Darrin J. Trask, 1,2 William R. Ledoux, 1,3,4 Eric C. Whittaker, 1 Grant C. Roush, 1,4 Bruce J. Sangeorzan 1,3

Transcription

1 Second Metatarsal Osteotomies for Metatarsalgia: A Robotic Cadaveric Study of the Effect of Osteotomy Plane and Metatarsal Shortening on Plantar Pressure Darrin J. Trask, 1,2 William R. Ledoux, 1,3,4 Eric C. Whittaker, 1 Grant C. Roush, 1,4 Bruce J. Sangeorzan 1,3 1 Department of Veterans Affairs RR&D Center of Excellence for Limb Loss Prevention and Prosthetic Engineering, VA Puget Sound Health Care System, Seattle, Washington 98108, 2 School of Medicine, University of Washington, Seattle, WA 98195, 3 Department of Orthopaedics & Sports Medicine, University of Washington, Seattle, Washington 98195, 4 Department of Mechanical Engineering, University of Washington, Seattle, Washington Received 20 June 2012; accepted 21 October 2013 Published online 14 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI /jor ABSTRACT: Symptom relief of recalcitrant metatarsalgia can be achieved through surgical shortening of the affected metatarsal, thus decreasing plantar pressure. Theoretically an oblique metatarsal osteotomy can be oriented distal to proximal (DP) or proximal to distal (PD). We characterized the relationship between the amount of second metatarsal shortening, osteotomy plane, and plantar pressure. We hypothesized that the PD osteotomy is more effective in reducing metatarsal peak pressure and pressure time integral. We performed eight DP and eight PD second metatarsal osteotomies on eight pairs of cadaveric feet. A custom designed robotic gait simulator (RGS) generated dynamic in vitro simulations of gait. Second metatarsals were incrementally shortened, with three trials for each length. We calculated regression lines for peak pressure and pressure time integral vs. metatarsal shortening. Shortening the second metatarsal using either osteotomy significantly affected the metatarsal peak pressure and pressure time integral (first and third metatarsal increased, p < 0.01 and <0.05; second metatarsal decreased, p < 0.01). Changes in peak pressure (p ¼ ) and pressure time integral (p ¼ ) were more sensitive to second metatarsal shortening with the PD osteotomy than the DP osteotomy. The PD osteotomy plane reduces plantar pressure more effectively than the DP osteotomy plane. Published 2013 by Wiley Periodicals, Inc. on behalf of the Orthopaedic Research Society. J Orthop Res 32: , Keywords: second metatarsal; plantar pressure; metatarsalgia; lesser metatarsal osteotomies; gait simulation Metatarsalgia is defined as pain, often during weight bearing, of the plantar aspect of the foot under and related to the lesser metatarsal heads. Primary metatarsalgia is associated with biomechanical insufficiencies, and secondary metatarsalgia is associated with systemic conditions. 1 Initially, metatarsalgia is managed with rest, stretching exercises, cushioning, plantar callosity shaving, and anti-inflammatory medications, but data confirming their effectiveness is limited. 2 If symptoms persist, surgery is employed to correct the alignment of the metatarsals and/or muscle/ligament balance. Numerous surgical treatments are used, each with its benefits and complications. 2 6 Surgical treatment redistributes pressure under the metatarsal heads 7 by either dorsally displacing the head or by shortening the length of the metatarsal. 3 The amount of shortening is determined by the surgeon s experience considering the preoperative length of the metatarsals. 4,8 In this study, we compared the effects of two different osteotomy planes and the overall amount of second metatarsal shortening on plantar pressure. Insight into the relationship between osteotomy plane, metatarsal shortening, and plantar pressure contributes to a more effective treatment of metatarsalgia by Grant sponsor: Department of Veterans Affairs Rehabilitation Research and Development Service A4843C; Grant sponsor: University of Washington Medical Student Research Training Program. Correspondence to: William R. Ledoux (T: þ ; F: þ ; wrledoux@uw.edu) Published 2013 by Wiley Periodicals, Inc. on behalf of the Orthopaedic Research Society. This article is a U.S. Government work and is in the public domain in the USA. providing the surgeon with greater knowledge of biomechanical principles that can be used to make intraoperative osteotomy decisions. The oblique distal to proximal (DP) sliding osteotomy and the oblique proximal to distal (PD) sliding osteotomy are designed to reduce second metatarsal pressure, but with different mechanisms. The DP osteotomy is a common treatment for metatarsalgia, 6,9,10 consisting of an oblique cut through the metatarsal neck from the dorsal, distal surface directed proximally and inferiorly relative to the metatarsal at an angle parallel to the plantar surface of the foot. The detached metatarsal head is moved proximally and reattached. The DP osteotomy is often successful at reducing forefoot pain, but complications occur. 