- Email: [email protected]
Elevated circulating concentrations of platelet activating factor in preeclampsia
Elevated circulating concentrations of platelet activating factor in preeclampsia Barbara L. Rowland, MD, Stephen T. Vermillion, MD, and William E. Roudebush, PhD Charleston, South Carolina OBJECTIVE: The aim of this study was to determine whether any association exists between preeclampsia and circulating platelet activating factor levels. STUDY DESIGN: We performed a cross-sectional observational study of circulating platelet activating factor concentrations in nonpregnant women, normotensive pregnant women in the third trimester, women with preeclampsia in the third trimester, and normotensive men. Platelet activating factor concentrations were measured with a commercially available platelet activating factor–specific radioimmunoassay (NEN Life Science Products, Inc, Boston, Mass). The primary outcome measure was the difference in mean platelet activating factor concentrations among the 4 study groups. Preeclampsia was determined according to the criteria of The American College of Obstetricians and Gynecologists. Data were analyzed with the Student t test, the χ2 test, the Fisher exact test, analysis of variance, and the Tukey test for pairwise multiple comparisons, with significance established at P < .05. RESULTS: The mean (±SD) circulating concentration of platelet activating factor was significantly higher in the group with preeclampsia (338.1 ± 26.9 ng/mL) than in either the normotensive pregnant group (217.9 ± 25.9 ng/mL; P < .05) or the nonpregnant female group (237.9 ± 20.9 ng/mL; P < .05). The 2 pregnant groups were similar with respect to selected demographic characteristics and gestational age at time of collection. There were no significant differences in the mean platelet activating factor concentrations between the group with preeclampsia and the normotensive male group or between the normotensive pregnant female group and the nonpregnant female group. CONCLUSION: Circulating platelet activating factor concentrations were increased in women with pregnancies complicated by preeclampsia with respect to those in normotensive pregnant women and normotensive nonpregnant women. Platelet activating factor may therefore serve as a marker for the risk of preeclampsia. (Am J Obstet Gynecol 2000;183:930-2.)
Key words: Platelet activating factor, preeclampsia
Platelet activating factor (PAF) is a phospholipid that has many biologic activities, including the ability to cause thromboxane production and platelet aggregation and destruction.1 The clinical manifestations of PAF have been hypothesized to include hypertension related to vasoconstriction from increased thromboxane production, thrombocytopenia related to platelet aggregation and destruction, and proteinuria.2 These clinical manifestations of PAF are similar to the clinical profile associated with preeclampsia. The association between preeclampsia and PAF remains controversial, because the few available studies have yielded varying results and used older and less sensitive measurement techniques.3-6 The From the Department of Obstetrics and Gynecology, Medical University of South Carolina. Presented at the Twentieth Annual Meeting of the Society for MaternalFetal Medicine, Miami Beach, Florida, January 31–February 5, 2000. Reprint requests: Barbara L. Rowland, MD, Department of Obstetrics and Gynecology, Medical University of South Carolina, 96 Jonathan Lucas St, PO Box 250619, Charleston, SC 29401-9901. Copyright © 2000 by Mosby, Inc. 0002-9378/2000 $12.00 + 0 6/6/108894 doi:10.1067/mob.2000.108894
930
objective of this study was therefore to determine with more sensitive measurement techniques whether an association exists between preeclampsia and circulating PAF concentration. Material and methods We performed a cross-sectional observational analysis of collected PAF concentrations from 4 distinctly different study populations. The studied subjects were categorized into 4 groups as follows: (1) patients with preeclampsia in the third trimester, (2) normotensive pregnant women in the third trimester, (3) normotensive nonpregnant women, and (4) normotensive men. The normotensive pregnant patients served as a control group for the preeclampsia study group, the normotensive nonpregnant women served as a control group for both pregnant groups, and the normotensive men provided the control group for the 3 female groups. The study was approved by the Institutional Review Board of the Medical University of South Carolina (Charleston). Any person 17 to 45 years of age was eligible for enroll-
Rowland, Vermillion, and Roudebush 931
Volume 183, Number 4 Am J Obstet Gynecol
Table I. Demographic characteristics of pregnant women with and without preeclampsia Odds ratio
Variable Maternal age (y, mean ± SD) Maternal race (No.) Black White Estimated gestational age (wk, mean ± SD) Multiparous (No.)
