This retrospective single‑center study included 139 patients with pathologically or clinically confirmed Langerhans cell histiocytosis (LCH) admitted between October 2009 and March 2025. The cohort comprised 76 males and 63 females. Median age at diagnosis was 21 years with an interquartile range of 6 to 39 years.
Patients were categorized by the presence or absence of central diabetes insipidus (CDI) into a CDI group (n = 20) and a non‑CDI group (n = 119). Multivariate logistic regression was used to evaluate factors associated with CDI after adjusting for age, sex, and body mass index (BMI).
Single‑system single‑lesion LCH was the most common presentation, observed in 94 of 139 patients (67.6%). Multi‑system involvement occurred in 39 patients (28.0%). Among organ involvement patterns, bone was the most frequently affected site (107/139, 77.0%).
Central nervous system (CNS) involvement was identified in 24 patients (17.3% of the cohort).
Of the 24 patients with CNS involvement, 20 presented with CDI. Among those 20 CDI cases, 10 (50.0%) had concurrent anterior pituitary dysfunction, indicating that pituitary hormone deficits commonly coexisted with CDI in this subgroup.
Overall, CDI affected 20 of 139 patients in this series.
Comparative analysis between the CDI and non‑CDI groups showed distinct patterns of disease distribution. The CDI group had:
These differences were statistically significant (all P < 0.05) in unadjusted comparisons reported in the source.
After adjusting for age, sex and BMI, multivariate logistic regression identified several independent associations with the occurrence of CDI in patients with LCH. Reported effect estimates were:
Multisystem involvement without high‑risk organ involvement: OR = 6.868 (95% CI: 2.305–20.289). This pattern was positively associated with CDI.
Skin and soft tissue involvement: OR = 6.318 (95% CI: 1.109–35.983). Skin/soft tissue disease was independently associated with higher odds of CDI.
Thyroid involvement: OR = 7.212 (95% CI: 1.053–49.395). Thyroid involvement showed a positive association with CDI.
Factors associated with lower odds of CDI included:
Single‑system single‑lesion involvement: OR = 0.323 (95% CI: 0.118–0.885). This presentation was inversely associated with CDI.
Bone involvement: OR = 0.082 (95% CI: 0.027–0.253). Isolated bone disease was linked to substantially lower odds of CDI.
The reported effect sizes and confidence intervals are drawn directly from the source abstract.
These findings suggest that within this single‑center cohort of LCH patients, disease patterns involving multiple systems (but not the high‑risk organs liver, spleen, or bone marrow), the skin/soft tissues, or the thyroid gland were associated with markedly higher odds of developing central diabetes insipidus. In contrast, isolated bone disease and single‑system single‑lesion presentations carried lower odds of CDI.
Clinically, awareness that patients with multisystem disease without high‑risk organ involvement or with skin/thyroid involvement may be at elevated risk for CDI could inform surveillance strategies and earlier endocrine evaluation, particularly when CNS involvement is suspected.
Limitations: the source is an abstract of a retrospective study and does not provide full methodological details in the available text. Specific diagnostic criteria for CDI and LCH, timing of onset, treatment regimens, duration of follow‑up, and other potential confounders were not reported in the abstract. All reported statistics and subgroup counts are those presented in the source document.
All authors declared no conflicts of interest in the published report.