6,7,10 14 An improper osteotomy plane can shift the MTPJ center of rotation, causing the interosseous muscles to act as dorsiflexors causing dorsiflexion contracture. 12 Additional complications such as penetrating hardware, soft tissue infections, fixation irritation, transferred metatarsalgia, MTPJ stiffness, floating toe deformity, and return to initial symptoms have been reported. 6,7,10,11,13 Furthermore, when the DP osteotomy angle is not parallel to the plantar surface, that is, when anatomical or technical issues preclude ideal alignment of the saw cut, the metatarsal head can move inferiorly increasing plantar pressure. 14 Removing a wedge of bone is a technique that can be used to address this problem. 15 An alternative is the PD osteotomy, which is made from a more proximal, dorsal point and directed in a distal, inferior direction. The detached head is moved superiorly and reattached. Helal introduced a similar 385

2 386 TRASK ET AL. procedure that was effective at relieving metatarsalgia symptoms, 16,17 however mal-union and nonunion complications were considerable, 8,16,17 possibly due to a lack of internal fixation. 8 In this cadaveric study, we quantified the relationship between second metatarsal shortening and plantar pressure distribution for a given osteotomy plane. We hypothesized that the PD osteotomy results in a greater reduction in second metatarsal peak pressure (PP) and pressure time integral (PTI) than the DP osteotomy. We also hypothesized that shortening the second metatarsal decreases the second metatarsal PP and PTI and increases first and third metatarsal head PP and PTI across surgery type. METHODS The robotic gait simulator (RGS) is a custom designed device that generates prosthetic and cadaveric gait simulations It can generate dynamic in vitro simulation of normative tibial motion, tendon forces, and vertical ground reaction forces (vgrf). The basic idea behind the system is to move the ground relative to the fixed tibia using inverse kinematics obtained from a motion analysis laboratory. The RGS consists of an R degree of freedom parallel robot (Mikrolar, Inc., Hampton, NH) on which a force plate (Kistler Instrument Corp., Amherst, NY) and a pressure plate (novel GmbH, Munich, Germany) are mounted in series; this hardware consists of the ground that moves relative to the foot with the tibia rigidly mounted to a stiff frame (Fig. 1). An electromechanical force-controlled tendon actuation system with nine DC linear tendon force actuators (Exlar Corp., Chanhassen, MN) in series with nine load cells (Transducer Techniques, Inc., Temecula, CA) generates the muscle forces, while a real time PXI controller (National Instruments Corp., Austin, TX) and a custom PC user interface is used to guide the interaction between the R2000, the tendon actuators, and the foot. A 6-camera motion analysis system (Vicon, Lake Forest, CA) is used to track the motion of the bones of the foot. Eight pairs of fresh frozen cadaveric feet (age ¼ 82 8 years, range 69 95, body weight (BW) ¼ N) were used. An orthopedic surgeon evaluated each foot to ensure that no major anatomic abnormalities, evidence of arthritis, or prior foot surgery were present. Specimens were transected at the mid-shaft of the tibia and thawed for use. The medullary canal was drilled to 12.7 mm diam. A custom steel post was fitted into the tibia to connect the foot to the RGS tibial mounting frame. The post was attached to the tibia with PMMA (Lang Dental Mfg. Co., Wheeling, IL). About 10 cm of each of the nine extrinsic ankle tendons (Achilles [Ach], peroneus longus [PL], peroneus brevis [PB], extensor hallucis longus [EHL], extensor digitorum longus [EDL], tibialis anterior [TA], tibialis posterior [TP], flexor digitorum longus [FDL], and flexor hallucis longus [FHL]) was dissected superior to the medial and lateral malleoli. The tendons were coupled to the actuators with custom clamps. After the second metatarsal osteotomy was completed as described below, the cadaveric specimen was mounted into the RGS (Fig. 1). We performed eight DP and eight PD second metatarsal osteotomies, with each respective osteotomy performed on each pair of feet. The respective osteotomies were alternated for a total of four right and four left feet for each. We assumed that at baseline, the groups were similar, as