Evaluation of cases of severe statin-related transaminitis within a large health maintenance organization

The American Journal of Medicine (2005) 118, 618 624 CLINICAL RESEARCH STUDY Evaluation of cases of severe statin-related transaminitis within a large health maintenance organization Evguenia C. Charles, PharmD, a,b Kari L. Olson, BSc(Pharm), PharmD, a,b Brian G. Sandhoff, PharmD, a,b David L. McClure, MSc, c John A. Merenich, MD a,d a Clinical Pharmacy Cardiac Risk Service, Kaiser Permanente of Colorado; b Clinical Assistant Professor, University of Colorado School of Pharmacy, Denver, Colorado; c Clinical Research Unit, Kaiser Permanente of Colorado; d Clinical Associate Professor of Medicine, University of Colorado Health Sciences Center, Denver, Colorado. KEYWORDS: 3-hydroxy-3- methylglutaryl coenzyme reductase inhibitors; Alanine aminotransferase; Liver ABSTRACT PURPOSE: To describe the rate, potential causes, symptoms, time to onset, and time to resolution of severe transaminitis associated with increased 3-hydroxy-3-methylglutaryl coenzyme reductase inhibitor ( statin ) usage in a large group model health maintenance organization. SUBJECTS AND METHODS: Health plan members 18 years of age and older, not receiving chemotherapy, who had received at least 1 statin prescription between January 1, 1997, and December 31, 2001 were eligible. All eligible patients with an alanine aminotransferase greater than 10 times the upper limit of normal at any time during the study period were identified using computerized laboratory records. Medical records were subsequently reviewed in order to determine whether the elevation was attributable to statin therapy. RESULTS: Alanine aminotransferase had never been measured in 2334 of 25 334 (9%) of eligible patients. In the remaining 23 000 patients, 62 (0.3%) were identified with an alanine aminotransferase greater than 10 times the upper limit of normal during the 5-year study period. Of these, 17 (0.1% of 23 000 patients) had severe transaminitis deemed directly attributable to statin use. All except 4 of these 17 cases were associated with drug interactions. In 16 cases, transaminitis resolved upon statin discontinuation. CONCLUSIONS: In the observed study sample, statin-related severe transaminitis occurred infrequently. These findings support less frequent liver enzyme monitoring for most patients on statins. Continued monitoring remains warranted for patients on concomitant medications or those with comorbid conditions. 2005 Elsevier Inc. All rights reserved. Since lovastatin, the first 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor ( statin ) was launched onto the market in 1987, statins have become one of the most Requests for reprints should be addressed to Kari L. Olson, BSc (Pharm), PharmD, BCPS, Clinical Pharmacy Cardiac Risk Service, 16601 East Centretech Parkway, Aurora, CO 80011. E-mail address: kari.olson@kp.org. frequently prescribed classes of medications. Although statins appear to be safe, animal and pre-marketing clinical trials have shown signs of liver toxicity, mostly manifesting as minor elevations in alanine aminotransferase and aspartate aminotransferase concentrations. 1 In large clinical trials, alanine aminotransferase elevations of greater than 3 times the upper limit of normal occurred in less than 2% of patients. 2-6 The prevalence of severe transaminitis (defined 0002-9343/$ -see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.amjmed.2005.02.008

Charles et al Statin-related Transaminitis 619 as an alanine aminotransferase level greater than 10 times the upper limit of normal) was even lower. 2-6 With statin use increasing, determining the prevalence of severe transaminitis in a real-world setting and whether regular monitoring of liver enzyme levels in all patients is necessary becomes an important clinical question. The objective of this study was to examine the rate, potential causes, symptoms, time to onset, and time to resolution of severe transaminitis associated with statin usage in a large sample of patients in a group model health maintenance organization. Methods Setting This retrospective review was conducted at Kaiser Permanente of Colorado, a large group model health maintenance organization in the western United States with approximately 385 000 enrolled members. Approval to conduct the study was obtained from the Institutional Review Board of Kaiser Permanente Northern