Birth weight, ototoxic medication, and surgical history predict individual hearing loss risks: a systematic review and meta-analysis

Background Hearing loss (HL) impairs sound perception and includes sensorineural, conductive, and mixed subtypes. Compared with healthy newborns, infants admitted to the neonatal intensive care unit (NICU) are at substantially increased risk of congenital anomalies and exposure to HL-related risk factors. However, the specific determinants of neonatal HL remain controversial. Objective This systematic review and meta-analysis seeks to identify risk factors linked to HL in neonates admitted to the NICU. Methods PubMed, the Cochrane Library, Embase, and Web of Science were systematically searched from March 26, 1996, to February 25, 2025. Eligible studies were English-language retrospective studies employing multivariate logistic regression to evaluate potential risk factors for HL in NICU neonates. Meta-analyses were conducted using STATA, and pooled estimates were reported as odds ratios or relative risks (OR/RR) with 95% confidence intervals (CI). Results This study included 21 retrospective studies with a total of 21,143 participants. Meta-analysis indicated that exposure to ototoxic drugs (aminoglycosides, loop diuretics), history of the surgical ligation of patent ductus arteriosus (PDA), craniofacial anomalies, family history of HL, and TORCH infections were significantly associated with HL in NICU neonates (all P ≤ 0.05). In contrast, low Apgar score, prematurity, low birth weight, duration of vancomycin exposure, sex, and sepsis were not significantly correlated. The findings were robust, and no evident publication bias was detected. Conclusion Statistically significant risk factors for HL included craniofacial anomalies, family history of HL, TORCH infections, and surgical ligation of PDA, as well as hyperbilirubinemia, exposure to loop diuretics and other ototoxic drugs, and mechanical ventilation (all P ≤ 0.05). In contrast, low Apgar score, preterm birth, and sepsis were not significantly linked to HL. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/search , PROSPERO CRD420250653476.