Outcomes following minimally invasive dorsal cheilectomy for hallux rigidus: A systematic review

Abstract

BACKGROUND

Cheilectomy of the 1^st metatarsophalangeal joint (MTPJ) is one of the most common procedures for the management of hallux rigidus. However, there is no consensus regarding outcomes following minimally invasive dorsal cheilectomy (MIDC) for the management of hallux rigidus.

AIM

To evaluate outcomes following MIDC for the management of hallux rigidus.

METHODS

During November 2023, the PubMed, EMBASE and Cochrane Library databases were systematically reviewed to identify clinical studies examining outcomes following MIDC for the management of hallux rigidus.

RESULTS

Six studies were included. In total, 348 patients (370 feet) underwent MIDC for hallux rigidus at a weighted mean follow-up of 37.9 ± 16.5 months. The distribution of patients by Coughlin and Shurna's classification was recorded in 4 studies as follows: I (58 patients, 27.1%), II (112 patients, 52.3%), III (44 patients, 20.6%). Three studies performed an additional 1^st MTPJ arthroscopy and debridement following MIDC. Retained intra-articular bone debris was observed in 100% of patients in 1 study. The weighted mean American orthopedic foot and ankle society score improved from a preoperative score of 68.9 ± 3.2 to a postoperative score of 87.1. The complication rate was 8.4%, the most common of which was persistent joint pain and stiffness. Thirty-two failures (8.7%) were observed. Thirty-three secondary procedures (8.9%) were performed at a weighted mean time of 8.6 ± 3.2 months following the index procedure.

CONCLUSION

This systematic review demonstrated improvements in subjective clinical outcomes together with a moderate complication rate following MIDC for the management of hallux rigidus at short-term follow-up. A moderate re-operation rate at short-term follow-up was recorded. The marked heterogeneity between included studies and paucity of high quality comparative studies limits the generation of any robust conclusions.

Key Words: Minimally invasive dorsal cheilectomy; Hallux rigidus; The first metatarsophalangeal joint; Cheilectomy; Minimally invasive surgery

Core Tip: This systematic review found improvements in patient reported outcome measurements together with a moderate complication rate following minimally invasive dorsal cheilectomy (MIDC) for the management of hallux rigidus at short-term follow-up. A moderate re-operation rate at short-term follow-up was recorded. The marked heterogeneity between included studies and dearth of high quality comparative studies limits the generation of any robust conclusions. This current systematic review demonstrates that MIDC may be a safe, efficacious procedure in the setting of low-to-moderate grade hallux rigidus.

INTRODUCTION

Osteoarthritis of the first metatarsophalangeal joint (MTPJ) is the most common arthritic condition of the foot, affecting 2.5% of patients over 50 years old[1]. Classic presentation entails 1^st MTPJ pain and progressive stiffness exacerbated on toe-off[1]. For patients with low-to-moderate grade hallux rigidus as graded by the Coughlin and Shurnas’ classification[2], initial management entails the use of nonsteroidal anti-inflammatory drugs, intra-articular corticosteroid injections, physical therapy, and shoe wear modification[1]. For patients with mild-to-moderate recalcitrant hallux rigidus, operative management using a dorsal metatarsal cheilectomy is often recommended[3]. DuVries first described the open dorsal cheilectomy in 1959, which involves resection of dorsal osteophytes from the metatarsal head and proximal phalanx and debridement of pathological synovitis. This procedure has produced satisfactory outcomes in the short-term and is considered the gold standard treatment option for moderate hallux rigidus[3].

Minimally invasive dorsal cheilectomy (MIDC) is a joint-preserving procedure that has become an increasingly popular alternative to an open approach[4-9]. The approach involves a dorsomedial stab incision and percutaneous insertion of instrumentation, with the dorsal osteophyte resected with a burr under fluoroscopic guidance[5]. Several recent studies have reported numerous benefits to this approach: smaller incisions, reduced soft-tissue disruption, reduced postoperative morbidity, accelerated recovery time, and improved aesthetics post-procedure[8,9]. Despite reports of good postoperative outcomes, there is no consensus regarding subjective clinical outcomes, radiographic outcomes, complication rates, and recurrence rates following MIDC for hallux rigidus.

