Objective
Women who develop certain common pregnancy complications have a greater chance of developing cardiovascular disease (CVD) later in life. However, most health care providers do not provide postpartum cardiovascular risk counselling or follow-up. The Maternal Health Clinic was established to address this gap in care. It targets women at increased risk of CVD to inspire lifestyle changes, encourage long-term follow-up, and initiate primary prevention. Here, we summarize results from the first 17 months of completed clinic visits.
Study Design
Patients experiencing at least one relevant complication in their index pregnancy were referred to the Maternal Health Clinic through standard postpartum order sheets. Patients underwent a complete assessment including screening history, physical examination, fasting bloodwork, and urinalysis. Lifetime and 30-year CVD risk estimates, along with a metabolic syndrome calculation, were determined for each patient.
Results
Complications most commonly leading to referral were gestational diabetes or impaired glucose tolerance (32.7%), preeclampsia (29.3%), preterm birth (29.3%), and gestational hypertension (19.6%). The clinic analysis group (n = 92) was compared with a healthy control group from the PreEclampsia New Emerging Team study (n = 118). Patients in the clinic analysis group had significantly increased lifetime and 30-year CVD risk estimates compared with healthy controls ( P < .0001). Furthermore, 17.4% of the clinic analysis group had metabolic syndrome, compared with 6.78% of healthy controls ( P < .05).
Conclusion
This study demonstrates that the Maternal Health Clinic accurately identifies postpartum patients that have underlying cardiovascular risks which make them susceptible to CVD. The clinic may serve as an effective primary prevention strategy.
Cardiovascular disease (CVD) is the leading cause of death in women worldwide. Despite greater awareness of this issue, mortality because of coronary heart disease in women aged 35 to 54 years is increasing. This trend underscores the need for earlier recognition of those at risk and implementation of primary prevention strategies.
Pregnancy is a physiologic stress test that identifies women at risk of future CVD. Large epidemiologic and cohort studies have demonstrated that patients who develop certain pregnancy complications have an elevated risk of CVD and CVD-related mortality. For instance, index pregnancies complicated either by hypertensive disorders, gestational diabetes, intrauterine growth restriction, or preterm delivery have been associated with a 50-300% increased risk of CVD. Furthermore, in women who experience multiple complications within a single pregnancy or have concomitant traditional risk factors such as obesity or dyslipidemia, the risk of CVD is further amplified. The obstetric history may therefore act as a more sensitive screening tool for CVD risk than traditional prediction models, such as the Framingham risk score, which rarely recognizes young women at risk.
Taken together, these studies suggest that pregnancy complications can identify high-risk women early enough to implement primary preventative strategies. However, few health care providers carry out postpartum CVD risk counseling and data on risk modification initiatives for this population is lacking. In a recent issue of Hypertension , Spaan et al propose the implementation of a structured postpartum cardiovascular screening program for women who have experienced a hypertensive disorder of pregnancy, citing the postpartum period as a time when women are motivated to modify their lifestyle. The Maternal Health Clinic at Kingston General Hospital (KGH), Queen’s University, follows many aspects of the suggested framework. To our knowledge, this is the first initiative of its kind in North America. The clinic aims to see women who have experienced a pregnancy-related cardiovascular risk (CVR) indicator in order to discuss their increased risk of CVD, identify comorbidities, encourage lifestyle changes, and ultimately ensure long-term follow-up and/or specialist referral. Our clinic methodology has been described in detail in a previous article, thus will only be briefly summarized in this report. In support of what was proposed in Hypertension, the objective of this article is to present the results of the first 17 months of completed visits at the Maternal Health Clinic.
Materials and Methods
Subjects
To be eligible for referral to the Maternal Health Clinic, patients had to experience one or more of the following pregnancy-related CVR indicators in their index pregnancy: gestational hypertension (GH), preeclampsia (PE), gestational diabetes (GDM), gestational impaired glucose tolerance (GIGT), idiopathic preterm birth, placental abruption, or intrauterine fetal growth restriction (IUGR). Women who had traditional CVD risk factors such as chronic prepregnancy hypertension or diabetes were eligible for referral to Maternal Health Clinic only if they experienced 1 of the relevant pregnancy complications. Diagnoses were validated by a research assistant through review of prenatal records. Patients were excluded if they lived outside of our local health integration network. All scheduled clinic visits from Nov. 2011 to March 2013 were considered in this analysis. Blood and urinalysis results received by April 30, 2013, were included. The Research Ethics Board at Queen’s University approved the study and all patients gave informed consent before participation.
