Proximal femoral valgus osteotomy for the treatment of developmental coxa vara: a systematic review of the literature

Abstract

Background

Developmental Coxa Vara (DCV) consists on a pathological reduction in head-shaft angle (HSA) and increased femoral retroversion. Several case series reported outcomes on proximal femoral valgus osteotomy (PFVO), but no evidence synthesis had been conducted. This systematic review aimed to (1) analyze success rate and complications, (2) report the degree of correction according to the HSA and the Hilgenreiner Epiphyseal Angle (HEA), compare success rate and degree of correction of subtrochanteric (SVO) vs intertrochanteric (IVO) osteotomy, and (4) difference in success rate and correction between patients in which an internal (IF) or external fixation (EF) technique was used.

Methods

Four databases (PubMed, Scopus, Embase, and Cochrane Database of Systematic Reviews) were searched until February 20th^, 2024 according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. Studies presenting outcomes on PFVO for DCV with >2 years follow-up and >5 patients were included. Review articles, language other than English and coxa vara secondary to other conditions were excluded. Study quality was evaluated through MINORS criteria.

Results

10 case series (level of evidence IV) were located for a total of 153 patients and 192 hips. Baseline results were as follows: male/female ratio was 1.45, mean age at surgery was 6.7 ± 1.5 and follow-up 5.1 ± 5.5 years. Overall success rate was 88.8%, with failure considered as any indication to revision surgery. Complications included loss of correction (10.9%), deep (1.0%) or superficial wound infection (2.6%). Revision surgery was performed in 18 hips (9.4%). Average correction was measured through HSA (preoperative 94.6 ± 8.1, postoperative 134.4 ± 10.2, change 38.2 ± 7.5°, p < 0.001) and HEA (preoperative 71.9 ± 5.5, postoperative 31.7 ± 5.7°, change 33.7 ± 10.5°, p < 0.001). Success rate was similar between osteotomy techniques (SVO: 91.0%; IVO: 94.1%; p = 0.48) and fixation strategy (IF: 85.4%; EF 95.8%; p = 0.096).

Conclusions

PFVO presented satisfactory results for the treatment of DCV, with similar outcomes concerning the osteotomy site and fixation technique used. HSA and HEA correction were correlated to PFVO success rate. However, coxa vara is a tridimensional deformity, thus other parameters such as posterior sloping angle, mechanical axial deviation and proximal femoral offset should be included in future studies.

1. Introduction

Coxa vara refers to a pathological reduction in the neck-shaft angle¹ with an increase in femoral retroversion. Coxa vara can be secondary to trauma, infections or hormone imbalances; associated with congenital femoral deficiency; or developmental (DCV) as a result of disruptions during the growth and maturation of the hip joint.² This condition alters the biomechanics of the hip joint, resulting in limping, hip instability, pain, and limited mobility.^1,3,4 With an estimated incidence of 1 in 25,000 live births, DCV is still a rather rare condition. Around 30%–50% of patients had bilateral involvement.⁵ Physical examination often indicates length discrepancy in the femoral segment, a relatively prominent and proximally located greater trochanter, and limitations in abduction and internal rotation. The clinical presentation, the patient's symptoms, the head-shaft angle (HSA), and—perhaps most crucially—the Hilgenreiner epiphyseal angle (HEA) all serve as criteria for surgical correction.⁶ The purpose of surgery is to correct varus neck malalignment, therefore altering the loading characteristics from shear to compression forces. The improvement in hip biomechanics may induce ossification and healing of the impaired femoral neck with restoration of the HSA.⁷,⁸ Proximal femoral valgus osteotomy (PFVO) either intertrochanteric (IVO)⁷ or subtrochanteric (SVO)⁹ was documented. Osteosynthesis can be achieved through internal fixation (IF)¹⁰,¹¹ or external fixation (EF).¹²

The primary outcome of the present study is to report the (1) success rate and complications, and (2) the degree of correction according to the HSA and the HEA angles of PFVO for treatment of DCV. As secondary outcomes, we assessed the differences in failure rate and the degree of correction based (3) on the osteotomy technique used (SVO versus IVO); and (4) on the osteosynthesis technique (IF versus EF).

