The present study showed that there is greater chance of developing moderate to severe BPD in extremely premature newborns after 36 days under mechanical ventilation16. It is important though to make it clear that BPD is a multifactorial disease, and its development is also associated to factors such as intrauterine infections, growth restriction and nicotine exposure16. Due to the pathophysiology of the disease, BPD remains the most frequent morbidity in extremely premature infant17, and the diagnosis of moderate to severe BPD is associated with worse prognosis. A previous study comparing infants with different disease severity classifications showed that the group with moderate to severe BPD had a higher proportion of subjects with grade 3 and 4 intraventricular hemorrhage, periventricular leukomalacia, NEC, late sepsis, home oxygen use and death after discharge, in addition to the increased incidence of neurological impairment, worse mental and psychomotor development, blindness and hearing impairment18. In the present study, infants with moderate to severe BPD had worse ROP, higher prevalence of PH, and longer length of hospital stay, corroborating the findings in the literature that preterm infants have worse outcomes.
Studies have shown a high incidence of moderate to severe BPD in extremely premature infants (PMA up to 27-29w). Stoll et al. reported an incidence of 41% in their population19. Another study with extremely low birth weight infants (401-1000 grams) found that 52% of the population had diagnosis of moderate to severe BPD18. In the present study, more than half of the preterm infants included (53%) had moderate to severe BPD (Figure 1).
Recent studies have enhanced the harms of exposure for extremely preterm infants to MV for long periods of time. Yossef et al. observed that a group of extremely premature infants that were submitted to more than 56 days of MV had a higher incidence of moderate to severe BPD20. Choi et al. found an almost three-fold increase in the risk of mortality of extremely low birth weight preterm infants who required MV for 15 to 28 days. Furthermore, that study also associated cumulative duration of MV with ROP requiring surgical correction, neurological impairment, BPD, PH, and length of stay6. The major cause of the development of ROP is the exposure of preterm infants to supplemental oxygen therapy at high concentrations, therefore its relationship with prolonged MV is already expected11. Although the number of days under MV found in our study is different from the aforementioned studies, our findings are mostly similar to the literature. It reinforces the importance to avoid invasive MV, as well as to limit the duration that the extremely preterm infants are exposed to this treatment.
The mild BPD group had more individuals who used CPAP before MV, and it is an expected result. In centers where CPAP is used during stabilization in the delivery room, the incidence of severe BPD is 3.3%21, and intubation for surfactant administration alone (without MV) has shown a significant reduction in the use of MV in extremely low birth weight preterm infants22. In addition, compared with intubation right after birth, CPAP reduces the incidence of BPD and death at 36 weeks of PMA23.
Laughon et al. in their large study (n=3629 preterm infants) to develop an instrument for predicting BPD risk and death through clinical information, six variables were relevant to the model: PMA at birth, birth weight, ethnicity, gender, respiratory support and inspiratory oxygen fraction on specific days of hospitalization10. Factors such as PMA and weight were also related to longer time under MV20. In our search for predictors for development of moderate to severe BPD, in addition to time on MV greater than 36 days, the presence of PH appeared as significant in the regression model. It is associated with prolonged exposure to MV6, and the association between PH and BPD is related to both its pathogenesis (similarly associated to complications of prematurity) and secondary PH development in BPD24. There was no difference between groups regarding PDA, NEC, use of VAD, chorioamnionitis and maternal use of tobacco, unlike other studies about both prolonged time under MV and higher severity of BPD6,20, which might have occurred due to the relatively small sample size. Additionally, a reasonable number of BPD patients did not survive until they completed 36 weeks PMA, which may also have contributed to this difference in outcomes.
Controversies regarding the definition and classification of BPD in literature may have limited our results. One premature newborn died before completing 36 weeks PMA and 41 died before 28 days of age. According to the most recent workshop on BPD, those infants could be classified as BPD type IIIa (early death from lung disease and respiratory failure)25, and this could have influenced our findings. Our study involved preterm infants from a single institution and included a small sample compared to similar ones. Additionally, long-term outcomes such as neurodevelopment were not analyzed. However, almost all variables that may influence the development and severity of BPD were analyzed, except genetic factors. Despite the weaknesses, most of the findings of the present study, which is a pilot study, are in line with the findings from the literature.
Despite the limitations, it is noteworthy that this study is the first to investigate a cutoff point of time under MV to identify the chance of developing more severe forms of BPD. In addition, the result found was validated in an independent sample. Our study was able to define 36 days as a time of MV exposure that is associated with the development of moderate to severe BPD and, as a consequence, may have worse outcomes, such as greater severity of ROP and length of hospital stay. Therefore, it is suggested that MV should be avoided whenever possible, using strategies such as prophylactic CPAP and less invasive surfactant admnistration26,27. Moreover, MV should be interrupted as soon as possible, preferably before 36 days to reduce the risk of developing moderate to severe BPD, as well as to avoid impairments associated with the severity of the disease. Furthermore, a multicenter study should be performed to verify whether similar results are observed with a larger sample of patients. Indeed, this characterizes the next step for a future study in which we will include other hospitals from the Brazilian neonatal research network.