Outcomes of Distal Femoral Osteotomy for Valgus Malalignment: A Systematic Review and Meta-analysis of Closing Wedge Versus Opening Wedge Techniques

Abstract

Background:

Valgus knee deformity increases the risk for lateral articular chondral damage, contributing to earlier onset and accelerated progression of osteoarthritis. Distal femoral osteotomy (DFO) unloads the lateral joint compartment and can be performed using closing wedge (CW) or opening wedge (OW) techniques.

Purpose:

To perform a systematic review and meta-analysis for patients with valgus knee deformity undergoing DFO to determine differences in patient-reported outcome measures (PROMs), complications, and survival rates, comparing CW versus OW DFO.

Study Design:

Systematic review, Level of evidence, 4.

Methods:

A literature review was performed according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines utilizing PubMed, Cochrane Database, Ovid/MEDLINE, and Scopus. Inclusion criteria consisted of studies reporting outcomes in patients undergoing CW or OW DFO for the treatment of valgus knee deformities with symptomatic lateral compartment pathology with a minimum 2-year follow-up. PROMs and complications were analyzed using random-effects modeling to identify differences in outcomes as a function of surgical technique. Long-term survival data, defined as conversion to total knee arthroplasty, were analyzed using a multiple metaregression model as a function of individual study follow-up time points and surgical technique.

Results:

In total, we included 23 retrospective studies (n = 619 knees), of which 10 studies (n = 271 knees) reported outcomes after CW DFO and 13 studies (n = 348 knees) reported on OW DFO outcomes. Good to excellent clinical outcomes were reported in PROMs when compared with preoperative values with both techniques, while no significant differences between techniques were appreciated on functional Knee Society Scores and Tegner scores. No significant differences were appreciated in the incidence of complications reported in patients undergoing CW (20%) versus OW (33%) DFO (P = .432). Pain requiring hardware removal was the most commonly reported complication in both groups. The survival rate for CW DFO was 81.5% (mean follow-up, 8.8 ± 4.3 years) compared with 90.5% for OW DFO (mean follow-up, 4.5 ± 1.5 years). Multiple metaregression demonstrated that patient follow-up (P < .001) was significantly associated with knee survival, while surgical technique (P = .810) was not a predictor of clinical failure.

Conclusions:

Both CW and OW DFO techniques were associated with good to excellent clinical outcomes with no significant differences in PROMs based on technique. Pain requiring hardware removal was the most common complication in both techniques, while long-term survivability was found to be a function of follow-up and not surgical technique. Technique selection should be based on shared patient-physician decision making with an emphasis on surgeon preference and technique familiarity.

Keywords

distal femoral osteotomy valgus opening wedge closing wedge

Weight distribution within the knee is dependent on the mechanical alignment of the lower extremity, with malalignment increasing the risk for chondral damage due to abnormal joint loading.¹ Patients with a valgus knee deformity have been shown to be at greater risk for both earlier onset and accelerated progression of osteoarthritis within the lateral joint compartment.¹⁷ Treatment is challenging for young patients with valgus knee malalignment. Realignment via total knee arthroplasty (TKA) is not recommended in younger patients because of the high risk for revision surgery secondary to implant longevity.⁶ Furthermore, patients with valgus malalignment >15° are contraindicated for unicompartmental knee arthroplasty.³ Performance of a distal femoral osteotomy (DFO) has been shown to alleviate symptoms and slow the progression of chondral damage within the lateral knee compartment by restoring the mechanical axis, effectively redistributing load away from the diseased lateral joint compartment.³⁵

Valgus knee malalignment has been shown to develop through a variety of mechanisms, including lateral femoral condyle hypoplasia, trauma, lateral meniscal deficiency, and medial collateral ligament laxity, as well as a variety of metabolic disorders affecting bone formation and growth during skeletal development.^5,35 Symptomatic valgus deformity is generally corrected through the femur, allowing for larger joint correction^7,8 while maintaining native tibial alignment.⁷ DFO correction can be performed using an opening wedge (OW) technique along the lateral femur or a closing wedge (CW) technique along the medial femur. While the OW DFO surgical exposure is farther away from critical neurovascular structures, this technique carries the potential risk of fracturing through the medial cortex along the hinge, as well as the potential for bone healing complications (nonunion, malunion, delayed union).²³ The CW DFO surgical technique, while more technically challenging, offers the advantage of faster healing time as a result of primary bone healing via direct bony opposition, with a lower risk of fracture along the lateral cortex.^32,39 Meanwhile, the advantages of the OW DFO include the capacity for intraoperative alignment adjustments, utilization of a more familiar surgical exposure, and access to the lateral aspect of the knee.^1,4 However, clinical outcomes and knee survivorship after CW versus OW DFO remain largely unknown.

The purpose of this investigation was to perform a systematic review of the literature evaluating patient-reported outcome measures (PROMs), the incidence of complications, and knee survival rates after DFO while comparing outcomes between CW and OW DFO techniques. The authors hypothesized both DFO techniques would result in improved PROMs when compared with preoperative values, with low complication rates and high survival rates without any significant differences between techniques.

Methods

Data Sources and Searches

A systematic review was conducted according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines using a PRISMA checklist. All literature pertaining to the performance of CW and OW DFO from inception to January 2021 was identified. Two authors (C.C.D. and A.K.) independently conducted a literature search in February 2021 using the following databases: PubMed, Cochrane Database, Ovid/MEDLINE, and Scopus. Each search included various combinations of the following terms, including Medical Subject Headings (MeSH): distal AND (“femur” MeSH Terms OR femur OR femoral) AND (“osteotomy” MeSH Terms OR osteotomy) AND (valgus or valgum).

