e45(1) C OPYRIGHT Ó 2014
BY
T HE J OURNAL
OF
B ONE
AND J OINT
S URGERY, I NCORPORATED
Biomechanical Assessment of Flexible Flatfoot Correction Comparison of Techniques in a Cadaver Model Diego H. Zanolli, MD, Richard R. Glisson, BS, James A. Nunley II, MD, and Mark E. Easley, MD Investigation performed at the Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina
Background: Options for surgical correction of acquired flexible flatfoot deformity involve bone and soft-tissue reconstruction. We used an advanced cadaver model to evaluate the ability of key surgical procedures to correct the deformity and to resist subsequent loss of correction. Methods: Stage-IIB flatfoot deformity was created in ten cadaver feet through ligament sectioning and repetitive loading. Six corrective procedures were evaluated: (1) lateral column lengthening, (2) medial displacement calcaneal osteotomy with flexor digitorum longus transfer, (3) Treatment 2 plus lateral column lengthening, (4) Treatment 3 plus ‘‘pants-overvest’’ spring ligament repair, (5) Treatment 3 plus spring ligament repair with use of the distal posterior tibialis stump, and (6) Treatment 3 plus spring ligament repair with suture and anchor. Correction of metatarsal dorsiflexion and of navicular eversion were quantified initially and periodically during postoperative cyclic loading. Results: Metatarsal dorsiflexion induced by arch flattening was initially corrected by 5.5° to 10.6°, depending on the procedure. Navicular eversion was initially reduced by 2.1° to 7.7°. The correction afforded by Treatments 1, 3, 4, 5, and 6 exceeded that of Treatment 2 initially and throughout postoperative loading. Inclusion of spring ligament repair did not significantly enhance correction. Conclusions: Under the tested conditions, medial displacement calcaneal osteotomy with flexor digitorum longus tendon transfer was inferior to the other evaluated treatments for stage-IIB deformity. Procedures incorporating lateral column lengthening provided the most sagittal and coronal midfoot deformity correction. Addition of spring ligament repair to a combination of these three procedures did not substantially improve correction. Clinical Relevance: An understanding of treatment effectiveness is essential for optimizing operative management of symptomatic flatfoot deformity. This study provides empirical evidence of the advantage of lateral column lengthening and novel information on resistance to postoperative loss of correction.
Peer Review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. It was also reviewed by an expert in methodology and statistics. The Deputy Editor reviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication. Final corrections and clarifications occurred during one or more exchanges between the author(s) and copyeditors.
T
ibialis posterior tendon dysfunction is a common cause of adult-acquired flatfoot deformity1-3. Surgical options for stage-II dysfunction, which involves tibialis posterior tendon elongation, hindfoot valgus, forefoot abduction, and arch collapse, include osseous realignment procedures and combinations of osseous correction and soft-tissue reconstruction1,2,4-7.
Contemporary realignment procedures include medial displacement calcaneal osteotomy and lateral column lengthening. Medial displacement calcaneal osteotomy shifts the Achilles tendon insertion relative to the subtalar joint axis, reducing hindfoot valgus deforming forces8-10. It is commonly combined with tenodesis of the flexor digitorum longus tendon to the compromised tibialis posterior tendon or to the medial aspect
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
J Bone Joint Surg Am. 2014;96:e45(1-8)
d
http://dx.doi.org/10.2106/JBJS.L.00258
e45(2) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
OF
F L E X I B L E F L AT F O O T
Materials and Methods Specimen Preparation
TABLE I Correction Procedures Treatment
BIOMECHANICAL ASSESSMENT CORRECTION
Description
1
Lateral column lengthening
2
Medial displacement calcaneal osteotomy and flexor digitorum longus transfer to the posterior tibialis tendon
3
Medial displacement calcaneal osteotomy and flexor digitorum longus transfer to the posterior tibialis tendon plus lateral column lengthening
4
Treatment 3 plus spring ligament repair by a pants-over-vest technique
5
Treatment 3 plus spring ligament repair by use of the distal posterior tibialis stump
6
Treatment 3 plus spring ligament repair by use of suture and anchor
of the navicular in an effort to restore transverse tarsal joint adduction and subtalar joint inversion11-13. Lateral column lengthening simultaneously adducts the foot at the talonavicular joint, plantar flexes the midfoot, and derotates the hindfoot out of valgus, restoring the medial longitudinal arch14-16. Combining lateral column lengthening with flexor digitorum longus transfer and medial displacement calcaneal osteotomy decreases pressure across the lateral column of the foot compared with lateral column lengthening alone8,17. When medial arch collapse is accompanied by attenuation or rupture of the spring ligament and uncovering of the talonavicular joint, spring ligament reconstruction may be performed18. Several techniques have been described, including ‘‘pants-over-vest’’ repair19,20, augmentation with use of the distal stump of the tibialis posterior tendon21, and peroneus longus autograft tendon transfer18. There is, however, no consensus regarding indications and technique for spring ligament reconstruction18. Because of the range of surgical options, it is critical that the surgeon understands the extent of flatfoot correction that each option may be expected to achieve and maintain. The orthopaedic literature offers only ‘‘fair’’ and ‘‘insufficient’’ levels of evidence for flatfoot correction by medial displacement calcaneal osteotomy and lateral column lengthening, respectively22, and is limited to two comparative clinical analyses of these procedures23,24. To address the lack of biomechanical studies directly comparing the effects of the hindfoot osteotomies, the present study was designed to objectively measure the correction achieved by common surgical procedures, utilizing an advanced cadaver model in which stage-IIB (flexible) flatfoot deformity was created. We hypothesized that treatments incorporating lateral column lengthening would provide better correction of medial longitudinal arch flattening and navicular eversion relative to the talus compared with treatments not incorporating such lengthening. We further hypothesized that spring ligament reconstruction would increase the correction achievable by the bone realignment procedures.
