This prospective single-center observational study was conducted at the Jiangsu Province Hospital of Chinese Medicine (Nanjing, China) from May 2017 to February 2022.

A total of 185 patients with a confirmed diagnosis of Crohn’s disease (CD) according to World Health Organization criteria (WHO) were enrolled in the experimental cohort. All CD patients were aged between 16 and 40 years for females, or 16 and 45 years for males, and were diagnosed and managed by the specialized Inflammatory Bowel Disease Center of our hospital. For baseline comparison, 58 non-CD individuals were recruited from the Department of Colorectal Surgery during the same period to match the demographic profile. These controls underwent routine health check-ups or were hospitalized for benign, non-inflammatory anorectal conditions (e.g., hemorrhoids, anal fissures) and had no history or evidence of inflammatory bowel disease or other chronic gastrointestinal inflammatory disorders.

Inclusion criteria for all participants were provision of written informed consent and age as specified above. The exclusion criteria (applied to both groups) were: (1) Pregnancy, lactation, or any known condition that could substantially affect sex hormone levels (e.g., endocrine disorders such as pituitary adenoma, hypothyroidism, or Cushing’s syndrome). (2) History of severe cardiac, hepatic, renal dysfunction, or major cardiovascular disease prior to enrollment. (3) Use of medications known to significantly influence serum prolactin (PRL) levels within 3 months before enrollment, including dopamine receptor antagonists (e.g., antipsychotics, metoclopramide), certain antidepressants (e.g., SSRIs, tricyclic antidepressants), opioids, and oral contraceptives.

Additionally, for CD patients, we specifically verified that no factors existed (as listed above) that could cause significant fluctuations in sex hormones prior to the diagnosis of CD, ensuring the observed hormonal changes were more likely associated with the disease itself. Participants who voluntarily withdrew from the study during follow-up were also excluded from the final analysis.

The study protocol was approved by the Ethics Committee of Jiangsu Province Hospital of Traditional Chinese Medicine (approval no. 2019NL-063-030), and all procedures were conducted in accordance with the Declaration of Helsinki.

The general data from patients in the experimental and control groups, including age, sex, and reproductive history, were collected. Additional information regarding the patients in the experimental group was collected from the Electronic Medical Records System of the Jiangsu Province Hospital of Chinese Medicine, which included the Montreal Classification, CRP, ESR, FC, Perianal Disease Activity Index (PDAI), Crohn’s Disease Activity Index (CDAI), body mass index (BMI), serum albumin, hemoglobin, folate, surgical history, and history of biological agents.

Blood samples were collected in the early morning (8:00–10:00 a.m.), and for female participants, on the third day of menstruation. Serum prolactin (PRL) levels were determined using a chemiluminescence immunoassay (Access Prolactin assay, Beckman Coulter) on an Access/UniCel DxI series system. This assay is a two-site immunoenzymatic (“sandwich”) assay employing paramagnetic particles coated with a mouse monoclonal anti-PRL antibody and a goat polyclonal anti-PRL antibody conjugated to alkaline phosphatase. It primarily detects monomeric prolactin. According to the manufacturer’s insert, the established reference range for premenopausal women (<50 years) is 3.34–26.72 ng/mL. Considering that the female participants in our study were relatively young (aged 16–40 years), the sex-specific upper reference limits validated and used by our institutional clinical laboratory for this population were applied: >13.1 ng/mL for men and >25 ng/mL for women. PRL levels exceeding these limits were classified as elevated.

Among all enrolled Crohn’s disease patients, those rehospitalized within 14–30 weeks for treatment or diagnostic purposes were included in follow-up assessments. At each visit, disease activity, treatment regimen, and laboratory indices (PRL, CRP, Hb, albumin, FC) were reassessed. Dynamic changes in PRL were analyzed alongside inflammatory indicators and clinical scores (CDAI and PDAI) to explore whether PRL could serve as a potential predictor of disease activity and therapeutic efficacy.

All patients included in the experimental group were diagnosed with CD using a specialized medical structure and met the World Health Organization criteria (16) for the diagnosis of CD. Patients with CD were classified based on the universal classification criteria for the Montreal typology (17) and the PDAI (18, 19).

