Translate this page into:
Total knee replacement in severe genu recurvatum. High prosthetic survival rate without deformity recurrence, using rotating hinge prosthesis
⁎Corresponding author: Leonel Perez Alamino. dr.lperezalamino@gmail.com
-
Received: ,
Accepted: ,
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Abstract
Genu recurvatum can originate from neurological and non-neurological causes. Its presence associated with knee osteoarthritis represents 1 % of the causes of total knee replacement (TKR). The aim of this retrospective study was to show the functional outcomes, recurrence rate and prosthetic survival in patients with genu recurvatum treated with articulated rotating hinge prostheses.
between January of 2008 and December of 2021, we performed 3224 primary TKR, of which 34 (1 %) had recurvatum. This was defined as hyperextension ≥10°. Inclusion crirteria were patients with primary TKR and recurvatum that were treated with a rotating hinge prosthesis (Endo Model, Walderman Link, Ham, Germany), and completed a minimum follow-up of 24 months. Final series consisted in 23 patients with 25 TKR with 15 (65.22 %) females and a mean age of 66.8 ± 14.2 years old. The median follow-up was 5.0 (range 2.11–11.5). Clinical analysis was performed with the use of Knee Society Score (KSS) and Western Ontario & McMaster Universities Osteoarthritis Index (WOMAC). Prosthetic survival rate was calculated with a Kaplan-Meier curve considering as the endpoint of the analysis revision due to any reason. Recurrence was defined as the presence of hyperextension ≥ 5° after surgery.
of the 23 patients, 9 (39.13 %) had history of neurological disorders. Median recurvatum observed was 20.12° (range 13.0° - 30.2°). A median of 19.6° (range 10.1° - 55.4°) valgus misalignment was observed in 19 patients and 10.7° (range 2.0–23.5°) of varus was observed in 6 (24 %) patients. Clinical and functional KSS improved significantly after 12 months (from 34.2 ± 9.8 and 40.1 ± 12.6 to 82.1 ± 7.5 and 89.2 ± 6.8 respectively; p < 0.01) and at the end of the study (84.3 ± 7.5 and 90.0 ± 11.3; p < 0.01). Final mean WOMAC value was 24.1 ± 9.6. One (4.0 %) complication was registered after surgery (superficial infection). Four (16.0 %) patients had demarcation. Prosthetic survival rate was 100 % at the end of the study.
The rotating hinge prosthesis restores joint function in patients with severe knee osteoarthritis with and without neurological disorders, without recurrence of misalignment and with excellent survival rates.
Keywords
Rotaing hinge prosthesis
Knee osteoarthritis
Joint replacement
Total knee arthroplasty
1 Introduction
Genu recurvatum can be observed in patients with and without neuromuscular disorders.1. Among the first ones, poliomyelitis stands out, although it can also occur in patients with diverse neurological sequels where the quadriceps strength is impaired.2. To walk, these patients need to hyperextend the knee and force the posterior capsule and ligaments until recurvatum begins to appear.1–3.
Among the non-neurological causes, it has been described as most frequently associated with misalignment in the coronal plane, fundamentally fixed valgus. In these cases, the iliotibial band contracture during full extension lies in front of the knee rotation axis, forcing hyperextension. Also, it has been reported in patients with ligament laxity, as in Rheumatoid Arthritis.1–4.
The presence of severe genu recurvatum (>10°) associated with gonarthrosis represents approximately 1 % of patients undergoing total knee replacement (TKR).4,5. At the time of arthroplasty, different surgical strategies have been proposed for its correction. Among the most reported are the under resection of the bone ends, the placement of the femoral component slightly flexed, the increase of the tibial slope, the collateral ligament transfer, and the plication of the posterior capsule.1,6,7.
Regarding the level of prosthetic constraint used in these cases, different authors have reported significant functional improvements in the short and medium follow-up using posterior cruciate-retaining and posterior stabilized implants '2–4]. The recurvatum recurrence rate with these implants has been reported to be up to 30 %.8. On the other hand, the use of rotating hinged prostheses in genu recurvatum has also been described, mainly in TKR revisions, showing excellent functional outcomes, without recurrence of the deformity, but with inconsistent medium and long-term survival rates.9,10.
