A retrospective analysis on the use of gemfibrozil in general outpatient clinics

Kenny Kung 龔敬樂, Augustine Lam 林璨, Philip K T Li 李錦滔

HK Pract 2005;27:404-410

Summary

Objective: To assess the current use of lipid lowering medication in general outpatient clinics (GOPCs) and explore the need for introduction of alternative modalities of treatment. Design: Retrospective analysis.

Design: Retrospective analysis.

Subjects: All patients prescribed with gemfibrozil in New Territories Cluster's GOPCs.

Main outcome measures: Pre-treatment cardiovascular disease risk levels according to Joint British Societies cardiovascular disease risk charts, pre-treatment triglyceride (TG) levels.

Results: 1157 patients were prescribed with gemfibrozil over the eight weeks of data collection. 15% did not have any pre-treatment TG levels. Only 13.1% fit pre-treatment criteria for initiating gemfibrozil. 43% of those taking gemfibrozil have co-existing atherosclerotic disease or diabetes warranting the use of alternative lipid lowering therapy. On the one hand, the current daily cost of inappropriate prescription of gemfibrozil in these 1157 patients is HK$853. On the other hand, statins at a cost of HK$2.38 per dose would result in cost neutrality with an expected improvement in cardiovascular disease risk in the same group of patients.

Conclusion: This study has highlighted the overuse of gemfibrozil for a population who may benefit from alternative therapy.

Keywords: Gemfibrozil, fibrates, dyslipidaemia, triglyceridaemia, primary care, statins.

摘要

目的： 評估目前普通科門診(GOPCs)使用降脂藥的程度及是否需要使用其他不同類型降脂藥物。

設計： 回顧式分析。

研究對象： 新界區普通科使用gemfibrozil的所有病人。

主要測量內容： 按照英國聯合會心血管疾危險因素圖表，量度治療心血管危險因素的程度，並量度治療前甘油三酯的血濃度。

結果： 8個星期數據收集期內，共用1157位病人被處方gemfibrozil。其中15%未做治療前甘油三脂測量； 只有13.1%符合使用此藥指引的標準：43%使用gemfibrozil的病人患有動脈粥樣硬化或糖尿病，需要使用其他類型降脂藥。 在這1157位病人每日處方gemfibrozil的花費為853元，而相等價錢的他汀類藥物(每份為2.38元)可以幫助這群病人降低心血管類危險指數。

結論： 一部份病人被過量處方gemfibrozil，他們可能通過使用其他類型降脂藥而得益。

詞彙： gemfibrozil，貝特類降血脂藥，血脂異常，甘油三酸酯過高，基層醫療，他汀類降血脂藥。

Introduction

Hyperlipidaemia is not uncommon in daily general practice. A recent epidemiological study in China found an incidence of raised total cholesterol (TC) and triglyceride (TG) to be 21% and 29.3%¹ respectively. The traditional goal in lipid lowering therapy is the correction of a raised low-density lipoprotein (LDL) cholesterol which has been demonstrated by a number of clinical trials on statins.^2-5 However, despite a significant reduction in LDL levels (by up to 55%), the reduction in cardiovascular event rates have only been 22-35%.

Results from the Framingham Heart Study⁶ and the United Kingdom Prospective Diabetes Study⁷ highlighted the association between high-density lipoprotein (HDL) cholesterol and cardiovascular disease risk. Indeed, it is now evident that HDL is a powerful independent inverse predictor of coronary artery disease.⁸ The third report of the Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (ATP III)⁹ defined low HDL as a level less than 1.0 mmol/l, but it does not specify a target level for raising HDL. Raising HDL levels using gemfibrozil in those with coronary heart diseases (CHD) has been shown to significantly reduce the risk of death from CHD or non-fatal myocardial infarction,¹⁰ with an additional benefit of reducing TG levels. These results occurred despite a lack of a LDL lowering effect.

Hypertriglyceridaemia has also been shown to be a risk factor for CHD. After adjustment for HDL-cholesterol, which is often reduced in patients with hypertriglyceridaemia, high TG values were a significant but modest independent predictor of cardiovascular disease in both men and women (relative risk 1.14 and 1.37, respectively).¹¹

Fibrates have traditionally been used to increase HDL and reduce TG levels. Furthermore some studies have already shown their usefulness in the reduction of cardiovascular events in patients with low HDL¹² or high TG levels.¹³ However, as with any medication, gemfibrozil is not without side effects. In a New Zealand study, gemfibrozil more frequently caused adverse effects leading to withdrawal of treatment than simvastatin.¹⁴ The most common side effects¹⁵ include gastrointestinal problems, - dyspepsia (20%), abdominal pain (10%), diarrhoea (7%), nausea/vomiting (3%) and constipation (1%), and fatigue (4%). Serious and life-threatening adverse effects include rhabdomyolysis, seizures, and thrombocytopenia. In addition, many other medications have been found to have interactions with gemfibrozil (for example, warfarin, propranolol, amiodarone, fluoxetine, glipizide, glimepiride, glitazones, sertraline, citalopram, diazepam).

