Risk factors of postoperative delirium in the knee and hip replacement patients: a systematic review and meta-analysis

Abstract

Background: The risk factors of postoperative delirium (POD), a serious while preventable complication, developed by patients undergoing knee and replacement surgery are still under investigation. In this systematic review and meta-analysis, we identified risk factors associated with POD in knee and hip replacement.

Methods: PubMed, Ovid MEDLINE, and Ovid EMBASE were used to identify original researches. The studies evaluating the risk factors of POD after knee and hip replacement were reviewed, and the qualities of the included studies were assessed with Newcastle-Ottawa Scale. Data were extracted, pooled, and a meta-analysis was completed RESULT: Twenty-two studies were finally included with a total of 11934 patients who underwent knee or hip replacement and 1841 developed POD with an incidence of 17.6% (95% confidential interval (CI) 13.2-22.0%). Eighteen significant risk factors were identified including advanced age (odds ratio (OR) 1.15 95% CI 1.08-1.22), cognitive impairment (OR 6.84, 95% CI 3.27-14.33), history of cerebrovascular events (OR 2.51, 95% CI 1.28-4.91), knee replacement (OR 1.42, 95% CI 1.00-2.02), blood loss (standardized mean difference (SMD) 0.30, 95% CI 0.15-0.44), dementia (OR 3.09, 95% CI 2.10-4.56), neurologic disorders (OR 2.26, 95% CI 1.23-4.15), psychiatric illness (OR 2.74, 95% CI 1.34-5.62), and obstructive sleep apnea (OR 4.17, 95% CI 1.72-10.09) along with several comorbidity evaluation scores and laboratory markers.

Conclusion: We identified risk factors consistently associated with the incidence of POD in knee and hip replacement. Strategies and interventions should be implemented to the patients receiving knee or hip replacement with potential risk factors identified in this meta-analysis.

Keywords: Hip replacement; Knee replacement; Postoperative delirium; Risk factors.

Fig. 1

Flow chart of literature search and screening

Fig. 2

Forest plots of the meta-analysis of the incidence of POD. a Incidence of POD of overall patients. b Incidence of POD of knee or hip replacement groups

Fig. 3

Forest plots and funnel plot of meta-analysis of advanced age. a Forest plot of SMD of age. b Forest plot of OR of advanced age. c Funnel plot of published studies reporting advanced age as a prognostic factor for POD (Egger’s test p < 0.001)

Fig. 4

Forest plots of risk factors of cognitive impairment and preoperative MMSE evaluations. a Forest plot of cognitive impairment. b Forest plot of preoperative MMSE evaluation

Fig. 5

Forest plot of history of cerebrovascular events as a risk factor

Fig. 6

Forest plots of risk factors of medical comorbidities. a Forest plot of dementia. b Forest plot of neurological disorders. c Forest plot of psychiatric illness. d Forest plot of obstructive sleep apnea

Fig. 7

Forest plots of surgical and anesthetic risk factors. a Forest plot of knee replacement (compared with hip replacement). b Forest plot of duration of surgery. c Forest plot of blood loss. d Forest plot of spinal anesthesia (compared with general anesthesia)

Fig. 8

Forest plots of preoperative evaluation of comorbidities in predicting POD. a Forest plot of ASA score ≥ III. b Forest plot of CCI score

Fig. 9

Forest plots of laboratory test in predicting POD. a Forest plot of the difference of preoperative total protein. b Forest plot of the difference of preoperative albumin. c Forest plot of the difference of preoperative hemoglobin