Abstract
Background:
Symptomatic genu recurvatum (GR) continues to be a challenge for clinicians. Patients may present with pain, weakness, and instability, which can lead to significant functional impairment. Currently, there are few reports discussing the treatment options and clinical outcomes of patients with symptomatic GR.
Purpose/Hypothesis:
The main objective of this study was to present the radiographic and functional outcomes of tuberosity-sparing anterior opening-wedge tibial osteotomy (TAOWTO) for symptomatic GR. It was hypothesized that TAOWTO can adequately correct the deformity and allow patients to have symptomatic relief after the procedure without compromising patellofemoral joint function.
Study Design:
Case series. Level of evidence, 4.
Methods:
This retrospective study was performed at a single, tertiary referral center. All patients underwent a TAOWTO performed by a single surgeon between January 2016 and January 2021. Pre- and postoperative radiographs were analyzed for posterior proximal tibial angle (PPTA), recurvatum angle, pertinent lower extremity alignment parameters, and patellar height. All patients were clinically evaluated pre- and postoperatively for the Knee injury and Osteoarthritis Outcome Score (KOOS).
Results:
After exclusions, 30 patients were included in the study. The mean age at the time of surgery was 22.6 ± 3.5 years, and the mean follow-up duration was 25.2 ± 7.8 (median, 35; range, 20.1-27.5) months. The causes of recurvatum were posttraumatic (53.3%), posterior cruciate ligament insufficiency (16.7%), anterior epiphysiodesis (6.7%), and soft tissue laxity (23.3%). The PPTA (in degrees) before and after the surgery was 94.2 ± 2.7 and 85.8 ± 1.8, respectively (P < .0001). There were no significant changes in hip-knee angle, medial proximal tibial angle (in degrees) and Caton-Deschamps index. In addition, the recurvatum angle was significantly reduced (12.0 ± 2.9 vs 2.9 ± 1.7; P < .0001). All domains of KOOS (Pain, Symptoms, Activities of Daily Living, Quality of Life, and Sport and Recreation) were significantly improved after the surgery (P < .0001).
Conclusion:
TAOWTO was an effective surgical procedure for patients with symptomatic GR. It reliably corrected the PPTA and recurvatum angle, while maintaining the native coronal alignment and patellar height. Patients also had significant symptomatic relief and functional improvement in both daily and sports activities.
Keywords
Symptomatic genu recurvatum (GR) continues to be a challenging problem for clinicians. 6 The prevalence of GR has been reported to be around 0.8% in primary school–aged children, and <1% of these patients are symptomatic. 10 It is characterized by hyperextension of the knee, and patients can present with symptoms such as pain, weakness, and instability. 6 Symptomatic GR can be caused by congenital conditions or be secondary to trauma, anterior epiphyseal arrest, and medical comorbidities such as stroke.6,29,35 Symptomatic GR occurs most commonly in adolescent girls and obese children, which can cause significant morbidities. 19 GR can be classified as either (1) soft tissue laxity, (2) pure osseous deformity, or (3) a mixed-type deformity due to both soft tissue and osseous abnormalities. 7 Based on the main pathological origins, patients are treated with soft tissue management (eg, capsular tensioning), osseous correction with an osteotomy, or a combination of both. 6 Soft tissue procedures are aimed at tensioning the elongated posterior capsuloligamentous structures,30,31 while osseous correction can be achieved through various opening-wedge18,27,35 and closing-wedge tibial osteotomies.3,6 To achieve successful correction, restoration of the tibial anatomic axis, preservation of the patellar height, and adequate bone fixation are all essential.18,27 Currently, there is no consensus for the ideal osseous procedure. Several techniques including proximal tibial closing-wedge and opening-wedge osteotomies have been described to correct the hyperextension deformity.3,7,18,31,32 Regardless of the type of osteotomy, these slope-increasing osteotomies are meant to correct the hyperextension, which reduces stress on the posterolateral capsuloligamentous structures, and thus improve pain and stability of the knee. 33 In this study, patients who underwent a tuberosity-sparing anterior opening-wedge tibial osteotomy (TAOWTO) for symptomatic GR were studied. The main objective of this study was to present the radiographic and functional outcomes for patients who underwent this procedure. It was hypothesized that TAOWTO can adequately correct the sagittal deformity and allow patients to have symptomatic relief while preserving the coronal plane alignment and patellar height.
