PURPOSE: Hospitalization for ARF due to COPD is associated with high mortality. There is limited information regarding the predictors of mortality from this region which would help in prioritizing ICU care and focused approach.
METHODS: Hundred patients (mean age 59.6±10.2 Yrs.; M:F = 75:25) with ARF due to acute exacerbation of COPD admitted to medical ICU were studied. Essential criteria included PaO2 < 55 mm Hg or PaCO2 > 45 mmHg.Patients with know cardiac, renal and hepatic diseases were excluded. Various clinical and biochemical parameters were compared between survivors and non-survivors. Multiple logistic regressions were applied to correlate these factors with mortality. ROC curves were used to find sensitivity and specificity of predictors.
RESULTS: Seventy three patients required mechanical ventilation.36 patients died during hospital stay. First day APACHE-II score (16.44±6.86 vs 11.61±3.21; p-0.00), blood urea (82.25±54.72 mg/dl vs 44.33±22.12 mg/dl; p-0.000), and creatinine (1.63±0.51 mg/dl vs 1.03±0.39 mg/dl; p-0.020), respectively were significantly higher in non-survivor group. Non-survivor group had significantly lower serum albumin (3.42±0.51gm/dl vs 3.14±0.48 gm/dl; p-0.010) levels. Duration of ICU (13.19±7.23 days vs 9.91±6.00 days; p-0.093) and hospital stay (20.03±11.73 days vs 15.48±8.72 days; p-0.047) and mechanical ventilation (14.25±10.09 days vs 8.12±5.60 days; p-0.003) were longer in non survivor group. Patient having lung infiltrates on X-ray had higher chances of dying (p-0.010). The best predictor of mortality was day-3 APACHE-II score. Using a cut off of >12.5 it showed a sensitivity and specificity of 97.3% and 97%, respectively. First day pH (7.25±0.15 vs 7.24±0.16; p-0.732), PaO2 (68.62±25.15 mmHg Vs 68.10±26.13 mmHg; p-0.921), and PaCO2 (80.05±25.86 mmHg vs 78.73±21.37 mmHg; p-0.795) among nonsurvivors and survivors respectively were not significantly different.
CONCLUSION: Acute exacerbation of COPD is associated with significant mortality. Day-1serum albumin, creatinine, blood urea, and presence of lung infiltrates on x-ray and APACHE-II score on day-3 are useful parameters in predicting mortality in these patients.
CLINICAL IMPLICATIONS: These predictors of mortality may be used for prioritizing intensive care. The studies evaluating their role in treatment decision are desired.
DISCLOSURE: Gopi Khilnani, No Financial Disclosure Information; No Product/Research Disclosure Information