SURVIVAL AND PREDICTION OF SUCCESSFUL VENTILATOR WEANING IN COPD PATIENTS REQUIRING MECHANICAL VENTILATION FOR MORE THAN 21 DAYS

We studied survival and failure or success of weaning from mechanical ventilation (MV) in 42 consecutive chronic obstructive pulmonary disea se (COPD) patients requiring prolonged MV (more than 21 days) after an episode of acute respiratory failure requiring admission to our Inter mediate Intensive Care Unit (IICU). Parameters including arterial bloo d gases, pulmonary function tests, respiratory muscle force, neuromusc ular drive, and nutritional status were recorded during a phase of cli nical stability, in order to identify the features related to survival and weaning. All the patients were submitted to a comprehensive rehab ilitation programme. Successful weaning from MV was defined as complet e respiratory autonomy for at least 48 h. Twenty three patients were s uccessfully weaned from MV after an average period of 44 days (Group A ), requiring no further MV during their stay in hospital, whilst the r emaining 19 patients were not able to be disconnected from the ventila tor (Group B). The discriminant analysis showed that weaning from MV w as significantly associated with arterial carbon dioxide tension (Paco (2)), neuromuscular drive (P-0.1), maximal inspiratory pressure (MIP), arterial oxygen tension (Pao(2)), the ratio of respiratory frequency to tidal volume (f/VT) and the serum protein level. Other pulmonary fu nction tests (forced expiratory volume in one second (FEV(1))), FEV(1) /forced vital capacity (FVC), anthropometric data, nutritional status, number of pulmonary exacerbations during MV and evidence of cor pulmo nale, were similar in the two groups. The discriminant equation consid ering Paco(2) and MIP could separate the two groups with an accuracy o f 84%. The overall survival at 2 yrs was 40%; in Group B it was signif icantly lower than in Group A (22 vs 68%). Most of the deaths occurred within the first 120 days after intubation. None of the parameters co nsidered could significantly predict the survival rate. We conclude th at more than half of COPD patients requiring prolonged MV (more than 2 1 days) could be successfully weaned. A simple equation including Paco (2) and MIP could correctly identify the weaning success or failure in about 84% of patients. The survival rate at 2 yrs is poor in those wh o could not be weaned, but is not associated with any of the indices c onsidered.