Correction of Clawed Hallux Deformity: Comparison of the Jones Procedure and FHL Transfer
A clawed hallux is a deformity of the great toe deﬁned as the extension of the ﬁrst metatarsophalangeal (MTP) joint combined with ﬂexion of the interphalangeal (IP) joint.10 The aberrant mechanics of this ﬁrst ray deformity can lead to increased pressure on the plantar surface of the ﬁrst metatarsal and the dorsal surface of the hallux, which could potentially lead to ulceration in neuropathic patients (Figure 1).1,13 The imbalance is thought to be a result of a functional “overpull” (a disproportionate load) of one or more of the three extrinsic muscles of the ﬁrst ray:
The role of each of these muscles has been explored previously in a cadaver model of the clawed hallux.10
The most commonly used operative procedure for the treatment of a pes cavus foot with a clawed hallux deformity was originally described by Jones in 1916.6 The original procedure consisted of transfer of the EHL tendon to the neck of the ﬁrst metatarsal but has evolved to include fusion of the IP joint as to minimize ﬂexion caused by the unopposed FHL tendon. This often is referred to as the modiﬁed Jones procedure.2
A retrospective study of the modiﬁed Jones procedure for correction of clawed hallux deformities demonstrated that tendon regeneration of the distal EHL stump was a contributing factor in the recurrence of the clawed hallux deformity.7 Additionally, another frequent complaint (though minor) of the modiﬁed Jones procedure was catching of the great toe when walking barefoot.2 In this study, 48 of 81 patients reported catching of the great toe. Nevertheless, 74 patients (including all 48 who reported catching of the great toe) were ‘very pleased’ or ‘pleased with reservations’ with their surgery. Hansen5 noted that patients are not always pleased with the lack of dorsiﬂexion and the stiff great toe that can result from the modiﬁed Jones. He suggested an alternative transfer of the FHL tendon from the ﬁrst distal phalanx to the base of the proximal phalanx. This procedure preserves the function of the EHL and does not require fusion of the IP joint (Figure 2). Further, these two procedures can be used to correct different pathologies.
The purpose of this study was to assess the mechanical effectiveness of the FHL transfer compared to the modiﬁed Jones procedure in correcting both the clawed hallux deformity and its mechanical consequences. We were interested in studying the overall efﬁcacy of operative correction (independent of surgery type), as well as whether or not there were differences between the two procedures in the resulting joint angles and plantar pressures. We examined the angular changes at the MTP and IP joints, as well as the plantar pressures beneath the ﬁrst metatarsal and the distal hallux. We hypothesized that there would be significant differences in joint angles and plantar pressure for the following:
Nine fresh-frozen cadaver feet (84 ± 3.5 years of age) were obtained for this institutional review board-approved study. All tissue and handling protocols for this study were in accordance with the University of Washington Human Subjects Division guidelines for biohazardous materials. Five feet were randomly selected for the modiﬁed Jones procedure (EHL transfer into the dorsal ﬁrst metatarsal and fusion of the great toe joint) and four for transfer of the FHL tendon from the distal to the proximal phalanx.
The feet were tested using a loading frame capable of statically simulating different phases of the gait cycle.10 Loads were applied to feet held in an acrylic frame through pneumatic cylinders and servovalves. In preparation for preoperative testing, the tibia and ﬁbula of each foot were transversely sectioned 12 cm proximal to the ankle, and the extrinsic tendons (anterior tibial, posterior tibial, peroneus brevis, extensor digitorum longus, ﬂexor digitorum longus, and Achilles, in addition to the PL, EHL, and FHL) were dissected free to the superior extensor retinaculum. The intramedullary canal of the tibia was reamed with a 1.27 cm drill bit to allow insertion of a compressive acrylic rod that provided sufﬁcient lateral support. Using a ﬂuoroscope for guidance, 4-mm holes were drilled and carbon ﬁber pins (4 mm diameter, approximately 5 cm length) were embedded into the distal phalanx, proximal phalanx, and ﬁrst metatarsal. Polymethylmethacrylate was used to tightly anchor the pins in place. Plastic tendon clamps were attached to the free tendons of the extrinsic musculature. Nylon lines connected these clamps to six 1.91-cm-diameter bore tensile pneumatic cylinders (the peroneus brevis and extensor digitorum brevis were not tested). As the pull on the Achilles tendon was much larger than the other extrinsic tendons, a larger plastic clamp was attached to the Achilles tendon and connected to a 9.53-cm-diameter bore tensile cylinder with a thin metal cable. Coarse sand paper was wrapped around the Achilles tendon to prevent slippage.
For the modiﬁed Jones procedure, the IP joint of the great toe was exposed through an L-shaped incision on the medial side of the foot (Figure 3).8 The tendon of the EHL was dissected and cut transversely 1 cm proximal to the IP joint. The synovial sheath around the EHL was excised. A transverse 3-mm hole was drilled into the inferomedial aspect of the ﬁrst metatarsal neck and continued along the long axis of the bone to the dorsolateral aspect of the neck. The tendon was passed through the hole and sutured to itself and the periosteum using 3-0 vicryl for reinforcement. The IP joint was fused using a 2-mm diameter carbon ﬁber rod that was inserted through a hole drilled anterior to posterior through the distal phalanx into the proximal phalanx.
For the FHL transfer, a medial midline incision was made over the MTP joint and carried distally to the IP joint (Figure 4).5 The FHL tendon was exposed under the proximal phalanx. The incision was carried distally to expose the FHL attachment at the distal phalanx. The tendon was cut at its most distal attachment to the distal phalanx. The synovial sheath of the FHL was excised. A 3-mm transverse hole was made through the base of the proximal phalanx. The tendon was passed through the hole and sutured to itself and the periosteum using 3-0 vicryl for reinforcement.
