- Email: [email protected]
Dyslipidemia in early second trimester is mainly a feature of women with early onset pre-eclampsia
British Journal of Obstetrics and Gynaecology October 2001, Vol. 108, pp. 1081–1087
Dyslipidemia in early second trimester is mainly a feature of women with early onset pre-eclampsia Torun Clausen a,*, Srdjan Djurovic b, Tore Henriksen c Objective To investigate whether hypertriglyceridemic dyslipidemia is a risk factor for either early or late onset pre-eclampsia. Design Prospective cohort study and nested case–control study. Setting Aker Hospital: a university hospital with all levels of obstetric care. Participants 2157 Caucasian pregnant women. Methods Blood samples were obtained from non-fasting subjects at 18 weeks of gestation. All samples were analysed for triglycerides, total-cholesterol, high density lipoproteins cholesterol and non-high density lipoproteins cholesterol. ApoB-100 were analysed in pre-eclamptic women and in 3:1 matched controls. The cohort data were analysed by multiple logistic regression and the case–control data by conditional logistic regression. Main outcome measures Adjusted odds ratios of early and late onset pre-eclampsia according to early second trimester serum concentration levels of lipids and ApoB-100. Results Eighteen women developed early onset pre-eclampsia and 53 women developed late onset pre-eclampsia. In the cohort model, women with triglycerides above 2.4mmol/L had increased risk (OR 5.1; 95% CI 1.1– 23.1) of early onset pre-eclampsia compared with those with triglycerides levels # 1.5mmol/L. For women with high triglycerides: non-high density lipoproteins cholesterol ratios (.90 centile) the OR (95% CI) for early onset pre-eclampsia was 7.1 (2.3–22.0) compared with those with low ratios (# 50 centile). Similar associations were found in the case control model. We found no associations between plasma lipids and risk of late onset pre-eclampsia. Conclusions Hypertriglyceridemic dyslipidemia before 20 weeks of gestation is associated with the risk of developing early but not late onset pre-eclampsia, giving support to the contention that these two variants of the disease are at least partly pathogenically different.
INTRODUCTION Pre-eclampsia is a hypertensive complication of pregnancy associated with well-documented risks for the mother and the fetus 1. Although hypertension and proteinuria are the simple clinical criteria for diagnosis of preeclampsia the pathophysiological mechanisms that lead to the disorder are by all evidence very diverse 2. Preeclampsia develops in a particular woman following an unfortunate combination of maternal (trophoblast-independent) risk factors and an excessive maternal response to the trophoblast/trophoblast-derived factors 2,3. The maternal risk factors include disorders associated with
a
Departments of Obstetrics and Gynaecology, Aker and Ullevaal University Hospitals, Oslo, Norway b Institute for Medical Genetics and Department of Medical Genetics, Ullevaal University Hospital,Oslo, Norway c Department of Obstetrics and Gynaecology, National Hospital, Oslo, Norway * Correspondence: Dr T. Clausen, Department of Obstetrics and Gynaecology, Ullevaal University Hospital, 0407 Oslo, Norway. q RCOG 2001 British Journal of Obstetrics and Gynaecology PII: S 0306-545 6(01)00247-9
endothelial dysfunction such as chronic hypertension, diabetes, kidney disease and dyslipidemia, as well as more specific defects such as protein C or protein S deficiencies, activated protein C resistance, anticardiolipin antibodies and hyperhomocysteinemia 1,4. Early onset pre-eclampsia exhibits features that are unusual in late onset disease 5. For example, the recurrence risk of pre-eclampsia is relatively high in women with early onset disease, the proportion of women with chronic hypertension is higher, fetal growth restriction is much more common and multi-organ involvement, as reflected in the HELLP syndrome, is more frequent 5–7. Thus early and late onset pre-eclampsia seem, at least partly, to be pathophysiologically different. A further clarification of these differences is of clinical interest because early onset disease is associated with a higher prevalence of iatrogenic prematurity, intrauterine growth restriction and severe maternal disease 5,8. In normal pregnancy the circulating concentrations of triglycerides and cholesterol increases progressively 9–12. In women with pre-eclampsia the serum lipids show a shift towards a dyslipidemic profile: The serum triglycerides are further elevated, there is a higher proportion of low density lipoproteins of small sizes, high density lipowww.bjog-elsevier.com
1082 T. CLAUSEN ET AL.
protein decreased and serum free fatty acids are increased 9,13–15. The pre-eclamptic dyslipidemia is not a consequence of the disease because it is present long before the disorder becomes clinically overt and tends to be present after pregnancy 14,16–18. Because the pathogenesis of early and late onset pre-eclampsia seems to differ slightly we wanted to investigate if hypertriglyceridemic dyslipidemic patterns are risk factors for both these two variants of the disorder.
METHODS The setting of the present study was Aker University Hospital in Oslo, which until December 1999 covered defined geographical areas of Oslo City, representing all socio-economic classes. Approximately 95% of the pregnant women in these areas gave birth at this hospital. All pregnant women in the areas of Aker Hospital were offered an ultrasound scan at 17–19 weeks of gestation. In a period of 21 months in 1995–1996 Caucasian nondiabetic women were asked to provide a blood sample at the time of the scan. Blood samples from 2434 women were obtained (i.e. from 93.4% of the recruited women), and 2190 of the women who provided blood samples were delivered at Aker Hospital. After excluding women with twin pregnancies, 2157 women remained for the study. Diagnosis of pre-eclampsia required the presence of proteinuria and pregnancy-induced hypertension. Proteinuria was defined as $ 11 on dipsticks (300mg/24 h) found twice at least six hours apart. Pregnancy-induced hypertension was defined either as blood pressure $ 140/ 90mmHg (n ¼ 59) or as an increase in diastolic pressure of $ 15mmHg (n ¼ 12), compared with the average before 20 weeks of gestation. In both cases two measurements at least six hours apart were required. The preeclamptic women were divided into groups of early and late onset pre-eclampsia (i.e. delivered before and after 37 weeks of gestation, respectively). Blood samples were drawn in non-fasting state and allowed to coagulate before centrifugation at 400g for 10 minutes. The serum samples were transferred on ice to a -708C freezer within 140 minutes after venipuncture. In the whole cohort (n ¼ 2157) we measured serum concentrations of triglycerides, total cholesterol and high density lipoprotein cholesterol. Non-high density lipoproteins cholesterol and the ratio of triglycerides: high density lipoproteins cholesterol and triglycerides: non-high density lipoproteins-cholesterol were calculated. ApoB-100 was measured in the 71 women with subsequent pre-eclampsia and in 3:1 matched controls consisting of healthy pregnant women from the study population. The controls were matched for age, parity and body mass index. The rational for choosing the parameters above was based on previous observations: fasting triglycerides
but not cholesterol at 18 weeks of gestation are increased in women with subsequent pre-eclampsia indicating enhanced production and/or impaired degradation of triglyceride-rich lipoproteins 14,19,20. The selected parameters may reflect such disturbances in the metabolism of triglyceride-rich lipoproteins. Similar kinds of metabolic disturbances are seen in individuals with ‘metabolic syndrome’ 21. The typical features of the plasma lipids in these individuals include high triglycerides and low highdensity lipoproteins cholesterol. Total-cholesterol and non-high density lipoproteins cholesterol may be increased or within normal range 22. High values of the ratios of triglycerides: non-high density lipoproteins cholesterol and triglycerides: ApoB may reflect accumulation of triglyceride-rich particles. The study protocol was approved by the Regional Medical Ethics Committee and written informed consent was obtained from each woman. Statistical analysis For data analysis we used the statistical program 8.0 SPSS for Windows (SPSS Inc, Chicago, Illinois, USA). Proportions were compared using the x 2 test. Women with and without pre-eclampsia were compared using the Mann U Whitney test. Odds ratio for pre-eclampsia was evaluated by multiple logistic regression analyses (cohort data) and conditional logistic regression (case–control data). The odds ratios in the multiple logistic regression models are shown with adjustment for parity, maternal age, body mass index and smoking, whereas the odds ratios in the matched case control analyses are presented with adjustment for smoking. Stratified Cox regression models were used to fit the conditional logistic regression model, because these two methods give the same likelihood functions and thus the same estimates 23. Triglycerides and the ratios triglycerides: nonhigh density lipoproteins cholesterol and ApoB: nonhigh density lipoproteins cholesterol were categorised according to the 50 th and 90 th centiles of their distribution to allow examination of dose–response relationship for extreme upper categories. High density lipoproteins cholesterol and the ratio ApoB: triglycerides were categorised in the same manner according to the 50 th and 10 th centiles to allow for examination of dose–response relationship for extreme lower categories. The number of cases below the 10 th centile was, however, too small for statistical analysis and we therefore categorised those variables in two groups only: # 50 th centile and .50 th centile. RESULTS Baseline characteristics of the cohort are presented in Table 1. Eighteen women developed early onset preeclampsia whereas 53 women developed late onset q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
Table 1. Clinical characteristics of the cohort and serum lipid concentrations at 18 weeks of gestation. Values are given as median (25-75 th centile) unless otherwise shown. BMI ¼ body mass index; PE ¼ preeclampsia; SBP ¼ systolic blood pressure; DBP ¼ diastolic blood pressure; TG ¼ triglycerides; HDL ¼ high density lipoproteins; CH ¼ cholesterol.
a b
Mann-Whitney U test. x 2 test.
