Risk factors associated with unplanned ICU admission following hip or knee arthroplasty: a systematic review and meta-analysis

Abstract

Background: Lower limb arthroplasty is not without risk with a significant proportion of patients requiring unplanned ICU admission. A collated summary of the various risk factors would contribute towards developing risk stratification methods aimed at reducing this perioperative outcome.

Objective: To perform a systematic review of published data on risk factors for unplanned ICU admissions following hip/knee arthroplasty.

Method: We conducted a systematic review with pre-defined eligibility criteria used to identify relevant manuscripts. Quality assessment was achieved using the Newcastle-Ottawa score. Study characteristics and identified risk factor data were extracted from eligible manuscripts. A meta-analysis was performed on the available data.

Results: Twelve published studies were included in our review. The pooled rate of ICU admission was 4.55% (95% CI 3.04–6.05%). Smoking history (OR 19.92, 95% CI 2.78 - 142.68, I2 = 47.9%), Charlson comorbidity index (OR 1.97, 95% CI 1.79 – 2.17, I2 = 0%), COPD (OR 3.25, 95% CI 1.12 – 9.4, I2 = 88.2%), coronary artery disease (OR 3.04, 95% CI 1.10 – 8.45, I2 = 81.9%), revision surgery (OR 2.17, 95% CI 1.35 – 3.50, I2 = 0%), excessive blood loss (OR 3.76, 95% CI 1.09 – 12. 93, I2 = 93.3%) and use of vasopressors (OR 5.00, 95% CI 2.22 – 11.27, I2 = 52.6%) were associated with an increased risk of ICU admission. Most estimates showed significant heterogeneity with I2of >50%.

Conclusion: Multiple patient and operative characteristics confer increased risk for unplanned ICU admission following hip or knee arthroplasty and can be used as markers of elevated perioperative risk facilitating optimal clinical decision making and guiding informed consent.