22: Hyperemesis gravidarum

Hyperemesis gravidarum

Veronica Gillispie and Sherri Longo

Ochsner Health System, New Orleans, LA, USA


Nausea and vomiting begins in the first trimester, usually starting at six to eight weeks gestation. Most frequently the symptoms peak at nine weeks gestation and dissipate by 12 weeks [1]. Hyperemesis gravidarum (HG) is a more severe form of nausea and vomiting associated with pregnancy.

The definition for HG is variable but most commonly accepted as intractable vomiting of at least three episodes in a day, a weight loss of greater than 5% of prepregnancy weight, acute starvation manifested as ketonuria, and electrolyte imbalance most commonly hypokalemia [24]. HG is a multifactorial condition involving gastrointestinal, hormonal, and genetic factors.

Severe nausea and emesis in pregnancy is common with an incidence of 50–80%. The incidence of HG, however, is rare and occurs in about 0.3–2% of pregnancies [2]. Most patients with HG have resolution of symptoms by 20th week of gestation. HG is the most common reason for hospitalization in the first trimester of pregnancy and is the second most common reason to be hospitalized throughout the entire pregnancy [5, 6]. The hospital readmission rate for HG is 25% [7]. In the past, the mortality rate of HG was about 10%. Now, however, maternal death is rare due to intervention with intravenous (IV) fluid [8].


The etiology of HG is unknown. The underlying pathogenesis is not well understood but theories include hormonal, genetic, psychological, and environmental components. Hormonally, HG is thought to be caused by high levels of estrogen, low levels of prolactin, and high levels of human chorionic gonadotropin (hCG). hCG is thought to contribute to HG by stimulating the ovary to produce estrogen, and estrogen increases nausea and emesis [2]. hCG’s contribution to HG is further supported by the fact that hCG levels are higher in multiple gestations and molar pregnancies which have a higher incidence of HG [9]. It has also been hypothesized by several investigators that various hCG isoforms secreted by different ethnic groups may affect HG pathogenesis [10, 11].

A large prospective study by Torgersen et al. suggested that women with the purging type of bulimia nervosa had a higher risk for nausea and vomiting during pregnancy than women without eating disorders [12]. Family history and a history of HG in a prior pregnancy are also considered risk factors for HG. Smoking seems to decrease the risk for HG [13].

Helicobacter pylori is a Gram‐negative flagellated spiral bacterium that increases the risk for gastric pathology by colonizing the stomach. H. pylori is known to be a factor in gastrointestinal diseases. Pregnant women are thought to have a predisposition for H. pylori because elevated hCG causes a shift in pH, decreased gastrointestinal motility, and the altered cell mediated immune system [13].

There have been many studies that suggest a significant positive association between H. pylori and HG. In 2009, Sandven et al. published a systematic review and meta‐analysis of case‐control studies and demonstrated that the presence of H. pylori is associated with an increased risk of HG [14]. Lin Li et al. published a meta‐analysis that included 32 articles with an overall conclusion that there is an association between H. pylori and HG. The mechanism by which H. pylori causes HG remains unclear, however some possible reasons include hormonal changes in early pregnancy resulting in a shift in gastrointestinal pH, mood changes affecting immune system with an increased risk for infection, and prolonged gastric emptying [15].

However, it remains unclear as to whether treating H. pylori will improve the symptoms of nausea and vomiting of pregnancy [16].

Biochemical thyroid axis abnormalities are frequently seen in HG patients and most commonly referred to as gestational transient thyrotoxicosis but, clinical hyperthyroidism is rare in this population. hCG is a structural homology to thyroid stimulating hormone (TSH) and suppresses TSH release. The hCG effect at the TSH receptor sites may contribute to the hyperthyroidism [8].

Over the years, investigators have proposed that there are underlying psychological or social factors that lead to HG. There has been reported a possible psychosomatic etiology to HG with patients having higher levels of anxiety with HG [17]. Some authors suggest that HG is more common in patients that are immature, dependent, depressed or hysterical but this hypothesis has not been studied [18]. There is no quality data to support the resurfacing hypothesis of a primary psychological basis for HG.