Without PE n ¼ 2086
PE ^ 37 weeks n ¼ 53
Pa
PE ,37 weeks n ¼ 18
Pa
30 (27-33) 22.1 (20.5-24.4) 51.2 22.7 117 (110-125) 75 (67-80) 40.0 (39.0-41.0) 3600 (3250-3930)
30 (27-33) 22.1 (20.5-24.4) 50.4 22.9 115 (110-124) 73 (67-80) 40.0 (39.0-41.0) 3605 (3270-3930)
29 (26-34) 22.8 (21.0-25.0) 69.8 19.2 140 (135-150) 95 (90-100) 39.0 (38.0-40.0) 3390 (3145-3930)
0.50 0.13 0.005 0.54 ,0.001 ,0.001 ,0.001 0.15
27 (24-30) 21.4 (19.9-23.7) 83.3 16.7 152 (145-173) 100 (95-110) 34.0 (29.25-35.0) 1900 (1060-2298)
0.02 0.46 0.005 0.78 ,0.001 ,0.001 ,0.001 ,0.001
1.50 (1.20-1.90) 5.30 (4.80-5.90) 1.75 (1.51-2.00) 2.80 (2.30-3.30) 0.86 (0.63-1.18) 0.54 (0.43-0.66)
1.50 (1.20-1.90) 5.30 (4.80-5.90) 1.75 (1.51-2.01) 2.80 (2.30-3.30) 0.86 (0.63-1.18) 0.54 (0.43-0.66)
1.50 (1.20-1.90) 5.10 (4.60-5.83) 1.72 (1.52-1.89) 2.75 (2.32-3.40) 0.88 (0.68-1.22) 0.56 (0.45-0.70)
0.81 0.63 0.24 0.98 0.35 0.46
1.85 (1.30-2.60) 4.90 (4.50-5.40) 1.65 (1.53-1.92) 2.45 (1.90-2.60) 1.11 (0.72-1.64) 0.75 (0.50-1.14)
0.14 0.02 0.29 0.003 0.125 0.008
DYSLIPIDEMIA AND EARLY ONSET PRE-ECLAMPSIA 1083
Age, (years) BMI (kg/m 2) Nulliparae, (%) b Smoking (%) b SBP at delivery (mmHg) DBP at delivery (mmHg) Gestational age (weeks) Birthweight (g) Lipid parameters TG (mmol/l) Total-CH (mmol/L) HDL-CH (mol/L) NonHDL-CH (mol/L) TG/HDL-CH TG/nonHDL-CH
Cohort n ¼ 2157
1084 T. CLAUSEN ET AL. Table 2. Clinical characteristics and serum lipid concentrations at 18 weeks of gestation in women with late ($ 37 weeks) and early (,37 weeks) onset preeclampsia (PE) and in matched controls. Values are given as median (25-75 th centile) unless otherwise shown. BMI ¼ body mass index; TG ¼ triglycerides; HDL ¼ high density lipoproteins; CH ¼ cholesterol. PE $37 weeks n ¼ 53 Age (years) BMI (kg/m) 2 Nullipara, n (%) b Smoking n (%) b Lipid parameters TG (mmol/L) Total CH (mmol/L) HDl-CH (mmol/L) NonHDL-CH (mmol/L) TG/HDL TG/nonHDL-CH ApoB-100 mg/100 mL ApoB-100/TG ApoB-100/nonHDL a b
Controls n ¼ 149
29.0 (26.0-34.0) 22.8 (21.0-25.0) 37 (69.8) 10 (19.2) 1.50 (1.20-1.90) 5.10 (4.60-5.83) 1.72 (1.52-1.89) 2.75 (2.32-3.40) 0.88 (0.68-1.22) 0.56 (0.45 (0.70) 125.0 (106.0-150.0) 84.4 (68.8-98.3) 45.4 (42.2-49.5)
29.0 (26.0-34.0) 22.6 (21.0-24.9) 102 (68.5) 37 (25.3) 1.60 (1.30-2.10) 5.30 (4.80-5.90) 1.71 (1.52-1.89) 2.90 (2.40-3.40) 0.93 (0.70-1.34) 0.57 (0.47-0.67) 134.0 (112.0-148.25) 80.0 (70.3-95.5) 45.8 (43.4-48.4)
Pa
PE ,37 weeks n ¼ 18
Controls n ¼ 51
Pa
0.97 0.93 0.86 0.37
27.0 (24.0-30.0) 21.4 (19.9-23.7) 15 (83.3) 3 (16.7)
27.0 (24.0-31.0) 21.6 (20.7-23.7) 42 (82.4) 15 (30.0)
0.95 0.55 1.00 0.27
0.24 0.35 0.78 0.57 0.65 0.77 0.45 0.51 0.62
1.85 (1.30-2.60) 4.90 (4.50-5.40) 1.65 (1.53-1.92) 2.45 (1.90-2.60) 1.11 (0.72-1.64) 0.75 (0.50-1.14) 116.5 (105.0-129.0) 66.8 (44.6-92.1) 50.4 (45.4-55.8)
1.40 (1.10-2.00) 5.30 (4.90-5.90) 1.78 (1.53-2.06) 2.80 (2.40-3.50) 0.76 (0.58-1.27) 0.50 (0.38-0.70) 130.0 (112.0-153.0) 91.6 (71.1-108.3) 45.9 (43.1-48.9)
0.15 0.03 0.24 0.01 0.12 0.01 0.02 0.02 0.02
Mann-Whitney U test. x 2 test.
well as of the ApoB: triglycerides ratio. Furthermore, the ratios triglycerides: non-high density lipoproteins cholesterol and ApoB: non-high density lipoproteinscholesterol were increased in the women who developed early onset pre-eclampsia. We then estimated the risk of developing early and late onset pre-eclampsia in relation to increasing serum levels of concentrations of triglycerides and to the ratio triglycerides: non-high density lipoproteins cholesterol using multiple logistic regression analyses (cohort data, Table 3). Women with serum triglycerides at 18 weeks of gestation above 2.4mmol/L had increased risk of developing early onset pre-eclampsia (OR 5.0, 95% CI 1.5–16.6) compared with levels of 1.5mmol/L or lower. For women with ratio triglycerides: non-high density lipoproteins cholesterol above 53, the odds ratio for developing early onset pre-eclampsia was 7.1 compared with ratios at
disease. No difference was found in early second trimester serum concentrations of lipids between women without pre-eclampsia and the ones who subsequently developed late onset disease. In the women with eventual early onset pre-eclampsia, however, the serum concentrations of total-cholesterol and non-high density lipoproteins cholesterol were significantly lower and the ratio triglycerides: non-high density lipoproteins cholesterol was higher than in the women without pre-eclampsia. Table 2 shows the baseline characteristics in the case– control part of the study. The serum lipids and ApoB concentrations of women with late onset pre-eclampsia and the matched controls were similar. Compared with controls the women with early onset disease had, however, decreased serum levels of total-cholesterol, non-high density lipoproteins cholesterol and ApoB as
Table 3. Adjusted odds ratios (ORs) for late and early onset pre-eclampsia (PE) in relation to serum lipid levels at 18 weeks of gestation. (cohort study). TG ¼ triglycerides; HDL ¼ high density lipoproteins; CH ¼ cholesterol. Late onset PE
Triglycerides (mmol/L) # 1.5 1.5-2.4 . 2.4 HDL-CH (mmol/L) # 1.75 . 1.75 TG/nonHD-CH # 46.0 46.0-53.0 . 53.0
Early onset PE a
n
n
%
OR (95% CI)
n
%
OR a (95%CI)
1148 834 175
29 19 5
2.5 2.3 2.9
1.0 0.9 (0.5-1.6) 1.3 (0.5-3.5)
8 4 6
0.7 0.5 3.4
1.0 0.7 (0.2-2.5) 5.0 (1.5-16.6)
1096 1061
31 22
2.8 2.1
1.0 0.7 (0.4-1.2)
13 5
1.2 0.5
1.0 0.4 (0.1-1.3)
1077 855 223
25 25 2
2.3 2.9 0.9
1.0 1.4 (0.8-2.5) 0.5 (0.1-2.0)
6 4 8
0.6 0.5 3.6
1.0 1.0 (0.3-3.4) 7.1 (2.3-22.0)
a
Adjusted for age, body mass index, nulliparous (yes/no) and smoking (yes/no). 226 of the women in the cohort, including one with early onset and two with late onset pre-eclampsia, were excluded in the analyses because of missing background data. The individuals with missing data were distributed proportionally in the categories of plasma lipids.
