Frank Andersohn

Systematic review: agranulocytosis induced by nonchemotherapy drugs.

Context Drug-induced agranulocytosis is a rare but potentially serious adverse event. Contribution This systematic review of case reports involving 980 patients with possible nonchemotherapy druginduced agranulocytosis found 125 drugs that definitely or probably caused agranulocytosis. More than half of the definite or probable cases involved the following drugs: carbimazole, clozapine, dapsone, dipyrone, methimazole, penicillin G, procainamide, propylthiouracil, rituximab, sulfasalazine, and ticlopidine. More patients with a neutrophil count nadir less than 0.1109 cells/L had fatal complications than did those with a nadir of 0.1109 cells/L or greater (10% vs. 3%). Caution Case reports do not describe the frequency of drug-induced complications and sometimes emphasize atypical features and outcomes. The Editors Nonchemotherapy druginduced agranulocytosis is a rare adverse reaction that is characterized by a decrease in peripheral neutrophil count to less than 0.5109 cells/L due to immunologic or cytotoxic mechanisms (1). Discontinuation of treatment with the causative drug is crucial to avoid continuation of hazardous exposure, but its identification may be challenging in patients with polypharmacotherapy. In a subgroup of patients, detection of drug-dependent antineutrophil antibodies can prove drug causality (2), but this method lacks standardization and is not appropriate for drugs that cause agranulocytosis by nonimmunologic mechanisms. The prescribing information is also often not helpful in identifying these drugs because leukopenia and neutropenia are listed as possible adverse drug reactions for many drugs, but no information is provided on the severity of these blood count changes. Reports on fatal cases of drug-induced acute agranulocytosis differ materially and range from 0% to 23% (39). A decrease in the case fatality rate over time has been reported for dipyrone (4), but whether it applies to drug-induced acute agranulocytosis in general is unknown. Granulocyte colony-stimulating factor (G-CSF) and granulocyte macrophage colony-stimulating factor (GM-CSF) are proven treatments for chemotherapy-induced neutropenia, but results on their effectiveness in nonchemotherapy druginduced agranulocytosis are conflicting. To establish a comprehensive update of drugs that cause agranulocytosis, we conducted a systematic review of published case reports. We reevaluated the evidence of causality by applying the standardized World Health Organization criteria for causality assessment of adverse drug reactions to each report. In addition, we analyzed the reports with respect to several clinical characteristics, including the efficacy of therapy with G-CSF or GM-CSF. Methods We followed a set protocol to identify case reports and extract clinical information (all included case reports are listed at www.adverse-effects.com/agranulocytosis/case_reports.html). We plan to update this database every 2 years. Literature Sources and Search Terms We searched MEDLINE for case reports of drug-induced agranulocytosis that were published from January 1966 to December 2006 by using the Medical Subject Heading terms agranulocytosis and chemically induced. The search was limited to articles involving the terms human and case reports and to articles written in German or English. We searched EMBASE for all articles that were published in German or English between 1989 and December 2006 and were described with the terms agranulocytosis-side-effect in the major medical descriptors field and case-report in a subheading field. From all articles that were identified in MEDLINE and EMBASE, 1 reviewer selected the reports to be reviewed, and another reviewer conducted the causality assessment and manually checked the references of all retrieved case reports and case series to identify articles that were not found during the primary search. Selection of Case Reports Articles were eligible if they reported on a case of nonchemotherapy druginduced acute agranulocytosis or neutropenia with an absolute neutrophil count less than 0.5109 cells/L. We excluded articles from further review if they fulfilled at least 1 of the following criteria: no individual patient reported; insufficient clinical or laboratory data to permit further evaluation of the report; absolute neutrophil count of 0.5109 cells/L or greater; suppression of bone marrow caused by cytotoxic drugs; diagnosis of aplastic anemia; intentional or accidental drug overdose; congenital agranulocytosis, cyclic neutropenia, or neutropenia due to underlying diseases or causes other than drugs (for example, malignant infiltration of bone marrow); or agranulocytosis in patients younger than 18 years of age (because the spectrum of drugs, comorbid conditions, and outcome of agranulocytosis in children may differ from that in adults). Data Abstraction We extracted several characteristics from the cases, including the patients sex, age, medical