To our knowledge, no systematic review examining outcomes following MIDC for hallux rigidus has been performed at present. The purpose of this systematic review was to assess the clinical, functional and radiographic outcomes following MIDC. In addition, the level of evidence (LOE) and quality of evidence of the included studies were analysed.

MATERIALS AND METHODS

In November 2023, a systematic review of the MEDLINE, EMBASE and Cochrane Library databases was performed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines[10]. The following search terms were used: [(hallux and rigidus) or (metatarsophalangeal joint) and (cheilectomy)]. Table 1 displays the inclusion and exclusion criteria. Following retrieval of the data, the titles, abstracts and full text articles were screened by 2 independent reviewers of all searched studies by applying the aforementioned criteria. A senior author was consulted to arbitrate any disagreements that arose.

Table 1 Inclusion and exclusion criteria.

Inclusion criteria	Exclusion criteria
Clinical studies reporting outcomes after minimally invasion cheilectomy (arthroscopic or mini-open) for hallux rigidus	Review articles
Follow up ≥ 12 months	Case reports
Published in a peer-reviewed journal	Cadaveric studies
Article written in English	Fewer than 5 patients per cohort

Assessment of LOE and methodological quality

The LOE was assessed using previously accepted criteria. The risk of bias and methodological quality of clinical evidence for non-randomised studies was assessed using the Risk Of Bias In Non-randomised Studies-of Interventions tool by 2 independent reviewers[11]. The options for a domain-level risk-of-bias judgement are ‘low’, ‘moderate’, ‘serious’ or ‘critical’ risk of bias, with an additional option of ‘no information’. Discrepancies were ameliorated by a senior author who assessed the data until a consensus was found.

Data extraction and evaluation

Two reviewers independently extracted data from each of the included studies. Data regarding patient demographics and the characteristics of the surgical procedure were collected. Objective and subjective outcomes, 1^st MTPJ range of motion, postoperative imaging, complications, failures, and reoperations were evaluated.

Statistical analysis

SAS software version 9.3 (SAS Institute, Inc., Cary, NC, United States) was used to complete statistical analyses. Descriptive statistics were calculated for all continuous and categorical variables. Continuous variables were reported as weighted mean and standard deviation, whereas categorical variables were reported as frequencies with associated percentages.

RESULTS

The search generated 928 studies. Of these, 6 met the inclusion and exclusion criteria (Figure 1). The 6 studies were published between 2014 and 2021.

Figure 1  Preferred reporting items for systematic reviews and meta-analyses flow diagram.

Study characteristics & patients demographics

Study characteristics and patient demographic data are listed in Table 2. One study was LOE III[8] and 5 studies were LOE IV[4-7,9]. All studies were moderate risk of bias (Figure 2). In total, there were 348 patients (370 feet) that underwent minimally invasive cheilectomy for hallux rigidus with a weighted mean age of 51.1 ± 5.4 years (range: 41.8-54). There were 66 males (30.7%) and 149 females (69.3%). The weighted mean postoperative follow up time was 37.9 ± 16.5 months (range: 17-56.4 months). Fifty-eight feet (27.1%) were classified as Grade I, 112 feet (52.3%) classified as Grade II, and 44 feet (20.6%) classified as Grade III.

Figure 2  Risk of bias in non-randomised studies-of interventions.

Table 2 Summary of study characteristics and outcomes.

Ref.	LOE	Patients (n)	Feet (n)	Mean follow-up (months)	Mean age (yr)	Sex (M/F)	Coughlin and Shurna's grade
Glenn et al[4], 2021	4	20	20	16.5	52	M = 6; F = 14
Hickey et al[5], 2020	4	36	36	56.4	50	M = 10; F = 26	I = 5; II = 27; III = 1
Levaj et al[6], 2021	4	29	29	31.2	41.8	M = 19; F = 10	I = 11; II = 12; III = 6
Pastides et al[7], 2014	4	41	54	17	43	M = 6; F = 35	I = 9; II = 19; III = 26
Stevens et al[8], 2020	3	133	133	36	54
Teoh et al[9], 2019	4	89	98	50	54	M = 25; F = 64	I = 33; II = 54; III = 11

M/F: Male/female; LOE: Level of evidence.

