Preeclampsia is a common, yet incompletely understood, complication of pregnancy. Its clinical features, including hypertension, proteinuria, and varying degrees of ischemic end-organ damage, are caused by widespread endothelial dysfunction.1 The etiology of endothelial dysfunction in preeclampsia is not known, but it has been postulated to be part of an exaggerated maternal inflammatory response to pregnancy.2
Endothelial dysfunction is also associated with atherosclerosis, an inflammatory disease that has several other features in common with preeclampsia.1,3 Epidemiologic risk factors for preeclampsia, such as obesity, diabetes, and hypertension, are also important risk factors for atherosclerosis.4,5 Metabolically, both preeclampsia and atherosclerosis are associated with insulin resistance, dyslipidemia, and hypercoagulability.4,6 Furthermore, placentas from pregnancies complicated by preeclampsia show atherosis of spiral arterioles, an atherosclerotic-like lesion characterized by foam cell invasion and intravascular fibrin deposition.1
C-reactive protein is a sensitive index of systemic inflammation that predicts adverse atherosclerotic events, including myocardial infarction, stroke, peripheral vascular disease, and death.7–12 Elevated C-reactive protein levels are correlated with obesity, an association that might explain part of the excess cardiovascular risk linked to obesity.13–15 Although systemic inflammation has been implicated in the pathogenesis of preeclampsia, whether elevated C-reactive protein levels, measured early in pregnancy, are associated with the subsequent development of preeclampsia is not known. It is also not known whether C-reactive protein and obesity are similarly associated in pregnancy, and if so, whether inflammation might be one pathway through which obesity predisposes to preeclampsia.
MATERIALS AND METHODS
The Massachusetts General Hospital Obstetrical Maternal Study (MOMS) cohort was developed in 1998 to study prospectively early-gestation risk factors for hypertensive disorders of pregnancy. Women who receive first-trimester prenatal care at Massachusetts General Hospital and affiliated health centers are eligible for inclusion in the cohort. The Massachusetts General Hospital obstetrics service provides community-based obstetric care for women from the metropolitan Boston area as well as high-risk obstetric care for women referred from throughout New England. Seventy percent of eligible women agree to enroll in the cohort, and study participants have characteristics similar to those of nonparticipants. The cohort represents a population of women from varied ethnic and socioeconomic backgrounds.
After providing written informed consent, first-trimester serum samples collected from eligible women were frozen at −80C for future analysis. The electronic medical record, which is the medical record used by the clinical staff, provides clinical and demographic data that detail the events of pregnancy through the early postpartum period. Specific information obtained from the electronic medical record included patient age, race, medical history, prior pregnancy history, smoking, height, weight, blood pressure (BP), gestational age and fetal weight at delivery, pregnancy outcome, and laboratory values. Pregnancy outcome and other variables obtained from the electronic medical record were verified using the hospital paper record and the hospital laboratory records.
For this study, nulliparous women with singleton gestations delivered after 20 weeks were eligible for inclusion. We performed a sample size calculation using our pilot data to estimate the difference in mean C-reactive protein levels between cases and controls (μ1 − μ2), the pooled standard deviation (s), and the standardized difference (μ1 − μ2/s). Based on a standardized difference of 0.56, a 1:2 case:control mix, and a two-sided type I error rate of .05, we calculated that studying 40 cases and 80 controls would yield greater than 80% power to detect a difference between cases and controls, should one exist.
Consecutive cases of preeclampsia were selected. Cases were defined by BP elevation greater than 140/90 mmHg after 20 weeks' gestation, in association with proteinuria, either 2+ or more by dipstick or greater than 300 mg per 24 hours in the absence of urinary tract infection. For each case of preeclampsia, two nulliparous controls were selected, matched by age ± 2 years and month of entry into the cohort. Women with preexisting chronic hypertension, defined as BP greater than 140/90 mmHg or need for antihypertensive medications before pregnancy or before 20 weeks' gestation, were excluded. Women with preexisting chronic renal disease and preexisting inflammatory conditions, such as connective tissue disease or cystic fibrosis, were also excluded. Finally, women with a history of diabetes or those in whom gestational diabetes developed were excluded. Our institution's human subjects committee approved the study.