history, other concomitant cytopenias or immunologic drug reactions (for example, exanthema), and neutrophil count nadir; duration, complications, treatment, and outcome of agranulocytosis; latency between start of treatment and onset of acute agranulocytosis; and results of bone marrow examination, if available. We used the World Health Organization categories certain, probable, possible, and unlikely for causality assessments (10) but did not use the category unclassifiable because articles that provided insufficient clinical or laboratory data were excluded. We assessed 4 aspects of the suspected drug reaction: 1) the existence of a plausible time relationship to drug administration, 2) the absence of concurrent disease or other drugs that may have caused acute agranulocytosis, 3) the existence of a reasonable response to drug withdrawal (dechallenge), and 4) the existence of a satisfactory rechallenge procedure or of a definitive pharmacologic explanation for acute agranulocytosis (for example, confirmation of causality by detecting drug-dependent antineutrophil antibodies). For a reaction to be rated certain, all criteria had to be met. A reaction was rated as probable if criteria 1, 2, and 3 were fulfilled and possible if only criterion 1 was met and information on criterion 3 was lacking or unclear. A reaction was rated as unlikely if criterion 1 was not met and if other drugs, chemicals, or underlying disease provided a plausible explanation for acute agranulocytosis. According to the Council for International Organizations of Medical Sciences criteria of causality assessment of drug-induced cytopenias (11), a plausible time relationship to drug administration was suggested if acute agranulocytosis occurred during current therapy or within 1 month after drug withdrawal and if it did not resolve spontaneously during continuous therapy. This window was extended to 6 months for rituximab-induced, delayed-onset neutropenia, which has repeatedly been reported to occur later than 1 month and up to 6 months after the last rituximab dose. A reasonable response to drug withdrawal was defined as an increase in neutrophil count to more than 1.5109 cells/L within 1 month. After causality assessments, we assigned a level of evidence to each reported drug: Level 1 evidence had to have at least 1 definite case report, level 2 evidence had to have at least 1 probable report but no definite report, and level 3 evidence had to have at least 1 possible report but no definite or probable case report. Statistical Analysis We analyzed the proportion of fatal cases reported by time period and tested for trends across 1966 to 1980, 1981 to 1990, and 1991 to 2006. We repeated this analysis after excluding patients who were treated with hematopoietic growth factors. For drugs with more than 3 definite or probable reports of agranulocytosis and complete information on the investigated variables, we analyzed the median duration of treatment before disease onset and the median time between onset and normalization of neutrophil count. To assess the natural course of the disease, patients who were treated with G-CSF or GM-CSF were excluded in the calculation of the median duration of neutropenia. We compared the proportion of patients who developed localized infections or sepsis or who died in cases with a neutrophil count nadir less than 0.1109 cells/L with those who had a nadir of 0.1109 cells/L or greater. We repeated these comparisons in the subgroup of patients who were asymptomatic at diagnosis and explored relationships between treatment with G-CSF or GM-CSF and duration of neutropenia, infectious complications, and death by using reports published since 1990, when hematopoietic growth factors became widely available. We used SAS software, version 9.1 (SAS Institute, Inc., Cary, North Carolina) to perform statistical tests. We compared categorical variables by using the Fisher exact test. We used the CochranArmitage test to test for trends in binomial proportions across ordinal-scaled levels of a single explanatory variable and the Student t-test to compare a continuous variable (for example, duration of acute agranulocytosis) in different groups. If the data in 1 of the groups were not normally distributed, a Wilcoxon rank-sum test was used. A 2-sided P value less than 0.05 was considered statistically significant. Role of the Funding Source This study was not funded. Results We identified 1135 potentially relevant articles by searching MEDLINE and EMBASE and identified 293 additional articles by checking the bibliographies of retrieved articles (Figure 1). Of these articles, 672 reports on 980 patients met the inclusion criteria. Of the 980 reported cases of drug-induced agranulocytosis, 56 (6%) reactions were rated as definite, 436 (44%) were rated as probable, 481 (49%) were rated as possible, and 7 (1%) were rated as unlikely. Table 1 shows the clinical characteristics of all case reports that were at least possibly related to the suspected drug (n= 973). Thirt

