Objectives:
While appropriate analgesia use after knee arthroplasty remains an important aspect of patient management and rehabilitation, high opioid prescribing rates have led to an increase in opioid misuse disorders and opioid-related deaths worldwide. Orthopaedic surgery is the 3rd highest prescriber of opioids by specialty and a number of studies have indicated that long-term opioid use is associated with a poorer post-operative outcome. The American Academy of Orthopaedic Surgeons (AAOS) has recommended that opioid use should be minimised following surgery. Our centre has previously reported an increased rate of continuing opioid prescribing in patients who have been prescribed opioids prior to arthroplasty.

Our centre has instituted a policy of avoiding prescription of strong opioids on discharge from hospital following knee arthroplasty (including morphine, oxycodone, tramadol and fentanyl preparations) if these were not being used prior to admission. Instead, paracetamol, non-steroidal anti-inflammatory drugs (NSAIDs) or weak opioid preparations (codeine or dihydrocodeine) are prescribed. However, strong opioids may be prescribed in the community prior to admission and we hypothesised that these patients would have a higher rate of strong opioid prescription following knee arthroplasty.

Methods

The aim of our study was to analyse the prescription of opioid preparations pre-and post-knee arthroplasty. We retrospectively identified a series of patients undergoing primary knee arthroplasty over a 24-month period. Exclusion criteria included simultaneous or sequential bilateral arthoplasty or mortality during the study period. 741 consecutive knee arthroplasty patients (427 females, 314 males; mean age 70 years, range 46-90 years) were analysed. 39 patients underwent unicondylar knee replacement (UKR) with the remainder (702 patients) undergoing total knee replacement (TKR). Repeat prescriptions at the time of pre-operative assessment clinic visit were compared with discharge prescriptions following arthroplasty and then with primary care repeat prescriptions at a minimum of 12 months follow-up.

Results

At time of pre-op assessment, 103 patients (14%) were prescribed strong opioids (morphine, oxycodone, buprenorphine, fentanyl, tramadol) and 282 (38%) were prescribed weak opioids (codeine, dihydrocodeine). 577 patients (78%) were prescribed weak opioids on discharge. 60% of those on strong opioids (62 patients) pre-op were subsequently prescribed strong opioids for discharge after arthroplasty.

85 patients (11%) were prescribed strong opioids at follow-up, of whom 54 had been prescribed strong opioids pre-op. 217 patients (29%) were prescribed weak opioids at follow-up. The relative risk for strong opioid use at 12 months follow-up given pre-operative strong opioid use was 10.79 (95% confidence interval 7.31-15.93, p<0.001).

Conclusions

Both strong and weak opioids are commonly prescribed for patients awaiting knee arthroplasty, as well as on discharge following surgery. In keeping with previous similar studies, our series suggests that pre-operative use of strong opioids is associated with an increased rate of prolonged strong opioid prescribing in the community following primary knee arthroplasty.