California. Patient sample All patients 18 years of age and older enrolled in Kaiser Permanente of Colorado between January 1, 1997, and December 31, 2001, who had received at least 1 statin prescription within this period were considered for evaluation. Patients receiving anti-neoplastic therapy were excluded due to the potential for these agents to cause liver enzyme abnormalities. Patients who were no longer Kaiser Permanente members by the end of the study period were also excluded. For any excluded patients in this latter category with an alanine aminotransferase level greater than 10 times the upper limit of normal, the medial record was screened to identify any potential statin-related deaths. Procedures Computerized laboratory and pharmacy data were used to identify all patients with severe transaminitis who were receiving lovastatin, simvastatin, pravastatin, atorvastatin, fluvastatin, and cerivastatin. Severe transaminitis was defined as an alanine aminotransferase level greater than 10 times the upper limit of normal ( 420 IU/L) at any time after the first statin prescription was dispensed. 7 NCEP III guidelines recommend transaminase monitoring (alanine aminotransferase or aspartate aminotransferase) to assess liver function in patients on statin therapy. 8 Alanine aminotransferase, rather than aspartate aminotransferase, was chosen as the marker of severe transaminitis as it has demonstrated greater sensitivity and specificity for drug-induced liver dysfunction and is the preferred test at Kaiser Permanente for liver function monitoring in patients on statin therapy. 9,10 Medical records for each patient with alanine aminotransferase levels greater than 10 times the upper limit of normal were reviewed independently by 4 investigators (1 physician and 3 clinical pharmacy specialists). Alternative causes of alanine aminotransferase elevations were examined for each patient along with the presence of comorbid conditions known to affect alanine aminotransferase concentrations. Specifically, these included autoimmune or viral hepatitis, cholelithiasis, fatty liver, and/or pancreatitis. The presence of concomitant medications known to interact with statins was evaluated with particular attention given to medications metabolized through the same metabolic pathways as statins (ie, cytochrome P450 3A4). The timing of events including de-challenge/re-challenge data was evaluated for each case. All cases were classified as either unrelated or possibly related to statin therapy according to predetermined definitions. 10 Cases that were determined to be possibly related to statin therapy were then scored using the Naranjo probability scale for assessing drug-related adverse reactions. 11 The Naranjo criteria, which have been validated for the assessment of drug-induced adverse events, consider the onset, course of reaction, and possible disease and drug alternatives. 11 A score of greater than 9 indicates a high probability of drug-induced adverse events, 5-8 is probable, 1-4 is possible, and less than 1 is doubtful. 11 Cases in which alanine aminotransferase elevation preceded statin initiation were considered unrelated to statin therapy. Instances of negative re-challenge, defined as a retrial of the same or different statin after an initial alanine aminotransferase elevation in which subsequent alanine aminotransferase results remained within normal limits, were also considered unrelated to statin therapy. 10 However, if a re-challenge with the same or different statin resulted in an alanine aminotransferase elevation greater than 42 IU/L upon subsequent testing, it was considered related. 10 In cases in which an alanine aminotransferase elevation resulted in statin discontinuation, laboratory data were examined to determine the clinical course after drug cessation. In situations in which the alanine aminotransferase did not decrease by at least 50% 2 months after statin discontinuation, the elevation was considered unrelated to the statin. A period of 1 month is suggested for the evaluation of drug-induced liver disorders; however, we chose 2 months to allow sufficient time for laboratories to be completed. 10 Unanimous agreement among the reviewing investigators was required for an episode of transaminitis to be considered related to statin therapy. Analysis Results were summarized using common methods such as calculating medians, maximum and minimum values, means, standard deviations, and percentages. Percentages refer to the proportion of the total evaluable study sample