T
en normal, fresh frozen lower limbs sectioned at the midpoint of the tibia and fibula were used for this institutional review board-approved study. The mean age of the eight male and two female donors was fifty-seven years (range, forty-three to sixty-nine years). Proximal soft tissues were removed to 4 cm above the ankle, preserving the interosseous membrane and extrinsic tendons. The proximal ends of the tibialis posterior, tibialis anterior, peroneus longus, peroneus brevis, flexor digitorum longus, flexor hallucis longus, extensor digitorum longus, and extensor hallucis longus tendons were sutured to plastic rings to provide secure attachment points for subsequent tensile load application. Steel rods were cemented into the medullary canals of the tibia and fibula with use of polymethylmethacrylate cement. The tibial rod was clamped to the linear actuator of a servohydraulic testing machine (Model 1321; Instron, Norwood, Massachusetts) such that the tibia was vertical and the foot sole was horizontal. This neutral flexion position was maintained throughout testing. A calibrated spring applied a downward force to the fibula equaling approximately 15% of the axial load applied to the tibia during the ensuing 25,26 testing . The hindfoot and forefoot rested on separate glass plates isolated from the testing machine table by ball bearings, allowing the foot to establish its preferred transverse-plane orientation and minimizing foot deformation con27-30 straints during loading . Three 5-mm half-pins were inserted for attachment of electronic angular measurement devices. The first projected medially and superiorly from the talar neck, the second projected superiorly from the navicular, and the third projected superiorly from the first metatarsal. Electronic clinometers (Schaevitz AccuStar; Measurement Specialties, Hampton, Virginia) were attached to the pins (Fig. 1). The outputs of these devices corresponded to their rotation in the vertical plane, and they had an operating range of 90° and a linearity of ±0.1%. 31-35 Similar devices have previously been used to measure relative bone rotations . The clinometers attached to the talus and the first metatarsal were aligned in the sagittal plane parallel to the second metatarsal to document the inclination of
Fig. 1
A specimen mounted in the testing machine for cyclic axial loading with simultaneous tendon loading while key osseous relationships were monitored. The two outer clinometers attached to the talus and the first metatarsal monitored their relative angulation in the sagittal plane. The two inner clinometers in the photograph were attached to the talus and the navicular for measurement of their relative angulation in the coronal plane. Axial load was applied through the tibia, and a proportionally smaller load was applied to the fibula with use of a spring, allowing fibular motion during specimen loading.
e45(3) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
TABLE II Initial Correction Relative to the Flattened Condition* Correction Relative to Flattened Condition (deg)
Treatment
First Metatarsal Plantar Flexion Relative to Talus
Navicular Inversion Relative to Talus
1
10.6 ± 3.4
6.6 ± 4.5
2
5.5 ± 2.3
2.1 ± 1.8
3
8.8 ± 4.1
6.7 ± 3.2
4
8.8 ± 4.3
7.2 ± 4.0
5
8.8 ± 4.4
7.0 ± 4.4
6
9.1 ± 4.7
7.7 ± 4.7
*Measured on the first load cycle of the postoperative cyclic loading. The values are given as the mean and the standard deviation. For comparison, the mean flattened talus-first metatarsal angle was 8.3° of metatarsal dorsiflexion relative to normal, and the mean flattened talus-navicular angle was 6.0° of navicular eversion relative to normal.
the talus relative to the first metatarsal, corresponding to the primary clinical measure of medial longitudinal arch flattening. Two additional clinometers attached to the talus and the navicular in a plane perpendicular to the second metatarsal measured their relative coronal-plane inclination. Cables, pulleys, and dead weights were used to apply tensile loads to the extrinsic tendons during axial loading of the foot, similar to the methodology 25 described by Blackman et al. . Cables attached to each of the tendons passed over pulleys that aligned the tendons in an approximately physiologic line of pull. Because of the large forces applied to the Achilles tendon, it was grasped in a specially designed clamp that compressed the tendon tissue between textured 36 metal plates (see Appendix) . An appropriately sized weight was suspended from each cable (see Appendix), applying approximately 50% of the load transmitted through the 37,38 tendon at the midstance phase of gait . The weights were raised and lowered simultaneously by means of a modified hydraulic motorcycle jack to unload and load the tendons.
BIOMECHANICAL ASSESSMENT CORRECTION
OF
F L E X I B L E F L AT F O O T
to mimic insufficiency. The 1780-N peak load was 50% greater than the load applied during preconditioning (and later during evaluation of the correction methods), and it was based on preliminary testing to determine a load magnitude that would expeditiously produce flattening representative of stage-IIB deformity within 1000 load cycles. The minimum acceptable flattening was established as 4° in each plane on the basis of clinical measures of first metatarsal dorsiflexion in patients with flatfoot and navicular eversion measured by 42 Kido et al. in their computed tomography (CT) study of flatfoot deformity . A 1187-N load was then applied as before, and the sagittal-plane talus-first metatarsal angle and the coronal-plane talus-navicular angle were recorded to document the amount of flattening attained.
Correction Procedures The evaluated procedures are summarized in Table I. Treatments 1, 2, and 3 were performed in that order in each specimen, and Treatments 4, 5, and 6 were performed in a random order. Treatment 1 involved lateral column lengthening by means of a calcaneal osteotomy 1.5 cm proximal to the calcaneocuboid joint, preserving the medial cortex. The osteotomy was fixed with an 8-mm BIOFOAM Evans Wedge (Wright Medical Technology, Arlington, Tennessee) and a two-hole 3.5-mm CHARLOTTE CLAW Plate with two screws (Wright Medical Technology). Between Treatments 1 and 2, the lateral column lengthening was undone by removing the wedge and plate and stabilizing the osteotomy with a plate and two screws. Treatment 2 involved a medial displacement calcaneal osteotomy, consisting of a transverse osteotomy 1 cm posterior and parallel to the peroneal tendons, translation of the tuberosity 1 cm medially, and fixation with two 7.0-mm cannulated screws (Synthes, West Chester, Pennsylvania). In addition, the flexor digitorum longus tendon was transected proximal to the knot of Henry, then attached to the tibialis posterior tendon through a hole in the medial third of the navicular by means of number-5 ETHIBOND suture (Ethicon, Somerville, New Jersey). In Treatment 3, lateral column lengthening was reestablished and combined with the medial displacement calcaneal osteotomy and tendon transfer. Treatments 4, 5, and 6 involved the addition of spring ligament repair to Treatment 3. In Treatment 4, this repair was performed with use of a pants-over-vest tech19,20 nique and four number-5 ETHIBOND sutures . In Treatment 5, the repair was performed by suturing the distal posterior tibialis tendon stump to the proximal part of the spring ligament by means of a number-5 ETHIBOND 21 suture . In Treatment 5, the repair was performed by inserting one 3.5-mm
Preconditioning Each specimen was preconditioned by a cyclic axial (downward-directed) load applied through the tibia, accompanied by the constant upward-directed extrinsic tendon loads. One thousand load cycles oscillating between 808 N and 1187 N were applied with use of a displacement rate of 1.5 mm/sec. The 808-N lower load exceeded the sum of the opposing tendon loads by 50 N, thereby maintaining the sole of the foot in contact with the testing machine table. The 1187-N peak load equaled the sum of the tendon loads plus 50% of the vertical ground reaction force typical during the stance phase of gait, and was equivalent to approximately 125% 39,40 of body weight . The axial load and the four clinometer outputs were monitored and recorded during the 1187-N peak loading of the 1000th load cycle. These measures of normal bone alignment under load were used as a reference to quantify both the deformity induced during the ensuing flattening process and the correction achieved by each subsequent treatment.
