Semaglutide-induced Hyperemesis Gravidarum

Abstract

Long-acting glucagon-like peptide-1 receptor agonists have traditionally been used for glycemic control in type 2 diabetes mellitus. More recently, the medical community in the United States has used its most common side effect, weight loss, as an adjunct therapy for weight management in certain populations. Little is known about the effects of semaglutide in pregnancy. But in nonpregnant people, nausea, vomiting, decreased appetite, and abdominal cramping are common side effects. Here we present the first reported case of hyperemesis gravidarum (HG) induced by semaglutide. Our patient presented at 7 weeks’ gestational age with 4 days of extreme nausea, vomiting, abdominal cramping, and anorexia, symptoms that she did not experience in prior pregnancies. The patient was treated with little symptomatic improvement during her hospitalization. She later informed the medical team of her use of a high dose of subcutaneous semaglutide (2 mg) the day of symptom onset before discovering her pregnancy; semaglutide was obtained online without clinicians’ guidance. Her symptoms resolved 6 days after taking semaglutide, which closely coincides with semaglutide's half-life. This leads us to believe that semaglutide exaggerated the physiological nausea and vomiting of pregnancy causing HG.

Introduction

Semaglutide is a glucagon-like peptide receptor agonist (GLP-1 RA) used for glycemic control in type 2 diabetes mellitus (T2DM) via decreased gastric emptying, reduced food intake, and enhanced glucose-dependent insulin secretion [1]. Weight loss is a favorable, common side effect of GLP-1 RA, making its use in weight management in overweight and obese patients appealing. Weight loss is due to a combination of pathophysiology including decreased gastric emptying and nausea and vomiting [2].

Semaglutide might augment nausea and vomiting in pregnancy (NVP). NVP is common during the first trimester and is often termed “morning sickness” despite its occurrence at any time throughout the day. Depending on the severity and effect on quality of life, symptoms can be mild, moderate, or severe, with hyperemesis gravidarum (HG) considered a severe form [3]. HG negatively affects daily function and performance, triggers anxiety and depression, and prompts the elective termination of the pregnancy in extreme cases [4]. Here, we describe the first reported case of HG incited by semaglutide ever cited in the literature.

Case Presentation

A 34-year-old G5P1A3 woman with a medical history of anxiety, attention-deficit/hyperactivity disorder, and borderline personality disorder was admitted to the hospital at 7 weeks’ gestation for a 4-day history of nausea, bilious nonbloody vomiting (>10 episodes per day), right lower-quadrant abdominal pain, and poor oral intake.

The patient denied substantial NVP in her prior pregnancies, aside from the expected morning sickness symptoms. Her first pregnancy was an uncomplicated spontaneous vaginal delivery at 40.3 weeks with minimal to no nausea and vomiting. Her other pregnancies were missed abortions and elective abortions. Menarche occurred at age 14. Her menstrual cycle is approximately 28 days long, with menses lasting 5 to 6 days. The patient denied any history of metrorrhagia.

The patient works as a nurse. She initially denied the use of any prescription drugs, oral contraceptives, vitamins, or supplements. She has no known food or drug allergies. The patient denied any history of HG in first-degree relatives. She denied cigarette smoking but reported nicotine vaping (approximately 1 cartridge every 3 days for the past 2-3 years). The patient reported seldom alcohol use and denied recreational drug use. She denied any consumption of new foods or recently eating out. Her most recent sick contact was her daughter over 2 weeks ago who had a low-grade fever and constipation, but no nausea or vomiting.

Diagnostic Assessment

In the emergency department, the patient was vitally stable with an oral body temperature of 99.4 °F (37.4 °C) (normal range, 97-99 °F; 36.1-37.2 °C), a blood pressure of 129/86 mm Hg (normal <120/80 mm Hg), and a pulse of 76 beats per minute (bpm) (normal range, 60-100 bpm). BMI, calculated as weight in kilograms divided by height in meters squared, was 26.3 (normal, 18.5-24.9). Transabdominal ultrasound revealed a single live intrauterine gestation, a simple left ovarian cyst, and subchorionic hemorrhage. The right lower-quadrant ultrasound was unremarkable. β-Human chorionic gonadotropin (hCG) was 71 604 mIU/mL (normal range for 7 weeks’ GA, 4059-153 767 mIU/mL). Complete blood count and comprehensive metabolic panel were grossly unremarkable without evidence of acute kidney injury; lipase was within reference range, making the diagnosis of acute pancreatitis unlikely (Tables 1 and 2). Urinalysis with microscopy showed ketonuria (Table 3). A routine 6-panel urine drug screen was negative.

Table 1.