Methods
This was a retrospective study of 30 patients who underwent TAOWTO for symptomatic GR at the Institute for Locomotion's Center for Arthritis Surgery, Sainte-Marguerite Hospital, Marseille, Department of Orthopaedic Surgery between January 2016 and January 2021. The study included patients operated on by a single, fellowship-trained orthopaedic surgeon (M.O.) and the inclusion criteria were the following:
Patients with symptoms such as pain, walking disturbance, instability sensation, and inability to engage in sports and other activities
Hyperextension >0° on heel-height test
Posterior proximal tibial angle (PPTA) > 90° (which indicates a reverse slope) and/or recurvatum >10°
Failure of nonoperative treatment for 6 months, including a formal course of physical therapy, pain medications such as acetaminophen and nonsteroidal anti-inflammatories, and activity modification
Follow-up of ≥20 months
The exclusion criteria were as follows:
Severe coronal plane malalignment (>5° varus or valgus)
Advanced osteoarthritic changes (Ahlbäck grade >3)
Combined multiligamentous injury
Follow-up of <20 months
Descending osteotomy (proximal to distal cut)
Osteotomy involving the tibial tuberosity (TT) (non–tuberosity sparing osteotomy)
The patient flow diagram is shown in Figure 1.

Patient inclusion diagram. TAOWTO, tuberosity-sparing anterior opening-wedge tibial osteotomy; DLO, double-level osteotomy.
The causes of recurvatum were posttraumatic (53.3%; n = 16); posterior cruciate ligament (PCL) insufficiency, defined as a patient with >10-mm posterior translation upon posterior drawer examination without evidence of traumatic history (16.7%; n = 5); anterior epiphysiodesis (6.7%; n = 2); and soft tissue laxity (23.3%; n = 7) (Table 1).
Causes of Genu Recurvatum a
Data are presented as n (%). PCL, posterior cruciate ligament.
For radiographic evaluation, each patient had long-leg radiographs of bilateral lower extremities, as well as lateral lower leg radiographs of the affected limb, with the knee in full extension and with true posterior and inferior lateral views. 17 Long-leg radiographs were obtained because they could more accurately predict the PPTA than short radiographs.13,28 The PPTA was measured on lateral radiographs of the lower leg using the best-fit circle method, which was defined as the angle between the lateral posterior tibial plateau and the full tibial anatomic axis indicated by 2 to 3 circles that best fit within the anterior and posterior tibial cortices in the proximal, middle, and distal tibia 21 (Figure 2). The normal value of PPTA was 77° to 84° (mean of 81°), with a value >84° indicating decreased tibial slope and <77° indicating increased slope. 1

Radiographs before and after correction. (A) Preoperative lateral view of the left tibia, showing estimation of the posterior proximal tibial angle based on the best-fit circle method using the full anatomic axis of the tibia. (B) Postoperative lateral view of the left tibia, showing anterior opening-wedge tibial osteotomy fixed with the plate (Newclip Technics) and staples (Dupuy Synthes) as well as estimation of the posterior proximal tibial angle based on the best-fit circle method.
Clinical and Radiographic Evaluation
All patients were clinically evaluated preoperatively (≤3 months before surgery) and postoperatively (at final follow-up, ≥20 months after surgery) with the Knee injury and Osteoarthritis Outcome Score (KOOS). All patients had preoperative radiographs for planning and postoperative radiographs to evaluate the osteotomy union. All radiographic measurements were performed on 2 separate occasions at 3-week intervals using the same images by 2 senior surgeons (M.O., W.A.D.) who did not participate in the surgery. To measure patellar position, the Caton-Deschamps (C-D)4,34 index was used. Classification was performed based on the C-D index: baja < 0.9, 0.9 <[neutral]< 1.2, alta > 1.2. The mean values were used in statistical analysis.
