Hepatic venous Doppler waveforms: changes in pregnancy

Maternal posture-physiology interactions in human pregnancy: a narrative review

There are several well-known medical conditions in which posture and gravity interact with natural history, including pregnancy. In this review, we provide a comprehensive overview of interactions between maternal posture and maternal physiology and pathophysiology at rest during pregnancy. We conducted a systematic literature search of the MEDLINE database and identified 644 studies from 1991 through 2021, inclusive, that met our inclusion criteria. We present a narrative review of the resulting literature and highlight discrepancies, research gaps, and potential clinical implications. We organize the results by organ system and, commencing with the neurological system, proceed in our synthesis generally in the craniocaudal direction, concluding with the skin. The circulatory system warranted our greatest and closest consideration-literature concerning the dynamic interplay between physiology (heart rate, stroke volume, cardiac output, blood pressure, and systemic vascular resistance), pathophysiology (e.g., hypertension in pregnancy), and postural changes provide an intricate and fascinating example of the importance of the subject of this review. Other organ systems discussed include respiratory, renal, genitourinary, gastrointestinal, abdominal, and endocrine. In addition to summarizing the existing literature on maternal posture-physiology interactions, we also point out gaps and opportunities for further research and clinical developments in this area. Overall, our review provides both insight into and relevance of maternal posture-physiology interactions vis à vis healthcare's mission to improve health and wellness during pregnancy and beyond.

Preeclampsia: a gestational cardiorenal syndrome

It is generally accepted today that there are two different types of preeclampsia: an early-onset or placental type and a late-onset or maternal type. In the latent phase, the first one presents with a low output/high resistance circulation eventually leading in the late second or early third trimester to an intense and acutely aggravating systemic disorder with an important impact on maternal and neonatal mortality and morbidity; the other type presents initially as a high volume/low resistance circulation, gradually evolving to a state of circulatory decompensation usually in the later stages of pregnancy, with a less severe impact on maternal and neonatal outcome. For both processes, numerous dysfunctions of the heart, kidneys, arteries, veins and interconnecting systems are reported, most of them presenting earlier and more severely in early- than in late-onset preeclampsia; however, some very specific dysfunctions exist for either type. Experimental, clinical and epidemiological observations before, during and after pregnancy are consistent with gestation-induced worsening of subclinical pre-existing chronic cardiovascular dysfunction in early-onset preeclampsia, and thus sharing the pathophysiology of cardiorenal syndrome type II, and with acute volume overload decompensation of the maternal circulation in late-onset preeclampsia, thus sharing the pathophysiology of cardiorenal syndrome type 1. Cardiorenal syndrome type V is consistent with the process of preeclampsia superimposed upon clinical cardiovascular and/or renal disease, alone or as part of a systemic disorder. This review focuses on the specific differences in haemodynamic dysfunctions between the two types of preeclampsia, with special emphasis on the interorgan interactions between heart and kidneys, introducing the theoretical concept that the pathophysiological processes of preeclampsia can be regarded as the gestational manifestations of cardiorenal syndromes.

Maternal Venous Hemodynamic Dysfunction in Proteinuric Gestational Hypertension: Evidence and Implications

This review summarizes current knowledge from experimental and clinical studies on renal function and venous hemodynamics in normal pregnancy, in gestational hypertension (GH) and in two types of preeclampsia: placental or early-onset preeclampsia (EPE) and maternal or late-onset (LPE) preeclampsia, presenting at <34 weeks and ≥34 weeks respectively. In addition, data from maternal venous Doppler studies are summarized, showing evidence for (1) the maternal circulation functioning closer to the upper limits of capacitance than in non-pregnant conditions, with intrinsic risks for volume overload, (2) abnormal venous Doppler measurements obtainable in preeclampsia, more pronounced in EPE than LPE, however not observed in GH, and (3) abnormal venous hemodynamic function installing gradually from first to third trimester within unique pathways of general circulatory deterioration in GH, EPE and LPE. These associations have important clinical implications in terms of screening, diagnosis, prevention and management of gestational hypertensive diseases. They invite for further hypothesis-driven research on the role of retrograde venous congestion in the etiology of preeclampsia-related organ dysfunctions and their absence in GH, and also challenge the generally accepted view of abnormal placentation as the primary cause of preeclampsia. The striking similarity between abnormal maternal venous Doppler flow patterns and those observed at the ductus venosus and other abdominal veins of the intra-uterine growth restricted fetus, also invites to explore the role of venous congestion in the intra-uterine programming of some adult diseases.