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
DYSLIPIDEMIA AND EARLY ONSET PRE-ECLAMPSIA 1085
46 or lower. These risks were not found in the pregnancies with late onset disease (Table 3). In the case–control study using conditional logistic regression analyses similar associations between high serum triglycerides values or the ratio triglycerides: nonhigh density lipoproteins cholesterol and the risk of developing early onset pre-eclampsia were found (Table 4). In the group with late onset disease no similar associations were observed. In this study we could also estimate the associations between the ratios ApoB: non-high density lipoproteins cholesterol and ApoB: triglycerides and the risk of subsequent pre-eclampsia. The former ratio was positively associated with the risk of developing early onset disease with odds ratio of 5.1 if the ratio was above 0.54 compared with 0.20 or lower. This association was not found in the group with late onset pre-eclampsia.
DISCUSSION It is well documented that in all pregnancies plasma lipid levels change towards a hyperlipidemic lipid profile. Although total cholesterol and high density lipoproteins cholesterol are increased, hypertriglyceridemia is the most prominent characteristic 9,12. All lipoprotein fractions show increased content of triglycerides 9,10. Increased synthesis of triglyceride-rich lipoproteins as well as decreased degradation (lipolysis) seems to contribute to the gestational hyperlipidemia 9,19,20. In women with overt pre-eclampsia the hypertriglyceridemia is even more pronounced 9,14,24. The pre-eclamptic dyslipidemia-like pattern of plasma lipids is apparently not a simple exaggeration of the physiological changes
because there is usually no further increase in total cholesterol and high density lipoproteins compared with normal pregnancies 9. In fact, in women with overt preeclampsia, plasma high density lipoproteins cholesterol and apoprotein A-I are usually decreased 9,13. In the present study we found that dyslipidemia-like changes in the plasma lipids at 18 weeks of gestation are associated with early but not late onset pre-eclampsia. Because of the prospective design of the study we can conclude that the observed plasma lipid pattern was not a late consequence of the pathophysiological changes in preeclamptic women. The plasma lipid pattern associated with early onset pre-eclampsia was found in the cohort as well as in the case–control studies. Using non-fasting blood samples as in the present study leads to greater variation especially in the triglycerides simply because of intestinally derived triglyceriderich lipoproteins (chylomicrons) 25. Despite this, early and late onset pre-eclampsia could clearly be distinguished already at 18 weeks of pregnancy. In previous studies of plasma lipids of pre-eclamptic women, distinction between early and late onset disease has not been made. As demonstrated in the present study such a distinction had major impact on the results and provides support to the notion that early and late onset pre-eclampsia may differ pathophysiologically. There is no general agreement with respect to the definition of early and late onset pre-eclampsia. We defined a woman as having early onset disease if she was delivered because of pre-eclampsia before 37 weeks of gestation. It may be argued that the use of 34 weeks of gestation would be more in accordance with current obstetric practices. The sample size of pre-eclamptic pregnancies was
Table 4. Adjusted odds ratios (OR) of late (PE $37 weeks) and early onset pre-eclampsia (PE ,37 weeks) in relation to lipid and apoB-100 levels at 18 weeks of gestation (nested case–control study). Values are given as median (25-75 th centile) unless otherwise shown. HDL ¼ high density lipoproteins; CH ¼ cholesterol; TG ¼ triglycerides; PE ¼ pre- eclampsia.
Triglycerides (mmol/L) # 1.5 1.5-2.4 . 2.4 HDL-CH (mmol/L) # 1.75 . 1.75 TG/nonHDL-CH # 46.0 46.0-53.0 . 53.0 ApoB/TG # 81.5 . 81.5 ApoB/nonHDL-CH # 0.20 0.20-0.54 . 0.54 a
Control n
PE $37weeks
OR a (95% CI)
Control n
PE ,37weeks
OR a (95%CI)
68 65 17
29 19 5
1.0 0.8 (0.4-1.6) 0.7 (0.2-2.2)
30 16 5
8 4 6
1.0 1.0 (0.2-5.1) 6.9 (1.1-41.9)
85 64
31 22
1.0 0.8 (0.4-1.6)
23 28
13 5
1.0 0.3 (0.1-1.0)
61 74 14
25 25 2
1.0 0.8 (0.4-1.5) 0.3 (0.1-1.6)
29 15 7
6 4 8
1.0 1.8 (0.4-8.5) 10.3 (1.5-69.5)
81 68
24 27
1.0 1.3 (0.7-2.6)
18 32
11 7
1.0 0.2 (0.04-0.9)
79 59 11
27 11 2
1.0 1.0 (0.5-2.2) 0.8 (0.2-4.4)
25 20 5
5 7 6
1.0 1.8 (0.5-6.7) 4.8 (1.0-22.5)
Adjusted for smoking (yes/no).
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
1086 T. CLAUSEN ET AL.
too small in the present study to allow categorisation before and after 34 weeks. The lipid changes that were associated with increased risk of early onset pre-eclampsia were characterised by high plasma triglycerides in absolute terms and relative to non-high density lipoproteins-cholesterol and to ApoB100 and by increased ratio of ApoB-100: non-high density lipoproteins cholesterol. We cannot exclude the fact that the special lipid patterns found in many of those with early onset pre-eclampsia developed during the first 18 weeks of pregnancy. However, the increasing evidence that pre-eclampsia is associated with athersclerotic cardiovascular disease later in life indicates that these women have changes in their lipid, glucose or energy metabolism that are independent of pregnancy. Insulin resistance may be involved because increased prevalence is found in pre-eclamptic women and because the lipid patterns presently found in women with early onset pre-eclampsia bear some similarities to those described in non-pregnant women with non-insulin dependent diabetes 21,26,27. The differences in the plasma triglycerides between the women with and without subsequent pre-eclampsia could either be due to higher endogenous (liver-produced) triglyceride-rich lipoproteins (ApoB-100 containing lipoproteins), to the intestinal B-48 dominated lipoproteins or to both. We have reported previously that fasting triglycerides are increased at 18 weeks of gestation in women with eventual pre-eclampsia 14. Thus at least parts of elevated plasma triglycerides are likely to be of endogenous origin. Traditionally, fasting plasma lipids have been used in studies of cardiovascular diseases. It is, however, evident that post-prandial lipid changes are also associated with an increased risk of atherosclerotic and prothrombotic diseases 28,29. The post-prandial increase in triglyceriderich lipoproteins are not only due to intestinal chylomicrons but also to accumulation of endogenous very low density lipoproteins 25. This seems mainly to be caused by a competition between very low density lipoproteins and chylomicrons for the lipoprotein lipase sites resulting in reduced degradation rate of very low density lipoproteins 25,28. As already suggested, the elevated triglycerides found among many women with early onset pre-eclampsia may be due to a combination of higher endogenous very low density lipoproteins and increased chylomicronbound triglycerides resulting from a competition for the binding sites at lipoprotein lipase. Alternatively, or in addition, the activity of lipoprotein lipase may be lower in women with early onset pre-eclampsia. The increased ratio of ApoB-100 to non-high density lipoproteins cholesterol may be tentatively explained by enhanced lipolysis of lighter low density lipoproteins fractions by hepatic lipase, which seems to have higher activity in pre-eclampsia 15. In this case the formation of small dense low density lipoproteins with a higher ApoB: cholesterol ratio would expectedly increase 29. Decreased
conversion of very low density lipoproteins to low density lipoproteins for the reasons discussed above may also contribute to the increased ratio of ApoB-100: non-high density lipoproteins cholesterol. We have recently reported that women with a high intake of energy and sucrose are at risk of developing early onset pre-eclampsia 30. This observation is in accordance with the findings in the present study. High intake of energy and sucrose is associated with increased liver production of very low density lipoproteins, which is one source of increased plasma triglycerides 31. There are several mechanisms by which hypertriglyceridemic dyslipidemia may contribute to pre-eclampsia. It is well established that dyslipidemic changes in plasma lipid in general may induce endothelial disturbances 32–35. In pregnancy a dyslipidemic lipoprotein profile may add to other factors that contribute to the endothelial dysfunction found in women with pre-eclampsia. Hypertriglyceridemic dyslipidemia could also augment development of atherotic changes in the spiral arteries as well as promote prohrombotic mechanisms 33,36. Finally, lipids or lipid derivates may interfere with trophoblast invasion 37. The present study shows that a hypertriglyceridemic dyslipidemic pattern of non-fasting plasma lipids at 18 weeks of gestation is associated with the increased risk of developing early but not late onset pre-eclampsia. These observations give further support to previous reports that abnormalities in plasma lipids or lipoproteins contribute to the development of pre-eclampsia. The finding that hypertriglyceridemic dyslipidemia is associated only with early onset pre-eclampsia underscores the notion that early and late onset disease may not share all risk factors as well as being pathophysiologically different 5–7,38. Finally, the increasing prevalence of obesity and the associated dyslipidemia may counteract efforts to reduce early onset pre-eclampsia 39.
