Abstract
Objective:
N-terminal pro–brain natriuretic peptide (NT-proBNP), a marker of ventricular dysfunction, varies by body mass index (BMI) outside of pregnancy. This study aimed to determine whether obesity affects NT-proBNP levels in pregnancy.
Study design:
This was a prospective observational study of healthy pregnant people in the 3rd trimester (3TM) and postpartum (PP). Subjects were excluded if they had significant medical comorbidities or if their fetuses had anomalies, growth restriction or aneuploidy. NT-proBNP was measured at 28 weeks (3TM), prior to delivery (PD), 1–2 days PP (IPP), and 4–6 weeks PP (DPP). LogNT-proBNP levels were analyzed using linear mixed effects models, including BMI < or ≥30, time, and time-by-BMI interactions.
Results:
Fifty-five people (28 [51%] with BMI ≥ 30 and 27 [49%] with BMI < 30) were enrolled. A greater proportion of obese than non-obese subjects developed hypertensive disorders of pregnancy (50% vs 15%, p=0.010) and obese subjects had higher systolic blood pressures at all time points (p<0.05). NT-proBNP levels (median [IQR] in pg/mL) were 18 (6–28) vs 26 (17–48) at 3TM, 16 (3–38) vs 43 (21–60) at PD, 58 (20–102) vs 63 (38–155) at IPP, and 33 (27–56) vs 23 (8–42) at DPP for obese compared to non-obese subjects. In linear mixed effects models, logNT-proBNP was lower in obese subjects at 3TM (β=−0.89 [95% confidence interval −1.51, −0.26]) and PD (β=−1.05 [95% CI −1.72, −0.38]). The logNT-proBNP trends over time differed by BMI category, with higher values in obese subjects at both postpartum time points compared to the 3TM (IPP β=1.24 [95% CI 0.75, 1.73]); DPP β=1.08 [95% CI 0.52, 1.63]), but only IPP for non-obese subjects (β=0.87 [95% CI 0.36, 1.38]).
Conclusions:
Obese subjects had lower NT-proBNP levels than non-obese subjects during pregnancy, but not postpartum. The prolonged postpartum elevation in NT-proBNP in obese subjects suggests that their postpartum cardiac recovery may be more prolonged.
Keywords: Pregnancy, amino-terminal brain natriuretic peptide, brain natriuretic peptide, heart failure, obesity, peripartum cardiomyopathy
Introduction
Brain natriuretic peptide (BNP) is a cardiac neurohormone that is synthesized in and secreted by the cardiac ventricles in response to ventricular overload.1 Outside of pregnancy, measurement of brain natriuretic peptide (BNP) or amino-terminal pro-brain natriuretic peptide (NT-proBNP), a cleavage fragment of the prohormone proBNP, is recommended in the evaluation of acute dyspnea and in the diagnosis of heart failure.2 With rising rates of maternal obesity, a significant proportion of people who present with dyspnea in pregnancy are obese and may be at higher risk for adverse cardiovascular events.3 In one retrospective Swedish cohort, the odds of heart failure in late pregnancy and postpartum was 2.5-fold higher in obese people compared to lean people.3 In the setting of dyspnea in pregnancy and postpartum, a screening modality to help categorize obese pregnant people as high or low probability for heart failure would be helpful. During pregnancy, elevated BNP has been associated with adverse cardiovascular events 4, while normal BNP levels have been observed in healthy pregnancies.5 While BNP levels outside of pregnancy have been shown to be lower in obese patients, variations in BNP levels by body mass index (BMI) in pregnancy have not been studied.6,7 Before BNP can be deemed a reliable screening test for heart failure or ventricular dysfunction in obese pregnant people, normal BNP values in the setting of obesity and pregnancy need to be investigated. We, therefore, proposed the following specific aim: to compare plasma NT-proBNP levels in the third trimester, immediately postpartum, and at 4–6 weeks postpartum between obese pregnant people without cardiovascular disease and non-obese pregnant people without cardiovascular disease. We hypothesized that plasma NT-proBNP levels in obese pregnant people without pre-existing cardiovascular disease would be significantly lower than levels in non-obese pregnant people at all pregnancy and postpartum timepoints. The results of this study could aid in providing an understanding of the baseline trends and variation in NT-proBNP levels in obese and non-obese pregnant people.