Surgical Planning and Technique
The planning was completed with a digital software planning system (PeekMed). Briefly, the surgical planning was performed with the lateral full-length tibial radiograph using the PeekMed software. The best-fit circle method was utilized to measure the PPTA (Figure 2A). The anticipated correction angle and width (in mm) of the osteotomy wedge was calculated. The targeted slope after surgical correction is recommended to be between 84° and 86°.
Starting with an anterior longitudinal incision centered over the TT, the proximal tibia was exposed medially through subperiosteal stripping of the deep medial collateral ligament and laterally by incising the tibialis anterior fascia and detaching the most proximal part of the muscle off the Gerdy tubercle (Figure 3). 23

The surgical approach in a left knee. (A) The knee position before surgery. The patient had a posterior sag sign, which raised concerns for posterior cruciate ligament insufficiency. (B) A 10-cm incision was planned starting just proximal to the tibial tuberosity (black star indicates patellar tendon). (C) The patellar tendon (black star) was dissected from the medial and lateral capsule. Then, subperiosteal flaps were raised to expose the proximal tibia medially and laterally.
Next, biplanar fashioned tuberosity-sparing osteotomy was performed in all cases, starting with a vertically ascending TT cut in the sagittal plane posterior to the TT. 26 Under fluoroscopic guidance, 2 parallel pins were inserted on either side of the patellar tendon slightly inferior to the Gerdy tubercle to avoid any disturbance to the iliotibial tract, directed toward the posterior bony hinge (tibial insertion site of the PCL) (Figure 4).

Anterior opening-wedge osteotomy cuts. (A) Fluoroscopic image, showing 2 guide wires inserted to template high tibial osteotomy under fluoroscopy and directed from distal to proximal toward the hinge point. The posterior hinge was distal to the insertion of the posterior cruciate ligament on the tibia. (B) The osteotomy cut was performed under the guide of the pins, the black star indicated the hinge protective wire. (C) A fluoroscopic image shows the cuts guided by the pins. The hinge protective wire (black star) was inserted from the distal tibial tuberosity to the hinge point.
After executing the cuts guided by the pins, the osteotomy site was opened anteriorly. This was achieved by using chisels with variable thickness (5 mm–20 mm) to achieve a satisfactory opening. An appropriately sized metal block (depending on our desired opening size) was then inserted in the osteotomy site (Figure 5A).

Intraoperative fluoroscopic images of osteotomy and plate fixation. (A) The opening of the osteotomy site with a metal block. (B) Insertion of an allograft bone block into the gap. (C) The locked plate fixation was applied on the medial side and staple fixation on the lateral side.
A femoral head allograft was then used to prepare the desired wedged allograft in the back table. This wedge allograft, corresponding to the preoperatively planned width (in mm) was filled into the gap manually and with the use of the bone tamp. Typically, this is performed with the knee in 60° to 90° of flexion. The osteotomy site was fixed by a locking plate system (Newclip Technics) on the medial side and staples (Dupuy Synthes) on the lateral side (Figure 5C). Adequate PPTA correction was confirmed by intraoperative fluoroscopy. Figure 6 demonstrates a patient from preoperatively to after correction and plate removal.

(A, B) Anteroposterior and lateral radiographs of a single patient preoperatively. (C, D) Six months postoperatively; and (E, F) 18 months postoperatively after hardware removal.
Additional Intraoperative Procedures
On the basis of intraoperative arthroscopic findings, 2 patients (6.7%) underwent PCL reconstruction. Patient-specific instruments16,26 were used for 8 cases (26.7%), while the other 22 cases (73.3%) were performed without one. In terms of fixation methods, 15 patients (50.0%) had a locking plate fixation on the medial side with a staple fixation on the lateral side, while 11 (36.7%) and 4 (13.3%) patients were fixed with double plates and staples, respectively (Table 2).
