Abstract
Over the last decade, the number of total knee arthroplasty (TKA) has risen over 125%. Numerous studies have established a relationship between obesity and complications in primary TKA. Interestingly, few studies have investigated medical and orthopaedic complications in obese patients undergoing revision TKA (RTKA). With the increasing number of RTKA and with the prevalence of obesity still on the rise, the study of obesity on complications following RTKA is indicated.
We retrospectively reviewed 180 RTKA performed by a single surgeon done between August 2008 and June 2017. All patients who underwent RTKA were included, but revisions done with simultaneous extensor mechanism reconstruction and/or distal femur replacement were excluded since these procedures are technically more demanding. 154 revisions met our inclusion/exclusion criteriaes and were included in the final analysis. 81 patients were included in the non-obese group (BMI .
The total number of orthopaedic complications in the obese group (46.6%) was significantly higher than in the non-obese group (27.2%) (OR=1.71) (p = 0.01). The number of infection was higher in the obese group (11%) than in the non-obese (6.2%) but this was not statistically significant (OR=1.77) (p = 0.28). Reoperation rate was also higher in the obese group (23.3%) than in the non-obese group (16%) but this did not reach statistical significance (OR=1.46) (p = 0.26) (Table 3). Medical complications were higher in the obese group (31.5% vs 19.8%) (OR=1.59) (p=0.09). According to the Dindo-Clavien classification, the obese group demonstrated a significantly higher rate of grade 3 or higher complications (p = 0.01).
Obesity significantly increases the occurrence of orthopaedic complications following RTKA. Obesity also seems to increase the number of medical complications following RTKA. The obese patient should be informed prior to revision TKA that there is an increased risk of complications when compared with the non-obese patient. Further research with higher power would seem advisable to confirm this trend.
For any figures or tables, please contact the authors directly.
