Modified Brostrom-Gould surgical procedure for chronic lateral ankle instability compared with other operations: a systematic review and meta-analysis

Abstract

Background

This study performed a randomized trial data meta-analysis to assess The Modified Brostrom-Gould (MBG) for proven chronic lateral ankle instability (CLAI).

Methods

All published randomized clinical trials comparing MBG and other operations were found by searching the Cochrane Library, EMBASE, and PubMed databases. The Review Manager 5.4 software was used to compare the two groups regarding postoperative functional score, ankle stability, and complications. Risk Ratio (RR) and Mean Differences (MD) were used in meta-analyses.

Results

8 experiments are suitable for it, 426 patients were enrolled, and 222 patients underwent other operations surgery. Among the six outcome indicators, in terms of FAOS scores, the other operations group has an advantage, 6.53 points higher than MBG; others show no significant differences.

Conclusions

Based on this meta-analysis, the authors believe that other surgical groups can achieve better outcomes than MBG in some aspects of CLAI treatment.

Introduction

Ankle sprains are one of the most prevalent sports injuries, accounting for 15% to 20% of all injuries [1]. While most patients can return to their everyday lives after nonsurgical therapy, some persons suffer CLAI. Patients with sprained ankles often undergo extensive rehabilitation training before being diagnosed. However, more than half of the patients will eventually develop CLAI, leading to talar cartilage injury and early ankle osteoarthritis [2].

The repair of ligaments in postoperative patients depends on the tissues that have been injured; Such conditions require surgical treatment to improve clinical symptoms, improve the ability of daily living, and improve prognosis. MBG is a surgical method for treating CLAI and has long been the first-line choice. However, in recent years, for the treatment of this type of patient, some new surgical methods have emerged, such as LARS, Suture-tape augmentation, Arthroscopic suture-tape internal bracing, the Chrisman-Snook procedure, etc. [3,4,5,6,7]. Many scholars have compared these surgical methods with MBG and conducted RCTs for many years. For example, in 2015 and 2019, Porter M compared the two surgical methods of LARS and MBG through randomized controlled trials. Through up to 2 years and 5 years of follow-up, he found that LARS performed better in FAOS and Tenger activity scores [4, 7]; Kulwin in the 2021 article, showed that through a two-year randomized controlled trial, it was found that the suture tape can reach preinjury level of activity (RTPAL) faster and have a lower chance of complications compared with MBG. In this article, we grouped them into other surgical groups to explore how these surgical modalities differed from the MBG group. The specific surgical steps are shown in Table 1. Many scholars have compared these surgical methods with MBG and conducted RCTs for many years. The trial period included in the previous meta-analysis was relatively long. There is still a lack of evidence to support whether MBG can be the first choice for the treatment of CLAI. New randomized controlled trial evidence has emerged recently, so a new meta-analysis is needed. Analyze Other operations in-depth to give evidence-based support for medical decisions.

Table 1 The specific surgical procedure

This study aims to demonstrate that other operations can achieve better outcomes than MBG in treating certain CLAIs, not only insisting on the idea that MBG is the gold standard for treating CLAI. To our knowledge, this is the first meta-analysis to incorporate recent randomized controlled trials evaluating both MBG and many procedures.