To our knowledge, there is scarce information regarding the outcomes of primary TKR with rotating hinges in patients with severe recurvatum. Therefore, this study aimed to show the functional outcomes, recurvatum recurrence rate, and prosthetic survival in patients with genu recurvatum >10°, with and without neuromuscular disorders, treated with rotating hinged prostheses.
2 Methods
According to the retrospective analysis of our department database between January 2008 and December 2020, we consecutively performed 3224 primary TKR. Of these, 34 (1 %) cases presented a preoperative genu recurvatum.
We defined genu recurvatum as a hyperextension deformity greater than 10° [Mesnard]. We indicated a rotating hinge in patients who presented, in addition to the recurvatum above with knee osteoarthritis, one or more of the following conditions: significant coronal misalignment, instability in the coronal plane due to either ligament insufficiency or notorious ligament laxity, or bony defects.
In the present analysis, we included primary TKRs performed in patients with genu recurvatum >10° treated with a rotating hinged prosthesis (EndoModel, Walderman Link, Ham, Germany), with a minimum clinical radiological follow-up of 24 months. We excluded patients with a history of previous surgery in the knee to be operated on and/or a history of knee infection. Out of 34 TKRs initially identified, nine were excluded: two for having a history of previous surgeries in or around the knee (one tibial plateau fracture, one osteotomy), four for having used a prosthetic model other than the one mentioned, two for not fulfill the minimum follow-up and one for not having the records available for the analysis.
The study series comprised 25 TKRs in 23 patients (two staged bilateral TKRs). Fifteen (65.22 %) patients were female and 8 (34.78 %) males, with a mean age of 66.8 ± 14.2 years and a median follow-up of 6 (range 2–11) years. Of the 23 patients, nine (39.13 %) had a neurological history with motor sequelae in the limb to be operated on, of which none presented ankle flexion contracture. Table 1 describes the history and preoperative muscle values of the quadriceps and ankle strength, according to The Medical Research Council clinical scale.11.
| Neuromuscular history (frecuency) | Motor values (quadriceps – ankle flexion – ankle extension) |
| Stroke (3) | (M3 – M4 – M4 x 3) |
| Lumbar spine surgery (3) | (M3-M5-M5) (M4-M5-MM5 x 2) |
| Cervical spine surgery (2) | (M3-M5-M5) (M4-M5-M5) |
| Non-progressive brain encephalopaty (1) | M4-M5-M5 |
We assessed the femorotibial alignment with standing anteroposterior and lateral (in extension) radiographs. We also used stress radiographs to assess ligament competency based on the deformity present (varus, valgus, and hyperextension). The measurement of the misalignments was performed with Synapse software (FujiFilm). The median recurvatum was 20.19° (range 13°-30°); additionally, 19 (76 %) knees presented valgus alignment with a median of 19.6° (range 5-55°) and 6 (24 %) in varus with a median of 10.8° (range 3-23°). The description of the series, discriminating patients with and without neurological history, is summarized in Table 2.