Flattening Process To facilitate flattening, the talocalcaneal interosseous, plantar naviculocuneiform, plantar first metatarsocuneiform, and anterior superficial deltoid liga25,41 ments were sectioned . The superomedial and inferomedial portions of the spring ligament were sectioned at the midsubstance. Axial loading between 808 N and 1780 N was then applied for 1000 cycles, accompanied by extrinsic tendon loading as described above but excluding the tibialis posterior tendon
TABLE III Initial Correction Relative to the Normal, Intact Condition* Correction Relative to Normal, Intact Condition (deg) Treatment
Talus-First Metatarsal Angle, Sagittal
Talus-Navicular Angle, Coronal
1
12.3 ± 4.6
10.6 ± 4.4
2
22.8 ± 4.2
23.9 ± 1.9
3
10.5 ± 6.1
10.8 ± 3.4
4
10.5 ± 6.2
11.2 ± 4.4
5
10.6 ± 6.0
11.0 ± 4.8
6
10.9 ± 6.3
11.7 ± 5.0
*Measured on the first load cycle of the postoperative cyclic loading. The values are given as the mean and the standard deviation. Positive mean numbers indicate slight overcorrection; negative mean numbers indicate failure to achieve complete correction.
e45(4) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
TABLE IV Loss of Sagittal-Plane and Coronal-Plane Correction* Loss of Initial Correction During Cyclic Loading (deg) Talus-First Metatarsal Angle, Sagittal
Talus-Navicular Angle, Coronal
Treatment
Mean
Std. Dev.
Mean
Std. Dev.
1
2.7
1.2
0.8
1.0
2
3.8
1.4
1.7
0.7
3
3.1
1.1
0.6
1.6
4
2.6
1.1
0.6
1.1
5
2.5
0.9
0.7
0.8
6
2.4
0.8
0.7
1.0
*Between postoperative load cycles 1 and 1000.
Corkscrew FT Suture Anchor (Arthrex, Naples, Florida) into the navicular and suturing it to the proximal part of the spring ligament.
Assessment of Resistance to Loss of Correction After each correction procedure, the cyclic loading regimen used to precondition the specimens (1000 cyclic loads to 1187 N with accompanying tendon loads) was repeated. The clinometer outputs at the first load peak indicated the initial correction achieved relative to both the normal, intact condition and the flattened condition. To document resistance to loss of correction, clinometer data were also recorded on load cycles 2, 3, 4, 5, 10 through 90 in 10-cycle increments, and 100 through 1000 in 100-cycle increments. Flattening relative to the intact state was calculated at each time point and was plotted graphically. Repeated-measures analysis of variance (ANOVA) with Bonferroni post hoc testing was used to identify significant performance differences among the six treatments at postoperative load cycles 1 (reflecting initial correction), 100, 500, and 1000.
BIOMECHANICAL ASSESSMENT CORRECTION
OF
F L E X I B L E F L AT F O O T
angle correction (p = 0.001) at load cycle 1. Post hoc testing indicated that the initial correction achieved by Treatment 1 (lateral column lengthening), Treatment 3 (lateral column lengthening plus medial displacement calcaneal osteotomy plus flexor digitorum longus transfer transfer), and Treatments 4, 5, and 6 (Treatment 3 plus spring ligament repair) all significantly exceeded that achieved by Treatment 2 (medial displacement calcaneal osteotomy plus flexor digitorum longus transfer) (p < 0.05). With the group sizes used in this study, no significant differences in initial sagittal correction among Treatments 1, 3, 4, 5, and 6 were demonstrable. Similarly, there was a significant interaction effect between treatment and the extent of talus-navicular angle correction in the coronal plane (p < 0.001). As in the case of the talus-first metatarsal angle, the initial correction achieved by Treatments 1, 3, 4, 5, and 6 each significantly exceeded that achieved by Treatment 2 (p < 0.001). Resistance of Correction to Reflattening For each treatment, gradual loss of some of the talus-first metatarsal angle correction occurred as postoperative cyclic loading progressed (Fig. 2, Table IV). The correction achieved by Treatment 2 deteriorated almost completely by load cycle 1000. The correction maintained by Treatments 1, 3, 4, 5, and 6 each exceeded that of Treatment 2 at load cycles 100, 500, and 1000 (p < 0.02). With the available group sizes, no significant differences in correction maintenance were demonstrable among Treatments 1, 3, 4, 5, and 6. The addition of spring ligament repair to Treatment 3 provided only slight, statistically nonsignificant, additional resistance to loss of correction. There was no evidence of suture or tissue failure at the sites of the spring ligament repairs or tendon transfers, and there was no evidence of structural failure at the osteotomy sites.
Source of Funding There was no external funding for this investigation.
Results Flattening lattening resulted in reduction of the sagittal-plane talusfirst metatarsal angle and eversion of the navicular relative to the talus, paralleling changes seen clinically in moderate flatfoot deformity4,5,13,42. The mean medial longitudinal arch flattening was 8.3° (range, 4.8° to 16.3°). The mean change in the coronal talus-navicular angle was 6.0° (range, 4.2° to 8.5°).
F
Initial Deformity Correction The initial correction achieved by each of the six treatments was substantial (Table II). Talus-first metatarsal angle correction ranged from 5.5° to 10.6°, and talus-navicular angle correction ranged from 2.1° to 7.7°, depending on the procedure. With the exception of Treatment 2 (medial displacement calcaneal osteotomy), all correction methods initially slightly overcorrected both the talus-first metatarsal and talus-navicular deformities (Table III). The ANOVA results indicated a significant interaction effect between treatment and the extent of talus-first metatarsal
Fig. 2
Progression of the talus-first metatarsal (MT1) angle over the course of 1000 cycles of postoperative loading. The angular change is relative to the normal, intact condition; positive values on the y axis indicate overcorrection. Data points represent the means of all tested specimens. Treatment 2 approached complete loss of correction by cycle 1000, as the mean sagittal-plane flattening resulting from the flattening process was 8.3°. Note that the x axis is nonlinear prior to load cycle 100.
e45(5) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
Fig. 3
Progression of the talus-navicular angle over the course of 1000 cycles of postoperative loading. The angular change is relative to the normal, intact condition; positive values on the y axis indicate overcorrection. Data points represent the means of all tested specimens. Treatment 2 approached the initial coronal flattening of 6.0° by load cycle 1000. Note that the x axis is nonlinear prior to load cycle 100.
