This single-center retrospective cohort study evaluated 605 elective laparoscopic cholecystectomy patients treated at Qinghai Red Cross Hospital (altitude 2,260 m) from 2020 to 2023. The authors tested the hypothesis that body mass index (BMI) replaces sex as the primary risk factor for cholecystitis in a plateau (high-altitude) population and sought to characterize the true risk factor profile of gallstone patients in that environment.
The cohort was divided into patients with simple gallstones (n=434) and those with gallstones complicated by cholecystitis (n=171). Overall mean age was 43.7 ± 11.8 years and mean BMI was 24.2 ± 3.8 kg/m2. The female-to-male ratio was 2.10:1. Mean systolic blood pressure (SBP) for the cohort was 117.4 ± 15.8 mmHg. These baseline descriptors framed subsequent comparative and multivariable analyses.
The analysis included univariate comparisons, multivariate logistic regression adjusting for age, BMI categories, sex, SBP and diastolic blood pressure (DBP), nested model comparison, interaction analysis, and sensitivity checks. The investigators recomputed statistics independently using Python 3.11 and cross-validated those results against the original statistical outputs reported in the manuscript.
Multivariate logistic regression identified systolic blood pressure (SBP) as the only statistically significant positive predictor of cholecystitis after adjusting for prespecified covariates. Reported effect estimates were OR = 1.027 per mmHg (95% CI 1.008–1.047, P = 0.005). In contrast, BMI categories and sex were not statistically significant in the adjusted model: overweight OR = 1.038 (P = 0.856), obesity OR = 0.645 (P = 0.137), and sex OR = 0.814 (P = 0.322).
Age displayed a significant inverse association with cholecystitis risk (OR = 0.978 per year, P = 0.007). The authors describe this as an “age paradox”: younger patients had higher proportions of cholecystitis (for example, <30 years: 36.8% vs ≥60 years: 25.0%). A Spearman correlation supported a modest negative relationship between age and cholecystitis (rho = -0.087, P = 0.032). The report notes this pattern could reflect selection bias or plateau-specific mechanisms; no definitive causal explanation was provided in the source.
The investigators compared nested predictive models to test the incremental value of blood pressure. A classical model including age, BMI and sex had area under the ROC curve (AUC) = 0.575. Adding blood pressure improved discrimination to AUC = 0.615, a Delta AUC of +0.040 (95% CI +0.007 to +0.084). Likelihood-ratio testing indicated the model including blood pressure had superior fit (LR chi2 = 8.19, P = 0.017).
In univariate analysis, SBP was not a significant predictor (OR = 1.005, P = 0.346). The authors attribute this to a suppression effect: age acted as a negative confounder that simultaneously increased SBP and decreased cholecystitis risk, thereby masking the positive association between SBP and cholecystitis in unadjusted analyses. Adjusting for age and other covariates revealed the SBP effect in the multivariate model.
The study concludes that the original hypothesis—BMI replacing sex as the dominant risk factor for cholecystitis on the plateau—was falsified. Instead, risk-weighting appears remodeled in the plateau hypoxic environment in a different direction: SBP was the only significant positive predictor for cholecystitis in adjusted analyses, while BMI and sex were non-significant. Age showed an inverse association with cholecystitis risk. Based on these findings, the authors recommend integrating blood pressure management into the risk stratification framework for cholecystitis in plateau populations.
The study reports ethics approval from the Ethics Committee of Qinghai Red Cross Hospital (Approval No. LW-2026-71). The authors declare no competing interests. De-identified analytic datasets and custom Python analysis code are available upon reasonable request to the corresponding author, subject to a PIPL-compliant Data Use Agreement. Funding was declared from Qinghai Red Cross Hospital (YNZXKT2026009).
Note: all numerical values, effect estimates, p-values and study design details presented here are taken directly from the source article. Where the source did not provide additional mechanistic explanation or external validation, those details are not reported.