Stored samples were thawed and tested for high-resolution C-reactive protein using the N latex C-reactive protein mono assay (Quest Diagnostics Inc., Cambridge, MA). Samples were stored at −80C for less than 1 year, which is within the manufacturer's range of optimal assay performance. Assays were performed on batched samples, consisting of combinations of cases and controls, by technicians blinded to case or control status. A subset of randomly selected samples was sent in duplicate for quality control. The intraassay and interassay coefficients of variation were 4.4% and 5.7%, respectively. Similar assays have been used in recent studies to predict cardiovascular disease in women.7
Continuous variables were compared using the two-sample t test or Wilcoxon rank sum test, and categoric variables were compared using the Fisher exact test. Because of the rightward skew of the C-reactive protein distributions, we compared median C-reactive protein levels among cases and controls using the Wilcoxon rank sum test. Side-to-side box plots were used to display the distribution of C-reactive protein among cases and controls. Next, all study subjects were divided into quartiles on the basis of the distribution of C-reactive protein in the controls. Odds ratios (OR) and 95% confidence intervals (CI) for developing preeclampsia were calculated, with the lowest quartile serving as the reference group. We determined whether there was a linear trend of increasing ORs among the four quartiles using the Mantel-Haenszel test.16 We used conditional logistic regression to control for potential confounding variables. Two-tailed P values < .05 were considered statistically significant. Results are reported as mean ± standard deviation. Analyses were performed with the SAS (SAS Institute, Cary, NC) and STATA (STATA Corporation, College Station, TX) statistical packages.
The baseline and delivery characteristics of all participants are presented in Table 1. Subjects were matched for age and month of entry into the cohort. There was no difference in race, gravidity, and smoking status among cases and controls. Compared with women with normal gestation, baseline body mass index (BMI) and systolic and diastolic BPs were significantly higher among women who subsequently developed preeclampsia. Women with preeclampsia delivered neonates with significantly lower birthweight and younger gestational age, and they were more likely to deliver by cesarean.
The distribution of C-reactive protein levels among the cases and controls is presented in Figure 1. First-trimester C-reactive protein levels were significantly higher among women in whom preeclampsia developed compared with women who had normal pregnancies (4.6 compared with 2.3 mg/L, P = .04). When study participants were divided into C-reactive protein quartiles, the OR of being in the highest quartile of C-reactive protein was 3.2 (95% CI 1.1, 9.3) among cases of preeclampsia compared with controls (P for trend = .02).
The association between C-reactive protein and risk of preeclampsia was then examined in more detail. Body mass index, systolic and diastolic BP, and smoking were included in a multivariable model together with quartiles of C-reactive protein. In this full model, the highest quartile of C-reactive protein was no longer associated with an increased risk of developing preeclampsia (OR 1.1, 95% CI .3, 4.3, P = .94). This loss of association was explained almost entirely by the addition of BMI to the model. In the full model, only BMI (OR 1.2 per unit increase, 95% CI 1.1, 1.5, P < .01) and systolic BP (OR 1.1 per unit increase, 95% CI 1.0, 1.1, P = .04) were significantly associated with preeclampsia. When the same multivariable model was reexamined without BMI, the highest quartile of C-reactive protein remained significantly associated with preeclampsia (OR 3.5, 95% CI 1.3, 9.5, P = .01). Body mass index had the strongest correlation with C-reactive protein (r = 0.4, P < .01).
In this prospective study of nulliparous women we found a significant association between elevated first-trimester C-reactive protein levels and subsequent development of preeclampsia. After adjusting for BMI, however, the association between C-reactive protein and preeclampsia was mitigated. This latter finding suggests a potential association between BMI and markers of inflammation that has not been studied in detail in pregnancy.
