Lung function in young adults born small for gestational age at term

Abstract

Moderate to extreme prematurity is associated with lower lung function in adults1 while evidence is poorer and controversial for late prematurity.2 Likewise, the potential long-term impact on adult lung function of being born small for gestational age (SGA) at term is not well established since most previous studies in this field have been done in groups with participants enrolled by birthweight and not by SGA per se.2 This may be important because not all infants born SGA have experienced intrauterine growth restriction (IUGR) and, the other way round, early IUGR does not necessarily bring fetal growth down below the 10th percentile (the definition of SGA). We recently showed that young adults born SGA at term had markedly reduced exercise capacity, mostly of cardiovascular origin.3 In particular, they showed lower maximal workload, peak oxygen consumption and oxygen pulse, as well as higher minute ventilation/carbon dioxide production equivalent at the anaerobic threshold, than age-matched controls.3 Here, we extend and complement these previously published observations3 with the analysis of pulmonary physiology (spirometry and carbon monoxide diffusing capacity [DLCO]) and the measurement of circulatory markers of abnormal lung development, including surfactant protein A and D (SP-A and SP-D) and club cell protein 16 (CC16). We conducted an ambispective, controlled, cohort study whose detailed methodology has been published elsewhere.3 Briefly, from the birth records of our institution, we identified individuals born at term (≥37 weeks of gestation) between 1975 and 1995 with either appropriate weight for gestational age (AGA) or SGA (<10th centile for gestational age). Distinction between constitutionally small and growth-restricted participants was not possible since Doppler ultrasound, which is the tool allowing to differentiate between these two groups, was not widely used when these individuals were born, so their medical records do not include this data. Exclusion criteria were twins, congenital malformations, genetic syndromes, macrosomia, mental disorder, current pregnancy or professional sports practice. Demographics and medical history were recorded. Forced spirometry and DLCO were determined following international standards. Reference values were those of Roca et al.4, 5 The serum concentrations of SP-A, SP-D and CC16 were determined using ELISA or Luminex following manufacturer's instructions. Results are presented as mean ± SD, median [interquartile range—IQR] or number (%). Comparisons of perinatal and demographic data were performed using the Student's t-test, Wilcoxon rank-sum test, Chi-square or Fisher's exact tests, as appropriate. Comparisons of pulmonary results were adjusted by age, sex, body surface area and smoking exposure using multivariate linear regression models. A p-value < 0.05 was considered statistically significant. Analyses were performed using Stata/IC 15.1. We compared individuals born at term with SGA (n = 61) or AGA (n = 66) in their early thirties (Table 1). All participants were of Caucasian origin and were born in Barcelona (Spain). By design, birthweight was lower in SGA. The number of males and females was similar in both groups. In adulthood, weight was similar but height was significantly lower in SGA. The proportion of smokers was also similar in both groups (Table 1). There were three AGA-born and six SGA-born individuals with asthma at the time of study (one and three requiring treatment, respectively) but this difference was not statistically significant. Although absolute values of forced expiratory volume in the first second (FEV1) were significantly lower in the SGA group, when expressed as % of reference values, spirometric and DLCO values were similar and normal in both groups. Differences in absolute DLCO values did not reach statistical significance but sample size was smaller. The circulating levels of SP-A, SP-D and CC-16 were also similar in both groups. The results of the multivariate analysis did not show any further differences in the analysed parameters.