Method
The Maternal Health Clinic has been adopted as standard of care in the Department of Obstetrics and Gynecology at KGH. All patients who deliver at KGH are considered for referral. At the time of discharge from hospital, those eligible are automatically referred through the use of standard postpartum order sheets. Patients are seen in the Maternal Health Clinic at 6 months postpartum, where they undergo a history and physical examination and are subsequently sent for fasting glucose, lipid profile, and urinalysis including microalbumin and creatinine. If a patient experienced GDM or GIGT, they are also asked to complete a 2-hour oral glucose tolerance test. Blood pressure measurements are obtained as previously described. Patients also meet with the obstetrician to discuss their pregnancy complication, its relation to CVD, and suggested lifestyle modifications to facilitate CVR reduction. A follow-up letter and clinical data collection form is sent to both the patient and her family doctor, and further consults to endocrinology, cardiology, and nephrology are made based on the patient’s individual profile.
Patients scheduled for a visit during our analysis period of Nov. 2011 to March 2013 delivered at KGH approximately 6 months prior, between May 2011 and Sept. 2012. Over that interval, 3015 women delivered at KGH and were screened for eligibility ( Figure ). Of these, 387 were invited to clinic based on inclusion/exclusion criteria, and 157 (40.6%) attended the clinic. At the time of analysis, 102 were considered as they had undergone a full assessment, which included their clinic visit and having completed fasting blood work and urinalysis. However, data are presented for 92 women. Ten women with complete visits and blood work were omitted because their reason for clinic referral was excess weight gain during their index pregnancy. This was originally considered in the inclusion criteria for referral to the Maternal Health Clinic but has since been removed. Data from these women was therefore omitted such that our analysis reflects the current population of individuals seen at the Maternal Health Clinic.
Analysis
Patients in the Maternal Health Clinic analysis group (n = 92) were compared with a normotensive postpartum control group (n = 118) from the Preeclampsia New Emerging Team (PE-NET) prospective longitudinal cohort, previously described. Normotensive women with a prior history of hypertension, diabetes, renal disease, or cardiovascular disease were excluded from the PE-NET cohort. Descriptive statistics were used to analyze the distribution of baseline characteristics in the Maternal Health Clinic analysis group and the PE-NET control group. Risk of CVD was determined for individual patients by 3 methods: a lifetime CVD risk estimate, a 30-year CVD risk estimate, and metabolic syndrome calculation. These measures were chosen because they help to better communicate risk to postpartum women who may have low short-term risk despite having significant risk factors for future CVD. The multiple comparison χ 2 test was used to compare the percentage of individuals in the Maternal Health Clinic analysis group and PE-NET control group who had optimal, non-optimal, elevated, or major lifetime risk estimates. The risk factors included in the lifetime CVD risk estimate included total cholesterol, systolic blood pressure, diastolic blood pressure, elevated fasting glucose or a previous diagnosis of diabetes, and smoking status. The Mann-Whitney U test was used to compare the mean 30-year risk estimate in each group. Finally, the χ 2 test was used to compare the prevalence of metabolic syndrome between groups. Ante hoc power analysis was not performed. Statistical analysis was performed with R version 2.13.2 ( www.r-project.org/ ) and OpenEpi version 2.3.1 ( www.openepi.com/v37/Menu/OE_Menu.htm ).
Results
Baseline demographics are presented in Table 1 . The mean age of patients in the Maternal Health Clinic analysis group was 32.9 years with a prepregnancy body mass index (BMI) of 26.7kg/m 2 ; 6 women had chronic hypertension and none had diabetes before pregnancy. There were no differences in parity, annual household income, educational attainment, prepregnancy BMI, or breastfeeding practices between the Maternal Health Clinic analysis group and PE-NET healthy control group. However, patients in the Maternal Health Clinic analysis group had more varied ethnicity, were more likely to have gained excess weight in pregnancy, and to have elevated postpartum BMI, waist circumference, and pregnancy weight retention ( P < .05).
