Minimally Invasive vs. Open Surgery for Hallux Valgus: A Meta-Analysis

Purpose In recent years, minimally invasive surgery (MIS) for hallux valgus has emerged and gained popularity. To date, evidence on the benefits of MIS for hallux valgus is still controversial. This updated meta-analysis aimed to comprehensively evaluate the efficiency of MIS vs. open surgery for hallux valgus. Methods A systematic literature search of PubMed, Embase, and the Cochrane Library was performed. Two independent reviewers conducted data extraction and analyzed data with R software. Data were presented with risk ratio (RR) and standardized mean difference (SMD) along with 95% confidence interval (CI). Results A total of 22 studies in which there were 790 ft treated with the MIS procedure and 838 ft treated with an open procedure were included. The correction of sesamoid position was better in the MIS group. The post-operative distal metatarsal articular angle (DMAA) of the MIS group was lower. There was less pain at the early phase in the MIS group. The MIS group had a shorter surgery time and shorter hospitalization time compared with the open group. Our meta-analysis revealed no statistically significant difference in hallux valgus angle (HVA), first intermetatarsal angle (IMA), the first metatarsal shortening, the American Orthopedic Foot and Ankle Society (AOFAS) score, visual analog scale (VAS) score at the final follow-up or complication rate (when all studies were considered). When taking into consideration only randomized controlled trial (RCT), the AOFAS score was higher in the MIS group while HVA, IMA, DMAA, and complication rate remained no significance. Post-operative IMA of the MIS group was significantly lower when only studies reporting the second-generation (2G) MIS were included. When just studies adopting the third-generation (3G) MIS were included, the HVA and DMAA were lower in the MIS group. Conclusion The MIS procedures were more effective than open surgeries in the treatment of hallux valgus. Moreover, the MIS group achieved better radiologic and clinical outcomes compared with the open group.


INTRODUCTION
Hallux valgus is a common forefoot disorder involving a lateral deviation of the hallux and medial deviation of the first metatarsal (1). Hallux valgus is often associated with first metatarsophalangeal joint osteoarthritis, and it has been linked to notable health problems, such as disability, a greater risk of falling, impaired balance, and lesser quality of daily life (2). Symptomatic hallux valgus is usually treated with surgeries such as chevron osteotomy, Lapidus procedure, and scarf osteotomy (3)(4)(5)(6)(7). There are over 150 open surgical procedures for hallux valgus, but none of them have been proven to be better than others. In recent years, minimally invasive surgery (MIS) has become popular for hallux valgus because of its shorter operation time, smaller scar, and shorter recovery time compared with open surgeries. So far, three generations of minimally invasive techniques have been developed. The first generation was the Reverdin-Isham technique (8), which used angular medial closed wedge osteotomy without osteosynthesis. The second generation (2G) was the Bösch osteotomy (9), which was a modification of Hohmann osteotomy. Temporary internal fixation with Kirschner wires were used after distal transverse osteotomy. The third generation (3G) was minimally invasive chevron and akin (MICA) (10), which used percutaneous osteotomies and was fixed with compression screws.
Recently, two meta-analyses (11,12) have been carried out to determine the effects of MIS vs. open surgery, showing no significant difference in radiological outcomes and functional scores. However, relatively a few outcome measures (OMs) had been included in Singh's meta-analysis (11). Lu et al. conducted a meta-analysis in which most of the included studies were of low or moderate quality (12). In the last 3 years, several new studies (13)(14)(15)(16)(17)(18)

MATERIALS AND METHODS
This meta-analysis was conducted in accordance with the preferred reporting items for systematic reviews and metaanalyses (PRISMA) statement (19) and the Cochrane Handbook guidelines (20).

Inclusion and Exclusion Criteria
All studies included in this meta-analysis need to meet the following criteria: (1) comparative studies reporting comparisons of MIS vs. open surgery for hallux valgus; (2) patients with hallux valgus; (3) at least a 6-month follow-up; and (4) OMs including at least one of the following: hallux valgus angle (HVA), first intermetatarsal angle (IMA), the American Orthopedic Foot and Ankle Society (AOFAS) score, visual analog scale (VAS) score, operating time, and complications. The exclusion criteria were: (1) unpublished data; (2) case series, case reports, reviews, and proceedings of meetings; (3) biomechanical research; and (4) no available data describing OMs mentioned earlier.

Data Extraction and Quality Assessment
Duplicates were initially excluded using Endnote Version X9. Two investigators independently screened titles and abstracts of the remaining studies. Then, full texts of the remaining studies were reviewed for eligibility according to the inclusion and exclusion criteria. Two independent reviewers performed data extraction from the included studies. The following data were extracted: the time of publication, country, study design, sample size, patients' age, technique, the duration of follow-up, HVA, IMA, AOFAS, and complications. Two authors used the Cochrane Handbook for Systematic Reviews of Interventions 5.2.0. (20) to assess the methodological quality and risk of bias of randomized controlled trials (RCTs), while the methodological qualities and risk of bias of non-RCTs were evaluated by the methodological index for non-randomized studies (MINORS) (21). MINORS score > 14 was set as the level of inclusion. Disagreements were resolved by a discussion to reach a consensus.