Materials and Methods
This was a prospective observational study of pregnant people receiving prenatal care through the Duke University Health System. The study was approved by the Duke University School of Medicine Institutional Review Board (Protocol # Pro00101213) and was registered with ClinicalTrials.gov (ClinicalTrials.gov ID# NCT04049136).
Potential study subjects were approached by trained research staff at routine prenatal visits from July 2019 to February 2020. Subjects were included if they were approaching the third trimester of pregnancy and due to have routine third trimester laboratory evaluation completed, ≥18 years old, and English-speaking. Exclusion criteria included chronic hypertension (diagnosis pre-dating pregnancy or blood pressure reading ≥ 140/90 at less than 20 weeks’ gestation), autoimmune disorder (systemic lupus erythematosus, rheumatoid arthritis, or Sjogren’s syndrome), baseline renal disease with creatinine >1.0, pre-existing diabetes mellitus, history of cardiomyopathy or heart failure, or history of myocardial infarction or cardiac arrest. Potential subjects were also excluded if their fetus was suspected to have a congenital anomaly, intrauterine fetal demise, severe fetal growth restriction at the time of enrollment (based on the contemporaneous institutional definition of severe fetal growth restriction as an estimated fetal weight <5th percentile), or aneuploidy. Subjects with pre-pregnancy BMI ≥ 30 were considered obese, while subjects with BMI <30 were considered non-obese.
Once written informed consent to participate in the study was obtained and a patient deemed eligible to participate, plasma N-terminal pro-brain natriuretic peptide levels were drawn, centrifuged, and the plasma analyzed for NT-proBNP by Duke Clinical Laboratories using electrochemiluminescence immunoassays on the Roche Cobas e411 analyzer at the following four timepoints: at the start of the third-trimester with routine third trimester blood work (referred to as 3rd-trimester or 3TM); on admission to labor and delivery for labor, rupture of membranes, or scheduled delivery (referred to as “pre-delivery” or PD); within 24–48 hours postpartum (referred to as “immediate postpartum” or IPP), and 4–6 weeks postpartum to coincide with postpartum visit (referred to as “delayed postpartum” or DPP). The reference range for NT-proBNP at Duke Clinical Laboratories for females <55 years of age is ≤190 pg/mL.
The following additional data were collected from the electronic medical record: maternal age at delivery, maternal height, maternal pre-pregnancy weight and weight at each time point, best obstetric estimate of due date, gestational age or days postpartum for each phlebotomy draw, patient-reported race and ethnicity, gravidity, parity, smoking status, mode of delivery, delivery date, maximum systolic blood pressure at each timepoint, hemoglobin and hematocrit at third trimester phlebotomy draw and admission, diagnosis of hypertensive disorders of pregnancy (gestational hypertension, preeclampsia, or HELLP syndrome), total intravenous fluids received in labor, and blood loss at delivery.
Based on a presumed 19.6% difference in mean plasma BNP levels between obese and non-obese women8 and an estimated mean BNP of 33±8 pg/mL in healthy pregnant women in the 3rd trimester9, a sample size of 25 obese pregnant people and 25 non-obese pregnant people was calculated as sufficient to satisfy power calculations with a two-sided alpha of 0.05 and beta=0.20 to reject a null hypothesis that mean plasma NT-proBNP levels do not differ between obese and non-obese pregnant people.8 To account for potential loss to follow-up of 10%, a sample size of 56 subjects was planned.
Demographics, pregnancy, and delivery characteristics were summarized as mean ± standard deviation (SD), median (interquartile range), or frequency (percentage) by BMI category. Continuous variables were compared between BMI classes using ANOVA or Wilcoxon rank sum tests, while categorical variables were compared using the Chi-Square test or Fisher’s exact tests. Log-transformed NT-proBNP values were used for the normality and homoscedasticity assumptions. Original-scale NT-proBNP values are provided for clinical context. Linear mixed effects models were fit with a random intercept to account for within-subject correlations induced by repeated measurement of the outcome over the four study visits. The unadjusted model included obese versus non-obese, study time points, and their interaction terms. The overall interaction effect was assessed using the likelihood-ratio test. If the test was significant (p<0.05), contrast statements were used to estimate the comparisons of interest. The final adjusted model included race and hypertensive disorders of pregnancy. All analyses were performed in SAS 9.4 (SAS Institute, Cary, NC) using the PROC MIXED procedure.