While some people have proposed a psychiatric component to HG, some studies suggest otherwise. D’Orazio et al. published in 2011 a pilot study that suggested pregnant women with HG were not more likely to have psychiatric symptoms or disturbances worse than pregnant women with normal nausea and vomiting [19]. More research is needed to determine if there is an association between psychiatric illness and HG.

Effects of hyperemesis gravidarum

Though the short‐term and long‐term effects of HG on the fetus have not been well established, a large meta‐analysis conducted by Veenendaal et al. did identify several trends [20].

Effects on the fetus

Fetal growth restriction can result as a consequence of HG. In the meta‐analysis, when low birth weight (LBW) was defined as a weight of less than 2500 g, 6.4% of HG pregnancies experienced LBW infants, compared to 5% of pregnancies not affected by HG. Not surprisingly, findings are similar when comparing the rate of small for gestational age (SGA) infants. 17.9% of pregnancies affected by HG gave birth to an infant that was SGA, compared to 12.7% of pregnancies not affected. The etiology of LBW and SGA is unclear but is most likely a result of poor maternal weight gain during the pregnancy. It is also unclear if the trend for LBW and SGA is associated with adverse outcomes as the rate of perinatal death does not seem to differ in pregnancies affected by HG versus pregnancies not affected [20].

Effects on the pregnancy

In the large meta‐analysis, having HG during pregnancy was found to be associated with an increased risk of preterm delivery. 7.4% of HG pregnancies resulted in preterm delivery. This was compared to 5.8% of pregnancies not affected by HG. Again, the clinical significance, however, is unclear but the increase in preterm deliveries may also contribute to the increased rate of LBW and SGA infants in pregnancies affected by HG [20].

Effects on the mother

The effect of HG on the mother is related to the severity of symptoms. The most detrimental side effect of HG is Wernicke’s encephalopathy [21]. Though rare, there have been case reports of Wernicke’s encephalopathy in pregnancy related to HG. Wernicke’s encephalopathy is a condition in which patients experience altered mental status due to severe thiamine deficiency. The treatment is to give high doses of thiamine but if left untreated, can lead to irreversible neurologic damage.

Because of the persistent vomiting experienced by pregnant women with HG, additional potential side effects requiring medical attention are splenic avulsion, esophageal rupture, Mallory‐Weiss tears, pneumothorax or peripheral neuropathy secondary to decreased Vitamins B6 and B12 [22].

Though there are several studies that show the effect of nutrition and weight gain in pregnancy on long‐term maternal health, to date, there are no studies evaluating the long‐term health risks to women that experienced HG during pregnancy.

Differential diagnoses

The diagnosis of HG is made by clinical evaluation and supported by laboratory evaluation. The clinical symptoms associated with HG are intractable nausea and vomiting with an inability to tolerate solids or liquids resulting in weight loss. Laboratory evaluation often shows electrolyte disturbances including a metabolic acidosis. In addition to clinical symptoms, the diagnosis can be made by a validated questionnaire such as the Pregnancy Unique Quantification of Emesis (PUQE) questionnaire [23]. This is a three question self‐administered questionnaire that quantifies the frequency of symptoms for pregnant women experiencing nausea and vomiting. The scoring system then stratifies the symptoms into mild, moderate, or severe. The results can be used to assist in distinguishing nausea and vomiting in pregnancy from HG.

Because the symptoms associated with HG can be non‐specific, it is important to consider other diagnoses that can cause nausea and vomiting in pregnancy. A logical way to consider these etiologies is to divide them into systems (See Tables 22.122.3, and 22.4).

Table 22.1 Differential diagnosis of persistent vomiting in pregnancy

Source: Goodwin (1998) [24].

Biliary tract disease
Intestinal obstruction
Peptic ulcer disease
Genitourinary tract
Kidney stones
Degenerating uterine leiomyoma
Diabetic ketoacidosis
Addison’s disease
Neurologic disorders
Pseudotumor cerebri
Vestibular lesions
Migraine headaches
Tumors of the central nervous system
Drug toxicity or intolerance
Pregnancy‐related conditions
Acute fatty liver of pregnancy

Table 22.2 Laboratory abnormalities in hyperemesis

Source: Goodwin (1998) [24].