Renal Interlobar Vein Impedance Index as a First-Trimester Marker Does Not Predict Hypertensive Disorders of Pregnancy

OBJECTIVES

The purpose of this study was to examine whether the maternal renal interlobar vein impedance index as assessed by first-trimester sonography is able to predict the later development of hypertensive disorders of pregnancy.

METHODS

Venous Doppler parameters of both maternal kidneys were studied in 214 pregnant women at gestational ages of 11 weeks to 13 weeks 6 days. Patients were classified according to outcomes related to hypertensive disorders. Detection rates and areas under receiver operating characteristic curves were determined for the maternal renal interlobar vein impedance index as a first-trimester predictor of preeclampsia and gestational hypertension.

RESULTS

Among the 214 patients, 22 (10.3%) developed preeclampsia; 10 (4.7%) developed gestational hypertension; and 182 were unaffected by hypertensive disorders (controls; 85.0%). In the overall study population, there was no difference in the impedance index between the right (0.44; 95% confidence interval, 0.35-0.50) and left (0.43; 95% confidence interval, 0.35-0.53) sides (P = .86). The average impedance index did not differ among women destined to develop preeclampsia (0.46; 95% confidence interval, 0.38-0.57), gestational hypertension (0.39; 95% confidence interval, 0.33-0.46), or pregnancies uncomplicated by hypertensive disease (0.42; 95% confidence interval, 0.37-0.50; P = .15). Low detection rates and the area under the curve analysis demonstrated that the impedance index was not predictive of hypertensive disorders of pregnancy.

CONCLUSIONS

The maternal renal interlobar vein impedance index should not be considered a first-trimester marker of hypertensive disorders of pregnancy.

Laparoscopic Sleeve Gastrectomy in the Postpartum Period: Increased Risk of Bleeding, a Challenging Situation

The female population represents three-fourths of patients undergoing a bariatric procedure and could be scheduled for surgery in their postpartum period. We report a difficult case of a female patient who underwent a laparoscopic sleeve gastrectomy 6 weeks postpartum. The postpartum period is accompanied by pronounced vasodilatation with transient portal hypertension. Most of the hemodynamic alterations occurring during pregnancy return to baseline within 6-8 weeks after delivery. Bariatric surgery in the postpartum period should be avoided in order for the cardiovascular system to regain its normality.

Hepatic vein and portal vein Doppler ultrasound of maternal liver in normal pregnancy

OBJECTIVE

The purpose of our study was to evaluate the changes occurring in maximum flow velocity and waveform patterns of the portal and hepatic veins during pregnancy in second and third trimesters.

METHODS

A total of 204 patients were randomly divided into the study and control groups. The control group was classified as group I; the study group was divided into two groups: group II, second trimester pregnancy (14-26 weeks, n = 91) and group III, third trimester pregnancy (26-40 weeks, n = 63). Vein waveforms were classified as triphasic, biphasic and monophasic.

RESULTS

Maximum flow velocities of both right hepatic vein and main portal veins were lower in group III compared with group I and group II (p < 0.05). On the basis of hepatic vein and portal vein waveforms, the incidence of biphasic pattern was prominent in group II (p < 0.05) whereas the monophasic pattern was prominent in group III (p < 0.05).

CONCLUSION

The results demonstrate that as gestational age progresses, maximum flow velocities of both right hepatic vein and main portal veins intend to be decreased with an increased incidence of monophasic wave form pattern, and may be accepted as sensitive parameters of indicators of physiological adaptations related to pregnancy.

Spectral Doppler of the Hepatic Veins in Noncardiac Diseases: What the Echocardiographer Should Know

In most instances, the flow profile in the hepatic veins (HVs) reflects the fluctuation of pressure within the right atrium. Thus, interrogation of blood flow in the HVs is highly useful for the evaluation of right heart hemodynamics and has become an integral part of any routine echocardiographic examination. However, flow in the HVs is also affected by the state of the liver parenchyma and by the fluctuation of pressure within the thoracic cavity. Therefore, liver and pulmonary pathologies influence the flow pattern in the HVs and may lead to its dissociation from right heart hemodynamics. Echocardiographers should familiarize themselves with the findings on HV Doppler in noncardiac diseases to avoid misinterpretation and incorrect diagnosis.