2. Material and methods

2.1. Search strategy

This systematic review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.¹³ To find studies concerning PFVO for the treatment of DCV, we performed a literature search on PubMed, Scopus, Embase, and Cochrane Database of Systematic Reviews databases. The following keywords, in association with the Boolean operators "AND","OR," were used in the search query to include all studies up to February 20th^, 2024: "congenital coxa vara," "developmental coxa vara," "valgus osteotomy," "coxa vara," "Hilgenreiner's epiphyseal angle," and "head-shaft angle”, “subtrochanteric valgus osteotomy”, “intertrochanteric valgus osteotomy”. See Appendix for details in data extraction.

2.2. Inclusion and exclusion criteria

Studies were eligible if the mean follow-up was no fewer than two years and if they provided detailed data on the occurrence of complications, the failure rate of osteotomies, the correction of HEA and HSA. Studies written in a language besides English, case series with less than five patients, and review articles were all excluded. Studies that involved patients who had coxa vara secondary to other conditions were also excluded from this review, including those with slipped capital femoral epiphysis, Perthes disease, trauma, infection, metabolic dysplasia, skeletal dysplasia and congenital femoral deficiency.

2.3. Data collection and extraction

Data were gathered and organized using Microsoft Excel spreadsheets by two authors (LB and GC). Throughout the process of gathering data, to determine inclusion, two authors (LB, GC) independently read the titles, abstracts, and full texts that came from the electronic database review. If there was a disagreement on a study's inclusion, a third author (AA) was consulted to decide whether to include the study in the systematic review. A reverse inclusion process was completed by looking over all the included articles' references.

2.4. Quality assessment

The results from each study were tabulated following a quality assessment using the 2011 Oxford CEBM Levels of Evidence.¹⁴

2.5. Statistical analysis

Categorical variables are portrayed through their frequencies and corresponding percentages, while continuous variables are illustrated by their respective means. Statistical significance was assessed by considering a P-value of less than 0.05, indicating a notable threshold for determining the presence of meaningful associations or differences.

A meta-analysis could not be performed due to the high heterogeneity (I²>75%) observed across the included studies.

3. Results

3.1. Study characteristics

An initial title and abstract evaluation of 122 studies were performed, followed by an overview of 12 full-text studies, and nine of these publications were included in the final analysis after reviewing the inclusion and exclusion criteria: four retrospective case series and five prospective case series (Fig. 1).¹³ A total of 153 patients (192 hips) were included in the study, no patients died or were lost to follow-up. Data on gender were available in eight studies (140 patients)¹⁰,¹¹,¹⁵–,²⁰ there were 83 men (59.3%) and 57 women (40.7%). The average age at the time of surgery was 6.7 years (range = 3.1 to 10.1; SD = 1.5), while the average follow-up was 5.1 years (range = 1.7 to 20; SD = 5.5). In five studies a subtrochanteric valgus osteotomy was performed in all cases (78 hips, 40.6%),¹⁰,¹²,¹⁵,¹⁶,¹⁸ while in four studies (68 hips, 35.4%) an intertrochanteric valgus osteotomy was routinely performed,¹¹,¹⁷,¹⁹,²¹ in one study both techniques were used (46 hips, 24.0%).²⁰ An internal fixation with plate and screws was used to fix the osteotomy in six studies (144 hips, 75.0%)¹⁰,¹¹,¹⁵,¹⁷,¹⁹–,²¹ while an external fixation was adopted in three studies (48 hips, 25.0%)¹²,¹⁶,¹⁸ (Table 1).

Fig. 1.

PRISMA flow diagram. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Table 1.

Included studies and general characteristics.