Laboratory abnormality % Usual range from given limit of normal
Free T4 Index elevated 60 13–40
Free T3 Index elevated 10 225–350
TSH suppressed 60 <0.4 mU ml−1
Sodium low 30 125–134
Potassium low 15 2.3–3.1
Chloride low 25 80–98
Bicarbonate high 15 27–34
Bicarbonate low 8 14–22
ALT or AST high 40 41–324
T. Bili > 1.0 20 1.1–5.3
Amylase high 10 151–391
Lipase high 10 70–200

TSH, thyrotropin; ALT, alanine amino transferase; AST, aspartate amino transferase; T. Bili, total bilirubin.

Table 22.3 Randomized trials of antiemetics in pregnancy

Source: Goodwin (1998) [24].

Agent Number of trials Benefit
Nausea or vomiting
Bendectin 3 +
Pyridoxine 6 +
Meclizine 4 +
Promethazine 2 +
Hydroxyzine 1 +
Timethobenzamide 1 +
Thielpyrazine 1 +
Mepyramine 1 +
Dimenhydrinate 1 +
Hyperemesis gravidarum
Powdered ginger 1 +
Ondansetron 1

ACTH, adrenocorticotropic hormone.

Table 22.4 Medrol dosing schedule

Source: Goodwin (1998) [24].

Day Morning (mg) Midday (mg) Bedtime (mg)
1 16 16 16
2 16 16 16
3 16 16 16
4 16 8 16
5 16 8 8
6 8 8 8
7 8 4 8
8 8 4 4
9 8 4
10 8 4
11 8
12 8
13 4
14 4

The differential diagnoses include gastrointestinal conditions (ex. appendicitis, small bowel obstruction, cholecystitis, and pancreatitis), endocrine disorders (ex. diabetic ketoacidosis, thyrotoxicosis, and hyperparathyroidism), neurologic conditions (pseudotumor cerebri, migraines, and vestibular lesions), biliary tract disease, hepatitis, pyelonephritis, and other pregnancy‐related conditions such as acute fatty liver disease of pregnancy or pre‐eclampsia [1, 25].

Laboratory Evaluation

As discussed above, the diagnosis of HG is made by clinical suspicion. Laboratory results demonstrating a metabolic acidosis help to support the diagnosis. Laboratory evaluation including a complete blood count, liver function tests, electrolytes, thyroid function tests, and urinalysis are important for ruling out other causes of nausea and vomiting in pregnancy. Other tests such as amylase and lipase may need to be added if concerned for other possible causes of the nausea and emesis. An ultrasound should be performed to rule out other pregnancy related causes of HG such as multiple gestation and molar pregnancy.

Treatment options: outpatient versus inpatient treatment

The location of treatment of HG is dependent on the severity but usually results in hospitalization. There have been some reports of patients being managed outpatient using home health services but from a patient safety standpoint in‐patient care is the best [26].

Treatment of nausea and vomiting in pregnancy can be addressed in a variety of modalities. The first priority in treating HG is to treat the most detrimental manifestations. Dehydration is a hallmark of HG and must be treated first. This also includes addressing associated electrolyte disturbances. Because of the inability to tolerate oral intake, most patients will require intravenous hydration. Although there are no randomized controlled trials (RCTs) that delineate one type of fluid over another as superior, there are some key concepts regarding hydration that should be remembered. Sodium chloride 0.9% intravenous fluid is preferable to dextrose containing fluids as to not precipitate Wernicke’s encephalopathy [27]. Because thiamine requirements increase in pregnancy and approximately 60% of patients with HG will have thiamine deficiency, they are at increased risk for Wernicke’s encephalopathy [28]. Intravenous hydration treated with glucose containing fluid can worsen thiamine deficiency and increase the risk for development of Wernicke’s encephalopathy. The rate of hydration with normal saline, as well as avoidance of higher concentrations of sodium, should also be taken into consideration as rapid correction of hyponatremia can lead to central pontine myelinolysis.

In addition to intravenous hydration, electrolyte abnormalities including thiamine deficiency, should be addressed. Listed below are formulas for correcting sodium and potassium deficiencies: [29]


Only gold members can continue reading. Log In or Register to continue

Jul 19, 2020 | Posted by in GYNECOLOGY | Comments Off on 22: Hyperemesis gravidarum
Premium Wordpress Themes by UFO Themes