Does preoperative risk for delirium moderate the effects of postoperative pain and opiate use on postoperative delirium?

Abstract

Objectives: To investigate whether preoperative risk for delirium moderates the effect of postoperative pain and opioids on the development of postoperative delirium.

Design: Prospective cohort study.

Setting: University medical center.

Participants: Patients 65 years of age or older scheduled for major noncardiac surgery.

Measurements: A structured interview was conducted preoperatively and postoperatively to determine the presence of delirium, defined using the Confusion Assessment Method. We first developed a prediction model to determine which patients were at high versus low risk for the development of delirium based on preoperative patient data. We then computed a logistic regression model to determine whether preoperative risk for delirium moderates the effect of postoperative pain and opioids on incident delirium.

Results: Of 581 patients, 40% developed delirium on days 1 or 2 after surgery. Independent preoperative predictors of postoperative delirium included lower cognitive status, a history of central nervous system disease, high surgical risk, and major spine and joint arthroplasty surgery. Compared with the patients at low preoperative risk for developing delirium, the relative risk for postoperative delirium for those in the high preoperative risk group was 2.38 (95% confidence interval: 1.67-3.40). A significant three-way interaction indicates that preoperative risk for delirium significantly moderated the effect of postoperative pain and opioid use on the development of delirium. Among patients at high preoperative risk for development of delirium who also had high postoperative pain and received high opioid doses, the incidence of delirium was 72%, compared with 20% among patients with low preoperative risk, low postoperative pain, and those who received low opioid doses.

Conclusions: High levels of postoperative pain and using high opioid doses increased risk for postoperative delirium for all patients. The highest incidence of delirium was among patients who had high preoperative risk for delirium and also had high postoperative pain and used high opioid doses.

Keywords: Delirium; aging; opioids; pain; postoperative; preoperative risk stratification; surgery.

Copyright © 2013 American Association for Geriatric Psychiatry. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1. The interaction of preoperative risk predictors and postoperative pain and opioids on postoperative delirium

Low risk is defined when patients have 3 risk points or fewer and high risk with 4 or more risk points. The risk points are defined by giving 1 point to each of the independent risk factors identified in the model where: Female gender = 1 point A history of central nerve system disorder = 1 point Preoperative TICs score (35 = 0 point) Surgical risk (high = 1 point, low or intermediate = 0 point) Surgery type (orthopedic and spine surgery = 1 point, others = 0 point) TICS = telephone interview of cognitive status, CNS = central nervous system, mod = moderate See text for details.

Figure 2. Delirium Rate by Preoperative Risk Group, Pain, and Opioid

The incidence of postoperative delirium (%) is shown, stratified by whether the patients were in the low vs. high preoperative risk categories. A logistic regression was performed in each preoperative risk group. Comparisons were made on the effects of different combinations of pain status and opioid doses on incident delirium. The reference group for this comparison was patients with low pain and low opioid dose, within each risk category. The test statistics and P-values are as follows: P1: χ2(df=1) = 12.40, P = .0004. P2: χ2(df=1) = 0.87, P = .350. P3: χ2(df=1) = 0.10, P = .752. P4: χ2(df=1) = 4.66, P = .031. P5: χ2(df=1) = 3.20, P = .074. P6: χ2(df=1) = 6.26, P = .012. POD = postoperative day, NRS = numerical rating scale of pain assessment