Study: Postoperative delirium triples risk of complications, death in older patients

Editor's Note

Postoperative delirium significantly worsens outcomes for older adults undergoing major noncardiac surgery, according to research published July 8 in JAMA Network Open.

Specifically, findings showed patients who developed postoperative delirium had 3.5 times the odds of death or major complications, 2.8 times the odds of 30-day mortality, and 4 times the odds of being discharged to a facility rather than home. The study also found that risk varied markedly by hospital, suggesting quality improvement opportunities tied to delirium prevention and detection.

The retrospective cohort analysis focused on 5,530,054 Medicare beneficiaries aged 65 years and older who underwent major noncardiac surgery at 3,169 US hospitals between 2017 and 2020. Using ICD-10 codes for delirium and encephalopathy, researchers identified 197,921 cases of postoperative delirium (3.6%). Outcomes were assessed using multivariable logistic regression adjusting for a wide range of patient, surgical, and hospital factors.

Key adjusted findings include:

• Death or major complications: aOR 3.47 (95% CI, 3.41–3.53; P < .001)
• 30-day mortality: aOR 2.77 (95% CI, 2.71–2.83; P < .001)
• Nonhome discharge: aOR 3.96 (95% CI, 3.88–4.04; P < .001)

In unadjusted analyses:

• Delirium was associated with a 5.5-fold increased risk of death or major complications
• 2% of patients with delirium died within 30 days, compared to 2.1% without (OR 4.54)
• 7% of patients with delirium required nonhome discharge, compared to 24.9% without (OR 6.09)

The odds of developing postoperative delirium also differed by hospital. After controlling for patient risk factors, the median odds ratio (OR) for delirium incidence between high- and low-incidence hospitals was 1.53 (95% CI, 1.50–1.56). Patients treated in hospitals with historically high delirium rates in 2019 were more likely to experience delirium in 2020 (aOR, 1.46), while those treated at historically low-incidence hospitals had lower odds (aOR, 0.52).

Delirium incidence rose sharply with patients’ estimated 30-day mortality risk. Compared to patients with a mortality risk <0.25%, the odds of delirium were:

• 10 for those with a 0.25–0.49% risk
• 81 for 0.50–1.99%
• 90 for 2.00–4.99%
• 00 for 5.00–9.99%
• 50 for ≥10% (all P < .001)

The authors caution that claims-based identification may underreport delirium, and variation in coding practices could partly explain hospital differences.