Spirometric values reflecting airflow are consistently lower in survivors of BPD at any age compared to controls born at term; those with BPD have substantial airway obstruction and alveolar hyperinflation [2, 55–58]. In most studies [2, 57, 59, 60], the mean forced expiratory volume in 1 s (FEV₁) values in those with BPD range from normal to values indicative of significant airflow limitation, reflecting the heterogeneity in the functional expression of the disease. These data should be interpreted with caution, however, since they are not generally applicable to the whole population of survivors, and especially not to those with new BPD or mild neonatal pulmonary disease. Patients who were born prematurely but did not have BPD usually fare better [61], but they too may have airflow limitation at school age [2, 58, 59, 62] and into adolescence and adulthood [3, 63, 64].

Air trapping has been reported in premature children with BPD [65–67] secondary to obstructive airway disease, impaired alveolarization, and/or abnormal lung growth or injury. In contrast, normal FRC has also been reported in former preterm adolescents with BPD [63]. The presence of residual lung function abnormalities, mainly airflow obstruction and progressive static hyperinflation, raise the question as to whether chronic lung disease of infancy may ultimately affect pulmonary aging, leading to the development of chronic obstructive pulmonary disease (COPD) [68].

DLCO in school-age children born prematurely has been noted to be impaired in several studies [64, 65]. The decrease of DLCO/Va may reflect a deficit in the alveolar number leading to increased airspace and a reduction in the gas exchange-surface area [69]. Due to the challenges to performing DLCO during infancy, there are no longitudinal studies tracking DLCO from infancy into childhood and adulthood.

The degree of airflow limitation in the first years of life also predicts later pulmonary function. In a small group of infants with severe BPD who were followed from birth, VmaxFRC at 2 years of age was closely related to FEV₁ at 8 years, suggesting tracking of lung function with time and negligible “catch-up” growth of the lung [27]. This finding highlights an irreversible early airway-remodeling process.

There are limited longitudinal outcome data in preterm children born in the postsurfactant era. Most studies demonstrate that lung function deficits persist into adulthood and do not worsen over time. This finding has been reported even in those with severe BPD requiring home ventilation [70]. Two small studies reported some improvement in airway obstruction [71] or lung hyperinflation [72] in adolescents with BPD. According to a meta-analysis and follow-up study, the adverse effects of prematurity on pulmonary function were still detectable in school-aged children prematurely born in comparison to sex-matched controls born at term [73]. On the other hand, Doyle et al. [63] reported that survivors of BPD may have a substantial decline in pulmonary function over time, on the basis of data from a large cohort with a birth weight of < 1500 g who were followed from 8 to 18 years of age [63]. Overall, most studies suggest that airway growth in preterm infants is maintained with somatic growth, but does not “catch up.”

The chest radiograph (CXR) is the simplest and most widely available method for imaging preterm children. The original case description of BPD included CXR abnormalities such as linear fibrotic opacities and hyperexpanded regions of lung parenchyma [74], thereby leading to the bronchopulmonary dysplasia terminology describing radiological changes [74]. Several scoring systems were subsequently developed [75]. These severe changes on CXR are rarely seen in today’s “new” BPD, and in general, CXRs in contemporary preterm cohorts demonstrate minimal abnormalities. The modern clinical diagnosis of BPD does not incorporate a radiological component and instead focuses on the clinically assessed need for supplemental oxygen. Although chest radiographs have the advantages of accessibility and simplicity, the modality is considered only marginally useful for diagnosis. While chest films may only provide limited information to guide care, this modality is performed routinely to assess progression of disease.

Chest computed tomography (CT) provides more detailed imaging of the lung parenchyma and airways compared to CXR. CT images can be obtained during quiet breathing or at full inspiration. Expiratory images are useful when identifying air trapping. This includes the interpretation of mixed attenuation, where either the area of high attenuation can represent a parenchymal abnormality or a low attenuation can be due to air trapping. There are often indications of air trapping on inspiratory images, but expiratory images may identify air trapping not seen on inspiratory images in the same patient. Young or uncooperative children may need to be sedated due to the difficulty with interpreting images due to motion artifact, which can further be decreased using a controlled ventilation technique [76, 77].

Chest CT is considered the most sensitive imaging modality to detect structural abnormalities in patients with BPD [78]. Chest CT may provide insight into BPD pathophysiology; the early neonatal course may be predictive of later impairment noted on imaging [79]. Several CT protocols and scoring methods have been described over the last 30 years to characterize and quantify the structural abnormalities of preterm patients with BPD. Most of them assessed the clinical severity of BPD (mild, moderate, severe) and reported an increase in CT abnormalities within those with more severe BPD [79–85]. CT findings of BPD patients have been compared with control patients, which were either healthy term-born [81] or preterm without BPD [79, 86, 87]. These studies report higher or worse CT scores in those with BPD. In CTs of survivors of old BPD, persistent radiological abnormalities have been reported in a majority of patients [88]. Respiratory mechanical measurements and functional residual capacity during infancy have been reported to be associated with structural disease on chest CT; [85, 89] however, diffusing capacity and forced expiratory flows were reported to not correlate with the structural disease [81]. Furthermore, chest CT has been reported to be more sensitive to identifying disease during infancy in those with chronic lung disease of infancy compared to diffusing capacity and forced flows [81]. Wong et al. [90] described a definitive appearance of emphysema in a group of young adults with a history of moderate to severe old BPD, and the extent of emphysema on CT was inversely related to their FEV₁ z-scores. Aukland et al. [79] studied two cohorts of BPD survivors, one from 1982–1985 and the other from 1991–1992, using inspiratory and expiratory high-resolution CT images. Participants were evaluated at a mean age of 10 and 18 years. Abnormalities were reported on chest CT scans in 86 % of the participants; the majority of the findings were linear/triangular opacities. Although the CT scores were higher or worse in the older cohort, the difference between the two groups was not statistically significant. A higher or worse HRCT score was associated with worse lung function as assessed through spirometric variables as well as the ratio of residual volume (RV) and total lung capacity (TLC). It is important to note that, to date, there are no validated and universally accepted CT scoring systems for quantifying structural changes [91]. Given the wide variations in the visual appearance of lung parenchyma observed in this population, the potential for imaging-based phenotyping is a possibility. These quantitative imaging biomarkers could be further refined to phenotype BPD and inform clinical care (Fig. 2) [92].