To ensure that observed prolactin (PRL) alterations were more specifically associated with Crohn’s disease rather than confounded by other common systemic conditions, we proactively collected and scrutinized additional laboratory parameters and medical history. Renal function was assessed by serum creatinine, and hematologic diseases were screened via white blood cell (20, 21). Furthermore, a detailed review of medical history, imaging reports (where available), and clinical presentations was conducted to exclude participants with known or suspected pituitary lesions, such as prolactinomas, in accordance with current diagnostic consensus (22). This comprehensive approach aimed to minimize the influence of renal, hematologic, and primary pituitary disorders on serum PRL levels. Given that endocrine dysfunction, including abnormalities in the thyroid and other axes, has been documented even in pediatric CD populations (14), our stringent exclusion of known endocrine disorders was crucial to isolate the relationship between PRL and intestinal inflammation.

However, it is important to acknowledge the study’s limitations in this regard. Due to the observational nature of this study and its primary focus on Crohn’s disease activity, systematic pituitary imaging (e.g., MRI) and dynamic testing of other endocrine axes (such as thyroid-stimulating hormone, TSH) were not part of the standardized protocol for all participants. While our meticulous history review aimed to exclude clinically apparent endocrine disorders, we recognize that the absence of these objective measures means subclinical pituitary or other endocrine pathologies could not be definitively ruled out as potential confounders for PRL levels. This limitation is considered in the interpretation of our findings.

All statistical analyses and graphical representations were performed using SPSS (version 25.0; SPSS, Chicago, IL, USA) and GraphPad Prism 9 (GraphPad Software, San Diego, USA).

Categorical variables were expressed as numbers (percentages) and continuous variables as mean ± SD or median (IQR). Chi-square tests and t-tests or Mann–Whitney U tests were used for group comparisons. Binary logistic regression was applied to identify predictors of PRL elevation, with results presented as odds ratios (ORs) and 95% confidence intervals (CIs).

All analyses were performed using SPSS 25.0 and GraphPad Prism 9.0, with p < 0.05 considered statistically significant. The optimal probability cutoff for logistic regression models was determined by maximizing the Youden index.

Prolactin concentrations are reported to one decimal place in accordance with the analytical precision of the immunoassay.

A total of 185 patients with Crohn’s disease (134 males [72%] and 51 females [28%]) were enrolled in the experimental group, while 58 non-CD individuals (39 males [67%] and 19 females [33%]) served as controls. The median age was 26.6 years (IQR 16–44) in the CD group and 29.8 years (IQR 16–45) in the control group. Regarding reproductive history, 129 CD patients (69.7%) and 16 controls (27.6%) were nulliparous, showing a significant difference between groups (p < 0.001). The median serum prolactin (PRL) level was 18.3 ng/mL (IQR 5.1–99.9) in CD patients and 12.3 ng/mL (IQR 4.6–33.8) in controls, also significantly higher in the CD group (p < 0.001). A graphical comparison of serum prolactin levels between groups and subgroups was presented in Figure 1. Additionally, renal function and routine hematological parameters were comparable between the two groups: serum creatinine levels (CD: 70.21 ± 13.54 μmol/L, control: 71.48 ± 12.52 μmol/L, p = 0.570) and white blood cell counts (CD: 6.45 ± 1.77 × 10⁹/L, control: 6.06 ± 1.69 × 10⁹/L, p = 0.182) showed no significant difference. The vast majority of participants had values within the normal reference range for both parameters (normal creatinine: 100% in both groups; normal WBC: 96.1% in CD group, 97.3% in control group). Overall, 112 participants (46%) in the total cohort had elevated PRL levels, whereas 131 (54%) showed normal PRL concentrations (Table 1). Notably, approximately 95% of the CD patients in this cohort had a history of perianal surgery, which is much higher than reported in typical CD populations. This indicates that our study population is heavily enriched for patients with severe, perianal-dominant Crohn’s disease, and thus the findings should be interpreted with caution regarding generalizability to broader CD populations.

Notably, the control group had a significantly higher proportion of participants with a history of childbirth compared to the CD group (72.4% vs. 30.3%, p < 0.001). This difference reflects a disparity in reproductive stage that may influence PRL levels and is consistent with the well-documented epidemiology of Crohn’s disease, where patients—particularly those with active, severe, or surgically treated disease—often have lower fertility rates and fewer pregnancies compared to the general population (23, 24).