| overall n = 25 | without Neuro n = 16 | with Neuro n = 9 | |
| Female gender fr(%) | 15 (60.0 %) | 9 (56.2 %) | 6 (66.7 %) |
| Age mean-SD | 66 ± 14.5 | 71.8 ± 10.23 | 57.7 ± 16.1 |
| Charlson CI fr(%) | 2.8 ± 1.5 | 3.1 ± 1.1 | 2.3 ± 2.0 |
| ASA I-II fr(%) | 15 | 7 | 8 |
| DM fr(%) | 4 | 3 | 1 |
| BMI >30 fr(%) | 2 | 1 | 1 |
| Motor mucle values∗ fr(%) | |||
| Quadriceps | |||
| M5 | 16 | 16 | – |
| M4 | 4 | – | 4 |
| M3 | 5 | – | 5 |
| Ankle FE | |||
| M5 | 22 | 16 | 6 |
| M4 | 3 | – | 3 |
| Knee ROM | |||
| Extension gap >5° fr(%) | 12 (48.0 %) | 3 (18.75 %) | 9 (100 %) |
| Flexion mean-SD | 101 ± 17.1° | 96 ± 15.0° | 111.6 ± 16.9° |
| KSS mean-SD | |||
| Cílinical | 32.6 ± 12.3 | 38.2 ± 12.2 | 22.2 ± 7.5 |
| Functional | 35.0 ± 12.8 | 38.7 ± 13.9 | 28.3 ± 7.7 |
| Alignment fr(%) | |||
| Recurvatum | 20.19° (13–30) | 21.2° (14–30) | 17.7° (11–30) |
| Valgus | 19.5° (5-55) | 22.7° (5-55) | 11.4° (7-15) |
| Varus | 10.8° (3-23) | 3° y 23° | 9.7° (5-15) |
| Overall n = 25 | With nueromuscular n = 16 | without neuromuscular n = 9 | |
| ROM | |||
| Extensión gap fr(%) | 9 (36.0 %) | 0 | 8 (88.89 %) |
| Flexión mean-SD | 109.6 ± 10.1° | 110 ± 11.0° | 109.1 ± 9.0° |
| KSS 1 yearmean-SD | |||
| Clínical | 79.6 ± 10.9 | 84.1 ± 7.6 | 67.7 ± 7.9 |
| Functional | 78.4 ± 9.4 | 83.2 ± 7.5 | 68.6 ± 4.1 |
| End of studymean-SD | |||
| Clínical | 79.4 ± 9.6 | 85.3 ± 8.2 | 70.7 ± 6.1 |
| Functional | 78.6 ± 8.7 | 84.2 ± 6.0 | 68.6 ± 3.7 |
| WOMACmean-SD | 24.1 ± 9.6 | 27.4 ± 9.8 | 18.2 ± 5.9 |
| Complicationsfr(%) | |||
| Superficilal infection | 1 (4 %) | 1 (6.25 %) | – |
| Radilucent linesfr(%) | 4 (16 %) | 3 (18.75 %) | 1(11.11 %) |
| Follow-upmedian-range | 6 (2–11) | 6.5 (2–11) | 5.5 (3–10) |
2.1 Surgical technique
All patients were operated on in a laminar flow operating room with spinal hypotensive anesthesia (see Table 3). We used the anterior approach with internal patellar arthrotomy. The soft tissue release and balancing were performed before the bone cuts. In severe valgus, before the arthrotomy, we performed the iliotibial band release at the level of the superior pole of the patella, and then, if necessary, we also performed an osteotomy of the lateral epicondyle. In the varus, we release the proximal and internal area of the tibia. After the bone cuts, we proceeded to release the collateral ligaments on demand until the trial components were reduced, and intraoperative full extension was achieved. The patella was replaced according to the attending surgeon's criteria.
Prosthetic fixation was cemented with cemented femoral and tibial stems using a cement restrictor. The postoperative rehabilitation consisted of sitting and isometric exercises on the first postoperative day, walking with canes or walker, and progressive flexion-extension exercises on the second postoperative day. A single cane was allowed after the second or third week (depending on the case), and flexion-extension and strengthening exercises were prescribed. Clinical radiological postoperative visits were performed at 3 and 6 weeks, at 3 and 6 months, followed by annual controls.
2.2 Variables assessment
We analyzed the (pre-and postoperative) range of motion (ROM) discriminating between active and passive extension using a goniometer.12. For clinical-functional assessment, we used the Knee Society Clinical Rating System scores preoperatively, at one year and at the time of the last follow-up.13. To assess the patient's perception of the impact of arthroplasty on their quality of life, we used the Western Ontario and McMaster Universities Arthritis Index score (WOMAC 0 best score and 96 worst), recorded at the last follow-up.14.
