Impact of Statin Dosing Intensity on Transaminase and Creatine Kinase

The American Journal of Medicine (2007) 120, 706-712

CLINICAL RESEARCH STUDY

Impact of Statin Dosing Intensity on Transaminase and Creatine Kinase Krista M. Dale, PharmD,a,c,d C. Michael White, PharmD,a,c,d Nickole N. Henyan, PharmD,a,c,d Jeffrey Kluger, MD,b,c Craig I. Coleman, PharmDa,c,d a d

The University of Connecticut Schools of Pharmacy and bMedicine, cStorrs and Farmington; and the Divisions of Cardiology and Drug Information, Hartford Hospital, Hartford, Conn. ABSTRACT PURPOSE: Higher intensity statin therapy reduces cardiovascular events more than lower intensity therapy, but the safety impact of higher intensity therapy is unknown. We performed a meta-analysis of randomized controlled trials comparing higher versus lower intensity therapy on liver and muscle safety. METHODS: A systematic literature search through January 2006 was conducted to identify randomized trials comparing higher versus lower intensity statin therapy meeting our criteria. Weighted averages were reported as relative risks (RRs) with 95% confidence intervals (random-effects model). Statistical heterogeneity scores were assessed with the Q statistic and L’Abbe plots. Publication bias was assessed with the Egger weighted regression and funnel plots. RESULTS: Higher intensity statin therapy increased the incidence of transaminase elevations (RR 3.10 [95% Confidence Interval [CI], 0.88-7.85]) versus lower intensity statin therapy. When studies of hydrophilic and lipophilic statins were evaluated separately, higher intensity hydrophilic statin therapy increased the risk for transaminase elevations (RR 3.54 [95% CI, 1.83-6.85]), but higher intensity lipophilic therapy did not (RR 1.58 [95% CI, 0.81-3.08]). The risk of creatine kinase (CK) elevations showed a trend toward an increase (RR 2.63 [95% CI, 0.88-7.85]) with higher intensity therapy. No occurrences of CK elevations occurred in studies evaluating hydrophilic statins, whereas lipophilic statins showed an increased risk with higher intensity therapy (RR 6.09 [95% CI, 1.36-27.35]). CONCLUSIONS: More aggressive statin therapy increases the incidence of transaminase elevations in clinical trials versus lower intensity therapy. Increases in transaminases may be more problematic when hydrophilic statins are used aggressively, whereas CK elevations are more problematic with higher intensity lipophilic statin therapy. © 2007 Elsevier Inc. All rights reserved. KEYWORDS: Creatine kinase; HMG CoA reductase inhibitor; Safety; Statin; Transaminase

Clinical trials have firmly established the cardiovascular benefits of higher intensity lipid-lowering therapy with statins. The Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis in Myocardial Infarction 22 study examined the effects of aggressive versus moderate statin therapy in patients with acute coronary syndrome, whereas the Treating to New Targets study sought to determine the Requests for reprints should be addressed to Craig I. Coleman, PharmD, Assistant Professor of Pharmacy, University of Connecticut, and Director, Pharmacoeconomics and Outcomes Studies Group, Hartford Hospital, Hartford, CT 06102-5037. E-mail address: [email protected]

0002-9343/$ -see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.amjmed.2006.07.033

benefits of aggressive statin therapy in patients with stable coronary disease. Aggressive statin therapy dramatically decreased death and cardiovascular events in both studies.1,2 However, the impact of using higher intensity therapy on liver or muscle toxicity is not known. Clinicians may be cautious about starting higher intensity therapy in their patients because of safety concerns or may try more aggressive therapy in a patient only to find an elevation in transaminase or creatine kinase (CK) concentrations. Therefore, we performed a meta-analysis of randomized controlled trials evaluating the impact of using higher versus lower intensity therapy on liver and muscle safety end points.