Selection Criteria

Predefined eligibility criteria consisted of articles written in English or articles with English translation reporting on outcomes in human patients undergoing CW or OW DFO for the treatment of valgus deformities with symptomatic lateral compartment disease with a minimum 2-year clinical follow-up. Exclusion criteria consisted of non-English articles, animal studies, biomechanical studies, patients undergoing osteotomy for cosmetic concerns, and review and surgical technique articles.

Data Extraction

After the 2 independent authors’ search of the literature, 981 articles were considered for further evaluation. Sequential screening of the articles was performed by the 2 authors in the following systematic approach: assessment of duplicate articles, content within the article title, content of the abstract, and full-text review. Full-text review was performed during the study selection process, if necessary, to determine if the articles satisfied inclusion and exclusion criteria. The search process is shown in the flow diagram (Figure 1). After application of the inclusion and exclusion criteria, 23 studies were identified for further analysis. No disagreement in study selection was appreciated between authors. To ensure that all available studies were identified, the authors cross-referenced references cited in the included articles for inclusion if they were overlooked during the initial search, during which no further studies were identified.

Data recorded from the included studies included level of evidence, number of patients, length of follow-up, osteotomy technique (CW vs OW), method of fixation (if applicable), the use of any adjuvant bone grafting, PROMs, the incidence of postoperative complications, and knee survival rates. Survival of the native knee was defined as the time between DFO and conversion to TKA. Primary study endpoints were PROMs, the incidence of complications related to the DFO procedure, and knee survival rates.

Statistical Analysis

On exploratory analysis, studies were expected to have high heterogeneity because of diverse patient populations and limited high-level evidence. Thus, the DerSimonian-Laird method was utilized to calculate pooled effect sizes.^10-12 Heterogeneity was evaluated using I² values, and all pooled statistics were reported with 95% CIs. Binary outcomes were assessed using a random-effects meta-analysis of proportions, while continuous outcomes were assessed using a random-effects meta-analysis of effect differences between postoperative and preoperative time points. Results were summarized as forest plots including subgroup analysis by surgical technique and the overall results of either technique. Subgroup analysis by technique was limited to functional Knee Society Score (KSS) and the Tegner score, as these were the only PROMs reported for both OW and CW DFO. Refer to Supplementary Figures (available in the online version of this article) for results pertaining to the International Knee Documentation Committee score, Hospital for Special Surgery Score, International Knee Score, Knee Injury and Osteoarthritis Outcome Score, Lysholm score, Objective Knee Society Score, and Oxford score. Clinical outcomes with insufficient data, including lack of current data reporting on both surgical techniques, were summarized using forest plots without subgroup analysis.

Based on predefined clinical relevance, a multiple metaregression model was built utilizing a stepwise entry method of follow-up time point and surgical technique variables. Collinearity of these variables (C = 0.53) was below the exclusion threshold, indicating multiple metaregression investigating these clinically defined features was appropriate.¹⁴ Multiple metaregression was performed utilizing maximum likelihood estimators for τ² calculations and Knapp-Hartung adjustments for regression coefficient to account for the relatively small number of studies included.²⁸ The alpha level was set to 0.05. Model fit was assessed via R², F test, and Akaike information criterion (AIC) measures. Metaregression models were assessed utilizing the likelihood ratio test after each stepwise addition of predictors. All statistical analyses were conducted in R Version 3.6.2 (R Foundation for Statistical Computing).

Results

Patient Characteristics

A total of 23 studies reporting on 619 knees, consisting of 10 studies (n = 271 knees) evaluating outcomes after CW DFO and 13 studies (n = 348 knees) reporting on OW DFO outcomes, were identified (Table 1). Indications for DFO within individual studies are summarized in Table 2. The mean age at the time of DFO was significantly longer in knees undergoing CW (49.4 ± 5.1 years) versus OW (41.7 ± 8.3 years) DFO (P = .021). The mean follow-up was significantly greater in knees treated using CW DFO (8.8 ± 4.3 years) versus OW DFO (4.5 ± 1.5 years) (P = .017).