all feet were screened for abnormalities and each group had one foot from each pair. An incision was made along the length of the second metatarsal. The EDL and extensor digitorum brevis tendons were retracted to expose the second metatarsal from the proximal base to the proximal aspect of the extensor expansion. The DP osteotomies were made proximal to the second metatarsal head as an oblique cut through the neck from the dorsal surface directed proximally at an angle parallel to the plantar surface of the foot (Fig. 2). The PD osteotomies were performed on the distal half of the second metatarsal epiphysis and made from a proximal, dorsal location directed distally and plantarly (Fig. 2). A custom aluminum bracket, previously used in a similar manner by our group to explore the effect of a long second metatarsal, 21 stabilized the osteotomies, simulated healing, and allowed for incremental shortening. Prior to the osteotomy, the bracket was positioned along the superior aspect of the metatarsal with the distal aspect of the bracket set in alignment with the proximal extent of the MTP joint capsule. The bracket was fixed to the second metatarsal with two distal bone screws and two proximal hanger bolts. Their locations and the osteotomy window were marked. The respective osteotomy was made in the window, and the bracket was then fixed at the original second metatarsal length determined via linear calipers. When the nuts on the Figure 1. The robotic gait simulator (RGS). A, force plate and pressure plate; B, mobile platform; C, tendon cables; D, 6-camera Vicon motion analysis system; E, support frame; F, freeze clamp system for Achilles tendon frame; G, second metatarsal lengthening bracket; H, medial malleolus marker.

3 SECOND METATARSAL OSTEOTOMY PLANES 387 Figure 2. Medial-lateral radiographs of the second metatarsal osteotomies, oblique distal to proximal sliding osteotomy (DP, top) and the oblique proximal to distal sliding osteotomy (PD, bottom). Green line, drawn over the in situ saw blade, indicates angle of cut. proximal bolts were loosened, the proximal slot on the bracket allowed unimpeded shortening of the metatarsal along the plane of the osteotomy (Fig. 3). The target shortening was in 2 mm increments from 0 to 6 mm. The actual amount of shortening was determined via linear calipers. The literature reports metatarsal shortening ranging from <4 up to 8 mm. 4,6,8,23 25 The bracket was low profile and did not interfere with lesser toe extensor tendon excursion. To remove interoperator variability, the same individual made the osteotomies and placed the bracket for all feet. For clarity, we generated a schematic of the bracket position for each surgery (Fig. 4). During gait simulation, the tibia was held fixed while the R2000 moved the force and pressure plates to simulate the motion of the tibia relative to the ground. Stance phase tibia kinematics and vgrf were prescribed from 10 healthy patients who underwent 4 5 gait cycle trials. 20 Target muscle forces were estimated from the literature using an EMG to force model with a 42 ms electromechanical delay. 31 The stance phase was scaled to 50% BW and 4.09 s. Loads were scaled to reduce the risk of catastrophic specimen failure (fracture of the talar neck or Achilles tendon ruptures, both of which happened previously on specimens with higher loads) while speed was decreased to reduce noise from mechanical vibrations on the force plate. All tendons except Ach and TA were controlled by a real time PID force controller. A fuzzy logic controller tracked the in vitro vgrf by altering the target TA and Ach tendon forces (real time) and tibial kinematics (iteratively) to track the target vgrf. 20 For each foot, three successful trials were obtained at either three or four target second metatarsal lengths. A successful trial consisted of: the first and second peaks of the trial vertical GRF within 10% of the target vgrf; during the entire stance phase, the in vitro tendon forces of the extrinsic muscles, excluding TA and Ach, within 10 N RMS error of their prescribed forces; and the plantar pressure profile began at the heel, moved through the lateral foot and onto the medial metatarsal heads and phalanges. If these criteria were not achieved, we adjusted the initial position of the force plate. Prior to each successful trial, three learning trials were performed to allow the plantar tissue to be preconditioned and the fuzzy logic controller to adjust the inferior/superior position of the force plate to achieve the target vgrf. The target metatarsal lengths were 0, 2, 4, and 6 mm of shortening. second metatarsal length was measured with digital calipers in reference to the fixation bracket placement prior to the osteotomy, after the osteotomy without shortening, and at each shortening target. Pressure masks were created in novel Multimask from preoperative radiographs using a previously described method. 