Material and method

Study selection and search strategy

This meta-analysis was implemented by the PRISMA guidelines (the preferred reporting item for systematic reviews and meta-analysis) [8]. the ID is CRD42021248704. As of March 1, 2021, we searched various electronic databases, including the Cochrane Library, PubMed, and Embase. Each database was independently searched by two reviewers (Qifan Yang and Jing Liu) using the following research methodology: ((((((Ankle[MESH]) OR (Ankle[Title/Abstract])) OR (Lateral Ligament, Ankle[MESH])) OR (Lateral Ligament, Ankle[Title/Abstract])) OR (ankle*[Title/Abstract] AND (injur*[Title/Abstract] OR joint*[Title/Abstract] OR ligament*[Title/Abstract] OR sprain*[Title/Abstract] OR strain*[Title/Abstract] OR inversion*[Title/Abstract] OR rupture*[Title/Abstract] OR tear*[Title/Abstract] OR torn)[Title/Abstract])) AND ((Ligaments, Articular[MESH]) OR (Ligaments, Articular[Title/Abstract]))) AND (((((instability[Title/Abstract] OR unstable[Title/Abstract] OR lax*[Title/Abstract] OR recurrent[Title/Abstract] OR chronic*[Title/Abstract]) OR (Joint Instability[MESH])) OR (Joint Instability[Title/Abstract])) OR (Chronic Disease[MESH])) OR (Chronic Disease[Title/Abstract])) AND (randomized controlled trial[Filter]). The senior author will resolve any disagreements (Dong Zhu). We looked at the title and abstract of each article we found and then read the complete text of the studies that fulfilled the criteria. Also, I looked through the reference list of the included literature to see whether there was any research that met the criteria. There are no limitations on the location of study, the type of research, or the status of publications.

Eligibility criteria

All prospective randomized controlled trials evaluating the clinical outcomes of other operations vs. MBG in the treatment of CLAI are included. (1) Randomized, double-blind, placebo-controlled trials; (2) Studies with at least one of the following clinical outcomes: Ankle Outcome Score (FAOS), Ankle Joint Stability, Complications;(3) Research in full text is available for reading. Studies with patients who have had any prior ankle surgery, studies without full text, and studies with more than 20% of patients lost to follow-up are excluded.

Study quality assessment

Two reviewers (Qifan Yang and Jing Liu) independently evaluated the quality of identified RCTs using the Cochrane Library's risk of bias [9], which included seven items: random number generation, concealment of allocation schemes, blinding of experimenters and participants, non-conforming result data, selective reporting, and other deviations. The studies were all categorized as low, unclear, or high risk. The senior author will decide on any differences between the two reviewers.

Data extraction

Two reviewers (Qifan Yang and Jing Liu) independently collected all relevant data from the study in a predetermined way. Data in other formats (such as mean and range) should be translated to mean ± SD using the Cochrane Handbook Standard Deviation guidelines [10]. The article data includes the first author's name, the year of publication, the study title, the sample size, and the follow-up date. The most prevalent demographic data points are average age, gender, and diagnosis. Clinical prognostic indicators include Foot and Ankle Prognosis Score (FAOS), ankle joint stability and complications. The senior author decides whether there is a disagreement between the two reviewers.

Statistical analysis

Review Manager (Revman, Version 5.4, The Cochrane Collaboration) was used to analyze the data. For continuous data, mean difference (MD) and 95% CI intervals (95% CI) were calculated, whereas, for dichotomous data, odds ratios (ORs) and 95% CI were used. I² tests were employed to determine heterogeneity, with I² > 50% and P < 0.10 indicating considerable heterogeneity. When significant heterogeneity was identified, a random-effect model was used; otherwise, a fixed-effect model was used. Statistical significance was defined as a P value of less than 0.05.

Results

Litterateurs review

After an initial search of relevant databases, 130 papers meeting the essential screening criteria were found, as shown in Fig. 1. The inclusion and exclusion criteria were applied to 22 research, and duplicate papers were deleted. Then 22 full texts were assessed for eligibility, and 8 RCTs [4, 5, 7, 10,11,12,13,14] with several 426 patients were included in this meta-analysis.

Fig. 1

Study selection pipeline diagram for PRISMA (Preferred Reporting Items for Systemic Meta-Analyses)

Studies characterization

Table 2 illustrates the general information about the studies that were included. All of the randomized controlled trials were published between 1994 and 2021, with a one minimum follow-up time. A total of 426 patients with CLAI were included in the study, with 222 receiving other operations treatment and 204 receiving MBG treatment. The average age of the two groups was not significantly different (P > 0.05). Figure 2 summarizes the findings of the quality assessment.