Table 1

Overview of Included Studies^a

Lead Author (Year)	No. of Patients/ Knees	Age, y	Follow-up, y	Mean Preoperative Valgus	Mean Postoperative Valgus	No. of Patients Converted to TKA	Adjuvant Treatments	Concurrent Procedures
Closing Wedge Distal Femoral Osteotomy
Backstein (2007)²	38/40	44.1	10.25	ATFA, 11.6°	ATFA, –1.2°	8	Autograft	0
Finkelstein (1996)¹⁸	20/21	56	11.08	N/R	ATFA, ~0°	7	N/R	0
Sternheim (2011)³⁷	41/45	46.2	13.3	N/R	ATFA, 0°	14	Autograft	0
Kosashvili (2010)²⁵	31/33	45.5	15.1	N/R	N/R	15	N/R	0
Forkel (2015)¹⁹	22/22	47	2.5	MATFA, 77.3%	MATFA, 42.6%^b	0	N/R	Microfracture (n = 17)
Wang (2006)⁴¹	30/30	53	8.25	ATFA, 18.2°	ATFA, ~0°	3	Autograft	Lateral compartment subchondral drilling (n = 30) Lateral release + patellofemoral subchondral drilling (n = 6) Proximal patellar realignment (n = 1) Distal patellar realignment (n = 1)
McDermott (1988)²⁹	24/24	53	4	N/R	ATFA, ~0°	1	Autograft	0
Healy (1988)²⁰	21/23	56	4	ATFA, 18°	ATFA, 2°	2	Autograft	0
Learmonth (1990)²⁶	11/12	40	3.41	N/R	ATFA, 0.46°	0	N/R	0
Stähelin (2000)³⁶	19/21	57	5	ATFA, 12°	ATFA, 1.7°	0	N/R	0
Opening Wedge Distal Femoral Osteotomy
Das (2008)⁸	12/13	55	2.83	ATFA, 16°	ATFA, 5°	2	Allograft	0
Dewilde (2013)¹³	19/19	47	5.66	MATFA, 5.3°	MATFA, –1.3°	2	Calcium phosphate cement	ACI (n = 1)
Jacobi (2011)²²	14/14	46	3.75	N/R	Mean correction, 5.8°	0	Cortical iliac crest autograft (n = 7/14)	0
Thein (2012)³⁹	6/7	46.7	6.5	ATFA, 13.5°	ATFA, 1.6°	0	Tricortical iliac crest allograft	0
Zarrouk (2010)⁴³	20/22	53	4.5	MATFA, 14.5°	MATFA, 1.5°	1	None	0
Ramanathan (2014)³¹	74/78	33	3.6	N/R	N/R	7	N/R	Cartilage restoration procedures in 60.3%
Saithna (2014)³³	18/21	41	4.5	MATFA, 75%^b	MATFA, 37%^b	4	Bone graft for wedge opening >12 mm	ACI to lateral femoral condyle (n = 1) MAT (n = 1) Microfracture (n = 3)
Cameron (2015)⁴	30/31	AG, 41 (n = 19) JPG, 26 (n = 12)	5	MATFA, 6.2°	MATFA, –2.0°	6	Iliac crest autograft, cancellous allograft, or both	OCA (n = 6) Meniscetomy (n = 3) MAT (n = 2) Chondroplasty (n = 1) Partial meniscetomy + ACLR (n = 1)
Madelaine (2016)²⁷	27/29	44.4	6.7	MATFA, 7.8°	MATFA, 0.4°	5	Iliac crest autograph	Lateral facetectomy (n = 7) Knee arthroscopy (n = 1)
Elattar (2017)¹⁶	28/41	44	2.2	MAD, 25.3 mm^c	MAD, 8 mm^d	0	Allograft cancellous chips + DBM (n = 41/41); BMAC (n = 6/41)	Sagittal plane corrective procedure (n = 8) Lateral retinacular release (n = 8) HTO (n = 4) Knee arthroscopies (n = 3) TTO (n = 2)
Ekeland (2016)¹⁵	24/24	48	7.9	ATFA, 11.5°	ATFA, 2.0°	6	Iliac crest autograph	0
Kolb (2019)²⁴	22/23	23.7	5.4	MAD, 34.8 mm^c	MAD, 1.9 mm^d	0	Bone graft substitute (n = 5/23)	N/R
de Carvalho (2014)⁹	26/26	48.6	4	ATFA, 12°	ATFA, 0°	0	N/R	Lateral partial meniscectomy (n = 6) Lateral compartment chondroplasty (n = 6)

All included studies had a level of evidence of 4 and were retrospective case series in design. ACI, autologous chondrocyte implantation; ACLR, anterior cruciate ligament reconstruction; AG, arthritic group (patients undergoing osteotomy for arthritic pain); ATFA, anatomic tibiofemoral angle; BMAC, bone marrow aspirate concentrate; DBM, demineralized bone matrix; HTO, high tibial osteotomy; JPG, joint preservation group (patients undergoing osteotomy for joint preservation); MAD, mechanical axis deviation; MAT, meniscal allograft transplantation; MATFA, mechanical axis tibiofemoral angle; N/R, not recorded; OCA, osteochondral allograft; TKA, total knee arthroplasty; TTO, tibial tubercle osteotomy.

Measured from medial to lateral.

Measured from lateral to midline.

Measured from medial to midline.

Table 2

Indications for Closing and Opening Wedge Distal Femoral Osteotomies in Included Studies^a