To determine the presence of demarcation around the implants (radiolucent lines <1 mm) and/or prosthetic loosening, we analyzed follow-up radiographs (anteroposterior and lateral loaded and Merchant projection). To locate these lines, we used the radiographic system of the Knee Society Total Knee Arthroplasty Roentgenographic Evaluation and Scoring System.15. We defined prosthetic loosening as the presence of radiolucent lines at the prosthesis cement or bone cement interface > 2 mm, or showing progression on successive follow-up radiographs, together with the presence or not of changes in component position.16. Two authors performed these measurements independently, and differences were resolved by consensus. We also recorded the recurrence rate of recurvatum (defined as the clinical and radiological presence ≥5° of hyperextension). For this, we evaluated the full extension lateral postoperative radiographic and hyperextension maneuvers performed by the treating surgeon in successive postoperative controls. We also recorded any intraoperative or postoperative complications until the end of the study. Finally, we calculated prosthetic survival, taking revision for any cause as the endpoint of the analysis.
2.3 Ethics
This study was performed after approval by our hospital's ethics and institutional review committee (Protocol 11324).
3 Results
In the last postoperative control registered, the active flexion showed significant improvements compared to the preoperative range (101 vs 109; p = 0.01). All patients achieved full passive extension, while 4/12 (33.33 %) recovered the preoperative active extension gap, three of the group without neurological history and one of those with neurological history (Table 2). At the end of the study, all patients had a stable knee, with no clinical or radiological recurrence of recurvatum.
Regarding the global clinical and functional KSS scores, we observed a significant improvement in the scores at one year (32.6 ± 12.3 versus 78.6 ± 10.9 p < 0.01; and 35.0 ± 12.8 versus 79.4 ± 9.4 p < 0.01) and at the end of the study (79.4 ± 9.6 and 79.6 ± 8.7, p < 0.01). The mean WOMAC was 24.1 ± 9.6 points at the end of the follow-up.
3.1 Complications, demarcation, and revisions
We recorded one (4.0 %) complication. One patient presented a superficial wound infection in the immediate postoperative period that resolved with systemic antibiotic treatment.
Four (16.0 %) cases showed radiolucent lines around the implant. Three are around the tibial component (zone 1 in 2 cases and zone 1,2 in one), and one is around the femoral (zone 1). No case showed radiolucent lines around the stems. No loosening or revisions were recorded at the study end.
4 Discussion
The main finding of our study was that in patients with knee osteoarthritis and genu recurvatum, with or without a neurological history, TKR with a rotating hinged prosthesis resulted in improved function, with no hyperextension recurrence and excellent mid-term prosthetic survival.
Using a hinged implant in a primary TKR is a topic of ongoing discussion.5,17. It arises because early prosthetic designs are associated with high failure rates, primarily related to constriction.5,17. However, improvements in the design of the current third-generation rotating hinge implants have shown good survival rates in the medium and long-term follow-up, both in primary and revision surgeries.18.
In recent years, the concept that rotating hinges are more constraining than condylar-constrained prostheses (CCK) is shifting. A recent biomechanical study showed that rotating hinges have lower stress rates throughout the range of motion, closer to normal knee biomechanics than CCK models. This is due to the greater rotational freedom of the rotating hinges. It may support the improved survival of currently available rotator hinges.19,20.
Regarding the type of prosthesis used in patients with recurvatum without neuromuscular disorders, different authors reported using posterior cruciate-retaining and posterior-stabilized prostheses. Mesnard et al.4 reported the results of 32 posteriorly stabilized TKRs in patients with recurvatum >10°, in which they performed under-resection of the distal femur. At 7.4 years of follow-up, the authors reported 15 % of complications, 65 % of recurvatum recurrence, and 9.4 % of revisions originated by component malpositioning or oversizing.4. Meding et al.1 reported the results of 57 posterior cruciate-preserving TKRs in patients with recurvatum >5° and valgus misalignment from 0 to 20°. They reported a recurvatum recurrence rate of 3.5 % and medial instability in 5.2 % of cases, with no reported revisions at 4.5 years of follow-up.1. While in patients who, in addition to recurvatum, present instability in the coronal plane, Kearns et al.9 reported the use of rotating hinges in 14 primary TKRs (7 for deformities in the coronal plane, 6 recurvatum, and one Charcot arthropathy) and 65 revisions. Although they do not discriminate the results between both groups of patients, they report 38 % complications. They mention extensor apparatus allograft failures, infections, and periprosthetic fractures as the most common. They also report 3 (3.8 %) cases of capture mechanism failure and one (1.2 %) recurrence of recurvatum with global instability.
