Page images

John Gorringe

25 years Dr John A. L. Gorringe was involved in International Clinical Research at Parke, Davis & Co. and from 1962 until his retirement in 1983 he was its Director.

As such, he was concerned with the development of a wide range of pharmaceutical products, including anti-infective agents, anti-convulsants, anti-inflammatories, anaesthetics, oxytocics, analgesics, anti-virals and beta-blockers. In addition, however, he was one of the leading forces behind the development of a comprehensive programme of studies on lipid-lowering agents, and particularly gemfibrozil.

This collection of papers, Further Progress with Gemfibrozil, has two purposes. Firstly, it sets out to describe the current state of investigation with this particular lipid-lowering agent. In addition, however, it is meant to represent a small tribute to John Gorringe from those investigators, and friends, who had the pleasure of collaborating with him during the course of this work.

Influence of gemfibrozil on lipoprotein composition: triglyceride removal capacity and fatty acid composition of the

plasma lipid esters


Department of Geriatrics, University of Uppsala, Uppsala, Sweden


Seventeen hyperlipidaemic patients on long-term treatment with gemfibrozil were studied before and 2 months after cessation of treatment. Serum triglycerides and serum total cholesterol increased significantly on withdrawal, largely because of an increase in very low density lipoproteins, while high density lipoprotein cholesterol decreased by 21 per cent (p<0.001). Serum apolipoprotein (apo) B concentration increased ( + 12 per cent; p<0.01) and apo A-1 and apo A-II decreased by 6 per cent (p<0.05) and 11 per cent (p<0.05) respectively. Triglyceride reduction during gemfibrozil treatment seems at least partly due to an increased clearance of triglycerides from the circulation. The fractional removal rate (K2) during the intravenous fat tolerance test decreased by 21 per cent (p<0.01), corresponding to decreased lipoprotein lipase activities in skeletal muscle (significantly lowered in males) and in adipose tissue (significant decrease in females) when the treatment was stopped. Only slight changes in plasma lipid fatty acid composition were observed, and there were no changes in blood glucose, glucose tolerance, or serum insulin concentrations.

Gemfibrozil has been shown to lower serum triglyceride levels as well as raising high density lipoprotein (HDL)-cholesterol concentrations (Nikkilä et al. 1976, Olsson et al. 1976, Vessby et al. 1976). However, information on the mechanisms by which gemfibrozil influences triglyceride metabolism in hypertriglyceridaemia is rather sparse.

The present study was undertaken to investigate the effects of gemfibrozil on triglyceride removal capacity, as measured by the intravenous fat tolerance test and by determination of tissue lipoprotein lipase activities. The fatty acid composition of

Further Progress with Gemfibrozil, edited by C. Wood, 1986: Royal Society of Medicine Services International Congress and Symposium Series No. 87. published by Royal Society of Medicine Services Limited.

[blocks in formation]


B. Vessby et al.

the plasma lipid esters was also studied during and after treatment to investigate any possible changes occurring during gemfibrozil treatment.

Materials and methods

Patient population

Seventeen patients, 13 males (mean age 60 years; range 46-76) and four females (mean age 68 years; range 62-78) were investigated in an open study, during and after cessation of gemfibrozil treatment. All patients had been on gemfibrozil (600 mg twice daily) for several years when the treatment was terminated. Laboratory tests were performed during the final days of treatment and 2 months after finishing gemfibrozil. All patients had been on long-term lipid lowering diets and they were instructed to keep their dietary habits and physical activity unchanged during the wash-out period. Some patients were receiving additional treatment with other drugs which were kept unchanged during the study period. All laboratory analyses were performed in the morning following an overnight fast.