The specimens were tested before and after surgery in the loading frame (Figure 5). Each foot was carefully positioned and centered in the frame so that the compressive load was applied to the tibia by an acrylic rod attached to the compressive cylinder (10.16-cm-diameter bore). Midstance during the gait cycle was simulated by positioning the foot in 7° of dorsiﬂexion by using a wooden block. Plantar pressure was measured using a PEDAR pressure measurement system. Measurement of the three-dimensional orientation of the bones of interest was obtained by using a Fastrak electromagnetic motion analysis system. The transmitter was mounted on the base of the frame with each of the four sensors attached to the carbon ﬁber pins embedded into the bones of interest. The loading protocol developed by Olson et al.10 was used to simulate the clawed hallux through overpull of the PL, FHL, and EHL (Figure 6).
Linear mixed effects models were used to assess changes group was examined by testing for main effect signiﬁcance for each of these variables. Post hoc multiple comparison tests by individual muscle-pull group were conducted by using Tukey’s HSD (honest signiﬁcant difference) method.
There was a signiﬁcant difference between preoperative and postoperative ﬁrst metatarsal pressures (p = 0.015), with pressure decreasing after surgery by an average of 3.2 N/cm2. The change in pressure was signiﬁcant across all groups. There was no signiﬁcant difference in change in pressure among muscle-pull groups, indicating that all of the overpulls had similar results (p = 0.11). There was no signiﬁcant relationship between change in pressure and surgery type (p = 0.7).
There was no overall signiﬁcant difference between preoperative and postoperative hallux pressure (p = 0.5), with pressure decreasing after surgery by an average of only 0.5 N/cm2. There was a signiﬁcant relationship between change in hallux pressure and muscle-overpull group (p = 0.0004), with a signiﬁcantly decreased pressure for the FHL overpull after surgery (2.4 N/cm2, p = 0.014), while there was no signiﬁcant change for the other three muscle-overpull groups. Post-hoc analysis demonstrated that mean hallux pressure was signiﬁcantly higher for FHL overpull than the other three groups (p < 0.05). Feet that had the modiﬁed Jones procedure had lower mean differences in pressure than those that had FHL transfer, but this difference was not signiﬁcant (p = 0.16).
There were signiﬁcant differences between the preoperative and postoperative MTP joint angles (p = 0.037), with the angle decreasing after surgery by an average of 1.4°. The signiﬁcant differences occurred mainly in the EHL and FHL muscle-pull groups (both groups had a mean change of 2.1°, p = 0.023) with no change in the angle for the PL group. Although there was no overall statistical signiﬁcance (p = 0.068), there were apparent differences in the mean MTP joint angle among muscle types. Post hoc analysis demonstrated that the PL overpull was signiﬁcantly less than with the EHL and FHL overpulls (p = 0.02). There were no signiﬁcant differences between MTP joint change and surgery type (p = 0.5). There were signiﬁcant differences between preoperative and postoperative IP joint angle (p = 0.0020) with the angle increasing after surgery by an average of 4.6°. As with MTP joint angle, the signiﬁcant differences occurred mainly with EHL and FHL muscle-pull groups (with mean increases of 6.4° and 7.3°, and p = 0.0009 and 0.0003, respectively). Unlike the MTP joint angle, the differences in change in IP joint angle were signiﬁcantly different among muscle-pull groups (p = 0.0010) with mean EHL and FHL angle differences signiﬁcantly higher than that for PL group (p < 0.05). There were no signiﬁcant differences in IP joint angle change and surgery type (p = 0.2).
Our results showed that there was a signiﬁcant decrease in hallux pressure after surgery only when the FHL was overpulled, while there was no signiﬁcant change for the other three muscle-pull groups. Both surgeries were equally effective at reducing plantar pressure under the distal hallux. It should be noted that the protocol by Olson et al.10 called for normalizing the pressure data to midstance, while this study did not. However, the trends for the preoperative data, both angular and pressure, were constant with those reported by Olson et al.10
Clinically, there is more than one etiology for clawed hallux deformity.3,4,9,11,12 However, we demonstrated that the modiﬁed Jones procedure and FHL transfer are equally effective at reducing the angular deformity at the MTP and IP joints. If increased plantar pressure under the ﬁrst metatarsal is signiﬁcant, both the modiﬁed Jones procedure and the FHL transfer will yield similar results. The indication for choosing one procedure over the other may lie in whether or not a patient is a good candidate for fusion of the IP joints. If increased plantar pressure under the ﬁrst metatarsal is signiﬁcant, both the modiﬁed Jones procedure and the FHL transfer will yield similar results. The indication for choosing one procedure over the other may lie in whether or not a patient is a good candidate for fusion of the IP joint. Often the lack of dorsiﬂexion in a fused great toe is a negative consideration after surgery.5 Usually, younger patients prefer to have mobility of the IP joint, and, in this situation, the FHL transfer is recommended because similar angular reductions at the MTP and IP joints were achieved without the potential for increased morbidity with a fused hallux. Further investigation regarding transfer of the PL to the peroneus brevis in combination with the FHL transfer is warranted to determine if it can provide better results in diabetic patients with a clawed hallux deformity and foot ulcers.
To read the full project description, please see:
Correction of Clawed Hallux Deformity: Comparison of the Jones Procedure and FHL Transfer in a Cadaver Model. Elias FN, Yuen TJ, Olson SL, Sangeorzan BJ, Ledoux WR. Foot Ankle Int. 2007 Mar;28(3):369-76. PMID: 17371661.