Compliance to Recommended Liver Function ... - Wiley Online Library

RESEARCH

Compliance to Recommended Liver Function Monitoring in Patients on Statin Therapy Heather Leaver, Tiong Keng Lim, Philip Thomson, Joanne Leaver, Anna M. J. Choy & Chim C. Lang Division of Medicine and Therapeutics, University of Dundee, Dundee, Scotland, UK

Keywords Adverse drug reactions; Drug safety monitoring; Hepatoxicity; Statins. Correspondence Professor Chim C. Lang, M.D., F.R.C.P., F.A.C.C., Division of Medicine and Therapeutics, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, Scotland, UK Tel:. 01382-632180; Fax: 01382-644972; E-mail: [email protected]

doi: 10.1111/j.1755-5922.2009.00082.x

Product information and national guidelines consistently recommend the estimation of serum alanine aminotranferase (ALT) before and after approximately 3 months as part of risk management following initiation of statin therapy. The aim of this study was to determine compliance to the recommendations for monitoring ALT in patients initiated on statin therapy. The prevalence of abnormal serum ALT levels was also evaluated and compared with current data. We performed a retrospective observational audit of the Medicine’s Monitoring Unit (MEMO) prescribing record-linkage database, which includes evidence of prescribing and biochemistry results of all patients in Tayside, Scotland. We examined patients with first-time statin prescriptions and evaluated all laboratory liver function tests recorded before and after prescriptions from 1st January 2002 to 31st December 2002. Of the 5717 patients identified, 54% were male and the subjects had a mean age of 62 ± 14 standard deviation (SD) years. The prescriptions yielded five types of statin and included the start date, dose, and number of tablets. In total, 1455 (25%) patients had no biochemistry data at all from the year 2002, and despite the 75% of patients who had an ALT measurement, over 90% of patients did not follow the current recommendations. The prevalence of elevated transaminases was consistent with published data in that significant elevation occurred in 1.9% of subjects. Adherence to the recommended laboratory monitoring for patients first prescribed with statin therapy is low in Tayside. Adverse drug reactions need to be more closely monitored by all healthcare professionals involved in prescribing. Monitoring can highlight abnormalities and therefore reduce potentially damaging effects.

Introduction Adverse drug reactions (ADRs) are a major cause of morbidity and mortality. In the United States, ADRs affect nearly 2 million people and cause 100,000 deaths each year [1]. In a prospective analysis of 18,800 hospital admissions in the United Kingdom, Pirmohamed and colleagues [2] found that the prevalence of hospital admissions related to an ADR was about 6.5%, with the ADR directly leading to the admission in 80% of cases. ADRs accounted for 4% of the hospital bed capacity, and the projected annual cost of such admissions to the National Health Service (NHS) was 466 million (€706 million, $847 million). Indeed, the cost related

96

to ADR may be higher than the total cost of cardiovascular or diabetes care [3]. A number of strategies have been proposed to limit the burden of ADR, including regular review of prescriptions, the use of computerized prescribing, and the involvement of pharmacists in assessing prescribing behavior. One strategy is by monitoring drug treatment to anticipate or detect ADR before they become clinically harmful [4]. Indeed, monitoring is recommended for more than half of the entries in the Medicines Compendium [Association of British Pharmaceutical Industry: electronic Medicines Compendium. http://emc.medicines.org.uk]. However, it is not clear whether these recommendations are adhered to in clinical practice. In this study, we have chosen to evaluate

c 2009 Blackwell Publishing Ltd Cardiovascular Therapeutics 27 (2009) 96–100

Compliance to Guidelines in Drug Monitoring

H. Leaver et al.

the compliance to recommendations of drug safety monitoring for statins, the most widely prescribed class of cardiovascular medications. Product information [5] and clinical practice guidelines [6,7] consistently advocate measurement of serum transaminases such as alanine aminotranferase (ALT) before and approximately 3 months after starting statin therapy. However, it should be noted that the guidelines differ in recommendations subsequent to the 3-month measurement, varying from 6 months to annually. Therefore, the aim of this study was to conduct a retrospective observational audit of the Medicine’s Monitoring Unit (MEMO) by comparing prescribing data with the recorded liver biochemistry of new statin users in Tayside, Scotland. This permitted evaluation of compliance to both product information and guideline recommendations and, additionally, the prevalence of abnormal ALT results in the cohort of patients studied.