620 The American Journal of Medicine, Vol 118, No 6, June 2005 Table 1 Baseline characteristics of the 23 000 patients who had alanine aminotransferase measured at least once after initiation of a statin Patient Characteristics n 23 000 Average age (years) 62.4 11.5 Age range (years) 21 to 99 Women (%) 44.9 Average statin duration (years) 3.9 1.6 Median statin dose mg per day (min, max) Lovastatin 10 (10,40) Simvastatin 40 (5,80) Atorvastatin 20 (10,80) Pravastatin 20 (10,40) Fluvastatin 20 (20,40) Median number of days to first ALT* 88 (0,2526) (min, max) *ALT: alanine aminotransferase. Figure 1 Patient disposition. ALT alanine aminotransferase; ULN upper limit of normal. that had alanine aminotransferase measurements available unless otherwise noted. Means are reported with accompanying standard deviations. Results A total of 30 830 patients received at least 1 prescription for a statin during the 5-year study period (Figure 1). Of these patients, a total of 4651 (15%) were no longer active Kaiser Permanente members at the end of the study period and 845 (3%) were receiving antineoplastic therapy and were excluded. Of the remaining 25 334 evaluable patients, 2334 (9%) had no alanine aminotransferase measurements. The average duration of statin use in this group was 2.8 1.8 years; range 0.08 to 5.0 years. Table 1 describes the baseline characteristics of the remaining 23 000 patients. There were 361 patients (2%) with alanine aminotransferase levels between 3 and 10 times the upper limit of normal (average 192 72 IU/L) during the 5-year study period (Figure 1). Elevated alanine aminotransferase values returned to normal upon subsequent testing in 235 (65%) of these patients but remained elevated (average 174.9 54 IU/L) in 91 (25%). Thirty-five of these patients (10%) did not have repeat alanine aminotransferase testing. Sixty-two patients (0.3%) with alanine aminotransferase greater than 10 times the upper limit of normal were identified and their medical records were reviewed. Elevated alanine aminotransferase levels in 46 of the 62 (74%) patients were determined to be unrelated to statin use. Eleven of the 46 patients were not on a statin at the time alanine aminotransferase was elevated. One or more of the following causes identified as responsible for the alanine aminotransferase elevation was present in a further 29 patients: cholelithiasis (n 15), viral or autoimmune hepatitis (n 8), fatty liver (n 1), pancreatitis (n 1), severe congestive heart failure (n 1), triamterene/ hydrochlorothiazide use (n 1), sertraline use (n 1), and cancer (n 1). Medical records were unavailable in two patients. In 4 cases the precipitating factor was unclear due to complicated history such as diverticulitis, suspected but unconfirmed cholelithiasis, gaps in patient s medical record or the lack of re-challenge data after statin discontinuation. The remaining 16 (26% of cases) patients had severe transaminitis considered to be directly related to statin use. Table 2 provides a summary for these 16 cases and an additional case of a patient who expired shortly after starting a statin. Baseline alanine aminotransferases prior to initiation of statin therapy were normal ( 42 IU/L) for 14 patients; two patients had values slightly higher than normal (46 and 54 IU/L); and for one patient the baseline values were not available due to new membership status. Using the Naranjo scale, we determined that 16 of the 17 cases were probable or highly probable for statin-related toxicity. In only one case was the alanine aminotransferase determined not likely to be related to the statin (Table 2). Review of patient cases Drug-drug interactions Of the 16 cases considered related to statin therapy, 13 (81%) were associated with drug interactions (Table 2). Seven patients had potential drug-drug interactions caused by agents that share common metabolic pathways with statins (ie, cyclosporine, erythromycin, cimetidine, and indinavir). Two patients were taking drugs for which liver enzyme elevation has been reported as an adverse effect in the literature (ie, propoxyphene with acetaminophen, naproxen, and codeine with acetaminophen) in addition to the statin therapy. Eight patients were on a combination of statin and another agent with either the potential to elevate liver enzymes (ie, propafenone) or