Table 2 Baseline data from the studies that were included

Fig. 2

The Cochrane Library for RCTs' Risk of Bias test was used to assess the quality of the RCTs that were included. a Author judgments on each risk of bias item for each included study, and b percentages for each risk of bias item across all included studies

Meta-analysis of outcome

FAOS – Pain, symptoms, activities of daily living (ADLs), sport, quality of life (QOL), and total scores

Pain

The FAOS pain score with SD was studied in four trials [4, 10, 14, 17], totaling 204 participants. The results showed that there was no considerable disparity (P = 0.43, I² = 0%). Hence the data was analyzed using the random-effects model. There was a significant difference in pain ratings between the other operations and MBG groups (MD = -4.00, 95% CI: -6.08— -1.91, P = 0.0002) in the summary data. The other operations group had a better clinical score influence; Then, according to follow-up time, the other operations group had a better clinical scoring effect. The data was broken down into subgroups and analyzed. First, 149 individuals in three studies [4, 14, 17]reported the FAOS pain score after being monitored for a year. Since there was no significant heterogeneity (P = 0.16, I² = 46), the findings were analyzed using the random-effects model. The pain scores of the Other operations and MBG groups were significantly different (MD = -6.24, 95% CI: -9.32- -3.16, P < 0.0001); When the follow-up time was two years, three items were significantly different. The study [10, 14, 17] included 143 patients who had FAOS with SD pain scores. The findings show no significant heterogeneity (P = 0.31, I² = 13%), and the data analysis is conducted using a random-effects model. The Other operations group outperformed the MBG group (MD = -4.46, 95% CI: -6.90- -2.02, P = 0.0003) (Fig. 3a). The authors believe that the difference in FAOS scores is due to the use of tendons or suture tape to mimic the torn ligament's original physiological role and enhance ankle stability with more muscular tissue or material, thereby causing the difference in FAOS results.

Fig. 3

Between MBG and Other operations, a forest plot of different sections of the Foot and Ankle Outcome Score (FAOS) was created. In the elements of Pain (a), Symptoms (b), Activities of daily living (ADL) (c), Sports (d), Quality of life (QOL) (e), and Total scores (f), the forest plot demonstrates that the summary points are slanted towards other operations

Symptoms

The symptom elements of the FAOS score were reported as MD ± SD in four studies (204 individuals) [4, 10, 14, 17]. Figure 3b depicts the results. Heterogeneity is statistically significant (P < 0.00001, I² = 83%). The data analysis is completed using the random-effects model, and there is a distinction between table the two groups (MD = -7.91, 95%CI: -15.07- -0.76, P = 0.03). Subgroup analysis based on follow-up time was used to determine the FAOS symptom score and SD; two studies [4, 17] with 87 patients reported the FAOS symptom score and SD at one year. Because there was no significant heterogeneity (P = 1.00, I² = 0%), the data were analyzed using the random-effects model. According to the summary data, there was a significant difference in symptom scores between the Other operations and MBG groups (MD = -9.50, 95%CI: -11.96- -7.03, P < 0.00001). After a two-year follow-up period, 143 patients were involved in three trials [4, 10, 17]. Because of the substantial heterogeneity (P = 0.0007, I² = 86%), the data were analyzed using the random-effects model. The Other operations group had a higher score (MD = -8.20, 95% CI: -14.45- -1.95, P = 0.01) than the MBG group.

ADLs

Four articles [4, 10, 14, 17] included 204 patients and provided the results of the FAOS ADL component in the form of MD ± SD for analysis. Due to the strong heterogeneity of the results (P = 0.003, I² = 66%), the random-effects model was used in the analysis, and the results revealed that the difference between the groups was not statistically significant (MD = -4.29, 95% CI: -10.29- 1.72, P = 0.18). Following that, a subgroup analysis was performed. After a year of follow-up, 88 patients in two trials [4, 17] revealed no significant heterogeneity in the ADLS score of FAOS (P = 1.00, I² = 0%), and the data analysis was randomized Effect model. The ADLS scores of the Other operations and MBG groups were substantially different (MD = -7.60, 95%CI: -3.89- -11.71, P < 0.00001), according to the summary results. Three trials [4, 10, 17] involved 143 patients when the period was two years. Because the data is highly heterogeneous (P = 0.03, I² = 72%), the random-effects model is used to examine it. According to the summary data, the difference in ADLS scores between the MBG and Other operations groups was not statistically significant (MD = -4.26, 95%CI: -10.41- 1.90, P = 0.18). (Fig. 3c).

