Gait Function at 20 Years or More After Rotationplasty Shows Pseudo Knee Motion, Decreased Walking Speed, and...

Rotationplasty is a surgical procedure primarily used for malignant bone tumors around the knee when conventional limb salvage is not feasible because of tumor extent or patient preference. The procedure repurposes the ankle to function as a pseudo knee. While early outcomes are generally good, long-term gait performance in adulthood has not been well described.

This study addresses the following questions about patient outcomes 20 years or more after rotationplasty: (1) What are the walking speed and energy cost as well as the spatiotemporal and gait parameters compared with a control group without lower limb diseases? (2) How are walking speed and energy cost related to age, follow-up duration, and gait parameters? (3) Does thigh-shank length discrepancy correlate with the energy cost of walking and gait parameters?

Between 1980 and 2002, a total of 70 patients underwent rotationplasty (all Winkelmann Type A1) at two centers in Amsterdam. Of these, 37% (26) died, 4% (3) underwent amputation, and 9% (6) could not be traced. Of the remaining 35 patients, 6% (2) lived abroad, 9% (3) declined participation, and 3% (1) had a nonfitting prosthesis, leaving 83% (29 of 35) of patients available for evaluation at a median (IQR) follow-up time of 33 years (29 to 35). Rotationplasty was performed for osteosarcoma in 76% (22 of 29) of patients, Ewing sarcoma in 7% (2), other malignancies in 10% (3), hemangioma in 3% (1), and femoral deficiency in 3% (1), all by a single orthopaedic surgeon. This cohort included 52% (15 of 29) male and 48% (14 of 29) female patients. The control group, obtained from the institutional normative database, included 38 participants for analyses of the energy cost of walking and 27 participants for analyses of gait without lower limb diseases or systemic conditions affecting gait or energy cost, and was comparable with the rotationplasty group in age, sex, and BMI. Function was evaluated by walking speed, energy cost of walking, and gait parameters for spatiotemporal kinematics and kinetics and compared with the same parameters for the control group. Furthermore, outcomes were compared between patients with and without measurable thigh-shank length discrepancy. Statistical analyses included independent t-tests, statistical parametric mapping, and Pearson correlations.

Compared with the control group, patients after rotationplasty walked slower (mean ± SD 1.2 ± 0.2 versus 1.4 ± 0.1 m/s, mean difference -0.2 [95% confidence interval (CI) -0.3 to -0.1]; p < 0.001) with a higher energy cost (4.4 ± 0.7 versus 3.5 ± 0.4 J/kg/m, mean difference 0.9 [95% CI 0.6 to 1.2]; p < 0.001). Cadence was lower (100 ± 7 versus 108 ± 7 steps per minute, mean difference -8 [95% CI -11 to -4]; p < 0.001) and stride length shorter (1.4 ± 0.2 versus 1.5 ± 0.1 m, mean difference -0.1 [95% CI -0.2 to -0.1]; p = 0.01). The double-support phase from the contralateral leg to the rotationplasty leg was longer (at 13.1% ± 1.9% versus 11.8% ± 1.2% of the gait cycle, mean difference 1.4% [95% CI 0.5% to 2.3%]; p = 0.003). Statistically significant differences were observed at specific phases of the gait cycle: Patients with rotationplasty lacked knee flexion during loading response and midstance, whereas the intact leg showed increased ankle dorsiflexion and greater knee and hip flexion during stance, accompanied by increased ground-reaction force in loading response and push-off. Walking speed increased very strongly as cadence increased (r = 0.7 [95% CI 0.6 to 0.8]; p < 0.001) and increased moderately as stride length increased (r = 0.3 [95% CI 0.1 to 0.5]; p < 0.01). As walking speed increased, walking energy cost decreased moderately (r = -0.4 [95% CI -0.6 to -0.2]; p < 0.001). With the numbers available, no associations were found between thigh-shank length discrepancy and walking speed, energy cost, joint angles, or joint moments.

More than two decades after rotationplasty, patients demonstrated lower walking speed and higher energy cost of walking than the control group, which is expected given the magnitude of the procedure. Compared with previous reports, the energy cost was not higher at longer follow-up and approached normal values, exceeding those typically observed after transfemoral amputation. Although the absence of knee flexion during stance was associated with compensatory strategies, mainly in the contralateral limb, overall gait performance remained functional, with preserved pseudo knee mechanics. No differences were observed between patients with and without thigh-shank length discrepancy, providing no clear guidance for surgical length-correction strategies. Despite biomechanical deviations from normal gait, long-term function after rotationplasty appears durable and energy efficient, which can help clinicians to counsel patients on realistic expectations and functional potential. Further studies comparing rotationplasty with transfemoral amputation and limb-salvage surgery are warranted to better inform shared decision-making.

Level III, therapeutic study.