Effect of plantar incision for metatarsal head resection arthroplasty of the small toes

Objective: To evaluate the clinical outcome of metatarsal head resection arthroplasty of the small toes using a plantar approach in patients with severe forefoot deformities. Methods: Twelve patients (15 feet), 10 females and two males, age 53 to 81 years old, with diabetes and rheumatoid arthritis were evaluated between January 2014 and September 2017. All patients underwent the same surgical technique – metatarsal head resection arthroplasty of the small toes via a plantar approach. The evaluation was based on pain according to the visual analogue scale, the type of footwear used and the index of patient satisfaction with the surgery. Results: In the series studied, all 12 patients (15 feet) presented significant pain improvement. Regarding wound healing, only one patient had a calloused scar, but without painful or functional impairment. Eleven individuals (14 feet) could wear all types of footwear, and only one had to use custom footwear. Conclusion: Metatarsal head resection arthroplasty of the smaller toes using a plantar approach is a safe, reliable and effective technique for the treatment of patients with severe forefoot deformities. Level of Evidence IV; Therapeutic Studies; Case Series.

Fourteen patients who underwent metatarsal head re section arthroplasty of the smaller toes, performed bet ween January 2014 and September 2017, were invited for clinical evaluation.
Patients were asked which type of footwear they were able to wear: any model, only custom shoes or no type of shoe. The visual analogue scale (VAS) was used to mea sure pain.
On physical examination, the wound healing aspects, such as hypertrophy or callosity and the distance between the digitoplantar fold of the third toe and the surgical scar, were observed.
The following method was used as the outcome crite rion: patients with minimal or absent pain who could wear any type of footwear presented an excellent outcome (VAS score less than 3); the outcome was good if pain was minimal or absent but the patient only used custom shoes (VAS score less than 3); the outcome was fair if the patient

INTRODUCTION
Metatarsal head resection arthroplasty is a procedure used to treat complex forefoot deformities, especially when there is dislocation of the metatarsophalangeal joints of the smaller toes (13) . This technique was first described by Hoffmann in 1912, in which a transverse plantar approach just below the digitoplantar fold is used to extract the metatarsal heads (4) .
When describing where the incision should be perfor med, Hoffmann notes that a distal approach to the meta tarsal heads, outside the loading area, would avoid pres sure on the surgical scar during walking, thus minimizing complications ( Figure 1).
The more distal incision, however, makes the dissection more aggressive because it is necessary to expose the neck of the metatarsals, where the surgical section for extraction of the heads is performed. When performing the incision below the metatarsal heads and consequently within the loading area, the dissection is less aggressive because the neck is nearer. Often, patients undergoing forefoot re construction surgery have diabetes or rheumatoid arthri tis, which are systemic diseases associated with surgical wound complications. Our hypothesis is that performing the incision at the level of the metatarsal heads minimizes these complications given the easier and less aggressive dissection of the anatomical structures involved.
The aim of this study is to present a case series of patients who underwent resection arthroplasty through a plantar in cision below the metatarsal heads of the smaller toes, hence more proximal than the incision described by Hoffman but less aggressive on the soft tissues, and to evaluate the possi ble consequences of the approach to the loading area.

METHODS
This study was approved by the Ethics Committee with registration in the Brazil Platform under CAAE number: 80591917.1.0000.5404. This is a retrospective, observational study in which all patients signed an informed consent form and the study met all human rights requirements. had moderate pain but showed an improvement compa red to the preoperative level, regardless of the shoe type (VAS score between 3 and 7); the outcome was considered poor if the patient had severe pain regardless of the shoe type (VAS score greater than 7).
At the end of each evaluation, the patients were asked about the degree of satisfaction with the surgery and whether or not they would undergo the surgery again.

Surgical technique
The patient was placed in a supine position, and an Esmarch tourniquet was applied to the middle third of the limb to be operated. A curved plantar incision approxima tely 3cm below the digitoplantar fold of the third toe was performed under the metatarsal heads, extending from the second to the fifth toe ( Figure 2).
Next, dissection of the plantar fat was performed, follo wed by haemostasis with electrocautery and resection of the cystic hygromas found. The flexor tendons were located and preserved. Through longitudinal incisions, arthrotomies were performed to expose the metatarsal heads ( Figure 3). The metatarsal necks were osteotomised at 45° ("flute beak") after isolation and protection of the adjacent neurovascular bundles. With the aid of a Backhaus forceps, the metatarsal heads were resected respecting the metatarsal formula, keeping the second toe larger than the third, the third larger than the fourth, and so on. The plantar fat was repositioned and the fascia retensioned when necessary. Subcutaneous suturing was performed with 3.0 absorbable sutures and skin suturing with 4.0 non absorbable sutures.
Toes were aligned with 1.5mm Kirschner sutures through the phalange following the medullary canal of the metatarsals, or simply by manual osteoclasia of the toes, maintaining the position with bandages. Regardless of the technique used for toe alignment, the patients used Ba rouk shoes after surgery, with a maximum load restriction during the first two weeks. When the stitches were remo ved, gradual loading was allowed until the fourth week, and when the Kirschner sutures (when used) were remo ved, full loading was allowed with the use of rigid footwear. Thereafter, the patients were followedup monthly through the first six months (Figure 4).

