Arch Gynecol Obstet (2011) 284:843847 DOI 10.

1007/s00404-010-1759-8

MATERNO-FETAL MEDICINE

Role of Helicobacter pylori in the pathogenesis of hyperemesis gravidarum

Ghada M. Mansour Ehab H. Nashaat

Received: 9 September 2010 / Accepted: 29 October 2010 / Published online: 16 November 2010 Springer-Verlag 2010

Abstract Purpose To evaluate the role of Helicobacter pylori (H. pylori) in the pathogenesis of hyperemesis gravidarum (HG) and the value of adding a non teratogenic regimen for its treatment in intractable cases. Methods Eighty hyperemesis gravidarum cases were recruited from Ain Shams University out patient clinics. A complete history was taken including history of medical disorders and chronic medications intake as non steroidal anti-inammatory drugs. After general and local examination, ultrasound was done for all cases to exclude obstetric causes of hyperemesis. Eighty normal pregnant women acted as control. Serum test for H. pylori IgG antibody titre was done for all patients and controls. Results Seventy-one cases among the 80 HG cases and twenty-four out of the 80 controls were H. pylori positive. Eight HG cases developed severe intractable vomiting. Three of them developed attacks of hematemesis. Gastroscopy done for the eight cases revealed antral gastritis and duodenitis. Gastric and duodenal erosions were found in two cases. The eight patients received a non teratogenic regimen for treatment. Attacks of vomiting decreased and pregnancy continued till delivery of healthy newborns. Conclusion Screening for H. pylori should be added to the investigations of hyperemesis gravidarum cases. Non

teratogenic treatment can be considered in intractable cases. Keywords Helicobacter pylori Hyperemesis gravidarum

Introduction Nausea and vomiting (morning sickness) is a major complaint in 7080% of pregnancies [1, 2]. Severe nausea and vomiting associated with weight loss, ketonemia, electrolyte imbalance (hyponatraemia and hypokalaemia), metabolic hypochloraemic alkalosis and elevated liver enzymes in pregnancy is called hyperemesis gravidarum (HG) [3]. It complicates 0.32% of all pregnancies. Its cause is unknown but there are some hypothesis like hormonal mechanisms, emotional factors and Helicobacter pylori infection. H. pylori is one of the common bacterial infections world wide and accepted as the cause of chronic active gastritis (type B). Most patients continue through life with a chronic supercial gastritis while some develop either duodenal or gastric ulcer [4]. H. pylori is a gram negative, spiral shaped, microaerophilic bacteria, with a prevalence rate higher in developing countries than developed countries [5]. The possible transmission route may be oral-oral, feco-oral iatrogenic transmission and vectorial spread [6]. H. pylori infection is associated with enhanced gastrin release from human antrum, and increasing evidence suggests a major role of cytokines in the pathogenesis of H. pylori associated gastritis and peptic ulcer disease, in addition to a defective bicarbonate secretion, which normally occurs in response to duodenal acidication. The nal result is an increased duodenal acid load in cases of H. pylori infection [7]. The increased

G. M. Mansour (&) Department of Obstetrics and Gynecology, Ain Shams University, 5th El Gahez street, Seventh area, Nasr city, Cairo, Egypt e-mail: gmansour@hotmail.com E. H. Nashaat Department of Internal Medicine, Ain Shams University, Cairo, Egypt

123

844

Arch Gynecol Obstet (2011) 284:843847

serum steroid level and human chorionic gonadotrophin change the pH of gastrointestinal tract in addition to the pregnancy induced GIT dysmotility and altered humoral, and cell mediated immunity in pregnancy all favor activation of H. pylori infection.

Aim of the work To evaluate the role of H. pylori in the pathogenesis of hyperemesis gravidarum and the value of adding a non teratogenic regimen for its treatment in intractable cases.