For each treatment, progressive loss of some of the initial talus-navicular deformity correction occurred during postoperative loading, with the correction achieved by Treatment 2 again deteriorating almost completely by load cycle 1000 (Fig. 3, Table IV). The correction maintained by Treatments 1, 3, 4, 5, and 6 exceeded that of Treatment 2 at 100, 500, and 1000 load cycles (p < 0.001). No significant difference among Treatments 1, 3, 4, 5, and 6 was demonstrable. The addition of spring ligament repair to Treatment 3 did not provide significant further correction, but there was a consistent trend of slightly increased resistance to reflattening of the talus-first metatarsal and talusnavicular angles. Discussion o our knowledge, the current literature is limited to two clinical comparisons of medial displacement calcaneal osteotomy and lateral column lengthening accompanied by flexor digitorum longus transfer. Marks et al.23 demonstrated improvements in gait and radiographic measures with both procedures. The medial displacement calcaneal osteotomy resulted in better first ray plantar flexion and varus, whereas lateral column lengthening resulted in greater heel inversion. Bolt et al.24 reported more effective radiographically assessed correction with flexor digitorum longus transfer accompanied by lateral column lengthening than with tendon transfer and medial displacement calcaneal osteotomy. Because of interest in correction procedures for flatfoot deformity, laboratory models have been developed in an effort to reproduce the condition and assess treatments. Most of these models have limited clinical applicability for reasons including overly simplistic specimen loading without tensioning of relevant extrinsic tendons28,43-46, severe axial loading limitations imposed by electromagnetic bone tracking systems25,27-30,45,47, artificial constraint or nonloading of the fibula27-30,43-51, grossly subphysiologic loading of the Achilles tendon25,27,28,30,47-51, and
T
BIOMECHANICAL ASSESSMENT CORRECTION
OF
F L E X I B L E F L AT F O O T
lack of documentation of correction loss during postoperative loading challenges25,27-30,43-48,51. In 1997, Kitaoka et al.27 introduced simultaneous axial and tendon loading during quantification of experimental foot flattening and correction. Although the method was groundbreaking, the loads applied (111 N for the axial load; 284 N for the sum of the loads on the tendons, including the Achilles tendon) were small and the fibula was unnaturally constrained by rigid embedding. The loading limitations were the consequence of the plastic loading apparatus necessitated by use of an electromagnetic motion measurement system that was incompatible with the metals from which higher-capacity loading devices are typically fabricated. More recently, Blackman et al. examined Achilles tendon over-pull in the presence of experimentally created flatfoot in 200925. Their model incorporated several improvements over earlier efforts. However, although flattening was achieved with use of a conventional testing machine, loads during subsequent bone angle measurements were limited to only 25% of physiologic magnitude because of the plastic loading frame and tendon clamps dictated by the use of electromagnetic bone tracking. The scale of the induced osseous changes was correspondingly quite small. Our model incorporates desirable features of earlier models and allows application of fully 50% of physiologic stance-phase loading to each relevant extrinsic tendon, including the Achilles tendon, as well as dynamic application of 50% of physiologic axial load. The high axial loads are enabled by the novel use of electronic clinometers to monitor critical bone orientations. Unlike electromagnetic tracking systems, these devices are unaffected by the large amounts of metal inherent in standard testing machines. They are particularly useful for continuous monitoring of correction loss over time, and to our knowledge the current study is the first study of flatfoot deformity to include this technology. Postoperative loss of correction has been noted repeatedly in the clinical setting5,24,52-56. The gradual decrease in effectiveness of the examined procedures appeared to be the result of elongation of critical ligaments, but quantification of this phenomenon was beyond the scope of this study. The duration of each 1000-cycle postoperative load challenge (approximately one hour) was shorter than the time span over which a patient is likely to apply 1000 weight-bearing cycles. Therefore, viscoelastic ligament elongation in response to the applied loading would not be expected to duplicate that occurring in vivo. It should be noted that the clinometers can monitor bone angles only in the sagittal, coronal, and other vertical planes. Their design utilizes a dielectric fluid that acts as a pendulum that responds to gravitational force. Consequently, they cannot assess forefoot abduction or the transverse-plane talus-navicular angle referred to as the ‘‘talonavicular coverage angle’’ often included in clinical quantification of flatfoot progression. We measured flattening on the basis of the classic sagittal-plane talus-first metatarsal angle, shown by Chu et al.50 to be the most sensitive indicator of flattening, and on the basis of the less commonly used coronal-plane talus-navicular angle, shown by Kido et al.42 to be the most marked talonavicular joint change in patients with flatfoot.
e45(6) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
The order in which the evaluated correction procedures were performed was randomized to the greatest extent possible, but Treatments 1, 2, and 3 were performed in a fixed order for technical reasons. Specifically, robust fixation of the medial displacement calcaneal osteotomy (Treatment 2) was essential because of the large axial and Achilles tendon forces, and ‘‘undoing’’ and then securely reestablishing this osteotomy proved problematic. Therefore, it was performed after lateral column lengthening (Treatment 1) and before Treatments 3 through 6. The effects of this lack of randomization are unknown. Our study has other methodological limitations. The model cannot incorporate healing of ligament repairs and tendon transfers. Furthermore, despite improvements in specimen loading compared with earlier studies, the combined static tendon and dynamic axial loading only grossly approximates physiologic stance-phase loading. However, in the present study we have shown the method to be a useful means of clearly demonstrating performance differences among correction procedures. Although a broad spectrum of flatfoot deformity is seen clinically, it was necessary to focus on a specific common manifestation of the disorder for this study. Therefore, caution should be exercised in extrapolating the results to other flatfoot stages. Similarly, there were practical limitations to the number of corrective procedures that could be performed sequentially on one set of ten specimens. We recognize that other osteotomies, such as the medial cuneiform plantar flexion osteotomy, can be effective in correcting sagittal-plane deformity. However, Bluman et al. indicated that the most used correction techniques for stage-IIB deformity include the medial displacement calcaneal osteotomy, lateral column lengthening, and flexor digitorum longus transfer57. Hindfoot coronal-plane correction is important but was not specifically measured in the present study. There is evidence, however, that the correction procedures that were performed also address the hindfoot deformity. This was shown by Mosca58, who demonstrated that lateral column lengthening provides improved hindfoot valgus. The medial displacement calcaneal osteotomy provides hindfoot coronal correction as well, by correcting the valgus deforming moment of the Achilles tendon59. In contrast to our laboratory findings, combined flexor digitorum longus transfer and medial displacement calcaneal osteotomy (Treatment 2 in the present study) has had some clinical success. Myerson et al.60 reported a mean AOFAS (American Orthopaedic Foot & Ankle Society) Ankle-Hindfoot Scale score of 79 out of 100 in 129 patients, concluding that this technique could be used for patients with flexible flatfoot, negligible forefoot supination, and
