Early pregnancy loss is a common complication of clinically recognized pregnancy and can have lasting psychological consequences. A pooled analysis estimated that 15.3% of recognized pregnancies end in miscarriage, although the estimated frequency depends on how pregnancy and loss are ascertained [1]. Maternal age and previous miscarriage are established risk factors, but they provide limited information about the viability of an individual pregnancy [2]. Transvaginal ultrasonography remains central to the assessment of first-trimester viability. When initial findings are inconclusive, conservative diagnostic criteria and appropriately timed repeat imaging help avoid misclassifying a viable pregnancy [3,4]. These limitations have prompted investigation of serum biomarkers that might complement clinical and sonographic assessment.

Anti-Müllerian Hormone (AMH) is a glycoprotein in the transforming growth factor beta family. In the ovary, granulosa cells of preantral and small antral follicles produce AMH, which participates in regulating follicular recruitment and development [5]. Circulating AMH reflects the population of growing follicles and is widely used to assess ovarian reserve and anticipate ovarian response to stimulation [6,7]. Its association with follicle quantity provides a rationale for studying reproductive outcomes, although serum AMH does not directly measure oocyte competence or embryonic chromosomal status. Concentrations can also change during pregnancy, making the timing of blood collection relevant when comparing AMH measured before conception with concentrations obtained during early pregnancy [8].

Studies of AMH and miscarriage after spontaneous conception have yielded differing results. In a prospective cohort of women aged 30-44 years who conceived naturally, Lyttle Schumacher et al. found that women with preconception AMH concentrations of 0.4 ng/mL or lower had a higher miscarriage hazard than women with concentrations of at least 1.0 ng/mL [9]. By contrast, the Effects of Aspirin in Gestation and Reproduction study found no association between low or high preconception AMH and clinical pregnancy loss among women with one or two previous losses [10]. In a smaller study of first-trimester spontaneous pregnancies, Kostrzewa et al. [11] reported associations with pregnancy loss at both low and high AMH concentrations relative to an intermediate range.

Differences in participant age, reproductive history, sampling time and AMH thresholds limit direct comparison across studies. The timing of outcome assessment is particularly important. A blood sample collected around the ultrasound diagnosis can identify an association with contemporaneous pregnancy viability. Estimating the risk of a subsequent loss requires measurement before the outcome and follow-up of pregnancies that are viable at enrolment. A systematic review found an association between diminished ovarian reserve and miscarriage in some populations, particularly after assisted conception, but the available evidence from spontaneous pregnancies was too heterogeneous for pooling [12].

The present study aimed to compare maternal serum AMH concentrations between women with first-trimester pregnancy loss and women with viable intrauterine pregnancies following spontaneous conception. Pregnancy status was classified by transvaginal ultrasonography. The study also examined the distribution of low, intermediate and high AMH concentrations in each group. This comparison addresses whether circulating AMH differs according to pregnancy viability at early assessment and provides a basis for investigating the association in prospective studies with clearly defined sampling and follow-up.

Patients and Methods

This single-center observational study compared maternal serum Anti-Müllerian Hormone (AMH) concentrations between women with nonviable and viable first-trimester intrauterine pregnancies following spontaneous conception. Participants were recruited consecutively during routine antenatal visits or emergency assessments at the Department of Obstetrics and Gynecology, Kadhimiya Teaching Hospital, Baghdad, Iraq, from 1 January to 1 December 2025. Pregnancy status was determined by transvaginal ultrasonography at clinical assessment.

Study Population

Women were eligible if they had conceived spontaneously, had a sonographically confirmed intrauterine pregnancy and agreed to participate. The stated gestational age at enrolment was 6-9 weeks. Exclusion criteria included conception following ovulation induction or assisted reproductive technology, multiple pregnancy, known uterine anomalies, chronic systemic disease such as diabetes mellitus or autoimmune disease, current hormonal therapy and incomplete clinical or laboratory data.

The analysis included 63 women. The pregnancy-loss group comprised 30 women in whom an intrauterine embryo was visualized without detectable cardiac activity and the pregnancy was classified as nonviable. The comparison group comprised 33 women with a viable intrauterine embryo and detectable cardiac activity at assessment.

