Hyperemesis gravidarum and placental dysfunction disorders

Nausea and vomiting are common and usually benign symptoms of primarily the first trimester of pregnancy. The onset of nausea correlates with the level of human chorionic gonadotropin (hCG), which typically rises within 4 weeks after the last menstrual period, peaking at approximately 9 weeks of gestation [1]. Sixty percent of nausea cases resolve by the end of the first trimester and 91% by 20 weeks of gestation [1]. Hyperemesis gravidarum is at the severe end of the nausea spectrum and according to The International Statistical Classification of Disease and Related Health Problems (ICD-9), is defined as ‘persistent and excessive vomiting starting before the end of the 22nd week of gestation’. Hyperemesis gravidarum is clinically classified as mild or severe, depending on associated metabolic disturbances such as carbohydrate depletion, dehydration and electrolyte imbalance. Its incidence is estimated at 0.3 to 1.5% of all live births but is unevenly distributed on a global level [2–6]. Asian women, for instance are more likely to suffer from hyperemesis than Caucasian women [3, 5]. Hyperemesis greatly affects maternal well-being and quality of life [7, 8] and is among the most common reasons for pregnancy-associated hospitalization [6, 9, 10].

Relatively little is known about the etiology of hyperemesis [6]. Causal roles of sex hormones, thyroid hormones, H. pylori infections and paternal genes have been suggested, although consensus has not been reached [6, 11]. Levels of hCG are positively associated with occurrence and severity of hyperemesis complaints, as seen in multiple- or molar pregnancies [12, 13]. While benign nausea and vomiting in early pregnancy are closely related to temporarily increased hCG levels, it has been argued that in women with hyperemesis, the persistently high hCG level dysregulates normal stimulation of trophoblast migration, which consequently alters placentation [14–16]. Ultimately abnormal placentation could lead to placental dysfunction that clinically manifests as gestational hypertension, preeclampsia, as well as miscarriage, stillbirth and intra-uterine growth restriction (IUGR) [15, 17–19]. In particular, elevated hCG plasma levels in the second trimester are associated with development of these conditions [13, 20]. Thus, hyperemesis gravidarum, occurring in first and early second trimester, could be an early pregnancy indicator of a process that results in symptomatic placental dysfunction later.

There is limited evidence about the consequences of hyperemesis on maternal and offspring’s health. Two large cohort studies in Scandinavian countries showed that hyperemesis was associated with higher risk of preeclampsia, lower birth weight and shorter gestational duration [15, 21]. This was supported by several studies suggesting higher risks of low birth weight (LBW), small for gestational age (SGA), and preterm birth if mothers experienced hyperemesis [22–24]. However, another large study [25] and several smaller studies, [26, 27] did not show such associations.

Women who experience severe hyperemesis have a significantly reduced maternal caloric ‘intake’ and lose additional nutrients and electrolytes [28]. This state resembles fasting and often involves ketonuria, which is frequently tested by clinicians in women suspected of having hyperemesis [29, 30]. Previous studies have shown that placental efficiency changes in women exposed to famine. Increased placental weight in women who were pregnant during the Dutch Hungerwinter suggests that compensatory growth of the placenta can occur in situations where nutritional resources are lacking [31, 32]. The same compensation might occur in women who experience severe hyperemesis gravidarum, however evidence is lacking.

Both hyperemesis and placental dysfunction constitute substantial maternal and neonatal health threats, particularly in the low and middle income countries where health care resources are limited [33]. Therefore, further exploration of a relation between hyperemesis and such disorders is warranted.

The present study aimed to investigate the relation between hyperemesis and placental dysfunction disorders (gestational hypertension, preeclampsia, miscarriage, and stillbirth), and neonatal outcomes, including birth weight, small for gestational age (SGA), Apgar score and gestational age at delivery.

This study shows that severe hyperemesis gravidarum is associated with a significant decrease in birth weight, but no associations between HG and gestational hypertension, preeclampsia or other placental dysfunction disorders were observed. No links were found between hyperemesis gravidarum and neonatal outcomes, such as SGA and LBW.