Ultrasonographic study and Doppler flow velocimetry of maternal kidneys and liver in low-risk pregnancy

OBJECTIVE

Longitudinal study with B-mode ultrasonography and Doppler ultrasonography of maternal kidneys and liver in low-risk pregnancy, to establish and quantify normality parameters, correlating them with physiological changes.

MATERIALS AND METHODS

Twenty-five pregnant women were assessed and selected to participate in the study, each of them undergoing four examinations at the first, second, third trimesters and postpartum.

RESULTS

Findings during pregnancy were the following: increased renal volume, pyelocaliceal dilatation with incidence of 45.4% in the right kidney, and 9% in the left kidney; nephrolithiasis, 18.1% in the right kidney, 13.6% in the left kidney. With pyelocaliceal dilatation, mean values for resistivity index were: 0.68 for renal arteries; 0.66 for segmental arteries; 0.64 for interlobar arteries; 0.64 for arcuate arteries. Without pyelocaliceal dilatation, 0.67 for renal arteries; 0.64 for segmental arteries; 0.63 for interlobar arteries; and 0.61 for arcuate arteries. Portal vein flow velocities presented higher values in pregnancy, with mean value for maximum velocity of 28.9 cm/s, and 22.6 cm/s postpartum. The waveform pattern of the right hepatic vein presented changes persisting in the postpartum period in 31.8% of the patients. Cholelithiasis was observed in 18.1% of the patients.

CONCLUSION

Alterations in renal volume, pyelocaliceal dilatation, nephrolithiasis, cholelithiasis, changes in portal vein flow velocity, alterations in waveform pattern of the right hepatic vein, proved to be significant.

Is there a correlation between maternal venous hemodynamic dysfunction and proteinuria of preeclampsia?

OBJECTIVES

To evaluate in early and late preeclampsia a correlation of maternal venous Doppler flow characteristics with biochemical parameters in maternal serum and urine, or with gestational outcome.

STUDY DESIGN

In this observational cross-sectional study, renal interlobar vein impedance index (RIVI) was measured according to a standardised protocol for combined electrocardiogram-Doppler ultrasonography in 86 women with uncomplicated pregnancy, 78 women with late onset preeclampsia (≥34w) and 67 with early onset preeclampsia (<34w). For each group, maternal age, pre-gestational BMI and parity were recorded together with birth weight and -percentile. For both early onset and late onset preeclampsia, maternal serum was analysed for thrombocyte count and concentrations of creatinine, ASAT, ALAT and uric acid and 24h urine collections were analysed for creatinine clearance and proteinuria (mg/24h). A non-parametric Mann-Whitney U-tests was performed for continuous data and a Fisher's exact tests for categorical data. Significant linear dependence between variables was identified using Pearson's correlation coefficient at nominal level a=0.05.

RESULTS

Proteinuria was higher in early onset than in late onset preeclampsia (1756mg [838-6116mg] versus 877mg [416-1696mg], p<0.001), and this was also true for RIVI in both left (0.45 [0.40-0.55] versus 0.41 [0.35-0.45], p=0.001) and right kidney (0.45 [0.39-0.55] versus 0.38 [0.30-0.43], p<0.001). In our data set, there was a significant correlation between proteinuria and RIVI of left (correlation coefficient=0.172, p=0.036) and right kidney (correlation coefficient=0.218, p=0.009) in late onset but not early onset preeclampsia.

CONCLUSION

Maternal RIVI may correlate with proteinuria of late onset preeclampsia.

Intra-abdominal pressure measurements in term pregnancy and postpartum: an observational study

Objective

To determine intra-abdominal pressure (IAP) and to evaluate the reproducibility of IAP-measurements using the Foley Manometer Low Volume (FMLV) in term uncomplicated pregnancies before and after caesarean section (CS), relative to two different reference points and to non-pregnant values.

Design

Observational cohort study.

Setting

Secondary level referral center for feto-maternal medicine.

Population

Term uncomplicated pregnant women as the case-group and non-pregnant patients undergoing a laparoscopic assisted vaginal hysterectomy (LAVH) as control group.