| Characteristic | MHC analysis group | PE-NET control group | P value |
|---|---|---|---|
| Age (y), mean (SD), n (%) | 32.9 (5.4) | 31.5 (4.1) | < .05 |
| ≤29 | 26 (28.3) | 51 (43.2) | < .05 |
| 30-34 | 33 (35.9) | 48 (40.7) | |
| 35-39 | 21 (22.8) | 17 (14.4) | |
| ≥40 | 12 (13.0) | 2 (1.7) | |
| Marital status, n (%) | |||
| Married | 74 (80.4) | Data not collected | ND |
| Common-law | 17 (18.5) | ||
| Single | 1 (1.1) | ||
| Ethnicity, n (%) | |||
| White | 79 (85.9) | 118 (100.0) | < .05 |
| First nations, Inuit or Metis | 3 (3.3) | 0 (0.0) | |
| Asian | 8 (8.6) | 0 (0.0) | |
| Other | 2 (2.2) | 0 (0.0) | |
| Educational attainment, n (%) | |||
| ≥Some college | 83 (90.2) | 113 (95.8) | NS |
| High school | 7 (7.6) | 5 (4.2) | |
| Grade school | 1 (1.1) | 0 (0.0) | |
| Unknown | 1 (1.1) | 0 (0.0) | |
| Annual household income, n (%) | |||
| <$30,000 | 7 (7.6) | 8 (7.0) | NS |
| $30,001-60,000 | 14 (15.2) | 17 (14.9) | |
| $60,001-90,000 | 25 (27.2) | 30 (26.3) | |
| ≥$90,001 | 44 (47.8) | 59 (51.8) | |
| Unknown | 2 (2.2) | 4 (3.9) | |
| Parity, n (%) | |||
| 1 | 55 (59.8) | 73 (61.9) | NS |
| 2 | 25 (27.2) | 35 (29.7) | |
| 3 | 10 (10.8) | 9 (7.6) | |
| ≥4 | 2 (2.2) | 1 (0.8) | |
| Smoking at follow-up, n (%) a | |||
| Never smoked | 59 (64.1) | 113 (95.8) | ND |
| Previous smoker | 22 (23.9) | ||
| Current smoker | 11 (12.0) | 5 (4.2) | |
| Alcohol consumption at follow-up (per wk), mean (SD), n (%) | 1.2 (1.5) | ||
| ≥3 drinks | 7 (7.6) | Data not collected | ND |
| 1-2 drinks | 38 (41.3) | ||
| None | 47 (51.1) | ||
| Distance from the clinic, n (%) | |||
| <1 h from the clinic | 84 (91.3) | Data not collected | ND |
| ≥1 h from the clinic | 8 (8.7) | ||
| Prepregnancy BMI (kg/m 2 ), mean (SD), n (%) | 26.7 (6.0) | 25.4 (5.1) | NS |
| <18.5 | 2 (2.2) | 2 (1.7) | NS |
| 18.5-24.9 | 43 (46.7) | 64 (54.2) | |
| 25.0-29.9 | 26 (28.3) | 30 (25.4) | |
| ≥30.0 | 21 (22.8) | 22 (18.6) | |
| Excess weight gain in pregnancy (kg), mean (SD), n (%) | 14.6 (8.2) | 17.3 (5.3) | < .05 |
| Prepregnancy BMI <18.5, weight gain >18.0 | 1 (1.1) | 1 (0.8) | < .05 |
| Prepregnancy BMI 18.5-24.9, weight gain >16.0 | 16 (17.3) | 37 (31.4) | |
| Prepregnancy BMI 25.0-29.9, weight gain >11.5 | 18 (19.6) | 27 (22.9) | |
| Prepregnancy BMI ≥30.0, weight gain >9.0 | 11 (12.0) | 19 (16.1) | |
| No excess weight gain | 46 (50.0) | 33 (28.0) | |
| Unknown | 0 (0.0) | 1 (0.8) | |
| Pregnancy weight retention at follow-up (kg), mean (SD), n (%) | 5.2 (5.6) | 2.3 (6.3) | < .05 |
| ≥10.0 | 17 (18.5) | 11 (9.3) | NS |
| 7.0-9.9 | 13 (14.1) | 16 (13.6) | |
| 3.0-6.9 | 23 (25.0) | 25 (21.2) | |
| <3.0 | 39 (42.4) | 66 (55.9) | |
| Waist circumference at follow up (cm), mean (SD) | 90.7 (14.4) | 87.8 (14.3) | NS |
| Excess (>80 for Asian, >88 for all other races) | 48 (52.1) | 30 (25.4) | < .05 |
| Borderline (72-80 for Asian, 80-88 for all other races) | 26 (28.3) | 24 (20.3) | |
| Normal (<72 for Asian, <80 for all other races) | 17 (18.5) | 30 (25.4) | |
| Unknown | 1 (1.1) | 34 (28.9) | |
| BMI at follow-up (kg/m 2 ), mean (SD) , n (%) | 28.2 (6.3) | 26.3 (5.8) | < .05 |
| <18.5 | 1 (1.1) | 4 (3.4) | < .05 |
| 18.5-24.9 | 27 (29.3) | 54 (45.8) | |
| 25.0-29.9 | 31 (33.7) | 34 (28.8) | |
| ≥30.0 | 33 (35.9) | 26 (22.0) | |