Triglyceride and total cholesterol concentrations in serum and in the isolated lipoprotein fractions were determined by semi automated methods with a Technicon Autoanalyzer II (Rush et al. 1971) using a combination of preparative ultracentrifugation (Havel et al. 1955) and heparin-manganese precipitation (Burstein and Samaille 1960) to separate very low density lipoproteins (VLDL), low density lipoproteins (LDL) and HDL, as described previously (Vessby et al. 1980a,b). The apolipoprotein (apo) A-I, A-II and B concentrations in serum were determined by electro-immunoassay according to the methods of Vessby et al. (1980a,b). Serum apolipoprotein concentrations are expressed as arbitrary units (AU) relative to the concentrations in a reference serum (100 AU) obtained from a large pool of healthy blood donors.

Specimens of abdominal adipose tissue and of skeletal muscle tissue from the lateral vastus muscle were taken for determination of lipoprotein lipase activity as described by Lithell and Boberg (1977, 1978). A triolein-phospholipid emulsion (Nilsson-Ehle and Schotz 1976) and a heparin concentration of 2 g/l (about 240 IU/ml) were used for both adipose and skeletal muscle tissue. The activity is expressed in mU/g (1 mU = 1 nmol fatty acid released per min). The intravenous fat tolerance test was performed as described by Carlson and Rössner (1972).

For determination of the plasma lipid ester fatty acid composition, blood was drawn into heparinized test-tubes and the plasma lipids were extracted into chloroform containing 0·005 per cent butylated hydroxytoluene (BHT) as an antioxidant. The plasma lipid esters, cholesterol esters, triglycerides and phospholipids were separated by thin-layer chromatography (Boberg 1966). The lipid esters were transmethylated at 60 °C overnight after addition of 5 per cent H2SO4 in methanol. The fatty acid methyl esters were separated by gas chromatography on a 25 m open tubular glass capillary column coated with SP1000, using helium as carrier gas. The fatty acids were identified by comparing retention times with those of Nu Check Prep (Elysian Gemfibrozil lipoproteins and fatty acids


Minnesota) fatty acid methyl ester standards and Supelco (Bellefonte, Pennsylvania) PUFA-mix No. 2 (animal source, 22:5 w-6, 22:5 w-3). The following fatty acids are reported on: 16:0, 16:1 w-7, 18:0, 18:1 w-9, 18:2 w-6, 18:3 w-6, 18:3 w-3, 20:3 w-6, 20:5 w-3, 22:4 w-6, 22:5 w-3 and 22:6 w-3. Relative concentrations of individual fatty acids are expressed as a percentage of the sum of all fatty acids together. Other fatty acids identified but present in concentrations no greater than 0.1 per cent are not reported on. A detailed description of the separation procedure has been reported elsewhere (Boberg et al. 1985).

Blood glucose and urinary glucose concentrations were determined by the glucose oxidase method (Hjelm and deVerdier 1963). Serum insulin concentrations were measured using the Phadebas (Pharmacia, Uppsala, Sweden) test method based on a radio-immunological technique (Wide and Olsson 1974).


Means and standard errors of the mean were calculated by ordinary methods. The effect of the treatment has been tested for each variable using a three-way and a fourway analysis of variance model (Lindman 1974). In the three-way model the sources of variation were: sex, time and patient (according to sex) as main factors. In the four-way model the patients were also classified according to weight reduction during the follow-up period.


Effects on serum lipoprotein and apolipoprotein concentrations

On terminating gemfibrozil treatment there were, as expected, significant increases of triglyceride concentrations in VLDL, LDL and HDL while HDL cholesterol decreased significantly, by an average of 21 per cent (Table 1). Serum triglyceride and serum total cholesterol concentrations also increased significantly. Simultaneously there was a significant increase in serum apo B, while serum apo A-I and apo A-II concentrations both decreased.

The composition of serum lipoproteins (as expressed by the ratios of apolipoprotein to lipid concentration) was changed in the following respects. The serum apo B/LDLcholesterol ratio increased from 34.511.1 (mean + SEM) to 37.5 +1.6 (p<0.05) on cessation of treatment. Similarly, the serum apo A-II/HDL-cholesterol ratio increased from 106 3 to 118+ 3 (p<0.001). There was a relatively more pronounced decrease of the serum apo A-II concentration (-11 per cent) than of apo A-I (-6 per cent), causing a significant increase of the apo A-1/apo A-II ratio, from 0.90 +0.01 to 0.95 +0.01 (p<0.05).