Methods

IAP was measured in 23 term pregnant patients, before and after CS and in 27 women immediately after and 1 day after LAVH. The midaxillary line was used as zero-reference (IAP_MAL) in all patients and in 13 CS and 13 LAVH patients, the symphysis pubis (IAP_SP) was evaluated as additional zero-reference. Intraobserver correlation (ICC) was calculated for each zero-reference. Paired student's t-tests were performed to compare IAP values and Pearson's correlation was used to assess correlations between IAP and gestational variables.

Main outcome measures

ICC before and after surgery, IAP before and after CS, IAP after CS and LAVH.

Results

The ICC for IAP_MAL before CS was lower than after (0.71 versus 0.87). Both mean IAP_MAL and IAP_SP were significantly higher before CS than after: 14.0±2.6 mmHg versus 9.8±3.0 mmHg (p<0.0001) and 8.2±2.5 mmHg versus 3.5±1.9 mmHg (p = 0.010), respectively. After CS, IAP was not different from values measured in the LAVH-group.

Conclusion

IAP-measurements using FMLV is reproducible in pregnant women. Before CS, IAP is increased in the range of intra-abdominal hypertension for non-pregnant individuals. IAP significantly decreases to normal values after delivery.

Improving the reliability of venous Doppler flow measurements: relevance of combined ECG, training and repeated measures

The nature of venous Doppler waves is highly variable. An additional electrocardiogram (ECG) improves the interpretation of venous Doppler wave characteristics and allows measurement of venous pulse transit time. The purpose of this study was to assess the reproducibility of ECG-guided repeated measurements of venous Doppler flow characteristics before and after sonographer training and the inter- and intra-observer variability. In four groups of 25 healthy women, venous Doppler flow measurements were performed at the level of the kidneys and liver according to a standardized protocol. Intra-observer Pearson correlation coefficients of the renal interlobar vein Doppler indices were ≥ 0.80 with the addition of the ECG, which are higher than the results of a former study. The inter-observer correlation between an experienced ultrasonographer and an inexperienced ultrasonographer improved from ≥ 0.71 to ≥ 0.91 after training. The correlation range of all parameters between two independent observers improved when values were based on repeated measures. The addition of an ECG to the Doppler image, training and repeated measurements are helpful in improving venous Doppler wave interpretation.

Role of dysfunctional maternal venous hemodynamics in the pathophysiology of pre-eclampsia: a review

The venous compartment has an important function in regulation and control of cardiac output. Abnormalities of cardiac output have been found in early gestational stages of both early- and late-onset pre-eclampsia. The venous compartment also maintains the balance between circulating and non-circulating blood volumes and regulates the amount of reserve blood stored in the splanchnic venous bed. It is well known that adaptive regulation of maternal blood volume is disturbed in pre-eclampsia. Abnormal venous hemodynamics and venous congestion are responsible for secondary dysfunction of several organs, such as the kidneys in cardiorenal syndrome and the liver in cardiac cirrhosis. Renal and liver dysfunctions are among the most relevant clinical features of pre-eclampsia. Doppler sonography studies have shown that the maternal venous compartment is subject to gestational adaptation, and that blood flow characteristics at the level of renal interlobar and hepatic veins are different in pre-eclampsia compared with uncomplicated pregnancy. In comparison to late-onset pre-eclampsia, in early-onset pre-eclampsia venous Doppler flow abnormalities are more prominent and present up to weeks before clinical symptoms. This paper reviews the growing evidence that dysfunction of maternal venous hemodynamics is part of the pathophysiology of pre-eclampsia and may perhaps be more important than is currently considered. Doppler sonography is a safe and easily performed method with which to study maternal venous hemodynamics. Therefore, exploring the role of maternal venous hemodynamics using Doppler sonography is an exciting new research topic for those who are interested in cardiovascular background mechanisms, as well as prediction and clinical work-up of pre-eclampsia.

Understanding the spectral Doppler waveform of the hepatic veins in health and disease

Duplex Doppler sonography is a fundamental component of the complete ultrasonographic examination of the liver. Accurate interpretation of the spectral Doppler tracing from the hepatic veins is valuable, as it reflects important cardiac and hepatic physiology. Normally, there are four phases: A, S, V, and D; the S and D waves indicate flow in the antegrade direction toward the heart. In hepatic and cardiac disease, these normal waves may be absent, a finding indicative of flow in a nonphysiologic manner. In addition, transient patient factors such as phase of the respiratory cycle may influence the appearance of the spectral tracing. Familiarity with the normal and abnormal spectral Doppler waveforms from the hepatic veins and knowledge of their respective physiology and pathophysiology provide valuable insights. Systematic analysis of the direction, regularity, and phasicity of the spectral tracing and the ratio of the amplitudes of the S and D waves allows one to arrive at the correct differential diagnosis in most situations.