Dale et al

Statin Intensity and Safety

707

METHODS

enroll a minimum of 100 patients; and report data on the incidence of elevations in aspartate aminotransferase, alaA systematic literature search (search terms: “HMG-CoA nine aminotransferase, or CK. reductase inhibitor,” “HMG-CoA RI,” “statin,” “pravastaData were independently abstracted by 3 investigators tin,” “simvastatin,” “lovastatin,” “atorvastatin,” cerivasta(KMD, CMW, CIC) by using a standardized data abstractin,” “rosuvastatin,” and “fluvastatin”) was conducted to tion tool with disagreements reidentify randomized clinical trials solved by consensus. All end of statin use with a primary or points were treated as dichotosecondary end point analyzing heCLINICAL SIGNIFICANCE mous variables. Weighted averpatic or muscle toxicity: MEDages were reported as relative LINE from 1966 to January 2006; ● The number 1 and 2 prescribed medicarisks (RRs) with 95% confidence EMBASE from 1990 to January tions in the United States are statins. intervals calculated with StatsDi2006; CINAHL from 1982 to Jan● Physicians are concerned about safety rect statistical software version uary 2006; the Web of Science because National Cholesterol Education 2.4.5 (available at http://www. from 1994 to January 2006; the statsdirect.com) with a random-efProgram guidelines and clinical trials Food and Drug Administration fects model (DerSimonian and support lower low-density lipoprotein website (www.fda.gov); and the Laird methodology).3 CalculaCochrane Database. Searches goals and aggressive statin therapy. tions of RR are problematic for were limited to human studies ● On the basis of these data, lipophilic individual studies with an absence published in English. A manual statins may be associated with a greater of CK or transaminase elevations review of references was perrisk of muscle toxicity, and more hydroin their control groups. For these formed to identify additional relestudies, a nominal value (0.5) was philic statins may be associated with a vant studies. Potentially relevant added in all 2 ⫻ 2 cells to enable greater risk of liver function test articles were reviewed indecalculation of RR.4 Statistical hetpendently by 3 investigators abnormalities. erogeneity scores were assessed (KMD, CMW, CIC). To be inwith the Q statistic (P ⬍.1 reprecluded, studies had to be randomsenting significant statistical hetized trials of statins; compare erogeneity) and by visual inspection of L’Abbe plots. The higher versus lower intensity statin therapy; have a mean (or median) duration of patient follow-up of 48 weeks or more; Egger weighted regression method was used to statistically

Figure 1

Trial identification, inclusion and exclusion.

708 Table 1

Study

The American Journal of Medicine, Vol 120, No 8, August 2007 Study Characteristics

Subject Higher Intensity (n) Statin Therapy

Prove-It 4162 TNT 10,001 IDEAL 8888 A to Z 4497 ARBITER 161 REVERSAL 654 BELLES 614 Post-CABG 1351 ASAP 325

Atorvastatin 80 mg Atorvastatin 80 mg Atorvastatin 80 mg Simvastatin 40-80 mg Atorvastatin 80 mg Atorvastatin 80 mg Atorvastatin 80 mg Lovastatin 76 mg Atorvastatin 80 mg

Lower Intensity Statin Therapy

Jadad Duration Average Score (y) Age (y) Male (%)

Pravastatin 40 mg Atorvastatin 10 mg Simvastatin 20 mg Simvastatin 20 mg Pravastatin 40 mg Pravastatin 40 mg Pravastatin 40 mg Lovastatin 4 mg Simvastatin 40 mg

3 3 3 5 2 5 4 2 5

2 4.9 4.8 2 1 1.5 1 4.3 2

58 61 62 61 59 56 64 61 48

3251 8099 7187 3396 115 362 0 369 130

Previous Myocardial Infarction (%)

Diabetes (%)

(78) 769 (18) 734 (18) (81) 5833 (58) 1501 (15) (81) 8888 (100) 1069 (12) (76) 764 (17) 1059 (24) (71) N/A* 16 (10) (55) N/A* 95 (15) (0) 17 (3) 86 (14) (27) 197 (15) N/A† (40) N/A* 6 (2)