	Study	Indication
Closing Wedge DFO	Backstein et al²	All patients who underwent DFO for isolated lateral unicompartmental OA with valgus deformity between 1972 and 2002 were included.
	Finkelstein et al¹⁸	All patients who underwent DFO for isolated lateral unicompartmental OA with valgus deformity between 1972 and 1985 were included.
	Sternheim et al³⁷	All patients within the time frame were included. However, prerequisites for distal femoral varus osteotomy were a 90° range of motion in the knee, age <60 y for women and 65 y for men, and physical and mental capability of a long rehabilitation process lasting up to 9 mo.
	Kosashvili et al²⁵	33 consecutive knees that underwent DFO for isolated lateral unicompartmental OA with valgus deformity between January 1984 and January 1999 with a minimum follow-up of 10 y were included.
	Forkel et al¹⁹	23 consecutive patients with symptomatic grade 3-4 cartilage damage and valgus knee alignment were included. Prerequisites for undergoing the procedure included intact medial cartilage and meniscus, desire of the patient to continue with sporting activity, and age <55 y. Radiological definition of genu valgum was an intersection of the Mikulicz line with the tibia plateau at >60%.
	Wang and Hsu⁴¹	The indication for DFO was a painful deformity of the knee associated with a valgus tibiofemoral angulation ≥12° and narrowing of the lateral joint space. Contraindications included severe arthritis of the medial compartment of the knee, severe tricompartmental OA, and tibiofemoral subluxation.
	McDermott et al²⁹	Included patients had OA of the lateral compartment that was associated with a valgus deformity and a superolateral tilt to the joint line. They also had both clinical and radiographic deformity of the involved knee, a normal or minimally involved medial compartment, good vascular circulation, and flexion of the knee to ≥90°.
	Healy et al²⁰	18 knees with varying forms of arthritis that had a primary indication for distal femoral varus osteotomy because of noted painful valgus deformity of the knee with narrowing of the lateral joint space.
	Learmonth²⁶	All patients who underwent DFO for isolated lateral unicompartmental OA with valgus deformity were included.
	Stähelin et al³⁶	21 knees with OA secondary to valgus deformity that underwent DFO were included.
Opening Wedge DFO	Das et al⁸	Indication for an opening wedge osteotomy was mild to moderate lateral radiographic OA associated with genu valgum malalignment of >10° in a patient with a biological age <65 y.
	Dewilde et al¹³	Indication for the DFO was the presence of isolated lateral femorotibial OA associated with an underlying mild to moderate valgus deformity, in patients aged <55 y with an active lifestyle and normal stability and range of motion of the knee.
	Jacobi et al²²	All included patients had lateral unicompartmental OA with an associated valgus axis.
	Thein et al³⁹	Patients included in this study were aged <65 y with isolated OA of the lateral compartment, tibiofemoral angle >12° of valgus, knee flexion >90°, and a flexion contracture <10° and without ligamentous instability. The indication for surgery was knee pain due to evident OA on examination and standing radiographs that was not responsive to nonoperative treatment for at least 1 y.
	Zarrouk et al⁴³	The inclusion criteria for this study were complete radiological workup, minimum 3 y of follow-up, and a single surgical technique (ie, only the DFO was performed). All the patients had a unilateral or bilateral symptomatic valgus knee deformity.
	Ramanathan et al³¹	Young patients with severe unicompartmental knee OA and malalignment were included
	Saithna et al³³	Patients included had lateral unicompartmental OA with valgus deformities.
	Cameron et al⁴	Patients had symptomatic lateral compartment arthritis with clinical valgus deformity or a cartilage or meniscal defect in the lateral compartment with clinical valgus alignment.
	Madelaine et al²⁷	Patients included had symptomatic genu valgus with signs of lateral compartment arthritis with or without patellofemoral symptoms.
	Elattar et al¹⁶	The indication for osteotomy was lateral knee pain and genu valgum deformity.
	Ekeland et al¹⁵	Included patients had severe pain in an osteoarthritic knee with valgus malformation.
	Kolb et al²⁴	All patients with genu valgum associated with knee pain or limited knee function due to valgus deformity were included.
	de Carvalho et al⁹	Young active adults with symptomatic OA and valgus deformity were included.

DFO, distal femoral osteotomy; OA, osteoarthritis.

Patient-Reported Outcome Measures

Functional KSS

One study² (n = 30 knees) reported on functional KSS for CW DFO, while 3 studies^13,27,43 (n = 70 knees) reported on functional KSS for OW DFO (Figure 2). No significant difference was appreciated in mean postoperative improvement in functional KSS when comparing CW (31.60; 95% CI, –17.40 to 80.60) versus OW (24.96; 95% CI, 3.87 to 46.05) DFO (P = .7).

Tegner Score

One study¹⁹ (n = 22 knees) reported on Tegner scores for CW DFO, while 2 studies^9,33 (n = 47 knees) reported on Tegner scores for OW DFO (Figure 3). No significant differences in mean postoperative improvement in Tegner scores were reported between CW (0.70; 95% CI, –1.46 to 2.86) and OW (0.11; 95% CI, –1.97 to 2.19) DFO techniques (P = .734).

Hospital for Special Surgery Score

Three studies^20,36,41 (n = 74 knees) reported on Hospital for Special Surgery Scores for CW DFO. The mean improvement was 23.39 (95% CI, 7.69-39.09).

Objective Knee Society Score

Two studies^2,25 (n = 63 knees) reported on Objective Knee Society Score for CW DFO. The mean improvement was 42.66 (95% CI, –2.07 to 87.40).

International Knee Documentation Committee Score

Two studies^4,33 (n = 52 knees) reported on International Knee Documentation Committee scores for OW DFO. The mean improvement was 21.27 (95% CI, 3.34-39.20).

Knee injury and Osteoarthritis Outcome Score

Two studies^22,24 (n = 37 knees) reported on Knee injury and Osteoarthritis Outcome Scores for OW DFO. The mean improvement was 30.56 (95% CI, 1.85-59.37).

Lysholm Score

Two studies^9,33 (n = 47 knees) reported on Lysholm scores for OW DFO. The mean improvement was 16.00 (95% CI, –8.10 to 40.09).

Oxford Score

One study³⁹ (n = 7 knees) reported on Oxford Scores for OW DFO. The mean improvement was 12.90 (95% CI, –8.84, 34.61).

International Knee Score

One study⁴³ (n = 22 knees) reported on International Knee Scores for OW DFO. The mean improvement was 24.95 (95% CI, –13.27 to 63.17).

Overall Complications

Overall complications were aggregated from 7 studies^{18,19,20,26,29,36,41} investigating CW DFO and 13 studies^‡ investigating OW DFO. The pooled mean proportions of overall complications were 0.20 (95% CI, 0.05-0.41) for CW DFO and 0.33 (95% CI, 0.14-0.55) for OW DFO. There was no significant difference in overall complications between techniques (P = .432).