Intraoperative Procedures a
Data are presented as n (%). PCLR, posterior cruciate ligament reconstruction.
Postoperative Rehabilitation
Early rehabilitation focused on quadriceps activation. Partial weightbearing with crutches was allowed for 3 weeks, followed by a gradual increase in weightbearing based on radiographic evaluation of osteotomy healing. An anterior cruciate ligament brace was used for 3 weeks. Cycling was allowed at 6 weeks, running at 4 months, and return to high-impact sports at 6 months.
Statistical Analysis
Descriptive statistics are presented as mean ± SD, and categorical data are presented as percentages. Statistical analysis was performed with SPSS 12.0 (IBM Corporation). The accuracy of measurements was evaluated using the intraclass correlation coefficient for intraobserver and interobserver reliability. Two independent observers (T.F.A.-C., J.S.-A.) reviewed every preoperative and postoperative radiograph set twice, blinded and with a 3-week measurement interval: all intraclass and interclass correlation coefficients were >0.8. A normality plot was used to confirm normal distribution of the data. For comparison between preoperative and postoperative values, paired t test and Fisher exact test were used, and statistical significance was assumed at P values of <.05.
Results
Patient Demographics
The patient demographics are presented in Table 3. A total of 30 patients with a mean ± SD age of 22.6 ± 3.5 years and follow-up duration of 25.2 ± 7.8 months (median, 35; range, 20.1-27.5) were enrolled (Table 3).
Patient Demographics a
Data are presented as mean ± SD [range] or n (%).
There were 10 (33.3%) female and 20 (66.7%) male patients. The right knee was affected in 16 (53.3%) patients, while 14 (46.7%) patients were affected on the left.
Radiographic Assessment and Recurvatum Before and After Surgery
The radiographic measurements are presented in Table 4. Hip-knee angle (180° = neutral; value >180° considered valgus and <180° considered varus) before and after the surgery was 179.0 ± 2.8 (179.9-180.0) and 179.3 ± 1.5 (178.7-179.9), respectively (P = .53). PPTA (measured in degrees) before and after the surgery was significantly reduced, at 94.2 ± 2.7 (93.1-95.2) and 85.8 ± 1.8 (85.1-86.4), respectively (P < .0001). Recurvatum (in degrees) of the operated side before and after surgery was significantly reduced, at 12.0 ± 2.9 (10.9-13.1) and 2.9 ± 1.7 (2.2-3.6), respectively (P < .0001).
Radiographic Assessment and Recurvatum Before and After Surgery a
Data are presented as mean ± SD [range]. ICC, intraclass correlation coefficient.
Patellar Position Before and After Surgery
The patellar height (C-D index) before and after surgery is presented in Table 5.
Patellar Position Before and After Surgery a
Data are presented as mean ± SD [range] or n (%). Classification based on Caton-Deschamps index: baja < 0.9, 0.9 <[neutral] < 1.2, 1.2 < alta.
The C-D index before and after the surgery was 1.1 ± 0.1 (0.9-1.4) and 1.0 ± 0.2 (0.7-1.3) (P = .11). The classification of patellar position showed that the preoperative position (21 cases of neutral, 1 case of alta, and 8 cases of baja) moved to postoperative position (16 cases of neutral, 1 case of alta, and 13 cases of baja) without significant difference (P = .22).
Knee injury and Osteoarthritis Outcome Score (KOOS) Before and After Surgery
The KOOS score of this cohort is presented in Table 6.
Knee injury and Osteoarthritis Outcome Score Before and After Surgery a
Data are presented as mean ± SD [range] and scores range from 1 to 100. ADL, Activities of Daily Living; Sport/Rec, Sport and Recreation.
Statistically significant.
