BJOG: an International Journal of Obstetrics and Gynaecology December 2005, Vol. 112, pp. 1641 –1645
DOI: 10 .1111/ j.1 471-0528.2 005.007 65.x
Recurrence risk in hyperemesis gravidarum Lill I.S. Trogstad,a Camilla Stoltenberg,b Per Magnus,b Rolv Skjærven,c Lorentz M. Irgensc Objectives To compare the risk of hyperemesis gravidarum in second pregnancies in women with and without hyperemesis in their first pregnancy, and to determine if this risk changes with changes in paternity or with the interval between deliveries. Design Cohort study. Setting Data from the population-based Medical Birth Registry of Norway, 1967 –1998. Sample All women in the registry with records of their first and second singleton delivery, a total of 547,238 women. Methods The relative risk of hyperemesis in the second delivery was estimated as odds ratios (ORs) in logistic regression models, controlling for potential confounding factors. Main outcome measures The main outcome measure was the risk of hyperemesis in the second pregnancy according to hyperemesis in the first pregnancy, interval between deliveries and change in paternity. Results The risk of hyperemesis was 15.2% in the second pregnancy in women with and 0.7% in women without previous hyperemesis [OR ¼ 26.4, 95% confidence interval (CI) 24.2, 28.7]. The OR did not change after adjustment for maternal age, change in paternity, period of the first delivery and time interval between deliveries. After a change in paternity, the risk of recurrent hyperemesis was 10.9% compared with 16.0% in women without a change in paternity [adjusted OR (aOR) ¼ 0.60, 95% CI 0.39, 0.92]. The risk of hyperemesis in the second pregnancy increased with increasing time interval between deliveries, but only in women with no previous hyperemesis. Conclusions The primary finding was the high risk of recurrence observed in women with hyperemesis in the first pregnancy. The risk was reduced by a change in paternity. For women with no previous hyperemesis, a long interval between births slightly increased the risk of hyperemesis in the second pregnancy. Further studies are needed to explore the relative impact of genetic and environmental factors and their possible interactions in hyperemesis gravidarum. INTRODUCTION A continuous spectrum of the severity of nausea and vomiting in pregnancy ranges from the nausea and vomiting that occurs in most pregnancies to the severe disorder of hyperemesis gravidarum. Hyperemesis gravidarum is characterised by persistent nausea and vomiting associated with ketosis and weight loss (>5% of pre-pregnancy weight).1 The condition may cause altered electrolytes and depletion of the maternal blood volume, as well as hepatic and renal damage. Whether hyperemesis gravidarum represents an extreme degree of the common nausea and vomiting in pregnancy or is caused by different aetiologic factors remains unknown.
Studies of recurrence are useful for counselling as well as increasing the understanding of the causes of hyperemesis gravidarum. Nausea and vomiting in one pregnancy have been associated with an increased risk of nausea and vomiting in a subsequent pregnancy.2 Furthermore, about onethird of multigravidas with hyperemesis gravidarum admitted to hospital were reported to have been treated in hospital for the same condition in a previous pregnancy.1,3 However, the risk of hyperemesis in women who have already experienced hyperemesis once has, to the best of our knowledge, never been systematically studied. The aim of this population-based study was to estimate the risk of hyperemesis gravidarum in second pregnancies in women with and without hyperemesis in their first pregnancy, using data from the Medical Birth Registry of Norway.
a
Department of Obstetrics and Gynecology, Ulleva˚l University Hospital, Oslo, Norway b Division of Epidemiology, Norwegian Institute of Public Health, Oslo, Norway c The Medical Birth Registry of Norway, Bergen, Norway Correspondence: Dr L. I. S. Trogstad, Division of Epidemiology, Norwegian Institute of Public Health, P.O. Box 4404, Nydalen, 0403 Oslo, Norway. D RCOG 2005 BJOG: an International Journal of Obstetrics and Gynaecology
METHODS More than 1.8 million births of 16 weeks of gestation or over have been compulsorily notified and recorded in the Medical Birth Registry since 1967.4 Standardised forms are completed by midwives and the attending physician on the delivery ward within one week of the delivery. We studied www.blackwellpublishing.com/bjog
1642
L.I.S. TROGSTAD ET AL.
singleton pregnancies of 547,238 women registered with their first and second delivery between 1967 and 1998 (a total of 1,094,476 pregnancies) linked into sibships with the woman as the unit of analysis. Data on all variables were obtained from the Medical Birth Registry.4 The main outcome variable was hyperemesis in the second pregnancy diagnosed according to the guidelines of The Norwegian Society of Gynecology and Obstetrics. The diagnostic criteria for hyperemesis in Norway are persistent nausea and vomiting in pregnancy associated with ketosis and weight loss >5% of pre-pregnancy weight.5 Hyperemesis gravidarum is notified as a verbatim description and coded according to ICD-8 as 638.0, 638.9 or 784.1. The main exposure variable was hyperemesis in the first pregnancy, notified and registered as in the second pregnancy. Identifiable confounding variables were the interval between deliveries, maternal age, change in paternity and the period of the first delivery.
The mean interval between births in all women was 3.6 years. Previous hyperemesis did not influence the birth interval. Because women with birth intervals shorter than 12 months were few and may represent a special group of women, cutoff for the first category in the paper was set at 12 months. The vast majority of women had birth intervals between one and five years and was a natural reference group. The remaining intervals were divided into two categories. Thus, the final categories based on the dates of the first and second births were set at