Subjects and methods A prospective study included 80 pregnant women with a gestational age ranging from 10 to 16 weeks, with hyperemesis gravidarum. Cases were recruited from Ain shams university out patient clinics in the period from January 2004 till July 2010. Eighty normal pregnant ladies acted as control. Controls were randomly selected from pregnant women presenting to the outpatient clinics for routine antenatal care of the same gestational age, same age range and same socioeconomic standard as cases. All controls had no gastrointestinal symptoms and had the same exclusion criteria of the cases. Statistically, 80 cases in each group would allow a difference of 25% points in prevalence of H. pylori assuming that the rate is 67% in rst group with the power of 90%, using 5% signicance level (condence interval = 95%). The criteria for HG were severe vomiting (more than 3 times per day without any other obvious cause), weight loss more than 3 kg and the presence of at least one positive ketonuria. A complete history was taken from all cases, including history of medical disorders such as peptic ulcer and history of chronic medications intake as non steroidal anti-inammatory drugs (NSAIDs), hyperthyroidism, psychological disorders, hepatic disorders, renal disorders including urinary tract infection, biliary or pancreatic diseases, and intracranial disorders. Patients with any of the above criteria were excluded from the study. Patients who received antibiotics or proton pump inhibitors or H2 blockers in the preceding months were excluded from the study. Cases with previous diagnosis of H. pylori infection and who received treatment for this were excluded from the study. The study was approved by the ethics committee and informed consent was obtained from all participants. After general and local examination and exclusion of any associated medical disorders, lab tests were done including complete blood count, AST, ALT, BUN, serum creatinine, T3, T4 and TSH. An ultrasound was done for all cases including fetal biometry, placental site, amount of amniotic

uid, and exclusion of any obstetric cause for hyperemesis as twin pregnancy or molar pregnancy. Urine analysis for ketones was done for detection of starvation ketosis. Serum H. pylori IgG antibody titer using commercial ELISA method was measured for all patients and controls. IgG antibody titer \13 U/mL (which corresponds to optical density ratio \0.9) was considered negative, IgG titer [16.5 U/mL (which corresponds to optical density ratio [1.1) was regarded as positive and IgG level 1316.5 U/mL (which corresponds to optical density ratio 0. 91.1) was regarded as suspicious requiring repetition after 24 weeks [9]. Upper GIT endoscopy was done for eight cases with severe intractable vomiting. Histopathological examination of all biopsies was done. After tabulation, all data were analyzed using SPSS software, version 11.0 (SPSS, Chicago, IL, USA). The Pearson v2 test was used for nominal values and the paired t test and analysis of variance were used for numerical values. P \ 0.05 indicated statistical signicance.

Results Age of the cases ranged from 2535 years (28.9 3.9) and from 22 to 32 years (26.39 3.9) in the control group. Cases of HG included 62 primigravidae and 18 multigravidae. None of them had a history of previous peptic or duodenal ulcer. Ten multigravidae had a history of hyperemesis in previous pregnancies, three of them had previous induced abortions for severe intractable vomiting. No statistical signicant difference was found as regarding age, parity, demographic data and fetal biometry between cases and control group. H. pylori serum antibody test was positive in 71 out of 80 hyperemesis cases (88%) and 24 out of 80 (30%) in the control group (Table 1). A signicant older mean age was found among cases and controls with seropositive H. pylori in comparison to seronegative women (Table 2). Among the 80 HG cases, there were 8 patients with severe symptoms not responding to treatment in spite of intravenous uids, electrolyte replacement, anti-emetics and vitamin supplementation. Three cases among the severe cases experienced attacks of hematemesis. An upper endoscopy was done by the second author for the eight cases with intractable symptoms which revealed antral gastritis and duodenitis. Gastric and duodenal erosions were detected in two of them. H. pylori infection was proved in histopathological examination of all biopsies. Three cases of these cases had a previous history of induced abortion due to the severe intractable vomiting. All the eight patients had passed 12 weeks of pregnancy and agreed on a consent for treatment in the form of ranitidine (class b) 150 mg twice daily, metronidazole (class b) 500 mg twice daily and ampicillin

