摘要
To the Editor.—Bronchopulmonary dysplasia (BPD) is a chronic lung disease (CLD) that occurs in preterm infants receiving respiratory support with mechanical ventilation or prolonged oxygen supplementation. However, the molecular basis of BPD is poorly understood.In their article regarding polymorphism of tumor necrosis factor-α (TNF-α) and risk and severity of BPD among very low birth weight infants, Kazzi et al1 determined whether alleles of TNF-α play a role in the susceptibility and/or severity of BPD among very low birth weight infants and concluded that the adenine allele of TNF-238 may reduce the risk and severity of BPD. However, in the Kazzi et al study, birth weight, gestational age, premature rupture of membranes, prenatal steroid, Apgar score, existence of respiratory distress syndrome, dose of surfactant, and ventilation index were all significantly different between BPD and non-BPD; the association of the adenine allele of TNF-238 with BPD needs additional examination with logistic regression analysis to clarify these confounding factors.Genetic predispositions to BPD have been identified in antioxidant defenses (eg, less efficient isoforms of glutathione-S-transferase-P142) and surfactant proteins (eg, SP-B intron 4 variant allele 5,3,4 SP-A 6A6 polymorphism5). There is increasing interest in different cytokine gene polymorphisms that might have a role in BPD. Adcock et al6 investigated if polymorphisms of cytokine genes influence the risk of developing CLD by genotyping 178 mechanically ventilated very low birth weight infants for the TNF-308 G/A, transforming growth factor-β(1)+915 G/C, and monocyte chemoattractant protein-1-2518 A/G polymorphisms and found that these polymorphisms did not play a significant role in determining risk for CLD. Unfortunately, this study wasn't based on case-control study of infants with and without CLD. We investigated whether gene polymorphisms for TNF-α-308 G/A and interleukin (IL)-1 influence the risk of developing CLD in 224 ventilated, small, preterm infants (<30 weeks' gestation) by case-control study and concluded that there was no significant association between the genotype or the allelic frequency of the TNF-α-308, IL-1 receptor antagonist, or IL-1β exon 5 polymorphisms with CLD and the duration of intermittent mandatory ventilation supplement.7We believe that any association study obtained even by case-control study should be regarded as provisional and that replication in independent population studies is critical. Researchers need to caution against the acceptance of positive association of polymorphisms. The appropriate safeguards must be implemented to minimize spurious positive association from population stratification, multiple-hypothesis testing, and bias for preferential publication of positive results.It is likely that genetic variations involved in the pathogenesis of BPD will be recognized in the future. However, genetic and environment factors interfere with each other, and the association study of polymorphism is better examined with case-control studies to avoid the confounding factors. Because BPD is multifactorial in etiology, researchers in the association of different polymorphisms with BPD should be very careful with the inclusion criteria and data interpretation.