Long-Term Use of Antidepressants for Depressive Disorders and the Risk of Diabetes Mellitus

OBJECTIVE Use of antidepressants has been reported to cause considerable weight gain. The aim of this study was to assess the risk of diabetes mellitus associated with antidepressant treatment and to examine whether the risk is influenced by treatment duration or daily dose. METHOD This was a nested case-control study in a cohort of 165,958 patients with depression who received at least one new prescription for an antidepressant between January 1, 1990, and June 30, 2005. Data were from from the U.K. General Practice Research Database. Patients were at least 30 years of age and without diabetes at cohort entry. RESULTS A total of 2,243 cases of incident diabetes mellitus and 8,963 matched comparison subjects were identified. Compared with no use of antidepressants during the past 2 years, recent long-term use (>24 months) of antidepressants in moderate to high daily doses was associated with an increased risk of diabetes (incidence rate ratio=1.84, 95% CI=1.35-2.52). The magnitude of the risk was similar for long-term use of moderate to high daily doses of tricyclic antidepressants (incidence rate ratio=1.77, 95% CI=1.21-2.59) and selective serotonin reuptake inhibitors (incidence rate ratio=2.06, 95% CI=1.20-3.52). Treatment for shorter periods or with lower daily doses was not associated with an increased risk. CONCLUSIONS Long-term use of antidepressants in at least moderate daily doses was associated with an increased risk of diabetes. This association was observed for both tricyclic antidepressants and selective serotonin reuptake inhibitors.

Use of First- and Second-Generation Cyclooxygenase-2–Selective Nonsteroidal Antiinflammatory Drugs and Risk of Acute Myocardial Infarction

Background— The cardiovascular safety of cyclooxygenase (COX)-2–selective nonsteroidal antiinflammatory drugs (NSAIDs) has come under scrutiny after the withdrawal of rofecoxib and halting of the Adenoma Prevention with Celecoxib trial. Whether the newer second-generation COX-2 inhibitors (etoricoxib, valdecoxib) also increase the cardiovascular risk is unknown. Methods and Results— We performed a nested case-control study in a cohort of 486 378 persons registered within the United Kingdom General Practice Research Database with at least 1 prescription of an NSAID between June 1, 2000, and October 31, 2004. A total of 3643 cases with acute myocardial infarction (AMI) were matched to 13 918 controls on age, sex, year of cohort entry, and general practice. Rate ratios (RRs) of AMI associated with use of COX-2–selective and –nonselective NSAIDs were calculated. Current use of etoricoxib was associated with a 2.09-fold (95% confidence interval [CI], 1.10 to 3.97) risk of AMI compared with no use of NSAIDs during the prior year. Current use of rofecoxib (RR=1.29; 95% CI, 1.02 to 1.63), celecoxib (RR=1.56; 95% CI, 1.22 to 2.00), and diclofenac (RR=1.37; 95% CI, 1.17 to 1.59) also significantly increased the AMI risk. For current use of valdecoxib, the RR was 4.60 (95% CI, 0.61 to 34.51). RRs appeared to increase with higher daily doses of COX-2 inhibitors and were also increased in patients without major cardiovascular risk factors. Conclusions— Our study supports the hypothesis that the elevated risk of AMI is a class effect of COX-2 inhibitors. The increase in risk appears to be dose dependent, but further data are needed to verify this observation.

Use of antiepileptic drugs in epilepsy and the risk of self-harm or suicidal behavior

Background: A recent meta-analysis of randomized trials revealed that antiepileptic drugs (AEDs) as a class increase the risk of suicidal thoughts and behavior. We conducted an observational study with data from the United Kingdom General Practice Research Database to investigate if an increase in risk for different groups of AEDs is also evident in clinical practice. Methods: This was a nested case-control study in a cohort of 44,300 patients with epilepsy who were treated with AEDs. Patients with self-harm or suicidal behavior were identified by predefined codes. We included 453 cases and 8,962 age-matched and sex-matched controls. AEDs were classified into 4 groups: barbiturates, conventional AEDs, and newer AEDs with low (lamotrigine, gabapentin, pregabalin, oxcarbazepine) or high (levetiracetam, tiagabine, topiramate, vigabatrin) potential of causing depression. Adjusted odds ratios (OR) were calculated using conditional logistic regression. Results: Current use of newer AEDs with a high potential of causing depression was associated with a 3-fold increased risk of self-harm/suicidal behavior (OR = 3.08; 95% [CI] 1.22–7.77) as compared with no use of AEDs during the last year. Use of barbiturates (OR = 0.66; 95% CI 0.25–1.73), conventional AEDs (OR = 0.74; 95% CI 0.53–1.03), or low-risk newer AEDs (OR = 0.87; 95% CI 0.47–1.59) was not associated with an increased risk. Conclusions: Newer AEDs with a rather high frequency of depressive symptoms in clinical trials may also increase the risk of self-harm or suicidal behavior in clinical practice. For the most commonly used other groups of AEDs, no increase in risk was observed.