Preeclampsia group (n = 11)
Normotensive group (n = 11)
Value
95% Confidence interval
Statistical significance
27.2 ± 8.0
25.3 ± 2.6
0.77
0.09-2.53
P = .76
4 (36%) 7 (64%) 33.1 ± 5.1 5 (46%)
6 (55%) 5 (45%) 33.3 ± 5.3 8 (73%)
0.48 2.10 0.94 0.31
0.06-3.56 0.28-16.8 0.88-1.12 0.03-2.50
P = .67 P = .67 P = .97 P = .39
ment in the study. Patients included in the 2 pregnant groups were limited to those in the third trimester (28-42 weeks’ gestation). Preeclampsia (mild and severe) was determined according to criteria established by The American College of Obstetricians and Gynecologists.7 Any subjects with the diagnosis of chronic hypertension requiring antihypertensive therapy were excluded from the study, as were any subjects with recent or long-term exposure to nonsteroidal anti-inflammatory agents. After informed consent was obtained, a 3- to 5-mL sample of whole blood from each participant was collected into a serum separator tube. Samples were then transported to the storage laboratory within 15 minutes of collection to prevent degradation of PAF. These samples were then centrifuged, and the resultant supernatant was immediately stored at –80°C until batched radioimmunoassays were performed. Immediately before the radioimmunoassays were performed, the samples were thawed to room temperature. PAF concentrations were then determined by means of a previously validated commercially available PAF–specific radioimmunoassay (NEN Life Science Products, Inc, Boston, Mass). The PAF assay performance profile was as follows: sensitivity, 0.1 ng/mL; intra-assay coefficient of variation, 5.27 ± 0.15 ng/mL; and interassay coefficient of variation, 5.62 ± 0.40 ng/mL. All PAF samples from the 2 pregnant groups were obtained before the onset of labor. In addition, complete blood cell count and platelet count determinations were obtained for all participants with the diagnosis of preeclampsia. Categoric data were tested for significance with the χ2 test and the Fisher exact test. Continuous data were tested with the Student t test, analysis of variance, and the Tukey test for pairwise analysis. P < .05 for all 2-tailed tests was considered significant. Results A total of 45 patients were enrolled in the study. Each study group included 11 patients, with the exception of the nonpregnant female group, in which 12 patients were enrolled. According to a specific analysis the 2 pregnant groups were similar with respect to the demographic characteristics of maternal age, maternal race, parity, and
gestational age at time of enrollment (Table I). Analysis of variance demonstrated a significant difference in the mean PAF concentrations among the 4 groups (P = .01; Fig 1). Pairwise analyses demonstrated a significantly higher mean PAF concentration in the preeclampsia group than in either the normotensive pregnant group (P < .05) or the nonpregnant female group (P < .05; Fig 2). Specifically, the preeclampsia group had a 35% higher mean PAF concentration than the normotensive pregnant group and a 29% higher mean PAF concentration than the nonpregnant female group. All other pairwise differences were not determined to be statistically significant. Within the preeclampsia group 8 of the 11 participants (73%) had disease classified as severe. A separate analysis within the preeclampsia group was unable to demonstrate any significant differences between the patients with mild and severe preeclampsia in the mean (±SD) PAF concentration (333 ± 69 ng/mL vs 344 ± 87 ng/mL) or the mean (±SD) platelet count (142 ± 37 cells/mm3 vs 120 ± 73 cells/mm3). Comment In our study circulating concentrations of PAF were significantly elevated in women with pregnancies affected by preeclampsia with respect to those in both normotensive pregnant women and nonpregnant women. These findings are consistent with those of Benedetto et al,3 who demonstrated a decrease in serum inhibition of PAF in patients with preeclampsia. However, the measurement techniques used in that study differ from ours in that those investigators used biologic assays and indirectly measured the concentrations of PAF through inhibitory functions. Norris et al5 evaluated the effect of PAF on platelet aggregation in women with preeclampsia by means of biologic assays. They found that preeclampsia was associated with an increase in the PAF-mediated platelet aggregation. More recently, Zhang et al6 determined by means of enzyme-linked immunosorbent assays that women with pregnancies affected by pregnancyinduced hypertension had greater concentrations of PAF than did normotensive pregnant women. Our study had the advantage of measuring PAF directly and using the most sensitive method currently available, which may
932 Rowland, Vermillion, and Roudebush
Fig 1. PAF concentrations (in nanograms per milliliter) in all 4 groups. Bar heights, Mean; error bars, SD. P = .01, analysis of variance.