In our series, discriminating those patients without neuromuscular involvement, like the authors above, we observed a significant improvement in the functional scores. Only one complication was recorded, unrelated to the implant, with no recurrence of the deformity or revisions until the end of the study.
One point that characterizes our series was that nine (39 %) of the cases presented motor sequelae that affected quadriceps strength and, in 3 of them also, ankle flexion. Although different authors have mentioned marked quadriceps weakness as a relative contraindication for TKR, several reports in patients with poliomyelitis highlight significant improvements in functional scores and quality of life. In a recent systematic review of this group of patients, the authors recommended using constrained implants in the presence of quadricipital muscle deficits (M3-M4). It was based on the high recurrence rate of recurvatum (up to 28 %) at 3.5 years of follow-up, with CR, PS, or CCK implants.10. In our series, none of the patients with neurological impairment showed recurrence, and there were significant improvements in functional scores. We should mention that only one patient in this subgroup recovered full active extension. This means that during ambulation in the full extension phase, lacking quadriceps, the implant locks the knee in extension. We do not know if this will affect prosthetic survival. However, at the time of the study closure, none of the patients showed signs of aseptic loosing or instability, similarly behaving in this aspect to patients without neurological affection.
Regarding prostheses survival, at a median follow-up of 6 years, our series did not show signs of mechanical failure either by aseptic loosening reported in up to 17 %.9,18. We also did not record dislocations or failure of the prosthetic capture mechanism, which have been reported in 17 % and 6.8 %, respectively.5. Comparatively, our prosthetic survival rate was superior to that reported by Abdulkarim et al.5 in their systematic review and meta-analysis of the outcomes of rotating hinges in primary surgeries, 82 % at 6–10 years and 88 % beyond ten years of follow-up. It was like the 94 % at 13.5 years, as reported by Kendoff et al.18 evaluating the same prosthetic model used in our series in 238 primary TKRs.
Finally, although the objective of this study was not to compare patients with and without neurological disorders, we observed that patients with motor deficits were younger, with lesser femorotibial misalignment and lower preoperative functional scores. Indeed, at the end of the follow-up, neither group showed differences in prostheses survival or complications, with functional scores being higher in those patients without neurology, which seems logical. However, when analyzing the impact of surgery on their perception of health, the best scores were presented by patients with neurological impairment (18.2 vs. 27.4). We believe that this could be a consequence of the fact that patients with neurological disorders had worse preoperative functional scores (28.7 vs. 38.3). Therefore, by providing them with a stable knee even with a poorly functional quadriceps, this limitation was reduced, generating a greater impact on their perception of health. However, this should be analyzed more deeply and will be the subject of a future study.
The limitations of this study are related to its retrospective design, with a low number of patients. Another limitation, although it was not the study's aim, is the lack of a comparative group with patients treated with another prosthetic design with lower constriction levels, which could provide more strength to our results, especially regarding the deformity recurrence and prosthetic survival rate. Finally, the absence of previous reports using this prosthetic model in patients (with and without neuromuscular impairment) with this degree of recurvatum limited our ability to compare the results. Indeed, there was the absence of reports of patients with neurological conditions presenting genu recurvatum, most of the series patients with poliomyelitis in which only a percentage presented hyperextension, again limiting the comparative analysis. To our knowledge, this is the first report describing the results exclusively in patients with genu recurvatum >10°, with and without neurological conditions, treated with rotating hinges in primary TKR.
5 Conclusions
Our results suggest that in patients with knee osteoarthritis and genu recurvatum, with and without neurological history, primary arthroplasty with rotating hinged prosthesis allowed restoration of function, without recurrence of the deformity, and with excellent prosthetic survival in the medium term.