Acknowledgements The authors would like to thank Mrs K. Høimyr and Mrs V. Volden for their excellent technical assistance. This work was supported by Johan Throne Holst’s Fond, Helse og Rehabilitering and Freia Fond.
References 1. Walker JJ. Pre-eclampsia. Lancet 2000;356:1260–1265. 2. Ness RB, Roberts JM. Heterogenous causes constituting the single syndrome of preeclampsia: a hypothesis and its implications. Am J Obstet Gynecol 1996;175:1365–1370. 3. Redman CW, Sacks GP, Sargent IL. Preeclampsia: an excessive maternal inflammatory response to pregnancy. Am J Obstet Gynecol 1999;180:499–506. 4. Baker PN, Cunningham FG. Platelet and coagulation abnormalites. In: Lindheimer MD, Roberts JM, Cunningham FG, editors. Chesley’s
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
DYSLIPIDEMIA AND EARLY ONSET PRE-ECLAMPSIA 1087
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
Hypertensive Disorders in Pregnancy, Stamford: Appleton & Lange, 1999:349–373. Sibai BM, Mercer B, Sarinoglu C. Severe preeclampsia in second trimester: recurrence risk and long term prognosis. Am J Obstet Gynecol 1991;165:1408–1412. Ferrazzani S, Caruso A, De Carolis S, Martino IV M, ancuso S. Proteinuria and outcome of 444 pregnancies complicated by hypertension. Am J Obstet Gynecol 1990;162:366–371. Sullivan CA, Magann EF, Perry Jr KG, Roberts WE, Blake PG, Marin Jr JN. The recurrence risk of the syndrome of hemolysis, elevated liver enzymes, and low platelets (HELLP) in subsequent gestations. Am J Obstet Gynecol 1994;171:940–943. Sibai BM, Spinnato JA, Watson DL, Hill GA, Anderson GD. Pregnancy outcome in 303 cases with severe preeclampsia. Obstet Gynecol 1984;64:319–325. Potter JM, Nestel PJ. The hyperlipidemia of pregnancy in normal and complicated pregnancies. Am J Obstet Gynecol 1979;133:165–170. Fahraeus L, Larsson-Cohn U, Wallentin L. Plasma lipoproteins including high density lipoprotein subfractions during pregnancy. Obstet Gynecol 1985;66:468–472. Silliman K, Shore V, Forte TM. Hypertriglycerideemia during pregnancy is associated with the formation of small dense low-density lipoprotein and the presence of large buoyant high-density lipoprotein. Metabolism 1994;43:1035–1041. Sattar N, Greer IA, Louden J, et al. Lipoprotein subfraction changes in normal pregnancy: threshold effect of plasma triglyceride on appearance of small, dense low density lipoprotein. J Clin Endocrinol Metab. 1997;82:2483–2491. Rosing U, Samsioe G, Ølund A, Johansson B, Kallner A. Serum levels of apoprotein A-I. A-II and high density lipoproteins-cholesterol in second half of normal pregnancy and in pregnancy complicated by pre-eclampsia. Horm Metabol Res 1989;21:376–382. Lorentzen B, Drevon CA, Endresen MJ, Henriksen T. Fatty acid pattern of esterified and free fatty acids in sera of women with normal and pre-eclamptic pregnancy. Br J Obstet Gynaecol 1995;102:530– 537. Sattar N, Bendomir A, Berry C, Shepherd J, Greer IA, Packhard CJ. Lipoprotein subfraction concentrations in preeclampsia: parallels to atherosclerosis. Obstet Gynecol 1997;89:403–408. He S, Silveira A, Hamsten A, Blomback M, Bremme K. Haemostatic, endothelial and lipoprotein parameters and blood pressure levels in women with a history of preeclampsia. Thromb Haemost 1999;81:538–542. Barden AE, Beilin LJ, Ritchie J, Walters BN, Michael C. Does a predisposition to the metabolic syndrome sensitize women that develop pre-eclampsia. J Hypertens 1999;17:1307–1315. Laivuori H, Tikkanen MJ, Ylikorkala O. Hyperinsulinemia 17 years after preeclamptic first pregnancy. J Clin Endocrinol Metab 1996;81:2908–2911. Kinnunen PJ, Unnere´ s H-A, Ranta T, Enholm C, Nikkila¨ EA, Seppa¨ la¨ M. Activities of post-heparin lipase and hepatic lipase during pregnancy and lactation. Eur J Clin Invest 1980;10:469–474. Herrera E, Go´ mez-Coronado D, Lasuncio´ n MA. Lipid metabolism in pregnancy. Biol Neonate 1987;51:70–77. Innes KE, Wimsatt JH. Pregnancy-induced hypertension and insulin resistance: evidence for a connection. Acta Obstet Gynecol Scand 1999;78:263–284.
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
22. Grundy SM. ypertriglyceridemia, atherogenic dyslipidemia and the metabolic syndrome. Am J Cardiol 1998;81:188–258. 23. Le CT, Lindgren BL. Computational implementation of the conditional logistic regression model in the analysis of epidemiologic matched studies. Comput Biomed Res 1988;21:48–52. 24. Hubel CA, McLaughlin MK, Evans RW, Hauth BA, Sims CJ, Roberts JM. Fasting serum triglycerides, free fatty acids, and malondialdehyde are increased in preeclampsia, are positively correlated, and decrease within 48 hours post partum. Am J Obstet Gynecol 1996;174:975– 982. 25. Griffin BA, Packard CJ. Metabolism of very low density lipoproteins and low density lipoproteins subclasses. Curr Opin Lipidiol 1994;5:200–206. 26. Laakso M. Insulin resistance and coronary heart disease. Curr Opin Lipidol 1996;7:217–226. 27. Evans M, Khan N, Rees A. Diabetic dyslipidaemia and coronary heart disease: new perspectives. Curr Opin Lipidol 1999;10:387–391. 28. Goldberg IJ, Kako Y, Lutz EP. Repsonses to eating: lipoproteins, lipolytic products and atherosclerosis. Curr Opin Lipidol 2000;11:235– 241. 29. Chapman MJ, Bruckert E. The atherogenic role of triglycerides and small dense low density lipoproteins: impact of ciprofibrate therapy. Atherosclerosis 1996;124:S21–S28. 30. Clausen T, Solvoll K, Slott M, Drevon CA, Vollset SE, Henriksen T. Increased intake of energy, sugar and polyunsaturated fat before 20 weeks of gestation is associated with increased risk of preeclampsia. J Soc Gynecol Investig 2000;7:S146. 31. Frayn KN, Kingman SM. Dietary sugars and lipid metabolisms in humans. Am J Clin Nutr 1995;62:S250–S261. 32. de Man FH, Weverling-Rijnsburger AW, der Laarse A, Smelt AH, Jukema JW, Blauw GJ. Not acute but chronic hypertriglyceridemia is associated with impaired endothelium-dependent vasodilation. Arterioscl Thromb Vasc Biol 2000;20:744–750. 33. Lorentzen B, Henriksen T. Plasma lipids and vascular dysfunction in preeclampsia. Sem Reprod Endocrinol 1998;16:33–39. 34. Sattar N, Petrie JR, Jaap AJ. The atherogenic lipoprotein phenotype and vascular endothelial dysfunction. Atherosclerosis 1998;138:229– 235. 35. Hubel CA. Dyslipidemia, iron, and oxiditative stress in preeclampsia: assessment of maternal and feto-placental interactions. Sem Reprod Endocrinol 1998;16:75–92. 36. Silveira A, Karpe F, Johnsson H, Bauer KA, Hamsten A. In vivo demonstration in humans that large postprandial triglyceride-rich lipoproteins activate coagulation factor VII through the intrinsic coagulation pathway. Arterioscl Throm Vasc Biol 1996;16:1333–1339. 37. Staff AC, Ranheim T, Henriksen T, Halvorsen B. 8-Iso-prostaglandin f(2alpha) reduces trophoblast invasion and matrix metalloproteinase activity. Hypertension 2000;35:1307–1313. 38. Dekker GA, de Vries JI, Doelitzsch PM, et al. Underlying disorders associated with severe early-onset preeclampsia. Am J Obstet Gynecol 1995;173:1042–1048. 39. Denke MA, Sempos CT, Grundy SM. Excess body weight. An underrecognized contributor to dyslipidemia in white American women. Arch Intern Med 1994;154:401–410. Accepted 6 June 2001
Dyslipidemia in early second trimester is mainly a feature of women with early onset pre-eclampsia Torun Clausen a,*, Srdjan Djurovic b, Tore Henriksen c Objective To investigate whether hypertriglyceridemic dyslipidemia is a risk factor for either early or late onset pre-eclampsia. Design Prospective cohort study and nested case–control study. Setting Aker Hospital: a university hospital with all levels of obstetric care. Participants 2157 Caucasian pregnant women. Methods Blood samples were obtained from non-fasting subjects at 18 weeks of gestation. All samples were analysed for triglycerides, total-cholesterol, high density lipoproteins cholesterol and non-high density lipoproteins cholesterol. ApoB-100 were analysed in pre-eclamptic women and in 3:1 matched controls. The cohort data were analysed by multiple logistic regression and the case–control data by conditional logistic regression. Main outcome measures Adjusted odds ratios of early and late onset pre-eclampsia according to early second trimester serum concentration levels of lipids and ApoB-100. Results Eighteen women developed early onset pre-eclampsia and 53 women developed late onset pre-eclampsia. In the cohort model, women with triglycerides above 2.4mmol/L had increased risk (OR 5.1; 95% CI 1.1– 23.1) of early onset pre-eclampsia compared with those with triglycerides levels # 1.5mmol/L. For women with high triglycerides: non-high density lipoproteins cholesterol ratios (.90 centile) the OR (95% CI) for early onset pre-eclampsia was 7.1 (2.3–22.0) compared with those with low ratios (# 50 centile). Similar associations were found in the case control model. We found no associations between plasma lipids and risk of late onset pre-eclampsia. Conclusions Hypertriglyceridemic dyslipidemia before 20 weeks of gestation is associated with the risk of developing early but not late onset pre-eclampsia, giving support to the contention that these two variants of the disease are at least partly pathogenically different.