Recommendations have been proposed for the management of abnormal TG, along with guidance on the appropriate use of fibrates.⁹ At TG levels between 2.2-5.6mmol/L, treatment with fibrates should be initiated if patients are at high risk. Otherwise one should only start fibrate use when levels are greater than 5.6mmol/L. Gemfibrozil is frequently prescribed as a major lipid-lowering agent at the general outpatient clinics (GOPCs) which are providing about 15% of the primary care services in Hong Kong. This study aims to identify the characteristics of patients who were prescribed gemfibrozil, as well as assess the appropriateness of the prescriptions.

Method

Study locations and patients selection

General outpatient clinics within the New Territories East Cluster (namely Ma On Shan Family Medicine Centre, Yuen Chau Kok Clinic, Shatin Clinic, Lek Yuen Clinic [Shatin District]; Wong Siu Ching Family Medicine Centre, Tai Po Jockey Club Clinic [Tai Po District]; Shek Wu Hui Clinic[North District] ) were included as recruitment centres. Data from patients were analysed if they were prescribed with gemfibrozil and attended these centres within the eight weeks period from 6^th September to 30th October 2004.

Data collection

Doctors working at the above GOPCs were asked to collect the data (see below) of all patients currently prescribed with gemfibrozil through review of case notes (paper-records or computer records):

	Age
	Gender
	Blood pressure at the time of starting gemfibrozil
	Pre-treatment lipid levels (LDL, HDL, TC and TG)
	Presence of risk factors:
	Evidence of left ventricular hypertrophy Diabetes mellitus (DM) Overt atherosclerotic diseases (such co-existing peripheral vascular disease, ischaemic stroke, or myocardial infarction) (AS) Concurrent smoking

Data analysis

Collected data were input into SPSS 12.0 for further analysis. Patient's risks for cardiovascular events were calculated using the Joint British Societies Cardiovascular Disease Risk Prediction Chart and its associated risk calculator programme. This method incorporates the data collected above to calculate a patient's cardiovascular disease risk. Data from patients with pre-existing atherosclerotic diseases and DM were analysed separately because of their co-existing high cardiovascular disease risks. Those patients with a cardiovascular disease risk greater than 20% over 10 years are considered to be at high risk.

Results

Patient demographics

During the eight weeks data collection period, 1157 patients were identified to be taking gemfibrozil. Table 1 shows the characteristics of these 1157 patients within the different clinics and district areas.

The characteristics of patients from the Shatin and Tai Po regions were similar in terms of age and gender distribution. The North District has a slightly higher proportion in the elderly age group, as well as a much higher female-to-male ratio. This finding coincides with the higher incidence of LVH and DM, and the higher cardiovascular disease risks in the North District.

Triglyceride levels of patients with diabetes and atherosclerotic diseasesTriglyceride levels of patients with diabetes and atherosclerotic diseases

In total there were 25 patients with both DM and AS, 447 patients with DM alone and 65 patients with AS alone.

Among the 472 diabetic patients, 15% (71) did not have any TG levels taken prior to initiating gemfibrozil (Figure 1). Only 18.6% (88) had TG levels greater than 5.6mmol/l. The remaining 66.3% (313) had levels less than 5.6mmol/l.

Among the 90 patients with pre-existing atherosclerotic diseases, 24.4% (22) did not have any TG levels taken before starting gemfibrozil (Figure 2). Only 7.8% (7) had TG levels greater than 5.6mmol/l. The remaining 67.8% (61) had levels less than 5.6mmol/l.

Triglyceride levels of patients without diabetes or atherosclerosis

Among the non-diabetic, non-atherosclerotic patients with a calculated cardiovascular disease risk of less than 10% over 10 years, only 4% have TG greater than 5.6mmol/l prior to starting gemfibrozil. In those with risk levels between 10-20%, 10.9% have TG greater than 5.6mmol/l. In those with risk levels greater than 20%, 21.3% have TG greater than 5.6mmol/l (Figure 3). Table 2 summarises the distribution of the study population's cardiovascular disease risk levels.