| Breastfeeding for the recent birth, n (%) | |||
| Breastfeeding ≥6 mo | 50 (54.3) | 37 (31.4) | NS |
| Breastfed for <6 mo | 23 (25.0) | 16 (13.6) | |
| No breastfeeding | 19 (20.7) | 8 (6.8) | |
| Unknown | 0 (0.0) | 57 (48.3) | |
| Frequency of activity at follow-up (times per wk), mean (SD), n (%) | 4.0 (1.8) | ||
| ≥5 | 19 (20.7) | Data not collected | ND |
| 3-4 | 25 (27.2) | ||
| 1-2 | 10 (10.8) | ||
| None | 38 (41.3) |
a For the PE-NET smoking data was only captured as Yes or No, information on previous smoking habits was not collected.
A comparison of eligible patients who attended their booked Maternal Health Clinic appointment (n = 119, excludes 21 patients with blood work pending at the time of analysis) to those who declined or did not show for their appointment (n = 172) demonstrated that patients who did not attend a clinic were younger, less educated, and of higher parity. They also had a higher prepregnancy BMI and were more likely to have gained excess weight over the course of pregnancy (data not shown; P < .05). No difference in distance from the hospital, marital status, or ethnicity was found (data not shown; P > .05).
Table 2 demonstrates the index pregnancy complications that lead to referral to the Maternal Health Clinic. The most common pregnancy-related complication for women who had complete clinic visits and blood work was gestational diabetes or impaired glucose tolerance (32.7%). This was followed by preeclampsia (29.3%), idiopathic preterm birth (29.3%), and gestational hypertension (19.6%). The distribution of such complications was similar in eligible patients who did not attend a clinic. Abruption was the least common pregnancy-related complication in women who did and did not attend a clinic, accounting for 5.4% and 4.7% of Maternal Health Clinic referrals, respectively.
| Variable | Maternal Health Clinic analysis group | Women invited to clinic who did not attend |
|---|---|---|
| Preeclampsia, n (%) | 27 (29.3) | 29 (16.7) |
| Gestational hypertension, n (%) | 18 (19.6) | 29 (16.7) |
| Gestational diabetes or impaired glucose tolerance, n (%) | 30 (32.7) | 62 (36.0) |
| Abruption, n (%) | 5 (5.4) | 8 (4.7) |
| Preterm birth, n (%) | 27 (29.3) | 52 (30.2) |
| IUGR, n (%) | 14 (15.2) | 26 (15.1) |
Table 3 represents the physical and biochemical parameters of women in the Maternal Health Clinic analysis group. Although mean systolic and diastolic blood pressures were normal at 119.9 mm Hg and 81.73 mm Hg, respectively, 16.3% of patients had systolic pressures over 140 mm Hg or were taking antihypertensive medication, and 28.3% had diastolic pressures over 90 mm Hg or were taking antihypertensive medication. Lipid values were abnormal in a considerable proportion of patients: high density lipoprotein (HDL) was low in 35.9% (normal ≥1.3 mmol/L), low density lipoprotein (LDL) was high in 53.3% (normal <2.6 mmol/L), and total cholesterol (TC) was high in 20.7% (normal <5.16 mmol/L). Microalbumin/creatinine ratio was high in 13.2% (normal <2.8 mg/mmol). Fasting glucose levels, however, were normal in most (97.8%) patients, and only 1 of 30 women in the Maternal Health Clinic analysis group who were asked to complete an oral glucose tolerance test had abnormal results.
