Value of routine daily chest radiography in a non-university ICU: results of a controlled prospective study

Background

A strategy of daily-routine chest radiographs (CXR) is practiced in many ICUs, although its efficacy remains controversial. The aim of this study was to evaluate the diagnostic and therapeutic efficacy, as well as the costs of the daily-routine CXR, and compare this with CXR that were judged clinically necessary ('on-demand CXR').

Materials and methods

In this prospective, blinded, controlled study, daily-routine CXR were obtained from all patients in a mixed surgical–medical ICU in a university-affiliated teaching hospital. CXR were evaluated by trained radiologists (to score for the presence of predefined items such as progressive or new infiltrates, pneumothorax, malposition of the tube/lines); CXR were not accessible for intensivists. In addition to these 'daily-routine CXR', the intensivist ordered 'on-demand CXR' if deemed necessary. In all these cases, a specific form had to be completed with reasons for CXR and suspected abnormalities. Considerable worsening according to predefined criteria on the 'routine CXR', but not clinically recognized or suspected, was communicated daily with the intensivist. From these data, diagnostic efficacy (the number of CXR with significant abnormality/total number of CXR) and therapeutic efficacy (number of CXR leading to an intervention/number of CXR) were calculated. Chi-squared analysis was used to test differences.

Results

During 4 months, 1063 CXR in 153 patients were obtained (725 'daily-routine' CXR and 338 'on-demand' CXR). Diagnostic efficacy of 'daily-routine' CXR was 6.3%; diagnostic efficacy of 'on-demand CXR was 21.9% (P < 0.0001). Therapeutic efficacy of 'daily-routine CXR' was 2.6%; therapeutic efficacy of 'on-demand' CXR was 21.3% (P < 0.0001). The most frequent interventions on the basis of CXR findings were the administration of diuretics (20%) and repositioning of the tube (18%). A potential CXR volume reduction of 36% was observed when the 'routine CXR' strategy would have been replaced by an 'on-demand CXR' approach. This amounts to a potential cost reduction of €82,000 per year.

Conclusion

The value of the 'daily-routine CXR' is low. Based on these preliminary data, daily-routine CXR should probably be abandoned for ICU patients.