Neonatal total parenteral nutrition and the risk of childhood autoimmune diseases: A nationwide population-based study of 2.3 million children.
We evaluated the long-term risk of childhood autoimmune diseases following neonatal total parenteral nutrition (TPN) exposure using a massive nationwide cohort.
Using the South Korean National Health Insurance Service database (2005-2010), we analyzed 2,362,290 neonates. To neutralize "sick kid bias" and ensure temporal separation, a 1-year landmark design with entropy balancing was used to align 101 maternal and neonatal covariates (standardized mean difference <0.01). Five major autoimmune conditions-inflammatory bowel disease, juvenile arthritis, systemic lupus erythematosus, psoriasis, and type 1 diabetes mellitus-were evaluated using weighted Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).
Among 8601 TPN-exposed neonates, neonatal TPN exposure was not significantly associated with the long-term risk of childhood autoimmune diseases (Weighted HR 1.00; 95% CI, 0.87-1.15; P = 0.996). This lack of independent association remained consistent across all individual conditions, including inflammatory bowel disease (HR 1.17; P = 0.404) and type 1 diabetes mellitus (HR 1.01; P = 0.970). Furthermore, subgroup analyses confirmed the immunological safety of TPN across clinically vulnerable populations, such as infants requiring mechanical ventilation (HR 0.98; P = 0.917) or NICU admission (HR 0.90; P = 0.638), and across both infant sexes (male sex: HR 0.61, P = 0.085; female sex: HR 1.25, P = 0.342).
Neonatal TPN exposure was not independently associated with childhood autoimmune diseases. These findings strongly reassure clinicians that perceived long-term immunological risks reflect baseline clinical severity rather than the TPN intervention itself.
Using the South Korean National Health Insurance Service database (2005-2010), we analyzed 2,362,290 neonates. To neutralize "sick kid bias" and ensure temporal separation, a 1-year landmark design with entropy balancing was used to align 101 maternal and neonatal covariates (standardized mean difference <0.01). Five major autoimmune conditions-inflammatory bowel disease, juvenile arthritis, systemic lupus erythematosus, psoriasis, and type 1 diabetes mellitus-were evaluated using weighted Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).
Among 8601 TPN-exposed neonates, neonatal TPN exposure was not significantly associated with the long-term risk of childhood autoimmune diseases (Weighted HR 1.00; 95% CI, 0.87-1.15; P = 0.996). This lack of independent association remained consistent across all individual conditions, including inflammatory bowel disease (HR 1.17; P = 0.404) and type 1 diabetes mellitus (HR 1.01; P = 0.970). Furthermore, subgroup analyses confirmed the immunological safety of TPN across clinically vulnerable populations, such as infants requiring mechanical ventilation (HR 0.98; P = 0.917) or NICU admission (HR 0.90; P = 0.638), and across both infant sexes (male sex: HR 0.61, P = 0.085; female sex: HR 1.25, P = 0.342).
Neonatal TPN exposure was not independently associated with childhood autoimmune diseases. These findings strongly reassure clinicians that perceived long-term immunological risks reflect baseline clinical severity rather than the TPN intervention itself.