Cyclooxygenase-2 Selective Nonsteroidal Anti-Inflammatory Drugs and the Risk of Ischemic Stroke: A Nested Case-Control Study

Background and Purpose— Several randomized trials and a large number of epidemiological studies have provided evidence of an increased risk of acute myocardial infarction associated with the use of cyclooxygenase (COX)-2 selective nonsteroidal anti-inflammatory drugs (NSAIDs). Few data are available concerning the risk of ischemic stroke associated with COX-2 inhibitors. Methods— We performed a nested case-control study in a cohort of 469 674 patients registered within the UK General Practice Research Database (GPRD), who had at least 1 prescription of an NSAID between June 1, 2000 and October 31, 2004. A total of 3094 cases with ischemic stroke were identified and 11 859 controls were matched on age, sex, year of cohort entry and general practice. Odds ratios (ORs) of ischemic stroke associated with the use of COX-2 selective NSAIDs were calculated by conditional logistic regression. Results— Current use of rofecoxib (OR=1.71; 95% CI, 1.33 to 2.18), etoricoxib (OR=2.38; 95% CI, 1.10 to 5.13), but not of celecoxib (OR=1.07; 95% CI, 0.79 to 1.44) was associated with a significantly increased risk of ischemic stroke. For rofecoxib and etoricoxib, ORs tended to increase with higher daily dose and longer duration of use and were also elevated in patients without major stroke risk factors. Conclusions— Our study suggests that COX-2 selective NSAIDs differ in their potential to cause ischemic cerebrovascular events. An increased risk of ischemic stroke may be influenced by additional pharmacological properties of individual COX-2 inhibitors.

Signals of progressive multifocal leukoencephalopathy for immunosuppressants: a disproportionality analysis of spontaneous reports within the US Adverse Event Reporting System (AERS).

Progressive multifocal leukoencephalopathy (PML) is a rare demyelinating disease of the central nervous system that has been reported as rare adverse drug reaction (ADR) of immunosuppressive drugs. We aimed to study signals of PML for immunosuppressants using a disproportionality analysis of spontaneous adverse event reports.

Drug-induced immune thrombocytopaenia: results from the Berlin Case–Control Surveillance Study

PurposeDrug-induced immune thrombocytopaenia is a rare, serious condition that can be triggered by numerous medications. To characterize the spectrum of drugs associated with immune thrombocytopaenia (ITP) in the Berlin Case–Control Surveillance Study (FAKOS).MethodsAdult hospitalized patients with new onset idiopathic, secondary or drug-induced acute ITP and hospital control patients were ascertained by active surveillance in 50 Berlin hospitals (>180 clinical departments) between 2000 and 2009. Drug exposures were obtained in a personal interview. Chronic cases were excluded in a follow-up after 6 or more months. A standardized causality assessment was conducted for each ITP patient to assess possible drug aetiology. Drug risks were quantified in a case–control design with unconditional logistic regression analysis.ResultsNinety out of 169 validated cases of acute ITP were assessed as being at least possibly drug-related (n = 85 different drugs overall, n = 30 drugs with certain or probable causality). Drugs involved in ≥2 cases with a probable or certain relationship were tirofiban (n = 10 cases), abciximab (n = 4), trimethoprim/sulphamethoxazole (n = 4), influenza vaccine (n = 3), and citalopram (n = 2). Pneumococcal and poliomyelitis vaccine were assessed as probably causing ITP in one case each. In the case–control analyses, significantly increased risks were observed for tirofiban, abciximab, trimethoprim/sulphamethoxazole, gentamicin, triamterene/hydrochlorothiazide, drospirenone/ethinylestradiol, and influenza vaccination.ConclusionsOur study confirms known ITP risks for glycoprotein IIb/IIIa receptor antagonists and sulphonamides and generates signals for several other drugs and vaccines. New onset of ITP should not only direct attention to drugs as possible aetiological agents, but also to vaccines that are known to cause autoimmune phenomena.