Table 2 # Sex/age (yr) Summary of 17 cases with alanine aminotransferase greater than 10 times the upper limit of normal Statin and daily dose (mg) Concomitant medications Comorbid conditions ALT (IU/L) AST (IU/L) Bilirubin (mg/dl) Resolved (weeks) Signs or symptoms Recurrence after re-challenge 1 M/71 Simvastatin 40 Cyclosporine Renal transplant 449 N/A N/A 2 Yes No, on pravastatin 8 2 F/36 Lovastatin 80 Cyclosporine Renal transplant 829 912 0.5 6 Yes N/A 7 3 F/31 Lovastatin 20 Cyclosporine Renal transplant 753 131 0.8 8 No N/A 7 4 M/64 Lovastatin 80 Cyclosporine Renal transplant 439 N/A N/A 5 Yes N/A 7 5 M/66 Simvastatin 20 Erythromycin Heart disease, heart 444 109 1.6 4 Yes No, on simvastatin 7 failure* 6 F/67 Atorvastatin 80 Propoxyphene/APAP Heart disease 1481 1961 1.3 5 No No, on simvastatin 9 7 F/67 Simvastatin 60 Gemfibrozil, Heart disease, 679 881 0.2 8 Yes No, on simvastatin 7 diltiazem diabetes, chronic renal insufficiency 8 M/73 Lovastatin 20 None Heart disease, 879 624 0.5 2 Yes No, on simvastatin 7 diabetes Yes, on lovastatin 9 F/74 Simvastatin 80 None Diabetes 787 N/A N/A 8 No No, on lovastatin 8 10 F/71 Atorvastatin 80 None None 478 233 1.1 2 Yes Yes, on atorvastatin 9 No, on lovastatin 11 M/70 Simvastatin 20 Propafenone, Heart disease, heart 430 N/A N/A 3 No N/A 7 digitalis failure 12 M/86 Simvastatin 40 Fluoxetine Heart disease, 662 N/A N/A Unknown No N/A 7 diabetes, chronic renal insufficiency 13 M/44 Simvastatin 40 Naproxen, APAP, None 685 N/A N/A 8 No N/A 7 codeine 14 F/74 Simvastatin 40 Verapamil, Heart disease, heart 492 N/A N/A 4 Yes Yes, on lovastatin 10 gemfibrozil failure 15 F/66 Simvastatin 40 Indinavir Heart disease, 750 670 0.8 7 Yes No, on pravastatin 7 diabetes, human immunodeficiency virus 16 M/55 Lovastatin 20 Cimetidine None 522 117 1.3 4? No, on lovastatin 7 No, on simvastatin 17 F/80 Simvastatin 40 None Heart failure 4300 7200 2.9 Death Yes N/A 3 *Heart disease denotes coronary artery disease; heart failure denotes congestive heart failure. Bilirubin (n 0.1 0.9 mg/dl), ALT alanine aminotransferase (N 42 IU/L), AST aspartate aminotransferase (N 36 IU/L). Time for alanine aminotransferase to return to less than the upper limit of normal. Yes denotes recurrence (alanine aminotransferase more than the upper limit of normal), No denotes absence of recurrence after re-challenge, and N/A denotes re-challenge was not attempted. A statin used for re-challenge is specified in this column. In some cases the patient was re-challenged with more than 1 statin. Naranjo score of greater than 9 indicates a high probability of drug-induced adverse events, 5-8 is probable, 1-4 is possible, and less than 1 is doubtful. Naranjo score Charles et al Statin-related Transaminitis 621

622 The American Journal of Medicine, Vol 118, No 6, June 2005 Table 3 Time to onset of alanine aminotransferase greater than 10 times the upper limit of normal from statin initiation/dose change or addition of concomitant medication Case Sex/age (yr) Statin and daily dose (mg) Time to onset after statin initiation or dose increase 1 M/71 Simvastatin 40 1 month (dose increased from 20 to 40 mg daily) Time to onset after addition of concomitant medication 2 weeks 2 F/36 Lovastatin 80 2 months Prior to statin 3 F/31 Lovastatin 20 4 years 3 days 4 M/64 Lovastatin 80 4 years 3 weeks 5 M/66 Simvastatin 20 1 year 4 weeks 6 F/67 Atorvastatin 80 1 month N/A 7 F/67 Simvastatin 60 1 month Prior to statin 8 M/73 Lovastatin 20 5 years N/A 9 F74 Simvastatin 80 1.5 months (dose increased from 40 to N/A 80 mg daily) 10 F/71 Atorvastatin 80 1 week N/A 11 M/70 Simvastatin 20 2 months 1.5 months 12 M/86 Simvastatin 40 2 years 1 month 13 M/44 Simvastatin 40 4 years Prior to statin 14 F/74 Simvastatin 40 5 months Prior to statin 15 F/66 Simvastatin 40 3 weeks Prior to statin 16 M/55 Lovastatin 20 10 months 2 days 17 F/80 Simvastatin 40 3 weeks N/A inhibit the CYP 3A4 metabolic pathway (ie, verapamil, diltiazem, fluoxetine, digoxin, or gemfibrozil). 