Clinical and Ultrasonographic Assessment

Maternal age, gestational age at presentation, gravidity, parity, previous miscarriage, body mass index and ultrasound findings were recorded using a structured clinical form. Experienced sonographers performed the transvaginal examinations. The crown-rump length criteria, indications for repeat scanning and intervals between scans used to confirm nonviability should be specified from the original clinical records [3,4]. Viability in the comparison group refers to the status documented at the ultrasound examination.

Blood Sampling and Serum AMH Measurement

Approximately 5 mL of venous blood was collected into a plain tube and allowed to clot at room temperature. Samples were centrifuged at 3,000 rpm for 10 minutes. The serum was separated and stored until analysis. Serum AMH was measured in the same laboratory using an Enzyme-Linked Immunosorbent Assay (ELISA) according to the manufacturer’s instructions and expressed in ng/mL.

AMH was evaluated as a continuous measure and classified into three categories using thresholds previously reported by Kostrzewa et al. [11]. Concentrations below 1.1 ng/mL were classified as low, concentrations from 1.1 to 4.5 ng/mL as intermediate and concentrations above 4.5 ng/mL as high. The intermediate category served as the reference for categorical comparisons.

Statistical Analysis

Analyses were performed using Statistica version 12 (StatSoft, USA). Continuous variables were summarized as mean±standard deviation or median and interquartile range, as applicable. Categorical variables were summarized as frequencies and percentages. Normally distributed continuous variables were compared between groups using an independent-samples t test.

The multivariable logistic regression used pregnancy status as the outcome, with viable pregnancy as the reference. The reported model included two indicators for AMH category, maternal age of 30 years or older and a history of miscarriage. Associations were expressed as adjusted odds ratios with 95% confidence intervals. Receiver operating characteristic analysis was used to assess how well serum AMH distinguished the two groups at assessment. Discrimination was summarized by the area under the curve and its 95% confidence interval. Statistical significance was defined as p<0.05.

The analysis included 63 women, comprising 30 with first-trimester pregnancy loss and 33 with viable intrauterine pregnancies at assessment. The mean maternal age was 27.80 years in the pregnancy-loss group and  27.20 years  in  the  viable-pregnancy  group.

Previous miscarriage was reported in 11 of 30 women (36.67%) in the pregnancy-loss group and 6 of 33 women (18.18%) in the viable-pregnancy group (p = 0.098). None of the characteristics listed in Table 1 differed significantly between groups at the 0.05 level.

Table 1: Demographic and Obstetric Characteristics of the Study Groups

The reported mean serum anti-Müllerian hormone (AMH) concentration was lower in women with pregnancy loss than in women with viable pregnancies (1.62±1.54 Vs 2.94±1.21 ng/mL, p<0.001). The reported median concentrations were 1.20 and 2.60 ng/mL, respectively (Table 2 and Figure 1).

Table 2: Serum AMH Concentrations by Pregnancy Viability

Low AMH was more frequent among women with pregnancy loss than among women with viable pregnancies (36.67% Vs 9.09%). Intermediate AMH was the most frequent category in the viable-pregnancy group (75.76%). High AMH occurred in 23.33% of the pregnancy-loss group and 15.15% of the viable-pregnancy group. A Pearson chi-square test recalculated from the six reported cell counts gave χ²(2) = 9.35, p = 0.0093 (Table 3 and Figure 2).

Table 3: Distribution of Serum AMH Categories by Pregnancy Viability

Using intermediate AMH as the reference, the odds of pregnancy loss were higher in the low-AMH category (crude Odds Ratio [OR] 7.64, 95% Confidence Interval [CI] 1.79-32.58). The estimate for high AMH was less precise and its confidence interval included 1.00 (OR 2.92, 95% CI 0.77-11.12) (Table 4).

Table 4: Crude Odds of Pregnancy Loss by Serum AMH Category

The study reported adjusted ORs of 4.38 for low AMH and 3.21 for high AMH, compared with intermediate AMH. The reported confidence intervals for maternal age ≥30 years and previous miscarriage included 1.00. These estimates are reproduced in Table 5 pending verification against the original regression output.

Table 5: Reported Multivariable Logistic Regression Estimates for First-Trimester Pregnancy Loss

The study reported an area under the receiver operating Characteristic Curve (AUC) of 0.79 (95% CI 0.68-0.90). At the stated threshold of AMH <1.1 ng/mL, the counts in Table 3 yielded a sensitivity of 36.67% and a specificity of 90.91%. The reported and recalculated values are shown separately in Table 6.