INTRODUCTION Pre-eclampsia is a hypertensive complication of pregnancy associated with well-documented risks for the mother and the fetus 1. Although hypertension and proteinuria are the simple clinical criteria for diagnosis of preeclampsia the pathophysiological mechanisms that lead to the disorder are by all evidence very diverse 2. Preeclampsia develops in a particular woman following an unfortunate combination of maternal (trophoblast-independent) risk factors and an excessive maternal response to the trophoblast/trophoblast-derived factors 2,3. The maternal risk factors include disorders associated with
a
Departments of Obstetrics and Gynaecology, Aker and Ullevaal University Hospitals, Oslo, Norway b Institute for Medical Genetics and Department of Medical Genetics, Ullevaal University Hospital,Oslo, Norway c Department of Obstetrics and Gynaecology, National Hospital, Oslo, Norway * Correspondence: Dr T. Clausen, Department of Obstetrics and Gynaecology, Ullevaal University Hospital, 0407 Oslo, Norway. q RCOG 2001 British Journal of Obstetrics and Gynaecology PII: S 0306-545 6(01)00247-9
endothelial dysfunction such as chronic hypertension, diabetes, kidney disease and dyslipidemia, as well as more specific defects such as protein C or protein S deficiencies, activated protein C resistance, anticardiolipin antibodies and hyperhomocysteinemia 1,4. Early onset pre-eclampsia exhibits features that are unusual in late onset disease 5. For example, the recurrence risk of pre-eclampsia is relatively high in women with early onset disease, the proportion of women with chronic hypertension is higher, fetal growth restriction is much more common and multi-organ involvement, as reflected in the HELLP syndrome, is more frequent 5–7. Thus early and late onset pre-eclampsia seem, at least partly, to be pathophysiologically different. A further clarification of these differences is of clinical interest because early onset disease is associated with a higher prevalence of iatrogenic prematurity, intrauterine growth restriction and severe maternal disease 5,8. In normal pregnancy the circulating concentrations of triglycerides and cholesterol increases progressively 9–12. In women with pre-eclampsia the serum lipids show a shift towards a dyslipidemic profile: The serum triglycerides are further elevated, there is a higher proportion of low density lipoproteins of small sizes, high density lipowww.bjog-elsevier.com
1082 T. CLAUSEN ET AL.
protein decreased and serum free fatty acids are increased 9,13–15. The pre-eclamptic dyslipidemia is not a consequence of the disease because it is present long before the disorder becomes clinically overt and tends to be present after pregnancy 14,16–18. Because the pathogenesis of early and late onset pre-eclampsia seems to differ slightly we wanted to investigate if hypertriglyceridemic dyslipidemic patterns are risk factors for both these two variants of the disorder.
METHODS The setting of the present study was Aker University Hospital in Oslo, which until December 1999 covered defined geographical areas of Oslo City, representing all socio-economic classes. Approximately 95% of the pregnant women in these areas gave birth at this hospital. All pregnant women in the areas of Aker Hospital were offered an ultrasound scan at 17–19 weeks of gestation. In a period of 21 months in 1995–1996 Caucasian nondiabetic women were asked to provide a blood sample at the time of the scan. Blood samples from 2434 women were obtained (i.e. from 93.4% of the recruited women), and 2190 of the women who provided blood samples were delivered at Aker Hospital. After excluding women with twin pregnancies, 2157 women remained for the study. Diagnosis of pre-eclampsia required the presence of proteinuria and pregnancy-induced hypertension. Proteinuria was defined as $ 11 on dipsticks (300mg/24 h) found twice at least six hours apart. Pregnancy-induced hypertension was defined either as blood pressure $ 140/ 90mmHg (n ¼ 59) or as an increase in diastolic pressure of $ 15mmHg (n ¼ 12), compared with the average before 20 weeks of gestation. In both cases two measurements at least six hours apart were required. The preeclamptic women were divided into groups of early and late onset pre-eclampsia (i.e. delivered before and after 37 weeks of gestation, respectively). Blood samples were drawn in non-fasting state and allowed to coagulate before centrifugation at 400g for 10 minutes. The serum samples were transferred on ice to a -708C freezer within 140 minutes after venipuncture. In the whole cohort (n ¼ 2157) we measured serum concentrations of triglycerides, total cholesterol and high density lipoprotein cholesterol. Non-high density lipoproteins cholesterol and the ratio of triglycerides: high density lipoproteins cholesterol and triglycerides: non-high density lipoproteins-cholesterol were calculated. ApoB-100 was measured in the 71 women with subsequent pre-eclampsia and in 3:1 matched controls consisting of healthy pregnant women from the study population. The controls were matched for age, parity and body mass index. The rational for choosing the parameters above was based on previous observations: fasting triglycerides
but not cholesterol at 18 weeks of gestation are increased in women with subsequent pre-eclampsia indicating enhanced production and/or impaired degradation of triglyceride-rich lipoproteins 14,19,20. The selected parameters may reflect such disturbances in the metabolism of triglyceride-rich lipoproteins. Similar kinds of metabolic disturbances are seen in individuals with ‘metabolic syndrome’ 21. The typical features of the plasma lipids in these individuals include high triglycerides and low highdensity lipoproteins cholesterol. Total-cholesterol and non-high density lipoproteins cholesterol may be increased or within normal range 22. High values of the ratios of triglycerides: non-high density lipoproteins cholesterol and triglycerides: ApoB may reflect accumulation of triglyceride-rich particles. The study protocol was approved by the Regional Medical Ethics Committee and written informed consent was obtained from each woman. Statistical analysis For data analysis we used the statistical program 8.0 SPSS for Windows (SPSS Inc, Chicago, Illinois, USA). Proportions were compared using the x 2 test. Women with and without pre-eclampsia were compared using the Mann U Whitney test. Odds ratio for pre-eclampsia was evaluated by multiple logistic regression analyses (cohort data) and conditional logistic regression (case–control data). The odds ratios in the multiple logistic regression models are shown with adjustment for parity, maternal age, body mass index and smoking, whereas the odds ratios in the matched case control analyses are presented with adjustment for smoking. Stratified Cox regression models were used to fit the conditional logistic regression model, because these two methods give the same likelihood functions and thus the same estimates 23. Triglycerides and the ratios triglycerides: nonhigh density lipoproteins cholesterol and ApoB: nonhigh density lipoproteins cholesterol were categorised according to the 50 th and 90 th centiles of their distribution to allow examination of dose–response relationship for extreme upper categories. High density lipoproteins cholesterol and the ratio ApoB: triglycerides were categorised in the same manner according to the 50 th and 10 th centiles to allow for examination of dose–response relationship for extreme lower categories. The number of cases below the 10 th centile was, however, too small for statistical analysis and we therefore categorised those variables in two groups only: # 50 th centile and .50 th centile. RESULTS Baseline characteristics of the cohort are presented in Table 1. Eighteen women developed early onset preeclampsia whereas 53 women developed late onset q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
Table 1. Clinical characteristics of the cohort and serum lipid concentrations at 18 weeks of gestation. Values are given as median (25-75 th centile) unless otherwise shown. BMI ¼ body mass index; PE ¼ preeclampsia; SBP ¼ systolic blood pressure; DBP ¼ diastolic blood pressure; TG ¼ triglycerides; HDL ¼ high density lipoproteins; CH ¼ cholesterol.
a b
Mann-Whitney U test. x 2 test.