Perioperative Complications

Thromboembolism

Perioperative thromboembolic events were aggregated from 5 studies^{18,19,26,29,36} investigating CW DFO and 7 studies^{4,8,16,24,27,31,39} investigating OW DFO. The pooled mean proportions of perioperative thromboembolism were 0.02 (95% CI, 0-0.06) for CW DFO and 0 (95% CI, 0-0.01) for OW DFO. There was no significant difference in perioperative thromboembolism between techniques (P = .090) (Figure 4).

Infection

Perioperative infections were aggregated from 5 studies^{18,19,26,29,36} investigating CW DFO and 7 studies^{4,8,15,24,31,33,39} investigating OW DFO. The pooled mean proportions of perioperative infection were 0.02 (95% CI, 0-0.06) for CW DFO and 0.01 (95% CI, 0-0.03) for OW DFO. There was no significant difference in perioperative infection between techniques (P = .438) (Figure 5).

Postoperative Complications

Nonunion

Nonunion events were aggregated from 5 studies^{19,20,26,36,41} investigating CW DFO and 11 studies^§ investigating OW DFO. The pooled mean proportions of nonunion were 0.02 (95% CI, 0-0.07) for CW DFO and 0.01 (95% CI, 0-0.03) for OW DFO. There was no significant difference in nonunion between techniques (P = .472) (Figure 6).

Delayed Union

Delayed union events were aggregated from 4 studies^19,26,36,41 investigating CW DFO and 12 studies^¶ investigating OW DFO. The pooled mean proportions of delayed union were 0 (95% CI, 0-0.01) for CW DFO and 0.02 (95% CI, 0-0.06) for OW DFO. There was no significant difference in delayed union between techniques (P = .174) (Figure 7). Estimates of the proportion of delayed union were highly heterogeneous among OW and CW groups (I² = 65%).

Arthrofibrosis

Arthrofibrosis complications were aggregated from 4 studies^18,20,26,29 investigating CW DFO and 8 studies^{4,8,9,15,16,27,31,39} investigating OW DFO. The pooled mean proportions of arthrofibrosis were 0.03 (95% CI, 0-0.08) in CW DFO and 0.03 (95% CI, 0.01-0.08) in OW DFO. There was no significant difference in arthrofibrosis between techniques (P = .969) (Figure 8).

Loss of Correction

Postoperative correction loss was aggregated from 5 studies^{18,19,26,29,36} investigating CW DFO and 7 studies^{8,13,15,27,31,33,39} investigating OW DFO. The pooled mean proportions of postoperative correction loss were 0.04 (95% CI, 0.01-0.09) for CW DFO and 0.04 (95% CI, 0.01-0.07) for OW DFO. There was no significant difference in postoperative correction loss between techniques (P = .791) (Figure 9).

Pain Requiring Surgical Hardware Removal

Pain requiring surgical hardware removal was aggregated from 3 studies^19,20,26 investigating CW DFO and 9 studies^{4,8,9,13,22,24,27,31,33} investigating OW DFO. The pooled mean proportions of pain requiring surgical hardware removal were 0.20 (95% CI, 0-0.74) for CW DFO and 0.34 (95% CI, 0.14-0.56) for OW DFO. There was no significant difference in pain requiring surgical hardware removal between techniques (P = .598) (Figure 10). Estimates of the proportion of pain requiring surgical hardware removal were highly heterogeneous among CW and OW groups (I² = 92%).

Long-Term Survivability

Survivorship was reported in all 10 CW DFO studies (n = 271 knees). At a mean final follow-up of 8.8 ± 4.3 years, 18.5% (n = 50/271 knees) of knees in the CW group required conversion to TKA, resulting in an overall survival rate of 81.5%. At 4-year follow-up, knee survival rates ranged from 83%²⁰ to 100%,³⁶ whereas survivorship reported in studies with a 10-year minimum follow-up ranged from 64%¹⁸ to 89.9%.³⁷ At 15 years of follow-up, CW survival rates ranged from 45%² to 78.9%.^25,37 Sternheim et al³⁷ reported a survival rate of 21.5% at 20 years of follow-up.

Survivorship was reported in all 13 OW DFO studies. At a mean final follow-up of 4.5 ± 1.5 years, 9.5% (n = 33/348) knees were converted to TKA, yielding a survival rate of 90.5%. Survival rates reported at a 5-year follow-up ranged from 74%²⁴ to 100%.³⁹ Ekeland et al¹⁵ reported the longest mean follow-up of 7.9 years in 24 patients with survival rates of 88% at 5 years and 74% at 10 years.

Utilizing a stepwise approach to metaregression analysis, the initial metaregression modeling clinical failure as a function of follow-up time point demonstrated that a longer follow-up time point was associated with higher rates of clinical failures (F_1,18 = 38.33; P < .001; R² = 83.27%; AIC = 42.56). The follow-up time point accounted for a large proportion of the heterogeneity observed in clinical failures across studies. Inclusion of surgical technique in the metaregression model demonstrated that surgical technique was not associated with clinical failure (P = .810) and produced an overall slightly inferior model fit (F_2,17 = 18.20; P < .001; R² = 83.49%; AIC = 44.50). Inclusion of surgical technique in the multiple metaregression model did not result in a significant improvement in model performance (χ² = 0.06; P = .806), indicating that surgical technique was not a reliable predictor of clinical failure (Figure 11).

Discussion

The main findings of this study were that both CW and OW DFO techniques resulted in improved functional KSSs and Tegner activity scores when compared with preoperative values while no significant differences in the incidence of complications between techniques was appreciated. The most common complication after DFO was pain requiring surgical hardware removal. Conversion to TKA after DFO was a function of follow-up (P < .001) and not surgical technique (P = .806).

