According to the Windsor definition, hyperemesis gravidarum (HG) is defined as a condition characterized by severe nausea and/or vomiting, and restriction of eating or drinking during the day, which begins during the early week of pregnancy (before the sixteenth week of pregnancy) [1]. The prevalence of HG during pregnancy is approximately 0.3–3%, with variations attributable to differing diagnostic criteria and ethnic differences among studies [2]. HG can cause abnormalities in hematological and biochemical parameters [3]. The most commonly cited diagnostic criteria for HG include persistent unexplained vomiting, ketonuria, and significant weight loss, typically defined as at least a 5% reduction from pre-pregnancy body weight [2]. Liver dysfunction occurs in up to 3% of pregnancies and can be associated with pregnancy-related liver damage, exacerbation of a pre-existing liver condition, or coincidental events occurring during pregnancy. Even in the absence of a true liver disease, HG can lead to elevated liver enzyme levels [4]. A combination of liver biochemical tests, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), and bilirubin, is typically used to assess liver health [5]. Bile acids (BAs) are naturally present in the enterohepatic circulation and play a crucial role in bile formation, bile lipid excretion, and lipid absorption from the intestines [6]. BAs, which are significant signaling molecules, are recognized as playing a crucial role in the development of liver damage [7]. Elevated levels of BAs induce the production of pro-inflammatory mediators in hepatocytes, attracting immune cells, initiating inflammation in the liver, and ultimately causing liver damage [8]. Both fasting and postprandial serum BA concentrations are generally more sensitive than bilirubin, AST, and ALT in detecting hepatobiliary disease [9]. This study hypothesized that fasting bile acid (FBA) levels could be a significant marker for HG. To test this hypothesis, we evaluated FBA levels, ALT, AST, total and direct bilirubin, and urine ketone levels in pregnant women diagnosed with HG. Ultimately, the study sought to investigate the relationship between HG and FBA levels.

This retrospective cohort study was conducted using blood samples obtained from healthy pregnant women and those diagnosed with HG who presented to the Department of Obstetrics and Gynecology at Istanbul Training and Research Hospital. The study received approval from the Ethics Committeefor Non-Interventional Clinical Studies of Istanbul Training and Research Hospital on March 10, 2023 (Approval Number: Ref No:55). HG can be diagnosed when nausea and/or vomiting are severe enough to prevent normal eating and drinking and significantly restrict activities of daily living. Dehydration symptoms are crucial for diagnosis. Patients presenting in the first trimester of pregnancy with severe vomiting and/or nausea that prevent normal eating and drinking, significantly limit daily activities and show signs of dehydration are diagnosed with HG according to the Windsor definition [1].

The study included a total of 50 pregnant women in the first trimester of pregnancy, comprising 25 diagnosed with HG and 25 healthy control subjects, all aged between 20 and 45 years. The study excluded individuals with systemic diseases, those currently taking medications, and those who experienced multiple pregnancies. Women diagnosed with HG received hydration therapy before having their blood collected, which was done before any prescriptions were administered. For each participant, serum levels of AST (measured by photometric pyridoxal phosphate (PLP) method, intra-assay coefficient of variation (CV): 1.9 inter-assay CV: 2.9, normal range: 0–32 U/L), ALT (measured by PLP method, intra-assay CV: 2.1; inter-assay CV: 2.7, normal range: 0–33 U/L), total bilirubin (TBS) (measured by photometric diazo method, intra-assay CV: 1.03; inter-assay CV: 2.59, normal range: 0–1.1 mg/dL), direct bilirubin (measured by photometric dihydro pyrimidine dehydrogenase (DPD) method, intra-assay CV: 3.82%; inter-assay CV: 4.27, normal range: 0–0.3 mg/dL), urine ketone levels (measured by quantitative colorimetric method, intra-assay CV: 2%; inter-assay CV: 4%, 5 is negative result), and FBA (measured by enzyme cycling method, intra-assay CV: 3.9; inter-assay CV: 2.9, normal range: 0–10 µmol/L) levels were assessed.

