A comparison of seriously ill patients with or without AIDS requiring prolonged mechanical ventilation

Chronic Comorbid Illnesses Predict the Clinical Course of 866 Patients Requiring Prolonged Mechanical Ventilation in a Long-Term, Acute-Care Hospital

OBJECTIVE

To determine whether burdens of chronic comorbid illnesses can predict the clinical course of prolonged mechanical ventilation (PMV)patients in a long-term, acute-care hospital (LTACH).

METHODS

Retrospective study of 866 consecutive PMV patients whose burdens of chronic comorbid illnesses were quantified using the Cumulative Illness Rating Scale (CIRS). Based on increasing CIRS scores, 6 groups were formed and compared: group A (≤25; n = 97), group B (26-28; n = 105), group C (29-31; n = 181), group D (32-34; n = 208), group E (35-37; n = 173), and group F (>37; n = 102).

RESULTS

As CIRS scores increased from group A to group F, rates of weaning success, home discharges, and LTACH survival declined progressively from 74% to 17%, 48% to 0%, and 79% to 21%, respectively (all P < .001). Negative correlations between the mean score of each CIRS group and correspondent outcomes also supported patients' group allocation and an accurate prediction of their clinical course (all P < .01). Long-term survival progressively declined from a median survival time of 38.9 months in group A to 3.2 months in group F (P < .001). Compared to group A, risk of death was 75% greater in group F (P = .03). Noteworthy, PMV patients with CIRS score <25 showed greater ability to recover and a low likelihood of becoming chronically critically ill. Diagnostic accuracy of CIRS to predict likelihood of weaning success, home discharges, both LTACH and long-term survival was good (area under the curves ≥0.71; all P <.001).

CONCLUSIONS

The burden of chronic comorbid illnesses was a strong prognostic indicator of the clinical course of PMV patients. Patients with lower CIRS values showed greater ability to recover and were less likely to become chronically critically ill. Thus, CIRS can be used to help guide clinicians caring for PMV patients in transfer decisions to and from postacute care setting.

Relationship of the Duration of Ventilator Support to Successful Weaning and Other Clinical Outcomes in 437 Prolonged Mechanical Ventilation Patients

OBJECTIVE

To investigate the relationships between durations of ventilator support and weaning outcomes of prolonged mechanical ventilation (PMV) patients.

METHODS

Cohort study of 957 PMV patients sequentially admitted to a long-term acute care hospital (LTACH). The study population was 437 PMV patients who underwent weaning, having achieved ≥4 hours of sustained spontaneous breathing. They were divided into tertiles of mechanical ventilation (MV) durations and compared for differences (tertile A: 21-58 days, n = 146; tertile B: 59-103 days, n = 147; and tertile C: ≥104 days, n = 144).

RESULTS

Tertiles showed comparable weaning success rates and survival. As MV durations increased, LTACH postweaning days became progressively greater, whereas decannulations and discharge physical function diminished, and home discharges decreased while nursing facility discharges increased (all P < .001). Patients with lower physical function before critical illness or greater burdens of comorbidities were least likely to be weaned (all P < .001). Younger ages, lower comorbidity burdens, neurological diagnoses, higher admission prealbumin levels, and successful weaning, each independently reduced the risk of death (all P < .01).

CONCLUSION

Durations of MV did not affect weaning success or survival, although deleterious effects were found in discharges, decannulations, LTACH postweaning days, and discharge physical function. Durations of MV alone should not guide transfer decisions for subsequent continuing care.

Etiologies, clinical features and outcome of cardiac arrest in HIV-infected patients

BACKGROUND

Compared to many other cardiovascular diseases, there is a paucity of data on the characteristics of successfully resuscitated cardiac arrest (CA) patients with human immunodeficiency virus (HIV) infection. We investigated causes, clinical features and outcome of these patients, and assessed the specific burden of HIV on outcome.

METHODS

Retrospective analysis of HIV-infected patients admitted to 20 French ICUs for successfully resuscitated CA (2000-2012). Characteristics and outcome of HIV-infected patients were compared to those of a large cohort of HIV-uninfected patients admitted after CA in the Cochin Hospital ICU during the same period.

RESULTS

99 patients were included (median CD4 lymphocyte count 233/mm(3), viral load 43 copies/ml). When compared with the control cohort of 1701 patients, HIV-infected patients were younger, with a predominance of male, a majority of in-hospital CA (52%), and non-shockable initial rhythm (80.8%). CA was mostly related to respiratory cause (n=36, including 23 pneumonia), cardiac cause (n=33, including 16 acute myocardial infarction), neurologic cause (n=8) and toxic cause (n=5). CA was deemed directly related to HIV infection in 18 cases. Seventy-one patients died in the ICU, mostly for care withdrawal after post-anoxic encephalopathy. After propensity score matching, ICU mortality was not significantly affected by HIV infection. Similarly, HIV disease characteristics had no impact on ICU outcome.

CONCLUSIONS

Etiologies of CA in HIV-infected patients are miscellaneous and mostly not related to HIV infection. Outcome remains bleak but is similar to outcome of HIV-negative patients.

Prolonged mechanical ventilation in 540 seriously ill older adults: effects of increasing age on clinical outcomes and survival

OBJECTIVES

To investigate effects of older age, comorbidities, and physiological measures on outcomes of elderly adults requiring prolonged mechanical ventilation (PMV).

DESIGN

Retrospective cohort study.

SETTING

Public long-term acute care hospital (LTACH) with an active program for ventilator weaning from PMV.

PARTICIPANTS

Chronically seriously ill individuals with PMV aged 65 and older divided into six cohorts (65-69, 70-74, 75-79, 80-84, 85-89, ≥ 90) for comparative purposes (n = 540).

MEASUREMENTS

Main outcomes were weaning criteria met, weaning success, discharge dispositions, and long-term survival. Other outcomes included weaning duration, LTACH days, discharge physical function, tracheostomy decannulation, and relapses to ventilator support. Weaning success was defined as 4 weeks or longer entirely free from mechanical ventilator support.

RESULTS

The main finding from age cohort comparisons was that the likelihood of meeting weaning criteria (P = .001) and subsequent successful weaning (P = .002) decreased with age. Best predictors for weaning success in multivariable analysis were lower comorbidity burden (P < .001) and less-severe illness (P = .001). Other clinically important predictors were more-normal values in the respiratory physiology measures of rapid shallow breathing (P = .001) and static compliance (P = .003). Successful weaning was also associated with a 62% lower risk of death (P < .001).

CONCLUSION

Although meeting weaning criteria and being successfully weaned decreased with increasing age, age was not the dominant factor in predicting outcomes. More importantly, individuals with PMV with better respiratory physiology and lower comorbidity burdens were more likely to be weaned and have longer survival, no matter their age.