KOOS Pain scores (0-100) before and after surgery were 50.9 ± 16.5 (44.7-57.0) and 93.3 ± 1.4 (92.8-93.8). Symptom scores (0-100) before and after surgery were 68.5 ± 20.7 (60.7-76.2) and 92.3 ± 6.8 (89.7-94.9). Activities of Daily Living (ADL) scores (0-100) before and after the surgery were 42.7 ± 11.4 (38.4-46.9) and 66.6 ± 10.6 (62.6-70.6). Quality of Life scores (0-100) before and after surgery were 56.6 ± 21.0 (48.8-64.5) and 96.9 ± 0.7 (96.6-97.1). Sport and Recreation scores (0-100) before and after surgery were 13.6 ± 12.4 (9.0-18.3) and 81.4 ± 9.6 (77.7-85.0). All subscale scores of KOOS were significantly improved after the surgery (P < .0001).
Complications
Of the 30 patients, 17 patients (56.7%) had anterior knee pain related to symptomatic hardware and required removal of hardware. This procedure was completed at a mean of 19.2 months (range, 16-24 months) postoperatively. One patient (3.3%) developed an operative-site infection at 1 week after the surgery. This patient was managed with surgical debridement of the wound and antibiotic treatment for 2 weeks. No other complications were noted in this cohort.
Discussion
The most significant finding of the current study was that patients treated with TAOWTO had significant improvement in radiographic and functional outcomes after the surgery at a mean follow-up of 25.2 months. Specifically, there was satisfactory correction of the PPTA and reduction of recurvatum without the modification of preoperative coronal alignment and patellar height. In addition, there was significant improvement in all sections of KOOS (Pain, Symptoms, ADL, Quality of Life, and Sport and Recreation activity).
The modern proximal tibial osteotomy technique was first described by Lecuire et al, 20 which performed TAOWTO detaching TT and fixing the opening using the staples. In recent years, modifications in this procedure allowed tuberosity-sparing osteotomies (eg, ascending TT osteotomy).18,27,32 In this series, 30 patients were treated successfully by biplanar fashioned tuberosity-sparing osteotomy with ascending TT cut.
With this current algorithm, it was possible to maintain a normal hip-knee angle and medial proximal tibial angle indicating minimal anatomic translation after TAOWTO. In a biomechanical analysis of cadaveric specimens, the authors noted a slight increase in external rotation of the tibia with increasing size of the opening wedge and fixation device (eg, plate), particularly when the osteotomy was >10 mm. 25 However, there was no increased strain in the medial and lateral collateral ligaments, confirming that a true sagittal plane corrective osteotomy can be performed without affecting the coronal alignment, which is in agreement with the results of this study. In the current study, the mean of 8.4° correction in the PPTA, and the decrease of 9.1° of the recurvatum angle were achieved, which are in line with other reported series.18,24,27,33
Patellar malposition (eg, patella baja) can lead to symptomatic patellofemoral incongruence and early osteoarthritis. 12 In addition, this can be can be problematic when performing surgical procedures that involve alignment and joint line correction procedures such as total knee replacement. 36 Currently, some studies have advocated for a more distalized osteotomy below the TT to reduce risk of patella baja in open-wedge high tibial osteotomy (HTO).2,9,14 Although better control of the patellar height may be achieved when performing a descending or detaching TT cut during slope-changing osteotomy, 22 Dejour et al 8 found no significant change in patellar height based on C-D index after anterior closing-wedge tibial osteotomy with ascending biplanar osteotomy. In current literature, there are very few reports that have discussed TAOWTO and the effects on patellar height. In this study, the mean patellar height measured with the C-D index was maintained postoperatively, and there were no significant differences (P > .05) before and after the surgery (Table 5).
Several fixation devices have been proposed to stabilize the osteotomy site. This can be completed with staples, plates and screws, external frames (eg, Ilizarov), or a combination of the above. 6 In a recent systematic review, Dean et al 6 identified 5 studies that performed osseous procedures for slope correction. Three studies performed an isolated proximal tibial osteotomy,18,27,35 while 2 performed progressive distraction using the Ilizarov technique after the initial proximal tibial osteotomy.5,24 All studies showed an increase in PPTA, while 4 reported a decrease in recurvatum. Using the single anteromedial plate system and staple in the current study allowed for rigid fixation of the osteotomy site and permitted early weightbearing after the surgery.