Statistical Analysis
The data analysis was performed with Review Manager (Version 5.3; The Cochrane Collaboration, Oxford, UK) and the statistical software R 4.0.3 with the meta package. For dichotomous data, the risk ratio (RR) along with 95% confidence intervals (CIs) was calculated. For continuous data, standardized mean difference (SMD) with 95% CIs was calculated. p < 0.05 indicated a statistical significance. I 2 -test was used to evaluate the heterogeneity between studies. When I 2 > 50% and p < 0.10, the heterogeneity was significant, and a random effect model was used. Otherwise, a fixed-effects model was applied in the metaanalysis. Subgroup meta-analysis was conducted according to study design (RCT or non-RCT) and technique (2G or 3G MIS). The publication bias was assessed by the Egger' test.

Characteristics of the Included Studies
As presented in Figure 1, a total of 537 articles were identified after the primary literature search. After the removal of duplicates, 276 articles remained. During the screening of titles and abstracts of the remaining studies, 150 irrelevant articles and 81 articles without a comparison group were excluded, and a total of 45 articles remained. Then, full-text reviewing according to inclusion and exclusion criteria was conducted. A total of 23 articles (including 18 case series, 3 reviews, and 2 biomechanical studies) were eliminated. Eventually, 22 studies were included in this meta-analysis, in which 790 ft were treated with MIS procedure and 838 ft with the open procedure. The characteristics of the included studies are depicted in Table 1. Follow-up times were highly variable, ranging from 6 months to 7 years. Among the included studies, 8 studies (15,17,24,26,28,29,31,35) were RCTs, 2 studies (4,14) were prospective comparative studies, and 12 studies (5, 13, 16, 18, 22, 23, 25, Figure 2. The MINORS score of comparative studies was 17.4 ± 2.0 (ranged from 15 to 21) ( Table 2), which indicated a high quality of the included studies.

Hallux Valgus Angle
A total of 21 studies reported post-operative HVA. When evaluating all studies, the pooled SMD of HVA at the final follow-up was not significant between MIS and open groups ( Figure 3A). When only studies reporting 3G MIS were included, the post-operative HVA was significantly lower in the MIS group (SMD: −0.4; 95% CI −0.68-0.13; p = 0.004; I 2 = 69%;), but this significance was lost when comparing the just studies reporting 2G MIS ( Table 3). Eight RCTs evaluated HVA, which did not reach a statistical significance (SMD: −0.08; 95% CI −0.50-0.34; p = 0.53; I 2 = 76%; Table 3).

The First Metatarsal Shortening
Three studies assessed the first metatarsal shortening, in which there were 98 ft treated with the MIS procedure and 84 ft treated with the open procedure. As shown in Figure 3D, the pooled SMD was also not significant between these two groups.  Figure 3E).

VAS Score
Eight studies were incorporated in the analysis of VAS score at the final follow-up (38). The overall results indicated no significant difference between MIS and open groups ( Figure 4B). Besides, 5 studies assessed the VAS score within post-operative 2 weeks.

Satisfaction Rate
A total of 7 studies documented the satisfaction rate. The pooled results indicated that the satisfaction rate was remarkably higher in the MIS group (RR: 1.15; 95% CI: 1.05-1.27; Z = 3.09; p = 0.002; I 2 = 0%; Figure 4D).

Complication Rate
There were some post-operative complications in MIS and open surgery for hallux valgus, such as screw irritation, recurrence, and non-union. In total, 20 studies reported complication rates. According to the pooled results, there was no difference between MIS and open groups with respect to the complication rate ( Figure 4E). Subgroup analysis of study design (RCT or non-RCT) and techniques (2G or 3G MIS) demonstrated no significance difference in the complication rate between the two groups ( Table 3).

Duration of Surgery
The duration of surgery was available in 7 studies. Figure 5A showed that the pooled SMD was statistically significant (SMD:

Publication Bias
The Egger' test of HVA suggested that there was no obvious publication bias (p = 0.58).

