Predictors of in-hospital mortality and need for invasive mechanical ventilation in elderly COPD patients presenting with acute hypercapnic respiratory failure

• Main Authors: Ashraf Mokhtar Madkour, Nermien Naim Adly
• Format: Article
• Language: English
• Published: Wolters Kluwer Medknow Publications 2013-07-01
• Series: Egyptian Journal of Chest Disease and Tuberculosis
• Subjects: Elderly COPD; Hypercapnic respiratory failure; Predictors; Mortality; Invasive mechanical ventilation; Discriminant equation
• Online Access: http://www.sciencedirect.com/science/article/pii/S0422763813001416

Predictors of in-hospital mortality and need for invasive mechanical ventilation (IMV) in chronic obstructive pulmonary disease (COPD) patients presenting with acute hypercapnic respiratory failure (AHRF) are yet lacking, but warranted in elderly population. Formulating a scoring system may aid prognostication. A prospective study including all elderly patients with COPD and AHRF admitted to a Geriatric and Respiratory ICU over one year period. Different clinical and laboratory variables were recorded at the time of admission and were compared between IMV and non-IMV patients and survivor and non-survivor patients upon discharge. Thirty patients were enrolled. Mean (SD) age of patients was 68.2 ± 7.95; 56.7% had IMV. The inhospital morality was 13.3% all of whom died in ICU and had IMV. Renal impairment, previous history of hypertension, pneumonia, sepsis and lower glucose level were significant predictors of mortality and discriminant equation for the prediction of mortality had accuracy for the area under ROC curve of 94.2% having sensitivity, specificity, positive and negative predictive values for predicting death of 100%, 89%, 57% and 100%, respectively. Sepsis, pack year smoking, severe current exacerbation, Simplified Acute Physiology Scores II and lower pH were significant predictors of IMV and discriminant equation for the prediction of IMV had accuracy of 87.8% for the area under ROC curve having sensitivity, specificity, positive and negative predictive values for predicting need for IMV of 76%, 92%, 93% and 75%, respectively. The equations derived by combining these parameters are useful for identifying patients that are likely to have a poor outcome or need for IMV.