J. Maurer et al., Morphological changes in chest radiographs of patients with acute respiratory distress syndrome (ARDS), INTEN CAR M, 24(11), 1998, pp. 1152-1156
Objective: To determine whether the quality of infiltrations in chest radio
graphs can accurately predict the histological extent of fibrotic change in
patients with acute respiratory distress syndrome (ARDS).
Design: Retrospective clinical investigation.
Setting: Intensive care unit (ICU) of a university teaching hospital.
Patients and methods: Of 47 patients treated with extracorporeal membrane o
xygenation (ECMO) for severe ARDS over a 5-year period, 23 patients underwe
nt open lung biopsy at thoracotomy for treatment, mostly of pneumothorax. C
hest films obtained by portable chest roentgenography preceding the operati
on were reviewed retrospectively and compared to the histomorphological res
ults of the lung specimen.
Results: Chest radiographs displayed mixed alveolar-reticular opacification
in 60.2 %, alveolar patterns in 22.9 % and reticular opacities in 10.5 %.
In 0.4 % there were no infiltrates, 6% could not be evaluated because of in
sufficient quality. There was no relevant difference between the right and
left lungs,; Subdividing patients into two groups according to the histolog
ical results of either absent or mild (1) or severe (2) lung fibrosis, we f
ound an alveolar haziness in 12.3 % in group I compared with 28.2 % in grou
p 2, while reticular characteristics were identified in 13 % and 11 %,respe
ctively.
Conclusions: The most common opacity in chest radiographs of patients with
severe ARDS treated with ECMO is mixed alveolar-reticular opacification. Se
vere lung fibrosis is not positively correlated with a reticular radiograph
ic pattern. ECMO does not lead to specific radiological changes in conventi
onal radiographs, contrary to clinical findings that treatment with ECMO mi
ght induce pleural or pulmonic haemorrhage, especially in the earlier days
when systemic heparinization had to be used instead of the heparin-coated t
ube-surfacing.