Table 6: Reported ROC Results and Count-Derived Performance at the Low-AMH Threshold

At the threshold of AMH <1.1 ng/mL, 11 of the 14 women below the threshold had pregnancy loss, yielding a positive predictive value of 78.57% in the study sample. Among the 49 women at or above the threshold, 30 had viable pregnancies, yielding a negative predictive value of 61.22%. Overall, 41 of 63 women were correctly classified (65.08%) (Table 7).

Table 7: Positive and Negative Predictive Values and Accuracy at AMH <1.1 ng/mL

Among 63 women with spontaneously conceived intrauterine pregnancies, the reported mean serum anti-Müllerian Hormone (AMH) concentration was lower in those with first-trimester pregnancy loss than in those with viable pregnancies at assessment. Low AMH was also more frequent in the pregnancy-loss group. The clearest finding was therefore the difference at the lower end of the AMH distribution. The evidence for an association at higher concentrations was less precise. This distinction matters because early pregnancy loss has diverse causes and the interpretation of a biomarker depends on how pregnancy viability was established. In this study, group assignment rested on transvaginal ultrasonography, the principal investigation for assessing early intrauterine pregnancy viability [1-4].

Maternal age, gestational age at presentation and gravidity were similar on the reported comparisons. Previous miscarriage was more frequent in the pregnancy-loss group, affecting 36.67% of women compared with 18.18% of those with viable pregnancies. Although the difference did not reach statistical significance, its magnitude warrants attention. Previous reproductive performance is relevant to miscarriage risk and a small sample may yield imprecise estimates even when a characteristic differs meaningfully between groups. Reproductive ageing is also relevant because maternal age relates to both ovarian reserve and pregnancy outcome. The absence of a statistically significant age difference offers limited reassurance about residual confounding, particularly when age is reduced to a binary variable in the adjusted model [5-7].

The thesis reported mean AMH concentrations of 1.62 ng/mL in the pregnancy-loss group and 2.94 ng/mL in the viable-pregnancy group. AMH signalling is involved in ovarian follicular development and circulating concentrations reflect the pool of growing follicles. This provides a biological rationale for examining AMH in relation to reproductive outcomes. Its established applications in ovarian reserve assessment and assisted reproduction concern follicle quantity and ovarian response. A lower serum concentration in a pregnancy that has already been classified as nonviable cannot establish impaired oocyte competence or explain the cause of embryonic demise. The study did not include embryonic genetic testing or other measurements that could distinguish those possibilities [8-13].

The studies cited in the thesis also distinguish ovarian reserve from reproductive outcomes. Investigations of natural fecundability have examined whether AMH relates to the probability of conception, while assisted-conception studies have evaluated outcomes such as implantation and clinical pregnancy. These outcomes concern different stages of reproduction. An association with the response to ovarian stimulation or the probability of conception offers limited evidence about whether a spontaneously conceived pregnancy will remain viable. Age-specific AMH distributions further complicate comparisons across populations. These differences in population, timing and outcome definition limit the extent to which findings from fertility cohorts can explain the present comparison [14-18].

The category counts help clarify the pattern. Low AMH occurred in 11 of 30 women with pregnancy loss and 3 of 33 with viable pregnancies. The crude odds ratio for low versus intermediate AMH was 7.64, with a wide 95% confidence interval of 1.79-32.58. High AMH occurred in seven and five women, respectively. Its crude odds ratio was 2.92, with a confidence interval of 0.77-11.12. Although the point estimates for both extremes exceeded one, the high-category comparison was based on few observations. The thesis’s discussion of polycystic ovarian morphology and the relationship between AMH and adiposity suggests possible sources of heterogeneity. The available results do not establish that either factor explained the high-AMH findings in this sample [19-22].

Timing is central to interpretation. The study compared women after their pregnancies had been classified by ultrasound and the thesis did not specify when blood was drawn relative to the diagnosis of loss. AMH concentrations can vary during early pregnancy, so differences in the timing of sampling could affect the comparison [23,24]. Women assigned to the viable group had detectable embryonic cardiac activity at assessment. The thesis does not establish how many remained pregnant at a defined follow-up point. The observed difference therefore concerns viability around the time of assessment. Testing whether AMH predicts a subsequent loss would require sampling while viability is documented and following each pregnancy to a prespecified endpoint.

