Risk of venous thromboembolism after knee, hip, or shoulder arthroplasty : a retrospective cohort study of 476,620 procedures using linked hospital and primary care data.

The overall VTE incidence within six months of arthroplasty was 1.5% across 476,620 procedures.

• Knee arthroplasty had the highest VTE incidence at 1.6%
• Hip arthroplasty had a VTE incidence of 1.4%
• Shoulder arthroplasty had the lowest VTE incidence at 1.0%
• The study period spanned October 2007 to September 2019
• Data were drawn from the Clinical Practice Research Datalink (CPRD) Aurum, the largest primary care database in the UK, linked with hospital and death data

More than 30% of VTE events were identified only through linked primary and secondary care data, highlighting the inadequacy of hospital data alone.

• Previous studies have largely relied solely on hospital data
• The linkage of primary care (CPRD Aurum) with hospital and death data captured VTE events that would have been missed using hospital records alone
• This finding underscores the importance of linked datasets for accurately measuring postoperative VTE rates

VTE rates declined significantly over time for both hip and knee arthroplasty.

• The decline in VTE for hip arthroplasty was statistically significant (p = 0.004)
• The decline in VTE for knee arthroplasty was statistically significant (p = 0.001)
• Trends were assessed using the Cox-Stuart trend test
• The declining trend was observed across the study period from 2007 to 2019

Introduction of NICE clinical guidelines in 2010, 2012, and 2018 did not significantly change VTE rates for knee or hip arthroplasty.

• The impact of clinical guidelines was assessed via logistic regression with a polynomial function
• NICE guidelines were introduced in 2010, 2012, and 2018 in the UK
• No statistically significant change in VTE rates was detected at the time of guideline introductions for knee or hip arthroplasty
• Seasonality was also accounted for in the regression model

Shoulder arthroplasty showed a unique 9% decrease in VTE rates following the introduction of the 2018 NICE guidelines.

• This was the only joint type to show a statistically notable response to a specific guideline introduction
• The 2018 guidelines appeared to have a differential effect on shoulder arthroplasty compared to knee and hip procedures
• The authors note this unique response warrants further investigation
• Overall shoulder arthroplasty VTE incidence was 1.0%, the lowest of the three joint types

Seasonal variation in VTE risk was observed and differed by joint type.

• Seasonality was incorporated into the logistic regression model alongside clinical guideline effects
• The pattern of seasonal variation was not uniform across knee, hip, and shoulder arthroplasty
• The authors call for future research to investigate the factors driving the seasonal association in VTE rates
• The mechanism underlying joint-specific seasonal differences in VTE risk was not determined in this study