The purpose of this study was to prospectively compare parameters that might predict successful translaryngeal extubation and tracheostomy tube decannulation. Irrespective of ventilatory function, 62 extubation/decannulation attempts were made on 49 consecutive patients with primarily neuromuscular ventilatory insufficiency who satisfied criteria. Thirty-four patients required 24-h ventilatory support. Noninvasive intermittent positive pressure ventilation (IPPV) was substituted as needed for IPPV via translaryngeal or tracheostomy tubes. Successful decannulation was defined as extubation or decannulation and site closure with no consequent respiratory symptoms or blood gas deterioration for at least 2 weeks. Failure was defined by the appearance of respiratory distress and decreases in vital capacity and oxyhemoglobin saturation despite use of noninvasive IPPV and assisted coughing. The independent variables of age, extent of predecannulation ventilator use, vital capacity, and peak cough flows (PCF) were studied to determine their utility in predicting successful extubation and decannulation. Only the ability to generate PCF greater than 160 L/min predicted success, whereas inability to generate 160 L/min predicted the need to replace the tube. All 43 attempts on patients with PCF greater than 160 L/min succeeded; all 15 attempts on patients with PCF below 160 L/min failed; and of 4 patients with PCF of 160 L/min, 2 succeeded and 2 failed. We conclude that the ability to generate PCF of at least 160 L/min is necessary for the successful extubation or tracheostomy tube decannulation of patients with neuromuscular disease irrespective of ability to breathe.