Prove-It ⫽ Intensive versus Moderate Lipid Lowering with Statin after Acute Coronary Syndromes; TNT ⫽ Intensive Lipid Lowering with Atorvastatin in Patients with Stable Coronary Disease; IDEAL ⫽ High-Dose Atorvastatin vs Usual-Dose Simvastatin for Secondary Prevention After Myocardial Infarction, The IDEAL Study: A Randomized Controlled Trial; A to Z ⫽ Early Intensive vs a Delayed Conservative Simvastatin Strategy in Patients With Acute Coronary Syndromes: Phase Z of the A to Z Trial; ARBITER ⫽ Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol: A Randomized Trial Comparing the Effects of Atorvastatin and Pravastatin on Carotid Intima Medial Thickness; REVERSAL ⫽ Effect of Intensive Compared With Moderate Lipid-Lowering Therapy on Progression of Coronary Atherosclerosis: A Randomized Controlled Trial; BELLES ⫽ Aggressive Versus Moderate Lipid-Lowering Therapy in Hypercholesterolemic Postmenopausal Women: Beyond Endorsed Lipid Lowering With EBT Scanning; Post-CABG ⫽ The Post Coronary Artery Bypass Graft Trial Investigators. The Effect of Aggressive Lowering of Low-Density Lipoprotein Cholesterol Levels and Low-Dose Anticoagulation on Obstructive Changes in Saphenous-Vein Coronary Artery Bypass Grafts; ASAP ⫽ Effect of aggressive versus conventional lipid lowering on atherosclerosis progression in familial hypercholesterolemia. *History of myocardial infarction was not reported in the ASAP, REVERSAL, or ARBITER trials. †History of diabetes was not reported in the Post-CABG trial.

assess publication bias (P ⬍.05 was considered representative of significant statistical publication bias). Publication bias also was assessed by visual inspection of funnel plots. Studies of poorer methodologic quality, such as unblinded or open-labeled trials, may exhibit exaggerated treatment effects. Jadad scores were used to evaluate the methodologic quality of the included trials.5 Including only higher quality studies may result in increased interval validity but could reduce external validity of the analysis. In addition, the selection of a random-effect versus a fixedeffect model in meta-analyses is controversial. The use of a random-effect model in the calculation of confidence intervals results in wider intervals and thus a more conservative estimate of treatment effect when compared with a fixedeffect model. To reconcile these issues, sensitivity analysis was conducted whereby the meta-analysis was reanalyzed excluding studies with a Jadad score less than 3 and using a fixed-effects model (Mantel-Haenszel methodology).6

RESULTS Our initial search yielded 8849 potential literature citations. The search was narrowed to 1752 studies by limiting the search to “humans,” “English language,” and “clinical trial.” Of these, 1592 were excluded through review of abstracts, leaving 160 studies for full publication review. An additional 151 of these were excluded for various reasons as depicted in Figure 1. Nine studies (n ⫽ 21,765 subjects) were found to conform to our inclusion criteria (Table 1).1,2,7-11Elevations of transaminases were defined as elevations in aspartate aminotransferase and/or alanine amino-