Results
Subject enrollment is outlined in Figure 1. A total of 56 subjects were enrolled. One subject was subsequently withdrawn from the study due to unexpected non-obstetric complications prior to obtaining the third trimester NT-proBNP. Fifty-five subjects were included in the final analysis, of whom 27 had a BMI<30 and 28 had a BMI≥30. An additional 56 subject was unable to be enrolled to replace the subject that was withdrawn due to cessation of research activity during the COVID-19 pandemic. The mean pre-pregnancy BMI for obese subjects was 41.0 ± 8.5 kg/m2 and for non-obese subjects was 24.8 ± 2.8 kg/m2. As shown in Table 1, obese and non-obese subjects were similar in terms of age (31.0 ± 4.1 vs. 33.2 ± 5.1 years, p=0.079), use of public insurance (39.3% vs. 37.0%, p=0.864), median parity (1 (IQR 0–2) vs. 1 (IQR 0–2), p=0.361), and tobacco use (10.7% vs. 14.8%, p=0.705). Obese subjects were significantly more likely to be of black race (75.0% vs. 29.6%, p=0.002).
Figure 1.
Subject enrollment diagram.
Table 1.
Baseline, pregnancy, and delivery characteristics compared between obese (BMI ≥30) and non-obese (BMI<30) pregnant women without cardiovascular disease.
| Characteristic | BMI ≥ 30 N=28 |
BMI < 30 N=27 |
p-value |
|---|---|---|---|
| Baseline characteristics | |||
| Maternal age at delivery, mean ± SD | 31.0 ± 4.1 | 33.2 ± 5.1 | 0.079 |
| Public insurance, n(%) | 11 (39.3%) | 10 (37.0%) | 0.864 |
| Pre-pregnancy BMI, kg/m2, mean ± SD | 41.0 ± 8.5 | 24.8 ± 2.8 | <0.0001 |
| Self-reported race, n(%) | 0.002 | ||
| White | 6 (21.4%) | 15 (55.6%) | |
| Black | 21 (75.0%) | 8 (29.6%) | |
| Asian | 0 (0%) | 2 (7.4%) | |
| Other | 1 (3.6%) | 2 (7.4%) | |
| Self-reported Hispanic ethnicity, n(%) | 2 (7.1%) | 2 (7.4%) | 1.00 |
| Parity, median (IQR) | 1 (0–2) | 1 (0–2) | 0.361 |
| Tobacco use, n(%) | 3 (10.7%) | 4 (14.8%) | 0.705 |
| Pregnancy characteristics | |||
| SBP (mmHg) at 3rd-trimester visit, median (IQR) | 119 (112, 125) | 111 (105, 118) | 0.042 |
| Maximum SBP (mmHg) prior to delivery, median (IQR) | 146 (138, 154) | 135 (126, 146) | 0.006 |
| SBP (mmHg) post-delivery, median (IQR) | 123 (114, 136) | 113 (106, 125) | 0.019 |
| SBP (mmHg) at postpartum visit, median (IQR) | 120 (113.0, 131.0) | 115 (105.0, 121.0) | 0.022 |
| 3rd trimester hematocrit (%), median (IQR) | 34.4 (32.1, 35.2) | 33.9 (32.6, 36.3) | 0.444 |
| Hematocrit at delivery admission (%), median (IQR) | 33.6 (32.5, 36.4) | 36.1 (33.9, 37.8) | 0.103 |
| Hypertensive disorder of pregnancy, n(%) | 14 (50%) | 4 (15%) | 0.010 |
| Delivery characteristics | |||
| Gestational age at delivery (w), median (IQR) | 38.7 (37.3, 39.3) | 39.1 (37.9, 39.4) | 0.245 |
| Mode of delivery, n(%) | 1.00 | ||
| Spontaneous vaginal | 15 (54%) | 14 (54%) | |
| Assisted vaginal | 2 (7%) | 2 (7%) | |
| Cesarean | 11 (39%) | 10 (39%) | |
| Type of labor, n(%) | 0.461 | ||
| Induced | 15 (54%) | 11 (42%) | |
| Spontaneous | 6 (21%) | 10 (39%) | |
| Planned cesarean | 7 (25%) | 5 (19%) | |
| Duration of labor (h), median (IQR) | 7.0 (0,15.5) | 5.0 (2.0,12.0) | 0.877 |
| Quantitative blood loss at delivery (mL), median (IQR) | 632 (225, 1013) | 398 (250, 779) | 0.216 |
| Total IV fluids in labor (L), median (IQR) | 2.0 (1.5, 2.4) | 1.5 (0.9, 2.0) | 0.102 |
BMI = body mass index; IQR = interquartile range; SD = standard deviation; SBP = systolic blood pressure
Footnote: Demographics, pregnancy, and delivery characteristics are presented as mean ± standard deviation (SD), median (IQR), or frequency (%) by BMI category. Continuous variables were compared between BMI classes using ANOVA or Wilcoxon rank sum tests, while categorical variables were compared using the Chi-Square test or Fisher’s exact tests.
