Hum. Reprod. Advance Access originally published online on May 23, 2006
Human Reproduction 2006 21(9):2440-2442; doi:10.1093/humrep/del166
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hyperemesis gravidarum: is an ultrasound scan necessary?
1 Early Pregnancy, Gynaecological Ultrasound and MAS Unit and 2 Fetal Medicine Unit, St George’s, University of London, London, UK
3 To whom correspondence should be addressed at: Early Pregnancy, Gynaecological Ultrasound and MAS Unit, St George’s, University of London, London, UK. E-mail: ejkirk{at}hotmail.co.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
BACKGROUND: Traditionally, in cases of hyperemesis gravidarum (HG), an ultrasound evaluation is recommended to confirm viability and to exclude multiple pregnancies and gestational trophoblastic disease (GTD). Our aim was to perform a case–control study to evaluate the incidence of these findings. METHODS: Each case of HG was matched for gestational age with the next ultrasound examination performed in an asymptomatic pregnancy. The findings were compared between the two groups. RESULTS: Two hundred and eighty-six cases of HG were matched with 286 asymptomatic women. The total number of viable pregnancies was higher in the HG group (280/286, 97.9%) than that in the control group (265/286, 92.6%; P = 0.006). The incidence of twins was 3.1% in each group (P > 0.999). The incidence of early pregnancy failure was 0.7% in women with HG compared to 7.0% in asymptomatic women (odds ratio 0.09, 95% CI 0.01–0.04, P < 0.0001). The one case of GTD was in the HG group; however, this case also presented with vaginal bleeding. CONCLUSIONS: Pregnancies complicated by HG had a similar risk of twin pregnancy, and a lower risk of early pregnancy failure compared to controls. In the absence of vaginal bleeding, there was no increase in GTD in women with HG. We conclude that an ultrasound scan is not clinically necessary in women presenting with HG, other than for maternal reassurance.
Key words: gestational trophoblastic disease/hyperemesis gravidarum/pregnancy failure/twin pregnancy/ultrasound
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Nausea and vomiting is a common symptom of early pregnancy, affecting up to 80% of women (Gadsby et al., 1993
; Lacroix et al., 2000). Hyperemesis gravidarum (HG), or severe protracted vomiting appearing for the first time before the 20th week of pregnancy that is not associated with other coincidental conditions and is of such severity as to require the patients’ admission to hospital, affects only 0.3–1.5% of pregnancies (Fairweather, 1968; Kallen, 1987; Tsang et al., 1996).
Epidemiological studies have demonstrated that HG is more common in twin pregnancies (Kallen, 1987; Basso and Olsen, 2001; Fell et al., 2006). For example, the study by Basso and Olsen (2001) identified 6084 births with HG in Denmark between 1980 and 1996 and found that twin births were twice as common as singletons in women with HG. The incidence of HG is also said to be higher in gestational trophoblastic disease (GTD). Felemban reported an incidence of HG of 28% in 71 cases of GTD diagnosed between 1984 and 1995 (Felemban et al., 1998). It has been suggested that HG may be a favourable prognostic sign in early pregnancy with a decreased risk of spontaneous miscarriage (Bashiri et al., 1995).
It has been recommended that in pregnancies affected by HG, an ultrasound examination should be performed to exclude multiple pregnancies and GTD (Nelson-Piercy, 2002). Our aim was to perform a case–control study to evaluate the incidence of multiple pregnancies, GTD or pregnancy failure in pregnancies affected by HG and in gestation-matched controls.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This was a retrospective case–control study. Women referred to the Acute Gynaecology Unit with a history of HG between January 2001 and January 2006 were included. For the purpose of this study, HG was defined as vomiting sufficient to cause ketosis as detected on urinalysis and necessitate referral to hospital. All women underwent a transvaginal or transabdominal ultrasound examination (Aloka SSD 900, 2000 or 4000, Keymed, Southend, UK; Aloka, Tokyo, Japan). Gestational age was documented for each woman according to her last menstrual period (LMP). If there was a discrepancy of more than 7 days between dates according to the LMP and the ultrasound findings, the pregnancy was re-dated according to fetal crown-rump length. During the scan, the presence or absence of an intrauterine sac, fetal cardiac activity, number of sacs/fetuses and features suspicious of GTD were documented.
If no pregnancy was visualized on scan at the time of the initial examination, this was described as a pregnancy of unknown location (PUL), serial hCG measurements were taken, and a further ultrasound examination was repeated in 7–14 days as indicated (Condous et al., 2004). If an intrauterine gestation was seen, but viability could not be confirmed on the initial scan, a further ultrasound examination was repeated in 7–14 days for viability to be determined.
As a control group, each case of HG was matched with the next ultrasound examination performed in an asymptomatic pregnancy, matched for gestational age (±7 days). This group consisted of asymptomatic women presenting to the Early Pregnancy Unit for dating or reassurance scans.