transferase more than 3 times the upper limit of normal in 7 studies1-2,7-9,13 and more than 2 times the upper limit of normal in 1 study.11 CK elevations were defined as more than 3 times the upper limit of normal in 2 studies9,10 and more than 10 times the upper limit of normal in 6 studies.2,7,9,10,13 Jadad scores ranged from 2 to 5. Patient enrollment ranged between 161 and 10,001 patients. The mean patient age ranged between 48.5 and 61.7 years, and participants were mostly male (78%).1,2,7-13 Three studies used the same agent (atorvastatin, simvastatin, or lovastatin)2,7,9 in both groups but with differing doses, and 6 trials compared high-dose atorvastatin with moderate doses of simvastatin or pravastatin.1,8-10,12,13 Hydrophilic agents (pravastatin and atorvastatin) were evaluated in 5 studies,1,2,7-9 and lipophilic agents (simvastatin and lovastatin) were evaluated in 2 studies.7-11 The effect of aggressive versus conventional lipid lowering on atherosclerosis progression in familial hypercholesterolemia and High-Dose Atorvastatin versus Usual-Dose Simvastatin for Secondary Prevention After Myocardial Infarction, The IDEAL Study trials compared a hydrophilic statin (atorvastatin) with a lipophilic statin (simvastatin); thus, they were not used in the subgroup analysis for relative lipophilicity or hydrophilicity.12,13 Duration of patient follow-up ranged from 1 year to 5 years in the included studies.1,2,7-13 The incidence of transaminase elevations was greater among those receiving higher versus lower intensity statin therapy (RR 3.10 [95% CI, 1.72-5.58]). This corresponds to a 1.5% incidence in the higher intensity statin group versus 0.4% in the lower intensity statin group. There was a trend

Dale et al

Statin Intensity and Safety

Figure 2

709

L’Abbe plots for creatine kinase (top) and AST/ALT (bottom).

toward increased incidence of elevations in CK with higher intensity statin therapy (RR 2.63 [95% CI, 0.88-7.85]). This corresponds with a 0.1% incidence in the higher intensity statin group versus 0.02% in the lower intensity statin group. Statistical heterogeneity was found for transaminase analysis (P ⫽ .002) but not for CK analysis (P ⫽ .89). However, there was no evidence of statistical heterogeneity on the L’Abbe plots for either the transaminase or CK analysis (Figure 2). There was no evidence of publication bias for transaminase elevations with the Egger weighted regression method (P ⫽ .82), but it could not be ruled out through visual inspection of the funnel plot (Figure 3). Publication

bias could not be ruled out for CK elevations with the Egger weighted regression method (P ⬍.0001) or by visual inspection of the funnel plot (Figure 3). When studies of hydrophilic and lipophilic statins were evaluated separately, higher intensity hydrophilic statin therapy increased the risk for transaminase elevations versus lower intensity therapy (RR 3.54 [95% CI, 1.83-6.85]), but no cases of CK elevations occurred in either group (Figure 4). Higher intensity lipophilic statin therapy did not significantly impact the incidence of transaminase elevations (RR 1.58 [95% CI, 0.81-3.08]) but increased the incidence of CK elevations (RR 6.09 [95% CI, 1.36-27.35]) versus lower intensity therapy.

710

The American Journal of Medicine, Vol 120, No 8, August 2007

Figure 3

Funnel plots for creatine kinase (top) and AST/ALT (bottom).

Statistical heterogeneity was not observed in any of the subgroup analyses (P ⱖ .11 for all). On sensitivity analysis (Table 2), when a fixed-effect model was used, the incidence of elevations in transaminase was not altered (RR 3.35 [95% CI, 2.56-4.40]). The incidence of CK elevations was significantly higher in patients receiving more intense statin therapy (RR 3.20 [95% CI, 1.17-8.75]). The results were not significantly altered when only trials with a Jadad score of 3 or more were included.

DISCUSSION Our meta-analysis of randomized, long-term, clinical trials directly comparing statins suggests that choosing more

modest statin therapy may reduce the occurrence of transaminase or CK elevations compared with higher intensity therapy. Although the risk of CK elevations only showed a trend toward significance with a random-effects model evaluating higher versus lower intensity therapy, a significantly elevated risk was seen with the fixed-effect model. However, our study also suggests that choosing a higher intensity lipid-lowering strategy with hydrophilic statins is associated with more transaminase elevations but does not increase the occurrence of CK elevations versus more modest hydrophilic therapy. In contrast, higher intensity lipophilic statin therapy greatly increases the risk of CK elevations but does not alter the risk of transaminase eleva-

Dale et al

Statin Intensity and Safety

711

Figure 4

Subgroup analysis.