have provided greater accuracy than either of these previously reported studies. Our findings differ from those of the study of Maki et al.4 In that study the investigators measured the level of PAF indirectly by recording the activity of PAF acetylhydrolase, which degrades PAF. They demonstrated that PAF acetylhydrolase activity was increased in patients with preeclampsia, thus signifying decreased concentrations of PAF. We were unable to detect any significant differences between participants with severe and mild preeclampsia in either the mean PAF concentration or the mean platelet count. However, the statistical power of the subanalysis within the preeclampsia group was limited by the small sample size of the subgroup of participants with mild preeclampsia. We are currently conducting research to specifically address a potential association between PAF concentration and the severity of preeclampsia. We recognized the potential for ethnic influences on PAF concentrations, so we compared the mean PAF concentrations between African American and white participants. We were unable to detect any difference between the two racial groups. PAF is a phospholipid that is a mediator of inflammatory and allergic responses, including vasoconstriction and platelet aggregation. The similarities of these properties of PAF to the clinical manifestations of preeclampsia have led to studies of an association between PAF and preeclampsia. Previous investigations have reported varying results with respect to the association between PAF and preeclampsia.3-6 On the basis of the findings in our
October 2000 Am J Obstet Gynecol
Fig 2. Pairwise analyses of PAF concentrations (in nanograms per milliliter) in female groups. Bar heights, Mean; error bars, SD. P < .05, preeclampsia group versus normotensive pregnant group; P < .05, preeclampsia group versus normotensive nonpregnant female group.
study, however, we conclude that PAF concentrations are increased in pregnancies affected by preeclampsia. Furthermore, PAF may serve as a marker for the increased thromboxane production, with its resultant vasoconstriction and platelet aggregation, that is associated with preeclampsia. Because an association between elevated circulating concentration of PAF and preeclampsia exists, a longitudinal cohort study to evaluate serial PAF concentrations could perhaps establish PAF as an early marker for preeclampsia before the onset of clinical symptoms. REFERENCES
1. Snyder F. Platelet-activating factor and related acetylated lipids as potent biologically active cellular mediators. Am J Physiol 1990;259:C697-708. 2. Rodrigues FA, Hofmeyr GJ, Sonnendecker EW. Hypothesis: a central role of platelet activating factor in the pathogenesis of preeclampsia. Med Hypotheses 1989;4:285-6. 3. Benedetto C, Massobrio M, Bertini E, Abbondanza M, Enrieu N, Tetta C. Reduced serum inhibition of platelet activating factor activity in preeclampsia. Am J Obstet Gynecol 1989;160:100-4. 4. Maki N, Magness RR, Miyaura S, Gant NF, Johnston JM. Platelet activating factor acetylhydrolase activity in normotensive and hypertensive pregnancies. Am J Obstet Gynecol 1993;168:50-4. 5. Norris LA, Gleeson N, Sheppard BL, Bonnar J. Whole blood platelet aggregation in moderate and severe preeclampsia. Br J Obstet Gynaecol 1993;100:684-8. 6. Zhang S, Zhang Y, Liu B. [Determination of platelet activating factor and platelet associated IgG in pregnancy-induced hypertension.] Chung Hua Fu Chan Ko Tsa Chih 1996;31:296-8. 7. American College of Obstetricians and Gynecologists. Hypertension in pregnancy. Washington: The College; 1996. ACOG Technical Bulletin No.: 219.
Key words: Platelet activating factor, preeclampsia
Platelet activating factor (PAF) is a phospholipid that has many biologic activities, including the ability to cause thromboxane production and platelet aggregation and destruction.1 The clinical manifestations of PAF have been hypothesized to include hypertension related to vasoconstriction from increased thromboxane production, thrombocytopenia related to platelet aggregation and destruction, and proteinuria.2 These clinical manifestations of PAF are similar to the clinical profile associated with preeclampsia. The association between preeclampsia and PAF remains controversial, because the few available studies have yielded varying results and used older and less sensitive measurement techniques.3-6 The From the Department of Obstetrics and Gynecology, Medical University of South Carolina. Presented at the Twentieth Annual Meeting of the Society for MaternalFetal Medicine, Miami Beach, Florida, January 31–February 5, 2000. Reprint requests: Barbara L. Rowland, MD, Department of Obstetrics and Gynecology, Medical University of South Carolina, 96 Jonathan Lucas St, PO Box 250619, Charleston, SC 29401-9901. Copyright © 2000 by Mosby, Inc. 0002-9378/2000 $12.00 + 0 6/6/108894 doi:10.1067/mob.2000.108894
930
objective of this study was therefore to determine with more sensitive measurement techniques whether an association exists between preeclampsia and circulating PAF concentration. Material and methods We performed a cross-sectional observational analysis of collected PAF concentrations from 4 distinctly different study populations. The studied subjects were categorized into 4 groups as follows: (1) patients with preeclampsia in the third trimester, (2) normotensive pregnant women in the third trimester, (3) normotensive nonpregnant women, and (4) normotensive men. The normotensive pregnant patients served as a control group for the preeclampsia study group, the normotensive nonpregnant women served as a control group for both pregnant groups, and the normotensive men provided the control group for the 3 female groups. The study was approved by the Institutional Review Board of the Medical University of South Carolina (Charleston). Any person 17 to 45 years of age was eligible for enroll-
Rowland, Vermillion, and Roudebush 931
Volume 183, Number 4 Am J Obstet Gynecol
Table I. Demographic characteristics of pregnant women with and without preeclampsia Odds ratio
Variable Maternal age (y, mean ± SD) Maternal race (No.) Black White Estimated gestational age (wk, mean ± SD) Multiparous (No.)