No significant difference in native knee survival was found between CW and OW DFO groups at any of the observed time points, while a decline in knee survivorship was appreciated with increased follow-up regardless of technique. The CW DFO technique had a slightly lower native knee survival rate (81.5%) compared with the OW group (90.5%). The minor variation in knee survival between CW and OW DFO groups may be attributable to the significantly longer follow-up in the CW DFO cohort (P = .017), a factor found to be associated with increasing clinical failure rates via multiple metaregression (P < .001). This association is evident in studies reporting on individual survival analysis, such as Sternheim et al,³⁷ who reported a survival rate of 78.9% at 15 years, decreasing to 21.5% at the 20-year follow-up. Backstein et al² reported survival rates of 82% at 10 years and 45% at 15 years. Furthermore, 88% of reported TKAs in the CW group occurred in studies reporting >10-year follow-up, while no studies of OW DFO reported on survival beyond a 10-year follow-up.^2,18,25,37

Another possible explanation for the slight difference in clinical failures between DFO techniques is age at the time of intervention.³⁰ The mean age in the CW DFO cohort was significantly older than that of the OW DFO cohort (P = .0214), and thus there may have been a lower threshold for conversion to TKA in the CW DFO group. Older patients with residual pain are more likely to undergo TKA when compared with younger populations because of concerns for implant longevity and high rates of TKA revision reported in patients aged <55 years.³⁴ Conversely, younger patients undergoing OW DFO may be more physically active, which may accelerate articular degeneration and increase risk for TKA.⁴⁰ Based on current evidence, 50% of patients undergoing DFO postpone knee replacement >15 years.³⁷ Moreover, patient selection may also account for the difference in knee survival. Bone density and patient age may have played a role in determining which patients were indicated for CW versus OW DFO. Namely, CW DFO allows for quicker healing at the osteotomy site because of bone-to-bone apposition.³⁵ As such, patients undergoing CW DFO may have been indicated for such a procedure because of concerns for a higher risk of clinical failure inherent to OW DFO.

The heterogeneity of reported PROMs limited direct comparison of OW versus CW DFO with functional KSSs and Tegner scores. Improvements in functional KSS were significant (mean, 26) likely because of a more comprehensive assessment of activity level, including activities of daily living.^21,38 Improvement in Tegner scores after DFO was modest (mean, 0.4); however, the Tegner score does not capture changes in activities of daily living.³⁸ Further high-quality studies reporting homogeneous PROMs are warranted to allow a better understanding of outcomes after DFO based on surgical technique.

The most common complication after DFO was pain requiring surgical hardware removal, which was reported to occur in 30% of knees analyzed. The estimated proportion of this complication was highly heterogeneous (I² = 92), which limits the strength of evidence for this outcome. The proportion of knees requiring surgery to remove hardware was larger in the OW group (mean, 0.34); however, this was not statistically different from knees undergoing CW DFO (mean, 0.20; P < .598). Of note, this finding was appreciably higher than the incidence reported (12.1%) in a previous systematic review performed by Wylie et al⁴² examining 16 articles, consisting of 372 knees. The increased rate of hardware removal after OW DFO may be because of hardware friction against the iliotibial band, serving as a pain generator.²²

No significant differences were appreciated in the incidence of complications reported between CW and OW DFO techniques. Previous studies by Kim et al²³ and Chahla et al⁵ similarly reported no significant differences based on complication incidences when comparing CW versus OW DFO techniques. This investigation found perioperative complication rates of arthrofibrosis (CW, 3%; OW, 3%), nonunion (CW, 2%; OW, 1%), delayed union (CW, 0%; OW, 2%), and thromboembolism (CW, 2%; OW, 0%) to be comparable with those reported in the systematic review by Wylie et al,⁴² in which the authors cited a 3.2% rate of nonunion, 3.8% rate of delayed union, 0.5% rate of arthrofibrosis, and 0.5% rate of thromboembolism.

A notable trend was appreciated in the increasing incidence of studies reporting on outcomes after OW DFO when compared with CW DFO. This may be related to increasing awareness of the advantages of the OW technique, requiring a single bone cut without violation of the medial cortex, allowing for increasing correction accuracy via incremental spreading while potentially reducing the risk of complications.^4,16 Meanwhile, despite predictable bony healing,³⁷ the technical challenges inherent with a medially based approach, as well as reported complications involving hardware failure, overcorrection, and undercorrection after CW DFO, may have contributed to increasing surgeon preference for using the OW technique.⁵ Further investigations evaluating complication incidence, as well as patient satisfaction and variables associated with surgeon preference, such as the use of orthobiologics, are necessary to better understand the factors responsible for the observed trend.

Limitations

This study was not without limitations. There remains a lack of high-level, standardized evidence studies reporting on outcomes after DFO, limiting the studies included in this review to retrospective case series. Furthermore, the data reported on outcomes after CW versus OW DFOs were heterogeneous, further limiting the ability to perform any meaningful statistical analyses on a number of variables while likely underpowering others. Namely, the influence of different approaches to osteotomy fixation, gap wedge filling, degree of lateral compartment degeneration, size of angle of correction, utilization of bone grafting and other adjuvant therapies, rehabilitation protocols, and disease burden at intervention were infrequently reported, and their effect on the outcomes reported in this study cannot be extrapolated. Analyses of Tegner score and functional KSS improvements were based on small sample sizes and were likely underpowered, limiting the ability to detect any meaningful statistical differences in outcomes based on the DFO technique. Patients undergoing CW DFO were found to be significantly older when compared with patients undergoing OW DFO, potentially confounding knee survivorship analysis. The heterogeneity of PROMs reported in the current literature limited the compilation and aggregation of data such that outcomes were not quantitatively comparable among studies. Discussion toward a common consensus of PROM instrument for patients with valgus knee malalignment pathology would facilitate more rigorous transitive investigations between individual studies.

