Is the Duration of Mechanical Ventilation Predictable?
In order to identify the need for prolonged intubation and perform tracheotomy at the earliest appropriate time, accurate prediction of the length of mechanical ventilation (MV) is essential. A prospective cohort study of 203 episodes of respiratory failure in 195 adult patients in a surgical ICU was carried out to assess whether a predictive scheme based on clinical characteristics, diagnoses, clinical scoring systems, and physiological variables could be used to identify patients requiring more than 15 days of intubation. The studied factors were collected prospectively in all intubated patients admitted to the SICU or requiring intubation during SICU treatment. Patients were divided into two groupsthose requiring more than 15 days of intubation (LONG) and those requiring less (SHORT). One hundred seventy-seven episodes of MV in 169 patients comprised the LONG group, and 26 episodes in 26 patients comprised the SHORT group.
The mortality was numerically but not statistically different between the groups: 37% (LONG) vs. 50% (SHORT). The groups were different in SICU length-of-stay, emergency intubation, indication for MV, sepsis score at admission, organ system failures at admission and at intubation, LIS at admission and intubation, and serum albumin concentration. However, only the LIS and the need for emergency intubation were independent predictors of LONG or SHORT. An LIS of 1 or more had a sensitivity of 0.61 and a specificity of 0.66 (positive predictive value of 0.22 and negative predictive value of 0.93); adding emergency intubation decreased sensitivity to 0.46 but increased specificity to 0.81.
The LIS predictor was tested prospectively in another 128 patients. Twenty-three of 26 patients requiring MV for more than 15 days had an LIS of 1 or more (sensitivity = 0.88); however, 29 of the 102 patients not requiring MV for more than 15 days had an LIS of less than 1 (specificity = 0.28). The negative predictive value was 0.91, and only three patients with an LIS of less than 1 required more than 15 days MV. Unfortunately, the only recommendation supported by these data was that if the LIS is less than 1, it is unlikely that MV will be needed longer than 15 days. Tracheotomy should be delayed in this group. Identification of individual patients who required prolonged intubation and might benefit from early tracheotomy was poor. (Troche G, Moine P. Chest 1997;112:745-751.)
COMMENT BY CHARLES G. DURBIN, Jr., MD, FCCM
Troche and Moine asked the question, "Is the duration of mechanical ventilation predictable?" and the answer in their patients was "No." Despite the collection of a large number of data elements and the application of sophisticated analysis techniques, it was not possible to identify on admission or at the time of intubation those patients who would eventually require a long period of mechanical ventilation. Patients with a low lung injury score (< 1) could be expected not to need a long period of MV, although several of the patients meeting this criterion experienced a period of MV greater than two weeks.
This conclusion is not surprising. More important than the acute need for mechanical ventilation is the progress or remission of the pathologic processes leading to this need. The rapidity of change correlates with outcome, good or bad. With acute disease, failure to improve is a bad prognostic observation. The concept of changing physiology has been used with other predictive indices. In the APACHE III system, the rate of improvement (or worsening) of the physiology components of the score is used to calculate daily predictions of death for specific patients. This approach could be used to predict the need for continuing MV and a decision for earlier tracheotomy could be entertained, but a much larger group of patients and daily measurements of important factors would be necessary.
Currently, it remains difficult to determine whether a patient will require prolonged MV, although as the disease process continues, observant clinicians become more accurate in identifying these patients. Instead of delaying the decision for tracheotomy until the end of the third week, it should be possible to perform this procedure in most patients within two weeks of intubation. With the availability of percutaneous tracheotomy, the risks of an unnecessary procedure are minimal and the potential benefits from a shortened length of stay (and perhaps length of MV) are considerable.
The need for more than 15 days of mechanical ventilation:
a. is easily predicted by the lung injury score on the first day.
b. can be predicted only in medical patients after emergency intubation.
c. is poorly predicted by the lung injury score on the first day.
d. never occurs in surgical patients.
e. is an indication for a larger endotracheal tube.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.