KOOS was utilized for functional assessment in this study. Previous studies have validated the KOOS score as an assessment tool for patients after HTO, identifying the minimal clinically important difference as 15.4 for KOOS Pain, 15.1 for KOOS Symptoms, 17 for KOOS ADL, 11.2 for KOOS Sport and Recreation, and 16.5 for KOOS Quality of Life after HTO.11,15 In the current study, all patients saw improvement in all 5 parameters after surgery. Of note, sports activity saw the greatest improvement, with a mean improvement of 67.8 at a mean follow-up of 25.2 months. This is most likely because by correcting the PPTA to a more anatomic position, the surrounding capsuloligamentous structures (eg, cruciate ligaments, menisci, and joint capsule) are tensioned, which restores the native knee biomechanics. 8 By restoring the anatomy, patients can theoretically have better functional performance, which was evident in this case series. Although patients in this cohort saw significant improvement in knee pain, postoperative anterior knee pain was noted in a subgroup of patients (n = 17). This was most prominent in the patients who practiced frequent kneeling activities and was reported by other studies as well. 18 This was most likely secondary to hardware irritation, and the patients had symptomatic relief after removal of hardware. Nonetheless, patients who have engage in kneeling activities should be informed of this risk before surgery. Surgeons should be aware of this complication.
As symptomatic GR can be a multifactorial disease (eg, bony malalignment and soft tissue laxity, etc), a single procedure may not address all pathologies; therefore, the decision on whether additional procedures need to be performed is dependent on patient symptoms, physical examination during intraoperative examination under anesthesia, and arthroscopic evaluation. In 2 patients, we performed both a PCL reconstruction and an osteotomy correction. The decision was made based on intraoperative examination after osteotomy correction, as there was residual posterior laxity with >10 mm of posterior translation with a posterior drawer exam. Additional long-term studies will be needed to elucidate whether a single procedure can lead to satisfactory results in 5 to 10 years after the surgery.
Although this was not the first case series evaluating a TAOWTO for patients with symptomatic recurvatum, this study adds several important findings to the current literature. (1) This study had a high patient number (N = 30) in a relatively rare cohort; (2) a thorough evaluation of pre- and postoperative evaluation of both coronal and sagittal alignment was performed to confirm that this procedure only corrected sagittal malignment with minimal effects on patellar height and coronal alignment; and (3) the fixation method of plate, screws, and staples along with a femoral head allograft was utilized as the primary bone graft, which provided satisfactory union rates and patient outcome.
Limitations
This study has several limitations. First, given that there is no current gold standard for the treatment of symptomatic GR, we were unable to have a comparison or control group. Finally, the follow-up time was a mean of 25.2 months, which can only provide insights into the short-term outcome after the procedure. Long-term changes such as the development of arthritis or recurrence of deformity and/or symptoms will require additional studies to elucidate.
Conclusion
TAOWTO was an effective surgical procedure for symptomatic GR. It reliably corrected the PPTA and recurvatum angle, while preserving a normal coronal alignment and patellar height. Patients had excellent symptomatic relief and significant improvement in daily and sports-related activities.
Footnotes
Final revision submitted March 5, 2025; accepted April 7, 2025.
One or more of the authors has declared the following potential conflict of interest or source of funding: W.A.D. is a consultant for New Clip Technics and Arthrex, has received nonconsulting fees from Arthrex, consulting fees from Arthrex, and education payments from Supreme Orthopedic Systems. M.O. is a paid consultant and receives royalties from Stryker and New Clip Technics. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
The ethics committee confirmed waiver MR004 to IRB regulation before initiation of the study (and local PADS was obtained at Institut du mouvement et de l'appareil locomoteur (No. 2022-245).