Uterine venous blood flow in normal and complicated pregnancies: a methodological study

OBJECTIVE

To investigate the possibility of recording Doppler flow signals from the maternal uterine veins (UtVs) during pregnancy and to assess the relationship between UtV signals and other Doppler parameters as well as pregnancy outcomes.

METHODS

Transabdominal Doppler ultrasound examination of the UtVs on both sides of the uterus was performed in 40 normal and 44 high-risk singleton pregnancies at 23-39 weeks' gestation. The UtV was identified using color Doppler imaging and the flow velocity signals of the UtV and uterine artery (UtA) were recorded. Morphological examination of the placenta was carried out in 45 of the pregnancies (14 uncomplicated and 31 high-risk pregnancies).

RESULTS

Flow-velocity signals of the UtVs were recorded from at least one side of the uterus in all patients (success rate of 81 and 89% for the right and left UtV, respectively). Three types of flow-velocity pattern were identified: continuous non-pulsatile flow (Type I, n = 70), pulsatile flow with a notch touching the zero line (Type II, n = 6) and pulsatile flow with absent flow signals for part of the heart cycle (Type III, n = 8). The UtA pulsatility index was significantly higher in women with UtV Types II and III than in those with UtV Type I (P = 0.039). Similarly, UtV Types II and III were more often found in pregnancies with bilateral UtA notching (P = 0.013) and with UtA score 3-4 (P = 0.024) than in those with normal UtA. No statistically significant association was found between the UtV flow pattern and abnormal histopathological findings in the placenta, or between the UtV and umbilical artery findings.

CONCLUSION

It is possible to record Doppler signals from the UtVs in the late second and third trimesters of pregnancy. Pulsatile flow-velocity patterns of the UtVs are associated with abnormal UtA Doppler findings.

Maternal hepatic vein Doppler velocimetry during uncomplicated pregnancy and pre-eclampsia

Changes of Doppler velocity measurements of distinct hepatic vein (HV) Doppler wave components were evaluated during uncomplicated pregnancy (UP) as a reference to measurements in pre-eclampsia (PE). Women with UP (n = 13) were submitted to standardised duplex scanning of HV at 11 stages of pregnancy between 10 and 38 weeks. For each stage, mean +/- SD was calculated for HV A-, X-, V- and Y-peaks. Women with PE (n = 30) were evaluated once, and mean +/- SD was calculated for pregnancies <32 weeks, 32-34(+6) weeks and > or =35 weeks. PE and UP values at corresponding gestational age were compared statistically using t-test. HV A-velocity measurements changed markedly from negative values in early uncomplicated pregnancy, converting around 22-24 weeks to positive values until term. Changes throughout gestation were less prominent for HV X-, V- and Y-velocities. HV A-velocity measurements were significantly lower in PE than in UP, the difference being more pronounced at 30 weeks (-3.59 +/- 3.41 vs. 6.12 +/- 3.43, p = 0.0001) than at 37 weeks (2.35 +/- 4.54 vs. 5.32 +/- 1.92, p = 0.04). From our results, we conclude that HV Doppler velocimetry shows a gradual shift from central venous reversed flow during atrial contraction in uncomplicated early pregnancy to constantly forward moving flow until term. HV A-velocities are significantly lower in PE than in UP, the differences being more pronounced in late second trimester than near term.