Surgical characteristics

All 6 studies described using a stab incision at the dorsomedial aspect of the first MTPJ, followed by debridement of osteophytes with a burr to complete cheilectomy. Five studies reported using a 3.1 mm burr[4,5,7-9], while the remaining study did not report burr size[6]. Three studies performed 1^st MTPJ arthroscopic evaluation and debridement following resection of the dorsal osteophyte[4-6]. One study reported arthroscopic findings and found that 100% of patients had bone debris within the 1^st MTPJ despite irrigation following resection of the osteophyte[4]. In addition, the authors reported that 100% of patients had evidence of synovitis, 10% had loose bodies within the 1^st MTPJ, and 30% had a loose cartilage flap identified on arthroscopic evaluation.

Clinical outcomes

Table 3 summarises the clinical and functional outcomes. Four different subjective clinical scoring tools were recorded across the 6 studies. The most commonly utilized patient reported outcome measure (PROM) was the American Orthopedic Foot and Ankle (AOFAS) score in 2 studies[5,7]. The weighted mean preoperative AOFAS score was 68.9 ± 3.2 (range, 66.6-71.1). One study recorded postoperative AOFAS score, with mean score of 87.1[7]. Two studies used visual analog scale (VAS) pain scores[4,9]. The weighted mean preoperative VAS score was 7.8 ± 0.6 (range, 7.1-8.0) which improved to a weighted mean postoperative score of 2.6 ± 1.6 (range, 0.8-3.0).

Table 3 Summary of clinical and functional outcomes, complications, failures and secondary surgical procedures.

Ref.	Patients (n)	Feet (n)	PROM	Score		1st MTPJ ROM (°)		Complications	Failures	Secondary surgical procedures
				Preop	Postop	Preop	Postop
Glenn et al[4], 2021	20	20	VAS	7.1	0.8	47	67	0	0	0
Hickey et al[5], 2020	36	36	AOFAS	66.6	n/r	31.9	72.7	EHL rupture = 1, neuropathic pain = 3, joint stiffness = 1	0	Repeat arthroscopic cheilectomy = 1, manipulation under anaesthetic = 3
Levaj et al[6], 2021	29	29	n/r			0	0	Joint stiffness = 1	1	Open revision = 1
Pastides et al[7], 2014	41	54	AOFAS	71.1	87.1	39	48	Neuropathic pain = 2	2	MTPJ fusion = 2
Stevens et al[8], 2020	133	133	n/r			0	0	EHL rupture = 1, neuropathic pain = 3, superficial wound problems = 2, joint stiffness/pain = 7, ectopic bone in wound = 2	17	MTPJ arthrodesis = 9, open cheilectomy = 3, MIS cheilectomy-2, interposition arthroplasty = 1, hallux valgus correction = 2
Teoh et al[9], 2019	89	98	VAS	8	3	11.3	69.1	Neuropathic pain = 4, superficial wound problems = 2, wound infection = 2	12	Repeat cheilectomy = 4, open removal of loose bone = 1, 1st MTPJ arthrodesis = 7

VAS: Visual analog scale; AOFAS: American orthopedic foot and ankle society; Preop: Preoperative; Postop: Postoperative; EHL: Extensor hallucis longus; MTPJ: Metatarsophalangeal joint; PROM: Patient reported outcome measurement; ROM: Range of motion; n/r: Not report; MIS: Minimally invasive surgery.

Postoperative range of motion (ROM) of the 1^st MTPJ joint was reported in 4 studies[4,5,7,9]. The weighted preoperative mean 1^st MTPJ ROM was 21.5° ± 20.5° (range, 11.3°-47°). The weighted postoperative mean 1^st MTPJ ROM was 42.8° ± 34.2° (range, 48°-72.7°). Four studies reported dorsiflexion[4,5,7,9]. The weighted mean dorsiflexion preoperatively was 26.8 ± 10.3 (range, 11.3-32), compared to 57.2 ± 16.3 (range, 39-72.7) postoperatively. Two studies reported plantarflexion[4,5]. The weighted preoperative mean plantarflexion was 11 ± 5.07 (range, 7-15), compared to postoperative was 14 ± 7.07 (range, 9-19).