Guardian/patient's consent
This was a retrospective study, so exception for patient's informed consent was given.
CRediT authorship contribution statement
Germán Garabano: Conceptualization, Methodology, Writing – original draft. Leonel Perez Alamino: Data curation, Writing – original draft, Formal analysis. Xavier Arturo Maya Nieto: Visualization, Investigation. Cesar Pesciallo: Supervision, Writing – review & editing, Validation.
Ethics
This study was performed after approval by our hospital's ethics and institutional review committee (Protocol 11324).
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References
- Genu recurvatum in total knee replacement. Clin Orthop Relat Res. 2003 Nov;416:64-67.
- [Google Scholar]
- Total knee replacement in patients with genu recurvatum. Clin Orthop Relat Res. 2001 Dec;393:244-249.
- [Google Scholar]
- Outcomes of total knee arthroplasty in degenerative osteoarthritic knee with genu recurvatum. Knee. 2018 Jan;25(1):167-176.
- [Google Scholar]
- Posterior-stabilized TKA in patients with severe genu recurvatum Achieves good clinical and radiological results at 5-year minimum follow-up: a case-Controlled study. J Arthroplasty. 2021 Sep;36(9):3154-3160.
- [Google Scholar]
- Rotating-hinge knee prosthesis as a viable option in primary surgery: Literature review & meta-analysis. Orthop Traumatol Surg Res. 2019 Nov;105(7):1351-1359.
- [Google Scholar]
- Recurvatum deformity complicating performance of total knee arthroplasty. A brief note. J Bone Joint Surg Am. 1990 Feb;72(2):268-271.
- [Google Scholar]
- High recurrence rates of hyperextension after total knee arthroplasty in asian patients with preoperative knee hyperextension: a prospective observational study. Cureus. 2023 Aug 10;15(8)
- [Google Scholar]
- Rotating hinge implants for complex primary and revision total knee arthroplasty. J Arthroplasty. 2018 Mar;33(3):766-770.
- [Google Scholar]
- Outcome of total knee arthroplasty in patients with poliomyelitis: a systematic review. EFORT Open Rev. 2018 Jun 6;3(6):358-362.
- [Google Scholar]
- History of standard scoring, notation, and summation of neuromuscular signs. A current survey and recommendation. J Peripher Nerv Syst. 2005 Jun;10(2):158-173.
- [Google Scholar]
- Reliability of intraoperative knee range of motion measurements by goniometer compared with robot-assisted arthroplasty. J Knee Surg. 2019;32(3):233-238.
- [Google Scholar]
- The knee society total knee arthroplasty roentgenographic evaluation and scoring system. Clin Orthop Relat Res. 1989;248:9-12.
- [Google Scholar]
- Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol. 1988;15(12):1833-1840.
- [Google Scholar]
- Development of a modern knee society radiographic evaluation system and methodology for total knee arthroplasty. J Arthroplasty. 2015;30(12):2311-2314.
- [Google Scholar]
- Complications of total knee arthroplasty: standardized list and definitions of the Knee Society. Clin Orthop Relat Res. 2013 Jan;471(1):215-220.
- [Google Scholar]
- Patient-reported outcomes after primary rotating hinge total knee arthroplasty: a multi-centre clinical cohort study. Int Orthop. 2021 Nov;45(11):2893-2897.
- [Google Scholar]
- Management of gonarthrosis with a rotating hinge prosthesis: minimum 10-year follow-up. Clin Orthop Surg. 2020 Dec;12(4):464-469.
- [Google Scholar]
- The knee prosthesis constraint dilemma: biomechanical comparison between varus-valgus constrained implants and rotating hinge prosthesis. A cadaver study. J Orthop Res. 2021 Jul;39(7):1533-1539.
- [Google Scholar]
- Rotating hinge knee causes lower bone-implant interface stress compared to constrained condylar knee replacement. Knee Surg Sports Traumatol Arthrosc. 2019 Apr;27(4):1224-1231.
- [Google Scholar]