Strengths of this study comprise the prospective design and the number of included women. This allowed for assessing both first and second trimester effects of hyperemesis gravidarum on neonatal outcomes and particularly birth weight, to give an impression of effects on the placenta, and evaluate early pregnancy weight change as a possible mediator of the effects. However, given the low incidence of placental dysfunction disorders, our study may have been too small for definite inference. Of participants, 18.9% had some form of hyperemesis, which is very high, but our estimates of severe hyperemesis do fit previous reports. Notably, women were recruited from a referral institute for mother and child health, and indeed some with hyperemesis were referred earlier in pregnancy. Although the data about indication for referral was not available, we believe that specific referrals for hyperemesis, particularly those in early pregnancy, were for that complaint only, and not for some associated expectation of higher risk for adverse maternal or neonatal outcome, such that selection bias is unlikely. This was supported by our finding that hyperemetic pregnant women have comparable risks for obstetric complications as compared to women without hyperemesis, which include pre-pregnancy BMI, chronic hypertension and type 2 diabetes, and previous obstetric complications. Missing information increased with longer follow-up, but was largely due to women who were temporarily provided antenatal care services and then referred back to primary care, a routine that we consider unrelated to the association of interest. Our self-report questionnaire information may contain measurement error, but as patients were unaware of the study aim, such error was likely random.

Our results do not show that severe hyperemesis gravidarum increases the risk of placental dysfunction disorders. A link between hyperemesis gravidarum and placental dysfunction disorders was first suggested in a 1991 case-control study reporting a 1.6 times higher risk for preeclampsia in women with severe vomiting [43]. Consistent with our findings, previous studies also showed no increased risks of placenta dysfunction disorders with exposure to hyperemesis [44, 45]. The largest cohort to date with 1,155,033 pregnancies, of which 13,287 were complicated by hyperemesis leading to hospital admission, showed a slightly increased risk for preeclampsia, and a higher risk for pre-term preeclampsia when hyperemesis occurred in the second trimester [15]. It has been suggested that differences in outcome in hyperemetic pregnancies are explained by maternal characteristics such as (gestational) hypertension, (gestational) diabetes, and primiparity [46]. A general problem in comparing findings on relations between hyperemesis and placental dysfunction disorders is the lack of a widely accepted definition of ‘severe’ hyperemesis. Working criteria range from hyperemesis requiring hospitalization or vomiting with associated metabolic disturbances to classifications purely based on caregiver’s diagnosis. In this study, the presence of (maternal) weight loss of more than 5% (compared to the weight prior to diagnosis) was used as criterion to classify severity of hyperemesis. The measured relative weight loss provided a more objective cut off for hyperemesis severity which is also generalizable to all pregnant women of different weight.

We showed a (adjusted) 172 g lower birth weight in offspring of women with severe hyperemesis, which was not explained by gestational duration. This was in accordance with several previous reports [22, 46–49]. In this study, we did not find any association between hyperemesis and SGA or LBW, although these appeared more common in women with severe hyperemesis. Although exposed babies were born smaller, it appeared that their chances of passing the threshold of the 10th percentile for the gestational age (SGA) or 2500 g for birth weight (LBW) were equal to the non-exposed. This agrees with several previous reports, including the largest study on neonatal outcome to date [21, 27, 50]. Differences in exposure definition and classification of its severity may contribute to the variation of findings.

Previous research on the association between hyperemesis and placental measures is limited. In the present study, we detected no effects on placental weight and placenta-weight-to-birth weight ratio (PW/BW ratio). Women with severe hyperemesis, however, had heavier placentas and higher of PW/BW ratio, although not statistically significant. Heavier placentas and higher PW/BW ratio were previously reported, but only for female offspring [28]. Studies on famine and placental weight and PW/BW ratio suggest that with low caloric intake, the placenta compensatorily grows, probably to maintain adequate fetal nutrition [31, 32]. This might also occur with hyperemesis and, therefore, further research on the placenta is warranted; especially since there is evidence that compensatory growth of the placenta is associated with cardiovascular problems in later life [28, 51].

Further research requires clear HG definition and severity criteria. Large cohort studies will be needed to estimate relations between hyperemesis and placental dimensions, and rare outcomes, such as eclampsia and stillbirth. Conducting follow up studies of women with severe hyperemesis gravidarum will give a better insight in possible long term effects of hyperemesis on the health of children born out of hyperemetic pregnancies.

Hyperemesis gravidarum is an invalidating disease in early pregnancy, associated with hospitalizations, use of medication and a poorer quality of life. However, our findings do indicate no relevant impact of hyperemesis gravidarum on placental dysfunction disorders.