32 PP and PTI were calculated in the novel software suite for the defined plantar mask areas for the first, second, and third metatarsal heads. PP (a measure of the greatest pressure) and PTI (a measure of the pressure dosage) were chosen as plantar pressure is presumed to be related to the etiology of metatarsalgia and the goal of the surgical treatments is the redistribution of pressure under the heads. 3 To determine if the slope of the second metatarsal PP or PTI versus amount of shortening differed by surgery type, linear mixed effects regression was used with PP or PTI as the dependent variable, while independent fixed effect covariates included surgery type, amount of shortening, and surgery type amount of shortening interaction. Foot pair was a random effect. A significant interaction term would suggest that the two slopes differed by surgery type. To determine if the slope of the first, second, and third metatarsal PP or PTI versus amount of shortening 6¼ 0, linear mixed effects regression was used with PP or PTI as the dependent variable. Amount of shortening was the independent fixed effect, and random effects included foot pair, individual foot ID, and foot pair amount of shortening interaction. The coefficient for amount of shortening represented the slope, the significance of which indicated that the slope 6¼ 0. A negative slope represented a decrease in PP or PTI with second metatarsal shortening. All analyses were done with R RESULTS Data from 14 feet (8 PD, 6 DP) were included. Shortening was not achieved in two DP osteotomies, due to the improper postoperative fixation. These feet were excluded; the bracket prevented shortening because the osteotomy angle was made too steep relative to the plantar surface. All eight PD feet completed the 6 mm trial, but only three of six DP feet achieved the 6 mm trial without hitting the bracket; the other three only reached the 4 mm trial. The average peak second metatarsal shortening was mm for the PD and mm the DP. In general, both osteotomies decreased second metatarsal pressure, but increased the first and third metatarsal pressures. The PD osteotomy resulted in a larger decrease in second metatarsal

4 388 TRASK ET AL. Figure 3. (A) Second metatarsal fixation with custom aluminum bracket. Medial/lateral X-ray of the (B) distal to proximal (DP) and (C) proximal to distal (PD) procedures. PP and PTI (Fig. 5) than the DP osteotomy based on regression analysis of second metatarsal shortening. The slope of the second metatarsal PP and second metatarsal shortening regression was 56% greater for the PD (Table 1). In addition, the slope of the second metatarsal PTI and second metatarsal shortening regression was 93% greater for the PD osteotomy (Table 1). Regardless of the type of osteotomy, there were significant changes in PP and PTI of the first, second, and third metatarsals (Fig. 5). Both the first and third metatarsal PP and PTI were positively correlated with second metatarsal shortening, whereas the second metatarsal PP and PTI were negatively correlated with second metatarsal shortening (Table 1). The fuzzy logic vgrf controller tracked the vgrf to within 1 SD of the mean in vivo data for most of stance phase (Fig. 6) across all 42 trials (3 trials 14 specimens). The first and second peaks of the vgrf, representing weight acceptance and push off, were within 1 SD of those found in vivo, but deviated from the 1 SD range in several locations, most notably just prior to midstance. This undershoot, indicating a necessary adjustment to the controller, led to a brief moment (a few percent of stance phase or 0.1 s with the slower simulation speed) of overshoot as the Ach

5 SECOND METATARSAL OSTEOTOMY PLANES 389 Figure 4. Schematic of the distal to proximal (DP) and proximal to distal (PD) procedures before and after shortening with the custom fixation bracket: metatarsal (A), custom fixation bracket (B), two hanger bolts in proximal segment (C), two bone screws in distal segment (D), and line of cut (E). The gap between the bracket and bone in the shortened diagrams was minimized by including minor sagittal plane rotation. As seen in the shortened DP diagram, the tip of the proximal segment was removed to allow for shortening without obstruction by the bracket. tendon control became active. However, across all 42 trials for the entire gait cycle, the average RMS error between the target