Prolactin normalization was defined as a reduction from an elevated level at baseline to within the normal reference range at the follow-up assessment. Patients with CD showed a significant increase in PRL levels (T = 5.91, p < 0.001) and a significant decrease in the reproductive history (X² = 32.54, p < 0.001). No significant differences were observed regarding sex (X² = 0.57, p = 0.51) or female age (p = 0.289). After dividing all patients into two groups according to the PRL level, a significant difference in sex, age (T = −3.94, p < 0.001), and CD (X² = 36.31, p < 0.001) was observed in the group with abnormal PRL levels. PRL values were assigned as normal or abnormal depending on the reference PRL levels for men and women. A total of 106 patients with CD (57.2%) in the experimental group had abnormal PRL levels, while only 7 patients (12%) in the control group had abnormal PRL levels (Figures 1, 2).

Baseline data for all 185 patients with CD were recorded in quantitative or qualitative detail, including sex, age, Montreal staging, indicators of specific inflammation, and nutritional status, as well as previous modalities of treatment received (Table 2) for inclusion in correlation and single multifactorial analyses.

Table 3 shows the clinical characteristics of the 185 patients with CD based on normal or non-normal PRL levels. Significant differences were found in age (p < 0.001), sex (p = 0.016), reproductive history (p = 0.001), FC (p = 0.017), and folic acid (p = 0.035) levels between patients with CD with normal and abnormal PRL levels. However, the two groups showed similar distributions of hematocrit, CRP, CDAI, PDAI, hemoglobin, albumin, BMI, age at diagnosis, Montreal staging, treatment options for biological agents, and surgery (all p > 0.05). Possible clinical characteristics (p < 0.20) were incorporated into a logistic multifactorial analysis. The analysis revealed that older age (OR = 0.898, p = 0.011) and increased FC levels (OR = 0.999, p = 0.012) were significantly associated with a lower likelihood of having abnormal PRL levels. In the ROC validation curve after logistic prediction based on the above clinical characteristics, the area under the curve was 0.742 (Figure 3).

Information on 99 patients (male, 71; female, 28) was collected at the 14–30 weeks follow-up. Sixty-seven patients comprised the abnormal PRL group, whereas the other 32 patients had normal PRL levels. During the follow-up period, 6 patients underwent abdominal surgery (6.1%), 96 patients underwent perianal surgery (97%), and 95 patients were in the course of therapy (95.9%) during their hospitalization.

Serum prolactin and other biomarkers were measured at baseline and reassessed at a single follow-up visit within 14–30 weeks after treatment initiation. After 14 to 30 weeks of treatment cycles, patients in the abnormal PRL group had significantly lower CRP levels than those in the normal PRL group (p = 0.002). Furthermore, FC levels were significantly lower in the abnormal PRL group (p = 0.014), whereas other markers of inflammation and indicators of nutritional status did not show any significant differences between the two groups after treatment (all p > 0.05) (Figure 4).

To visualize the individual response patterns, longitudinal trajectories of serum PRL levels are presented in Figure 5. Among the 61 patients analyzed, 14 were female and 48 were male. A striking disparity was observed: 12 of 14 female patients (85.7%) achieved PRL normalization at follow-up, and all of them showed a clear downward trend. In contrast, only 13 of 47 male patients (27.6%) achieved normalization. Notably, the male cohort as a whole did not demonstrate a consistent pattern of decline, with many trajectories showing fluctuation or persistent elevation above the gender-specific upper limit of normal (ULN). This graphical analysis underscores a pronounced gender difference in both the rate of PRL normalization and the pattern of its change during therapy. The differential trajectories suggest that the administered treatments did not uniformly suppress PRL levels. Instead, PRL normalization appeared to be intrinsically linked to a more favorable anti-inflammatory response, which was markedly more prevalent in female patients (Figure 5).

Focusing on the 61 patients with abnormal baseline PRL, PRL levels returned to normal after treatment in 25 patients (40.9%), while levels remained elevated in the remaining 36 patients (59.1%). Logistic regression analysis of the two groups revealed that high CRP levels (OR = 1.034, p = 0.44) and high Hb values (OR = 1.078, p = 0.002) were risk factors for abnormal PRL maintenance (Table 4). The area under the curve in the logistically predicted roc validation curve based on the follow-up data was 0.757 (Figure 6).