Preeclampsia group (n = 11)
Normotensive group (n = 11)
Value
95% Confidence interval
Statistical significance
27.2 ± 8.0
25.3 ± 2.6
0.77
0.09-2.53
P = .76
4 (36%) 7 (64%) 33.1 ± 5.1 5 (46%)
6 (55%) 5 (45%) 33.3 ± 5.3 8 (73%)
0.48 2.10 0.94 0.31
0.06-3.56 0.28-16.8 0.88-1.12 0.03-2.50
P = .67 P = .67 P = .97 P = .39
ment in the study. Patients included in the 2 pregnant groups were limited to those in the third trimester (28-42 weeks’ gestation). Preeclampsia (mild and severe) was determined according to criteria established by The American College of Obstetricians and Gynecologists.7 Any subjects with the diagnosis of chronic hypertension requiring antihypertensive therapy were excluded from the study, as were any subjects with recent or long-term exposure to nonsteroidal anti-inflammatory agents. After informed consent was obtained, a 3- to 5-mL sample of whole blood from each participant was collected into a serum separator tube. Samples were then transported to the storage laboratory within 15 minutes of collection to prevent degradation of PAF. These samples were then centrifuged, and the resultant supernatant was immediately stored at –80°C until batched radioimmunoassays were performed. Immediately before the radioimmunoassays were performed, the samples were thawed to room temperature. PAF concentrations were then determined by means of a previously validated commercially available PAF–specific radioimmunoassay (NEN Life Science Products, Inc, Boston, Mass). The PAF assay performance profile was as follows: sensitivity, 0.1 ng/mL; intra-assay coefficient of variation, 5.27 ± 0.15 ng/mL; and interassay coefficient of variation, 5.62 ± 0.40 ng/mL. All PAF samples from the 2 pregnant groups were obtained before the onset of labor. In addition, complete blood cell count and platelet count determinations were obtained for all participants with the diagnosis of preeclampsia. Categoric data were tested for significance with the χ2 test and the Fisher exact test. Continuous data were tested with the Student t test, analysis of variance, and the Tukey test for pairwise analysis. P < .05 for all 2-tailed tests was considered significant. Results A total of 45 patients were enrolled in the study. Each study group included 11 patients, with the exception of the nonpregnant female group, in which 12 patients were enrolled. According to a specific analysis the 2 pregnant groups were similar with respect to the demographic characteristics of maternal age, maternal race, parity, and
gestational age at time of enrollment (Table I). Analysis of variance demonstrated a significant difference in the mean PAF concentrations among the 4 groups (P = .01; Fig 1). Pairwise analyses demonstrated a significantly higher mean PAF concentration in the preeclampsia group than in either the normotensive pregnant group (P < .05) or the nonpregnant female group (P < .05; Fig 2). Specifically, the preeclampsia group had a 35% higher mean PAF concentration than the normotensive pregnant group and a 29% higher mean PAF concentration than the nonpregnant female group. All other pairwise differences were not determined to be statistically significant. Within the preeclampsia group 8 of the 11 participants (73%) had disease classified as severe. A separate analysis within the preeclampsia group was unable to demonstrate any significant differences between the patients with mild and severe preeclampsia in the mean (±SD) PAF concentration (333 ± 69 ng/mL vs 344 ± 87 ng/mL) or the mean (±SD) platelet count (142 ± 37 cells/mm3 vs 120 ± 73 cells/mm3). Comment In our study circulating concentrations of PAF were significantly elevated in women with pregnancies affected by preeclampsia with respect to those in both normotensive pregnant women and nonpregnant women. These findings are consistent with those of Benedetto et al,3 who demonstrated a decrease in serum inhibition of PAF in patients with preeclampsia. However, the measurement techniques used in that study differ from ours in that those investigators used biologic assays and indirectly measured the concentrations of PAF through inhibitory functions. Norris et al5 evaluated the effect of PAF on platelet aggregation in women with preeclampsia by means of biologic assays. They found that preeclampsia was associated with an increase in the PAF-mediated platelet aggregation. More recently, Zhang et al6 determined by means of enzyme-linked immunosorbent assays that women with pregnancies affected by pregnancyinduced hypertension had greater concentrations of PAF than did normotensive pregnant women. Our study had the advantage of measuring PAF directly and using the most sensitive method currently available, which may
932 Rowland, Vermillion, and Roudebush
Fig 1. PAF concentrations (in nanograms per milliliter) in all 4 groups. Bar heights, Mean; error bars, SD. P = .01, analysis of variance.