Methods Subject Selection The data were extracted from the MEMO database, which includes both primary care and hospital records. It is a highly anonymized database recording all of the prescribing data for every patient in Tayside. The information received included patients’ gender, date of birth correct to one month, and full statin prescription (dose, strength, start date, and amount) Laboratory liver function tests data were also extracted that highlighted the test date and results. All laboratory liver function tests requested were included in the MEMO data set and so there was no underestimation of the extent of transaminases measurement. Only patients who were newly prescribed statins in the first 9 months of 2002 were included in this study; therefore, excluding those previously prescribed statins. Examination of data revealed subjects with both single or multiple statin prescriptions. Biochemistry results requested were ALT, bilirubin (BIL), alkaline phosphatase (ALP), and albumin (ALB), which are all considered as standard liver function tests. Reference intervals used in this study were those used in Ninewells Hospital laboratories. The data for this study were obtained after Ethics and Caldicott approval. Since this study involved only the reviews of records from the MEMO database in an anonymized way, no patient consent was required.

Stat1istical Analysis The results from normally distributed continuous data are shown as mean ± 2 standard deviation [SD]. Data were


converted into tables and analyzed using the software Microsoft Excel 2002 (Service Pack 3, Microsoft Corporation, Seattle, WA, USA). This tested the hypothesis that monitoring of liver function is not uniform or consistent with the guidelines proposed. Because of the large quantities of data, calculations were electronically generated using formulas and specialist functions in Excel. Total duration of statin prescriptions was calculated by adding the number of tablets onto the final prescription start date. This permitted exclusion from relevant calculations of those patients who were no longer prescribed statins. In order to evaluate whether monitoring was carried out within time periods, specialist formulas in Excel were used to define ALT results obtained between 3 and 4 months after initiation of treatment. A control sample of 50 random patients was then checked manually to ensure calculations were accurate. Particular attention was paid to ALT levels as elevation is the main abnormality associated with statin hepatotoxicity, and AST levels are not requested routinely with liver function screening. Other concomitant liver function abnormalities were merely noted at this time. A standard error (SE) was then calculated and expressed with an estimated 95% confidence interval. Statistical analysis was performed with Analyseit software for Microsoft Excel (version 1.62; Analyse-it Software Ltd., Leeds, UK).

Results Out of the 5717 patients who received a prescription for a statin in 2002, 3105 (54%) participants were male and 2612 (46%) were female. The majority of patients were aged between 51 and 74 years(mean age 62 ± 27 years, median age 64 years). A total of 5972 prescriptions were collected, including those on simvastatin (69%), atorvastatin (17%), fluvastatin (4%), pravastatin (10%), and cerivastatin (0.0003%). This included acute single prescriptions and duplicates for those patients with repeat prescriptions and/or change of statin. Overall, 44,937 biochemistry results (ALT, BIL, ALB, and ALP) from 2002 were amassed. A total of 4262 patients had records of ALT estimation carried out as seen in Table 1. Therefore, 25.5 ± 1.3% SE of those studied did not receive any liver function testing at any time within 2002. Table 2 shows that, in total, 223 (1.87% of recorded ALT measurements) entries were above three times the upper reference interval (129 U/L) and this correlates as 81 patients (1.9% of subjects with ALT results), with values ranging from 130 U/L to 5220 U/L. However, Table 2 also highlights that 39 patients had elevated ALT estimations prior to statin therapy. Other additional

97


H. Leaver et al.

Table 1 Time intervals of ALT measurements in ﬁrst-time statin users Time at which ALT is measured

Number of patients with ALT measurement

Total subjects on statin therapy at time of measurement

Percentage of total subjects on statins ± standard error (%)

At any point of therapy (including prior to initiation) Prior to therapy initiation Within 3–4 months of initiation Prior to and within 3–4 months of statin initiation

4262

5717

74.5 ± 1.13

3140 539

5717 3637

55.9 ± 1.74 14.8 ± 1.15

335

3637

9.21 ± 3.09

The total subjects decrease because of those excluded as no longer on statins. Percentages expressed with standard error.

Table 2 Evaluation of ALT levels taken before and after statin treatment

Before statin commenced Within 3–4 months of statin initiation All ALT results

Total number of laboratory test and patients

ALT within reference intervals (129 U/L)

6135 (3140) 680 (539) 11,895 (4262)

5300 (2906) 86.4% 590 (491) 86.6% 10,268 (4011) 86.3%

84 (39) 1.4% 20 (10) 2.9% 223 (81) 1.9%

Italic numbers show the number of laboratory results. Number shown in brackets is the number of patients.

findings included 1011 raised ALP entries and 945 raised BIL levels. The current guidelines state that a transaminase level should be obtained before and after initiation of statin therapy [6–8]. Therefore, the time elapsed between the initiation of statin treatment and the date of biochemistry results was studied. As seen in Table 1, 3140 (55%) patients had their first ALT measurement taken before statin treatment, and the remaining 45% had no baseline measure recorded. The number of participants who had their initial ALT measurement within 3–4 months was 539 out of a possible 3637 patients. Therefore, 85.2 ± 1.15% of this cohort did not have ALT measurement within 3–4 months after first initiation of statin therapy. The total number of individuals who adhered to the guidelines of baseline and approximately 3-month measures was 335 patients. This was a mere 9.2%, indicating that the remaining 90.8% of patient care management did not follow the recommendations.