The adjusted model reported higher odds of pregnancy loss at both AMH extremes after accounting for maternal age category and previous miscarriage. These estimates remain provisional because the original regression output and the number of participants included in the fitted model were unavailable. The study included 30 pregnancy-loss events overall and only three viable pregnancies fell into the low-AMH category. The illustrative curve in the thesis cannot establish the shape of the association across continuous AMH values. Potential influences discussed in the thesis include maternal adiposity, thyroid dysfunction, stress and immunological factors. Their roles in this sample could not be assessed from the available results [25-28].

The reported discrimination also requires reconciliation with the observed counts. The thesis gives an overall area under the receiver operating characteristic curve of 0.79, but the individual AMH values or original ROC output are needed to verify that estimate. At the stated threshold of <1.1 ng/mL, only 11 of the 30 women with pregnancy loss were below the threshold, giving a sensitivity of 36.67%. Thirty of the 33 viable pregnancies were at or above it, giving a specificity of 90.91%. The low threshold therefore missed 19 of the 30 losses in the study sample. The positive and negative predictive values also reflect the proportion of losses in this comparison. Clinical evaluation of early pregnancy already integrates symptoms, ultrasound findings and, where appropriate, serial human chorionic gonadotropin measurements [3,4,29,30]. The study did not test whether AMH improved classification beyond that assessment.

Several source discrepancies limit what can be concluded. The reported upper quartile of AMH in the pregnancy-loss group requires reconciliation with the number of women above 4.5 ng/mL. The reported gestational-age standard deviations also require verification against the stated 6-9-week eligibility interval. The relative risks reported in the thesis do not reproduce from the category counts, so the revised Results present crude odds ratios calculated from those counts. Verification against the original laboratory records and statistical output is needed before the adjusted estimates or ROC findings can be interpreted. The small sample, single-centre setting and limited documentation of the assay further restrict how far the results can be generalized.

The supplied results support a provisional association between lower maternal serum AMH and nonviability at first-trimester assessment. The high-AMH estimate remains uncertain and the available data cannot establish a clinically useful threshold. A prospective study that documents viability before sampling, standardizes the assay and follows pregnancies to a defined endpoint would address whether AMH adds information beyond the clinical and sonographic findings already available at presentation.

