BACKGROUND: Langerhans cell histiocytosis (LCH) is a rare histiocytic disorder of unknown etiopathogenesis. Central diabetes insipidus (CDI) is the most frequent endocrine manifestation and is a known risk factor for the development of further anterior pituitary hormone deficiencies (APD). However, not all CDI patients develop APD, as observed during prolonged periods of follow-up. AIM: To find predictors of developing APD in LCH children with CDI followed in our institution. METHODS: We retrospectively analysed 44 patients over a median period (quartiles) of 12.3 years (8.79-14.24). Patients were subdivided into group 1 and group 2, according to absence or presence of APD, respectively. The main variables studied were: (1) chronological age (CA) at LCH diagnosis, (2) the primary site of LCH at diagnosis: low risk (LR) and multisystemic risk organs, and (3) the presence of reactivation. RESULTS: Multivariate Cox regression analysis showed that APD was positively associated with CA at LCH diagnosis [relative risk (RR) 1.14, p < 0.01], the LR clinical form (RR 8.6, p < 0.03), and negatively associated with the presence of reactivations (RR 0.3, p < 0.01). CONCLUSIONS: Patients with older CA at LCH diagnosis, LR clinical forms, and fewer reactivation episodes might represent a subgroup of paediatric LCH CDI patients with a higher risk of developing APD.
BACKGROUND:Langerhans cell histiocytosis (LCH) is a rare histiocytic disorder of unknown etiopathogenesis. Central diabetes insipidus (CDI) is the most frequent endocrine manifestation and is a known risk factor for the development of further anterior pituitary hormone deficiencies (APD). However, not all CDIpatients develop APD, as observed during prolonged periods of follow-up. AIM: To find predictors of developing APD in LCH children with CDI followed in our institution. METHODS: We retrospectively analysed 44 patients over a median period (quartiles) of 12.3 years (8.79-14.24). Patients were subdivided into group 1 and group 2, according to absence or presence of APD, respectively. The main variables studied were: (1) chronological age (CA) at LCH diagnosis, (2) the primary site of LCH at diagnosis: low risk (LR) and multisystemic risk organs, and (3) the presence of reactivation. RESULTS: Multivariate Cox regression analysis showed that APD was positively associated with CA at LCH diagnosis [relative risk (RR) 1.14, p < 0.01], the LR clinical form (RR 8.6, p < 0.03), and negatively associated with the presence of reactivations (RR 0.3, p < 0.01). CONCLUSIONS:Patients with older CA at LCH diagnosis, LR clinical forms, and fewer reactivation episodes might represent a subgroup of paediatric LCH CDIpatients with a higher risk of developing APD.