Obese and non-obese subjects exhibited several differences in terms of pregnancy and delivery characteristics (Table 2). Systolic blood pressure was significantly higher in obese subjects at the 3rd-trimester visit (median 119 (IQR 112, 125) vs. 111 (IQR 105, 118) mmHg, p=0.042), pre-delivery (measured as the maximum systolic blood pressure during labor; 146 (138, 154) vs. 135 (IQR 126, 146), p=0.006), immediately postpartum (123 (IQR 114, 136) vs. 113 (IQR 106, 125), p=0.019), and at the postpartum visit (120 (IQR 113, 131) vs. 115 (IQR 105, 121), p=0.022). Hypertensive disorders of pregnancy were more common among obese subjects than non-obese subjects (50.0% vs. 15.4%, p=0.010). Obese and non-obese subjects were similar in terms of third trimester hematocrit, pre-delivery hematocrit, mode of delivery, duration and type of labor, quantitative blood loss at delivery, and total intravenous fluids received in labor.
Table 2.
Comparison of NT-proBNP and logNT-proBNP by BMI category.
| BMI ≥ 30 (N=28) |
BMI < 30 (N=27) |
p-value | |
|---|---|---|---|
|
| |||
| 3rd trimester NT-proBNP (pg/mL) | 0.010 | ||
| Missing, n(%) | 0 (0) | 0 (0) | |
| Median (IQR) | 18 (6, 28) | 26 (17, 48) | |
| 3rd trimester logNT-proBNP | 0.004 | ||
| Mean (SD) | 2.6 (1.0) | 3.3 (0.9) | |
|
| |||
| Pre-delivery NT-proBNP (pg/mL) | 0.005 | ||
| Missing, n(%) | 3 (11) | 6 (22) | |
| Median (IQR) | 16 (2, 38) | 43 (21, 60) | |
| Pre-delivery logNT-proBNP | 0.003 | ||
| Mean (SD) | 2.6 (1.0) | 3.6 (0.9) | |
|
| |||
| Immediate postpartum NT-proBNP (pg/mL) | 0.294 | ||
| Missing, n(%) | 2 (7) | 3 (11) | |
| Median (IQR) | 58 (20, 102) | 63 (38, 154) | |
| Immediate postpartum logNT-proBNP | 0.226 | ||
| Mean (SD) | 3.8 (1.2) | 4.2 (1.1) | |
|
| |||
| 4–6 weeks postpartum NT-proBNP (pg/mL) | 0.149 | ||
| Missing, n(%) | 10 (36) | 9 (33) | |
| Median (IQR) | 33 (27, 56) | 23 (8, 42) | |
| 4–6 weeks postpartum logNT-proBNP | 0.096 | ||
| Mean (SD) | 3.5 (0.9) | 2.9 (1.2) | |
BMI = body mass index; SD = standard deviation; IQR = interquartile range
Footnote: A comparison of NT-proBNP levels in pg/mL by BMI category is demonstrated here at the four study timepoints (3rd trimester, pre- delivery, immediate postpartum, and 4–6 weeks postpartum). NT-proBNP levels are provided for clinical correlation.
