Preterm infants are a vulnerable population at high risk for mortality, morbidity, and neurodevelopmental impairments that carry lifelong consequences. [2, 3, 13, 30, 31] Some authors have shown an improvement in survival rates and sequelae-free survival in recent decades, however these children are still at high risk of NDI.[3, 32] The assessment of neurodevelopment is important because of the impact it can have on the child's quality of life, especially in VPT. This knowledge by gestational age, as shown in our work, is relevant because we need to give parents accurate information about the risk to life and NDI in this group of NBs. In this study, we evaluate the rates of survival and NDI in a tertiary center in Portugal.
In our study, mortality < 32 weeks GA was 10.9%. This is slightly higher than that reported by other studies, and might be explained by the fact that we included deaths that occurred in the delivery room, which is not the case in other studies (eNewborn). A previous Portuguese study analyzing the same age group reported a mortality rate of 15%.[33] We verified a decline in mortality with increasing GA until 29w, and afterwards it remained below 5%, which is similar with what is reported in the eNewborn European Network database study, where Portuguese data is also included. [30]
Our follow-up rate of 88.1% is higher than what is reported in other studies and close to the value referred to as ideal in this age group by the guidelines of the American Academy of Pediatrics (90%).[3, 7, 34]
In our study, the survival rate without sequelae exceeded the number of deaths at 25 weeks of GA. Similar data was reported in a 2006 study (EPICure).[32] Our survival rate without moderate to severe NDI increased with GA, being over 85% from the 28w onwards. The survival rate without NDI by GA was higher in our study than what is reported in EPIPAGE: GA ≤ 26w: 66% in our study vs 48.5% on EPIPAGE and at 27-31w: 93% vs. 90%. [3]
Regarding NDI, in our study 6.2% of VPT had moderate to severe NDI at 24 months of PMA, which is lower than the 10% reported in Resende et al, a Portuguese study at the same hospital. Comparing the GMDS-II global score < 70 and CP, in the last decade we have obtained better results (GMDS-II global score < 70: 4.7% vs. 6.8% and CP: 3.3% vs. 6%).[28, 35]
In our study, the incidence of CP was lower (4.8% in GA ≤ 27w and 3% at 28–31w) than what is reported in surveillance of cerebral palsy in Europe: 14% at ≤ 27w of GA, 6% at 28–31w and < 1% at 32–36w. [6, 28]
Comparison with other studies in relation to the presence of NDI must be made with caution, since the methodology, the scale applied and the population studied differ in the various published studies. Picciolini et al reported that 9.2% of VPT had NDI; Vermont Oxford Network reported 34.2% (included only ELBW) and EPICure reported 11.8% (included only < 27w). [27, 32, 35, 36]
Currently, there are few studies in Portugal evaluating neurodevelopment in VPT. Furthermore, none of them stratify by GA and neither do they analyze the correlation with possible risk factors and serious mortality complications.
In our study, abnormal neurodevelopmental outcome was associated with inferior GA, being male and resuscitation with tracheal intubation. Lower GA is well described to be associated with neurodevelopment impairments and mortality. [3, 37–39] In our cohort, this result was clear, showing survival rates without serious sequelae of over 88% from 29 weeks of GA onwards. Because lower GA is related to more risk of morbidities, the literature is more centered in EPT, so there are limited studies about neurodevelopment outcomes in VPT. [1, 5, 7, 40, 41] Concerning sex, there are theories about differences in brain organization and genetic or hormonal factors predisposing males to be more vulnerable to injury. [2, 14, 42, 43] However, a systematic review, that included studies conducted later in childhood, revealed that the influence of sex on general cognition was largely diminished at five years old.[43] Resuscitation with tracheal intubation at birth is related with lower Apgar score, and in accordance to what is described in the literature, the need of advance reanimation at birth, in our study, was associated with moderate to severe NDI or death.[39]
The evaluation of maternal education aimed to infer about the socioeconomic status. As it is described in the literature, lower status is a risk factor for NDI.[3, 7, 35, 43] In our study, the bivariate analysis showed that a higher level of maternal education was a protective factor for NDI, however this was not demonstrated in the multivariate analysis. In a systematic review, unlike factors related to infant characteristics, the influence of parental education appeared to persist into middle childhood.[43]
In a previous study in the same NICU, it was verified that fetal growth restriction (FGR) was associated with a significantly increased risk of poor neurodevelopmental outcome at 24 months of PMA versus AGA infants.[44] However, in this study we only evaluated the weight percentile for GA and no differences were found. In the literature, there is often no clear definition of FGR and sometimes SGA is used, which can confuse the results and make comparison difficult.
When comparing singletons to multiples there were no difference in mortality and NDI, which is similar to the results reported in an Italian study and in a recent systematic review. [19, 45] In contrast to Taborda et al, who reported that monochorionic twins had an increased risk of severe neurodevelopmental delay.[20] The current results can be related to a better prenatal care with closer monitoring of twin pregnancies.
To our knowledge, this is the only Portuguese study to show neurodevelopment at 2 years of age in this population, with stratification by GA. A low rate of children who lost to follow-up, and the fact that it did not have significant differences comparing to our cohort, is one of the major strengths of this work. Nevertheless, since it’s a retrospective and single-center study, it has some limitations, as the findings may not be representative of other NICUs. Memory and registration biases can be minimized in this work since the data on this population is recorded by some of the authors at national level and international level (National Registry of Very Preterm Newborns and the eNewborn European Network database) and is systematically recorded from birth to discharge. [30] Another limitation is that children´s neurodevelopment is only monitored and assessed up to 24 months of PMA, as some neurodevelopment disorders may appear later. In this sense, a school evaluation would be important.