123

Arch Gynecol Obstet (2011) 284:843847 Table 1 H. pylori serum antibody positive and negative among cases and controls Groups Control (80) N H. pylori serum antibody test -ve (65) ?ve (95) v test Cases odds, 7.8; control odds, 0.42; relative risk (RR), 2.9; relative risk increase, 193%; absolute risk increase or reduction, 58% Table 2 Comparison between cases and controls regarding age and its relation to H. pylori serum antibody positivity and negativity Groups Control (80) N H. pylori serum antibody test -ve ?ve T P Students t test Cases with positive H. pylori test showed signicantly older age when compared to those with -ve test among cases and among control group P \ 0.001 24.13 30.57 3.40 0.001 2.27 0.57 25.00 29.42 14.73 \0.001 0.00 3.88 % Cases (80) N %
2

845

Odds ratio (95% CI) Cases (80) % N %

v2

56 24

70.0 30.0

9 71

11.3 88.8

18.41 (7.9342.74)

57.23

\0.001

(class b) 1,000 mg twice daily for 2 weeks. Ampicillin and ranitidine were given by parentral route and metronidazole by rectal route, till patients could receive oral therapy. Six patients showed marked improvement with reduced number of vomiting attacks which became from 0 to 2 attacks per day and reduced epigastric pain. The other two patients showed less improvement of the vomiting and pain, doses of antacids were increased and they also continued their pregnancy safely until delivery.

Discussion HG complicates from 0.3 to 2% of all pregnancies [8, 9]. Sometimes symptoms of HG become alarming, patients who manifest continuous weight loss and electrolyte disturbances may be at risk for growth restriction, fetal anomalies and decreased neonatal birth weight [10, 11]. Dehydration, electrolyte imbalance, and acidbase disturbances may lead to maternal renal and hepatic injury [12, 13]. Many theories postulated the aetiology of HG, Eliakim et al. [12], stated that Hyper emesis gravidarum involves pregnancy induced hormonal changes with concurrent gastrointestinal dysmotility and possible H. pylori

infection. H. pylori is a very common pathogen found in nearly 55% of population, 90% of peptic ulcer patients, and in 6080% in gastritis cases with no ulcers [14]. Many authors stated the association of HG and H. pylori infection [1519]. The prevalence rate of H. pylori is higher in developing countries than developed countries [5]. The possible transmission route may be oraloral, fecooral, iatrogenic transmission and vectorial spread [6]. H. pylori causes a noninvasive infection of the gastric epithelium and the mucous layer that coats this epithelium. It can cause duodenal ulcer, gastric ulcer, chronic gastritis, gastric adenocarcinoma, mucosa associated lymphoid tissue lymphoma and a few other rare upper gastrointestinal disorders [20]. It is reported that gastric mucosa infected by H. pylori almost always shows a combination of inammation and epithelial changes. The classical feature caused by this organism is chronic active gastritis. The inltrate generally consists of monocytes and neutrophils. The degree of inammation varies in severity from a minimal inammatory inltrate in the lamina propria with preserved architecture to severe gastritis with dense inammation. In very severe cases intraepithelial neutrophils can be detected in both the surface epithelium and in the gastric pits as micro abscesses [21]. In this study, incidence of H. pylori was signicantly higher in HG cases than normal group (Table 1) and that agree with many other studies who found an increased incidence of H. pylori infection in cases of HG [1620]. It is well known that nausea and vomiting are associating changes in hormonal levels following gestation, but the pathogenesis of HG is still unclear. With a rise in serum HCG level, symptoms of nausea and vomiting appear, and when serum HCG levels decrease, nausea and vomiting disappear. High incidence of hyperemesis is found in cases of molar pregnancies, and in twin pregnancies exhibiting high serum HCG concentrations, and nausea and vomiting immediately disappear when gestation is interrupted. Thus, most researchers believed that elevated serum HCG levels are associated with hyperemesis gravidarum. However, it is not easy to explain the fact that serum HCG levels are not directly proportional to the severity of the symptoms. Another proposition is that HG is