For the primary outcome, both NT-proBNP and log-transformed NT-proBNP levels are demonstrated in Table 3. The spaghetti plot in Figure 2 visually depicts the trend in NT-proBNP levels over the course of pregnancy and postpartum in obese and non-obese subjects. The most prominent trend is a peak in NT-proBNP levels immediately postpartum, followed by a decrease at 4–6 weeks postpartum. Levels remained below the upper limit of normal for the majority of subjects. For obese subjects compared to non-obese subjects, median (IQR) NT-proBNP levels were 18 (6, 28) vs. 26 (17, 48) pg/mL (p=0.010) in the 3rd trimester, 16 (3, 38) vs. 43 (21, 60) pg/mL (p=0.005) pre-delivery, 58 (20, 102) vs. 63 (38, 55) pg/mL (p=0.294) at the IPP timepoint, and 33 (27, 56) vs. 23 (8, 42) pg/mL (p=0.149) at the DPP timepoint.
Table 3.
Log-transformed NT-proBNP levels varied by BMI (< 30 and ≥ 30) and study timepoint.
| Group | Comparisons | Unadjusted | Adjusteda | ||
|---|---|---|---|---|---|
|
| |||||
| β (95% CI) | P-value | β (95% CI) | P-value | ||
|
| |||||
| 3rd trimester | ≥ 30 vs < 30 | −0.78 (−1.35, −0.21) | 0.007 | −0.89 (−1.51, −0.26) | 0.006 |
| Pre-delivery | ≥ 30 vs < 30 | −0.97 (−1.59, −0.36) | 0.002 | −1.05 (−1.72, −0.38) | 0.002 |
| Immediate postpartum | ≥ 30 vs < 30 | −0.42 (−1.01, 0.17) | 0.164 | −0.52 (−1.16, 0.13) | 0.114 |
| 4–6 weeks postpartum | ≥ 30 vs < 30 | 0.74 (0.05, 1.42) | 0.036 | 0.65 (−0.07, 1.38) | 0.076 |
|
| |||||
| BMI ≥ 30 | Delivery vs 3rd trimester | 0.0001 (−0.49, 0.50) | 1.000 | 0.01 (−0.49, 0.51) | 0.963 |
| Immediate postpartum vs 3rd trimester | 1.24 (0.75, 1.73) | <0.001 | 1.24 (0.75, 1.73) | <0.001 | |
| 4–6 weeks postpartum vs 3rd trimester | 1.07 (0.52, 1.62) | <0.001 | 1.08 (0.52, 1.63) | <0.001 | |
|
| |||||
| BMI < 30 | Delivery vs 3rd trimester | 0.19 (−0.34, 0.72) | 0.479 | 0.18 (−0.36, 0.71) | 0.513 |
| Immediate postpartum vs 3rd trimester | 0.88 (0.37, 1.38) | 0.001 | 0.87 (0.36, 1.38) | 0.001 | |
| 4–6 weeks postpartum vs 3rd trimester | −0.45 (−1.00, 0.11) | 0.113 | −0.46 (−1.02, 0.10) | 0.103 | |
Model adjusted for race and gestational hypertension or preeclampsia.
BMI = body mass index
Footnote: Differences in the primary outcome were compared using a linear mixed effects model. The β coefficient represents the adjusted and unadjusted difference in logNT-proBNP levels between each comparison group as indicated. A negative β coefficient indicates lower logNT-proBNP levels and a positive β coefficient indicates higher logNT-proBNP levels. The unadjusted model includes obese versus non-obese, study timepoints, and their interaction terms. The adjusted model also includes race and hypertensive disorders of pregnancy.
Figure 2. Spaghetti plot of NT-proBNP in pg/mL by BMI category and pregnancy timepoint.
This spaghetti plot shows the absolute values of NT-proBNP for all 55 patients over the course of the 4 study visits categorized by BMI. The gray lines represent subjects with BMI <30 and the black lines represent subjects with BMI ≥30. There appears to be a peak in NT-proBNP levels in the immediate postpartum period and a subsequent decrease at 4–6 weeks postpartum.
The results for the unadjusted and adjusted linear mixed effects models are shown in Table 3. After adjustment for race and hypertensive disorders of pregnancy, logNT-proBNP was significantly lower at the 3rd trimester (β=−0.89, 95% CI −1.51, −0.26; p=0.006) and pre-delivery timepoints (β=−1.05, 95% CI −1.72, −0.38; p=0.002), did not differ significantly IPP (β=−0.52, 95% CI −1.16, 0.13; p=0.114), and trended towards higher levels at the DPP timepoint (β=0.65, 95% CI −0.07, 1.38; p=0.076). In this model, the interaction term between BMI category and study timepoint was statistically significant (p<0.001), demonstrating that logNT-proBNP levels varied by study timepoint differently among obese and non-obese subjects. While logNT-proBNP levels were significantly higher IPP compared to the 3rd trimester in non-obese (β=0.87, 95% CI 0.36, 1.38; p=0.001) and obese (β=1.24, 95% CI 0.75, 1.73; p<0.001) subjects, this elevation persisted to 4–6 weeks postpartum in obese subjects only (β=1.08, 95% CI 0.52, 1.63; p<0.001).