C-reactive protein is an acute-phase reactant produced by the liver in response to the proinflammatory cytokines interleukin (IL)-6 and tumor necrosis factor (TNF)-α.15 Because it has a relatively short half-life, the serum C-reactive protein level is dependent almost entirely on the rate of hepatic synthesis; therefore, it is a sensitive index of systemic inflammation.17 Until recently, C-reactive protein was used clinically to monitor disease activity and response to treatment in patients with a variety of inflammatory diseases. With the advent of reproducible, high-resolution C-reactive protein assays, subtle differences in C-reactive protein levels, well within the reported normal range, can be quantified accurately and used to detect inflammation before clinical consequences become evident.
Recent prospective studies that used high-resolution assays found that elevated C-reactive protein levels were associated with atherosclerotic events, such as myocardial infarction, need for coronary revascularization, stroke, peripheral vascular disease, and death, in men and women with established atherosclerosis as well as among healthy men and women.7–12 Additionally, C-reactive protein testing improves the capacity of serum lipid screening to predict risk for future cardiovascular events in apparently healthy, middle-aged women.7 In pregnancy, C-reactive protein has been studied primarily as a potential early marker for chorioamnionitis in women with premature rupture of membranes and as a predictor of outcomes in preterm labor, including the likelihood of successful tocolysis.18–20 Additional studies have attempted to define normal reference ranges for C-reactive protein at various stages of normal gestation, delivery, and the puerperium.21–23 Most of those studies used qualitative assays for C-reactive protein, and in those that used quantitative assays, the lower thresholds for C-reactive protein detection were significantly higher than in currently available high-resolution assays. C-reactive protein has not been examined in detail in preeclampsia.
Preeclampsia is a disease of endothelial dysfunction, but the cause of the endothelial dysfunction is not known.1 Some authorities believe a placental factor that is secreted in response to placental ischemia induced by incomplete trophoblast invasion and uterine artery remodeling is toxic to endothelial cells.24 Others suggested that a systemic, maternal inflammatory response to pregnancy is responsible for the endothelial dysfunction.2 In either case, endothelial injury leads to abnormal vasomotor regulation, increased vascular permeability, and thrombosis, the foundations of the clinical and pathologic picture of preeclampsia.1,25 Abnormal vasomotor tone with inappropriate vasoconstriction causes hypertension and placental ischemia. Increased vascular permeability through injured endothelial cells leads to edema, and in the case of renal glomerular capillaries, proteinuria. Reduced expression of endogenous anticoagulants by activated endothelial cells renders them more susceptible to microthrombosis, resulting in systemic and placental infarcts.1
The association between first-trimester C-reactive protein levels and subsequent preeclampsia identified in our univariable analysis supports the hypothesis that systemic inflammation is involved in the pathogenesis of preeclampsia.2 In the multivariable model, the finding that BMI mitigated the association suggests that C-reactive protein is not associated with preeclampsia, or that BMI and C-reactive protein might share a common pathway linking obesity to preeclampsia.