Discussion

9.5% of patients (110/1157) did not have pre-treatment lipid TG levels available in their case notes. However, many patients attend GOPCs for treatment after the blood results done at private clinics. It would be more likely that such abnormal results were not filed or documented.

The indications for use of fibrates in the treatment of hyperlipidaemia have already been outlined.⁹ Patients who definitely need fibrate are those who have TG levels greater than 5.6mmol/l. In this study only 13.1% (151/1157) fit this criterion (Table 2, bold line).

Excluding those without TG levels prior to taking gemfibrozil, 72.3% (837/1157) have TG levels lower than 5.6mmol/l, a level that does not necessitate fibrate use (Table 2, dashed line).

30.9% (358/1157) of patients have AS or DM with initial TG levels less than 5.6mmol/l (Table 2, grey box). These patients will benefit from the use of alternative lipid lowering therapies (such as statins) as secondary prevention.

The costs of gemfibrozil and the various available statins (at equivalent doses^17,18) in the New Territories East Cluster are shown in Table 3. The daily cost of treating those having unnecessary gemfibrozil is HK$853 (HK$1.02 x 837). Among these 837 patients, 358 require the use of statins instead. In order to be cost-neutral, the price of statins will be around HK$2.38. This means that despite providing a more expensive treatment as secondary prevention, the use of certain statins will not incur a higher cost.

Drug costs should not be the only consideration in health economics. Indirect costs, such as hospitalization rates, patient morbidity and mortality rates, must be included in any health care economic analysis. Both fibrates and statins have been shown to provide benefits, each being recommended in different scenarios. However, the availability of statins in GOPCs is currently limited, thus propagating this overuse of gemfibrozil.

This study was limited by biased data collection, since it is possible that not all patients taking gemfibrozil were included by participating doctors. In addition, data collection is subject to individual doctor's incentive to participate, and the data inputs would vary depending on his/her interest in this area.

Those patients with TG levels between 2.2mmol/l and 5.6mmol/l may eventually benefit from the use of gemfibrozil, depending on the success of non-pharmacological interventions. However, as mentioned above, the primary treatment target for this subgroup of patients would be non-HDL cholesterol. Therefore the use of statins would not be an unreasonable suggestion if cardiovascular disease risks were high.

The 30.9% of patients requiring statin therapy by criteria in this study may in fact be overestimated if other factors such as biological age, coexisting diseases (e.g. dementia) and other daily function indices were considered together. The calculations above have demonstrated that a large proportion of patients could receive appropriate therapy without extra cost. This estimation has assumed that low statin doses would be adequate for the control of high cholesterol level among our study populaton. A recent local study has indicated that in Hong Kong Chinese, lower doses of statins can achieve the same LDL levels,¹⁸ therefore reinforcing the applicability of this assumption.

Health resources have been under constraint for many years. The appropriate allocation of such limited resources would help improving population health. Obviously, statins are expensive drugs. Furthermore, latest studies have proposed cholesterol targets that are well below previous goals,¹⁹ additionally demanding for more resources. Continuing medical education for doctors with regards to optimal and efficient lipid management, promotion of healthy lifestyle modifications and the rational prescription of lipid lowering agents in GOPCs will all ensure better health care services provision in Hong Kong.

Conclusion

Appropriate use of resources can translate into greater benefits for a larger population. Results from this retrospective analysis illustrate this concept, where the cost of using gemfibrozil can be shifted to another population, resulting in greater reductions in cardiovascular disease risk. Future directions will include implementing the appropriate use of both gemfibrozil and alternative therapies such as statins in GOPCs, and observing whether such measures will translate into lower costs and greater health benefits.

Acknowlegement

The authors would like to thank all participating doctors working in NTEC's GOPCs who helped in data collection. Special acknowledgements go to the following doctors: Dr. Kwan Chi Wa, Dr. Ngai Kwok Fung, Dr. Lai Chor Yat, Dr. Cheung Yu, Dr. Chan Lin, Dr. Hui Ming Tung, Dr. Su Whang, Dr. Lau Tze Leung, and Dr Sham Man Wai.

Key messages

1. There is an overuse of gemfibrozil in the GOPC setting.
2. A significant proportion of patients despite being prescribed with gemfibrozil are still at high cardiovascular disease risk. Such patients may benefit from alternative lipid lowering therapy.
3. Judicious use of alternative lipid lowering therapy such as statins together with the appropriate prescription of fibrates may provide a more cost-effective lipid lowering management in the general outpatient setting.