123

846

Arch Gynecol Obstet (2011) 284:843847

caused by rapidly increased estrogen levels. As a result of an increased accumulation of uid caused by elevated steroid hormones in pregnant women, a shift in pH may occur. A simple change of pH in the gastrointestinal tract could hypothetically result in the appearance of manifestation of a subclinical H. pylori infection, which can exacerbate gastrointestinal symptoms. Some researchers mentioned that H. pylori infects the gastric mucosa of more than half of all humans worldwide, but only 15% of those affected have clinical symptoms. It has been reported that the pathogenicity of H. pylori is related not only to its virulence but also to host gene susceptibility and environmental factors [22]. Bagis et al. obtained histopathologic diagnosis of H. pylori infection by endoscopic evaluation in 20 pregnant women with severe HG and 10 pregnant women without gastric complaints. H. pylori was diagnosed in 90% of HG patients and 50% of control subjects. They stated that the severity of H. pylori infection was higher in HG patients. It was suggested that the degree of gastric complaints may be associated with the density of H. pylori infection [23]. These data may explain the asymptomatic positive H. pylori women among controls in this study. We can also add the humoral and cellular immunity which may differ from one pregnant woman to another. Various epidemiologic studies have proved that the major risk factor of H. pylori infection is low socioeconomic level [24]. Cases and controls in this study were recruited from the out patient clinics of university hospitals where medical service is almost free. This explains the moderate to low socioeconomic standard of all women among cases and controls. Serum H. pylori IgG antibody was used in this study. It is an economic noninvasive and feasible screening for H. pylori infection. Among the 80 HG cases in this study, there were 8 patients with severe symptoms not responding to treatment inspite of intravenous uids, electrolyte replacement, anti-emetics and vitamin supplementation. Three of them experienced attacks of hematemesis. An upper endoscopy was done by the second author for the eight patients with intractable symptoms which revealed antral gastritis and duodenitis. Gastric and duodenal erosions were found in two of them. Histopathological examination proved H. pylori infection in all biopsies. All the eight patients had passed 12 weeks of pregnancy and agreed on a consent for treatment in the form of ranitidine (class b) 150 mg twice daily, metronidazole (class b) 500 mg twice daily and ampicillin (class b) 1,000 mg twice daily for 2 weeks. Ampicillin and ranitidine were given by parentral route and metronidazole by rectal route, till patients could receive oral therapy. Six patients showed marked improvement with declining of number of vomiting attacks which became from 0 to 2 attacks per day and improvement of epigastric pain. The other two patients showed less improvement of the

vomiting and pain, doses of antacids were increased and they also continued their pregnancy safely until delivery. It should be noted that the treatment in this study was given to the eight complicated cases and although a signicant response was noticed but still further studies with larger numbers of patients are needed to corroborate such ndings. Also more studies on the safety of proton pump inhibitors during pregnancy should be done, because proton pump inhibitors will give better results in these cases. Patients education for food safety is important. Careful food handling and hand washing are important to prevent introduction of foodborne pathogens to the diet of pregnant women [12].

Conclusion Screening for H. pylori should be added to the investigations of HG, especially in prolonged conditions that are refractory to conventional management and cases that extend to the second trimester. A non teratogenic regimen for treatment of H. pylori may be considered in intractable cases. Patients education for food safety should be considered to prevent the increasing numbers of infected cases especially in developing countries and lower socioeconomic levels. Although there was no evidence to prove the role of H. pylori infection in pathogenesis of HG but its association may aggravate the symptoms.
Conict of interest None.