First Author (YoP)	Patients/Hips	Gender (male/%)a	Age at the time of surgeryb(years)	Follow-up lengthb(years)	Osteotomy technique	Fixation method	Healing time (months)	HSA (preop/postop)	HEA (preop/postop)	Success of the procedure No (%)
Desai (1993)	12/20	6 (50%)	7 ± 2	20 (6–41)	SVO	IF	4.5	96°/130°	66°/30°	15 (75%)
Yang (1997)	4/6	n/a	7.8 ± 3.1	4,2 (2–8.3)	IVO	IF	n/a	96°/131°	66°/24°	6 (100%)
Sabharwal (2005)	5/6	3 (60%)	8.4 (5–11)	2.1 (0.4–3.8)	SVO	EF	n/a	86°/137°	74°/33°	6 (100%)
Abdelaziz (2012)	26/31	18 (69.2%)	5.6 (4–9)	4.6 (3–9)	IVO	IF	2	n/a	66.5°/33.5°	27 (87.1%)
Hefny (2012)	9/9	n/a	10.1 (6–16)	4.2 (2–8)	SVO	EF	4	82°/132.3°	75.2°/41.3°	8 (88.9%)
Elzohairy (2016)	18/18	12 (66.7%)	6.1 (5–7)	2.5 (2–3)	IVO	IF	2.4	93.7°/129.9°	78.2/27.8°	18 (100%)
Galal (2017)	30/33	16 (53.3%)	7.2 (3–15)	1.7 (1–2.9)	SVO	EF	3	87.5°/124°	70.6°/41.2°	32 (97%)
Rizk (2017)	7/13	4 (57.1%)	3.1 ± 1.2	2 ± 0.6	IVO	IF	2.9	86.9°/139.6°	84.1°/24.6°	13 (100%)
Bian (2019)	32/46	18 (56.3%)	7.3 (3–13)	4.7 (2–13)	SVO and IVO	IF	n/a	105.1°/147.6°	68.8°/29.8°	34 (73.9%)
El-sobki (2021)	10/10	6 (60%)	5.1 ± 4.9	2.5 (1.4–3.3)	SVO	IF	3.8	94°/120°	76.5°/26.5°	10 (100%)
Total/Average ± SD	153/192	83 (59.3%)	6.7 ± 1.5	5.1 ± 5.5	SVO (78; 40.6%), VIO (68; 35.4%), n/a (46; 24.0%)	IF (144; 75.0%) EF (48; 25.0%)	3.0 ± 2.0	94.6 ± 8.1°/134.4 ± 10.2	71.9 ± 5.5°/31.7 ± 5.7°	169 (88.0%)

YoP = Year of Publication; SVO = Subtrochanteric Valgus Osteotomy; IVO = intertrochanteric Valgus Osteotomy; IF = internal fixation; EF = external fixation.

^aHefny et al. and Yang et al. did not report gender distribution. The percentage of male/female was calculated on 140 patients.

^b(standard deviation or range).

There were four retrospective studies,¹⁵,¹⁶,²⁰,²¹ and five prospective studies¹⁰–¹²,¹⁷–.¹⁹ All the studies were Level of Evidence 4 according to the 2011 Oxford CEBM Levels of Evidence.¹⁴ The average MINORS score was 10.3. All the studies were categorized as “poor quality”, with a MINORS score <15 (Table 2).

Table 2.

Quality assessment and MINORS scores.

Author	A clearly stated aim	Inclusion of consecutive patients	Prospective collection of data	Endpoint appropriate for aim of study	Unbiased assessment of study endpoint	Follow-up period appropriate	Loss of follow-up less than 5%	Prospective calculation of study size	Ov.
Desai (1993)	2	1	1	2	0	2	2	0	10
Yang (1997)	1	1	1	2	0	2	2	0	9
Sabharwal (2005)	1	1	1	2	0	2	2	0	9
Abdelaziz (2012)	2	2	2	2	0	2	2	0	12
Hefny (2012)	1	2	2	2	0	2	2	0	11
Elzohairy (2016)	2	1	2	2	0	2	2	0	11
Galal (2017)	2	2	2	2	0	2	2	0	12
Bian (2019)	2	1	1	2	0	2	2	0	10
El-sobki (2021)	1	1	1	2	0	2	2	0	9
Rizk (2017)	2	1	2	2	0	2	2	0	11

Ov. = Overall.

Poor quality 9–14; High quality > 15.

3.2. Success rate and complications

The overall success rate of PFVO for the management of DCV was 88.8% (169/192 hips), considering failure as any indication for surgical revision (Table 3).

Table 3.

Reported complications and overall failure rate.