Conclusion

Both CW and OW DFO techniques are associated with good to excellent clinical outcomes with no significant differences in PROMs based on technique. Pain requiring hardware removal was the most common complication in both techniques, while long-term survivability was found to be a function of follow-up and not surgical technique. Technique selection should be based on shared patient-physician decision making with an emphasis on surgeon preference and technique familiarity.

Footnotes

Submitted March 15, 2021; accepted June 7, 2021.

One or more of the authors has declared the following potential conflict of interest or source of funding: J.C. has received consulting fees from Arthrex, Conmed Linvatec, Ossur, DePuy Synthes, and Smith & Nephew; education support and nonconsulting fees from Arthrex and Smith & Nephew; grant support from Arthrex; speaking fees from Linvatec; and hospitality payments from Medwest. B.F. has received a grant from Arthrex, personal fees from Elsevier, research support from Smith & Nephew, consulting fees and grants from Stryker, and education payments from Medwest and holds stock in Jace Medical. 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.

An online CME course associated with this article is available for 1 AMA PRA Category 1 Credit™ at http://www.sportsmed.org/aossmimis/Members/Education/AJSM_Current_Concepts_Store.aspx. In accordance with the standards of the Accreditation Council for Continuing Medical Education (ACCME), it is the policy of The American Orthopaedic Society for Sports Medicine that authors, editors, and planners disclose to the learners all financial relationships during the past 12 months with any commercial interest (A ‘commercial interest’ is any entity producing, marketing, re-selling, or distributing health care goods or services consumed by, or used on, patients). Any and all disclosures are provided in the online journal CME area which is provided to all participants before they actually take the CME activity. In accordance with AOSSM policy, authors, editors, and planners’ participation in this educational activity will be predicated upon timely submission and review of AOSSM disclosure. Noncompliance will result in an author/editor or planner to be stricken from participating in this CME activity.

ORCID iD

Ophelie Z. Lavoie-Gagne

‡

References 4, 8, 9, 13, 15, 16, 22, 24, 27, 31, 33, 39, .

§

References 4, 8, 9, 15, 22, 24, 27, 31, 33, 39, .

¶

References 4, 8, 9, 15, 16, 22, 24, 27, 31, 33, 39, .

References

Abdulmassih

McConkey

Amendola

. Distal femoral osteotomy. In: Cole

Sekiya

eds. Surgical Techniques of the Shoulder, Elbow & Knee in Sports Medicine. 2nd ed. Elsevier; 2013.

Backstein

Morag

Hanna

Safir

Gross

. Long-term follow-up of distal femoral varus osteotomy of the knee. J Arthroplasty. 2007;22(4)(suppl 1):2-6.

Borus

Thornhill

. Unicompartmental knee arthroplasty. J Am Acad Orthop Surg. 2008;16(1):9-18.

Cameron

McCauley

Kermanshahi

Bugbee

. Lateral opening-wedge distal femoral osteotomy: pain relief, functional improvement, and survivorship at 5 years. Clin Orthop Relat Res. 2015;473(6):2009-2015.

Chahla

Mitchell

Liechti

, et al. Opening- and closing-wedge distal femoral osteotomy: a systematic review of outcomes for isolated lateral compartment osteoarthritis. Orthop J Sports Med. 2016;4(6):2325967116649901.

Charette

Sloan

DeAngelis

Lee

. Higher rate of early revision following primary total knee arthroplasty in patients under age 55: a cautionary tale. J Arthroplasty. 2019;34(12):2918-2924.

Coventry

. Upper tibial osteotomy. Clin Orthop Relat Res. 1984;182:46-52.

Das

DHPW

Sijbesma

Hoekstra

van Leeuwen

. Distal femoral opening-wedge osteotomy for lateral compartment osteoarthritis of the knee. Open Access Surg. 2008;1:25-28.

de Carvalho

Jr Temponi

Soares

Goncalves

Costa

. Physical activity after distal femur osteotomy for the treatment of lateral compartment knee osteoarthritis. Knee Surg Sports Traumatol Arthrosc. 2014;22(7):1607-1611.

10.

DerSimonian

. Meta-analysis in the design and monitoring of clinical trials. Stat Med. 1996;15(12):1237-1252.

11.

DerSimonian

Laird

. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188.

12.

DerSimonian

Laird

. Meta-analysis in clinical trials revisited. Contemp Clin Trials. 2015;45(pt A):139-145.

13.

Dewilde

Dauw

Vandenneucker

Bellemans

. Opening wedge distal femoral varus osteotomy using the Puddu plate and calcium phosphate bone cement. Knee Surg Sports Traumatol Arthrosc. 2013;21(1):249-254.

14.

Dormann

Elith

Bacher

, et al. Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography. 2013;36(1):27-46.

15.

Ekeland

Nerhus

Dimmen

Heir

. Good functional results of distal femoral opening-wedge osteotomy of knees with lateral osteoarthritis. Knee Surg Sports Traumatol Arthrosc. 2016;24(5):1702-1709.

16.

