The 67gallium-transferrin pulmonary leak index in patients at risk for the acute respiratory distress syndrome.

Abstract

To determine the occurrence and clinical importance of increased pulmonary microvascular permeability in patients at risk for acute respiratory distress syndrome (ARDS). Prospective, cohort study. Medical and surgical intensive care unit (ICU) and department of nuclear medicine of a university hospital. Thirteen consecutive patients at risk for ARDS with sepsis (n = 4), multiple trauma (n = 3), hemorrhagic pancreatitis (n = 5), and near-drowning (n = 1), admitted into the ICU. All patients were intubated and mechanically ventilated in the course of the study. None. The 67gallium-transferrin pulmonary leak index (PLI) (upper limit of normal in patients scheduled for major surgery 14.7 x 10(-3)/min) was measured within 24 hrs of admission. In addition, radiographic, gas exchange, and ventilatory variables allowing calculation of the Lung Injury Score, were obtained on the day of the PLI (day 0) and from days 1 to 7. Patients were followed until discharge or death in the ICU. No patient developed ARDS (Lung Injury Score of >2.5 in the absence of congestive heart failure) and all but four patients survived. The PLI (median [range]) was 18.5 (7.0 to 81.9) x 10(-3)/min and was supranormal in 8 of 13 patients (p < .001 vs. normal). The PLI correlated with the Simplified Acute Physiology Score (SAPS) II (r[s] = .79, p < .01) and was higher in nonsurvivors than in survivors (p < .05), with a tendency for a higher SAPS II in the nonsurvivors. On day 0, the chest radiographic score of alveolar abnormalities was 0 (0 to 4), indicating absence of radiographic abnormalities in most patients. Also, on day 0, the median PaO2/FIO2 ratio was 257 (range 119 to 460). The Lung Injury Score on day 0 was 1.0 (0 to 2.0), which is consistent with mild acute lung injury in all but one patient. In the course of time, the oxygenation ratio was lower and the Lung Injury Score was higher in nonsurvivors than in survivors, particularly for the Lung Injury Score on days 1 and 2. The PLI correlated directly with the Lung Injury Score on days 0 and 1. On day 0, the PLI correlated inversely with the oxygenation ratio. On the last evaluable day of the first ICU week, the PLI correlated directly with the Lung Injury Score and the positive end-expiratory pressure level. The duration of mechanical ventilation tended to be longer in patients with a supranormal PLI. In this small population, 61% of patients at risk for ARDS and with only mild respiratory changes exhibited increased microvascular permeability in the lungs at ICU admission. Increased permeability may be an early marker of acute lung injury and its clinical features, and may be associated with a relatively complicated respiratory course during the first week after admission, even in the absence of progression to ARDS. Finally, increased pulmonary microvascular permeability may be associated with severe underlying disease and ultimate mortality in the ICU.