Discussion To the best of our knowledge, this is the first study that assesses the compliance to the current recommended lab-

98

oratory monitoring in patients on statin therapy. It had two main findings. First, that the majority (over 90%) of patients prescribed statins are not getting the recommended liver function testing at the appropriate time. In those who have had liver function assessed, it is mainly carried out for the first time between 3 and 9 months. Second, the prevalence of abnormal liver enzymes detected was low at 1.9%, which is consistent with current literature [9]. Although the cardioprotective effects of statins are well established, less is known about the side effects. Rhabdomyolysis, despite low incidence, is more commonly known as an adverse effect of statins because of its high fatality rate. Analyses conducted by the FDA have reported the rate of fatal rhabdomyolysis as 0.15 per 1 million statin prescriptions [10]. On the other hand, statinrelated myalgia is more common than rhabdomyolysis, but reported incidence ranges from 1.1% to 5.0% [11]. A similar incidence rate of 1–3% has also been reported with statin-induced elevations in hepatic transaminases [12]. In our study, we found that the prevalence of elevated ALT was 1.9%, which is in keeping with these reports. A previous study by Smith and colleagues, which reviewed computerized medical records of about 1200 patients in primary care, also found that 1.5% of the patients on statins had abnormal ALT, although none



H. Leaver et al.

of these abnormalities were directly related to statin use [13]. It should be noted that these elevations in hepatic transaminase rarely progress to liver failure [14] and are usually reversible with a dose reduction or termination of therapy [15]. At this time, despite the low incidence of these side effects, product information [5] and clinical practice guidelines [7,8] still advocate that serum transaminase estimation such as ALT be measured before and approximately 3 months after starting statin therapy. As this is an analysis of prescribing by Scottish physicians, it is worth noting that that the current Scottish Intercollegiate Guidelines Network guidelines state that until there is a change in the FDA-approved prescribing information for statins, it is appropriate to continue to measure transaminase levels before starting therapy, 12 weeks after initiating therapy, after a dose increase, and periodically thereafter [7]. The guidelines do, however, vary in monitoring recommendations for the frequency following the 3-month blood measurement. As rhabdomyolysis is exceedingly rare, the current guidelines only recommend screening for myostis if symptoms are reported in patients. In our study, we found that more than 90% of the patients prescribed statins in Tayside were not monitored in the manner as recommended by the product information or guidelines. In those patients with recorded transaminase levels, most samples were measured between 3 and 9 months of therapy. Forty-five percent patients had no record of liver function testing prior to commencement of statin therapy. An audit based on primary care discussing how abnormal liver function tests were managed revealed some inadequacies. Some liver function abnormalities were not always properly managed, and intervention in these cases in which treatment may have been of benefit was often missed [16]. Though the data were collected 6 years ago, it still remains unknown whether practice has changed in the ensuing 6 years. Nevertheless, the reasons for the poor compliance to recommended routine screening during statin therapy are not entirely clear and cannot be determined from this study. However, one can speculate on a number of possible reasons. First, it may be that at the time when a statin is being considered or initiated for a patient, there maybe limited or no opportunity to obtain the baseline blood measurement. A second possible reason might be the poor understanding of statin-induced liver abnormalities and of the actions that should be taken. It is unclear whether or when treatment should be stopped, as mild abnormalities of liver function have been reported to improve even with continued treatment with statin [17]. Third, it is possible that prescribers may be questioning the need for routine monitoring in statin use, as they may consider the frequency of these abnormalities to be low.


Indeed, a meta-analysis examining 112,000 person-years of exposure to pravastatin found the frequency of abnormal liver function tests (1.4%) to be similar in statin and placebo arms [18]. Furthermore, in the Heart Protection Study, treatment with statins at high dose (40-mg simvastatin) was found to be safe [19]. There may also be uncertainty about the timing of the monitoring, as the infrequent monitoring currently recommended is likely to miss most patients who develop the sudden idiosyncratic hepatic reactions. These findings may lead prescribers to think that regular monitoring may not be necessary. Indeed, this was the conclusion of a retrospective analysis of 1014 patients in primary care in which routine monitoring revealed no cases of significant or moderately abnormal transaminase values attributable to statins [13]. Although the nature of our study does not allow us to define the reasons for the poor compliance to the recommended routine monitoring, it does highlight the fact that the current routine drug safety monitoring is not effective. It can be seen that relying purely on the spontaneous activity of prescribers to monitor is generally not very effective. In the identification of ADRs, spontaneous reporting identifies only 5% of ADRs [20]. Alternative strategies should be identified. In the reporting of ADRs, computerized detection has been reported to be effective in both hospital and out-patient settings [21,22]. Indeed, some general practices are increasingly utilizing electronic systems that summon the patients with a letter to receive routine blood tests related to their medications. Therefore, it is a possibility that statins could be added to these lists for routine monitoring. In conclusion, we have shown that adherence to the recommended laboratory monitoring for patients first prescribed with statin treatment is low in Tayside, Scotland. Monitoring of the liver function throughout statin therapy requires more diligence from prescribers.