References
1. Klebanoff MA, Koslowe PA, Kaslow R, Rhoads GG (1985) Epidemiology of vomiting in early pregnancy. Obstet Gynecol 66(5):612616 2. Cunningham G, Gant NF, Kenneth JL, Gilstrap LC, Hauth JC, Wenstrom KD (2001) Williams obsteyrics, 21st edn. Mc GrawHill Company, USA, pp 12751276 3. Creasy R, Resnik R (1999) Maternal fetal medicine, 4th edn. WB Saunders company, Philadelphia, pp 10421044 4. Hunt H (1996) The role of Helicobacter pylori in pathogenesis: the spectrum of clinical outcomes. Scand J Gastroenterol Suppl 220:223 5. Soll AH (1996) Helicobacter pylori induced gastritis. In: Bennet JC, Plum F (eds) Cecil textbook of medicine, 20th edn. Philadelphia, WB Saunders, pp 659660 6. Cave D (1997) How is Helicobacter pylori transmitted? Gastroenterology 113(supp):S9S14 7. Olbe L, Fandriks L, Hamlet A (2000) Conceivable mechanisms by which H. pylori provokes duodenal ulcer disease. Clin Gastroenterol 14(1):1 8. Broussard CN, Richter JE (1998) Nausea and vomiting of pregnancy. Gastroenterol Clin North Am 27:123151 9. Hod M, Orvieto R, Kaplan B (1994) Hyperemesis gravidarum: a review. J Reprod Med 39(8):605612

123

Arch Gynecol Obstet (2011) 284:843847 10. Gross S, Librach C, Cecutti A (1989) Maternal weight loss association with hyperemesis gravidarum: a predictor of fetal outcome. Am J Obstet Gynecol 160:906909 11. Tincello DG, Johnstone MJ (1996) Treatment of hyperemesis gravidarum with the 5-HT3 antagonist ondansetron (Zofran). Postgrad Med J 72(853):688689 12. Eliakim R et al (2000) Hyperemeiss gravidarum: a current review. Am J Perinatol 17:207 13. Sylvia ES (2008) Vomiting, pernicious. In: Sylvia Escott-Stump (ed) Nutrition and DiagnosisRelated care, 6th edn. Lippincott Williams & Wilkins, Philadelphia, p 384 14. Wu CY, Tseng JJ, Chou MM et al (2000) Correlation between Helicobacter pylori infection and gastrointestinal symptoms in pregnancy. Adv Ther 17:152158 15. Frigo P, Lang C, Reisenberger K, Kolbl H, Hirschl AM (1998) Hyperemesis gravidarum associated with Helicobacter pylori seropositivity. Obstet Gynecol 91(4):615617 16. Kocak I, Akcan Y, Ustun C, Demirel C, Cengiz L, Yanik FF (1999) Helicobacter pylori seropositivity in patients with hyperemesis gravidarum. Int J Gynaecol Obstet 66(3):251254 17. Hayakawa S, Nakajima N, Karasaki-Suzuki M, Yoshinaga H, Arakawa Y, Satoh K, Yamamoto T (2000) Frequent presence of

847 Helicobacter pylori genome in the saliva of patients with hyperemesis gravidarum. Am J Perinatol 17(5):243247 Jacoby EB, Porter KB (1999) Helicobacter pylori infection and persistent hyperemesis gravidarum. Am J Perinatol 16(2):8588 Jamal A, Pooransari P, Andansari R (2004) Relationship between Helicobacter pylori seropositivity and hyperemesis gravidarum. Acta Medica Iranica 42(5):375378 Flook NW (1998) Helicobacter pylori primary care management from symptoms to cure. Can Fam Physician 44:14291430 Blecker U (1997) Helicobacter pylori-associated gastroduodenal disease in childhood. South Med J 90(6):570575 Meijer BC, Thijs JC, Kleibeuker JH et al (1997) Evaluation of eight enzyme immunoassays for detection of immunoglobulin G against Helicobacter pylori. J Clin Microbiol 35:292294 Bagis T, Gumurdulu Y, Kayaselcuk F et al (2002) Endoscopy in hyperemesis gravidarum and Helicobacter pylori infection. Int J Gynaecol Obstet 79:105109 Karadeniz RS, Ozdegirmenci O, Altay MM et al (2006) Helicobacter pylori seropositivity and stool antigen in patients with hyperemesis gravidarum. Infect Dis Obstet Gynecol 2006:73073

18. 19.

20. 21. 22.

23.

24.

123