DISCUSSION
More than 150 open surgical procedures have been described for the treatment of hallux valgus (39). There was a trend toward more minimally invasive surgery for hallux valgus, involving the theoretical advantages of less soft tissue dissection, reduced surgical time, and a faster recovery time (3). In this meta-analysis, MIS procedure showed better radiologic and clinical outcomes compared with the open group. Compared to the former meta-analysis conducted by Singh et al. (11), 13 new studies were included in this meta-analysis, making it more comprehensive. Most of the included studies in the meta-analysis performed by Lu et al. (12) were of low or moderate quality, and limited OMs were assessed due to incomplete information. Our meta-analysis included all new studies according to the inclusion and exclusion criteria and pooled 12 OMs.
Hallux valgus angle and IMA were the most common radiologic OMs to evaluate hallux valgus because they have been shown to determine the severity of hallux valgus. IMA correction was a marker to determine the corrective potential of a metatarsal osteotomy (40). Singh et al. (11) pooled 8 studies for HVA and IMA, showing no significant difference between MIS and open groups. Our meta-analysis found no significant difference in HVA and IMA between the two groups when all 22 studies were included. Nevertheless, post-operative IMA of the MIS group was significantly lower when only studies reporting 2G MIS were included. The 2G MIS was Bösch osteotomy or the SERI technique, which adopted an axial wire to displace and maintain the metatarsal head in the initial few weeks (5). When just studies adopted 3G MIS were included, the HVA were lower in the MIS group. Lai et al. (30) performed a comparison between the PECA and open scarf/akin technique, showing a better HVA correction but a comparable IMA correction in the MIS group. The 3G MIS included MICA and PECA. PECA is a modification from the MICA procedure as introduced by Vernois and Redfern (10). The MICA technique adopted phalangeal osteotomy with burr, which had a direct impact on the correction of HVA (35). Lim et al. (16) also demonstrated lower postoperative HVA in the MIS group, and they believed that it was possibly owing to the use of toe alignment splint after the MICA procedure. DMAA is the angle between the perpendicular line to the distal metatarsal articular surface and the first metatarsal  The diameter of burrs used in MIS procedures is about 2 mm, which may lead to shorter metatarsal compared with open techniques (5). The shortened first metatarsal will transfer more load to the lateral metatarsal heads, which could cause metatarsalgia (42). This meta-analysis indicated that MIS surgery do not cause more metatarsal shortening compared with open surgery.
The change of the sesamoid position was better in the MIS group than in the open group, which perhaps was due to more use of image intensifier. Okuda et al. (43) pointed out that insufficient sesamoid position correction was an important risk factor for hallux valgus recurrence, so the correction of the sesamoid position was needed.
The AOFAS score (37) was to evaluate the functional outcome, including pain (40 points), function (45 points), and alignment (15 points). Singh et al. (11) conducted a meta-analysis to demonstrate that the open group provided more improvement in the AOFAS score. Poggio et al. (27) reported that the open scarf technique showed more increases in the AOFAS score than the Kramer technique (also known as the 2G MIS technique). Radwan and Mansour (24) and Lee et al. (29) reported a trend toward a better improvement in the AOFAS score in the MIS group. This meta-analysis pooled 17 studies and found no difference in the improvement of the AOFAS score between the two groups. When taking into consideration only RCTs, the AOFAS score was higher in the MIS group. The VAS score of the early post-operative phase was lower in the MIS group, which reflected that the MIS surgery required a minor soft tissue dissection (15). The VAS score at the final follow-up was not found to be significant, which was in accordance with Singh's meta-analysis.
Singh's meta-analysis pooled 4 studies and found no significant difference in the satisfaction rate between MIS and open groups. Lu's meta-analysis pooled 3 studies and demonstrated a higher satisfaction level in the MIS group. Our meta-analysis included 7 studies, showing a higher satisfaction rate in the MIS group. The high satisfaction rate in the MIS was possibly due to the cosmetic result of surgery. Choi et al. reported 2G MIS for young female patients, and believed that the MIS technique could be considered for young female patients who desire a less visible scar (13). The scar length was significantly shorter in the MIS group. Because of the limited scar on the medial side, MIS is expected to result in fewer soft tissue complications, less stiffness, and a higher satisfaction rate.  (44). To minimize the learning curve, cadaveric training was recommended for any surgeon considering performing MIS surgery (30).
To our knowledge, this meta-analysis includes the largest number of studies (22 studies) evaluating the efficiency of MIS vs. open surgery for hallux valgus. There were some limitations in this meta-analysis. First, different surgical techniques were used in the MIS and open groups, involving a high heterogeneity. Secondly, non-randomized controlled studies were included; therefore, the results of this study must be interpreted with caution due to the natural defects of retrospective studies. Larger sample multicenter randomized controlled studies are needed to further verify the results of this meta-analysis.

CONCLUSION
The MIS procedures were more effective than open surgery in the treatment of hallux valgus. Moreover, the MIS group achieved better radiologic and clinical outcomes compared with the open group. The 2G MIS demonstrated better corrective power to IMA while the 3G MIS provided a stronger correction to HVA. The MIS procedures offered benefits mainly in the early post-operative period, including a shorter surgery time, a more cosmetic scar, a higher satisfaction rate, and a faster recovery time. These features of the MIS procedure make it a better choice for young female patients who favor a cosmetic scar and patients who are at high risk due to recurrent ulceration.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
MZ designed this study. KW and SD conducted the literature search and data extraction. SD and CS performed quality assessment of the included studies. LJ and SS performed a statistical analysis. LJ and KW wrote this manuscript. All authors have read and approved this final manuscript.