| |  Preeclampsia as a risk factor for cardiovascular disease later in life: validation of a preeclampsia questionnaire
Received 27 July 2007; received in revised form 6 September 2007; accepted 19 September 2007. published online 13 February 2008. ObjectiveThis study was undertaken to validate a self-administered questionnaire in verifying the diagnosis of preeclampsia, eclampsia, or toxemia in a group of women with a greater than 20-year history of preeclampsia. Study DesignQuestionnaires were mailed to a random sample of 144 women who received a diagnosis of any of these 3 conditions and 158 women who had normotensive pregnancies at Mayo Clinic, Rochester, Minnesota, from 1960-1979. ResultsA previous diagnosis of preeclampsia, eclampsia, or toxemia was verified with 80% sensitivity and 96% specificity. ConclusionOur validated questionnaire may be a useful research tool in identifying women with a previous history of preeclampsia. Women with a history of preeclampsia had a higher prevalence of future hypertension than those with a history of normotensive pregnancy. Preeclampsia, a pregnancy disorder characterized by hypertension and proteinuria, affects 3-5% of pregnancies.1 Eclampsia is characterized by the occurrence of seizures in a preeclamptic patient, whereas toxemia is a term that previously has been used for both conditions. The role of preeclampsia as a risk factor for cardiovascular disease (CVD) is increasingly recognized.2 Epidemiologic surveys testing this association depend on the ability of affected women to report this condition later in life. A study from Taiwan compared medical records with recall data obtained by telephone interview from women who delivered during 1995-2000, on average 6 years after the index delivery.3 The diagnosis of preeclampsia was reported by 2 of 3 women who had term pregnancies (sensitivity 66.7%), and by 23 of 29 with preterm pregnancies (sensitivity of 68.9%). With respect to the association between preeclampsia and future CVD, the accuracy of maternal recall becomes important during the postmenopausal years when there is an increased risk for CVD. A questionnaire, testing recollections of the diagnosis of preeclampsia more than 20 years after the affected pregnancies, has never been validated. The aim of this study was to design such a questionnaire and test its validity. In addition, we aimed to compare hypertension rates between women with a history of preeclampsia with those with a history of normotensive pregnancy. Materials and Methods  The institutional review board approved our protocol. All subjects consented to the use of their records for research. The Mayo diagnostic index was used to search for the diagnoses of preeclampsia, eclampsia, or toxemia, 1960-1979. One of 3 physicians (M.C.H., A.A.E., and V.D.G.) reviewed the chart of each patient. The information obtained from the medical records was the “gold standard” for the diagnosis of preeclampsia and used to confirm the diagnosis based on the current guidelines1: (1) the presence of hypertension (blood pressure 140/90 mm Hg or higher on 2 determinations performed at least 6 hours apart), and (2) proteinuria (urinary excretion of 0.3 g of protein or greater in a 24-hour urine collection or its 1+ dipstick equivalent). Rochester Mayo Clinic medical records were reviewed and 161 cases with these diagnoses were randomly selected. Nineteen cases were excluded after chart review (1 had denied research authorization, 2 were deceased, 16 did not fulfill the diagnostic criteria). The control group consisted of women who reached full-term delivery without hypertension or proteinuria during pregnancy and early postpartum. Women with normal index pregnancies, but with previous histories of preeclampsia, eclampsia, or toxemia and those who had hypertension develop in pregnancy but did not progress to preeclampsia or eclampsia were excluded. Our questionnaire (Table 1) contained all 3 terms, covering the spectrum of the disease and possible differences in terminology that existed between 1960-1979. The agreement between answers from each individual and the diagnosis from medical records was tested. A positive response (ie, verification of preeclampsia) was defined as a self-report of any of the following: 1.History of preeclampsia, eclampsia, or toxemia, with or without hypertension, during the index pregnancy. 2.Hypertension and proteinuria during, but not before the index pregnancy. 3.Hypertension and seizures during, but not before the index pregnancy.  | 1.Are you currently taking prescription medication to lower your blood pressure? 2.Have you had at least one pregnancy that lasted more than 6 mo? 3.During any of these pregnancies (which lasted more than 6 mo), did a physician ever tell you that you had high blood pressure or hypertension?a)Yesa.In the first pregnancy only b.Not in the first, but in a subsequent pregnancy c.In the first pregnancy and in at least 1 subsequent pregnancy b)Noa.During any of your pregnancies, did you have preeclampsia, eclampsia, or toxemia? 4.In what year, or how old were you, when the pregnancy-related high blood pressure first occurred?a. Year____ or Age __ 5.During any of the pregnancies in which you developed hypertension, did you have:a)Protein in the urine b)Seizures or convulsions c)Preeclampsia, eclampsia, or toxemia of pregnancy 6.Before the first pregnancy in which you developed hypertension, did you have:a)Protein in the urine b)Seizures or convulsions c)High blood pressure (hypertension) | | | | |