in vivo and actual in vitro GRF was 4.6% BW for the vgrf, 7.8% BW for the AP GRF, and 3.5% BW for the medial/lateral GRF, despite the fact that the latter two were not controlled for. The RGS could replicate the in vivo kinematics of the tibia with respect to the ground (i.e., the RGS) within 1 SD (Fig. 7). The departure from 1 SD during the last 10% of stance resulted from a decrease in speed of the RGS as it completed the kinematic trajectory. The RMS tracking error for the EHL, EDL, TP, FHL, FDL, PB, and PL tendons ranged from 3.1 N for FDL to 5.2 N for PL with a mean value of 4.2 N across all 42 trials. DISCUSSION In our study, the PD osteotomy achieved a greater reduction in second metatarsal PP and PTI than the DP osteotomy. However, independent of osteotomy plane, shortening the second metatarsal decreased second metatarsal head PP and PTI, while increasing first and third metatarsal head PP and PTI. Therefore, either osteotomy can decrease symptoms associated with primary metatarsalgia, as shown by the decrease in plantar pressure under the second metatarsal head. The increase in PP and PTI under the first and third metatarsals reflects the change in loading pattern created in the postoperative foot and illustrates how subtle change in the length of one metatarsal influences the surrounding metatarsals. The probably mechanism allowing the PD osteotomy to achieve greater reduction in second metatarsal PP and PTI is the difference in geometry of the osteotomies. The DP osteotomy leaves the metatarsal head to follow a trajectory inferior to the proximal metatarsal, allowing for shortening, but not elevation of the head. Conversely, the PD osteotomy allows the head to slide superiorly and away from the painful plantar tissue as the metatarsal is shortened. Not only was there a larger decrease in PP using the PD osteotomy (56% > DP), suggesting further symptom relief over the DP osteotomy, there was an even larger decrease in PTI using the PD osteotomy (93% > DP). The larger change in PTI suggests that the PD osteotomy lowers the amount of pressure that the metatarsal head is exposed to throughout the gait cycle. Since the slope of the PD osteotomy regression line was steeper than that of the DP osteotomy regression line, the same reduction of pressure can be achieved by performing the PD osteotomy with less shortening than the DP osteotomy. The clinical implication is that while either procedure can reduce pressure, the PD might be preferred because it can be employed with less overall metatarsal length change, which could lead to less disruption of surrounding soft tissue. However, the PD procedure could also more easily lead to aberrant first and third metatarsal pressure, so surgical care must be taken. Several studies measuring pressures associated with DP osteotomy showed a similar trend. 23,24,34 In a clinical study, Vandeputte et al. 23 showed similar pressure reductions post-op (88.3 kpa/5.9 mm shortening ¼ 15.0 kpa/mm) to the slopes of the second metatarsal PP and second metatarsal shortening regressions in our study. In a statically loaded cadaveric study, Khalafi et al. 24 reported a significant decrease in pressure under the operated second metatarsal of 36% ( kpa) in neutral position and 65% ( kpa) in 45 of plantar flexion and an increase in pressure under the first metatarsal in neutral position. Significant decreases in pressure also occurred under the third metatarsal, contrary to our study, possibly explained by axially loading the specimens and not simulating the entire gait cycle. In another cadaveric study, Lau et al. 34 shortened second

6 390 TRASK ET AL. Figure 5. Regression plots for two feet, comparing peak pressure (PP) and pressure time integral (PTI) during shortening for oblique distal to proximal sliding osteotomy (DP) and the oblique proximal to distal sliding osteotomy (PD). metatarsals by 5 mm and cyclically loaded the feet in 30 of dorsiflexion and reported averaged pressures under the second metatarsal head of kpa preoperatively and kpa postoperatively, without change in first and third metatarsal plantar pressure. In contrast, Snyder et al. 25 showed no difference in pressure between an intact second metatarsal and a DP osteotomy on the second metatarsal, perhaps due to the use of static loading. No other significant changes in pressure or load in other regions of the foot were seen in the DP osteotomy. These studies provided a better biomechanical understanding of the DP osteotomy; however, none examined the relationship between the amount of second metatarsal shortening and plantar pressure. The novel contribution of our study is that we explored changes in