Body mass index is a validated, independent risk factor for preeclampsia.5 To support the hypothesis that BMI and C-reactive protein share a common pathway to preeclampsia, BMI must also be linked to C-reactive protein. Indeed, BMI and C-reactive protein consistently have been positively correlated across various cardiovascular outcome studies.11,26 Furthermore, epidemiologic studies designed specifically to test the association between BMI and C-reactive protein have shown a graded increase in C-reactive protein with increasing BMI.13,14 From a physiologic standpoint, preeclampsia is characterized by elevated circulating levels of TNF-α and IL-6.24,27 In a recent report, investigators showed that placental TNF-α and IL-6 expression was not significantly different between women with preeclampsia and controls, suggesting that sources other than the placenta contribute to the elevated circulating levels of these cytokines.28 Maternal adipose tissue might be one such source. Adipose cells are a major source of basal IL-6 and TNF-α secretion, which are the principal determinants of hepatic C-reactive protein production.29 Therefore, there is evidence that C-reactive protein might be an intermediary in the pathway between BMI and preeclampsia; if so, controlling for both in the multivariable analysis might be inappropriate.16
Obesity is a validated risk factor for preeclampsia, but the mechanism of how it imparts increased risk is not completely understood.5 Obesity might act through its association with insulin resistance, a syndrome of metabolic derangement characterized by hyperinsulinemia, hyperlipidemia, hypertension, and endothelial dysfunction.4,6 The results of this study suggest that it also might act through an inflammatory mechanism. It has been proposed that subclinical inflammation, marked by elevated C-reactive protein levels, is another distinct feature of the insulin resistance syndrome, a feature that could account for the increased cardiovascular risk associated with obesity.15,30,31
The strength of our study is in its prospective design. Whereas cross-sectional studies are limited in their ability to infer causation, in this prospective analysis, evidence of systemic inflammation existed many weeks before the clinical syndrome manifested. Consequently, the conclusion that inflammation contributes to the pathogenesis of preeclampsia is more plausible. Furthermore, cross-sectional studies that examined inflammatory markers in preeclampsia did not examine their association with BMI.24,27
This study also has certain limitations. First, the estimates of effect size are limited by wide confidence intervals, a function of sample size. Second, this study is limited by its inability to conclude whether elevation of C-reactive protein levels occurred after conception or whether the elevated levels existed at baseline. As a result it is impossible to determine whether systemic inflammation was induced by specific pregnancy-related factors or by factors that predated pregnancy. That C-reactive protein levels among women with uncomplicated pregnancies were similar to those seen in other studies of healthy, nonpregnant women supports the latter possibility.7,13 Another important limitation of this study is that C-reactive protein was the only inflammatory marker sampled, and it was assayed only once during pregnancy. Therefore, we do not know how C-reactive protein interacts with other inflammatory markers, nor do we know how C-reactive protein levels change during the course of pregnancy in women with and without preeclampsia. Furthermore, we can not exclude the possibility that other non-inflammatory factors caused the elevation of first-trimester C-reactive protein levels. Further studies involving larger samples of women, in which C-reactive protein and other inflammatory markers are assayed at multiple time points, are needed to support our findings.
1. Roberts JM. Endothelial dysfunction in preeclampsia. Semin Reprod Endocrinol 1998;16:5–15.
2. Redman CW, Sacks GP, Sargent IL. Preeclampsia: An excessive maternal inflammatory response to pregnancy. Am J Obstet Gynecol 1999;180:499–506.
3. Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med 1999;340:115–26.
4. Thadhani R, Stampfer MJ, Hunter DJ, Manson JE, Solomon CG, Curhan GC. High body mass index and hypercholesterolemia: Risk of hypertensive disorders of pregnancy. Obstet Gynecol 1999;94:543–50.
5. Sibai BM, Ewell M, Levine RJ, Klebanoff MA, Esterlitz J, Catalano PM, et al. Risk factors associated with preeclampsia in healthy nulliparous women. The Calcium for Preeclampsia Prevention (CPEP) Study Group. Am J Obstet Gynecol 1997;177:1003–10.
6. Kaaja R. Insulin resistance syndrome in preeclampsia. Semin Reprod Endocrinol 1998;16:41–6.
7. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000;342:836–43.
8. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men [erratum appears in N Engl J Med 1997;337:356]. N Engl J Med 1997;336:973–9.
9. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Plasma concentration of C-reactive protein and risk of developing peripheral vascular disease. Circulation 1998;97:425–8.
10. Ridker PM, Buring JE, Shih J, Matias M, Hennekens CH. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. Circulation 1998;98:731–3.
11. Koenig W, Sund M, Fröhlich M, Fischer HG, Löwel H, Döring A, et al. C-reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: Results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation 1999;99:237–42.
12. Gussekloo J, Schaap MC, Fröhlich M, Blauw GJ, Westendorp RG. C-reactive protein is a strong but nonspecific risk factor of fatal stroke in elderly persons. Arterioscler Thromb Vasc Biol 2000;20:1047–51.