Without PE n ¼ 2086
PE ^ 37 weeks n ¼ 53
Pa
PE ,37 weeks n ¼ 18
Pa
30 (27-33) 22.1 (20.5-24.4) 51.2 22.7 117 (110-125) 75 (67-80) 40.0 (39.0-41.0) 3600 (3250-3930)
30 (27-33) 22.1 (20.5-24.4) 50.4 22.9 115 (110-124) 73 (67-80) 40.0 (39.0-41.0) 3605 (3270-3930)
29 (26-34) 22.8 (21.0-25.0) 69.8 19.2 140 (135-150) 95 (90-100) 39.0 (38.0-40.0) 3390 (3145-3930)
0.50 0.13 0.005 0.54 ,0.001 ,0.001 ,0.001 0.15
27 (24-30) 21.4 (19.9-23.7) 83.3 16.7 152 (145-173) 100 (95-110) 34.0 (29.25-35.0) 1900 (1060-2298)
0.02 0.46 0.005 0.78 ,0.001 ,0.001 ,0.001 ,0.001
1.50 (1.20-1.90) 5.30 (4.80-5.90) 1.75 (1.51-2.00) 2.80 (2.30-3.30) 0.86 (0.63-1.18) 0.54 (0.43-0.66)
1.50 (1.20-1.90) 5.30 (4.80-5.90) 1.75 (1.51-2.01) 2.80 (2.30-3.30) 0.86 (0.63-1.18) 0.54 (0.43-0.66)
1.50 (1.20-1.90) 5.10 (4.60-5.83) 1.72 (1.52-1.89) 2.75 (2.32-3.40) 0.88 (0.68-1.22) 0.56 (0.45-0.70)
0.81 0.63 0.24 0.98 0.35 0.46
1.85 (1.30-2.60) 4.90 (4.50-5.40) 1.65 (1.53-1.92) 2.45 (1.90-2.60) 1.11 (0.72-1.64) 0.75 (0.50-1.14)
0.14 0.02 0.29 0.003 0.125 0.008
DYSLIPIDEMIA AND EARLY ONSET PRE-ECLAMPSIA 1083
Age, (years) BMI (kg/m 2) Nulliparae, (%) b Smoking (%) b SBP at delivery (mmHg) DBP at delivery (mmHg) Gestational age (weeks) Birthweight (g) Lipid parameters TG (mmol/l) Total-CH (mmol/L) HDL-CH (mol/L) NonHDL-CH (mol/L) TG/HDL-CH TG/nonHDL-CH
Cohort n ¼ 2157
1084 T. CLAUSEN ET AL. Table 2. Clinical characteristics and serum lipid concentrations at 18 weeks of gestation in women with late ($ 37 weeks) and early (,37 weeks) onset preeclampsia (PE) and in matched controls. Values are given as median (25-75 th centile) unless otherwise shown. BMI ¼ body mass index; TG ¼ triglycerides; HDL ¼ high density lipoproteins; CH ¼ cholesterol. PE $37 weeks n ¼ 53 Age (years) BMI (kg/m) 2 Nullipara, n (%) b Smoking n (%) b Lipid parameters TG (mmol/L) Total CH (mmol/L) HDl-CH (mmol/L) NonHDL-CH (mmol/L) TG/HDL TG/nonHDL-CH ApoB-100 mg/100 mL ApoB-100/TG ApoB-100/nonHDL a b
Controls n ¼ 149
29.0 (26.0-34.0) 22.8 (21.0-25.0) 37 (69.8) 10 (19.2) 1.50 (1.20-1.90) 5.10 (4.60-5.83) 1.72 (1.52-1.89) 2.75 (2.32-3.40) 0.88 (0.68-1.22) 0.56 (0.45 (0.70) 125.0 (106.0-150.0) 84.4 (68.8-98.3) 45.4 (42.2-49.5)
29.0 (26.0-34.0) 22.6 (21.0-24.9) 102 (68.5) 37 (25.3) 1.60 (1.30-2.10) 5.30 (4.80-5.90) 1.71 (1.52-1.89) 2.90 (2.40-3.40) 0.93 (0.70-1.34) 0.57 (0.47-0.67) 134.0 (112.0-148.25) 80.0 (70.3-95.5) 45.8 (43.4-48.4)
Pa
PE ,37 weeks n ¼ 18
Controls n ¼ 51
Pa
0.97 0.93 0.86 0.37
27.0 (24.0-30.0) 21.4 (19.9-23.7) 15 (83.3) 3 (16.7)
27.0 (24.0-31.0) 21.6 (20.7-23.7) 42 (82.4) 15 (30.0)
0.95 0.55 1.00 0.27
0.24 0.35 0.78 0.57 0.65 0.77 0.45 0.51 0.62
1.85 (1.30-2.60) 4.90 (4.50-5.40) 1.65 (1.53-1.92) 2.45 (1.90-2.60) 1.11 (0.72-1.64) 0.75 (0.50-1.14) 116.5 (105.0-129.0) 66.8 (44.6-92.1) 50.4 (45.4-55.8)
1.40 (1.10-2.00) 5.30 (4.90-5.90) 1.78 (1.53-2.06) 2.80 (2.40-3.50) 0.76 (0.58-1.27) 0.50 (0.38-0.70) 130.0 (112.0-153.0) 91.6 (71.1-108.3) 45.9 (43.1-48.9)
0.15 0.03 0.24 0.01 0.12 0.01 0.02 0.02 0.02
Mann-Whitney U test. x 2 test.
well as of the ApoB: triglycerides ratio. Furthermore, the ratios triglycerides: non-high density lipoproteins cholesterol and ApoB: non-high density lipoproteinscholesterol were increased in the women who developed early onset pre-eclampsia. We then estimated the risk of developing early and late onset pre-eclampsia in relation to increasing serum levels of concentrations of triglycerides and to the ratio triglycerides: non-high density lipoproteins cholesterol using multiple logistic regression analyses (cohort data, Table 3). Women with serum triglycerides at 18 weeks of gestation above 2.4mmol/L had increased risk of developing early onset pre-eclampsia (OR 5.0, 95% CI 1.5–16.6) compared with levels of 1.5mmol/L or lower. For women with ratio triglycerides: non-high density lipoproteins cholesterol above 53, the odds ratio for developing early onset pre-eclampsia was 7.1 compared with ratios at
disease. No difference was found in early second trimester serum concentrations of lipids between women without pre-eclampsia and the ones who subsequently developed late onset disease. In the women with eventual early onset pre-eclampsia, however, the serum concentrations of total-cholesterol and non-high density lipoproteins cholesterol were significantly lower and the ratio triglycerides: non-high density lipoproteins cholesterol was higher than in the women without pre-eclampsia. Table 2 shows the baseline characteristics in the case– control part of the study. The serum lipids and ApoB concentrations of women with late onset pre-eclampsia and the matched controls were similar. Compared with controls the women with early onset disease had, however, decreased serum levels of total-cholesterol, non-high density lipoproteins cholesterol and ApoB as
Table 3. Adjusted odds ratios (ORs) for late and early onset pre-eclampsia (PE) in relation to serum lipid levels at 18 weeks of gestation. (cohort study). TG ¼ triglycerides; HDL ¼ high density lipoproteins; CH ¼ cholesterol. Late onset PE
Triglycerides (mmol/L) # 1.5 1.5-2.4 . 2.4 HDL-CH (mmol/L) # 1.75 . 1.75 TG/nonHD-CH # 46.0 46.0-53.0 . 53.0
Early onset PE a
n
n
%
OR (95% CI)
n
%
OR a (95%CI)
1148 834 175
29 19 5
2.5 2.3 2.9
1.0 0.9 (0.5-1.6) 1.3 (0.5-3.5)
8 4 6
0.7 0.5 3.4
1.0 0.7 (0.2-2.5) 5.0 (1.5-16.6)
1096 1061
31 22
2.8 2.1
1.0 0.7 (0.4-1.2)
13 5
1.2 0.5
1.0 0.4 (0.1-1.3)
1077 855 223
25 25 2
2.3 2.9 0.9
1.0 1.4 (0.8-2.5) 0.5 (0.1-2.0)
6 4 8
0.6 0.5 3.6
1.0 1.0 (0.3-3.4) 7.1 (2.3-22.0)
a
Adjusted for age, body mass index, nulliparous (yes/no) and smoking (yes/no). 226 of the women in the cohort, including one with early onset and two with late onset pre-eclampsia, were excluded in the analyses because of missing background data. The individuals with missing data were distributed proportionally in the categories of plasma lipids.
