Associations between obesity and knee osteoarthritis or complications after total knee arthroplasty (TKA) are well established. The procedure can significantly improve knee function, favoring weight loss, despite the risk of surgical complications. The main objective of the present study was to assess change in body mass index (BMI) after TKA in patients with severe or morbid obesity (BMI≥35kg/m²). The secondary endpoint was the rate of surgical revision. The hypotheses were that there is no significant change in BMI after TKA and that there is a significant rate of revision.

This retrospective descriptive study was conducted for the period June 2009 to December 2019. Thirty-three patients (48 knees) were included: 27 women, 6 men; mean age, 66.5 years (range, 55–80). Preoperatively, 11 patients had BMI 35–39.9kg/m² and 22 BMI≥40 (including 11 with BMI≥45kg/m²). The preoperative axis was in varus for 35 patients (73%, including 54%≥10°) and in valgus for 13 (27% including 33%≥10°). Radiological and clinical evaluation was carried out at 3 months and 1 year postoperatively. At≥2 years, change in BMI and EQ5D functional score were established by telephone survey; >5% change in BMI was considered significant.

Mean follow-up was 6.9±2.3 years (range, 2.9–10.5). Twenty-five patients (38 knees) were included for analysis. Mean postoperative BMI was 41±5kg/m², with a mean decrease of 1.2±3.6kg/m². At the last follow-up, BMI had increased in 8 patients (32%), including 3 by >5% (12%), and decreased in 16 (64%), including 7 by >5% (28%). The higher the baseline BMI, the greater the decrease: for BMI [35–39.9], –0.81 (range, –6.8; +4.3); for BMI [40–44.5],–1 (range, –9; +5.22); and for BMI>45, –1.54 (range, –3.97; +1.3). EQ5D averaged 0.75 at last follow-up. The higher the preoperative BMI, the more satisfactory the postoperative EQ5D: EQ5D for BMI [35–39.9]=0.71 (range, 0.36; 1); for BMI [40–44.5]=0.75 (range, 0.45; 1); and for BMI>45=0.80 (range, 0.48; 1). Four early surgical site infections (10.5%) and 2 isolated changes of the tibial component for early loosening (5.2%) required surgical revision.

Patients with severe or morbid obesity had a low tendency to lose weight after TKA, but this does not appear to us to be clinically relevant: the functional results were good. Nevertheless, this series showed a significant rate of revision (15%). TKA was feasible in patients with BMI≥35kg/m², but requires appropriate patient information.

IV.