Discussion
In this study, significantly lower NT-proBNP levels were demonstrated during pregnancy, but not postpartum, in otherwise healthy obese compared to non-obese people. NT-proBNP levels increased in both obese and non-obese subjects in the immediate 24–48 hours postpartum. Examination of NT-proBNP trends over time during and after pregnancy stratified by obesity suggests that NT-proBNP levels return to baseline or lower by 4–6 weeks postpartum among non-obese, but not obese, parturients. NT-proBNP levels remain elevated at 4–6 weeks postpartum compared to the third trimester baseline only among obese people. This study provides novel information about the impact of obesity on NT-proBNP levels across time during and after pregnancy.
As markers of ventricular dysfunction, NT-proBNP and BNP cardiac biomarkers are useful in distinguishing cardiac causes of dyspnea from non-cardiac causes and in monitoring cardiac disease progression and response to therapy.10 An NT-proBNP threshold of 125 pg/mL has been proposed by the American Heart Asssociation/American College of Cardiology and the European Society of Cardiology as the threshold above which further evaluation for heart failure should be conducted.11,12 The sensitivity of an NT-proBNP level >125 pg/mL for the detection of heart failure is 98%, suggesting a low chance of a false negative result with the use of this threshold.13 Median NT-proBNP levels in this study remained below the recommended threshold at all timepoints, supporting the use of a similar threshold for concern in pregnancy. Within subject variations over time are included in Figure 2 and demonstrate that some individuals did exceed this threshold.
Outside of pregnancy, NT-proBNP and BNP levels are significantly lower in obese patients both with and without heart failure.7,14–16 The use of a single, weight-blind cutoff may result in lower sensitivity for the diagnosis of heart failure or ventricular dysfunction in obese patients.17 The 2022 American Heart Association/American College of Cardiology/Heart Failure Society of America Guideline for the Management of Heart Failure suggests consideration of BMI in the interpretation of BNP and NT-proBNP levels in obese patients due to lower diagnostic sensitivity in this population.12 To date, trends in NT-proBNP levels related to obesity and pregnancy have not been examined. The findings from our study suggest that BMI should also be considered when interpreting NT-proBNP levels in pregnancy and a lower threshold for concern may be warranted.
In addition to comparing NT-proBNP levels between obese and non-obese pregnant subjects, trends in NT-proBNP levels over the course of the third trimester and postpartum period were also examined. NT-proBNP levels did not vary significantly from the early third trimester to delivery. However, both obese and non-obese pregnant people experienced a rise in NT-proBNP levels in the immediate postpartum period, consistent with previous studies.5 This increase is likely due to increased left ventricular volume in the setting of auto-transfusion that occurs immediately following delivery.5
Two interesting trends were observed in the postpartum period. First, in the immediate postpartum period, a trend towards lower NT-proBNP levels in obese subjects compared to non-obese subjects was noted, possibly reflecting the blunted NT-proBNP response seen in non-pregnant obese individuals in the setting of ventricular overload. Second, obese, but not non-obese, postpartum subjects had a prolonged elevation in their NT-proBNP levels out to 4–6 weeks postpartum. This finding could be attributed to the increased proportion of subjects in the obese cohort who experienced hypertensive disorders of pregnancy. However, there may also be an alternative explanation for this trend. It is possible that the persistent postpartum elevation of NT-proBNP in subjects with obesity is due to pre-pregnancy cardiac dysfunction in the setting of obesity, a phenomenon that has been proposed previously.18 Alternatively, a more prolonged return to baseline cardiac function may occur in obese compared to non-obese pregnant individuals, a phenomenon that has been observed in patients with peripartum cardiomyopathy.19 Future studies including pre-pregnancy assessment of cardiac function in obese individuals may help clarify whether postpartum findings are due to pre-existing cardiac dysfunction or pregnancy-induced changes.