Risk of ischemic stroke in patients with Crohn's disease: a population-based nested case-control study.

Background: Observational studies have linked Crohns disease (CD) to an increased risk of venous thromboembolic complications. Case reports of ischemic stroke in CD patients have raised the question of a similar association, but data from observational studies are lacking. Methods: Using data from the UK General Practice Research Database we conducted a nested case‐control analysis within a population‐based cohort of 8054 patients with and 161,078 patients without CD. A total of 1748 cases of ischemic stroke were identified to whom 17,348 controls were matched on age, sex, and year of cohort entry. Adjusted odds ratios (ORs) of ischemic stroke associated with CD were calculated by conditional logistic regression. Stratified analyses were performed for age and sex. Results: While CD was not associated with an overall increased risk of ischemic stroke (OR 1.10, 95% confidence interval [CI] 0.85–1.43), stratified analyses revealed an increase in risk in younger patients (<50 years: OR 2.93; 95% CI 1.44–5.98) but not in elderly patients (≥50 years: OR 0.99; 95% CI 0.75–1.30; P for interaction <0.01). The interaction with age remained statistically significant even after changing the cutoff value for the younger and older age group to 45, 55, or 60 years in a sensitivity analysis. There was no interaction with sex (P = 0.79). Conclusions: The study indicates that younger patients with CD may be under an increased risk of ischemic stroke. Inflamm Bowel Dis 2010

Drug induced immune haemolytic anaemia in the Berlin Case-Control Surveillance Study.

Drug‐induced immune haemolytic anaemia is a rare but serious condition. This study investigated the possibility of drug aetiology of immune haemolytic anaemia (IHA) in 134 patients with new onset of IHA who were identified in the Berlin Case‐Control Surveillance Study between 2000 and 2009. Single drugs related to IHA in three or more patients and assessed more than once as a certain or probable cause of IHA in a standardized causality assessment included diclofenac, fludarabine, oxaliplatin, ceftriaxone and piperacillin. In a case‐control study including all 124 IHA cases developed in outpatient care and 731 controls, significantly increased odds ratios (OR) were observed for beta‐lactam antibiotics (OR = 8·8; 95% confidence interval [CI] 3·2–25·2), cotrimoxazole (OR = 6·5; CI 1·1–37·9), ciprofloxacin (OR = 6·9, CI 1·3–38·5), fludarabine (OR = 22·2; CI: 2·8–454·5) and lorazepam (OR = 5·3; CI: 1·2–21·2). Excluding new onset cases with a chronic IHA disease course, an increased risk became also apparent for diclofenac with an OR of 3·1 (CI 1·3–7·0). This is the first case‐control study investigating drugs as risk factors for IHA. It corroborates an increased risk for several drugs that have been implicated as a cause of IHA in the standardized causality assessment of individual cases.

Priapism Associated With Antipsychotics: Role of α1 Adrenoceptor Affinity

Case reports indicate that antipsychotics can cause priapism, a persistent penile erection possibly leading to erectile dysfunction. The mechanism of antipsychotic-induced priapism is thought to be related to blockade of &agr;1 adrenergic receptors, but clinical data supporting this hypothesis are lacking. The aim of this study was to investigate if the presence of safety signals for antipsychotics and priapism is associated with their &agr;1 affinity. Spontaneous reports of adverse drug reactions contained in the US Adverse Event Reporting System database were used to calculate reporting odds ratios (RORs) of priapism for antipsychotics. In total, 426 cases of priapism with 144 of them attributed to antipsychotics were identified. For antipsychotics with high &agr;1 affinity, the adjusted ROR was markedly elevated (ROR = 9.9; 95% CI, 7.9-12.4), whereas a weaker signal was observed for antipsychotics with low/medium &agr;1 affinity (ROR = 3.6; 95% CI, 2.4-5.2). Signals were present for chlorpromazine, quetiapine, risperidone, ziprasidone, and aripiprazole. After restricting the analysis to cases with medical intervention or disability, the safety signal remained evident only for antipsychotics with high but not for those with low/medium affinity. The observed pattern of signals indicates a relationship between &agr;1 affinities of antipsychotics and the occurrence of priapism.