Sport

Data from four studies [4, 10, 14, 17] are provided for the FAOS sports scores. Because of the considerable heterogeneity (P < 0.00001, I² = 91%), we applied a random-effects model to examine the data and discovered that there was a statistical difference between the Other operations and MBG groups (MD = 7.69, 95%CI: 0.26- 14.93, P < 0.00001). For various periods, a subgroup assessment was conducted. A total of 88 patients in two investigations [4, 17] reported FAOS exercise scores after a year of follow-up, with no significant heterogeneity in the results (P = 1.00, I² = 0%). The data were analyzed using a random-effects model, which indicated a significant variation in Sports ratings between the Other operations and MBG groups (MD = 12.10, 95% CI: 10.00–14.20, P < 0.00001). The follow-up period was two years, with three items. There were 143 patients in the study [4, 10, 17]. It has significant heterogeneity (P < 0.00001, I² = 92%), and data analysis is done using the random-effects model. The Other operations group outperformed the MBG group in Sports scores (MD = 10.09, 95% CI: 1.54- 18.65, P = 0.02), according to the summary data (Fig. 3d). The authors believe that the difference in FAOS scores is due to the use of tendons or suture tape to mimic the torn ligament's original physiological role and enhance ankle stability with more muscular tissue or material, thereby causing the difference in FAOS results.

QOL

In four various studies, the FAOS QOL score was recorded. [4,5,6,7] Because the variability was significant (P < 0.00001, I² = 92%), a random-effects meta-analysis was used, which can be seen in Fig. 3e. There was a significant difference between the two groups (MD = 9.14, 95% CI: 0.87–17.41, P = 0.03). After a one-year follow-up period, two studies [4, 17]with 88 patients reported the FAOS QOL score with SD. The data were examined using a random-effect model because there was no substantial variation (P = 1.00, I² = 0%). There was a significant difference between the Other operations and MBG groups in terms of Sports score (MD = 12.10, 95%CI: 10.00–14.20, P < 0.00001). Three investigations [4, 10, 17] involving 143 patients were conducted after two years of follow-up. Because of the significant heterogeneity (P < 0.00001, I² = 93%), the data were analyzed using a random-effect model. The difference in QOL score between the other operations and MBG groups was considerable (MD = 10.16, 95%CI: 1.56- 18.75, P = 0.02), according to the pooled data.

Total scores

Four studies were selected for data synthesis in AFOS Total Scores [4, 10, 14, 17]. We pooled the data using a random-effect model (Fig. 3f) due to significant heterogeneity (P < 0.00001, I² = 87%) and discovered a significant difference (MD = 6.53, 95% CI: 0.08- 12.08, P = 0.05). After one year of follow-up, two studies [4, 17] involving 88 people revealed total FAOS scores with SD. The statistical analyses were conducted using a random-effect model (P = 1.00, I² = 0%). Three studies [4, 10, 17] involving 143 patients found that the difference between MBG and Other operations groups was significant in terms of Total scores (MD = 7.50, 95%CI: 6.06– 8.94, P < 0.00001) during one year follow-up period. The data were determined using a random-effect model due to the substantial heterogeneity (P = 0.0003, I² = 88%). The variation scores between the MBG and Other operations groups were notable (MD = 8.05, 95%CI: 2.90- 13.20, P = 0.002) during a two-year follow-up period.

Ankle stability–talar tilt angle and anterior talar translation

Talar tilt angle

The Talar tilt angle was reported in five investigations [5, 6, 13, 18]. There was no significant heterogeneity (P = 0.82, I² = 0%), and the distinction was not statistically significant (MD = -0.19, 95%CI: -1.25– 0.87, P = 0.72). Figure 4a shows that after a three-month follow-up term, two investigations [10, 12] involving 95 patients reported the talar tilt angle with SD. Because there was no significant variation (P = 0.28, I² = 13%), the data were collected using a random-effect model. Three investigations [4, 10, 14] comprising 156 patients found no significant variations in Talar tilt angle between MBG and Other operations groups after a year of follow-up (MD = -0.08, 95% CI: -2.03– 1.88, P = 0.94). The data were analyzed using the random-effect approach, and there was no considerable disparity (P = 0.54, I² = 0%). According to the combined results, there was no great disparity in Talar tilt angle between the MBG and Other operations groups (MD = -0.13, 95%CI: -1.56– 1.29, P = 0.85).