BY
T HE J OURNAL
OF
B ONE
AND J OINT
S URGERY, I NCORPORATED
Biomechanical Assessment of Flexible Flatfoot Correction Comparison of Techniques in a Cadaver Model Diego H. Zanolli, MD, Richard R. Glisson, BS, James A. Nunley II, MD, and Mark E. Easley, MD Investigation performed at the Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina
Background: Options for surgical correction of acquired flexible flatfoot deformity involve bone and soft-tissue reconstruction. We used an advanced cadaver model to evaluate the ability of key surgical procedures to correct the deformity and to resist subsequent loss of correction. Methods: Stage-IIB flatfoot deformity was created in ten cadaver feet through ligament sectioning and repetitive loading. Six corrective procedures were evaluated: (1) lateral column lengthening, (2) medial displacement calcaneal osteotomy with flexor digitorum longus transfer, (3) Treatment 2 plus lateral column lengthening, (4) Treatment 3 plus ‘‘pants-overvest’’ spring ligament repair, (5) Treatment 3 plus spring ligament repair with use of the distal posterior tibialis stump, and (6) Treatment 3 plus spring ligament repair with suture and anchor. Correction of metatarsal dorsiflexion and of navicular eversion were quantified initially and periodically during postoperative cyclic loading. Results: Metatarsal dorsiflexion induced by arch flattening was initially corrected by 5.5° to 10.6°, depending on the procedure. Navicular eversion was initially reduced by 2.1° to 7.7°. The correction afforded by Treatments 1, 3, 4, 5, and 6 exceeded that of Treatment 2 initially and throughout postoperative loading. Inclusion of spring ligament repair did not significantly enhance correction. Conclusions: Under the tested conditions, medial displacement calcaneal osteotomy with flexor digitorum longus tendon transfer was inferior to the other evaluated treatments for stage-IIB deformity. Procedures incorporating lateral column lengthening provided the most sagittal and coronal midfoot deformity correction. Addition of spring ligament repair to a combination of these three procedures did not substantially improve correction. Clinical Relevance: An understanding of treatment effectiveness is essential for optimizing operative management of symptomatic flatfoot deformity. This study provides empirical evidence of the advantage of lateral column lengthening and novel information on resistance to postoperative loss of correction.
Peer Review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. It was also reviewed by an expert in methodology and statistics. The Deputy Editor reviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication. Final corrections and clarifications occurred during one or more exchanges between the author(s) and copyeditors.
T
ibialis posterior tendon dysfunction is a common cause of adult-acquired flatfoot deformity1-3. Surgical options for stage-II dysfunction, which involves tibialis posterior tendon elongation, hindfoot valgus, forefoot abduction, and arch collapse, include osseous realignment procedures and combinations of osseous correction and soft-tissue reconstruction1,2,4-7.
Contemporary realignment procedures include medial displacement calcaneal osteotomy and lateral column lengthening. Medial displacement calcaneal osteotomy shifts the Achilles tendon insertion relative to the subtalar joint axis, reducing hindfoot valgus deforming forces8-10. It is commonly combined with tenodesis of the flexor digitorum longus tendon to the compromised tibialis posterior tendon or to the medial aspect
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
J Bone Joint Surg Am. 2014;96:e45(1-8)
d
http://dx.doi.org/10.2106/JBJS.L.00258
e45(2) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
OF
F L E X I B L E F L AT F O O T
Materials and Methods Specimen Preparation
TABLE I Correction Procedures Treatment
BIOMECHANICAL ASSESSMENT CORRECTION
Description
1
Lateral column lengthening
2
Medial displacement calcaneal osteotomy and flexor digitorum longus transfer to the posterior tibialis tendon
3
Medial displacement calcaneal osteotomy and flexor digitorum longus transfer to the posterior tibialis tendon plus lateral column lengthening
4
Treatment 3 plus spring ligament repair by a pants-over-vest technique
5
Treatment 3 plus spring ligament repair by use of the distal posterior tibialis stump
6
Treatment 3 plus spring ligament repair by use of suture and anchor
of the navicular in an effort to restore transverse tarsal joint adduction and subtalar joint inversion11-13. Lateral column lengthening simultaneously adducts the foot at the talonavicular joint, plantar flexes the midfoot, and derotates the hindfoot out of valgus, restoring the medial longitudinal arch14-16. Combining lateral column lengthening with flexor digitorum longus transfer and medial displacement calcaneal osteotomy decreases pressure across the lateral column of the foot compared with lateral column lengthening alone8,17. When medial arch collapse is accompanied by attenuation or rupture of the spring ligament and uncovering of the talonavicular joint, spring ligament reconstruction may be performed18. Several techniques have been described, including ‘‘pants-over-vest’’ repair19,20, augmentation with use of the distal stump of the tibialis posterior tendon21, and peroneus longus autograft tendon transfer18. There is, however, no consensus regarding indications and technique for spring ligament reconstruction18. Because of the range of surgical options, it is critical that the surgeon understands the extent of flatfoot correction that each option may be expected to achieve and maintain. The orthopaedic literature offers only ‘‘fair’’ and ‘‘insufficient’’ levels of evidence for flatfoot correction by medial displacement calcaneal osteotomy and lateral column lengthening, respectively22, and is limited to two comparative clinical analyses of these procedures23,24. To address the lack of biomechanical studies directly comparing the effects of the hindfoot osteotomies, the present study was designed to objectively measure the correction achieved by common surgical procedures, utilizing an advanced cadaver model in which stage-IIB (flexible) flatfoot deformity was created. We hypothesized that treatments incorporating lateral column lengthening would provide better correction of medial longitudinal arch flattening and navicular eversion relative to the talus compared with treatments not incorporating such lengthening. We further hypothesized that spring ligament reconstruction would increase the correction achievable by the bone realignment procedures.