Complete blood count

Parameter	Value	Normal range
White blood cell count	10.2×103/μL (10.2×109/L)	4.5-11.0×103/μL (4.5-11.0×109/L)
Red blood cell count	4.74×106/μL (4.72×1012/μL)	4.10-5.10×106/μL (4.10-5.10×1012/L)
Hemoglobin	14.4 g/dL (8.94 mmol/L)	12.0-16.0 g/dL (1.86-2.48 mmol/L)
Hematocrit	41.6% (0.146)	35.0%-48.0% (0.35-0.48)
Mean corpuscular volume	87.8 µm3 (87.8 fL)	80.0-100.0 µm3 (80-100 fL)
Mean corpuscular hemoglobin	30.4 pg/cell (1.89 fmol/cell)	25.4-35.6 pg/cell (0.39-0.54 fmol/cell)
Mean corpuscular hemoglobin concentration	34.6 g/dL (346.0 g/L)	31.0-36.0 g/dL (310.0-360.0 g/L)
Platelet count	375 (375×109/L)	140-450×103/μL (150-400×109/L)

Values in parenthesis are in International System of Units (SI). Values in bold font are abnormal.

Abbreviations: μL, microliters; g/dL, grams per deciliter; g/L, grams per liter; L, liters; mmol/L, millimoles per liter; µm³, cubic microns; fL, femtoliters; fmol/cell, femtomole per cell; pg/cell, picograms per cell.

Table 2:

Comprehensive Metabolic Panel

Parameter	Value	Normal range
Sodium	139 mEq/L (139.0 mmol/L)	136-145 mEq/L (136-145 mmol/L)
Potassium	3.8 mEq/L (3.80 mmol/L)	3.5-5.1 mEq/L (3.5-5.1 mmol/L)
Chloride	101 mEq/L (101.0 mmol/L)	98-107 mEq/L (98-107 mmol/L)
Glucose	87 mg/dL (4.83 mmol/L)	70-99 mg/dL (3.89-5.49 mmol/L)
BUN	22 mg/dL (7.85 mmol/L)	9-23 mg/dL (3.21-8.21 mmol/L)
Albumin	4.9 g/dL (49.0 g/L)	3.4-5.0 g/dL (34.0-50.0 g/L)
Total bilirubin	0.8 mg/dL (13.68 µmol/L)	0.2-1.3 mg/dL (3.42-22.24 µmol/L)
Calcium	10.1 mg/dL (2.52 mmol/L)	8.7-10.4 mg/dL (2.17-2.60 mmol/L)
Creatinine	0.80 mg/dL (70.72 µmol/L)	0.55-1.02 mg/dL (48.62-90.17 µmol/L)
Alkaline phosphatase	53 U/L (0.89 µkat/L)	46-116 U/L (0.77-1.94 µkat/L)
Protein, total	8.3 g/dL (83.0 g/dL) (H)	6.0-8.0 g/dL (60.0-80.0 g/L)
AST	20 U/L (0.33 µkat/L)	0-34 U/L (0.0-0.57 µkat/L)
Carbon dioxide	22 mEq/L (22.0 mmol/L)	20-31 mEq/L (20.0-31.0 mmol/L)
Anion gap	15 mEq/L (15.0 mmol/L)	5-15 mEq/L (5.0-15.0 mmol/L)
ALT	18 U/L (0.30 µkat/L)	10-49 U/L (0.17-0.82 µkat/L)
eGFR	≥60 mL/min/1.73 m2	≥60 mL/min/1.73 m2
Lipase	28 U/L (0.47 µkat/L)	12-53 U/L (0.27-0.89 µkat/L)

Values in parenthesis are in International System of Units (SI). Values in bold font are abnormal.

Abbreviations: µkat/L, microkatal per liter; µmol/L, micromoles per liter; ALT, alanine transaminase; AST, aspartate transaminase; BUN, blood urea nitrogen; eGFR, estimated glomerular filtration rate; g/L, gram per liter; mEq/L, milliequivalents per liter; mg/dL, milligram per deciliter; mL/min/1.73 m², milliliters per minute per 1.73 meters squared; mmol/L, millimole per liter; U/L, units per liter.

Table 3:

Urinalysis

Parameter	Value	Normal Range
Appearance	Clear	Clear
Color	Yellow	Straw or yellow
Glucose	Negative	Negative (mg/dL) (Negative [mmol/L])
Bilirubin	Negative	Negative
Ketones	≥ 80 (H)	Negative (mg/dL) (Negative [mmol/L])
Specific gravity	1.020	1.005-1.030
Blood	Negative	Negative
pH	7.0	4.0-8.0
Protein	Trace (H)	Negative (mg/dL) (Negative [g/L])
Nitrites	Negative	Negative
Leukocyte esterase	Small (H)	Negative

Values in parenthesis are in International System of Units (SI). Values in bold font are abnormal.