Fig. 4

The pooled results of ankle stability evaluated with stress radiographs: Forest plot of talar tilt angle (a) and anterior talar translation (b) showing no significant difference between two groups

Anterior talar translation

Data on anterior talar translation was gathered from 193 people in four studies [6, 10, 12, 14]. Considering the lack of heterogeneity (P = 0.11, I² = 50%), a fixed model was used to merge the data, as seen in Fig. 4b, and no statistical significance was found (MD = 0.26, 95%CI: -0.98– 1.51, P = 0.68). After a three month follow-up period, two studies [10, 12] involving 95 patients reported a Sports score of FAOS with SD. There was no considerable disparity (P = 0.24, I² = 29%). In terms of anterior talar translation, there was no massive distinction between the MBG and Other operations groups (MD = 0.72, 95% CI: -0.58– 2.02, P = 0.28). After a year of follow-up, three investigations [4, 12, 14] with 156 participants were completed. Because there was no significant heterogeneity (P = 0.44, I² = 0%), the data were analyzed using a random-effect model. There were no substantial distinction groups (MD = 0.65, 95%CI: -0.39–1.68, P = 0.22), according to the combined data.

Postoperative complications

For the meta-analysis, four complications were chosen: infection, recurrence, irritation, and nerve injury. Figure 5 depicts the final result. We found no significant discrepancies and completed the meta-analysis using the fixed model to account for all problems. When assessing each complication, no statistical significance was identified between the two groups (MD = 1.11, 95% CI: 0.43– 2.88, P = 0.83).

Fig. 5

Complications

Discussion

Lateral ankle sprain (LAS) is a common disease for physically active people. It also has a high incidence in the general population. Conservative treatment procedures such as bracing and plaster fixing can heal about 80% of acute ankle sprains. CLAI continues to affect 20% of individuals [19]. The expert panel agreed on five critical factors in terms of injury diagnosis. The following are some of them: (1) previous lateral ankle sprain; (2) etiology of injury; (3) weight-bearing status; (4)clinical evaluation of ligaments, and (5) clinical evaluation of bones [20]; Delahunt et al. expanded the under-defined inclusion criteria in CLAI to classify people with mechanically and functionally unstable ankles. They clearly stated that: to be classified as CLAI, residual symptoms ("withdrawal" and a feeling of ankle instability) should be present at least one year after the initial sprain [21]. The diagnostic criteria are the difference between the lateral ankle joint and the opposite side when the talus is tilted 10° or more. The radiographic criteria for chronic lateral ankle instability surgery were > 15° or a side-to-side difference > 10° Of tibiotalar tilt angle on varus stress radiographs and > 10 mm or a side-to-side difference of > 3 mm of anteriortalar translation on anterior drawer stress radiographs [22]. Due to the high recurrence rate, patients, their families, and society need to bear a substantial medical burden [23].

With the improvement of living standards, the number of people with heavy weights has gradually increased. After sprained ankle joints, higher requirements for its stability have been put forward. Ligament repair and reconstruction, are two of CLAI's ligament repair treatments. MBG surgery refers to suturing the upper bundle of IER to the lower fibular segment based on the original operation; it is one of the most often utilized surgical procedures to treat CLAI. Other operations covered in this article, such as the LARS、AST、the procedure as reported by Karlsson et al. and the Chrisman-Snook procedure.