First Author (YoP)	Hips	Osteotomy	Fixation Method	Heterotopic ossifications	Joint nail penetration	Deep Infection	Superficial wound infection	GT Apophyseodesis failure	Hypertermia	AVN	Femural fracture	Loss of correction	Revision surgery	Failure
Desai (1993)	20	SVO	IF	0 (0%)	1 (5%)	1 (5%)	0 (0%)	1 (5%)	1 (5%)	0 (0%)	0 (0%)	3 (15%)	5 (25%)	5 (15%)
Yang (1997)	6	IVO	IF	0 (0%)	n/a	n/a	n/a	n/a	n/a	n/a	n/a	0 (0%)	0 (0%)	0 (0%)
Sabharwal (2005)	6	SVO	EF	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Abdelaziz (2012)	31	IVO	IF	0 (0%)	0 (0%)	0 (0%)	3 (9.7%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	4 (12.9%)	3 (9.7%)	4 (12.9%)
Hefny (2012)	9	SVO	EF	0 (0%)	0 (0%)	1 (11.1%)	2 (22.2%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	1 (11.1%)	1 (11.1%)	1 (11.1%)
Elzohairy (2016)	18	IVO	IF	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Galal (2017)	33	SVO	EF	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	1 (3%)	1 (3%)	1 (3%)	1 (3%)
Rizk (2017)	13	IVO	IF	1 (7.7%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Bian (2019)	46	Both	IF	0 (0%)	n/a	n/a	n/a	n/a	n/a	n/a	n/a	12 (26.1%)	9 (19.6%)	12 (26.1%)
El-sobki (2021)	10	SVO	IF	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Total	192			1 (0.5%)	1 (0.5%)	2 (1%)	5 (2.6%)	1 (0.5%)	1 (0.5%)	0	1 (0.5%)	21 (10.9%)	18 (9.4%)	23 (12.0%)

A total of 23 hips (11.2%) were required surgical revision and were classified as failures: 21 hips (10.9%) experienced a loss of correction and 2 hips (1.0%) developed a deep infection. Within the first subgroup (hips with loss of correction), revision surgery was not performed on 3 hips due to refusal by patients or their families. In the second subgroup (hips with deep infection), one hip underwent debridement, antibiotics, and implant retention, while the other required the removal of an infected pin from the external fixator. Additionally, 5 cases (2.6%) of superficial wound infections were reported, with all instances treated with intravenous antibiotics. A single case of femoral fracture, managed with closed reduction and cast immobilization, joint nail penetration, and greater trochanter epiphysiodesis failure, was documented (0.6%). These cases did not necessitate revision surgery and did not affect the final correction. Moreover, no cases of avascular necrosis of the femoral head were reported.

3.3. Correction of the deformity

The average preoperative and postoperative HSA was reported in all except one study¹⁷ (161 hips, 83.9%). The average preoperative HSA was 94.6° (range = 82°–105.1°; SD = 8.1), while the average postoperative HSA was 134.4° (range = 120° to 147.6; SD = 10.2). The difference between preoperative and postoperative HSA was statistically significant (p < 0.001). The average change of HSA was 38.2° (range = 26°–52.7°; SD = 7.5).

The average preoperative and postoperative HEA was reported in all studies (192 hips, 100%). The average preoperative HEA was 71.9° (range = 66 to 84.1; SD = 5.5), while the average postoperative HEA was 31.7° (range = 24–41.3°; SD = 5.7). The difference between the preoperative and postoperative HEA was statistically significant (p < 0.001). The average change of HEA was 33.7° (range = 29.4°–59.5°; SD = 10.5).

3.4. Difference between SVO and IVO

In five studies a SVO was performed in all cases (78 hips, 40.6%),¹⁰,¹²,¹⁵,¹⁶,¹⁸ while in three studies (68 hips, 35.4%) an IVO was routinely performed,¹¹,¹⁷,¹⁹,²¹ in one study both techniques were used (46 hips, 24.0%) but the results were reported together, and this study was excluded from this analysis²⁰ (Table 4). Patients managed with SVO and IVO showed a similar success rate, respectively 91.0% (71 cases) and 94.1% (64 cases), the difference was not statistically significant (p = 0.48). The two groups differ significantly for the average age at the time of surgery (higher in SVO; p < 0.001), follow-up length (longer in SVO; p < 0.001), average postoperative HSA (higher in IVO; p < 0.001), change in HSA (higher in IVO; p < 0.001), average postoperative HEA (lower in IVO; p < 0.001), the average change of HEA (greater in IVO; p < 0.001), and healing time (shorter in IVO; p < 0.001). No other statistically significant differences were observed for the other variables (gender distribution, average preoperative HSA, average postoperative HSA, average preoperative HEA, and postoperative HEA).