After terminating the treatment there was a limited, but significant, increase in mean body weight from 77.812.3 to 79.0 +2.3 kg (p<0.01). Since body weight changes might themselves effect serum lipoprotein composition and thus constitute a confounding factor, a model for variance analysis was used to compare changes recorded in patients with a body weight increase > 1.0kg to those in patients with unchanged body weight (<1.0kg) during the wash-out period. The two groups generally showed similar changes after treatment. With regard to the effects on serum


B. Vessby et al.

Table 1
Serum lipoprotein and apolipoprotein concentrations during and after

termination of gemfibrozil treatment (mean + SEM; n = 17)

On gemfibrozil

After gemfibrozil

VLDL TG, mmol/l
VLDL CHOL, mmol/l
LDL TG, mmol/l
LDL CHOL, mmol/l
HDL TG, mmol/l
HDL CHOL, mmol/l
SERUM TG, mmol/l
SERUM CHOL, mmol/l
Apo B, AU
Apo A-I, AU
Apo A-II, AU

0.56 + 0.08
0.40 + 0.02
4.92 +0:37
0.22 +0.01
1.86 + 0.28
6.61 +0.36

165 + 12
104 + 3
115 + 3

3:41 +0.57**
1.61 +0.23***
0.68 +0.05***
5:22 +0.39ns
0.27 +0.01**
0.89 +0.04***
4.41 +0.58**
7.55 +0.33**
184 + 10**

98 + 3*
102 + 3*

*= p<0.05, 0.01 and 0.001, respectively, compared with gemfibrozil creatment
TG = serum triglyceride; CHOL = serum total cholesterol; Apo = serum apolipoprotein.
AU = arbitrary units

lipids and apolipoproteins, the only differences were in LDL-cholesterol and serum total cholesterol. Those patients who gained the most weight tended to show an increase in LDL-cholesterol and a more pronounced increase in serum total cholesterol concentrations than the group with stable body weight.

Effects on triglyceride removal capacity

There was a highly significant reduction (21 per cent; p<0.002) in the fractional removal rate (K? during the intravenous fat tolerance test (IVFTT) after cessation of treatment. During treatment K2 was 3.6+0.4 per cent, falling, after treatment, to 2.6+0.3 per cent. Concomitantly there were reduction of mean lipoprotein lipase activity in adipose tissue from 198 + 15 to 179 + 14 mU/g and in skeletal muscle tissue from 37 +4 to 29+ 4 mU/g. These changes (10 and 22 per cent, respectively) were not statistically significant in the group of 13 patients from whom adequate tissue samples were obtained. However, analysis of the two sexes separately showed a significant reduction of the adipose tissue lipoprotein lipase activity from 247 – 39 to 177 + 28 mU/g in females (n = 4; p<0.05) while the skeletal muscle lipoprotein lipase activity decreased in the male subjects from 34 + 4 to 25 + 3 mU/8 (n = 9; p<0.05). Reduction of lipoprotein lipase activity in adipose tissue was more pronounced in patients who gained most weight, but there was no significant relationship between changes in body weight and lipoprotein lipase activity ir skeletal muscle.

There were significant correlations between changes in skeletal muscle lipoprotein lipase activity and serum apolipoprotein B concentration (r=0.65; p<0.05). Although the mean value of LDL-cholesterol did not increase after treatment, individual changes of LDL-cholesterol were correlated with individual changes in skeletal muscle lipoprotein lipase activity (Fig. 1). Changes in adipose tissue lipoprotein lipase were negatively correlated with body weight changes (r= -0.56; p<0.05) but not significantly correlated with changes in serum triglyceride or VLDL triglyceride.

« PreviousContinue »