have provided greater accuracy than either of these previously reported studies. Our findings differ from those of the study of Maki et al.4 In that study the investigators measured the level of PAF indirectly by recording the activity of PAF acetylhydrolase, which degrades PAF. They demonstrated that PAF acetylhydrolase activity was increased in patients with preeclampsia, thus signifying decreased concentrations of PAF. We were unable to detect any significant differences between participants with severe and mild preeclampsia in either the mean PAF concentration or the mean platelet count. However, the statistical power of the subanalysis within the preeclampsia group was limited by the small sample size of the subgroup of participants with mild preeclampsia. We are currently conducting research to specifically address a potential association between PAF concentration and the severity of preeclampsia. We recognized the potential for ethnic influences on PAF concentrations, so we compared the mean PAF concentrations between African American and white participants. We were unable to detect any difference between the two racial groups. PAF is a phospholipid that is a mediator of inflammatory and allergic responses, including vasoconstriction and platelet aggregation. The similarities of these properties of PAF to the clinical manifestations of preeclampsia have led to studies of an association between PAF and preeclampsia. Previous investigations have reported varying results with respect to the association between PAF and preeclampsia.3-6 On the basis of the findings in our
October 2000 Am J Obstet Gynecol
Fig 2. Pairwise analyses of PAF concentrations (in nanograms per milliliter) in female groups. Bar heights, Mean; error bars, SD. P < .05, preeclampsia group versus normotensive pregnant group; P < .05, preeclampsia group versus normotensive nonpregnant female group.
study, however, we conclude that PAF concentrations are increased in pregnancies affected by preeclampsia. Furthermore, PAF may serve as a marker for the increased thromboxane production, with its resultant vasoconstriction and platelet aggregation, that is associated with preeclampsia. Because an association between elevated circulating concentration of PAF and preeclampsia exists, a longitudinal cohort study to evaluate serial PAF concentrations could perhaps establish PAF as an early marker for preeclampsia before the onset of clinical symptoms. REFERENCES
1. Snyder F. Platelet-activating factor and related acetylated lipids as potent biologically active cellular mediators. Am J Physiol 1990;259:C697-708. 2. Rodrigues FA, Hofmeyr GJ, Sonnendecker EW. Hypothesis: a central role of platelet activating factor in the pathogenesis of preeclampsia. Med Hypotheses 1989;4:285-6. 3. Benedetto C, Massobrio M, Bertini E, Abbondanza M, Enrieu N, Tetta C. Reduced serum inhibition of platelet activating factor activity in preeclampsia. Am J Obstet Gynecol 1989;160:100-4. 4. Maki N, Magness RR, Miyaura S, Gant NF, Johnston JM. Platelet activating factor acetylhydrolase activity in normotensive and hypertensive pregnancies. Am J Obstet Gynecol 1993;168:50-4. 5. Norris LA, Gleeson N, Sheppard BL, Bonnar J. Whole blood platelet aggregation in moderate and severe preeclampsia. Br J Obstet Gynaecol 1993;100:684-8. 6. Zhang S, Zhang Y, Liu B. [Determination of platelet activating factor and platelet associated IgG in pregnancy-induced hypertension.] Chung Hua Fu Chan Ko Tsa Chih 1996;31:296-8. 7. American College of Obstetricians and Gynecologists. Hypertension in pregnancy. Washington: The College; 1996. ACOG Technical Bulletin No.: 219.