Objective
The purpose of this study was to determine, with the use of cardiac magnetic resonance imaging, whether there is vertical displacement of the heart during pregnancy. Cardiopulmonary resuscitation guidelines during pregnancy recommend placing the hands 2-3 cm higher on the sternum than in nonpregnant individuals. This recommendation is based on the presumption that the heart is displaced superiorly by the diaphragm during the third trimester. Whether there is true cardiac displacement because of the expanding uterus in pregnancy remains unknown.
Study Design
A total of 34 healthy female volunteers 18-35 years old were enrolled prospectively from 2010-2012 at 2 tertiary care centers. The conditions of all participants were evaluated with cardiac magnetic resonance imaging in the one-half left lateral decubitus position during the third trimester of pregnancy and again at a minimum of 3 months after delivery (surrogate for the nonpregnant state). Superior displacement of the heart was determined by measurement of the distance between the inferior aspect of the clavicular heads and the coronary sinus at both time points.
Results
The study population included 34 women (mean age, 29 ± 3 years; body mass index, 24 ± 4 kg/m 2 ). The mean gestational age at third-trimester imaging was 237 ± 16 days (34 weeks ± 16 days); the mean number of days for postpartum imaging (baseline) was 107 ± 25 days (16 weeks ± 25 days). There was no statistical difference between the cardiac position at baseline (10.1 ± 1.2 cm) and during the third trimester (10.3 ± 1.1 cm; P = .22).
Conclusion
Contrary to popular assumption, there is no significant vertical displacement of the heart in the third trimester of pregnancy relative to the nonpregnant state. Accordingly, there is no need to alter hand placement for chest compressions during cardiopulmonary resuscitation in pregnancy.
Cardiac arrest in late pregnancy is an uncommon event that has been estimated to affect between 1:20,000 and 1:30,000 cases. However, cardiac disease in general represents the most common cause of increased maternal mortality rate in the developed world; it recently has overtaken maternal hemorrhage and now leads to more maternal deaths than preeclampsia, eclampsia, thrombosis, and/or thromboembolism combined. In addition, the incidence of maternal death from cardiac disease is on the rise because the average maternal age continues to increase. One recent review found that cardiac arrest now complicates up to 1:12,000 hospital admissions for delivery. Management of cardiac arrest in late pregnancy presents unique challenges and requires simultaneous care of both the mother and fetus and adjustment to the physiologic changes that occur during pregnancy. Maternal survival after cardiac arrest and attempted resuscitation has been reported at 15% in 1 case series, including in and out of hospital cardiac arrest, and as high as 59% in the in-hospital setting. The variability in maternal survival after cardiac arrest and the potential years of life gained by successful resuscitation highlights the need for optimizing resuscitation in late pregnancy.
Closed-chest cardiopulmonary resuscitation (CPR) was first described in 1960; the first set of CPR guidelines was released by the American Heart Association in 1966. Over the past 50 years, there has been an ongoing process of updating and refining these guidelines as new evidence has become available. The most recent guidelines, released in 2010, describe several modifications to resuscitation in pregnant patients that include modification in basic life support positioning and technique. The recommendations include adjustment for elevation of the diaphragm that is caused by the gravid uterus by placing the hands slightly higher on the sternum than normally recommended, although the degree of adjustment is not specified. It is also recommended to perform chest compressions in either the left lateral tilt position or to use manual leftward uterine displacement in the supine position to relieve aortocaval compression. Although there is a large body of evidence to support leftward uterine displacement during chest compressions, there is no corresponding evidence for the alteration of hand placement, which recently was stressed as an area that requires further research by the Society for Obstetric Anesthesia and Perinatology.
Current recommendations for hand placement for chest compressions in nonpregnant patients are that rescuers should place their hands in the middle of the lower one-half of the sternum, but that this location should be taught with the use of the more simplified “center of the chest.” Recently, the increased availability and usage of cross-sectional imaging has permitted noninvasive identification of lower substernal structures in the general population. However, no contemporary research that has identified changes in cardiac position with late pregnancy is available. Although previous research that has used cross-sectional imaging to correlate CPR hand placement with underlying anatomy has been performed with computed tomography, cardiac magnetic resonance imaging (CMR) permits safe imaging in all trimesters of pregnancy. The objective of the present study was to use CMR to measure vertical cardiac displacement objectively in late pregnancy.
Materials and Methods
A prospective cohort study was conducted with healthy volunteers recruited from prenatal clinics at 2 tertiary care centers. Eligibility criteria included age 18-35 years old at last normal menstrual period (LNMP), a healthy singleton pregnancy, and no previous pregnancy carried >14 weeks’ gestation. Exclusion criteria included any history of thyroid disease, hypertension, or diabetes mellitus; any history of cardiac disease or cardiac surgery; any significant pregnancy-related complication such as gestational hypertension, gestational diabetes mellitus, or preeclampsia; any contraindication to CMR that included pacemakers or surgical clips; and multiple gestation. The study protocol was approved by the University of Manitoba Human Research Ethics Board, and informed consent was obtained from all study participants.