T
en normal, fresh frozen lower limbs sectioned at the midpoint of the tibia and fibula were used for this institutional review board-approved study. The mean age of the eight male and two female donors was fifty-seven years (range, forty-three to sixty-nine years). Proximal soft tissues were removed to 4 cm above the ankle, preserving the interosseous membrane and extrinsic tendons. The proximal ends of the tibialis posterior, tibialis anterior, peroneus longus, peroneus brevis, flexor digitorum longus, flexor hallucis longus, extensor digitorum longus, and extensor hallucis longus tendons were sutured to plastic rings to provide secure attachment points for subsequent tensile load application. Steel rods were cemented into the medullary canals of the tibia and fibula with use of polymethylmethacrylate cement. The tibial rod was clamped to the linear actuator of a servohydraulic testing machine (Model 1321; Instron, Norwood, Massachusetts) such that the tibia was vertical and the foot sole was horizontal. This neutral flexion position was maintained throughout testing. A calibrated spring applied a downward force to the fibula equaling approximately 15% of the axial load applied to the tibia during the ensuing 25,26 testing . The hindfoot and forefoot rested on separate glass plates isolated from the testing machine table by ball bearings, allowing the foot to establish its preferred transverse-plane orientation and minimizing foot deformation con27-30 straints during loading . Three 5-mm half-pins were inserted for attachment of electronic angular measurement devices. The first projected medially and superiorly from the talar neck, the second projected superiorly from the navicular, and the third projected superiorly from the first metatarsal. Electronic clinometers (Schaevitz AccuStar; Measurement Specialties, Hampton, Virginia) were attached to the pins (Fig. 1). The outputs of these devices corresponded to their rotation in the vertical plane, and they had an operating range of 90° and a linearity of ±0.1%. 31-35 Similar devices have previously been used to measure relative bone rotations . The clinometers attached to the talus and the first metatarsal were aligned in the sagittal plane parallel to the second metatarsal to document the inclination of
Fig. 1
A specimen mounted in the testing machine for cyclic axial loading with simultaneous tendon loading while key osseous relationships were monitored. The two outer clinometers attached to the talus and the first metatarsal monitored their relative angulation in the sagittal plane. The two inner clinometers in the photograph were attached to the talus and the navicular for measurement of their relative angulation in the coronal plane. Axial load was applied through the tibia, and a proportionally smaller load was applied to the fibula with use of a spring, allowing fibular motion during specimen loading.
e45(3) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
TABLE II Initial Correction Relative to the Flattened Condition* Correction Relative to Flattened Condition (deg)
Treatment
First Metatarsal Plantar Flexion Relative to Talus
Navicular Inversion Relative to Talus
1
10.6 ± 3.4
6.6 ± 4.5
2
5.5 ± 2.3
2.1 ± 1.8
3
8.8 ± 4.1
6.7 ± 3.2
4
8.8 ± 4.3
7.2 ± 4.0
5
8.8 ± 4.4
7.0 ± 4.4
6
9.1 ± 4.7
7.7 ± 4.7
*Measured on the first load cycle of the postoperative cyclic loading. The values are given as the mean and the standard deviation. For comparison, the mean flattened talus-first metatarsal angle was 8.3° of metatarsal dorsiflexion relative to normal, and the mean flattened talus-navicular angle was 6.0° of navicular eversion relative to normal.
the talus relative to the first metatarsal, corresponding to the primary clinical measure of medial longitudinal arch flattening. Two additional clinometers attached to the talus and the navicular in a plane perpendicular to the second metatarsal measured their relative coronal-plane inclination. Cables, pulleys, and dead weights were used to apply tensile loads to the extrinsic tendons during axial loading of the foot, similar to the methodology 25 described by Blackman et al. . Cables attached to each of the tendons passed over pulleys that aligned the tendons in an approximately physiologic line of pull. Because of the large forces applied to the Achilles tendon, it was grasped in a specially designed clamp that compressed the tendon tissue between textured 36 metal plates (see Appendix) . An appropriately sized weight was suspended from each cable (see Appendix), applying approximately 50% of the load transmitted through the 37,38 tendon at the midstance phase of gait . The weights were raised and lowered simultaneously by means of a modified hydraulic motorcycle jack to unload and load the tendons.
BIOMECHANICAL ASSESSMENT CORRECTION
OF
F L E X I B L E F L AT F O O T
to mimic insufficiency. The 1780-N peak load was 50% greater than the load applied during preconditioning (and later during evaluation of the correction methods), and it was based on preliminary testing to determine a load magnitude that would expeditiously produce flattening representative of stage-IIB deformity within 1000 load cycles. The minimum acceptable flattening was established as 4° in each plane on the basis of clinical measures of first metatarsal dorsiflexion in patients with flatfoot and navicular eversion measured by 42 Kido et al. in their computed tomography (CT) study of flatfoot deformity . A 1187-N load was then applied as before, and the sagittal-plane talus-first metatarsal angle and the coronal-plane talus-navicular angle were recorded to document the amount of flattening attained.
Correction Procedures The evaluated procedures are summarized in Table I. Treatments 1, 2, and 3 were performed in that order in each specimen, and Treatments 4, 5, and 6 were performed in a random order. Treatment 1 involved lateral column lengthening by means of a calcaneal osteotomy 1.5 cm proximal to the calcaneocuboid joint, preserving the medial cortex. The osteotomy was fixed with an 8-mm BIOFOAM Evans Wedge (Wright Medical Technology, Arlington, Tennessee) and a two-hole 3.5-mm CHARLOTTE CLAW Plate with two screws (Wright Medical Technology). Between Treatments 1 and 2, the lateral column lengthening was undone by removing the wedge and plate and stabilizing the osteotomy with a plate and two screws. Treatment 2 involved a medial displacement calcaneal osteotomy, consisting of a transverse osteotomy 1 cm posterior and parallel to the peroneal tendons, translation of the tuberosity 1 cm medially, and fixation with two 7.0-mm cannulated screws (Synthes, West Chester, Pennsylvania). In addition, the flexor digitorum longus tendon was transected proximal to the knot of Henry, then attached to the tibialis posterior tendon through a hole in the medial third of the navicular by means of number-5 ETHIBOND suture (Ethicon, Somerville, New Jersey). In Treatment 3, lateral column lengthening was reestablished and combined with the medial displacement calcaneal osteotomy and tendon transfer. Treatments 4, 5, and 6 involved the addition of spring ligament repair to Treatment 3. In Treatment 4, this repair was performed with use of a pants-over-vest tech19,20 nique and four number-5 ETHIBOND sutures . In Treatment 5, the repair was performed by suturing the distal posterior tibialis tendon stump to the proximal part of the spring ligament by means of a number-5 ETHIBOND 21 suture . In Treatment 5, the repair was performed by inserting one 3.5-mm
Preconditioning Each specimen was preconditioned by a cyclic axial (downward-directed) load applied through the tibia, accompanied by the constant upward-directed extrinsic tendon loads. One thousand load cycles oscillating between 808 N and 1187 N were applied with use of a displacement rate of 1.5 mm/sec. The 808-N lower load exceeded the sum of the opposing tendon loads by 50 N, thereby maintaining the sole of the foot in contact with the testing machine table. The 1187-N peak load equaled the sum of the tendon loads plus 50% of the vertical ground reaction force typical during the stance phase of gait, and was equivalent to approximately 125% 39,40 of body weight . The axial load and the four clinometer outputs were monitored and recorded during the 1187-N peak loading of the 1000th load cycle. These measures of normal bone alignment under load were used as a reference to quantify both the deformity induced during the ensuing flattening process and the correction achieved by each subsequent treatment.