plantar pressure distribution with the PD osteotomy during dynamic testing under tendon loading and tibial kinematic conditions that generate the vgrf profile developed during gait. Our study was not without potential limitations. The small sample size for the DP osteotomy at 6 mm shortening might limit applicability at this shortening length. Certainly, the linear relationship should not be extrapolated outside this range. We were unable to reproduce screw fixation and bony healing as seen in vivo, but the aluminum bracket provided rigid fixation of the osteotomized second metatarsal. Clinically meaningful testing in the RGS required stable metatarsal fixation that could bear near physiologic conditions just after surgery. The same bracket was used for all osteotomies and should not have altered the second metatarsal shortening geometries. Reliance on fixation to simulate healing is an inherent imperfection of cadaveric studies, but the bracket was a good substitute for a healed metatarsal; we have used the

7 SECOND METATARSAL OSTEOTOMY PLANES 391 Table 1. Peak pressure (PP) and pressure time integral (PTI) means SDs for second metatarsal comparing the oblique proximal to distal sliding osteotomy (PD) and oblique distal to proximal sliding osteotomy (DP) surgeries (significance based on differences in slope by surgery); first, second, and third metatarsal data shown, with DP and PD surgeries grouped (significance based on difference in slope from 0) Initial Value Before Shortening Range From Initial to Max Shortening Slope (U/mm) p-value Second metatarsal (PD vs. DP) PP:PD (kpa) PP:DP (kpa) PTI:PD (kpa s) PTI:DP (kpa s) First metatarsal (all specimens, PD and DP) PP (kpa) PTI (kpa s) Second metatarsal (all specimens, PD and DP) PP (kpa) PTI (kpa s) Third metatarsal (all specimens, PD and DP) PP (kpa) PTI (kpa s) Negative indicates a decrease. Note Average second metatarsal shortening was mm (PD) and mm (DP). indicates significance based on difference in slope from 0. same device for a lengthening study. 21 Admittedly, we did not simulate pathology; the cadaveric feet were considered to have normal anatomic variation and not necessarily be predisposed to metatarsalgia. A study using pathologic feet could potentially alter the absolute plantar pressure characteristics of the osteotomy planes, but our comparisons were relative between the DP and PD procedures. It should be noted that intrinsic muscles were not actively include in the gait simulations, so only their passive characteristics were included. Also, we did not explore the effect of a wedge cut on DP osteotomy, which is another strategy sometimes employed to decrease inferior head translation. Another limitation was the reduced magnitude of some dynamic gait simulation parameters. The stance phase was completed in 4.09 s (1/6 physiologic velocity) to reduce noise from inertial forces on the force plate and at 50% BW to prevent specimen failure. Neither restriction was fatal as this was a comparative study, and both surgeries experienced the same conditions. Pressure data were decreased from typical in vivo loading, and the slow speed and decreased BW may have altered foot bone kinematics. However, realistic GRF curves, muscle loading patterns, and plantar pressure distributions for the two osteotomies gave confidence that the stance phase of gait was simulated well. In conclusion, we characterized the relationship between osteotomy plane, metatarsal shortening, and plantar pressure in the potential surgical treatment of metatarsalgia. We compared oblique sliding DP and the oblique sliding PD osteotomies, using a dynamic gait simulator. Our second metatarsal pressure data are largely consistent with other cadaveric studies and Figure 6. Left: Mean 1 SD in vitro vertical ground reaction force (vgrf) for all 14 feet (blue) compared to target in vivo vgrf (gray). Right: Mean 1 SD in vitro (blue) compared to in vivo (gray) medial/lateral GRF (top) and AP GRF (bottom) for all 14 feet with 3 trials/foot. Note the different scales for each figure.