Data were collected from Nov. 22, 2010, to Oct. 11, 2012, inclusive. CMR imaging was performed at 2 time points for all participants: (1) baseline images were obtained in the third trimester, and (2) additional images were acquired approximately 3 months after delivery as a surrogate for the nonpregnant state.
CMR scans were performed with a 1.5 Tesla Siemens Scanner (Avanto, Magnetom, Siemens, Erlangen, Germany). All patients were imaged in the one-half left lateral decubitus position at both time points in correspondence with the recommendation that CPR be performed in the left lateral tilt position or with leftward uterine displacement in third-trimester pregnancy to minimize aortocaval compression. An axial dark blood single-shot fast spin echo sequence (slice thickness, 6 mm) was performed with electrocardiographic gating to obtain images through the chest to the level of the diaphragm. No sedative or contrast agents were administered. To enable comparison between the 2 study time points, cardiac position was defined for the purposes of the present study as the distance between the easily reproducible landmarks of the inferior aspect of the clavicular heads and the coronary sinus ( Figure ). The clavicular heads were selected as an easily identifiable landmark that relates to the superior sternum. More inferior sternal landmarks, such as the xiphoid, were not included because of concerns regarding displacement with chest expansion during pregnancy. The coronary sinus was selected as a relatively inferior landmark with less mobility throughout the cardiac cycle than other potential inferior cardiac landmarks, such as the ventricular apex. Cardiac displacement was defined as the difference between the cardiac position in the nonpregnant state and cardiac position during the third trimester, for which positive values indicate superior displacement in pregnancy and negative values indicate inferior displacement in pregnancy.
Measurements were performed initially on a subset of 5 patients by both the staff cardiac radiologist coauthor (I.D.C.K.), who is a fellowship-trained cardiac and body radiologist with 10 years of practice experience, and by the radiology resident coauthor (S.H.). There was 100% concordance between measurements in this subset, and the remainder of the measurements were performed by the resident reviewer. Reviewers were not blinded to the antepartum or postpartum status of participants because of the obvious anatomic changes visible on acquired images.
All parametric data were reported as mean ± standard deviation (SD). Categoric data were reported as number (percentage). Comparison of nonpregnant and third-trimester cardiac position was performed with a paired samples 2-tail Student t test. Correlations between subject characteristics and cardiac position and displacement were performed with the Pearson correlation coefficient rather than with a regression model to maximize sensitivity to possible confounding variables. Statistical significance was defined at a probability value of < .05. SAS software (version 9.0; SAS Institute Inc, Cary, NC) was used to perform the analysis.
Results
The baseline and delivery-related clinical characteristics of the study population (n = 34 women) are summarized in Table 1 . The mean age at LNMP was 29 ± 3 years; the mean prepregnancy body mass index (BMI) of 24 ± 4 kg/m 2 . The mean gestational age at third-trimester imaging was 237 ± 16 days (33 weeks 6 days ± 16 days), and the mean number of days after delivery for baseline imaging was 107 ± 25 days. Mean gestational age at delivery was 278 ± 9 days (39 weeks 5 days ± 9 days).
| Characteristic | Mean ± SD |
|---|---|
| Maternal age at last normal menstrual period, y | 29 ± 3 |
| Prepregnancy body mass index, kg/m 2 | 24 ± 4 |
| Gestational age, d | |
| Third-trimester imaging | 237 ± 16 |
| Delivery | 278 ± 9 |
| Days after delivery at baseline imaging, d | 107 ± 25 |
Cardiac position and cardiac displacement parameters are outlined in Table 2 . In the third trimester and 3 months after delivery, the mean cardiac positions were 10.1 ± 1.2 cm and 10.3 ± 1.1 cm, respectively. The difference in cardiac position was not statistically significant ( P = .22).
| Variable | Data | |
|---|---|---|
| Mean ± SD, mm | 95% confidence interval | |
| Baseline (nonpregnant) cardiac position a | 103 ± 11 | 100–107 |
| Third-trimester cardiac position | 101 ± 13 | 97–106 |
| Cardiac displacement b | 3 ± 12 | −1–6 |
a Defined as the distance between the inferior aspect of the clavicular heads and the coronary sinus
b Defined as the difference between the cardiac position in the baseline state and in the third trimester, where positive values indicate superior displacement.
Correlations between subject characteristics and cardiac position and displacement are summarized in Table 3 . Cardiac displacement was not correlated significantly with gestation age at third trimester CMR (r = 0.14; P = .45) or days after delivery at baseline CMR (r = 0.12; P = .54). Cardiac displacement was also not significantly correlated with participant height r = –0.05; P = .77), age at LNMP (r = 0.16; P = .41), weight gain in pregnancy (r = 0.03; P = .90), or infant birthweight (r = 0.24; P = .20). Height was correlated significantly with both nonpregnant cardiac position (r = 0.38; P = .04) and third-trimester cardiac position (r = 0.36; P = .05). However, prepregnancy BMI was not significantly correlated with nonpregnant cardiac position (r = 0.29; P = .12), and predelivery BMI was not correlated significantly with third-trimester cardiac position (r = -0.09; P = .64). Nonpregnant cardiac position was correlated significantly with third-trimester cardiac position (r = 0.51; P = .004).