Flattening Process To facilitate flattening, the talocalcaneal interosseous, plantar naviculocuneiform, plantar first metatarsocuneiform, and anterior superficial deltoid liga25,41 ments were sectioned . The superomedial and inferomedial portions of the spring ligament were sectioned at the midsubstance. Axial loading between 808 N and 1780 N was then applied for 1000 cycles, accompanied by extrinsic tendon loading as described above but excluding the tibialis posterior tendon
TABLE III Initial Correction Relative to the Normal, Intact Condition* Correction Relative to Normal, Intact Condition (deg) Treatment
Talus-First Metatarsal Angle, Sagittal
Talus-Navicular Angle, Coronal
1
12.3 ± 4.6
10.6 ± 4.4
2
22.8 ± 4.2
23.9 ± 1.9
3
10.5 ± 6.1
10.8 ± 3.4
4
10.5 ± 6.2
11.2 ± 4.4
5
10.6 ± 6.0
11.0 ± 4.8
6
10.9 ± 6.3
11.7 ± 5.0
*Measured on the first load cycle of the postoperative cyclic loading. The values are given as the mean and the standard deviation. Positive mean numbers indicate slight overcorrection; negative mean numbers indicate failure to achieve complete correction.
e45(4) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
TABLE IV Loss of Sagittal-Plane and Coronal-Plane Correction* Loss of Initial Correction During Cyclic Loading (deg) Talus-First Metatarsal Angle, Sagittal
Talus-Navicular Angle, Coronal
Treatment
Mean
Std. Dev.
Mean
Std. Dev.
1
2.7
1.2
0.8
1.0
2
3.8
1.4
1.7
0.7
3
3.1
1.1
0.6
1.6
4
2.6
1.1
0.6
1.1
5
2.5
0.9
0.7
0.8
6
2.4
0.8
0.7
1.0
*Between postoperative load cycles 1 and 1000.
Corkscrew FT Suture Anchor (Arthrex, Naples, Florida) into the navicular and suturing it to the proximal part of the spring ligament.
Assessment of Resistance to Loss of Correction After each correction procedure, the cyclic loading regimen used to precondition the specimens (1000 cyclic loads to 1187 N with accompanying tendon loads) was repeated. The clinometer outputs at the first load peak indicated the initial correction achieved relative to both the normal, intact condition and the flattened condition. To document resistance to loss of correction, clinometer data were also recorded on load cycles 2, 3, 4, 5, 10 through 90 in 10-cycle increments, and 100 through 1000 in 100-cycle increments. Flattening relative to the intact state was calculated at each time point and was plotted graphically. Repeated-measures analysis of variance (ANOVA) with Bonferroni post hoc testing was used to identify significant performance differences among the six treatments at postoperative load cycles 1 (reflecting initial correction), 100, 500, and 1000.
BIOMECHANICAL ASSESSMENT CORRECTION
OF
F L E X I B L E F L AT F O O T
angle correction (p = 0.001) at load cycle 1. Post hoc testing indicated that the initial correction achieved by Treatment 1 (lateral column lengthening), Treatment 3 (lateral column lengthening plus medial displacement calcaneal osteotomy plus flexor digitorum longus transfer transfer), and Treatments 4, 5, and 6 (Treatment 3 plus spring ligament repair) all significantly exceeded that achieved by Treatment 2 (medial displacement calcaneal osteotomy plus flexor digitorum longus transfer) (p < 0.05). With the group sizes used in this study, no significant differences in initial sagittal correction among Treatments 1, 3, 4, 5, and 6 were demonstrable. Similarly, there was a significant interaction effect between treatment and the extent of talus-navicular angle correction in the coronal plane (p < 0.001). As in the case of the talus-first metatarsal angle, the initial correction achieved by Treatments 1, 3, 4, 5, and 6 each significantly exceeded that achieved by Treatment 2 (p < 0.001). Resistance of Correction to Reflattening For each treatment, gradual loss of some of the talus-first metatarsal angle correction occurred as postoperative cyclic loading progressed (Fig. 2, Table IV). The correction achieved by Treatment 2 deteriorated almost completely by load cycle 1000. The correction maintained by Treatments 1, 3, 4, 5, and 6 each exceeded that of Treatment 2 at load cycles 100, 500, and 1000 (p < 0.02). With the available group sizes, no significant differences in correction maintenance were demonstrable among Treatments 1, 3, 4, 5, and 6. The addition of spring ligament repair to Treatment 3 provided only slight, statistically nonsignificant, additional resistance to loss of correction. There was no evidence of suture or tissue failure at the sites of the spring ligament repairs or tendon transfers, and there was no evidence of structural failure at the osteotomy sites.
Source of Funding There was no external funding for this investigation.
Results Flattening lattening resulted in reduction of the sagittal-plane talusfirst metatarsal angle and eversion of the navicular relative to the talus, paralleling changes seen clinically in moderate flatfoot deformity4,5,13,42. The mean medial longitudinal arch flattening was 8.3° (range, 4.8° to 16.3°). The mean change in the coronal talus-navicular angle was 6.0° (range, 4.2° to 8.5°).
F
Initial Deformity Correction The initial correction achieved by each of the six treatments was substantial (Table II). Talus-first metatarsal angle correction ranged from 5.5° to 10.6°, and talus-navicular angle correction ranged from 2.1° to 7.7°, depending on the procedure. With the exception of Treatment 2 (medial displacement calcaneal osteotomy), all correction methods initially slightly overcorrected both the talus-first metatarsal and talus-navicular deformities (Table III). The ANOVA results indicated a significant interaction effect between treatment and the extent of talus-first metatarsal
Fig. 2
Progression of the talus-first metatarsal (MT1) angle over the course of 1000 cycles of postoperative loading. The angular change is relative to the normal, intact condition; positive values on the y axis indicate overcorrection. Data points represent the means of all tested specimens. Treatment 2 approached complete loss of correction by cycle 1000, as the mean sagittal-plane flattening resulting from the flattening process was 8.3°. Note that the x axis is nonlinear prior to load cycle 100.
e45(5) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
Fig. 3
Progression of the talus-navicular angle over the course of 1000 cycles of postoperative loading. The angular change is relative to the normal, intact condition; positive values on the y axis indicate overcorrection. Data points represent the means of all tested specimens. Treatment 2 approached the initial coronal flattening of 6.0° by load cycle 1000. Note that the x axis is nonlinear prior to load cycle 100.