1. Quenby, S. et al. "Miscarriage matters: The epidemiological, physical, psychological and economic costs of early pregnancy loss." The Lancet, vol. 397, no. 10285, 2021, pp. 1658-1667.
2. Annual Capri Workshop Group. "Early pregnancy loss: The default outcome for fertilized human oocytes." Journal of Assisted Reproduction and Genetics, vol. 37, no. 5, 2020, pp. 1057-1063.
3. American College of Obstetricians and Gynecologists' Committee on Practice Bulletins-Gynecology. "ACOG practice bulletin no. 200: Early pregnancy loss." Obstetrics and Gynecology, vol. 132, no. 5, 2018, pp. e207.
4. Jeve, Y. and R. Rana. "Accuracy of first-trimester ultrasound in the diagnosis of early embryonic demise: A systematic review." Ultrasound in Obstetrics and Gynecology, vol. 38, 2011, pp. 489-496.
5. Regan, L. et al. "Influence of past reproductive performance on risk of spontaneous abortion." BMJ, vol. 299, no. 6698, pp. 541-545.
6. Nelson, S.M. et al. "The ageing ovary and uterus: New biological insights." Human Reproduction Update, vol. 19, no. 1, 2013, pp. 67-83.
7. Hansen, K. et al. "A new model of reproductive aging: The decline in ovarian nongrowing follicle number from birth to menopause." Human Reproduction, vol. 23, 2008, pp. 699-708.
8. Hart, K.N. et al. "Structure of AMH bound to AMHR2 provides insight into a unique signaling pair in the TGF-β family." Proceedings of the National Academy of Sciences of the United States of America, vol. 118, no. 26, 2021.
9. di Clemente, N. et al. "Anti-Müllerian hormone in female reproduction." Endocrine Reviews, vol. 42, no. 6, 2021, pp. 753-782.
10. Moolhuijsen, L.M.E. and J.A. Visser. "Anti-Müllerian hormone and ovarian reserve: Update on assessing ovarian function." Journal of Clinical Endocrinology and Metabolism, vol. 105, no. 11, 2020, pp. 3361-3373.
11. Broer, S.L. et al. "Anti-Müllerian hormone: Ovarian reserve testing and its potential clinical implications." Human Reproduction Update, vol. 20, no. 5, 2014, pp. 688-701.
12. La Marca, A. et al. "Anti-Müllerian Hormone (AMH) as a predictive marker in Assisted Reproductive Technology (ART)." Human Reproduction Update, vol. 16, 2010, pp. 113-130.
13. Tremellen, K. et al. "Anti-Müllerian hormone: A marker of ovarian reserve." Australian and New Zealand Journal of Obstetrics and Gynaecology, vol. 45, 2005, pp. 20-24.
14. Hagen, C. et al. "Low concentration of circulating anti-Müllerian hormone is not predictive of reduced fecundability in young healthy women: A prospective cohort study." Fertility and Sterility, vol. 98, pp. 1602-1608.
15. Zarek, S. et al. "Is anti-Müllerian hormone associated with fecundability? Findings from the EAGeR Trial." Journal of Clinical Endocrinology and Metabolism, vol. 100, no. 11, 2015, pp. 4215-4221.
16. Steiner, A.Z. et al. "Antimüllerian hormone as a predictor of natural fecundability." Fertility and Sterility, vol. 95, no. 7, 2011, pp. 2423-2427.
17. Tal, R. et al. "Antimüllerian hormone as predictor of implantation and clinical pregnancy after assisted conception." Fertility and Sterility, vol. 104, no. 1, 2015, pp. 122-130.
18. Seifer, D.B. et al. "Age-specific serum anti-Müllerian hormone values for 17,120 women presenting to fertility centers." Fertility and Sterility, vol. 95, no. 2, 2011, pp. 747-750.
19. Dewailly, D. et al. "Definition and significance of polycystic ovarian morphology: A task force report." Human Reproduction Update, vol. 20, no. 3, 2014, pp. 334-352.
20. Kriseman, M. et al. "Antimullerian hormone levels are inversely associated with Body Mass Index (BMI) in women with polycystic ovary syndrome." Journal of Assisted Reproduction and Genetics, vol. 32, no. 9, 2015, pp. 1313-1316.
21. Bernardi, L.A. et al. "Relationship between obesity and anti-Müllerian hormone in reproductive-aged African American women." Obesity, vol. 25, no. 1, 2017, pp. 229-235.
22. Metwally, M. et al. "Body mass index and risk of miscarriage in women with recurrent miscarriage." Fertility and Sterility, vol. 94, no. 1, 2010, pp. 290-295.
23. Hamilton, K. et al. "Longitudinal changes in maternal serum concentrations of antimüllerian hormone in individual women during conception cycles and early pregnancy." Fertility and Sterility, vol. 106, 2016, pp. 1407-1416.
24. Stegmann, B. et al. "Changes in antimüllerian hormone levels in early pregnancy are associated with preterm birth." Fertility and Sterility, vol. 104, 2015, pp. 347-355.
25. Metwally, M. et al. "Does high body mass index increase the risk of miscarriage after spontaneous and assisted conception? A meta-analysis of the evidence." Fertility and Sterility, vol. 90, no. 3, 2008, pp. 714-726.
26. Maraka, S. et al. "Subclinical hypothyroidism in pregnancy: A systematic review and meta-analysis." Thyroid, vol. 26, no. 4, 2016, pp. 580-590.
27. Frazier, T. et al. "Weathering the storm: A review of pre-pregnancy stress and risk of spontaneous abortion." Psychoneuroendocrinology, vol. 92, 2018, pp. 142-154.
28. Deshmukh, H. and S.S. Way. "Immunological basis for recurrent fetal loss and pregnancy complications." Annual Review of Pathology, vol. 14, 2019, pp. 185-210.
29. Barnhart, K.T. et al. "Differences in serum human chorionic gonadotropin rise in early pregnancy by race and value at presentation." Obstetrics and Gynecology, vol. 128, no. 3, 2016, pp. 504-511.
30. Barnhart, K. "Early pregnancy failure: Beware of the pitfalls of diagnosis." Obstetrics and Gynecology, vol. 119, no. 2, 2012, pp. 360-367.