The incidence of viable and multiple pregnancies, early pregnancy failures and GTD were compared between the two groups. The diagnosis of GTD was based on histological confirmation following uterine curettage. The Mann–Whitney U-test was used to examine the gestational age range between the two groups, and chi-squared or Fisher’s exact test was used as appropriate to examine dichotomous variables. Two-sided P-values are reported.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Over the study period, there were 286 cases of HG. In 263 (92%) cases, HG was the primary indication for an ultrasound examination. It was the secondary indication in the remaining 23 cases, with the primary indication being abdominal pain in 16 cases (5.6%) and vaginal bleeding in 7 (2.4%) cases. The cases were matched for gestational age with 286 asymptomatic women who underwent an ultrasound examination for uncertain dates or anxiety in 193 (67.5%) and 93 (32.5%) cases, respectively. The study and control groups were well matched for gestational age, and the median gestational age (interquartile range) was 8+0 (6+6–9+4) weeks in both the HG and the control group (P = 0.89).
At the initial ultrasound examination, a viable pregnancy was diagnosed in 93.4% of cases of HG and in 83.6% of controls. The respective values were 5.2 and 11.2% for pregnancies where the gestation was visible but too early to assess viability, 0.7 and 2.8% for PUL and 0.3 and 2.4% for early pregnancy failure. There was one case of GTD, and this was in the HG group (overall 
2 = 16.80, df = 4, P = 0.0021, Table I).
|
In the pregnancies where the gestation was too early to assess viability and in those where there was a PUL, a further ultrasound evaluation was performed as described. Taking into account the outcomes that were subsequently established in these cases, the total number of women with a viable pregnancy was higher in the HG group (280/286, 97.9%) than that in the control group (265/286, 92.6%;
2 = 7.62, P = 0.006, Table I). There were 18 cases of twin pregnancies, nine in each group (3.1% in each, P > 0.999), and no higher order pregnancies. The incidence of early pregnancy failure was much lower in women with HG (0.7%) when compared with that in asymptomatic women (7.0%), with the former having a 10-fold decreased risk of early pregnancy failure (odds ratio 0.09, 95% CI 0.01–0.04, P < 0.0001). The one case of suspected GTD was confirmed on histology as being a complete hydatidiform mole and occurred in a woman whose primary indication for ultrasound was vaginal bleeding in pregnancy. |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Previous studies into HG have examined birth records of women with singleton or twin pregnancies and reported the incidence of previous HG. In this study, we have examined a large number of women at the time of presentation with HG to assess the status of the pregnancy. We compared these women with gestation-matched controls and have not been able to demonstrate a significant association between twin pregnancy and HG. Although the aetiology of HG remains unknown, the most likely cause is thought to be high levels of maternal serum hCG, which is more common in conditions such as GTD and multiple pregnancies. In an epidemiological study analysing 6084 births with HG using 76 084 births from asymptomatic women as a reference population, birth records and discharge diagnoses were examined for all women hospitalized during pregnancy with HG. In this study, the incidence of twins in births with HG was 2.4% compared to 1.2% in the reference population (odds ratio 2.00, 95% CI 1.67–2.39). The incidence of twin pregnancy in our population was much higher (3.1%), and this is in keeping with recent data showing that the often-quoted incidence of twin pregnancies is no longer 1 in 80 but in fact closer to 1 in 30 (Martin et al., 2003). In our study, there was no difference in the incidence of twins between cases and controls, although this may be due to the smaller sample size.
There was one case of GTD in this series of 572 pregnancies. This was a case of complete hydatidiform mole, and although there was a history of HG, the principal presenting complaint was vaginal bleeding. A recent retrospective study of the clinical presentation of 41 cases of complete hydatidiform moles showed that 58% presented with vaginal bleeding and just 2% with hyperemesis (Gemer et al., 2000). This change in the clinical presentation of GTD over the years was also reported by Soto-Wright et al. (1995) who showed that vaginal bleeding is the usual presenting feature, with classical signs and symptoms such as excessive uterine size, anaemia, pre-eclampsia, hyperthyroidism and HG being less common (Soto-Wright et al., 1995). The case of GTD reported in this study was diagnosed ultrasonographically and confirmed by histopathological examination after surgical evacuation. Ultrasound examination has, however, been shown to identify <50% of cases of GTD, with the majority of cases appearing as incomplete or delayed miscarriages on ultrasound scan (Fowler et al., 2006). Women in our study who underwent surgical evacuation for a delayed miscarriage all had products of conception sent for histopathological examination, and no further cases of GTD were identified in either the HG or the control group.
We have also shown that in women with HG there is a 10-fold reduction in the incidence of early pregnancy failure when compared with that in asymptomatic controls (0.7 versus 7%). In the study by Bashiri et al., the incidence of miscarriage in women with HG was reported as 3.1%. There was no control group for miscarriage in this study, and the rate was reported as being lower than previously reported rates in the general population of 15% (Bashiri et al., 1995).