We recognize several limitations to this study. First, this study did not correlate NT-proBNP levels with cardiac parameters on echocardiogram. This limitation circumscribes any conclusions that can be made regarding cardiac function in obese compared to non-obese women in pregnancy. Second, there was significant loss to follow-up, particularly later in the postpartum period, and the sample size did not meet the goal of 56 due to research restrictions in place during the COVID-19 pandemic. The use of the mixed effects model mitigated some of the effects of the loss-to-follow-up, however, as this model utilizes all available values for each individual subject. Lastly, this was not a randomized controlled trial and there was unequal representation of different racial groups in each cohort. The inverse relationship between BMI and brain natriuretic peptide levels has been confirmed in both Black20 and White populations.6–8,15,21 This may, at least in part, reflect racial disparities in the prevalence of obesity in the state of North Carolina, where this study was performed. In North Carolina, an estimated 44.8% of Black adults self-report obesity, compared to 29.9% of White adults, according to data from the Behavioral Risk Factor Surveillance System.22
In summary, NT-proBNP levels during pregnancy were different between obese and non-obese subjects, with obese subjects having significantly lower NT-proBNP levels than non-obese subjects. This finding suggests that providers may need to consider a lower threshold for abnormal NT-proBNP serum levels when evaluating obese patients in pregnancy. Moreover, NT-proBNP levels remained persistently elevated at 4–6 weeks postpartum compared to the 3rd trimester among obese, but not non-obese, subjects. The different temporal trends in NT-proBNP levels between obese and non-obese people in the postpartum period suggest that cardiac remodeling postpartum may be more prolonged in obese people. Further studies are needed to correlate these findings with echocardiographic parameters and to determine a sensitive and specific NT-proBNP threshold indicative of heart failure.
Key points:
NT-proBNP levels are lower in obese than non-obese subjects during pregnancy
Levels remain elevated in obese, but not non-obese, subjects up to 4–6 weeks’ postpartum
A lower threshold for concern regarding NT-proBNP levels may be needed in obese pregnant people
Acknowledgements:
This study was generously supported by funding from the Duke Department of Obstetrics and Gynecology Hammond Research Fund. The Duke BERD Methods Core’s support of this project was made possible in part by CTSA Grant (UL1TR002553) from the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH), and the NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NCATS or NIH.
Footnotes
Disclosures: The authors report no conflicts of interest.
Clinical Trial Registration:
URL: https://clinicaltrials.gov/ct2/show/NCT04049136
Unique identifier: NCT04049136
Conference presentations: This manuscript was presented in poster format at the Society for Maternal-Fetal Medicine’s 41st Annual Pregnancy Meeting (virtual), which was held January 25–30th, 2021.
References
- 1.Nakagawa O, Ogawa Y, Itoh H, et al. Rapid transcriptional activation and early mRNA turnover of brain natriuretic peptide in cardiocyte hypertrophy. Evidence for brain natriuretic peptide as an “emergency” cardiac hormone against ventricular overload. J Clin Invest. 1995;96(3):1280–1287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Yancy CW, Jessup M, Bozkurt B, et al. 2017 ACC/AHA/HFSA Focused Update of the 2013 ACCF/AHA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. Circulation. 2017;136(6):e137–e161. [DOI] [PubMed] [Google Scholar]
- 3.Barasa A, Rosengren A, Sandstrom TZ, Ladfors L, Schaufelberger M. Heart Failure in Late Pregnancy and Postpartum: Incidence and Long-Term Mortality in Sweden From 1997 to 2010. J Card Fail. 2017;23(5):370–378. [DOI] [PubMed] [Google Scholar]
- 4.Tanous D, Siu SC, Mason J, et al. B-Type Natriuretic Peptide in Pregnant Women With Heart Disease. Journal of the American College of Cardiology. 2010;56(15):1247–1253. [DOI] [PubMed] [Google Scholar]