12 Re-challenge data Of the 16 patients with alanine aminotransferase greater than 10 times the upper limit of normal attributable to statin use, 6 were not re-challenged. Ten patients were re-challenged with the same (n 7) or different statin (n 3) at least once. Out of 7 re-challenged with the same statin, 4 patients had no recurrent alanine aminotransferase elevation. Three had a positive re-challenge. These 3 patients and 3 others were tried on a different statin. Five out of the 6 patients tolerated the alternative statin without subsequent alanine aminotransferase elevation. One patient had severe alanine aminotransferase elevation when re-challenged with the initial statin as well as an alternative statin. This was a 74-year-old woman on multiple medications, including verapamil, digoxin, and warfarin. She was hospitalized twice with symptoms of myositis and alanine aminotransferase and creatine kinase elevations. At the time of the first hospitalization, she was on simvastatin 80 mg daily and gemfibrozil 600 mg twice a day; however, during the second hospitalization she was receiving lovastatin 80 mg daily as monotherapy. Both the alanine aminotransferase and creatine kinase elevations resolved when statins were discontinued. One death was identified as potentially, though not likely, statin related. The patient, an 80-year-old woman with no documented history of liver disease or prior statin usage, was started on simvastatin 40 mg daily as well as lisinopril 10 mg daily. Three weeks later, she was admitted to hospital with multiorgan failure, including myocardial infarction, liver failure, and severe coagulopathy. She expired after several days in the hospital. Time to onset, symptoms, and resolution Tables 2 and 3 describe time to alanine aminotransferase elevation, presence or absence of symptoms of liver damage, and the time for alanine aminotransferase to return to normal limits. Of the 17 patients, 11 (65%) experienced severe transaminitis within 2 days to 4 weeks, 3 (18%) within 8 weeks, and the remaining 3 (18%) between 16 weeks and more than 5 years after the changes in patient s drug regimen were made (Table 3). There did not appear to be any correlation between statin dosage and the time of onset of alanine aminotransferase elevation. Ten out of 17 patients were symptomatic at the time of abnormal laboratory findings with symptoms ranging from mild (yellow skin tint, itching, urine and stool discoloration) to severe (rhabdomyolysis, liver failure, coagulopathy). Six patients presented with no symptoms of liver damage. Data were not available for 1 patient. With the exception of the 1 fatality, alanine aminotransferase levels returned to normal for all patients within 2 to 8 weeks. Discussion Clinical trials with statins have demonstrated that approximately 1% to 3% of patients develop alanine aminotransferase greater than 3 times the upper limit of normal. 2-6 In agreement with that, our study found that of over 23 000 patients with alanine aminotransferase levels drawn while

Charles et al Statin-related Transaminitis 623 on statin therapy, only 1.6% developed alanine aminotransferase greater than 3 times the upper limit of normal, and fewer (0.3%) had levels greater than 10 times the upper limit of normal. Upon review of the 0.3% of patients, very few cases (less than 0.1% of the 23 000 with alanine aminotransferase levels drawn) were deemed to be actually related to statin therapy. Our data are consistent with other studies in the literature. 1,13-16 A recent review of animal toxicology as well as pre- and post-marketing clinical trials of lovastatin found no direct correlation between the increased use of lovastatin over the past 15 years and the incidence of drug-related serious liver disease. 1 In fact, serious liver disease due to lovastatin was less common than that associated with NSAIDs. 1 Smith et al reviewed the medical charts of 194 primary care patients receiving statins within a 1-year time period and reported that only 1% developed alanine aminotransferase greater than 3 times the upper limit of normal. 15 Moreover, none of these cases were deemed to be statinrelated. A recent study by Chalasani et al found that individuals with elevated baseline liver enzymes were not at increased risk for hepatotoxicity from statin therapy. 16 The elevation of liver enzymes observed in the clinical trials may actually be related to cholesterol-lowering per se rather than a direct hepatotoxic effect of statins. 1 Cholesterol lowering may increase hepatocyte membrane permeability, resulting in an increase in liver enzyme leakage. 1 Animal toxicology data and the fact that liver enzyme elevation has been reported with other classes of cholesterollowering medications support this theory. 