• Despite a good outcome for many patients, approximately 20% of patients experience chronic pain after total knee arthroplasty (TKA).

• Chronic pain after TKA can affect all dimensions of health-related quality of life, and is associated with functional limitations, pain-related distress, depression, poorer general health and social isolation.

• In both clinical and research settings, the approach to assessing chronic pain after TKA needs to be in-depth and multidimensional to understand the characteristics and impact of this pain. Assessment of this pain has been inadequate in the past, but there are encouraging trends for increased use of validated patient-reported outcome measures.

• Risk factors for chronic pain after TKA can be considered as those present before surgery, intraoperatively or in the acute postoperative period. Knowledge of risk factors is important to guide the development of interventions and to help to target care. Evaluations of preoperative interventions which optimize pain management and general health around the time of surgery are needed.

• The causes of chronic pain after TKA are not yet fully understood, although research interest is growing and it is evident that this pain has a multifactorial aetiology, with a wide range of possible biological, surgical and psychosocial factors that can influence pain outcomes.

• Treatment of chronic pain after TKA is challenging, and evaluation of combined treatments and individually targeted treatments matched to patient characteristics is advocated. To ensure that optimal care is provided to patients, the clinical- and cost-effectiveness of multidisciplinary and individualized interventions should be evaluated.

Cite this article: EFORT Open Rev 2018;3:461-470. DOI: 10.1302/2058-5241.3.180004

• Twenty randomized controlled trials comprising 1893 primary total knee replacements were included in this review.

• The subvastus approach conferred superior results for mean difference (MD) in time to regain an active straight leg raise (1.7 days, 95% confidence interval [CI] 1.0 to 2.3), visual analogue score for pain on day one (0.8 points on a scale out of 10, 95% CI 0.2 to 1.4) and total range of knee movement at one week (7°, 95% CI 3.2 to 10.7). The subvastus approach also resulted in fewer lateral releases (odds ratio 0.4, 95% CI 0.2 to 0.7) and less peri-operative blood loss (MD 57 mL, 95% CI 10.5 to 106.4) but prolonged surgical times (MD 9.7 min, 95% CI 3.9 to 15.6).

• There was no difference in Knee Society Score at six weeks or one year, or the rate of adverse events including superficial or deep infection, deep vein thrombosis or knee stiffness requiring manipulation under anaesthesia.

• This review demonstrates evidence of early post-operative benefits following the subvastus approach with equivalence between approaches thereafter.

Cite this article: EFORT Open Rev 2018;3:78-84. DOI: 10.1302/2058-5241.3.170030.