Table 4.

Comparison between subtrochanteric valgus osteotomy and intertrochanteric valgus osteotomy.

Variable	SVO (n = 66; hips = 78)	IVO (n = 55; hips = 68)	P-value	Significant
Age	7.3 ± 1.5	5.4 ± 1.5	<0.001	yes
Fu	6.8 ± 8.7	3.1 ± 1.4	<0.001	yes
Gender (M)	31 (54.%)	39 (66.6%)	0.19	no
Success rate	71 (91.0%)	64 (94.1%)	0.48	no
Preop HSA	89.8 ± 5.3	91.7 ± 4.4	0.61	no
Postop HSA	127.0 ± 5.4	133.5 ± 0.6	<0.001	yes
Change HSA	37.2 ± 8.2	41.8 ± 0.9	<0.001	yes
Preop HEA	71.0 ± 4.0	72.9 ± 8.5	0.08	no
Postop HEA	35.8 ± 6.7	29.5 ± 0.5	<0.001	yes
Change HEA	35.2 ± 7.5	43.5 ± 2.3	<0.001	yes
Healing time	3.7 ± 0.7	2.3 ± 0.1	<0.001	yes

n = patients; SVO = Subtrochanteric Valgus Osteotomy; IVO = Intertrochanteric Valgus Osteotomy; HSA = Head-shaft-angle; HEA = Hilgenreiner Epiphyseal Angle.

3.5. Difference between internal and external fixation

An internal fixation with plate and screws was used to fix the osteotomy in six studies (144 hips, 75.0%)¹⁰,¹¹,¹⁵,¹⁷,¹⁹–,²¹ while an external fixation was adopted in three studies (48 hips, 25.0%)¹²,¹⁶,¹⁸ (Table 5). Patients treated with an internal fixation device showed a lower success rate compared to patient managed with an external fixation device, which was respectively 85.4% (123 cases) and 95.8% (46 cases). However, the difference was not statistically significant (p = 0.096). The two groups differed significantly for the average age at the time of surgery (higher in EF, p < 0.001), follow-up length (longer in IF, p < 0.001), preoperative and postoperative HSA (greater in IF, p < 0.001), postoperative HEA (higher in EF, p < 0.001), and average change in HEA (higher in IF, p < 0.001). No statistically significant differences were reported for gender distribution, the average change in HSA, preoperative HEA, and healing time.

Table 5.

Comparison between internal fixations and external fixation osteosynthesis.

Variable	IF (n = 109; hips = 144)	EF (n = 44; hips = 48)	P-value	Significant
Age	6.2 SD 1.2	7.9 SD 1.4	< 0.001	yes
Fu	6.1 SD 6.1	2.1 SD 1.2	< 0.001	yes
Gender (M)	64 (61.0%)	19 (54.3%)	0.110	no
Success Rate	123 (85.4%)	46 (95.8%)	0.096	no
Preop HSA	98.0 SD 6.6	86.3 SD 2.6	< 0.001	yes
Postop HSA	136.3 SD 10.8	127.2 SD 6.0	< 0.001	yes
Change HSA	38.4 SD 7.5	40.9 SD 8.0	0.075	no
Preop HEA	70.9 SD 6.3	71.9 SD 2.9	0.290	no
Postop HEA	29.4 SD 3.0	40.2 SD 5.4	< 0.001	yes
Change HEA	41.5 SD 8.8	31.7 SD 4.8	< 0.001	yes
Healing time	4.2 SD 3.5	3.2 SD 0.6	0.0686	no

IF = Internal Fixation; EF = External Fixation; HSA = Head-shaft-angle; HEA = Hilgenreiner's Epiphyseal Angle.

4. Discussion

The primary outcomes of this study were to analyze the overall success rate and related complications of PFVO in DCV, and the degree of correction of HSA and HEA.

PFVO was performed in 153 patients for a total of 192 treated hips. Among them, 169 hips (88.0%) were considered successful with a mean follow-up of 5.1 years.