1,17 Considering the remarkable clinical benefits of statins, the widespread use of these agents, and the potentially nontoxic nature of liver transaminitis, our results question the clinical utility of routine liver function monitoring in all statin users. Previous studies have also questioned the value of routine monitoring of liver function tests and whether monitoring prevents serious liver damage. 1,15 Furthermore, given the fact that statins are being evaluated for over-thecounter status, the practice of routine liver enzyme monitoring will be less likely to occur in clinical practice. A good screening tool should be capable of detecting the target condition prior to the onset of symptoms. Current treatment guidelines recommend liver enzyme monitoring every 6 to 8 weeks after the initiation of statin therapy, every 3 to 6 months for the first year, and periodically thereafter. 8 In our study, the majority of patients (65%) with alanine aminotransferase greater than 10 times the upper limit of normal experienced severe liver enzyme elevation within the first month after statin initiation, dose adjustment, and/or addition of medications with potential to interact with statins. Ten out of 17 patients (58%) were symptomatic at the time of the test. Thus, in the majority of our patients, testing detected enzyme elevation only after the patient became symptomatic. Our results suggest that routine liver enzyme monitoring is likely most appropriate in patients at risk for severe transaminitis, such as those with multiple comorbid conditions and/or who are receiving concomitant medications. 12 We identified 1 death that was potentially, though not likely, simvastatin related. The patient s medical chart provided little detail other than the patient was an 80-year-old statinnaïve woman with history of congestive heart failure, who presented with liver failure and myocardial infarction 3 weeks after initiation of simvastatin 40 mg daily. Considering that this was a single case out of 30 000 statin users and the retrospective nature of this analysis, we cannot rule out other cases of liver failure and death, such as organ failure secondary to a myocardial infarction or a rare idiosyncratic reaction. Elevations of liver enzymes, alanine aminotransferase in particular, are not specific to statin therapy. Enzyme elevation can be associated with conditions such as gallbladder disease, infectious liver disease, passive hepatic congestion secondary to congestive heart failure, nonalcoholic fatty liver associated with diabetes, obesity, and dyslipidemia. In our study, 15 out of 17 cases of statin-related alanine aminotransferase greater than 10 times the upper limit of normal presented with medical comorbid conditions that could have been the contributing factor to alanine aminotransferase elevation. We chose to evaluate only patients with alanine aminotransferase greater than 10 times the upper limit of normal rather than patients with alanine aminotransferase greater than 3 times the upper limit of normal. Our decision was based on the fact that alanine aminotransferase greater than 10 times the upper limit of normal is the range that has been defined as serious. 7 In fact, it is possible that discontinuation of lipid-lowering therapy simply based upon mildly elevated alanine aminotransferase levels (ie, 1 to 3 times the upper limit of normal) could be more detrimental by increasing the risk of death from heart disease and strokes than simply continuing therapy. Future studies are needed to evaluate the impact of discontinuation of statins with alanine aminotransferase greater than 3 times the upper limit of normal and the rate of recurrent coronary events with this practice. Our study has some limitations. We could not assess the impact of other factors such as alcohol use or the use of over-the-counter and herbal products from this retrospective chart review. About 9% of patients started on statin never had alanine aminotransferase measured, suggesting that the actual number of patients with severe transaminitis may be even greater. However, it is likely that the number of patients missed is very small because most patients with this degree of transaminase elevation in our study were symptomatic. There were insufficient data to compare alanine aminotransferase elevations with different statins because the majority of our patients were on lovastatin or simvastatin. A study specifically designed to compare statins is needed to determine whether some agents have a lower risk of transaminitis than others.