Renal vein Doppler ultrasound of maternal kidneys in normal second and third trimester pregnancy

The flow pattern in intrarenal veins depends on renal parenchymal histology and cardiac physiology. The intrarenal venous impedance index obtained by Doppler ultrasound is related to compliance in vein, and can be helpful in the assessment of renal parenchymal compliance. The purpose of this study was to determine whether normal pregnancy has a significant effect on intrarenal venous blood flow, and assess if the physiological pyelocaliectasis causes a measurable reduction in venous impedance indexes in pregnant women. Doppler ultrasound of intrarenal veins was performed in 35 asymptomatic pregnant women in the second and third trimester of gestation, and in 24 non-pregnant healthy women. After grading the degree of hydronephrosis, venous impedance index was obtained from the interlobar veins. The venous waveforms in pregnant women showed diminished phasic oscillations owing to elevated pre-systolic flow. The mean venous impedance indexes in pregnant women were significantly lower than the values in non-pregnant subjects, 0.30+/-0.10 versus 0.44+/-0.06 in the right (p<0.001), and 0.36+/-0.11 versus 0.41+/-0.07 in the left kidney (p=0.03). There was an inverse correlation between the grade of pelvicalyceal dilatation and the venous impedance indexes in both kidneys in pregnant women (r=-0.62, p<0.001 for the right kidney, and r=-0.38, p=0.05 for the left kidney). An abnormally reduced venous impedance index in pregnant women can at least in part be explained by reduced vascular compliance from increased interstitial pressure subsequent to partial obstruction of ureters by the gravid uterus, and caution should be exercised in interpreting it as a sign of pathological ureteral obstruction.

Association between duplex Doppler sonographic flow pattern in right hepatic vein and various liver diseases

PURPOSE

The aim of this study was to evaluate the association between the Doppler sonographic waveforms in the right hepatic vein and various liver diseases.

METHODS

We performed Doppler sonography of the right hepatic vein in 225 individuals (189 patients with liver disease and 36 control subjects). Patients were categorized on the basis of their histologic diagnosis: cirrhosis (n = 122), fibrosis (n = 23), fatty liver disease (n = 11), metastatic liver disease (n = 8), and noncirrhotic liver disease with neither fatty infiltration nor metastases (n = 25). The waveforms of the hepatic veins were categorized as type 1, triphasic; type 2, biphasic; or type 3, flat.

RESULTS

Type 1 waveforms were found in 100% of the control subjects and in patients with each of the various liver diseases, including 40% of those with cirrhosis. Type 2 waveforms were not found in control subjects or in patients with either fatty infiltration or metastatic liver disease. Type 2 waveforms were, however, found in 16% of patients with cirrhosis, 13% of patients with fibrosis, and 4% of patients with other noncirrhotic liver diseases. Type 3 waveforms were found in 43% of patients with liver cirrhosis, 13% of patients with liver fibrosis, 27% of patients with fatty liver disease, and 50% of patients with metastatic liver disease. In contrast, type 3 waveforms were not found in any control subjects or in patients with other noncirrhotic liver diseases without fatty liver or metastases.

CONCLUSIONS

Type 3 waveforms in the right hepatic vein are suggestive of liver cirrhosis, although they may also occur in patients with fibrosis, fatty liver disease, or metastatic liver disease.

Evaluation of hepatic venous pulsatility and portal venous velocity with Doppler ultrasonography during the puerperium

OBJECTIVE

The aim of this study is to evaluate pregnancy-induced changes of hepatic venous pulsatility and portal venous velocity in the puerperium and to determine if these changes disappeared by the end of the puerperium.

METHODS AND MATERIAL

Healthy normal volunteers (90) were examined on the 2nd and 7th days of puerperium and between the 6th and 8th weeks postpartum. Doppler waveform patterns were obtained in the middle hepatic vein and main portal vein. The hepatic venous pulsatility was named as normal, damped or flat.

RESULTS

On the 2nd day postpartum, the hepatic vein pulsatility was shown as normal in 8 (26%), damped in 11 (37%) and flat in 11 (37%) cases. On the 7th day postpartum, 15 (50%) cases had normal, 9 (30%) cases had dampened, and 6 (20%) cases had still flat pattern. The majority of the cases (60%) displayed normal hepatic venous pulsatility in the 6th and 8th weeks of puerperium, whereas 23% had still dampened and 17% had flat patterns. There was a trend toward normal pulsatility with increasing puerperal age. The mean portal venous velocity was still higher than the non-pregnant levels and did not showed significant alterations during puerperium.

CONCLUSION

This study emphasised that, since pregnancy-induced alterations in hepatic venous pulsatility and portal venous velocity had not completely returned to normal in most cases until the end of the puerperium, these physiological changes should be considered whenever hepatic and portal systems are interpreted with Doppler sonography during the puerperal period.