13. Ford ES. Body mass index, diabetes, and C-reactive protein among U.S. adults. Diabetes Care 1999;22:1971–7.
14. Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB. Elevated C-reactive protein levels in overweight and obese adults. JAMA 1999;282:2131–5.
15. Yudkin JS, Stehouwer CD, Emeis JJ, Coppack SW. C-reactive protein in healthy subjects: Associations with obesity, insulin resistance, and endothelial dysfunction: A potential role for cytokines originating from adipose tissue? Arterioscler Thromb Vasc Biol 1999;19:972–8.
16. Rothman KJ, Greenland S. Modern epidemiology. Boston: Little, Brown, 1998:253–7.
17. Westhuyzen J, Healy H. Review: Biology and relevance of C-reactive protein in cardiovascular and renal disease. Ann Clin Lab Sci 2000;30:133–43.
18. Bek KM, Nielsen FR, Qvist I, Rasmussen PE, Tobiassen M. C-reactive protein (CRP) and pregnancy. An early indicator of chorioamnionitis. A review. Eur J Obstet Gynaecol Reprod Biol 1990;35:29–33.
19. Cammu H, Goossens A, Derde MP, Temmerman M, Foulon W, Amy JJ. C-reactive protein in preterm labour: Association with outcome of tocolysis and placental histology. Br J Obstet Gynaecol 1989;96:314–9.
20. de Villiers WJ, Louw JP, Strachan AF, Etsebeth SM, Shephard EG, de Beer FC. C-reactive protein and serum amyloid A protein in pregnancy and labour. Br J Obstet Gynaecol 1990;97:725–30.
21. Keski-Nisula L, Kirkinen P, Ollikainen M, Saarikoski S. C-reactive protein in uncomplicated parturients delivered by cesarean section. Acta Obstet Gynecol Scand 1997;76:862–7.
22. Nielsen FR, Bek KM, Rasmussen PE, Qvist I, Tobiassen M. C-reactive protein during normal pregnancy. Eur J Obstet Gynaecol Reprod Biol 1990;35:23–7.
23. Romem Y, Artal R. C-reactive protein in pregnancy and in the postpartum period. Am J Obstet Gynecol 1985;151:380–3.
24. Vince GS, Starkey PM, Austgulen R, Kwiatkowski D, Redman CW. Interleukin-6, tumour necrosis factor and soluble tumour necrosis factor receptors in women with preeclampsia. Br J Obstet Gynaecol 1995;102:20–5.
25. Thadhani R, Ecker JL, Kettyle E, Sandler L, Frigoletto FD. Pulse pressure and risk of preeclampsia: A prospective study. Obstet Gynecol 2001;97:515–20.
26. Rohde LE, Hennekens CH, Ridker PM. Survey of C-reactive protein and cardiovascular risk factors in apparently healthy men. Am J Cardiol 1999;84:1018–22.
27. Conrad KP, Miles TM, Benyo DF. Circulating levels of immunoreactive cytokines in women with preeclampsia. Am J Reprod Immunol 1998;40:102–11.
28. Benyo DF, Smarason A, Redman CW, Sims C, Conrad KP. Expression of inflammatory cytokines in placentas from women with preeclampsia. J Clin Endocrinol Metab 2001;86:2505–12.
29. Mohamed-Ali V, Pinkney JH, Coppack SW. Adipose tissue as an endocrine and paracrine organ. Int J Obes 1998;22:1145–58.
30. Festa A, D'Agostino R Jr, Howard G, Mykkanen L, Tracy RP, Haffner SM. Chronic subclinical inflammation as part of the insulin resistance syndrome: The Insulin Resistance Atherosclerosis Study (IRAS). Circulation 2000;102:42–7.
31. Pickup JC, Mattock MB, Chusney GD, Burt D. NIDDM as a disease of the innate immune system: Association of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia 1997;40:1286–92.