Blood samples were collected from outpatient HG patients and the healthy control group after a 9-hour fast and a new appointment. For each patient, data on age, gravidity, parity numbers, height, weight, body mass index (BMI), and weeks of gestation (measured via ultrasonography (USG)) were recorded. The laboratory work was conducted in the Medical Biochemistry Laboratory at Istanbul Training and Research Hospital. Blood samples, collected in monovette tubes after a 10–12 hour fast, were centrifuged at 4000 rpm for 10 minutes. The samples were then analyzed for total AST and ALT levels using the Roche Cobas 6000 integrated with the C501 Module (California, USA) and for urine tests using the Dirui FUS-200/H-800 (Changchun, Jilin, China). FBA levels were determined at reference laboratories. Blood collected in monovette tubes was centrifuged at 3000 rpm for 10 minutes and measured enzymatically using 3a-hydroxysteroid dehydrogenase. BA profiles were determined using the Wakefield-Vette HPLC-10 (Nashua, New Hampshire, USA). Any results that were incongruent or suspicious were resampled and re-analyzed.

The average age of the pregnant women participating in our study was 28.2 $\pm$ 4.9 years, and their BMI was 24.6 $\pm$ 2.0 kg/m². Of the participants, 27 (54.0%) were nulliparous, while 23 (46.0%) were multiparous. The participating pregnant women had no pre-existing diseases and were not on any medication before pregnancy. The average gestational age measured by USG was 8.3 $\pm$ 1.6 weeks. Among the 25 patients diagnosed with HG, 9 (18.0%) had a urine ketone value of I+, 9 (18.0%) had II+, and 7 (14.0%) had III+ (see Table 1). Table 1 shows the demographic characteristics of the study participants. No statistical tests were applied to the data presented in this table, so no specific tests are mentioned.

In our study, the gestational ages for the group diagnosed with HG and the healthy control group were 8.1 $\pm$ 1.7 weeks and 8.4 $\pm$ 1.5 weeks respectively. Both groups presented similar gestational ages as measured by USG. Moreover, there were no significant differences between the healthy control group and the HG group in terms of age, BMI, parity, AST (U/L), ALT (U/L), total bilirubin (mg/dL), and direct bilirubin (mg/dL) values (p $>$ 0.05) (Table 2).

In the group diagnosed with HG, the FBA levels were 2.6 $\pm$ 2.1 µmol/L, compared to 1.5 $\pm$ 1.3 µmol/L in the healthy control group. The FBA levels in the HG group were significantly higher than those in the healthy control group (p 0.05) (Table 2).

HG is a condition characterized by persistent nausea and vomiting, which can cause abnormalities in hematological and biochemical parameters during pregnancy. Gaba and Gaba [10], in a retrospective study, found ALT and AST levels to be significantly higher in patients admitted or undergoing day care for HG, excluding those who were not investigated and those with pre-existing or newly diagnosed liver disease. An increase was also observed in total serum bilirubin (TBS) levels [10]. In our study, serum ALT, AST, TBS, and other values were found within normal limits. Couse and Syed [11] found that in patients diagnosed with HG, the use of ondansetron (ODSN)—an antiemetic— was associated with increased serum ALT and AST levels. In our study, we recruited pregnant women diagnosed with HG who did not receive drug treatment and found that their ALT and AST levels remained within normal limits. Chraïbi et al. [12] in a retrospective study of 109 pregnant women diagnosed with HG, found that 39.6% of the patients had elevated liver function test (LFT), and 35.28% had elevated ALT levels. Unlike our study, they also examined prothrombin time and vitamin K levels and found elevated serum ALT and LFT levels. Pornchai et al. [13], reported increased levels of ALT and bilirubin, contrary to our study. Sari et al. [14] in their study on pregnant women diagnosed with HG, examined serum hemoglobin, hematocrit, white blood cells, platelets, fasting blood sugar, creatinine, urea, lipid profile, AST, ALT, sodium, potassium, thyroid stimulating hormone (TSH), and ischemia-modified albumin (IMA). They found higher IMA levels in patients with HG [14]. In our study, we examined serum LFT, FBA, and ketonuria levels in pregnant women with HG. Aslan et al. [3] examined ketones and hematological parameters in the urine of patients with HG. They compared the neutrophil-lymphocyte ratio (NLR) and platelet-lymphocyte ratio (PLR) based on the severity of ketonuria. They observed that both NLR and PLR increased with the severity of the ketonuria [3].