Conﬂict of Interest The authors have no conflict of interest.

References 1. Lazarou J, Pomeranz BH, Corey PN. Incidence of adverse drug reactions in hospitalized patients: A meta-analysis of prospective studies. JAMA 1998;279:1200–1205. 2. Pirmohamed M, James S, Meakin S, et al. Adverse drug reactions as cause of admission to hospital: Prospective analysis of 18 820 patients. BMJ 2004;329:15–19. 3. Johnson JA, Bootman JL. Drug-related morbidity and mortality. A cost-of-illness model. Arch Intern Med 1995;155:1949–1956.

99


4. Pirmohamed M, Ferner RE. Monitoring drug treatment. BMJ 2003;327:1179–1181. 5. The British National Formulary 2008. Available at: http://www.bnf.org (accessed November 2008). 6. Scottish intercollegiate guidelines guideline 40 lipids and the primary prevention of coronary heart disease. Available at: http://www.sign.ac.uk (accessed September 1999). 7. Scottish intercollegiate guidelines network 97. Risk estimation and the prevention of cardiovascular disease. Available at: http://www.sign.ac.uk/guidelines/fulltext/97/index.html (accessed February 2007). 8. Pasternak RC, Smith SC Jr., Bairey-Merz CN, Grundy SM, Cleeman JI, Lenfant C. ACC/AHA/NHLBI clinical advisory on the use and safety of statins. Circulation 2002;106:1024–1028. 9. Walley T, Folino-Gallo P, Schwabe U, van Ganse E. Variations and increase in use of statins across europe: Data from administrative databases. BMJ 2004;328:385–386. 10. Graham DJ, Staffa JA, Shatin D, et al. Incidence of hospitalized rhabdomyolysis in patients treated with lipid-lowering drugs. JAMA 2004;292:2585–2590. 11. Thompson PD, Clarkson P, Karas RH. Statin-associated myopathy. JAMA 2003;289:1681–1690. 12. Gotto AM Jr. Safety and statin therapy: Reconsidering the risks and benefits. Arch Intern Med 2003;163:657–659. 13. Smith CC, Bernstein LI, Davis RB, Rind DM, Shmerling RH. Screening for statin-related toxicity: The yield of transaminase and creatine kinase measurements in a primary care setting. Arch Intern Med 2003;163:688–692. 14. Pedersen TR, Tobert JA. Benefits and risks of hmg-coa

100

H. Leaver et al.

15. 16.

17.

18.

19.

20.

21.

22.

reductase inhibitors in the prevention of coronary heart disease: A reappraisal. Drug Saf 1996;14:11–24. Talbert RL. Safety issues with statin therapy. J Am Pharm Assoc (2003) 2006;46:479–488; quiz 488–490. Sherwood P, Lyburn I, Brown S, Ryder S. How are abnormal results for liver function tests dealt with in primary care? Audit of yield and impact. BMJ 2001;322:276–278. Dujovne CA. Side effects of statins: Hepatitis versus “Transaminitis”-myositis versus “Cpkitis”. Am J Cardiol 2002;89:1411–1413. Pfeffer MA, Keech A, Sacks FM, et al. Safety and tolerability of pravastatin in long-term clinical trials: Prospective pravastatin pooling (PPP) project. Circulation 2002;105:2341–2346. Heart Protection Study Collaborative Group. MRC/BHF heart protection study of antioxidant vitamin supplementation in 20,536 high-risk individuals. A randomised placebo-controlled trial. Lancet 2002;360:23–33. Cullen DJ, Bates DW, Small SD, Cooper JB, Nemeskal AR, Leape LL. The incident reporting system does not detect adverse drug events: A problem for quality improvement. Jt Comm J Qual Improv 1995;21:541–548. Jha AK, Kuperman GJ, Teich JM, et al. Identifying adverse drug events: Development of a computer-based monitor and comparison with chart review and stimulated voluntary report. J Am Med Inform Assoc 1998;5:305–314. Honigman B, Light P, Pulling RM, Bates DW. A computerized method for identifying incidents associated with adverse drug events in outpatients. Int J Med Inform 2001;61:21–32.