Elattar

Swarup

Lam

, et al. Open wedge distal femoral osteotomy: accuracy of correction and patient outcomes. HSS J. 2017;13(2):128-135.

17.

Felson

Niu

Gross

, et al. Valgus malalignment is a risk factor for lateral knee osteoarthritis incidence and progression: findings from the Multicenter Osteoarthritis Study and the Osteoarthritis Initiative. Arthritis Rheum. 2013;65(2):355-362.

18.

Finkelstein

Gross

Davis

. Varus osteotomy of the distal part of the femur: a survivorship analysis. J Bone Joint Surg Am. 1996;78(9):1348-1352.

19.

Forkel

Achtnich

Metzlaff

Zantop

Petersen

. Midterm results following medial closed wedge distal femoral osteotomy stabilized with a locking internal fixation device. Knee Surg Sports Traumatol Arthrosc. 2015;23(7):2061-2067.

20.

Healy

Anglen

Wasilewski

Krackow

. Distal femoral varus osteotomy. J Bone Joint Surg Am. 1988;70(1):102-109.

21.

Insall

Dorr

Scott

. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res. 1989;248:13-14.

22.

Jacobi

Wahl

Bouaicha

Jakob

Gautier

. Distal femoral varus osteotomy: problems associated with the lateral open-wedge technique. Arch Orthop Trauma Surg. 2011;131(6):725-728.

23.

Kim

Yang

Kim

, et al. Distal femoral varus osteotomy for valgus arthritis of the knees: systematic review of open versus closed wedge osteotomy. Knee Surg Relat Res. 2018;30(1):3-16.

24.

Kolb

Isak

Hobusch

Chiari

Windhager

. Distal femoral varus osteotomy: results of the lateral open-wedge technique without bone grafting. Int Orthop. 2019;43(10):2315-2322.

25.

Kosashvili

Safir

Gross

, et al. Distal femoral varus osteotomy for lateral osteoarthritis of the knee: a minimum ten-year follow-up. Int Orthop. 2010;34(2):249-254.

26.

Learmonth

. A simple technique for varus supracondylar osteotomy in genu valgum. J Bone Joint Surg Br. 1990;72(2):235-237.

27.

Madelaine

Lording

Villa

, et al. The effect of lateral opening wedge distal femoral osteotomy on leg length. Knee Surg Sports Traumatol Arthrosc. 2016;24(3):847-854.

28.

Makambi

. The effect of the heterogeneity variance estimator on some tests of treatment efficacy. J Biopharm Stat. 2004;14(2):439-449.

29.

McDermott

Finklestein

Farine

, et al. Distal femoral varus osteotomy for valgus deformity of the knee. J Bone Joint Surg Am. 1988;70(1):110-116.

30.

Nelson

Saleh

Kassim

, et al. Total knee arthroplasty after varus osteotomy of the distal part of the femur. J Bone Joint Surg Am. 2003;85(6):1062-1065.

31.

Ramanathan

Von Keudell

Minas

Gomoll

. Survivorship and complications of the distal femoral osteotomy. Orthop J Sports Med. 2014;2(7)(suppl 2):2325967114S00051.

32.

Sabbag

Woodmass

Krych

Stuart

. Medial closing-wedge distal femoral osteotomy with medial patellofemoral ligament imbrication for genu valgum with lateral patellar instability. Arthrosc Tech. 2017;6(6):e2085-e2091.

33.

Saithna

Kundra

Getgood

Spalding

. Opening wedge distal femoral varus osteotomy for lateral compartment osteoarthritis in the valgus knee. Knee. 2014;21(1):172-175.

34.

Santaguida

Hawker

Hudak

, et al. Patient characteristics affecting the prognosis of total hip and knee joint arthroplasty: a systematic review. Can J Surg. 2008;51(6):428-436.

35.

Sherman

Thompson

Clohisy

JCF

. Distal femoral varus osteotomy for the management of valgus deformity of the knee. J Am Acad Orthop Surg. 2018;26(9):313-324.

36.

Stähelin

Hardegger

Ward

. Supracondylar osteotomy of the femur with use of compression: osteosynthesis with a malleable implant. J Bone Joint Surg Am. 2000;82(5):712-722.

37.

Sternheim

Garbedian

Backstein

. Distal femoral varus osteotomy: unloading the lateral compartment: long-term follow-up of 45 medial closing wedge osteotomies. Orthopedics. 2011;34(9):e488-e490.

38.

Tegner

Lysholm

. Rating systems in the evaluation of knee ligament injuries. Clin Orthop Relat Res. 1985;198:43-49.

39.

Thein

Bronak

Thein

Haviv

. Distal femoral osteotomy for valgus arthritic knees. J Orthop Sci. 2012;17(6):745-749.

40.

Vannini

Spalding

Andriolo

, et al. Sport and early osteoarthritis: the role of sport in aetiology, progression and treatment of knee osteoarthritis. Knee Surg Sports Traumatol Arthrosc. 2016;24(6):1786-1796.

41.

Wang

Hsu

. Distal femoral varus osteotomy for osteoarthritis of the knee. J Bone Joint Surg Am. 2006;88(suppl 1, pt 1):100-108.

42.

Wylie

Jones

Hartley

, et al. Distal femoral osteotomy for the valgus knee: medial closing wedge versus lateral opening wedge: a systematic review. Arthroscopy. 2016;32(10):2141-2147.

43.

Zarrouk

Bouzidi

Karray

, et al. Distal femoral varus osteotomy outcome: is associated femoropatellar osteoarthritis consequential? Orthop Traumatol Surg Res. 2010;96(6):632-636.