This study aimed to investigate whether serum amyloid A (AA) level can be used as a biomarker in women with hyperemesis gravidarum (HEG). This prospective observational study was conducted at the Antalya Training and Research Hospital Gynecology and Obstetrics Clinic, Türkiye, between July and December 2023. Forty women diagnosed with HEG and 40 healthy women were included. No statistically significant differences were observed between the groups in terms of sociodemographic data such as age, body mass index, family history, educational status, economic level, place of residence, occupation, smoking and alcohol use, or drug habits. However, obstetric characteristics such as number of miscarriages, number of dilatation curettages, and gestational age and laboratory values including complete blood count, hematocrit, leukocyte, neutrophil, lymphocyte, platelet, free T4, albumin, alanine aminotransferase, aspartate aminotransferase, urea, creatinine, hs-C-reactive protein, and sodium (P > .05) all differed significantly. In addition, significant differences were observed between the HEG and healthy groups in terms of numbers of gravidities (2 [1-3] vs 1 [0-1], respectively, P < .001), numbers of parities (1 [0-1] vs 1 [0-1], P < .001), numbers of living children (1 [0-2] vs 1 [0-1], P < .001), presenting complaints (nausea 0 [0%], nausea + vomiting 0 [0%], none 40 [100.0%] vs nausea 27 [67.5%], nausea + vomiting 13 [32.5%], none 0 [0%], P < .001), serum thyroid-stimulating hormone (1.16 ± 0.56 vs 1.81 ± 0.624, P = .004), potassium (4.1 ± 0.7 vs 3.8 ± 0.2, P = .001), and AA values (7.29 ± 2.61 vs 10.74 ± 3.04, P < .001). At receiver operating characteristic analysis, the area under the curve (AUC: 0.881) was statistically significant for serum AA (P: <.001), with a cutoff value of ≥ 8.79 ([95% confidence interval] 0.743-0.919, sensitivity 87.4%, specificity 80.2%). The positive predictive value of serum AA was 81.1% and the negative predictive value was 80.4%. The study results showed that serum AA can be used as a diagnostic biomarker in HEG. Prospective studies involving more participants are now required to confirm our results.