Abbreviations: g/L, grams per liter; H, high; mg/dL, milligrams per deciliters; mmol/L, millimoles per liter.

Treatment

The patient initially received a 1-L bolus of normal saline (NS) followed by an infusion of NS at a rate of 125 mL/h, 3 doses of phenergan 25 mg suppository administered 6 hours apart, hydromorphone 0.5 mg intravenously (IV) twice, metoclopramide 10 mg IV once, prochlorperazine 25 mg suppository once, ondansetron 4 mg IV once, lorazepam 1 mg IV once, famotidine 20 mg IV once, and acetaminophen 100 mg IV once with no relief of symptoms. She eventually received pyridoxine 100 mg IV, oral doxylamine 25 mg, and an IV infusion of 5% dextrose with NS, thiamine, folate, and multivitamins without symptomatic improvement. As a last resort, a one-time dose of IV methylprednisolone 125 mg was given as an abortive agent for intractable nausea and vomiting. Her symptoms remained extraordinarily refractory to these interventions.

Outcome and Follow-up

On day 2 of admission, the patient disclosed that she self-administered 2 mg of semaglutide subcutaneously once for weight reduction 4 days before admission and before learning she was pregnant. She was a first-time user of the drug, obtained it from an online source without medical consultation, and did not uptitrate it from lower doses as clinically practiced. The primary team confirmed the medication and its dosage via her online receipt. Before taking semaglutide, the patient reported having no gastrointestinal symptoms. The day she took the medication, she developed nausea and vomiting, prompting her to take a urine pregnancy test, which was positive. Her symptoms were intolerable, causing her to present to the hospital. Two days after admission and 6 days after taking semaglutide, the patient reported complete resolution of her symptoms and was discharged home. Her symptoms resolved suddenly, and no particular medical intervention by the primary team could be attributed to her drastic clinical improvement.

Discussion

HG is intractable NVP that occurs early in pregnancy—typically before 9 weeks’ GA but usually resolving by the 20th week [3]. It is located on the extreme end of the NVP spectrum resulting in anorexia, severe limitations in daily life activities, and weight loss [3, 4]. This can result in volume depletion and dehydration, acute kidney injury secondary to contraction alkalosis, and ketonemia/ketonuria [3, 4]. While NVP is generally benign and common (up to 52%), HG is less common (0.3% to 3%) and pathological, leading to multiple hospitalizations and poor maternal outcomes such as limited functionality, mood disorders, suicide, and termination of pregnancy [4]. Possible, unfortunate outcomes in fetuses and offspring include intrauterine growth restriction and low birth weight, respectively [3].

There are no clear causes for HG. Positive correlations between the incidence of HG and levels of hCG have been inconsistently observed [3]. The gastroenterologic side effect profile of estrogen and estradiol and their heightened levels in early pregnancy have also been explored but yielded dissonant results [3, 5]. Risk factors include multiparity, history of migraines and kinetosis, personal history of HG, and history of HG in first-degree relatives [5]. Cigarette smoking is protective against HG as it decreases levels of estradiol and hCG levels; this is supported by the decreased incidence of HG in cigarette smokers [5]. However, cigarette and nicotine cessation are highly encouraged to avoid fetal, maternal, and newborn pathologies including orofacial cleft, placenta previa, low birth weight, maternal thyroid disease, increased risk of respiratory infections in the child, childhood obesity, etc [5]. Growth differentiation factor 15 (GDF15) and insulin-like growth factor binding protein 7 (IGFBP7), which promote placentation and activate the area postrema and are upregulated during pregnancy, are plausible players in the development of HG [3, 6].

HG management includes pharmacologics, fluid, vitamin and electrolyte resuscitation, and nonpharmacological therapies. Pyridoxine, with doxylamine, is a common vitamin used in early pregnancy to curb NVP. Antiemetics like ondansetron, promethazine, dimenhydrinate, and metoclopramide are also used. Methylprednisolone can be used as a last resort and has been shown to reduce rehospitalizations when compared to promethazine [5]. IV fluid resuscitation with NS, dextrose, vitamins, minerals, and electrolytes is highly effective in correcting dehydration, nutrient deficiencies, electrolyte derangements, and emetic symptoms [6].