We acknowledge that one of the limitations of our study is the possibility of selection bias. As an early first-trimester ultrasound scan is not performed routinely in our unit, it is difficult to obtain an unbiased control group. Although more than two-thirds of the controls underwent ultrasound because of uncertain menstrual dates, a further one-third of controls were on the basis of patient request. It is possible that this latter self-selected group includes more women with a previous poor pregnancy history, therefore increasing the likelihood of miscarriage in this group. However, as women with a previous history of miscarriage have been shown to have an increased chance of HG in subsequent pregnancies (Chin, 1989; Bashiri et al., 1995), we do not feel that this is likely to be significant.
From the results of this study, it would appear that there is no need to routinely scan all women presenting with HG to exclude GTD or multiple pregnancies. In addition, pregnancies in women with HG are about 10 times less likely to be non-viable when compared with pregnancies in women not complaining of HG. Unless there are other symptoms of possible poor pregnancy outcome, such as bleeding or abdominal pain, a scan need only be performed for maternal reassurance.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Bashiri A, Neumann L, Maymon E, Katz M. (1995) Hyperemesis gravidarum: epidemiologic features, complications and outcome. Eur J Obstet Gynecol Reprod Biol 63:135–138.[CrossRef][ISI][Medline]
Basso O and Olsen J. (2001) Sex ratio and twinning in women with hyperemesis or pre-eclampsia. Epidemiology 12:747–749.[CrossRef][ISI][Medline]
Chin RK. (1989) Antenatal complications and perinatal outcome in patients with nausea and vomiting-complicated pregnancy. Eur J Obstet Gynecol Reprod Biol 33:215–219.[CrossRef][ISI][Medline]
Condous G, Okaro E, Khalid A, Timmerman T, Lu C, Zhou Y, Van Huffel S, Bourne T. (2004) The use of a new logistic regression model for predicting the outcome of pregnancies of unknown location. Hum Reprod 19:1900–1910.
Fairweather DV. (1968) Nausea and vomiting in pregnancy. Am J Obstet Gynecol 102:135–175.[ISI][Medline]
Felemban AA, Bakri YN, Alkharif HA, Altuwaijri SM, Shalhoub J, Berkowitz RS. (1998) Complete molar pregnancy. Clinical trends at King Fahad Hospital, Riyadh, Kingdom of Saudi Arabia. J Reprod Med 43:11–13.[ISI][Medline]
Fell DB, Dodds L, Joseph KS, Allen VM, Butler B. (2006) Risk factors for hyperemesis gravidarum requiring hospital admission during pregnancy. Obstet Gynecol 107:277–284.
Fowler DJ, Lindsay I, Seckl MJ, Sebire NJ. (2006) Routine pre-evacuation ultrasound diagnosis of hydatidiform mole: experience of more than 1000 cases from a regional referral centre. Ultrasound Obstet Gynecol 27:56–60.[CrossRef][ISI][Medline]
Gadsby R, Barnie-Adshead AM, Jagger C. (1993) A prospective study of nausea and vomiting during pregnancy. Br J Gen Pract 43:245–248.[ISI][Medline]
Gemer O, Segal S, Kopmar A, Sassoon E. (2000) The current clinical presentation of complete molar pregnancy. Arch Gynecol Obstet 264:33–34.[CrossRef][ISI][Medline]
Kallen B. (1987) Hyperemesis during pregnancy and delivery outcome: a registry study. Eur J Obstet Gynecol Reprod Biol 26:291–302.[ISI][Medline]
Lacroix R, Eason R, Eason E, Melzack R. (2000) Nausea and vomiting during pregnancy. A prospective study of its frequency, intensity, and patterns of change. Am J Obstet Gynecol 182:931–937.[CrossRef][ISI][Medline]
Martin JA, Hamilton BE, Sutton PD, Menacker F, Munson ML. (2003) National vital statistics reports. Births: Final data for 2002(National Center for Health Statistics, Hyattsville, MD) vol. 52:no. 10.
Nelson-Piercy C. (2002) Gastrointestinal disease. Handbook of Obstetric Medicine 2nd edn. (Martin Dunitz Ltd, London, UK) pp. 218–240.
Soto-Wright V, Bernstein M, Goldstein DP, Berkowitz RS. (1995) The changing clinical presentation of complete molar pregnancy. Obstet Gynecol 86:775–779.[Abstract]
Tsang IS, Katz VL, Wells SD. (1996) Maternal and fetal outcomes in hyperemesis gravidarum. Int J Gynaecol Obstet 55:231–235.[CrossRef][Medline]
Submitted on February 22, 2006; resubmitted on March 12, 2006; accepted on April 7, 2006.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||