tions versus lower intensity lipophilic statin therapy. As such, the clinician treating a patient with a small body frame or other risk factor for muscle toxicity might opt for higher intensity therapy with a hydrophilic statin because it does not increase the risk of muscle toxicity compared with modest statin therapy. Clinicians with patients experiencing a transaminase elevation on a high-intensity hydrophilic statin regimen might try high-intensity lipophilic statin therapy after resolution of the elevated transaminases before opting for lower intensity statin therapy. Given the additional clinical benefits derived from higher intensity therapy, being able to maintain high-intensity statin therapy by switching drugs within the class to circumvent potential safety concerns may be important. Our results may make pharmacologic sense because the liver expresses OATP-C pumps to actively pump in statins, whereas muscle cells do not.14 As such, lipophilicity is not important in determining intrahepatocyte concentrations but is crucial for peripheral cell concentrations.15,16 A recent meta-analysis examined the safety of highversus low-dose atorvastatin treatment.17 The authors

Table 2

concluded that CK elevations were rare in patients receiving atorvastatin and did not seem to be dose-related.17 However, patients receiving high-dose atorvastatin therapy were more likely to experience transaminase elevations than patients receiving low-dose atorvastatin.17 These findings support the hypothesis that statins’ relative hydrophilicity or lipophilicity is related to the type and frequency of adverse reactions. Although we have divided statins into hydrophilic and lipophilic categories, there are actually 3 general groupings based on lipophilicity.18 The highly lipophilic agents include cerivastatin, simvastatin, and lovastatin.18 The modestly lipophilic statins include atorvastatin and fluvastatin.18 The low lipophilic statins include rosuvastatin and pravastatin.18 It would have increased the complexity of our analysis to use 3 groups, so we opted to break the groups into 2 categories. In addition, all of the studies evaluating statins with modest lipophilicity compared high-dose atorvastatin with moderate doses of pravastatin. There are no studies to date that compared 2 hydrophilic agents. As such, we decided to add the modestly lipophilic atorva-

Sensitivity Analysis

Safety Outcome

No. of Studies

Higher Intensity Statin Therapy

Lower Intensity Statin Therapy

Risk Ratio (95% CI)

Heterogeneity (P)

CK Random effects Fixed effects Jadad score ⱖ 3 AST/ALT Random effects Fixed effects Jadad score ⱖ 3 CI ⫽ confidence interval.

8 8 7

13/13,246 13/13,246 9/12,570

2/13,245 2/13,245 1/12,570

2.63 (0.88-7.85) 3.20 (1.17-8.74) 2.29 (0.65-8.09)

.89 .89 .83

8 8 7

225/15,185 225/15,185 205/14,509

66/15,143 66/15,143 49/14,468

3.10 (1.72-5.58) 3.86 (2.70-5.52) 3.80 (2.23-6.45)

.11 .11 .06

712 statin to the hydrophilic group on the basis of evidence that atorvastatin does not cross the blood-brain barrier as the more highly lipophilic agents do.19,20 According to postmarketing surveillance data, the incidence of rhabdomyolysis is similar among atorvastatin, fluvastatin, and pravastatin, but markedly lower among simvastatin, lovastatin, and cerivastatin.19,20 Although this analysis does provide useful information for physicians prescribing statins, there are some limitations. The clinical trials report the number of patients experiencing elevations in CK or transaminase, but additional information regarding the severity of the adverse effect or clinical outcomes is not routinely provided. Most of the studies used a cutoff point for transaminase elevations of more than 3 times the upper limit of normal. Patients with transaminase elevations of 3 times the upper limit of normal may experience a complete resolution of laboratory abnormalities and experience no long-term clinical effects on discontinuation of statin therapy, whereas patients with elevations significantly higher than 3 times the upper limit of normal may have substantial or irreversible hepatic damage. Many of the included trials had run-in periods during which patients experiencing significant adverse effects or laboratory abnormalities were excluded from the trial. As a result, the overall incidence of transaminase or CK elevations in the general population cannot be estimated by these data.