Here we present the case of a 34-year-old woman with 4 days of HG. Before discovering her pregnancy, she self-administered semaglutide. Semaglutide is a glucagon-like receptor analogue traditionally used as an add-on therapy for T2DM patients whose disease is uncontrolled (ie, hemoglobin A_1c >1.5% the acceptable target, or whose hemoglobin A_1c goal is not reached within 3 months of the previous medical regimen) and/or for those who have contraindications or intolerances to metformin [2, 7]. It is exceptionally reasonable and encouraged to start GLP-1 RA on T2DM patients with concurrent atherosclerotic cardiovascular disease or risk factors because of its cardioprotective effects; of note, only selected GLP-1 RAs, including semaglutide, are cardioprotective [2, 7]. It is administered as weekly subcutaneous injections of 0.25 mg, 0.5, 1, and 2 mg, or daily oral doses of 3, 7, and 14 mg [8, 9]. The risk of hypoglycemia with semaglutide is low [7]. Patients have experienced 12% loss of their body weight by the 28th week of use [8]. The subcutaneous form of semaglutide has a half-life of 7 days, a bioavailability of 89%, and reaches its maximum concentration in the body in 1 to 3 days [9]. Our patient's gastrointestinal symptoms resolved 6 days post exposure to the drug, coinciding closely with the half-life of subcutaneous semaglutide.

In June 2021, the Food and Drug Administration approved semaglutide's use for weight management. Specifically, it's been approved for overweight (BMI ≥27) with at least 1 weight-related comorbidity (T2DM, hyperlipidemia, hypertension) or obese (BMI ≥ 30) people [10]. The Food and Drug Administration mandated gradually increasing the dosage of semaglutide to 2.4 mg (maximum) over 16 to 20 weeks to minimize its gastrointestinal effects (nausea and vomiting [36.6%], diarrhea [15%], fatigue [6.3%], and abdominal pain [9%]) [8, 10].

Unfortunately, our patient obtained semaglutide from an online source and used it for the first time without medical supervision. Unbeknown to her, she was pregnant and pregnancy is a contraindication of semaglutide [2]. Not much is known about the side-effect profile of semaglutide in pregnant humans, but animal studies have shown risks of mortality, structural defects, and growth alterations in embryos and fetuses [1]. Therefore, the use of contraception and the discontinuation of semaglutide 2 months before conception are recommended due to its long washout period [1, 2, 9].

To conclude, this is the first case of semaglutide-induced HG reported in the literature. Semaglutide is an antidiabetic medication whose indication recently broadened to weight management in obesity, and overweight people with specific comorbidities. Its most notable side-effect profile is gastrointestinal disturbances (ie, nausea, vomiting, abdominal cramping). Therefore it is advised to uptitrate the dosage of semaglutide over 4 weeks to promote adherence; this is especially advised in those who have never been exposed to the medication (ie, treatment-naive) [9].

Our patient presented with HG that was resistant to guideline medical therapies. As a treatment-naive user, she injected a higher-than-recommended dose of semaglutide before knowing she was pregnant. The proposed positive relationship between hCG and NVP likely had an additive effect on the drug, resulting in intractable nausea, vomiting, and abdominal pain. Her symptom onset and resolution followed the pharmacodynamic and side-effect profile of semaglutide; her symptoms began 1 day after exposure to semaglutide and resolved suddenly 6 days after exposure.

Little is known about the use of semaglutide in pregnancy. There are no case studies on this matter to date. But the adverse effects displayed in animals do warrant precaution when considering its use in premenopausal women, and in pregnant patients with gestational or pregestational diabetes. The use of contraception, pregnancy screening, and alternative means for weight loss or diabetes management should be discussed between health-care providers and patients before initiating semaglutide.

Learning Points

• Semaglutide is an antidiabetic drug that is also indicated for weight management in obesity, and overweight people with specific comorbidities.

• Semaglutide has not been studied in human pregnancy studies and therefore is contraindicated in pregnancy.

• The prescription and careful titration by a licensed health-care provider is highly suggested to ensure semaglutide is being used in the intended patient population and monitored for common gastrointestinal disturbances.

Contributors

Each author has been individually involved in and has made substantial contributions to conceptions and designs, acquisition of data, analysis, interpretation of data, drafting, and editing the case report. S.C., I.O., and A.R.C. contributed to the diagnosis and management of this patient. I.O., A.R.C., S.C., and R.O. contributed to the designs, acquisition of data, analysis, interpretation of data, drafting, and editing of the case report. All authors reviewed and approved the final draft.

Abbreviations

GA

gestational age

GLP-1 RA

glucagon-like peptide-1 receptor agonist

hCG

β-human chorionic gonadotropin

HG

hyperemesis gravidarum

IV

intravenous

NS

normal saline

NVP

nausea and vomiting in pregnancy

T2DM

type 2 diabetes mellitus

Funding

No public or commercial funding. No funding is provided.

Disclosures

None to declare.

Informed Patient Consent for Publication

Signed informed consent could not be obtained from the patient or a proxy but has been approved by the treating institution.

Data Availability Statement

Data sharing is not applicable to this article as no data sets were generated or analyzed during the current study.