Do age and gender influence outcome from mechanical ventilation?

OBJECTIVES

To define the role of age and gender on outcome from mechanical ventilation.

DESIGN

Prospective analysis of consecutive patients.

SETTING

A military-referral medical center.

PATIENTS

A total of 240 consecutive patients requiring 246 episodes of mechanical ventilation.

MEASUREMENTS

Demographic, clinical, and outcome data.

RESULTS

Univariate statistical techniques comparing 13 variables between survivors and nonsurvivors of an episode of mechanical ventilation indicated that the following eight variables were significantly different (p < 0.05) between these two groups: age, length of intensive care unit stay, duration of MV, Acute Physiology and Chronic Health Evaluation (APACHE) II score, Age Removed APACHE II score, Organ System Failure Index (OSFI), patient diagnostic category, and presence of malignancy. Patient sex demonstrated a trend towards significance (p = 0.051). These nine variables were entered into a stepwise logistic regression analysis to develop a model to predict outcome from mechanical ventilation. This model showed that only the OSFI (p < 0.001) and the APACHE II score (p = 0.0281) independently predicted outcome from mechanical ventilation. Similar analyses were performed to identify outcome predictors after segregating the patients according to age (less than 70 years and equal to or greater than 70 years) and according to sex. In the first of these subsequent analyses the OSFI was found to be the sole independent predictor of outcome for both older and younger subgroups. In the second analysis the OSFI, length of hospital stay, and duration of mechanical ventilation predicted outcome for male patients while the OSFI alone independently predicted outcome for female patients. Fifty-two survivors of mechanical ventilation over the age of 70 years were observed after hospital discharge (average follow-up time 2.75 +/- 1.36 months) to assess their need for institutional care after hospital discharge. Five of these patients (9.6%) required long-term care in a nursing home or rehabilitation hospital until their deaths (four patients) or discharge to home (one patient). With a stepwise logistic regression analysis, only the intensive care unit length of stay was found to be significant independent predictor for this outcome.

CONCLUSIONS

These results suggest that the occurrence of derangements in organ function as described by the OSFI best predicts outcome from mechanical ventilation. Age and sex do not significantly contribute to patient outcome independent of organ system dysfunction.