Patients undergoing total hip or knee arthroplasty (THA/TKA) are commonly associated with high pain scores and narcotic use. Duloxetine is effective in relieving chronic pain. The aim of this study was to evaluate the safety and efficacy of duloxetine for pain management following THA/TKA.

Five major databases (PubMed, Embase, Scopus, Cochrane, and Web of Science) were searched for randomised controlled trials (RCTs) that compared duloxetine to placebo in patients that underwent THA/TKA. The primary outcome was pain reduction with rest and movement at short-term and long-term time intervals. Secondary outcomes were the use of analgesics, length of stay, and safety profile. The risk of bias was assessed using the Cochrane tool. Data were pooled using RevMan 5.4. The results were reported as mean difference (MD) or standardised MD (SMD) and 95% confidence intervals (CI).

Eight RCTs with 767 patients were included. 50.2% (n=385/767) of patients received duloxetine. After one day, duloxetine was superior to the control regarding pain reduction with rest (SMD= −0.22 [−0.41, −0.03], p=0.02) after sensitivity analysis and pain reduction at movement (SMD= −0.39 [−0.55, −0.24], p<0.001). Similarly, after 12 weeks, duloxetine significantly reduced pain with rest (SMD= −0.3 [−0.52, −0.09], p=0.006) and pain with movement (SMD= −0.52 [−0.87, −0.17], p=0.003). In addition, after sensitivity analysis, duloxetine was associated with less analgesic use after one day (MD= −4.65 [−7.3, −2.01], p<0.001) and two days (MD= −5.65 [−10.62, −0.67], p=0.03). Patients who received duloxetine also required fewer analgesics after three days. However, there was no significant difference between the duloxetine and control groups in analgesic use after one week, length of stay, and adverse events.

Duloxetine was superior to the placebo regarding short-term and long-term pain reduction with rest and movement following THA/TKA. Duloxetine reduced postoperative analgesic use. There was no significant difference between duloxetine and placebo regarding adverse events and length of stay.

The primary aim was to identify patient and injury factors independently associated with humeral diaphyseal fracture nonunion after non-operative management. The secondary aim was to determine the effect of management (operative/non-operative) on nonunion.

Over a ten-year period, 734 humeral diaphyseal fractures (732 consecutive patients) were retrospectively identified from a trauma database. Follow-up was available for 663 fractures (662 patients; median age 57yrs [16–96], 54% female [n=359/662]) which formed the study cohort. Patient and injury characteristics were recorded. There were 523 patients (79%) managed non-operatively and 139 (21%) managed operatively. Outcome (union/nonunion) was determined from medical records and radiographs.

Median follow-up was five months (1.2–74). Nonunion occurred in 22.7% of non-operatively managed injuries (n=119/524). Multivariate analysis demonstrated pre-injury NSAIDs (adjusted OR [AOR] 40.8, 95% CI 2.6–632.3; p=0.008), being underweight (BMI <18.5kg/m2; AOR 7.3, 95% CI 1.3 to 40.2; p=0.022), overweight (BMI 25–29.9kg/m2; AOR 2.5, 95% CI 1.1 to 5.9; p=0.034) and class II obesity (BMI 35–39.9kg/m2; AOR 4.5, 95% CI 1.4 to 15.5; p=0.014) were independently associated with an increased risk of nonunion. Operative fixation was independently associated with a lower risk of nonunion (2.9%, n=4/139) than non-operative management (AOR for non-operative/operative 11.0, 95% CI 2.8 to 43.6; p=0.001). Based upon these results, five patients would need to undergo operative fixation to avoid one nonunion.

Pre-injury NSAIDs and BMI were independently associated with nonunion following non-operative management of a humeral diaphyseal fracture. Operative fixation was the independent factor most strongly associated with a lower risk of nonunion.