Statistical analysis
Data distribution was tested using the ShapiroWilk test, and skewness and kurtosis were also evaluated. Student's ttest for paired samples was used to compare the subjec tive perception of pre and postoperative pain. The signi ficance level adopted was 5% (P<0.05). The analyses were conducted in SPSS (SPSS Inc., Chicago, USA).

RESULTS
Of the 14 patients who underwent surgery, two did not report for evaluation: one of them indicated impro vement and did not want to be followedup at the outpa tient clinic, and the other had moved from the municipali ty and would be followedup at an orthopaedic service in their new place of residence. Thus, the sample consisted of 12 patients, three with bilateral involvement, for a total of 15 operated feet.
Nine patients had rheumatoid arthritis and complex forefoot deformities, with dislocation of the metatarso phalangeal joints of the smaller toes and painful plantar callosities, and three patients had neuropathic feet se condary to diabetes and developed ulcers under the me tatarsal heads.
The descriptive data of the sample are presented in Table 1. After surgery, a significant reduction in pain was observed, as evidenced by a decrease of approximately 7 points (P <0.001) on the VAS ( Figure 5).
The distance between the digitoplantar fold proxi mal to the third toe and the surgical scar ranged from 3 to 3.5cm with a mean of 3.2cm. All patients had healed wounds, and only one patient presented a calloused scar, but without painful or functional impairment.  The mean preoperative pain intensity was 8 (5 to 9), which significantly (P<0.001) dropped to 0.9 (0 to 2) after the procedure.
When asked whether they would undergo surgery again, all patients were satisfied because they experien ced a substantial improvement in pain and deformity and could wear conventional or custommade footwear.

DISCUSSION
Surgical treatment in patients with severe forefoot deformities presents a challenge to the orthopaedic sur geon due to the potential for postoperative complications related to both the degree of deformity and the presence of underlying diseases such as diabetes and rheumatoid arthritis (5) .   Metatarsal head resection arthroplasty is a technique frequently used to correct these deformities. It can be per formed via a dorsal or plantar approach. Most authors use the dorsal approach and report complications such as re currence of deformity caused by softtissue retraction du ring healing, shortening and overlap of the phalanx over the adjacent metatarsals, suture dehiscence and postope rative infection (69) .
The recurrence of deformities is associated with the de velopment of painful callosities (1,3) , being present in 40 to 58% of cases with unsatisfactory outcomes (3) .
Ishie et al. (5) , Amin et al. (9) and Hamalainen (10) all used a plantar approach with satisfactory results for most patients, but none of the studies included a clear description of the effect of the scar on the outcome or patient satisfaction.
Although the dorsal approach is the most used, we cho se the plantar approach because it offers better exposure of and easy access to the metatarsal heads, especially if the me tatarsophalangeal joints are dislocated. The plantar approa ch at the level of the metatarsal heads allows retensioning the fascia and repositioning the plantar fat pad, producing a more anatomical reconstruction of the soft tissues.
However, many authors avoid this approach, claiming a risk of injury to the neurovascular bundle (plantar location), injury to the plantar fat pad and the risk of painful callosities associated with the incision in the forefoot loading area.
In a series of 45 rheumatoid feet subjected to dorsal reconstruction reported by Hulse et al. (3) , 15.5% of pa tients presented severe pain and 11% presented mode rate pain, with callus formation in 40% of the cases. The reappearance of callosities was considered a consequen ce of inadequate bone resection during the extraction of the metatarsal heads.
In the present study, patients had very satisfactory hea ling, with only one case showing a callosity, suggesting that the incision in the loading area does not seem to be related to the development of painful scars.
Regarding the presence of pain analysed through the VAS, the patients showed a significant decrease in pain: all 12 patients (15 feet) reported a significant decrease in pain after surgical treatment.
Other studies describe a plantar incision through an ascending access via the root of the fifth toe but do not precisely report the location of the surgical approach.
Although this is a retrospective study with few literature references for support and few cases in the analysed series, the plantar approach at the level of the metatarsal heads is a very satisfactory approach with a high rate of good results.
Comparative studies between plantar and dorsal inci sions, which could identify the safer and more efficient ap proach, have not yet been performed. Our case series shows that the plantar approach at the level of the metatarsal heads, even when performed in the loading area, presents satisfactory outcomes with few complications.

CONCLUSION
Although further studies are needed, the plantar inci sion used for metatarsal head resection of the smaller toes in severe forefoot deformities is a safe, reliable and effecti ve procedure with a high rate of good results.