When the residual ligament of the patient is small, the use of MBG alone cannot provide satisfactory postoperative results, so other surgical methods can be considered to meet the needs of the patient. Regarding ligament reconstruction surgery, such as LARS, AST, etc. In the study of Hong Li et al., there are no substantial differences in risk of complications between Suture Tape Augmented repair and BR surgery; according to the researchers, the Suture Tape Augmented repair procedure seems to be a safe and fast option [24]. However, in one study, it was found that the MBG had less anterior talar displacement and talar tilting than the Chrisman-Snook or the incised ligament groups at all forces. Many scholars have recently compared other surgical methods with MBG to explore the best treatment method, Tekin Kerem Ulku et al. [14] found that the lateral ankle ligament suture augmentation is comparable to arthroscopic MBG, with a shorter operation and operation time. No casts or braces are required. William L. Henrik et al. [16] concluded that the MBG treated group had the lowest morbidity among the patients observed. The frequently mentioned complications of nerve entrapment and excessive lateral ankle tightening were less common. According to Stephen H. Liu et al. [18, 20], for ClAI with extensive ligamentous laxity, other procedures augmented with suture tape are a practical option. Some scholars believe that MBG provides excessive tension on the repaired ligament compared with other surgeries, resulting in more significant mechanical constraints than other techniques. As a method to improve the clinical outcome of CLAI. In the treatment of CLAI, various techniques have been matured. The author believes that different treatment methods can achieve sound therapeutic effects, and there is no significant difference in postoperative complications. Therefore, other surgical techniques also can be used for treatment, not only MBG.

There is currently no consistent scoring method for evaluating the prognosis of ankle stabilization surgery [25]. However, most studies used the American Orthopedic Foot and Ankle Association Ankle Posterior Foot Scale (AOFAS) score, stress radiographs, the Sefton Grading system, and patient satisfaction scores, these scoring methods are covered in relevant randomized trials [4, 5, 7, 10,11,12,13,14]. Using the scoring methods described above, we compared the results of the other operations groups and the MBG group. The other operations groups had significant advantages over the MBG group in all aspects of the FAOS score. The authors believe that the difference in FAOS scores is due to the use of tendons or suture tape to mimic the torn ligament's original physiological role and enhance ankle stability with more muscular tissue or material, thereby causing the difference in FAOS results. There was no significant difference between the two groups in the talar tilt angle and Anterior talar translation, both of which restored the average angle of the ankle joint. According to the results, other procedures can obtain better prognostic indicators than MBG and have a more comprehensive application range and shorter postoperative fixation time. According to the authors, other surgeries could be a good option for treating CLAI.

For postoperative ankle stability evaluation, radiographs were used to measure the talar tilt angle under a 150 N varus stress and the anterior talar translation under the anterior drawer stress. The Talar Tilt Angle is the angle created by the distal tibia's articular surfaces and the talus's articular surface while under varus stress, the method has been proven to be an accurate and practical method [26]. The mean talar tilt angle and anterior talar translation of patients in the two groups were over 10° and 10 mm before surgery, indicating lateral ligament damage in the ankle joint. In both groups of patients, the talar tilt angle and anterior talar translation dropped to the normal range after surgery, suggesting that CLAI can be stabilized. Since Other operations apply to an extensive range of people and have a shorter postoperative fixed period, they can obtain better prognostic indicators than MBG. The author believes that other operations can be a good choice for treating CLAI.

After pooling all the results, the other surgery group had an advantage in the FAOS score; there are no differences between the two groups in evaluating ankle stability and complications. Hence, the authors considered that the other operation group was better than the MBG group. In treating CLAI, other suitable surgical methods can be adopted, and MBG is not stubbornly used as the only choice.

Study strengths and limitations

For postoperative ankle stability evaluation, radiographs were used to measure the talar tilt angle under a 150 N varus stress and the anterior talar translation under the anterior drawer stress. The Talar Tilt Angle is the angle created by the distal tibia's articular surface and the talus' articular surface under varus load. Before surgery, patients in both groups had a mean talar tilt angle of over 10° and anterior talar translation of over 10 mm, indicating lateral ligament injury in the ankle joint.

Conclusion

The current meta-analysis found no significant differences between other operations and MBG operations regarding postoperative ankle stability evaluation, or complications. Based on the findings, the author feels that other operations surgery has clinical advantages over MBG surgery. While the data is favorable, due to the low methodological quality of RCTs, it is not conclusive. More high-quality RCTs with a low risk of bias and adequate sample numbers are needed to demonstrate its genuine effects.