Primarily, surgical failure was attributed to the loss of correction, followed by two instances requiring revision due to deep infection. The risk of recurrence had been closely linked to the degree of undercorrection, as quantitatively assessed through the HEA.¹,⁷,¹⁵,22, 23, 24 Bian et al.²⁰ highlighted that 12 hips (26.1%) experienced a recurrence of varus deformity, necessitating reoperation in 9 cases. Their univariate analysis further established that both undercorrection, as determined by measurements of the HEA and head-shaft angle (HSA), and age below 6.5 years significantly contributed to the likelihood of recurrence.²⁰ Amstutz et al.²⁴ reinforced the observation that younger patients were particularly susceptible to higher recurrence rates and a greater necessity for subsequent surgeries. Reports indicated that older children, especially those with premature closure of the growth plate following valgus osteotomy, exhibited a reduced risk of recurrence.⁴,²⁰ This pattern underscored the critical influence of the physis in surgical outcomes, suggesting that the extensive capacity for bone remodeling present in younger children may have amplified the risk of recurrence when adequate HEA was not achieved initially.

However, radiographic outcomes were just analyzed through the HSA and HEA. Although these parameters have a renowned diagnostic and prognostic value when treating coxa vara,⁶ they are not able to describe the deformity in the axial plane. Bian et al.²⁰ included the Posterior Sloping Angle (PSA) in their analysis as 15% of patients had recurrence despite satisfactory HEA correction. Among these, 59% were found with an increased preoperative PSA. Therefore, valgus flexion osteotomy was indicated to address the antero-posterior deformity.

Moreover, any osteotomy for angular correction should include axial translation to avoid mechanical axial deviation (MAD).²² Concerning proximal femoral valgus osteotomy, the femoral diaphysis should be translated laterally to preserve as much proximal femoral offset as possible and to limit valgus deviation of the knee's joint line. Bian et al.²⁰ reported a higher risk of recurrence (75%) in patients with a negative proximal femoral offset. According to the Hueter-Volkmann law,²³ compressive forces tend to limit the physeal growth, while tension forces stimulate it. The medialization of the femoral diaphysis may alter the forces applied on the proximal femoral physis, thus compromising the angular correction obtained over time.

Future studies should include preoperative and postoperative PSA and femoral offset to allow comparative analysis and further investigate their prognostic role in PFVO for the treatment of coxa vara.

The secondary outcome was firstly to compare patients treated either by SVO or IVO. Even though patients did not differ concerning gender distribution, IVO was performed in significantly younger patients than SVO. Moreover, the degree of correction achieved through IVO (change of postoperative HSA and HEA) resulted significantly higher with respect to the SVO group. As stated above, higher correction in IVO group may be planned by the operating surgeon because of the risk of recurrence in younger populations. No patients required revision surgery for delayed or non-union at the site of osteotomy. However, bone healing resulted significantly faster in IVO group. Overall, success rate between IVO (94.1%) and SVO (91.0%) resulted non-significant at statistical analysis.

Then, patients treated either by IF or EF technique were compared. Patients in the IF group resulted significantly younger than in the EF group. As discussed in the previous paragraph, the younger age group (IF) reported significantly higher corrections of HSA and HEA for higher risk of recurrence. Even though EF group reported higher success rate (95.8%) than the IF group (85.4%) with faster healing time (EF = 3.2 months; IF = 4.2 months), no differences were found at statistical analysis.

While this systematic review provided valuable insights into the treatment of DCV with PFVO, several limitations must be acknowledged. A particular challenge was the existing lack of a standardized classification system. The decision to exclude patients with coxa vara secondary to other conditions was made to avoid introducing heterogeneity into our study sample. While this approach allowed for a focused analysis on primary DCV, our findings may not be applicable to coxa vara secondary to underlying pathologies. Moreover, the exclusion of non-English studies may have led to a language bias. Then, all the included studies were categorized as “poor quality” according to MINORS criteria, suggesting that the current literature may not have offered robust evidence. The reliance on case-series with Level 4 evidence¹⁴ underscored the need for high-quality studies on the subject. Lastly, only two studies reported pre- and postoperative PROMs.¹⁰,¹⁹ Therefore, it was decided not to include this data in the analysis.

