Preoperative predictors of health-related quality of life changes (EQ-5D and EQ VAS) after total hip and knee replacement: a systematic review

Caroline Schatz,Nina Klein,Antonia Marx,Peter Buschner

doi:10.1186/s12891-021-04981-4

Abstract

BackgroundPatient-reported outcomes are of ever-increasing importance in medical decision-making. The EQ-5D is one of the generic instruments measuring health-related quality of life (HRQoL) in arthroplasty. This review aimed to identify possible predictors of HRQoL changes for patients undergoing total knee replacements (TKR) or total hip replacements (THR).MethodsA systematic literature review according to the PRISMA guidelines was conducted, searching several databases. Preoperative to postoperative HRQoL changes were evaluated in patients undergoing THR or TKR, using the EQ-5D visual analog scale (VAS) or the preference-based EQ-5D Index were evaluated. Articles were considered with prospectively or retrospectively collected data, as well as registry data, each with statistical analyses of patient-related factors.ResultsEight hundred eighty-two articles were found, of which 21 studies met the inclusion criteria. Predictors were distinguished in alterable and non-alterable ones. The EQ-5D Index indicated a tendency towards beneficial improvements for patients with a high body mass index (BMI) (> 40) and no significant results for the VAS. Additionally, one study found that patient education and preoperative physiotherapy appeared to enhance HRQoL. Some evidence indicated that male gender was negatively associated with changes in the VAS and the EQ-5D Index, but one study reported the opposite. Changes in VAS and EQ-5D Index were lower for older patients, whereas a higher educational level seemed to be advantageous. A high Charnley class led to deteriorating changes in VAS, although a high Kellgren Lawrence classification was positively associated with the EQ-5D Index, in a limited number of studies. For all results, clinical relevance was calculated differently and mainly reported as uncertain or small.ConclusionsThe literature on this topic was weak and offers only limited guidance. Results for alterable predictors, such as the BMI, indicated valuable improvements for highly obese patients. Further, high-quality research is required to support medical decision-making.Level of evidenceLevel IV, according to the OCEBM Levels of Evidence Working Group.

Highlights

Patient-reported outcomes are of ever-increasing importance in medical decision-making
There are patients who remain dissatisfied with total knee replacement (TKR) or total hip replacement (THR) [2, 3]
The p-value in statistics was chosen by the researcher and determines only a decision whether to decline the hypothesis or not, but this did not imply whether the change in health-related quality of life (HRQoL) was meaningful for the patient [20]

Summary

Introduction

Patient-reported outcomes are of ever-increasing importance in medical decision-making. The EQ-5D is one of the generic instruments measuring health-related quality of life (HRQoL) in arthroplasty. Total joint replacements are a recommended surgical procedure for patients with advanced osteoarthritis (OA) These interventions are performed to reduce pain and improve the function of joints for patients [1]. Patient-reported outcome measures are tools for the measurement of health-related quality of life (HRQoL). The questionnaire exists in two versions, the EQ-5D-3L and EQ-5D-5L, where the EQ-5D-5L is considered as the successor to the EQ-5D-3L [5, 6] Both contain five dimensions (mobility, selfcare, usual activities, pain/discomfort, anxiety/ depression) with each having three problem levels in the former and five in the latter. The vertical visual analog scale (VAS) completes the questionnaire Patients evaluate their own, current health state by setting a mark between 0 being the worst health state and 100 being the best health state [4]. For THR, the EQ VAS scores were found to be highly correlated, and their estimates for different problem levels to be largely consistent between the 3 L and 5 L-versions [8,9,10]

Objectives

Methods

Results

Conclusion