Return to sport was recorded in 1 study[7]. The authors demonstrated that 73% of patients had returned to sporting activity by 6 wk after the procedure.

Radiographic data was not reported in any of the 6 studies.

Complications, secondary procedures and failures

Data regarding complications, secondary procedures, and failures is listed in Table 3. Complications in total, 31 complications (8.4%) were observed in the patient cohort. Complications included numbness and neuropathic pain, extensor hallucis longus tendon rupture, joint stiffness, ectopic bone in wound, superficial wound infection, and deep wound infection. The most common complication was neuropathic pain (3.2%). In total, 33 patients (8.9%) required secondary procedures at a weighted mean time of 8.6 ± 3.2 months following the index procedure. These included arthroscopic cheilectomy, steroid injection, open revision, arthrodesis, interposition arthroplasty, hallux valgus correction, and open removal of loose bone. The most common secondary surgical procedure was first MTPJ arthrodesis (6.9%). In total, there were 32 failures (8.7%) recorded at final follow-up.

DISCUSSION

This current systematic review found that MIDC produces improvements in subjective clinical outcomes and high patient satisfaction rates together with moderate complication rates (8.4%) at short-term follow up. Notably, a moderate reoperation (8.9%) and failure rate (8.7%) was observed. However, the heterogeneity amongst the included studies in addition to the poor quality of evidence and low LOE restricts the generation of meaningful cross-sectional analysis. This review illustrates that MIDC may be a safe, efficacious procedure in the setting of low-to-moderate grade hallux rigidus.

Hallux rigidus is a debilitating pathology which results in progressively worsening of 1^st MTPJ pain, decreased ROM and reduced quality of life[1]. The progressive, relentless nature of the degenerative changes at the 1^st MTPJ renders conservative treatment modalities ineffective in the long term. Thus, following failure of nonoperative management, surgical intervention is warranted. Joint sparing procedures including cheilectomy and interpositional arthroplasty are indicated in mild-to-moderate disease while joint sacrificing procedures such as 1^st MTPJ arthrodesis are indicated in more severe pathology[3]. Open cheilectomy is the most commonly utilized joint sparing procedure for the management of mild-to-moderate hallux rigidus, with good clinical outcomes and excellent survivorship reported at short-term follow-up in the literature[12]. Cetinkaya et al[12] conducted a retrospective review of 29 patients who underwent open cheilectomy as first line treatment of grade III hallux rigidus at 2-year follow-up[12]. The authors reported statistically significant improvement in AOFAS and VAS scores in addition to a 0% reoperation rate. There has been a growing interest in the use of less invasive cheilectomy techniques given the concerning complications rates associated with the traditional open procedures[13]. This current systematic review found that MIDC resulted in high patient satisfaction rates (range, 88%-100%) and significant improvements in subjective clinical outcomes regardless of the clinical scoring tool utilized. However, the most commonly utilized PROM was the AOFAS score, which is not validated for hallux rigidus. One of the few validated subjective clinical scoring tools, the FAOS score[14], was not utilized across any of the 6 studies in this review, thus, caution must be taken when interpreting the reported improvements in subjective clinical outcomes following MIDC in this review.

All studies included in this current review performed MIDC using a 3 mm dorsomedial stab incision followed by resection of any dorsal osteophyte using a 3.1 mm conical or wedge burr under fluoroscopic guidance. Furthermore, in 3 studies, the joint was accessed with a 2.7 mm arthroscope following resection of the dorsal osteophytes, and any identified synovitis, loose bodies, scar tissue, cartilage flaps and/or osteochondral lesions was treated as appropriate[4-6]. This systematic review found significant improvements in functional outcomes following MIDC, as evidenced by an increase in 1^st MTPJ ROM from preoperative ROM of 21.5° to postoperative ROM of 42.8°, which is comparable to prior case series for open cheilectomy. This indicates that despite the less aggressive surgical approach, MIDC leads to noninferior restoration of 1^st MTPJ range of motion in comparison to open cheilectomy. Furthermore, the use of the arthroscope provides excellent visualization of the joint and can identify any intra-articular pathology. Glenn et al[4] utilized an arthroscope during their procedures and found that 100% of their cases had intra-articular synovitis, 30% had large cartilage flaps contributing to 1^st MTPJ impingement and 10% had loose bodies, all of which were treated concurrently at the time of the procedure, underpinning both the diagnostic and therapeutic benefits of 1^st MTPJ arthroscopy during minimally invasive surgery cheilectomy[4].

