Risk of neuroblastoma among patients with pediatric Horner syndrome.

PURPOSE: To determine the risk of neuroblastoma in children diagnosed with Horner syndrome after infancy.

METHODS

Children (<18 years of age) with an incident diagnosis of Horner syndrome were identified in the TriNetX Research Network (January 1, 2005-January 1, 2022). Patients <1 year were excluded to minimize congenital cases. A comparison non-Horner cohort of children undergoing routine child health examinations was followed for 5 years from the index date. Neuroblastoma was identified in the record using diagnosis and oncology classification codes. Propensity score matching adjusted for age, sex, race, and ethnicity; hazard ratios (HRs) and adjusted HRs (aHRs) were calculated. Sensitivity analyses excluded patients with prior trauma, carotid dissection, or upper-lobe lung neoplasm.

RESULTS

The Horner cohort included 1,331 children (56% male) from 56 health care organizations (mean age, 3.0 years) with 2,962,242 children comprising the unmatched comparison cohort. Within 5 years, 107 of 1,331 children (8.04%) with Horner syndrome and 595 of 2,962,242 controls (0.02%) developed neuroblastoma (HR = 446.90; 95% CI, 363.75-549.06; P < 0.0001). After matching (n = 1,331 per group), 5-year risk remained higher in the Horner cohort (8.03% vs 0.75%; aHR = 119.03; 95% CI, 16.61-852.79). In sensitivity analysis excluding patients with history of trauma, carotid dissection, or upper lobe lung masses, 89 of 1,148 children with incident Horner syndrome (7.75%) developed neuroblastoma.

CONCLUSIONS

In our study cohort, noncongenital Horner syndrome was associated with a markedly increased risk of neuroblastoma (∼1 in 12 within 5 years). Neuroblastoma evaluation should be strongly considered for acquired or otherwise unexplained noncongenital pediatric Horner syndrome.