Results

Fifty four infants with chronic lung disease and 44 healthy preterm infants were studied. Table 1 shows the demographic and clinical characteristics of the two groups. Infants with chronic lung disease were significantly lighter and less mature at birth, and had a lower 5 minute Apgar score. A significantly higher proportion of the healthy infants studied were born after a multiple pregnancy.

A total of 248 Doppler examinations were performed: 150 studies in infants with chronic lung disease and 98 studies in healthy infants. Table 2 summarises the timing and results of the examinations performed in each group. The median number of examinations was 2 (range, 2–6) in infants with chronic lung disease and 2 (range, 2–4) in healthy preterm infants. There was no significant difference between the two groups in the overall median corrected age at which the examinations were performed: 12 (range, 38 weeks post menstrual age to 44 weeks) in infants with chronic lung disease versus 13 (range, 38 weeks post menstrual age to 39 weeks) in healthy infants (p = 0.77).

The overall median AT/RVET was significantly lower in infants with chronic lung disease (table 2). Figure 1 shows changes in AT/RVET with time. Median AT/RVET in infants with chronic lung disease tended to be lower than that found in healthy infants throughout the period of study. This difference was significant at all time periods except at 36 weeks post menstrual age and between 40 and 52 weeks of age corrected for prematurity.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Changes in the ratio between acceleration time and the right ventricular ejection time (AT: RVET) with corrected age. Data points are medians (interquartile ranges); p < 0.01.

Within individual healthy preterm infants, changes in AT/RVET were significantly correlated with changes in time (r = 0.67; p < 0.001). There was a similar degree of correlation between the same variables in infants with chronic lung disease (r = 0.55; p < 0.001). There was also no significant difference in median AT/RVET velocity between the two groups (table 2).

In the first multiple regression model, AT/RVET was significantly and independently associated with the presence of chronic lung disease, gestational age, and corrected age at examination (table 3). In the second multiple regression model, AT/RVET was significantly independently associated with age at examination and inversely with duration of supplemental oxygen treatment in infants with chronic lung disease. The association with gestational age was of borderline significance (table 3).

Discussion

Pulmonary hypertension is common in preterm infants with respiratory distress syndrome and PAP remains raised in those who subsequently develop chronic lung disease.9 ,10 We evaluated changes in PAP, as assessed by the pulmonary artery Doppler derived AT/RVET ratio, in infants with established chronic lung disease and healthy preterm infants after discharge from hospital during the 1st year of life. Pulmonary arterial pressure is inversely related to age in both healthy infants and those with chronic lung disease. However, although PAP falls at a similar rate during infancy in both groups, infants with established chronic lung disease have persistently raised PAP compared with healthy preterm infants.

In healthy term infants, intra-acinar arterial multiplication parallels alveolar proliferation, which occurs most rapidly in the first 2 years of life. Intra-acinar vessels also increase in size with growth during infancy.18 This increase in the total cross sectional area of the pulmonary vascular bed results in an overall reduction in pulmonary vascular resistance. In childhood, this allows PAP to remain unchanged despite an increase in cardiac output.19

Lung development in non-ventilated preterm infants is appropriate for post menstrual age, suggesting that preterm birth is not itself associated with poor postnatal lung growth.20 In contrast, impaired alveolar and arterial development has been demonstrated in studies of preterm infants dying after mechanical ventilation in the early neonatal period.4 ,20 In our study, AT/RVET was significantly directly related to gestational age, indicating that PAP is higher in more immature infants in the 1st year of life, independently of the presence of chronic lung disease.

Histological studies of the pulmonary circulation in infants and older children dying after the development of chronic lung disease have shown pronounced structural abnormalities, including increased medial thickness of small pulmonary arteries and distal extension of arterial smooth muscle.2 ,3 ,5 There is also a reduction in the number of peripheral arteries associated with a normal alveolar to arterial ratio, indicating failure of alveolar development.4 Disruption in growth and development of the pulmonary vascular bed might result in raised pulmonary vascular resistance and PAP in ventilated preterm infants who subsequently develop chronic lung disease.