624 The American Journal of Medicine, Vol 118, No 6, June 2005 The present study suggests that very few patients develop severe transaminitis as a direct result of statin therapy. The majority of patients who developed severe transaminitis either had comorbid conditions or were receiving concomitant medications with the potential to interact with the statin. These data would suggest that routine liver enzyme monitoring in patients without significant comorbid conditions or concomitant therapies may not be necessary. More aggressive monitoring of liver function tests and possibly earlier than recommended by current guidelines is likely warranted in patients on multiple medications with comorbid conditions or if a potentially interacting drug is added to a patient s drug regimen. In patients who develop elevations in liver function tests, the tests typically return to normal after the statin is discontinued. Furthermore, most patients who fail one statin due to liver enzyme elevation can be successfully re-challenged with the same or different statin. References 1. Tolman KG. The liver and lovastatin. Am J Cardiol. 2002;89:1374 1380. 2. Scandinavian Simvastatin Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383 1389. 3. Sacks FM, Pfeffer MA, Lemuel AM, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med. 1996;335:1001 1009. 4. Downs JR, Clearfield M, Weis, S, et al. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels. Results of AFCAPS/TexCAPS. JAMA. 1998;279: 1615 1622. 5. The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med. 1998;339:1349 1357. 6. Heart Protection Study Collaborative Group. MRC/BHF heart protection study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomized placebo-controlled trial. Lancet. 2002;360: 7 22. 7. PhRMA/FDA/AASLD Drug-induced hepatotoxicity white paper postmarketing considerations. Available from: http://www.fda.gov/cder/ livertox/. [cited 2004 Apr 29]. 8. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA. 2001;285:2486 2497. 9. Dujovne CA. Side effects of statins: hepatitis versus transaminitis myositis versus CPKitis. Am J Cardiol. 2002;89:1411 1413. 10. Benichou C. Criteria of drug-induced liver disorders. Report of an international consensus meeting. J Hepatol. 1990;11:272 276. 11. Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther. 1981;30: 239 245. 12. Williams D, Feely J. Pharmacokinetic-pharmacodynamic drug interactions with HMG-CoA reductase inhibitors. Clin Pharmacokinet. 2002;41:343 370. 13. Newman CB, Palmer G, Silbershatz H, Szarek M. Safety of atorvastatin derived from analysis of 44 completed trials in 9,416 patients. Am J Cardiol. 2003;92:670 676. 14. Ballantyne CM, Corsini A, Davidson MH, et al. Risk of myopathy with statin therapy in high-risk patients. Arch Intern Med. 2003;163: 553 564. 15. Smith CC, Bernstein LI, Davis RB, et al. Screening for statin-related toxicity. Arch Intern Med. 2003;163:688 692. 16. Chalasani N, Aljadhey H, Kesterson J, Murray MD. Patients with elevated liver enzymes are not at higher risk for statin hepatotoxicity. Gastroenterology. 2004;126:1287 1292. 17. MacDonald JS, Gerson RJ, Kornbrust DJ, et al. Pre-clinical evaluation of lovastatin. Am J Cardiol. 1988;62:16J 27J.