Study objectives: To determine whether there is any difference in diagnostic, therapeutic, and outcome efficacy between protocols utilizing daily (ie, routine) chest radiographs (CXRs) and those utilizing clinically indicated (ie, nonroutine) CXRs in mechanically ventilated patients.
Design: Prospective, randomized, observational study.
Setting: A 20-bed medical ICU at a university hospital.
Patients: Adult patients who had been receiving mechanical ventilation for > 48 h.
Measurements and results: Ninety-four patients who had been intubated for at least 48 h were randomized to receive either routine or nonroutine CXRs and were observed until extubation or death. The percentage of CXRs with new findings was significantly larger in the nonroutine CXR group (53.1%; 120 CXRs) compared to that in the routine CXR group (33.4%; 98 CXRs; odds ratio [OR], 1.59; 95% confidence interval [CI], 1.16 to 2.18; p = 0.004). The number of CXRs with new findings that resulted in interventions was significantly larger in the nonroutine CXR group (26.5%; 60 CXRs) compared to that in the routine CXR group (13.3%; 39 CXRs; OR, 2.0; 95% CI, 1.29 to 3.08; p = 0.002). There was no difference in the mean duration of mechanical ventilation (routine CXR arm, 7.93 days; nonroutine CXR arm, 6.76 days; p = 0.2606), length of ICU stay (routine CXR arm, 11.93 days; nonroutine CXR arm, 9.55 days; p = 0.1936), and total length of hospital stay (routine CXR arm, 19.34 days; nonroutine CXR arm, 16.45 days; p = 0.2199) between the two groups. Although patients in the nonroutine arm of the study received fewer CXRs, adverse outcomes (eg, length of mechanical ventilation, length of hospital stay, length of ICU stay, and mortality) did not increase (p = 0.818). Using the number of CXRs per patient as a surrogate, there was a statistically significant cost savings and decreased radiation exposure in the nonroutine CXR group.
Conclusions: For mechanically ventilated patients, a strategy calling for daily routine CXRs compared to CXRs obtained based on clinical indications alone was not associated with reduced ICU or hospital length of stay or with reduced mortality.