5. Conclusions

PFVO reported satisfactory outcomes in treating DCV with an overall success rate of 88.0% across 192 treated hips and an average follow-up of 5.1 years. Younger age and under correction of HSA and HEA were associated with a heightened risk of recurrence. Additionally, our findings revealed no significant differences in success rates between SVO or IVO techniques and between IF or EF osteosynthesis methods.

However, most studies reported only HEA and HSA. Future research should include parameters like the PSA, MAD, and proximal femoral offset to enrich our understanding of the three-dimensional aspects of the deformity and their impact on postoperative results. Furthermore, the incorporation of validated PROMs in future studies is crucial for a more thorough analysis of PFVO's clinical efficacy.

Ethics statement

Ethics approval was not required for this manuscript as it is a systematic review of published papers that have sought ethics approval from their respective committees/departments.

Funding and conflict of interest

No funding was received for this project and there are no conflicts of interest to declare.

RE patient consent

Consent was not required for this manuscript as it did not directly obtain any new patient information or data.

CRediT authorship contribution statement

Andrea Audisio: Methodology, Writing – original draft, Data curation, Investigation. Giorgio Cacciola: Conceptualization, Writing – original draft, Data curation, Investigation. Lorenzo Braconi: Data curation, Investigation. Carmelo Giudice: Writing – review & editing. Alessandro Massè: Supervision, Writing – review & editing. Alessandro Aprato: Project administration, Writing – review & editing, Formal analysis.

Declaration of Competing interest

The authors declare that they have no conflicts of interest related to this work and received no financial support or funding for the research, authorship, and/or publication of this article.

Appendix.

Medline/PubMed search strategy

Population

"congenital coxa vara" [MeSH Terms] OR "developmental coxa vara" [All Fields] OR "coxa vara" [All Fields]

Intervention

"valgus osteotomy" [All Fields] OR "subtrochanteric valgus osteotomy" [All Fields] OR "intertrochanteric valgus osteotomy" [All Fields]

Outcome

"Hilgenreiner's epiphyseal angle" [All Fields] OR "head-shaft angle" [All Fields]

Combined Search Strategy

(("congenital coxa vara" [MeSH Terms] OR "developmental coxa vara" [All Fields] OR "coxa vara" [All Fields]) AND ("valgus osteotomy" [All Fields] OR "subtrochanteric valgus osteotomy" [All Fields] OR "intertrochanteric valgus osteotomy" [All Fields])) AND ("Hilgenreiner's epiphyseal angle" [All Fields] OR "head-shaft angle" [All Fields])

Cochrane Library search strategy

Population

"congenital coxa vara" OR "developmental coxa vara" OR "coxa vara" OR MeSH descriptor: [Coxa Vara] explode all trees.

Intervention

"valgus osteotomy" OR "subtrochanteric valgus osteotomy" OR "intertrochanteric valgus osteotomy" OR MeSH descriptor: [Osteotomy] explode all trees.

Outcome

"Hilgenreiner's epiphyseal angle" OR "head-shaft angle"

Combined Search Strategy for the Cochrane Library

(("congenital coxa vara" OR "developmental coxa vara" OR "coxa vara" OR MeSH descriptor: [Coxa Vara] explode all trees) AND ("valgus osteotomy" OR "subtrochanteric valgus osteotomy" OR "intertrochanteric valgus osteotomy" OR MeSH descriptor: [Osteotomy] explode all trees)) AND ("Hilgenreiner's epiphyseal angle" OR "head-shaft angle")

Embase search strategy

Population

'congenital coxa vara'/exp OR 'developmental coxa vara' OR 'coxa vara'

Intervention

'valgus osteotomy'/exp OR 'subtrochanteric valgus osteotomy' OR 'intertrochanteric valgus osteotomy'

Outcome

'Hilgenreiner's epiphyseal angle' OR 'head-shaft angle'

Combined Search Strategy for Embase

(('congenital coxa vara'/exp OR 'developmental coxa vara' OR 'coxa vara') AND ('valgus osteotomy'/exp OR 'subtrochanteric valgus osteotomy' OR 'intertrochanteric valgus osteotomy')) AND ('Hilgenreiner's epiphyseal angle' OR 'head-shaft angle')