There is a growing body of evidence in favour of continuous adductor canal block (CACB) for total knee arthroplasty. However, there are no studies describing the optimal duration of the infusion. At our institution the usual practice was to stop the infusion on day three. Our hypothesis was that extending the infusion to five days would improve analgesia and quality of recovery. A prospective, non-blinded, randomised trial was undertaken. Patients received a continuous infusion of 0.2% ropivacaine via an adductor canal catheter for either three or five days. Primary outcome was pain while walking during the 24-hour period up to day five (numeric rating scale from 0 to 10). The minimum clinically important difference was set at 1.5 on the numeric rating scale. Secondary outcome measures included quality of recovery, mobility, pain while walking on postoperative day six, Oxford Knee Scores, and complications. Eighty-six patients were recruited with 43 randomised to each group. Seventy-eight were analysed. Median pain scores reported on day five were significantly better in the intervention group (1 versus 3, P=0.003). Furthermore, quality of recovery (QOR-15) scores were significantly better in the intervention group (133.6 versus 123.4, P=0.017). No statistically significant difference between groups was identified for other secondary outcome measures. CACB prolonged to five days provides superior analgesia and a higher quality of recovery on postoperative days four and five compared to a three-day infusion. This benefit did not extend beyond the period of infusion.