1990, The British Journal of Radiology, 63, 22-25

Asymmetrical pulmonary changes in premature infants with surgical closure of a persistent ductus arteriosus By J. D. J. Steinberg, MD, *S. Bambang Oetomo, MD, PhD and A. Martijn, MD, PhD Departments of Radiology and 'Paediatrics, University Hospital Groningen, Groningen, The Netherlands (Received March 1989 and in revised form August 1989)

Abstract. The chest radiographs of 57 premature infants admitted consecutively with hyaline membrane disease and receiving respirator therapy were reviewed, comparing right and left pulmonary abnormalities. Fifteen infants (Group I) did not develop a persistent ductus arteriosus (PDA). Forty-two infants developed a PDA, which was successfully treated with indomethacin in 17 infants (Group II), while in 25 infants (Group III) the ductus was closed surgically. Analysis of the three groups showed that infants who had undergone surgical closure of their PDA developed significantly more asymmetrical broncho-pulmonary damage. Statistically significant fewer radiological findings of broncho-pulmonary damage were found on the left side in comparison with the right side in the group treated surgically.

Premature infants with hyaline membrane disease (HMD) develop lung changes during respirator therapy. Northway et al (1967) described bilateral, rounded, Jucent areas alternating with areas of irregular density and chose the term broncho-pulmonary dysplasia (BPD) for this chronic lung syndrome. Swischuk (1977) distinguished between three different types of bubbles found in HMD: in the acute phase of HMD, HMD with pulmonary interstitial emphysema and in the chronic phase of HMD (BPD). We use the term bronchopulmonary damage pattern introduced by Caffey (1985) because of the overlap of the radiological patterns. Further reports by Sickles and Gooding (1976), Ablow and Orzalesi (1971), Moylan and Shannon (1979), Tchou et al (1972) and Watts et al (1977) described asymmetrical or localized lung involvement. It has twice been reported that complications such as atelectasis and pneumothoraces can influence the distribution of BPD (Sickles & Gooding, 1976; Watts et al, 1977). The purpose of this study is to investigate the radiological pattern of broncho-pulmonary damage pattern in premature infants with and without surgical closure of their PDA. Patients and methods

The records of 58 premature infants admitted consecutively to the Neonatal Intensive Care Unit of the University Hospital of Groningen with the diagnosis HMD and needing respirator therapy were reviewed. Of the 58 infants, 15 did not develop a PDA (Group I), while the remaining 43 infants showed this complication. The PDA was diagnosed clinically. In all cases of PDA, closure was attempted primarily with indomethacin, and was successful in 17 cases (Group II). One infant had to be excluded from this group because the ductus was patent post-mortem. Address reprint requests to: Dr A. Martijn, Department of Radiology, University Hospital Groningen, Oostersingel 59, 9711 XE Groningen, The Netherlands.

22

The remaining 25 infants (Group III) underwent surgical closure of their PDA. Gestational age, birth weight, total length of ventilation, peak pressure and peak oxygen concentrations were recorded. Because of the wide range of total length of ventilation, the data were compared after logarithmic transformation. The two groups that developed a PDA were compared with regard to the age at which the ductus was closed. The groups were compared with respect to the clinical data using a one-way analysis of variance. The chest radiographs were then reviewed for asymmetrical distribution of pulmonary abnormalities, and analysed using Fisher's exact test. Results Clinical data

Table I gives the mean values of the clinical data of the three groups. The mean values were 28.8 weeks for gestational age, 1191.7 g for birth weight and 21 days for the total length of ventilation, with peak pressures of 28.8 cm of water and peak oxygen concentrations of 86.6%. There were no statistically significant differences between the three groups with regard to birth weight, ventilation pressure, or oxygen concentrations. Infants in Groups II and III were younger than those in Group I (28.8 and 28.1 weeks against 30 weeks, respectively), and Group III was ventilated for a longer period of time (28.6 days against 15.8 and 15 days in Groups I and II, respectively). There was no significant difference between Groups II and III with regard to the age at which the ductus was closed, even though, in all cases, closure with indomethacin was attempted primarily. Evaluation of the radiographs was later in the two groups that had developed a PDA (Group II, 45.5 days; Group III, 44.5 days) than in Group I (30.4 days). At the time of evaluation, eight (53%) of the 15 infants in Group I were still alive as were 15 (88%) of the 17 infants in Group II and 17 (68%) of the 25 infants of Group III. The British Journal of Radiology, January 1990

Asymmetrical pulmonary changes in premature infants with PDA Table I. Composition of study groups

Number of infants Gestational age (weeks) Birth weight (g) Total length of mechanical ventilation (days)" Maximum ventilation pressure (cm water) Peak oxygen concentration (%) Age at closure of PDA (days) Day of scoring of radiographs Survivors (%)

Group I no PDA

Group II PDA closed with indomethacin

Group III PDA closed surgically

15 30 + 2.1 1254.7 + 356.7 15.8 + 12.5 29 + 8.1 92+14.2 — 30.4+1.8 8/15 (53)

17 28.8+1.6 1220 + 255.4 15+11.5 27.5 + 4.4 86.8+18.4 20.8+14.6 45.5± 13.7 15/17 (88)

25 28.1+2.2 1134.6 + 285.2 28 + 21 28.5 + 5.5 83.2+17.3 19.7+13.4 44.5 + 17.6 17/25 (68)

p value

0.01 0.05 >0.05 >0.05