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
DYSLIPIDEMIA AND EARLY ONSET PRE-ECLAMPSIA 1085
46 or lower. These risks were not found in the pregnancies with late onset disease (Table 3). In the case–control study using conditional logistic regression analyses similar associations between high serum triglycerides values or the ratio triglycerides: nonhigh density lipoproteins cholesterol and the risk of developing early onset pre-eclampsia were found (Table 4). In the group with late onset disease no similar associations were observed. In this study we could also estimate the associations between the ratios ApoB: non-high density lipoproteins cholesterol and ApoB: triglycerides and the risk of subsequent pre-eclampsia. The former ratio was positively associated with the risk of developing early onset disease with odds ratio of 5.1 if the ratio was above 0.54 compared with 0.20 or lower. This association was not found in the group with late onset pre-eclampsia.
DISCUSSION It is well documented that in all pregnancies plasma lipid levels change towards a hyperlipidemic lipid profile. Although total cholesterol and high density lipoproteins cholesterol are increased, hypertriglyceridemia is the most prominent characteristic 9,12. All lipoprotein fractions show increased content of triglycerides 9,10. Increased synthesis of triglyceride-rich lipoproteins as well as decreased degradation (lipolysis) seems to contribute to the gestational hyperlipidemia 9,19,20. In women with overt pre-eclampsia the hypertriglyceridemia is even more pronounced 9,14,24. The pre-eclamptic dyslipidemia-like pattern of plasma lipids is apparently not a simple exaggeration of the physiological changes
because there is usually no further increase in total cholesterol and high density lipoproteins compared with normal pregnancies 9. In fact, in women with overt preeclampsia, plasma high density lipoproteins cholesterol and apoprotein A-I are usually decreased 9,13. In the present study we found that dyslipidemia-like changes in the plasma lipids at 18 weeks of gestation are associated with early but not late onset pre-eclampsia. Because of the prospective design of the study we can conclude that the observed plasma lipid pattern was not a late consequence of the pathophysiological changes in preeclamptic women. The plasma lipid pattern associated with early onset pre-eclampsia was found in the cohort as well as in the case–control studies. Using non-fasting blood samples as in the present study leads to greater variation especially in the triglycerides simply because of intestinally derived triglyceriderich lipoproteins (chylomicrons) 25. Despite this, early and late onset pre-eclampsia could clearly be distinguished already at 18 weeks of pregnancy. In previous studies of plasma lipids of pre-eclamptic women, distinction between early and late onset disease has not been made. As demonstrated in the present study such a distinction had major impact on the results and provides support to the notion that early and late onset pre-eclampsia may differ pathophysiologically. There is no general agreement with respect to the definition of early and late onset pre-eclampsia. We defined a woman as having early onset disease if she was delivered because of pre-eclampsia before 37 weeks of gestation. It may be argued that the use of 34 weeks of gestation would be more in accordance with current obstetric practices. The sample size of pre-eclamptic pregnancies was
Table 4. Adjusted odds ratios (OR) of late (PE $37 weeks) and early onset pre-eclampsia (PE ,37 weeks) in relation to lipid and apoB-100 levels at 18 weeks of gestation (nested case–control study). Values are given as median (25-75 th centile) unless otherwise shown. HDL ¼ high density lipoproteins; CH ¼ cholesterol; TG ¼ triglycerides; PE ¼ pre- eclampsia.
Triglycerides (mmol/L) # 1.5 1.5-2.4 . 2.4 HDL-CH (mmol/L) # 1.75 . 1.75 TG/nonHDL-CH # 46.0 46.0-53.0 . 53.0 ApoB/TG # 81.5 . 81.5 ApoB/nonHDL-CH # 0.20 0.20-0.54 . 0.54 a
Control n
PE $37weeks
OR a (95% CI)
Control n
PE ,37weeks
OR a (95%CI)
68 65 17
29 19 5
1.0 0.8 (0.4-1.6) 0.7 (0.2-2.2)
30 16 5
8 4 6
1.0 1.0 (0.2-5.1) 6.9 (1.1-41.9)
85 64
31 22
1.0 0.8 (0.4-1.6)
23 28
13 5
1.0 0.3 (0.1-1.0)
61 74 14
25 25 2
1.0 0.8 (0.4-1.5) 0.3 (0.1-1.6)
29 15 7
6 4 8
1.0 1.8 (0.4-8.5) 10.3 (1.5-69.5)
81 68
24 27
1.0 1.3 (0.7-2.6)
18 32
11 7
1.0 0.2 (0.04-0.9)
79 59 11
27 11 2
1.0 1.0 (0.5-2.2) 0.8 (0.2-4.4)
25 20 5
5 7 6
1.0 1.8 (0.5-6.7) 4.8 (1.0-22.5)
Adjusted for smoking (yes/no).
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
1086 T. CLAUSEN ET AL.
too small in the present study to allow categorisation before and after 34 weeks. The lipid changes that were associated with increased risk of early onset pre-eclampsia were characterised by high plasma triglycerides in absolute terms and relative to non-high density lipoproteins-cholesterol and to ApoB100 and by increased ratio of ApoB-100: non-high density lipoproteins cholesterol. We cannot exclude the fact that the special lipid patterns found in many of those with early onset pre-eclampsia developed during the first 18 weeks of pregnancy. However, the increasing evidence that pre-eclampsia is associated with athersclerotic cardiovascular disease later in life indicates that these women have changes in their lipid, glucose or energy metabolism that are independent of pregnancy. Insulin resistance may be involved because increased prevalence is found in pre-eclamptic women and because the lipid patterns presently found in women with early onset pre-eclampsia bear some similarities to those described in non-pregnant women with non-insulin dependent diabetes 21,26,27. The differences in the plasma triglycerides between the women with and without subsequent pre-eclampsia could either be due to higher endogenous (liver-produced) triglyceride-rich lipoproteins (ApoB-100 containing lipoproteins), to the intestinal B-48 dominated lipoproteins or to both. We have reported previously that fasting triglycerides are increased at 18 weeks of gestation in women with eventual pre-eclampsia 14. Thus at least parts of elevated plasma triglycerides are likely to be of endogenous origin. Traditionally, fasting plasma lipids have been used in studies of cardiovascular diseases. It is, however, evident that post-prandial lipid changes are also associated with an increased risk of atherosclerotic and prothrombotic diseases 28,29. The post-prandial increase in triglyceriderich lipoproteins are not only due to intestinal chylomicrons but also to accumulation of endogenous very low density lipoproteins 25. This seems mainly to be caused by a competition between very low density lipoproteins and chylomicrons for the lipoprotein lipase sites resulting in reduced degradation rate of very low density lipoproteins 25,28. As already suggested, the elevated triglycerides found among many women with early onset pre-eclampsia may be due to a combination of higher endogenous very low density lipoproteins and increased chylomicronbound triglycerides resulting from a competition for the binding sites at lipoprotein lipase. Alternatively, or in addition, the activity of lipoprotein lipase may be lower in women with early onset pre-eclampsia. The increased ratio of ApoB-100 to non-high density lipoproteins cholesterol may be tentatively explained by enhanced lipolysis of lighter low density lipoproteins fractions by hepatic lipase, which seems to have higher activity in pre-eclampsia 15. In this case the formation of small dense low density lipoproteins with a higher ApoB: cholesterol ratio would expectedly increase 29. Decreased
conversion of very low density lipoproteins to low density lipoproteins for the reasons discussed above may also contribute to the increased ratio of ApoB-100: non-high density lipoproteins cholesterol. We have recently reported that women with a high intake of energy and sucrose are at risk of developing early onset pre-eclampsia 30. This observation is in accordance with the findings in the present study. High intake of energy and sucrose is associated with increased liver production of very low density lipoproteins, which is one source of increased plasma triglycerides 31. There are several mechanisms by which hypertriglyceridemic dyslipidemia may contribute to pre-eclampsia. It is well established that dyslipidemic changes in plasma lipid in general may induce endothelial disturbances 32–35. In pregnancy a dyslipidemic lipoprotein profile may add to other factors that contribute to the endothelial dysfunction found in women with pre-eclampsia. Hypertriglyceridemic dyslipidemia could also augment development of atherotic changes in the spiral arteries as well as promote prohrombotic mechanisms 33,36. Finally, lipids or lipid derivates may interfere with trophoblast invasion 37. The present study shows that a hypertriglyceridemic dyslipidemic pattern of non-fasting plasma lipids at 18 weeks of gestation is associated with the increased risk of developing early but not late onset pre-eclampsia. These observations give further support to previous reports that abnormalities in plasma lipids or lipoproteins contribute to the development of pre-eclampsia. The finding that hypertriglyceridemic dyslipidemia is associated only with early onset pre-eclampsia underscores the notion that early and late onset disease may not share all risk factors as well as being pathophysiologically different 5–7,38. Finally, the increasing prevalence of obesity and the associated dyslipidemia may counteract efforts to reduce early onset pre-eclampsia 39.