Niemeijer et al. [15] in their review meta-analysis of HG patients, which covered 81 biomarkers across 9 articles, found that ketonuria was reported in 65% of the studies. In 5 of the studies, no relationship was found between ketonuria and the severity of HG. Studies conducted on human chorionic gonadotropin (hCG), thyroid hormones, leptin, estradiol, progesterone, and white blood count have shown inconsistent relationships with HG. However, lymphocyte counts tended to be higher in women with HG [15]. In our study, no relationship was found between ketonuria and the severity of HG. Importantly, the present study identifies serum LFT and FBA values as potential biomarkers. Koot et al. [16] observed 148 patients with HG who exhibited ketonuria ranging from 1+ to 4+ in their study. They reported no relationship between ketonuria and the severity of HG [16]. Çintesun et al. [17], in their study on pregnant women diagnosed with HG, investigated the relationship between ketonuria and hematological parameters, including mean platelet volume (MPV), red cell distribution width (RDW), platelet distribution width (PDW), NLR, PLR, platelet crit (PCT), and complete blood count (CBC). Despite the controversial relationship between ketonuria and the severity of HG, no marker other than RDW was found to be related to the degree of ketonuria. Furthermore, they reported that as the degree of ketonuria increased, the RDW parameter also increased [17]. Soysal et al. [18] reported in their study that inflammation parameters such as monocyte-lymphocyte ratio (MLR), NLR, and PLR increased with ketonuria in patients diagnosed with HG. In our study, we examined serum LFT, FBA, and ketonuria in patients diagnosed with HG. We found that serum FBA levels increased in conjunction with the presence of ketonuria in the urine, while LFT values remained unchanged. Our literature review did not identify any studies that assessed FBA levels in pregnant women diagnosed with HG.

In the context of HG, the potential utility of FBA levels as an objective biomarker warrants further exploration. Elevations in FBA levels may hold significance for the early detection and targeted therapeutic intervention of hepatic anomalies. Early detection of hepatic anomalies through FBA-level monitoring could lead to reduced hospitalization durations, optimized healthcare resource allocation, and potential cost savings. Additionally, the role of FBA levels in the pathophysiology of HG presents a promising avenue for further investigation. Elucidating this relationship could provide insights into additional metabolic pathways and their interactions. Given the limited available literature on this topic, there is a compelling need for extensive research, including studies with larger patient cohorts, to provide more definitive insights to the clinical community.

The data sets generated and analyzed during the current study are available in the Istanbul Training and Research Hospital data repository. The datasets used and analyzed during the current study are available from the corresponding author upon reasonable request.

AKK, AC, CT, and DB designed the research study. AKK, AC, and CT performed the research. AKK and DB provided help and advice on the results. AKK and CT analyzed the data. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects.

Approval was obtained from the Ethics Committee for Non-Interventional Clinical Studies of Istanbul Training and Research Hospital, dated 10.03.23, with decision number 55. Informed consent has been obtained.

We would like to express our sincere gratitude to all those who contributed to this research and the preparation of this manuscript. We extend our thanks to the patients who participated in the study. We are grateful to the staff of the Istanbul Training and Research Hospital Gynecology and Obstetrics Clinic for their assistance in sample collection and data management. Special thanks to the Ethics Committee for Non-Interventional Clinical Studies of Istanbul Training and Research Hospital, for their approval of this study. We also acknowledge the valuable input and feedback from our colleagues and reviewers, which greatly improved the quality of this work.

This research received no external funding.

The authors declare no conflict of interest.