Although open cheilectomy results in excellent functional outcomes, there is concern regarding postoperative morbidity, with particular attention paid to postoperative wound complication rates which are reported to range from 3.5%-9.1%[13,15]. MIDC mitigates against the morbidity associated with open cheilectomy due to the minimal soft tissue disruption afforded by smaller stab incisions in comparison to the standard 5-6 cm incision performed during open cheilectomy[4,12]. Overall, the postoperative wound complication rate in the MIDC cohort was 1.6% with no cases of deep infections nor wound dehiscence, reinforcing that MIDC is a safe, effective surgical technique with low incisional morbidity. MIDC may be a particularly useful strategy in patients predisposed to wound complications including patients with a history of diabetes, smoking history, increasing age and prior corticosteroid injections[16,17].

MIDC is not without limitations, however. The most common complication observed in this systematic review was iatrogenic nerve injury, which was found in 3.2% of patients who primarily reported altered sensation in the distribution of the dorsal medial cutaneous nerve (DMCN). The stab incision during MIDC is typically placed over the dorsomedial aspect of the 1^st metatarsal just proximal to the medial eminence[9]. In comparison to a mid-medial incision, a dorsomedial stab incision provides superior access to the dorsal osteophyte and facilitates an ergonomic sweeping movement of the burr which optimises resection of the osteophyte, that can not be achieved with a mid-medial incision[9]. However, the variation in size of the dorsal osteophyte may lead to incorrect placement of the incision, thus placing the DMCN at risk of iatrogenic injury. Furthermore, a cadaveric study by Solan et al[18] found that a mid-medial incision was less likely to result in iatrogenic injury to the DMCN compared to a dorsomedial incision as the dorsomedial incision is positioned more adjacent to the nerve[18].

This current systematic review found a moderate secondary surgical procedure rate (8.9%) following MIDC in the short-term period. Teoh et al[9] postulated that a substantial proportion of the reoperations in their cohort could be attributed to the “learning curve” associated with MIDC, as 42% of their reoperations were due to a lack of technical proficiency, illustrated by the high rates of inadequate osteophyte resection[9]. Interestingly, the reoperation rate was higher in the studies that did not perform 1^st MTPJ arthroscopy (10.9%) compared to the studies that did perform 1^st MTPJ arthroscopy (2.4%). Glenn et al[4] found that, despite copious irrigation following resection of the osteophyte, 100% of patients had retained bone debris within the joint as visualized on arthroscopic examination of the 1^st MTPJ[4]. The large quantities of retained bone debris may not be sufficiently expelled through the small incisions and can lead to persistent 1^st MTPJ stiffness and pain, requiring further intra-articular injections and inevitable repeat cheilectomies and/or joint sacrificing procedures. The authors performed 1^st MTPJ arthroscopic debridement of any identified bone debris in their patient cohort, and reported no reoperations at final follow-up, reinforcing that 1^st MTPJ arthroscopy is crucial following MIDC to ensure adequate clearance of any debris from the joint to reduce the risk of symptomatic recurrence[4].

This systematic review is influenced by numerous limitations and/or biases. There was significant inconsistency in the reporting of data between studies, limiting the generation of robust conclusions. The search criterion was limited to the MEDLINE, EMBASE and Cochrane Library Database articles that were published exclusively in English. The included studies have a LOE that was graded as low with poor methodological quality, with no adequate control group included in any of these studies. The follow-up was short, thus the long-term longevity of MIDC could not be determined. Finally, the data extraction was not performed blindly, but was carried out by two independent reviewers and later confirmed by the lead author

CONCLUSION

This systematic review demonstrated improvements in subjective clinical outcomes together with a low complication rate following minimally invasive cheilectomy for the management of hallux rigidus at short-term follow-up. A moderate re-operation rate at short-term follow-up was recorded. However, the marked heterogeneity between included studies and paucity of high quality comparative studies limits the generation of any robust conclusions. Further research with a longer follow-up is warranted to determine the precise role for minimally invasive cheilectomy in the management of hallux rigidus.