- 5.Burlingame JM, Yamasato K, Ahn HJ, Seto T, Tang WHW. B-type natriuretic peptide and echocardiography reflect volume changes during pregnancy. Journal of Perinatal Medicine. 2017;45(5):577–583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mehra MR, Uber PA, Park MH, et al. Obesity and suppressed B-type natriuretic peptide levels in heart failure. Journal of the American College of Cardiology. 2004;43(9):1590–1595. [DOI] [PubMed] [Google Scholar]
- 7.Daniels LB, Clopton P, Bhalla V, et al. How obesity affects the cut-points for B-type natriuretic peptide in the diagnosis of acute heart failure Results from the Breathing Not Properly Multinational Study. American Heart Journal. 2006;151(5):999–1005. [DOI] [PubMed] [Google Scholar]
- 8.Wang TJ, Larson MG, Levy D, et al. Impact of Obesity on Plasma Natriuretic Peptide Levels. Circulation. 2004;109(5):594–600. [DOI] [PubMed] [Google Scholar]
- 9.Yoshimura T, Yoshimura M, Yasue H, et al. Plasma concentration of atrial natriuretic peptide and brain natriuretic peptide during normal human pregnancy and the postpartum period. Journal of Endocrinology. 1994;140(3):393–397. [DOI] [PubMed] [Google Scholar]
- 10.Madamanchi C, Alhosaini H, Sumida A, Runge MS. Obesity and natriuretic peptides, BNP and NT-proBNP: Mechanisms and diagnostic implications for heart failure. International Journal of Cardiology. 2014;176(3):611–617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.McDonagh TA, Metra M, Adamo M, et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021;42(36):3599–3726. [DOI] [PubMed] [Google Scholar]
- 12.Heidenreich P, Bozkurt B, Aguilar D, et al. 2022. AHA/ACC/HFSA Guideline for the Management of Heart Failure: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. (1558–3597 (Electronic)). [DOI] [PubMed] [Google Scholar]
- 13.Zaphiriou A, Robb S Fau - Murray-Thomas T, Murray-Thomas T Fau - Mendez G, et al. The diagnostic accuracy of plasma BNP and NTproBNP in patients referred from primary care with suspected heart failure: results of the UK natriuretic peptide study. (1388–9842 (Print)). [DOI] [PubMed] [Google Scholar]
- 14.Batal O, Faulx M, Krasuski RA, et al. Effect of Obesity on B-Type Natriuretic Peptide Levels in Patients With Pulmonary Arterial Hypertension. The American Journal of Cardiology. 2012;110(6):909–914. [DOI] [PubMed] [Google Scholar]
- 15.Bayes-Genis A, Lloyd-Jones DM, Kimmenade RRJv, et al. Effect of Body Mass Index on Diagnostic and Prognostic Usefulness of Amino-Terminal Pro–Brain Natriuretic Peptide in Patients With Acute Dyspnea. Archives of Internal Medicine. 2007;167(4):400–407. [DOI] [PubMed] [Google Scholar]
- 16.Khalid U, Wruck LM, Quibrera PM, et al. BNP and obesity in acute decompensated heart failure with preserved vs. reduced ejection fraction: The Atherosclerosis Risk in Communities Surveillance Study. International Journal of Cardiology. 2017;233:61–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Krauser DG, Lloyd-Jones DM, Chae CU, et al. Effect of body mass index on natriuretic peptide levels in patients with acute congestive heart failure: A ProBNP Investigation of Dyspnea in the Emergency Department (PRIDE) substudy. American Heart Journal. 2005;149(4):744–750. [DOI] [PubMed] [Google Scholar]
- 18.Giorgione VA-O, O’Driscoll J, Coutinho CA-O, et al. Peripartum echocardiographic changes in women with hypertensive disorders of pregnancy. (1469–0705 (Electronic)). [DOI] [PubMed] [Google Scholar]
- 19.Davis EM, Ewald G, Givertz MM, et al. Maternal Obesity Affects Cardiac Remodeling and Recovery in Women with Peripartum Cardiomyopathy. (1098–8785 (Electronic)). [DOI] [PubMed] [Google Scholar]
- 20.Fox ER, Musani SK, Bidulescu A, et al. Relation of Obesity to Circulating B-Type Natriuretic Peptide Concentrations in Blacks. Circulation. 2011;124(9):1021–1027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Das SR, Drazner MH, Dries DL, et al. Impact of body mass and body composition on circulating levels of natriuretic peptides: results from the Dallas Heart Study. Circulation. 2005;112(14):2163–2168. [DOI] [PubMed] [Google Scholar]
- 22.Centers for Disease Control and Prevention. Adult Obesity Prevalence Maps. https://www.cdc.gov/obesity/data/prevalence-maps.html. Published 2020. Updated 09/21/2020. Accessed 01/26/2021, 2021.