Acknowledgements The authors would like to thank Mrs K. Høimyr and Mrs V. Volden for their excellent technical assistance. This work was supported by Johan Throne Holst’s Fond, Helse og Rehabilitering and Freia Fond.
References 1. Walker JJ. Pre-eclampsia. Lancet 2000;356:1260–1265. 2. Ness RB, Roberts JM. Heterogenous causes constituting the single syndrome of preeclampsia: a hypothesis and its implications. Am J Obstet Gynecol 1996;175:1365–1370. 3. Redman CW, Sacks GP, Sargent IL. Preeclampsia: an excessive maternal inflammatory response to pregnancy. Am J Obstet Gynecol 1999;180:499–506. 4. Baker PN, Cunningham FG. Platelet and coagulation abnormalites. In: Lindheimer MD, Roberts JM, Cunningham FG, editors. Chesley’s
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
DYSLIPIDEMIA AND EARLY ONSET PRE-ECLAMPSIA 1087
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
Hypertensive Disorders in Pregnancy, Stamford: Appleton & Lange, 1999:349–373. Sibai BM, Mercer B, Sarinoglu C. Severe preeclampsia in second trimester: recurrence risk and long term prognosis. Am J Obstet Gynecol 1991;165:1408–1412. Ferrazzani S, Caruso A, De Carolis S, Martino IV M, ancuso S. Proteinuria and outcome of 444 pregnancies complicated by hypertension. Am J Obstet Gynecol 1990;162:366–371. Sullivan CA, Magann EF, Perry Jr KG, Roberts WE, Blake PG, Marin Jr JN. The recurrence risk of the syndrome of hemolysis, elevated liver enzymes, and low platelets (HELLP) in subsequent gestations. Am J Obstet Gynecol 1994;171:940–943. Sibai BM, Spinnato JA, Watson DL, Hill GA, Anderson GD. Pregnancy outcome in 303 cases with severe preeclampsia. Obstet Gynecol 1984;64:319–325. Potter JM, Nestel PJ. The hyperlipidemia of pregnancy in normal and complicated pregnancies. Am J Obstet Gynecol 1979;133:165–170. Fahraeus L, Larsson-Cohn U, Wallentin L. Plasma lipoproteins including high density lipoprotein subfractions during pregnancy. Obstet Gynecol 1985;66:468–472. Silliman K, Shore V, Forte TM. Hypertriglycerideemia during pregnancy is associated with the formation of small dense low-density lipoprotein and the presence of large buoyant high-density lipoprotein. Metabolism 1994;43:1035–1041. Sattar N, Greer IA, Louden J, et al. Lipoprotein subfraction changes in normal pregnancy: threshold effect of plasma triglyceride on appearance of small, dense low density lipoprotein. J Clin Endocrinol Metab. 1997;82:2483–2491. Rosing U, Samsioe G, Ølund A, Johansson B, Kallner A. Serum levels of apoprotein A-I. A-II and high density lipoproteins-cholesterol in second half of normal pregnancy and in pregnancy complicated by pre-eclampsia. Horm Metabol Res 1989;21:376–382. Lorentzen B, Drevon CA, Endresen MJ, Henriksen T. Fatty acid pattern of esterified and free fatty acids in sera of women with normal and pre-eclamptic pregnancy. Br J Obstet Gynaecol 1995;102:530– 537. Sattar N, Bendomir A, Berry C, Shepherd J, Greer IA, Packhard CJ. Lipoprotein subfraction concentrations in preeclampsia: parallels to atherosclerosis. Obstet Gynecol 1997;89:403–408. He S, Silveira A, Hamsten A, Blomback M, Bremme K. Haemostatic, endothelial and lipoprotein parameters and blood pressure levels in women with a history of preeclampsia. Thromb Haemost 1999;81:538–542. Barden AE, Beilin LJ, Ritchie J, Walters BN, Michael C. Does a predisposition to the metabolic syndrome sensitize women that develop pre-eclampsia. J Hypertens 1999;17:1307–1315. Laivuori H, Tikkanen MJ, Ylikorkala O. Hyperinsulinemia 17 years after preeclamptic first pregnancy. J Clin Endocrinol Metab 1996;81:2908–2911. Kinnunen PJ, Unnere´ s H-A, Ranta T, Enholm C, Nikkila¨ EA, Seppa¨ la¨ M. Activities of post-heparin lipase and hepatic lipase during pregnancy and lactation. Eur J Clin Invest 1980;10:469–474. Herrera E, Go´ mez-Coronado D, Lasuncio´ n MA. Lipid metabolism in pregnancy. Biol Neonate 1987;51:70–77. Innes KE, Wimsatt JH. Pregnancy-induced hypertension and insulin resistance: evidence for a connection. Acta Obstet Gynecol Scand 1999;78:263–284.
q RCOG 2001 Br J Obstet Gynaecol 108, pp. 1081–1087
22. Grundy SM. ypertriglyceridemia, atherogenic dyslipidemia and the metabolic syndrome. Am J Cardiol 1998;81:188–258. 23. Le CT, Lindgren BL. Computational implementation of the conditional logistic regression model in the analysis of epidemiologic matched studies. Comput Biomed Res 1988;21:48–52. 24. Hubel CA, McLaughlin MK, Evans RW, Hauth BA, Sims CJ, Roberts JM. Fasting serum triglycerides, free fatty acids, and malondialdehyde are increased in preeclampsia, are positively correlated, and decrease within 48 hours post partum. Am J Obstet Gynecol 1996;174:975– 982. 25. Griffin BA, Packard CJ. Metabolism of very low density lipoproteins and low density lipoproteins subclasses. Curr Opin Lipidiol 1994;5:200–206. 26. Laakso M. Insulin resistance and coronary heart disease. Curr Opin Lipidol 1996;7:217–226. 27. Evans M, Khan N, Rees A. Diabetic dyslipidaemia and coronary heart disease: new perspectives. Curr Opin Lipidol 1999;10:387–391. 28. Goldberg IJ, Kako Y, Lutz EP. Repsonses to eating: lipoproteins, lipolytic products and atherosclerosis. Curr Opin Lipidol 2000;11:235– 241. 29. Chapman MJ, Bruckert E. The atherogenic role of triglycerides and small dense low density lipoproteins: impact of ciprofibrate therapy. Atherosclerosis 1996;124:S21–S28. 30. Clausen T, Solvoll K, Slott M, Drevon CA, Vollset SE, Henriksen T. Increased intake of energy, sugar and polyunsaturated fat before 20 weeks of gestation is associated with increased risk of preeclampsia. J Soc Gynecol Investig 2000;7:S146. 31. Frayn KN, Kingman SM. Dietary sugars and lipid metabolisms in humans. Am J Clin Nutr 1995;62:S250–S261. 32. de Man FH, Weverling-Rijnsburger AW, der Laarse A, Smelt AH, Jukema JW, Blauw GJ. Not acute but chronic hypertriglyceridemia is associated with impaired endothelium-dependent vasodilation. Arterioscl Thromb Vasc Biol 2000;20:744–750. 33. Lorentzen B, Henriksen T. Plasma lipids and vascular dysfunction in preeclampsia. Sem Reprod Endocrinol 1998;16:33–39. 34. Sattar N, Petrie JR, Jaap AJ. The atherogenic lipoprotein phenotype and vascular endothelial dysfunction. Atherosclerosis 1998;138:229– 235. 35. Hubel CA. Dyslipidemia, iron, and oxiditative stress in preeclampsia: assessment of maternal and feto-placental interactions. Sem Reprod Endocrinol 1998;16:75–92. 36. Silveira A, Karpe F, Johnsson H, Bauer KA, Hamsten A. In vivo demonstration in humans that large postprandial triglyceride-rich lipoproteins activate coagulation factor VII through the intrinsic coagulation pathway. Arterioscl Throm Vasc Biol 1996;16:1333–1339. 37. Staff AC, Ranheim T, Henriksen T, Halvorsen B. 8-Iso-prostaglandin f(2alpha) reduces trophoblast invasion and matrix metalloproteinase activity. Hypertension 2000;35:1307–1313. 38. Dekker GA, de Vries JI, Doelitzsch PM, et al. Underlying disorders associated with severe early-onset preeclampsia. Am J Obstet Gynecol 1995;173:1042–1048. 39. Denke MA, Sempos CT, Grundy SM. Excess body weight. An underrecognized contributor to dyslipidemia in white American women. Arch Intern Med 1994;154:401–410. Accepted 6 June 2001