CONCLUSIONS More aggressive statin therapy increases the incidence of transaminase and CK elevations. Increases of transaminase may be more problematic when hydrophilic statins are used aggressively, whereas CK elevations are more problematic with higher intensity lipophilic statin therapy.

References 1. Cannon CP, Braunwald E, McCave Ch, et al. Intensive versus moderate lipid lowering with statin after acute coronary syndromes. N Engl J Med. 2004;350:1495-1504. 2. LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Eng J Med. 2005;352:1425-1435. 3. DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled Clin Trials. 1986;7:177-188. 4. Naylor AF. Small sample considerations in combining 2⫻2 tables. Biometrics. 1967;23:349-356.

The American Journal of Medicine, Vol 120, No 8, August 2007 5. Jadad AR, DPhil R, Moore A, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Controlled Clin Trials. 1996;17:1-12. 6. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst. 1957;22:719-748. 7. deLemos JA, Blazing MA, Wiviott SD, et al. Early intensive vs a delayed conservative simvastatin strategy in patients with acute coronary syndromes: phase Z of the A to Z Trial. JAMA. 2004;292: 1307-1316. 8. Taylor AJ, Kent SM, Flaherty PJ, et al. ARBITER: Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol: a randomized trial comparing the effects of atorvastatin and pravastatin on carotid intima medial thickness. Circulation. 2002;206:2055-2060. 9. Nissen SE, Tuzcu EM, Schoenhagen P, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomized controlled trial. JAMA. 2004; 291:1071-1080. 10. Raggi P, Davidson M, Callister TQ, et al. Aggressive versus moderate lipid-lowering therapy in hypercholesterolemic postmenopausal women: beyond endorsed lipid lowering with EBT scanning (BELLES). Circulation. 2005;112:563-571. 11. Post-CABG: The Post Coronary Artery Bypass Graft Trial Investigators. the effect of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous-vein coronary artery bypass grafts. N Engl J Med. 1997; 336:153-162. 12. Smilde TJ, van Wissen S, Wollersheim H, et al. Effect of aggressive versus conventional lipid lowering on atherosclerosis progression in familial hypercholesterolaemia (ASAP): a prospective, randomized, double-blinded, trial. Lancet. 2001;357:577-581. 13. Pedersen TR, Faergeman O, Kastelein, JJP, et al. High-dose atorvastatin vs usual-dose simvastatin for secondary prevention after myocardial infarction, the IDEAL Study: a randomized controlled trial. JAMA. 2005;294:2437-2445. 14. Tamai I, Nezu J, Uchino H, et al. Molecular identification and characterization of novel members of the human organic anion transporter (OATP) family. Biochem Biophys Res Commun. 2000;273:251-260. 15. McTavish D, Sorkin EM. Pravastatin: a review of its pharmacologic properties and therapeutic potential in hypercholesterolemia. Drugs. 1991;42:65-89. 16. Negre-Aminou P, Van Vliet AK, Van Erck M, et al. Inhibition of proliferation of human smooth muscle cells by various HMG CoA reductase inhibitors: comparison with other human cell types. Biochemica et Biophysica Acta. 1997;1345:259-268. 17. Newman C, Szarek M, Luo D, et al. Comparative safety of atorvastatin 80 mg versus 10 mg derived from analysis of 49 completed trials in 14,236 patients. Am J Cardiol. 2006;97:61-67. 18. White CM. A review of pharmacologic and pharmacokinetic aspects of rosuvastatin. J Clin Pharmacol. 2002;42:963-970. 19. White CM, Chow MSS. A review of HMG CoA reductase inhibitors. US Pharmacist. 1998;23:H19-HS30. 20. Staffa JA, Chang J, Green L. Cerivastatin and reports of fatal rhabdomyolysis. N Engl J Med. 2002;346:539-540.