology Clinic, National Jewish Health, Denver, Colo. E-mail: [email protected]. Disclosure of potential conflict of interest: The authors declare that they have no relevant conflicts of interest.

REFERENCES 1. Widal F, Abrami P, Lermoyez J. Anaphylaxie et idiosyncrasie. Presse Medicale 1922;30:189-93. 2. Samter M, Beers RF Jr. Concerning the nature of intolerance to aspirin. J Allergy 1967;40:281-93. 3. Berges-Gimeno MP, Simon RA, Stevenson DD. Long-term treatment with aspirin desensitization in asthmatic patients with aspirin-exacerbated respiratory disease. J Allergy Clin Immunol 2003;111:180-6. 4. Sweet JM, Stevenson DD, Simon RA, Mathison DA. Long term effects of aspirin desensitization—treatment for aspirin sensitive rhinosinusitis asthma. J Allergy Clin Immunol 1990;86:59-65. 5. Stevenson DD, Hankammer MA, Mathison DA, Christiansen SC, Simon RA. Long term ASA desensitization-treatment of aspirin sensitive asthmatic patients: clinical outcome studies. J Allergy Clin Immunol 1996;98:751-8. 6. Trivedi CD, Pitchumoni CS. Drug-induced pancreatitis. J Clin Gastroenterol 2005; 39:709-16. 7. Antonopoulos S, Mikros S, Kokkoris S, Protopsaltis J, Filioti K, Karamanolis D, et al. A case of acute pancreatitis possibly associated with combined salicylate and simvastatin treatment. JOP 2005;6:264-8. 8. Folch E, Closa D, Prats N, Gelpi E, Rosello-Catafau J. Leukotriene generation and neutrophil infiltration after experimental acute pancreatitis. Inflammation 1998;22: 83-93. 9. Cuzzocrea S, Rossi A, Serraino I, Di Paola R, Dugo L, Genovese T, et al. 5-lipoxygenase knockout mice exhibit a resistance to acute pancreatitis induced by cerulein. Immunology 2003;110:120-30. 10. Hirano T. Peptide leukotriene receptor antagonist diminishes pancreatic edema formation in rats with cerulein-induced acute pancreatitis. Scand J Gastroenterol 1997;32:84-8. 11. Leong RWL, Chan FFKL. Drug-induced side effects affecting the gastrointestinal tract. Expert Opin Drug Saf 2006;5:585-92. Available online January 10, 2012. doi:10.1016/j.jaci.2011.12.003