Alcohol misuse raises mortality, ventilation odds in ICU patients with pneumonia

Key takeaways:

• Alcohol misuse vs. no misuse increased the odds for several poor outcomes.
• Higher odds for mechanical ventilation were also found with stimulant and multiple substance misuse.

Patients in the ICU for pneumonia with vs. without alcohol misuse faced heightened odds for mortality, mechanical ventilation and extended hospital stays, according to results published in Annals of the American Thoracic Society.

“The more than threefold increase in cost dedicated to caring for patients with pneumonia who require ICU care compared with patients who do not require ICU level of care, coupled with the $13 billion cost associated with hospitalizations due to substance use disorders, emphasizes the importance of identifying potentially modifiable associations between substance misuse and outcomes in pneumonia and underscores the importance of our findings,” Paul M. Reynolds, PharmD, BCCCP, clinician pharmacy specialist at VA Eastern Colorado Health Care System, and colleagues wrote.

In a retrospective cohort study, Reynolds and colleagues analyzed 167,095 ICU patients with pneumonia as a primary or secondary diagnosis in the Premier Healthcare Database from 2010 to 2017 to evaluate differences in odds for in-hospital mortality when these patients misused alcohol, cannabis, opioids, stimulants or multiple substances.

All patients received antibiotics or antivirals within 1 day of admission.

Following assessment of substance misuse and mortality, researchers evaluated how different types of substance misuse were related to mechanical ventilation, vasopressor administration and continuous paralytic administration.

Researchers used multivariable mixed-effects logistic regression models that accounted for age, sex, race/ethnicity, revised combined comorbidity score, hospital clustering, urban vs. rural location, teaching vs. nonteaching and hospital bed size to find associations.

Of the total sample, 152,045 patients (91.1%) did not have identified substance misuse, and 25,402 patients (15%) died in the hospital. Among the different types of misuse, alcohol misuse was the most common in 8,417 patients (5%), followed by opioid misuse in 2,579 patients (1.5%), misuse of multiple substances in 2,051 patients (1.2%), stimulant misuse in 1,048 patients (0.6%) and cannabis misuse in 955 patients (0.6%).

Odds for in-hospital mortality

Prior to adjusting for the variables outlined above, researchers found heightened odds for in-hospital mortality with alcohol misuse vs. no misuse (OR = 1.12; 95% CI, 1.06-1.19), which also appeared in adjusted analysis (aOR = 1.31; 95% CI, 1.23-1.4).

On the other hand, lowered odds for mortality were linked to opioid misuse vs. no misuse in unadjusted analysis (OR = 0.46; 95% CI, 0.39-0.53), and this outcome was found again after adjusting for the outlined variables (aOR = 0.63; 95% CI, 0.54-0.73).

Researchers continued to see significant reduced odds for mortality in unadjusted analyses when assessing cannabis (OR = 0.67; 95% CI, 0.54-0.82), stimulant (OR = 0.55; 95% CI, 0.44-0.68) and multiple substance (OR = 0.62; 95% CI, 0.54-0.71) misuse compared with no substance misuse. However, in the adjusted analysis, researchers no longer observed these relationships.

Similar findings were observed in a model that additionally factored in the need for mechanical ventilation or vasopressors within a patients’ first 2 days in the hospital, demonstrating higher odds for mortality with alcohol misuse (aOR = 1.23; 95% CI, 1.16-1.32) and reduced odds with opioid misuse (aOR = 0.57; 95% CI, 0.49-0.67), stimulant misuse (aOR = 0.74; 95% CI, 0.59-0.92) and misuse of multiple substances (aOR = 0.83; 95% CI, 0.71-0.97).

Odds for secondary outcomes

Switching to evaluating mechanical ventilation, researchers found that alcohol (aOR = 1.55; 95% CI, 1.48-1.6), stimulant (aOR = 1.25; 95% CI, 1.1-1.43) and multiple substance (aOR = 1.42; 95% CI, 1.29-1.56) misuse elevated these odds in adjusted analysis.

Alcohol misuse vs. no misuse also raised the odds for mechanical ventilation after 2 days of hospitalization (aOR = 1.31; 95% CI, 1.24-1.39), whereas opioid and stimulant misuse lowered these odds (opioid, aOR = 0.88; 95% CI, 0.78-0.98; stimulant, aOR = 0.8; 95% CI, 0.67-0.96).

Besides opioid misuse, researchers observed significant increased odds for vasopressor use with alcohol misuse (aOR = 1.52; 95% CI, 1.45-1.6), cannabis misuse (aOR = 1.2; 95% CI, 1.04-1.4), stimulant misuse (aOR = 1.21; 95% CI, 1.05-1.39) and multiple substance misuse (aOR = 1.28; 95% CI, 1.16-1.41) vs. no substance misuse.

Similar findings were reported in the odds for continuous paralytic use, with higher odds for those with alcohol misuse (aOR = 1.46; 95% CI, 1.38-1.55), cannabis misuse (aOR = 1.3; 95% CI, 1.11-1.52), stimulant misuse (aOR = 1.33; 95% CI, 1.15-1.55) and multiple substance misuse (aOR = 1.5; 95% CI, 1.35-1.67) vs. no substance misuse.

Lastly, researchers found prolonged hospital stays among those who misused alcohol (least squares means, 8.2 days) and those who misused multiple substances (8.11 days) vs. those who did not misuse any substances (7.47 days).

“In evaluating and interpreting these findings, it is important to keep in mind that selection bias and residual confounding for severity of illness may have impacted our results,” Reynolds and colleagues wrote. “Our observations support a need for clinicians to consistently consider and characterize substance use habits in patients with pneumonia as comorbid conditions with the potential to impact outcomes and need for services provided in the ICU.”