For each treatment, progressive loss of some of the initial talus-navicular deformity correction occurred during postoperative loading, with the correction achieved by Treatment 2 again deteriorating almost completely by load cycle 1000 (Fig. 3, Table IV). The correction maintained by Treatments 1, 3, 4, 5, and 6 exceeded that of Treatment 2 at 100, 500, and 1000 load cycles (p < 0.001). No significant difference among Treatments 1, 3, 4, 5, and 6 was demonstrable. The addition of spring ligament repair to Treatment 3 did not provide significant further correction, but there was a consistent trend of slightly increased resistance to reflattening of the talus-first metatarsal and talusnavicular angles. Discussion o our knowledge, the current literature is limited to two clinical comparisons of medial displacement calcaneal osteotomy and lateral column lengthening accompanied by flexor digitorum longus transfer. Marks et al.23 demonstrated improvements in gait and radiographic measures with both procedures. The medial displacement calcaneal osteotomy resulted in better first ray plantar flexion and varus, whereas lateral column lengthening resulted in greater heel inversion. Bolt et al.24 reported more effective radiographically assessed correction with flexor digitorum longus transfer accompanied by lateral column lengthening than with tendon transfer and medial displacement calcaneal osteotomy. Because of interest in correction procedures for flatfoot deformity, laboratory models have been developed in an effort to reproduce the condition and assess treatments. Most of these models have limited clinical applicability for reasons including overly simplistic specimen loading without tensioning of relevant extrinsic tendons28,43-46, severe axial loading limitations imposed by electromagnetic bone tracking systems25,27-30,45,47, artificial constraint or nonloading of the fibula27-30,43-51, grossly subphysiologic loading of the Achilles tendon25,27,28,30,47-51, and
T
BIOMECHANICAL ASSESSMENT CORRECTION
OF
F L E X I B L E F L AT F O O T
lack of documentation of correction loss during postoperative loading challenges25,27-30,43-48,51. In 1997, Kitaoka et al.27 introduced simultaneous axial and tendon loading during quantification of experimental foot flattening and correction. Although the method was groundbreaking, the loads applied (111 N for the axial load; 284 N for the sum of the loads on the tendons, including the Achilles tendon) were small and the fibula was unnaturally constrained by rigid embedding. The loading limitations were the consequence of the plastic loading apparatus necessitated by use of an electromagnetic motion measurement system that was incompatible with the metals from which higher-capacity loading devices are typically fabricated. More recently, Blackman et al. examined Achilles tendon over-pull in the presence of experimentally created flatfoot in 200925. Their model incorporated several improvements over earlier efforts. However, although flattening was achieved with use of a conventional testing machine, loads during subsequent bone angle measurements were limited to only 25% of physiologic magnitude because of the plastic loading frame and tendon clamps dictated by the use of electromagnetic bone tracking. The scale of the induced osseous changes was correspondingly quite small. Our model incorporates desirable features of earlier models and allows application of fully 50% of physiologic stance-phase loading to each relevant extrinsic tendon, including the Achilles tendon, as well as dynamic application of 50% of physiologic axial load. The high axial loads are enabled by the novel use of electronic clinometers to monitor critical bone orientations. Unlike electromagnetic tracking systems, these devices are unaffected by the large amounts of metal inherent in standard testing machines. They are particularly useful for continuous monitoring of correction loss over time, and to our knowledge the current study is the first study of flatfoot deformity to include this technology. Postoperative loss of correction has been noted repeatedly in the clinical setting5,24,52-56. The gradual decrease in effectiveness of the examined procedures appeared to be the result of elongation of critical ligaments, but quantification of this phenomenon was beyond the scope of this study. The duration of each 1000-cycle postoperative load challenge (approximately one hour) was shorter than the time span over which a patient is likely to apply 1000 weight-bearing cycles. Therefore, viscoelastic ligament elongation in response to the applied loading would not be expected to duplicate that occurring in vivo. It should be noted that the clinometers can monitor bone angles only in the sagittal, coronal, and other vertical planes. Their design utilizes a dielectric fluid that acts as a pendulum that responds to gravitational force. Consequently, they cannot assess forefoot abduction or the transverse-plane talus-navicular angle referred to as the ‘‘talonavicular coverage angle’’ often included in clinical quantification of flatfoot progression. We measured flattening on the basis of the classic sagittal-plane talus-first metatarsal angle, shown by Chu et al.50 to be the most sensitive indicator of flattening, and on the basis of the less commonly used coronal-plane talus-navicular angle, shown by Kido et al.42 to be the most marked talonavicular joint change in patients with flatfoot.
e45(6) TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O L U M E 96-A N U M B E R 6 M A R C H 19, 2 014 d
d
d
The order in which the evaluated correction procedures were performed was randomized to the greatest extent possible, but Treatments 1, 2, and 3 were performed in a fixed order for technical reasons. Specifically, robust fixation of the medial displacement calcaneal osteotomy (Treatment 2) was essential because of the large axial and Achilles tendon forces, and ‘‘undoing’’ and then securely reestablishing this osteotomy proved problematic. Therefore, it was performed after lateral column lengthening (Treatment 1) and before Treatments 3 through 6. The effects of this lack of randomization are unknown. Our study has other methodological limitations. The model cannot incorporate healing of ligament repairs and tendon transfers. Furthermore, despite improvements in specimen loading compared with earlier studies, the combined static tendon and dynamic axial loading only grossly approximates physiologic stance-phase loading. However, in the present study we have shown the method to be a useful means of clearly demonstrating performance differences among correction procedures. Although a broad spectrum of flatfoot deformity is seen clinically, it was necessary to focus on a specific common manifestation of the disorder for this study. Therefore, caution should be exercised in extrapolating the results to other flatfoot stages. Similarly, there were practical limitations to the number of corrective procedures that could be performed sequentially on one set of ten specimens. We recognize that other osteotomies, such as the medial cuneiform plantar flexion osteotomy, can be effective in correcting sagittal-plane deformity. However, Bluman et al. indicated that the most used correction techniques for stage-IIB deformity include the medial displacement calcaneal osteotomy, lateral column lengthening, and flexor digitorum longus transfer57. Hindfoot coronal-plane correction is important but was not specifically measured in the present study. There is evidence, however, that the correction procedures that were performed also address the hindfoot deformity. This was shown by Mosca58, who demonstrated that lateral column lengthening provides improved hindfoot valgus. The medial displacement calcaneal osteotomy provides hindfoot coronal correction as well, by correcting the valgus deforming moment of the Achilles tendon59. In contrast to our laboratory findings, combined flexor digitorum longus transfer and medial displacement calcaneal osteotomy (Treatment 2 in the present study) has had some clinical success. Myerson et al.60 reported a mean AOFAS (American Orthopaedic Foot & Ankle Society) Ankle-Hindfoot Scale score of 79 out of 100 in 129 patients, concluding that this technique could be used for patients with flexible flatfoot, negligible forefoot supination, and