We contacted 144 women with a confirmed diagnosis of any of these 3 conditions and 158 time-matched controls, aiming for 100 completed questionnaires in each group (expected response rate 60-70%). A sample of 100 women in each group was chosen to allow comparable precision for the estimation of sensitivity and specificity. It was also a feasible sample size from the perspectives of chart review and survey effort. Results Response rates between cases (103/144, 71.5%) and controls (100/158, 63.3%) were not different (P = .13). Women who responded (n = 202) compared with those who did not (n = 99) were similar with respect to mean age (55 vs 55.9 years, respectively; P = .49), and median time interval from index pregnancy to survey (27 vs 27.8 years, respectively; P = .51). Women with a history of preeclampsia had fewer pregnancies and a significantly (P ≤ .0001) higher frequency (48.5%) of current hypertension compared with 22% for those who had normotensive pregnancies (Table 2). For the cases, a positive response was obtained from 82 women who verified that they had preeclampsia. Among these, 81 reported a history of preeclampsia, eclampsia, or toxemia with (n = 23) or without hypertension (n = 58); with a history of proteinuria during pregnancy (n = 28), seizures during pregnancy (n = 1) and both (n = 2). In only 1 case was the diagnosis ascertained based on the report of hypertension and proteinuria, but not preeclampsia, eclampsia, or toxemia. The sensitivity of our questionnaire, true positives/(true positives ± false negatives), was 79.6%, with a 95% CI of 71.8-87.4%. Four control subjects reported a history of preeclampsia. Accordingly, our specificity, true negatives/ (true negatives ± false positives), was 96%, with a 95% CI of 92.2-99.8%. The respective positive predictive value (PPV) and negative predictive value (NPV) for our questionnaire, given a commonly reported 5% prevalence of preeclampsia, were 51% and 99%, respectively. Conclusion Accumulating evidence suggests that preeclampsia is associated with an increased risk for CVD later in life. As CVD events in women typically occur during menopause, a maternal recall bias years after the affected pregnancies is of particular concern. Previous studies have tested the accuracy of maternal recall of preeclampsia, but not more than 6 years on average after the affected pregnancies. In our study, we have tested the accuracy of a questionnaire used to screen a history of preeclampsia 24.5 years after the index pregnancies. Our validated preeclampsia questionnaire demonstrated 80% sensitivity and 96% specificity in verifying a history of preeclampsia, with a PPV of 51%. Rigorous medical chart review to confirm the presence or absence of the diagnostic criteria of preeclampsia supports the validity of our estimates of sensitivity, specificity, and PPV. Maternal recall of a remote vs a more recent history of preeclampsia does not appear to be significantly different. A recent study reported a PPV of 59.2% of the women's own report of preeclampsia in a 6-month postpartum interview,4 which was administered to women in the Danish National Birth Cohort who gave birth in the years 1998-2000. In future studies, we will use our validated questionnaire to identify women with a history of preeclampsia and explore the association of preeclampsia with future CVD events, pre-clinical markers of atherosclerosis (such as microalbuminuria), and conventional, predisposing, and novel risk factors. Our current observations provide preliminary data to support our future studies: women with a history of preeclampsia had a higher prevalence of hypertension later in life compared with those with a history of normotensive pregnancy. In the current study, we were interested in identifying women who had ever had preeclampsia and we did not test the recall of preeclampsia in a specific index pregnancy. Consequently, our questionnaire will not be useful in identifying individuals born to preeclamptic mothers, and thus in studying the long-term effects of in utero exposure to preeclampsia. In addition, preeclampsia, being a systemic disease, may have different underlying mechanisms than other pregnancy disorders, and thus different implications for later CVD. Before future studies aiming to make distinctions among different hypertensive pregnancy disorders with respect to their long-term CVD outcomes, we plan to validate our questionnaire in a cohort of women with isolated hypertension. References 1. 1Anonymous. Report of the NHBPEP Working Group on High Blood Pressure in Pregnancy. Am J Obstet Gynecol. 2000;(183):S1–S22. 2. 2Garovic VD, Hayman SR. Hypertension in pregnancy: an emerging risk factor for cardiovascular disease. Nat Clin Pract Nephrol. 2007;3:613–622.
CrossRef
3. 3Sou SC, Chen WJ, Hsieh WS, Jeng SF. Severe obstetric complications and birth characteristics in preterm or term delivery were accurately recalled by mothers. J Clin Epidemiol. 2006;59:429–435. Abstract | Full Text |
Full-Text PDF (120 KB)
|
CrossRef
4. 4Klemmensen AK, Olsen SF, Osterdal ML, Tabor A. Validity of preeclampsia-related diagnoses recorded in a national hospital registry and in a postpartum interview of the women. Am J Epidemiol. 2007;166:117–124. MEDLINE |
CrossRef
a Division of Obstetrics and Gynecology, Mayo Clinic, Rochester, MN b Division of Nephrology and Hypertension, Mayo Clinic, Rochester, MN c Department of Medicine, Mayo Clinic, Rochester, MN d Biostatistics, Mayo Clinic, Rochester, MN.  Reprints not available from the authors. Address correspondence to Vesna D. Garovic, MD, Division of Nephrology and Hypertension, Mayo Clinic, 200 First Street SW, Rochester, MN 55905.
PII: S0002-9378(07)01182-9 doi:10.1016/j.ajog.2007.09.038 © 2008 Mosby, Inc. All rights reserved. | |
|
FULL TEXT