Several cross sectional echocardiographic studies have described raised PAP in infants with chronic lung disease.21-23 Pulmonary hypertension with raised pulmonary vascular resistance has also been confirmed by direct measurement at cardiac catheterisation.6 ,12 ,13 However, the natural history of abnormalities of pulmonary haemodynamics in infants with chronic lung disease is not known. Reports in a few infants who have undergone repeat catheterisation have demonstrated a reduction in PAP with time, although PAP might remain persistently raised into later childhood.6 ,24 One infant, not receiving supplemental oxygen despite raised PAP, showed progressive pulmonary hypertension in the 20 month period between catheter investigations.11 To our knowledge, there are no published studies of changes in PAP in healthy preterm infants or those with chronic lung disease during infancy after discharge from hospital.

Pulmonary arterial pressure and pulmonary vascular resistance in infants with chronic lung disease have been assessed non-invasively using M mode and Doppler echocardiography.21-23 25-27 The two methods most commonly used have been: (1) measurement of pulmonary flow indices, or (2) measurement of the maximal velocity of tricuspid regurgitation and application of the modified Bernoulli equation.28Measurement of maximal tricuspid regurgitation velocity is generally accepted as the method of choice when assessing PAP non-invasively.29 However, the only study that has used this technique in infants with chronic lung disease found that the incidence of measurable tricuspid regurgitation was only 44%, despite sedation.22 The same study also reported a close correlation between measurements of AT/RVET and tricuspid regurgitation derived PAP. Pulmonary flow indices were chosen in our study because measurements could be performed successfully in almost all infants without the need for sedation.

Unlike previous studies, our study has compared PAP in infants with chronic lung disease with healthy preterm infants, rather than using a reference population of preterm infants who had received respiratory support but not developed chronic lung disease. This inevitably led to demographic differences between chronic lung disease infants and our reference group. Although it might have been preferable to have arranged Doppler assessments at predefined times, this was impractical in view of the geographical distribution of the study populations within the Mersey region. Instead, all assessments in our study were made on an opportunistic basis, at the same time as an infant attended follow up clinic. Infants with chronic lung disease who continued to attend clinic throughout the study period are likely to have been the sickest infants. This might have influenced our results through over representation of infants with low AT/RVET values towards the end of our study.

Reactivity of the pulmonary vascular bed to oxygen in infants with chronic lung disease has been confirmed in direct and indirect studies.6 ,13 ,22 However, the precise relation between different values of SpO₂ and PAP has not been fully evaluated. This might reflect the difficulty in maintaining a steady SpO₂ value during an echocardiographic examination. Because of the inherent inaccuracy in defining true SpO₂ on a background of continuous variation with time and movement artefact, all infants in our study were maintained at SpO₂ values above 93%, rather than attempting to define a precise SpO₂value. It is possible that in some infants, maintaining an even higher SpO₂ value might have led to a reduction in pulmonary vascular resistance and PAP.22

We have shown previously that persistently raised PAP at 24 hours of age is an important early predictor for the development of chronic lung disease in ventilated preterm infants.30 Our study has demonstrated an indirectindependent relation between duration of supplemental oxygen treatment and AT/RVET. This suggests that PAP remains higher in infants who receive a longer period of oxygen treatment, and therefore presumably have more severe disease. Because none of our infants died within the period of our study, we were unable to examine the relation between PAP and mortality in our study group.

We have shown that although PAP is higher in infants with chronic lung disease than in healthy preterm infants, it tends to fall with increasing age during the 1st year of life at a similar rate in both groups. Pulmonary hypertension is also independently associated with a longer duration of supplemental oxygen treatment. Future research will need to consider the importance of pulmonary hypertension and the effects of pulmonary vasodilator treatment in influencing respiratory outcome in preterm infants with chronic lung disease.