8 392 TRASK ET AL. Figure 7. Mean 1 SD in vivo tibia with respect to ground angles (gray) compared to mean 1 SD in vitro tibia angles for the frontal (blue), transverse (red), and sagittal (green) planes for 14 feet with 3 trials/foot. one in vivo study. The PD osteotomy provides a more effective means to reduce PP and PTI. Our study suggests that the plane of the osteotomy matters in redistribution of plantar pressure. ACKNOWLEDGMENTS Jane B. Shofer, M.S. completed the statistical analysis. This study was supported by Department of Veterans Affairs Rehabilitation Research and Development Service grant no. A4843C and the University of Washington Medical Student Research Training Program. The funding sources played no role in this investigation. REFERENCES 1. Scranton PE Jr Metatarsalgia: diagnosis and treatment. J Bone Joint Surg Am 62: Espinosa N, Maceira E, Myerson MS Current concept review: metatarsalgia. Foot Ankle Int 29: Hamilton KD, Anderson JG, Bohay DR Current concepts in metatarsal osteotomies: a remedy for metatarsalgia. Tech Foot Ankle Surg 8: Hodgkins CW, O Malley MJ, Elliot AJ Lesser metatarsal osteotomies in metatarsalgia. Tech Foot Ankle Surg 5: Roukis TS Central metatarsal head-neck osteotomies: indications and operative techniques. Clin Podiatr Med Surg 22: , vi. 6. O Kane C, Kilmartin TE The surgical management of central metatarsalgia. Foot Ankle Int 23: Beech I, Rees S, Tagoe M A retrospective review of the weil metatarsal osteotomy for lesser metatarsal deformities: an intermediate follow-up analysis. J Foot Ankle Surg 44: Trnka HJ, Muhlbauer M, Zettl R, et al Comparison of the results of the Weil and Helal osteotomies for the treatment of metatarsalgia secondary to dislocation of the lesser metatarsophalangeal joints. Foot Ankle Int 20: Barouk LS Weil s metatarsal osteotomy in the treatment of metatarsalgia. Orthopade 25: Hofstaetter SG, Hofstaetter JG, Petroutsas JA, et al The Weil osteotomy: a seven-year follow-up. J Bone Joint Surg Br 87: Trnka HJ, Gebhard C, Muhlbauer M, et al The Weil osteotomy for treatment of dislocated lesser metatarsophalangeal joints: good outcome in 21 patients with 42 osteotomies. Acta Orthop Scand 73: Trnka HJ, Nyska M, Parks BG, et al Dorsiflexion contracture after the Weil osteotomy: results of cadaver study and three-dimensional analysis. Foot Ankle Int 22: Migues A, Slullitel G, Bilbao F, et al Floating-toe deformity as a complication of the Weil osteotomy. Foot Ankle Int 25: Grimes J, Coughlin M Geometric analysis of the Weil osteotomy. Foot Ankle Int 27: Hodgkins CW, O Malley MJ, Elliott AJ, et al Lesser metatarsal osteotomies in metatarsalgia. Tech Foot Ankle Surg 5: Helal B Metatarsal osteotomy for metatarsalgia. J Bone Joint Surg Br 57: Helal B, Greiss M Telescoping osteotomy for pressure metatarsalgia. J Bone Joint Surg Br 66: Aubin PM, Cowley MS, Ledoux WR Gait simulation via a 6-DOF parallel robot with iterative learning control. IEEE Trans Bio-Med Eng 55: Bayomy AF, Aubin PM, Sangeorzan BJ, et al Arthrodesis of the first metatarsophalangeal joint: a robotic cadaver study of the dorsiflexion angle. J Bone Joint Surg Am 92: Aubin PM, Whittaker EC, Ledoux WR A robotic cadaveric gait simulator with fuzzy logic vertial ground reaction force control. IEEE Trans Robot 28: Weber JR, Aubin PM, Ledoux WR, et al Second metatarsal length is positively correlated with increased pressure and medial deviation of the second toe in a robotic cadaveric simulation of gait. Foot Ankle Int 33: Whittaker E, Hahn ME, Ledoux WR Foot bone motion in cavus, neutral, and planus feet using an in vivo kinematic foot model. In: Proceedings of the second International Foot and Ankle Biomechanics Congress, Seattle, WA. 23. Vandeputte G, Dereymaeker G, Steenwerckx A, et al The Weil osteotomy of the lesser metatarsals: a clinical and pedobarographic follow-up study. Foot Ankle Int 21: Khalafi A, Landsman AS, Lautenschlager EP, et al Plantar forefoot pressure changes after second metatarsal neck osteotomy. Foot Ankle Int 26: Snyder J, Owen J, Wayne J, et al Plantar pressure and load in cadaver feet after a Weil or chevron osteotomy. Foot Ankle Int 26: Wickiewicz TL, Roy RR, Powell PL, et al Muscle architecture of the human lower limb. Clin Orthop 179: Ishikawa M, Komi PV, Grey MJ, et al Muscle-tendon interaction and elastic energy usage in human walking. J Appl Physiol 99: Pierrynowski MR A physiological model for the solution of individual muscle forces during normal human walking. British Columbia, Canada: Simon Fraser University. 29. Fukunaga T, Roy RR, Shellock FG, et al Specific tension of human plantar flexors and dorsiflexors. J Appl Physiol 80: Perry J Gait analysis: normal and pathological function. Thorfare, NJ: SLACK Incorporated.

9 SECOND METATARSAL OSTEOTOMY PLANES Zhou S, Lawson DL, Morrison WE, et al Electromechanical delay in isometric muscle contractions evoked by voluntary, reflex and electrical stimulation. Eur J Appl Physiol Occup Physiol 70: Ledoux WR, Hillstrom HJ The distributed plantar vertical force of neutrally aligned and pes planus feet. Gait Posture 15: Team RDC A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. 34. Lau JT, Stamatis ED, Parks BG, et al Modifications of the Weil osteotomy have no effect on plantar pressure. Clin Orthop 421: