531226
research-article2014
FAIXXX10.1177/1071100714531226Foot & Ankle InternationalHetsroni et al
Article
Plantar Pressure Anomalies After Open Reduction With Internal Fixation of High-Grade Calcaneal Fractures
Foot & Ankle International® 2014, Vol. 35(7) 712–718 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100714531226 fai.sagepub.com
Iftach Hetsroni, MD1,2, David Ben-Sira, PhD3, Meir Nyska, MD1,2, and Moshe Ayalon, PhD3
Abstract Background: Plantar pressure abnormalities after open reduction with internal fixation (ORIF) of intra-articular calcaneal fractures have been observed previously, but high-grade fractures were not selectively investigated and follow-up times were shorter than 2 years. The purpose of this study was to characterize plantar pressure anomalies in patients with exclusively high-grade calcaneal fractures after ORIF with a minimum 2 years of follow-up, and to test the association between plantar pressure distribution and the clinical outcome. Methods: The orthopaedic registry was reviewed to identify patients with isolated high-grade calcaneal fractures (Sanders types III-IV) who were operated on and had a minimum 2 years of follow-up. Sixteen patients were evaluated. Mean age was 47 years and follow-up was between 2 and 6 years. The Pedar-Mobile system was used to measure 3 loading and 3 temporal variables and compare these between the operated and the uninjured limbs. Results: Mean American Orthopaedic Foot and Ankle Society (AOFAS) score was 76 ± 7 at latest follow-up. Bohler’s angle was 5 ± 8 degrees before surgery and 25 ± 7 degrees at latest follow-up. Stance was shorter in operated limbs (P = .001). Timing of the peak of pressure was delayed in operated limbs under the hallux and the second toe (P ≤ .03). Peak pressure, force time integral, and pressure time integral were increased under the lateral midfoot (P ≤ .03) and decreased under the second metatarsal (P ≤ .03). Force time integral was decreased under the first metatarsal (P = .02) and under the hallux and the lateral toes (P ≤ .05). Increased loading under the lateral midfoot and decreased loading under the lateral toes were correlated with poorer clinical outcome (r = –.53, P < .05, and r = .63, P < .01, respectively). Conclusions: Side-to-side plantar pressure mismatch persisted at more than 2 years after ORIF of high-grade calcaneal fractures performed via lateral approach, despite improvement of Bohler’s angle. This was characterized by shortened stance phase, delayed timing of peak of pressure under the hallux and second toe, lateral load shift at the midfoot, and decreased toe pressures in operated limbs. Since loading abnormalities were correlated with the clinical outcome, modifications in treatment strategy that can improve foot loading may be desirable in these cases. Level of Evidence: Level III, case control. Keywords: calcaneal fracture, Bohler’s angle, plantar pressure, loading variables, temporal variables
Intra-articular displaced fractures of the calcaneus represent a complex injury, typically affecting the middle-aged active population.21 Although nonoperative management and open reduction with internal fixation (ORIF) are both considered acceptable treatment strategies with no definitive conclusion as to the superiority of either option,1,5-7,11,19 it is accepted that nonoperative treatment avoids the costs and risks of surgery (ie, operative wound healing problems, infection, sural nerve injuries), but ORIF can result in better calcaneal and heel morphology, leading to improved function16 and enabling more comfortable shoe wear.5-6,11 Yet, despite the potential benefits in function, abnormalities in gait pattern
after the operation are still noticeable, most commonly demonstrated through plantar pressure analyses.3,10,16
1
Department of Orthopedic Surgery, Meir General Hospital, Kfar Saba, Israel 2 Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel 3 Biomechanics Laboratory, Zinman College of Physical Education and Sport Sciences, Wingate Institute, Netanya, Israel Corresponding Author: Iftach Hetsroni, MD, Department of Orthopedic Surgery, Meir General Hospital, Tsharnichovski St 59, Kfar Saba 44281, Israel. Email: [email protected]
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
713
Hetsroni et al Interpreting available data about plantar pressure abnormalities in this respect is however subjected to several limitations. First, abnormalities in plantar pressures after surgery have not been adequately linked to specific types of intra-articular calcaneal fractures, but rather reported for mixed groups of fracture types by some investigators.10,20 Since increase in number of fracture elements is associated with poorer outcomes,7,16 possibly because of the lower energy of the primary trauma as well as the relative ease to obtain anatomic reduction with stable fixation during surgery in low-grade injuries (ie, Sanders type II) compared to high grade injuries (ie, Sanders types III and IV),22 outcome after ORIF of calcaneal fractures should ideally be investigated in specific subgroups of fracture type. Moreover, some investigators had inadequate follow-up times, which were shorter than 2 years,8,10,12,16 or did not confirm normalization of Bohler’s angle in all patients to ascertain that the immediate postoperative goal of the surgery was indeed achieved.10 Viewing Sanders types III and IV as a particular subset of complex intra-articular calcaneal fractures (as opposed to Sanders type II) where ongoing debate about the outcome and advisability of surgery compared to nonoperative management still exists,7 a selective analysis of outcome is desirable in these high-grade fractures specifically, with adequate follow-up and acceptable radiographic alignment after ORIF. Because abnormal plantar pressure distribution has been associated with foot pathologies and because normalization of plantar pressures after corrective surgery or application of orthotic devices has been associated with improvement in function,9,23 information about plantar pressure distribution can be used for seeking operative modifications and for planning orthotics, which may improve outcomes in these challenging cases. The purpose of this study was therefore to characterize plantar pressure distribution in patients with exclusively high-grade calcaneal fractures after ORIF with a minimum 2 years of follow-up and to test the association between plantar pressure abnormalities and the clinical outcome. We hypothesized that abnormalities in plantar pressure distribution with predominance of loading lateralization would be demonstrated within the hindfoot and midfoot in operated limbs despite the length of follow-up and normalization of Bohler’s angle, and that greater abnormalities in loading pattern would be associated with inferior clinical outcomes.
Methods Study Population The orthopaedic trauma registry relating to ORIF of closed calcaneal fractures performed in our institution between 2000 and 2004 was reviewed. Inclusion criteria were age 18 years or older, unilaterality of the injury, fracture classified
preoperatively with computed tomography scan as Sanders type III or type IV, and minimum 2 years of follow-up. After excluding 11 patients due to other accompanying lower limb and spine fractures that underwent internal fixation (5 patients), severe knee arthrosis (1 patient), postoperative osteomyelitis of the calcaneus (1 patient), and subtalar arthrodesis performed before this study due to subtalar arthrosis and pain (4 patients), 16 patients were contacted and volunteered to participate in this study. There were 13 men and 3 women. Mean age at the time of the injury was 47 years ± 11 years (range, 18-67 years). Five fractures were classified as Sanders type III and 11 fractures as Sanders type IV. Mean follow-up time was 3 years and 4 months (range, 2-6 years). The operation was performed in all cases within 2 weeks after the injury through a lateral approach by a team of 2 orthopaedic surgeons, specialized in foot and ankle surgery. Quality of reduction was assessed intraoperatively with lateral, axial, and Broden views. Postoperative management included non-weight-bearing for 8-10 weeks, followed by weight-bearing and range of motion and strengthening exercises. Patients were routinely observed in our outpatient clinic for 6 to 12 months after the operation. The latest follow-up visit was performed specifically for the purpose of this study, during which patients completed ankle-hindfoot score according to the American Orthopaedic Foot and Ankle Society (AOFAS)13 and had lateral radiograph to measure Bohler’s angle. The study was approved by our institutional review board, and all participants signed informed consent.
Plantar Pressure Analysis Protocol Plantar pressures were collected during walking on a level floor at a natural preferred cadence using Pedar-Mobile system insoles (Novel Gmbh, Munich, Germany) at a rate of 50Hz. The insoles were inserted under both operated and contralateral uninjured foot, using one shoe design for all participants. Ten zones of the plantar area of the foot were defined by specific subsets of 99 insole sensors. The zone definitions were designed to approximate relevant plantar structures as follows (Figure 1): medial heel, lateral heel, medial midfoot, lateral midfoot, first metatarsal, second metatarsal, lateral metatarsals, first toe (hallux), second toe, and lateral toes. Plantar pressure data were collected over 4 gait cycles for each participant and recorded at the same gait cycle for both the operated and the contralateral uninjured foot. For each step, 3 loading variables and 3 temporal variables were calculated for each foot zone, and the mean of 4 steps for each variable was used to compare the operated limb and the uninjured limb. The 3 loading variables included (1) peak pressure (PkP) [N*cm-2], representing the maximum pressure recorded under the specific foot zone, (2) force time integral (FTI) [N*sec], representing the
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
714
Foot & Ankle International 35(7)
Plantar Pressure Analysis
Figure 1. Plantar zone definitions.
overall amount of force that has been applied over time, and (3) pressure time integral (PTI) [N*cm-2*sec], representing the overall amount of pressure that had been applied over time. The 3 temporal variables included (1) contact time (CT) [msec], representing the duration of contact at the specific foot zone, (2) relative contact time (Relative CT) [%ST], representing the duration of contact at the specific foot zone as percentage of the entire stance time, and (3) timing of peak of pressure (PkPt) [%ST], representing the timing of maximal pressure at the specific foot zone as a percentage of the stance time.
Statistical Analysis Paired Student t tests were used to compare plantar pressure variables between the operated and the uninjured limbs. Pearson’s product–moment correlation coefficient was used to test the relationships between the AOFAS clinical outcome scores and the plantar pressure values. Level of significance was set at .05.
Mean natural gait velocity of the participants was 0.99 ± 0.15 m/s. Descriptive statistics of the loading variables with comparisons between operated and uninjured limbs are presented in Table 1. total force time integral (FTI) and total pressure time integral (PTI) were reduced in the operated limb compared to the uninjured limb by 7% (P = .02 for comparing FTI, and P = .06 for comparing PTI). Viewing specific foot subzones, symmetrical loading was recorded at the heel in the operated limb compared to the uninjured limb, but increased loading was recorded in all 3 variables at the lateral midfoot of the operated limb compared to the uninjured limb (P ≤ .03). Observing the forefoot, reduced loading values were recorded in the operated limb compared to the uninjured limb under the second metatarsal in all variables (P ≤ .03), and under the first metatarsal in FTI (P = .02). Reduced loading values were also recorded in the operated limb under the hallux, second toe, and lateral toes, with regard to FTI and PTI. Assessment of the relationships between AOFAS clinical outcome scores and loading variables that displayed significant differences between the operated and the uninjured limbs showed the following statistically significant findings: At the lateral midfoot, there was a statistically significant and moderate correlation between the AOFAS and FTI (r = –.52, P < .05) and between the AOFAS and PTI (r = –.53, P < .05) (Figure 2). At the lateral toes, there was a statistically significant and moderate correlation between the AOFAS and PkP (r = .61, P < .05), FTI (r = .66, P < .01), and PTI (r = .63, P < .01) (Figure 3). Descriptive statistics of the temporal variables and comparisons between the operated and uninjured limbs are presented in Table 2. Stance duration was reduced in the operated limb compared to the uninjured limb by 4.7% (705 msec vs 740 msec, P = .001). Viewing specific foot subzones, contact time was shorter under the medial heel (P = .05), and under all forefoot zones, including all metatarsal and all toe zones in the operated limb compared to the uninjured limb (P ≤ .04). Timing of peak of pressure (PkPt) was delayed in the operated limb compared to the uninjured limb under the hallux and under the second toe (P ≤ .03). There were no significant bilateral differences in total contact area or in either of the 10 foot subzones contact areas.
Results
Discussion
Mean preoperative Bohler’s angle was 5 ± 8 degrees (range, 0-20 degrees), and mean postoperative Bohler’s angle at latest follow-up was 25 ± 7 degrees (range, 15-35 degrees). Mean latest follow-up AOFAS ankle hindfoot score was 76 ± 7. Specifically, 4 patients had a poor result (AOFAS ≤ 69), 6 had a fair result (AOFAS 70-79), and 6 had a good result (AOFAS 80-89). There were no cases with an excellent result (AOFAS 90-100).
The current literature reveals that functional outcome in patients with intra-articular calcaneal fractures that undergo ORIF is suboptimal, and that deviations in gait pattern is a consistent finding after surgery.3,8,10,12,16,20 However, inadequate follow-up and nonspecific series of patients with regard to the severity of the injury suggest that additional data are desirable to characterize abnormalities in gait and function in these cases with greater injury specificity and
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
715
Hetsroni et al Table 1. Comparisons of Loading Variables Between Operated and Uninjured Limbs. Foot zone
Side
PkP [N*cm-2]
P
FTI [N*sec]
P
Lateral heel
Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured
198.3 ± 66.4 212 ± 47.8 187 ± 58.1 197.8 ± 29.5 140.6 ± 34.4 121.4 ± 19.9 107 ± 28.1 94.5 ± 37.3 208 ± 58.2 208 ± 54.8 207.8 ± 46.6 239.2 ± 50.2 179.8 ± 73.1 195.9 ± 47.1 68.4 ± 37.6 93.3 ± 40.2 103.3 ± 38.3 117.5 ± 49.1 195.8 ± 65.4 237.8 ± 85.8 279.5 ± 51 290.8 ± 52.1
.2
40.8 ± 15.5 43.3 ± 17.4 29 ± 10.1 30.7 ± 12.8 48 ± 20.9 39.5 ± 19.6 8.6 ± 5.3 8.5 ± 5.6 57.2 ± 18.4 57.5 ± 21.9 24.1 ± 7.6 31.4 ± 10.8 18 ± 9.5 23.3 ± 9.1 1.8 ± 1.6 3.1 ± 1.8 4.6 ± 3.1 6.2 ± 3.5 7.9 ± 4.2 11.5 ± 5.3 465.4 ± 119.8 501.6 ± 156.6
.46
Medial heel Lateral midfoot Medial midfoot Lateral MTT Second MTT First MTT Lateral toes Second toe Hallux Total foot
.37 .03 .12 1.0 .03 .4 .05 .22 .07 .49
.59 .03 .95 .96 .001 .02 .05 .07 .03 .02
PTI [N*cm-2*sec] 66 ± 36.3 71.3 ± 30.4 58.5 ± 16.7 63 ± 21 57.7 ± 18.5 49.1 ± 15 37.5 ± 13.8 33.1 ± 13.4 68.8 ± 21.1 72.6 ± 25.2 57.1 ± 17.6 74.6 ± 26.5 49.9 ± 31.6 61 ± 23.7 17.2 ± 10.7 28.3 ± 14.1 24.3 ± 13.7 34.1 ± 21.5 36.8 ± 18 57 ± 22.8 120.1 ± 27.1 129.3 ± 28.9
P .43 .37 .02 .09 .6 .001 .11 .02 .02 .01 .06
FTI, force time integral; MTT, metatarsal; PkP, peak pressure; PTI, pressure time integral. Values are presented as mean ± standard deviation. Statistically significant differences are presented in bold.
Figure 2. The correlation between AOFAS score and pressure time integral (PTI) at the lateral midfoot in operated limbs.
Figure 3. The correlation between AOFAS score and pressure time integral (PTI) at the lateral toes in operated limbs.
longer rehabilitation. The current study was designed to characterize plantar pressure anomalies specifically in patients with high-grade fractures of the calcaneus after ORIF that obtained normalization of Bohler’s angle and had a minimum 2 years of follow-up, and to test the relationships between pressure anomalies and clinical outcomes. Viewing the foot as a whole, reductions observed in total FTI, total PTI, and stance time under the operated limb
compared to the uninjured limb indicate that asymmetric gait persists in these cases despite adequate postoperative duration. This is consistent with other investigators that observed reduced contact time,10 reduced total impulse,20 and shorter pressure duration under the operated foot.3 Such gait asymmetry may be the mere consequence of persistent pain in these operated limbs, most likely resulting from abnormalities and degenerative changes in the subtalar joint.
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
716
Foot & Ankle International 35(7)
Table 2. Comparisons of Temporal Variables Between Operated and Uninjured Limbs. Foot zone
Side
CT [msec]
P
Lateral heel
Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured
596 ± 172 628 ± 162 562 ± 151 607 ± 173 644 ± 120 667 ± 115 524 ± 124 522 ± 160 677 ± 127 714 ± 132 631 ± 110 678 ± 158 586 ± 172 629 ± 166 382 ± 166 524 ± 158 442 ± 146 541 ± 165 469 ± 188 619 ± 147 705 ± 118 740 ± 128
.34
Medial heel Lateral midfoot Medial midfoot Lateral MTT Second MTT First MTT Lateral toes Second toe Hallux Total foot
.05 .23 .92 .01 .03 .04 .03 .01 .02
Relative CT [% ST] 83.6 ± 14.9 84.3 ± 13.2 79.2 ± 14 81 ± 15 91.3 ± 6.8 90.5 ± 8.1 73.9 ± 9.2 70 ± 20.3 95.8 ± 4 96.5 ± 3.9 89.9 ± 7.6 91 ± 7.9 82.3 ± 15.4 83.9 ± 10.6 55.4 ± 23.8 70.8 ± 17.2 63.8 ± 18.9 73.1 ± 17.3 65.8 ± 21.3 84.4 ± 15.8
P
PkPt [% ST]
P
.87
15 ± 4.1 15.5 ± 12.5 15.5 ± 4.7 12.8 ± 4.8 27.3 ± 14.5 26 ± 15.6 20.3 ± 4 22.6 ± 9 75.5 ± 4.6 73.9 ± 6.5 78.5 ± 4.2 76.3 ± 3.4 79.4 ± 4 75.6 ± 4.8 74 ± 14.2 74.9 ± 15.9 81.3 ± 3.8 78.8 ± 4.4 86.7 ± 3.5 82.5 ± 4.7 60.7 ± 27.2 70.4 ± 17.9
.89
.68 .76 .48 .64 .65 .6 .04 .04 .03
.001
.15 .61 .39 .18 .06 .02 .86 .004 .001 .15
CT, contact time; relative CT, contact time at a foot subzone as a percentage of the stance time; MTT, metatarsal; PkPt, = peak pressure timing; % ST, percentage of stance time. Values are presented as mean ± standard deviation. Statistically significant differences are presented in bold.
Previous studies showed that pain in operated limbs after ORIF indeed persisted and was comparable to the level of pain demonstrated in nonoperated limbs with similar fractures.6,11 These patients are therefore likely to spend less time on the affected limb to reduce loading that provokes pain. The fact that Bohler’s angle at latest follow-up indicated acceptable reduction of calcaneal morphology in our series (ie, range 15-35 degrees) suggest that re-creating calcaneal morphology during the operation with stable fixation cannot by itself be considered a sufficient measure to achieve symmetric gait pattern after surgery. The damaged cartilaginous articular surfaces at the subtalar joint, however, cannot be appreciated by using standard intra- or postoperative fluoroscopy or plain radiography. This intra-articular cartilage wear may provide 1 explanation for pain persistence in cases where bone alignment is satisfactory. In this regard, therapeutic alternatives that have been suggested to address cartilage wear, stimulate chondrogenesis, and improve function in affected joints including the use of mesenchymal stem cells17 and temporary distraction fixators.2 These alternatives, however, have not been implemented for cartilage wear at the subtalar joint, but may be tested in the future for this purpose. Viewing foot subzones, the most consistent finding in terms of foot loading was a lateral load shift observed at the midfoot in operated limbs compared to uninjured limbs.
Several mechanisms could have contributed to this finding. The most likely mechanism is a reduction in subtalar range of motion.10,12 Reduced normal subtalar eversion during the stance leads to relatively inverted foot positioning during foot loading, and this can result in a lateral load shift. Lateral load shift may also be related to dysfunction of the gastrocnemius, recorded in this type of injury.4 In case such weakness is secondary to Achilles shortening, it may accentuate as well an inverted foot position. A third associated factor to lateral load shift may be dysfunction of the peroneal tendons. Although we did not test the peroneal tendons strength and function specifically, we believe it is possible because the operative approach used for performing ORIF in all our cases involved a lateral L-shaped incision during which the peroneal tendons were temporarily translocated anteriorly away from their normal location around the peroneal tubercle to expose and then reduce and fix the fragments; it is possible that some dysfunction of these foot evertors persisted after surgery. The sinus tarsi approach, which avoids exposure of the peroneal tendons, is an operative alternative to the lateral approach used in our series, although it is unknown whether it can reduce the lateral load shift that is seen after surgery performed via lateral approach.14 Compartment syndrome of the foot has been recorded in a significant proportion of patients with intra-articular calcaneal fractures, leading to plantar muscle ischemia and
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
717
Hetsroni et al plantar muscle scarring.15 This scarring may result in reduced flexibility of the medial plantar arch, again contributing to a more inverted foot position. Greater surgeons’ awareness of this potential sequela, with intraoperative intracompartmental foot pressure assessment and then compartment release in case of increased pressure, may provide a solution to this problem. And last, the lateral L-shaped approach used in all our cases enabled us to assess with precision the reduction of the fragments at the lateral aspect of the calcaneus as well as subtalar joint congruency, but it may have underestimated medial calcaneal wall integrity in some cases. Thus, theoretically, an underreduced medial column may have been resulted to some extent, contributing to an inverted foot position as well. In regard to forefoot plantar pressure findings, the reductions observed in all loading variables under the second metatarsal and in FTI under the first metatarsal, again support a lateral load shift pattern and are consistent in this respect with findings of previous investigators.20 Reduced loading was observed under all toes in our series as well, and this again may represent toe clawing, resulting from compartment syndrome of the foot, plantar muscle ischemia and eventual scarring.15 As to heel loading pattern, we did not identify side-toside mismatch between the operated and the uninjured limb in any of the loading variables, which is in contrast to other investigators that showed decreased pressure at the medial heel.3,20 This may have been related to the fact Bohler’s angle was re-created in our series and patients were examined at a relatively longer follow-up, potentially enabling longer recovery and adaptations to occur, compared to previous investigations. Plantar pressure distribution was previously measured on normal healthy individuals of similar age-range as ours.18 No significant side-to-side differences in any of the foot loading variables were demonstrated.18 The asymmetric foot loading observed between the operated and the uninjured limb in our operated population should be therefore viewed as representing an abnormal loading pattern. Furthermore, assessment of the relationships between the clinical outcome (ie, AOFAS score) and the abnormal foot loadings showed in this study significant correlations. More pronounced pressure lateralization at the midfoot and reduction in toe pressures were correlated with poorer clinical outcomes (ie, lower AOFAS scores). Although this correlation does not necessarily imply causative relationships, it may strengthen the notion that modifications in treatment which contribute to improvements in loading pattern may as well have positive effects on the clinical outcome. In terms of fabricating orthotic support, which is one alternative to address this issue, reducing the elevated pressures at the lateral midfoot during midfoot support and redistributing pressures at the forefoot to improve the effectiveness of the push-off phase, possibly by introducing hard materials
at the transition zones between the midfoot and forefoot, may be desirable in this population. This requires differential use of materials with specific mechanical properties at the distinct foot areas with respect to both latitudinal and longitudinal distribution. Limitations of this study include the fact only fluoroscopy and plain radiographs were used to assess calcaneal morphology during the operation and at follow-up. While these are acceptable imaging modalities to assess fracture reduction during and after surgery in these cases, CT scans, and potentially 3D CT reconstructions, could have delineated with greater accuracy calcaneal morphology and subtalar joint congruency and indicated whether these cases should have been indeed referred to as postoperatively adequately aligned cases of calcaneal fractures. As CT exposes patients to additional irradiation, we felt it was not justified to perform follow-up CT imaging for the purpose of this study. In conclusion, side-to-side plantar pressure mismatch persisted at more than 2 years after ORIF of high-grade calcaneal fractures performed via lateral approach, despite recreation of Bohler’s angle. This was characterized by shortened stance phase, delayed timing of peak of pressure under the hallux and second toe, lateral load shift at the midfoot, and decreased toe pressures in operated limbs. Since loading abnormalities were correlated with the clinical outcome, modifications in treatment strategy that can improve foot loading may be desirable in these cases. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
References 1. Bajammal S, Tornetta P III, Sanders D, Bhandari M. Displaced intra-articular calcaneal fractures. J Orthop Trauma. 2005;19:360-364. 2. Barg A, Amendola A, Beaman DN, Saltzman CL. Ankle joint distraction arthroplasty: why and how? Foot Ankle Clin. 2013;18:459-470. 3. Besch L, Radke B, Mueller M, et al. Dynamic and functional gait analysis of severely displaced intra-articular calcaneus fractures treated with a hinged external fixator or internal stabilization. J Foot Ankle Surg. 2008;47:19-25. 4. Bozkurt M, Kentel BB, Yavuzer G, et al. Functional evaluation of intraarticular severely comminuted fractures of the calcaneus with gait analysis. J Foot Ankle Surg. 2004;43:374-379. 5. Bridgman S, Dunn K, McBride D, Richards P. Withdrawn: interventions for treating calcaneal fractures. Cochrane Database Syst Rev. 2008;4:CD001161.
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
718
Foot & Ankle International 35(7)
6. Bruce J, Sutherland A. Surgical versus conservative interventions for displaced intra-articular calcaneal fractures. Cochrane Database Syst Rev. 2013;1:CD008628. 7. Buckley R, Tough S, McCormack R, et al. Operative compared with nonoperative treatment of displaced intra-articular calcaneal fractures: a prospective, randomized, controlled multicenter trial. J Bone Joint Surg Am. 2002;84:1733-1744. 8. Davies MB, Betts RP, Scott IR. Optical plantar pressure analysis following internal fixation for displaced intra-articular os calcis fractures. Foot Ankle Int. 2003;24:851-856. 9. Hetsroni I, Ayalon M, Mann G, Meyer G, Nyska M. Walking and running plantar pressure analysis before and after resection of tarsal coalition. Foot Ankle Int. 2007;28:575-580. 10. Hirschmuller A, Konstantinidis L, Baur H, et al. Do changes in dynamic plantar pressure distribution, strength capacity and postural control after intra-articular calcaneal fracture correlate with clinical and radiological outcome? Injury. 2011;42:1135-1143. 11. Jiang N, Lin QR, Diao XC, Wu L, Yu B. Surgical versus nonsurgical treatment of displaced intra-articular calcaneal fracture: a meta-analysis of current evidence base. Int Orthop. 2012;36:1615-1622. 12. Kinner BJ, Best R, Falk K, Thon KP. Is there a reliable outcome measurement for displaced intra-articular calcaneal fractures? J Trauma. 2002;53:1094-1101, discussion 1102. 13. Kitoaka HB, Alexander IJ, Adelaar RS, et al. Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int. 1994;15:349-353. 14. Kline AJ, Anderson RB, Davis WH, Jones CP, Cohen BE. Minimally invasive technique versus an extensile lateral
approach for intra-articular calcaneal fractures. Foot Ankle Int. 2013;34:773-780. 15. Mittlmeier T, Machler G, Lob G, et al. Compartment syndrome of the foot after intraarticular calcaneal fracture. Clin Orthop Relat Res. 1991;269:241-248. 16. Mittlmeier T, Morlock MM, Hertlein H, et al. Analysis of morphology and gait function after intraarticular calcaneal fracture. J Orthop Trauma. 1993;7:303-310. 17. Orozco L, Munar A, Soler R, et al. Treatment of knee osteoarthritis with autologous mesenchymal stem cells: a pilot study. Transplantation. 2013;95:1535-1541. 18. Putti AB, Arnold GP, Cochrane L, Abboud RJ. The pedar inshoe system: repeatability and normal pressure values. Gait Posture. 2007;25:401-405. 19. Randle JA, Kreder HJ, Stephen D, et al. Should calcaneal fractures be treated surgically? A meta-analysis. Clin Orthop Relat Res. 2000;377:217-227. 20. Rosenbaum D, Lubke B, Bauer G, Claes L. Long-term effects of hindfoot fractures evaluated by means of plantar pressure analyses. Clin Biomech. 1995;10:345-351. 21. Sanders R. Displaced intra-articular fractures of the calcaneus. J Bone Joint Surg Am. 2000;82:225-250. 22. Sanders R, Fortin P, DiPasquale T, Walling A. Operative treatment in 120 displaced intraarticular clacaneal fractures. Results using a prognostic computed tomography scan classification. Clin Orthop Relat Res. 1993;290: 87-95. 23. Tsung BY, Zhang M, Mak AF, Wong MW. Effectiveness of insoles plantar pressure redistribution. J Rehabil Res Dev. 2004;41:767-774.
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
research-article2014
FAIXXX10.1177/1071100714531226Foot & Ankle InternationalHetsroni et al
Article
Plantar Pressure Anomalies After Open Reduction With Internal Fixation of High-Grade Calcaneal Fractures
Foot & Ankle International® 2014, Vol. 35(7) 712–718 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100714531226 fai.sagepub.com
Iftach Hetsroni, MD1,2, David Ben-Sira, PhD3, Meir Nyska, MD1,2, and Moshe Ayalon, PhD3
Abstract Background: Plantar pressure abnormalities after open reduction with internal fixation (ORIF) of intra-articular calcaneal fractures have been observed previously, but high-grade fractures were not selectively investigated and follow-up times were shorter than 2 years. The purpose of this study was to characterize plantar pressure anomalies in patients with exclusively high-grade calcaneal fractures after ORIF with a minimum 2 years of follow-up, and to test the association between plantar pressure distribution and the clinical outcome. Methods: The orthopaedic registry was reviewed to identify patients with isolated high-grade calcaneal fractures (Sanders types III-IV) who were operated on and had a minimum 2 years of follow-up. Sixteen patients were evaluated. Mean age was 47 years and follow-up was between 2 and 6 years. The Pedar-Mobile system was used to measure 3 loading and 3 temporal variables and compare these between the operated and the uninjured limbs. Results: Mean American Orthopaedic Foot and Ankle Society (AOFAS) score was 76 ± 7 at latest follow-up. Bohler’s angle was 5 ± 8 degrees before surgery and 25 ± 7 degrees at latest follow-up. Stance was shorter in operated limbs (P = .001). Timing of the peak of pressure was delayed in operated limbs under the hallux and the second toe (P ≤ .03). Peak pressure, force time integral, and pressure time integral were increased under the lateral midfoot (P ≤ .03) and decreased under the second metatarsal (P ≤ .03). Force time integral was decreased under the first metatarsal (P = .02) and under the hallux and the lateral toes (P ≤ .05). Increased loading under the lateral midfoot and decreased loading under the lateral toes were correlated with poorer clinical outcome (r = –.53, P < .05, and r = .63, P < .01, respectively). Conclusions: Side-to-side plantar pressure mismatch persisted at more than 2 years after ORIF of high-grade calcaneal fractures performed via lateral approach, despite improvement of Bohler’s angle. This was characterized by shortened stance phase, delayed timing of peak of pressure under the hallux and second toe, lateral load shift at the midfoot, and decreased toe pressures in operated limbs. Since loading abnormalities were correlated with the clinical outcome, modifications in treatment strategy that can improve foot loading may be desirable in these cases. Level of Evidence: Level III, case control. Keywords: calcaneal fracture, Bohler’s angle, plantar pressure, loading variables, temporal variables
Intra-articular displaced fractures of the calcaneus represent a complex injury, typically affecting the middle-aged active population.21 Although nonoperative management and open reduction with internal fixation (ORIF) are both considered acceptable treatment strategies with no definitive conclusion as to the superiority of either option,1,5-7,11,19 it is accepted that nonoperative treatment avoids the costs and risks of surgery (ie, operative wound healing problems, infection, sural nerve injuries), but ORIF can result in better calcaneal and heel morphology, leading to improved function16 and enabling more comfortable shoe wear.5-6,11 Yet, despite the potential benefits in function, abnormalities in gait pattern
after the operation are still noticeable, most commonly demonstrated through plantar pressure analyses.3,10,16
1
Department of Orthopedic Surgery, Meir General Hospital, Kfar Saba, Israel 2 Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel 3 Biomechanics Laboratory, Zinman College of Physical Education and Sport Sciences, Wingate Institute, Netanya, Israel Corresponding Author: Iftach Hetsroni, MD, Department of Orthopedic Surgery, Meir General Hospital, Tsharnichovski St 59, Kfar Saba 44281, Israel. Email: [email protected]
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
713
Hetsroni et al Interpreting available data about plantar pressure abnormalities in this respect is however subjected to several limitations. First, abnormalities in plantar pressures after surgery have not been adequately linked to specific types of intra-articular calcaneal fractures, but rather reported for mixed groups of fracture types by some investigators.10,20 Since increase in number of fracture elements is associated with poorer outcomes,7,16 possibly because of the lower energy of the primary trauma as well as the relative ease to obtain anatomic reduction with stable fixation during surgery in low-grade injuries (ie, Sanders type II) compared to high grade injuries (ie, Sanders types III and IV),22 outcome after ORIF of calcaneal fractures should ideally be investigated in specific subgroups of fracture type. Moreover, some investigators had inadequate follow-up times, which were shorter than 2 years,8,10,12,16 or did not confirm normalization of Bohler’s angle in all patients to ascertain that the immediate postoperative goal of the surgery was indeed achieved.10 Viewing Sanders types III and IV as a particular subset of complex intra-articular calcaneal fractures (as opposed to Sanders type II) where ongoing debate about the outcome and advisability of surgery compared to nonoperative management still exists,7 a selective analysis of outcome is desirable in these high-grade fractures specifically, with adequate follow-up and acceptable radiographic alignment after ORIF. Because abnormal plantar pressure distribution has been associated with foot pathologies and because normalization of plantar pressures after corrective surgery or application of orthotic devices has been associated with improvement in function,9,23 information about plantar pressure distribution can be used for seeking operative modifications and for planning orthotics, which may improve outcomes in these challenging cases. The purpose of this study was therefore to characterize plantar pressure distribution in patients with exclusively high-grade calcaneal fractures after ORIF with a minimum 2 years of follow-up and to test the association between plantar pressure abnormalities and the clinical outcome. We hypothesized that abnormalities in plantar pressure distribution with predominance of loading lateralization would be demonstrated within the hindfoot and midfoot in operated limbs despite the length of follow-up and normalization of Bohler’s angle, and that greater abnormalities in loading pattern would be associated with inferior clinical outcomes.
Methods Study Population The orthopaedic trauma registry relating to ORIF of closed calcaneal fractures performed in our institution between 2000 and 2004 was reviewed. Inclusion criteria were age 18 years or older, unilaterality of the injury, fracture classified
preoperatively with computed tomography scan as Sanders type III or type IV, and minimum 2 years of follow-up. After excluding 11 patients due to other accompanying lower limb and spine fractures that underwent internal fixation (5 patients), severe knee arthrosis (1 patient), postoperative osteomyelitis of the calcaneus (1 patient), and subtalar arthrodesis performed before this study due to subtalar arthrosis and pain (4 patients), 16 patients were contacted and volunteered to participate in this study. There were 13 men and 3 women. Mean age at the time of the injury was 47 years ± 11 years (range, 18-67 years). Five fractures were classified as Sanders type III and 11 fractures as Sanders type IV. Mean follow-up time was 3 years and 4 months (range, 2-6 years). The operation was performed in all cases within 2 weeks after the injury through a lateral approach by a team of 2 orthopaedic surgeons, specialized in foot and ankle surgery. Quality of reduction was assessed intraoperatively with lateral, axial, and Broden views. Postoperative management included non-weight-bearing for 8-10 weeks, followed by weight-bearing and range of motion and strengthening exercises. Patients were routinely observed in our outpatient clinic for 6 to 12 months after the operation. The latest follow-up visit was performed specifically for the purpose of this study, during which patients completed ankle-hindfoot score according to the American Orthopaedic Foot and Ankle Society (AOFAS)13 and had lateral radiograph to measure Bohler’s angle. The study was approved by our institutional review board, and all participants signed informed consent.
Plantar Pressure Analysis Protocol Plantar pressures were collected during walking on a level floor at a natural preferred cadence using Pedar-Mobile system insoles (Novel Gmbh, Munich, Germany) at a rate of 50Hz. The insoles were inserted under both operated and contralateral uninjured foot, using one shoe design for all participants. Ten zones of the plantar area of the foot were defined by specific subsets of 99 insole sensors. The zone definitions were designed to approximate relevant plantar structures as follows (Figure 1): medial heel, lateral heel, medial midfoot, lateral midfoot, first metatarsal, second metatarsal, lateral metatarsals, first toe (hallux), second toe, and lateral toes. Plantar pressure data were collected over 4 gait cycles for each participant and recorded at the same gait cycle for both the operated and the contralateral uninjured foot. For each step, 3 loading variables and 3 temporal variables were calculated for each foot zone, and the mean of 4 steps for each variable was used to compare the operated limb and the uninjured limb. The 3 loading variables included (1) peak pressure (PkP) [N*cm-2], representing the maximum pressure recorded under the specific foot zone, (2) force time integral (FTI) [N*sec], representing the
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
714
Foot & Ankle International 35(7)
Plantar Pressure Analysis
Figure 1. Plantar zone definitions.
overall amount of force that has been applied over time, and (3) pressure time integral (PTI) [N*cm-2*sec], representing the overall amount of pressure that had been applied over time. The 3 temporal variables included (1) contact time (CT) [msec], representing the duration of contact at the specific foot zone, (2) relative contact time (Relative CT) [%ST], representing the duration of contact at the specific foot zone as percentage of the entire stance time, and (3) timing of peak of pressure (PkPt) [%ST], representing the timing of maximal pressure at the specific foot zone as a percentage of the stance time.
Statistical Analysis Paired Student t tests were used to compare plantar pressure variables between the operated and the uninjured limbs. Pearson’s product–moment correlation coefficient was used to test the relationships between the AOFAS clinical outcome scores and the plantar pressure values. Level of significance was set at .05.
Mean natural gait velocity of the participants was 0.99 ± 0.15 m/s. Descriptive statistics of the loading variables with comparisons between operated and uninjured limbs are presented in Table 1. total force time integral (FTI) and total pressure time integral (PTI) were reduced in the operated limb compared to the uninjured limb by 7% (P = .02 for comparing FTI, and P = .06 for comparing PTI). Viewing specific foot subzones, symmetrical loading was recorded at the heel in the operated limb compared to the uninjured limb, but increased loading was recorded in all 3 variables at the lateral midfoot of the operated limb compared to the uninjured limb (P ≤ .03). Observing the forefoot, reduced loading values were recorded in the operated limb compared to the uninjured limb under the second metatarsal in all variables (P ≤ .03), and under the first metatarsal in FTI (P = .02). Reduced loading values were also recorded in the operated limb under the hallux, second toe, and lateral toes, with regard to FTI and PTI. Assessment of the relationships between AOFAS clinical outcome scores and loading variables that displayed significant differences between the operated and the uninjured limbs showed the following statistically significant findings: At the lateral midfoot, there was a statistically significant and moderate correlation between the AOFAS and FTI (r = –.52, P < .05) and between the AOFAS and PTI (r = –.53, P < .05) (Figure 2). At the lateral toes, there was a statistically significant and moderate correlation between the AOFAS and PkP (r = .61, P < .05), FTI (r = .66, P < .01), and PTI (r = .63, P < .01) (Figure 3). Descriptive statistics of the temporal variables and comparisons between the operated and uninjured limbs are presented in Table 2. Stance duration was reduced in the operated limb compared to the uninjured limb by 4.7% (705 msec vs 740 msec, P = .001). Viewing specific foot subzones, contact time was shorter under the medial heel (P = .05), and under all forefoot zones, including all metatarsal and all toe zones in the operated limb compared to the uninjured limb (P ≤ .04). Timing of peak of pressure (PkPt) was delayed in the operated limb compared to the uninjured limb under the hallux and under the second toe (P ≤ .03). There were no significant bilateral differences in total contact area or in either of the 10 foot subzones contact areas.
Results
Discussion
Mean preoperative Bohler’s angle was 5 ± 8 degrees (range, 0-20 degrees), and mean postoperative Bohler’s angle at latest follow-up was 25 ± 7 degrees (range, 15-35 degrees). Mean latest follow-up AOFAS ankle hindfoot score was 76 ± 7. Specifically, 4 patients had a poor result (AOFAS ≤ 69), 6 had a fair result (AOFAS 70-79), and 6 had a good result (AOFAS 80-89). There were no cases with an excellent result (AOFAS 90-100).
The current literature reveals that functional outcome in patients with intra-articular calcaneal fractures that undergo ORIF is suboptimal, and that deviations in gait pattern is a consistent finding after surgery.3,8,10,12,16,20 However, inadequate follow-up and nonspecific series of patients with regard to the severity of the injury suggest that additional data are desirable to characterize abnormalities in gait and function in these cases with greater injury specificity and
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
715
Hetsroni et al Table 1. Comparisons of Loading Variables Between Operated and Uninjured Limbs. Foot zone
Side
PkP [N*cm-2]
P
FTI [N*sec]
P
Lateral heel
Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured
198.3 ± 66.4 212 ± 47.8 187 ± 58.1 197.8 ± 29.5 140.6 ± 34.4 121.4 ± 19.9 107 ± 28.1 94.5 ± 37.3 208 ± 58.2 208 ± 54.8 207.8 ± 46.6 239.2 ± 50.2 179.8 ± 73.1 195.9 ± 47.1 68.4 ± 37.6 93.3 ± 40.2 103.3 ± 38.3 117.5 ± 49.1 195.8 ± 65.4 237.8 ± 85.8 279.5 ± 51 290.8 ± 52.1
.2
40.8 ± 15.5 43.3 ± 17.4 29 ± 10.1 30.7 ± 12.8 48 ± 20.9 39.5 ± 19.6 8.6 ± 5.3 8.5 ± 5.6 57.2 ± 18.4 57.5 ± 21.9 24.1 ± 7.6 31.4 ± 10.8 18 ± 9.5 23.3 ± 9.1 1.8 ± 1.6 3.1 ± 1.8 4.6 ± 3.1 6.2 ± 3.5 7.9 ± 4.2 11.5 ± 5.3 465.4 ± 119.8 501.6 ± 156.6
.46
Medial heel Lateral midfoot Medial midfoot Lateral MTT Second MTT First MTT Lateral toes Second toe Hallux Total foot
.37 .03 .12 1.0 .03 .4 .05 .22 .07 .49
.59 .03 .95 .96 .001 .02 .05 .07 .03 .02
PTI [N*cm-2*sec] 66 ± 36.3 71.3 ± 30.4 58.5 ± 16.7 63 ± 21 57.7 ± 18.5 49.1 ± 15 37.5 ± 13.8 33.1 ± 13.4 68.8 ± 21.1 72.6 ± 25.2 57.1 ± 17.6 74.6 ± 26.5 49.9 ± 31.6 61 ± 23.7 17.2 ± 10.7 28.3 ± 14.1 24.3 ± 13.7 34.1 ± 21.5 36.8 ± 18 57 ± 22.8 120.1 ± 27.1 129.3 ± 28.9
P .43 .37 .02 .09 .6 .001 .11 .02 .02 .01 .06
FTI, force time integral; MTT, metatarsal; PkP, peak pressure; PTI, pressure time integral. Values are presented as mean ± standard deviation. Statistically significant differences are presented in bold.
Figure 2. The correlation between AOFAS score and pressure time integral (PTI) at the lateral midfoot in operated limbs.
Figure 3. The correlation between AOFAS score and pressure time integral (PTI) at the lateral toes in operated limbs.
longer rehabilitation. The current study was designed to characterize plantar pressure anomalies specifically in patients with high-grade fractures of the calcaneus after ORIF that obtained normalization of Bohler’s angle and had a minimum 2 years of follow-up, and to test the relationships between pressure anomalies and clinical outcomes. Viewing the foot as a whole, reductions observed in total FTI, total PTI, and stance time under the operated limb
compared to the uninjured limb indicate that asymmetric gait persists in these cases despite adequate postoperative duration. This is consistent with other investigators that observed reduced contact time,10 reduced total impulse,20 and shorter pressure duration under the operated foot.3 Such gait asymmetry may be the mere consequence of persistent pain in these operated limbs, most likely resulting from abnormalities and degenerative changes in the subtalar joint.
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
716
Foot & Ankle International 35(7)
Table 2. Comparisons of Temporal Variables Between Operated and Uninjured Limbs. Foot zone
Side
CT [msec]
P
Lateral heel
Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured Operated Uninjured
596 ± 172 628 ± 162 562 ± 151 607 ± 173 644 ± 120 667 ± 115 524 ± 124 522 ± 160 677 ± 127 714 ± 132 631 ± 110 678 ± 158 586 ± 172 629 ± 166 382 ± 166 524 ± 158 442 ± 146 541 ± 165 469 ± 188 619 ± 147 705 ± 118 740 ± 128
.34
Medial heel Lateral midfoot Medial midfoot Lateral MTT Second MTT First MTT Lateral toes Second toe Hallux Total foot
.05 .23 .92 .01 .03 .04 .03 .01 .02
Relative CT [% ST] 83.6 ± 14.9 84.3 ± 13.2 79.2 ± 14 81 ± 15 91.3 ± 6.8 90.5 ± 8.1 73.9 ± 9.2 70 ± 20.3 95.8 ± 4 96.5 ± 3.9 89.9 ± 7.6 91 ± 7.9 82.3 ± 15.4 83.9 ± 10.6 55.4 ± 23.8 70.8 ± 17.2 63.8 ± 18.9 73.1 ± 17.3 65.8 ± 21.3 84.4 ± 15.8
P
PkPt [% ST]
P
.87
15 ± 4.1 15.5 ± 12.5 15.5 ± 4.7 12.8 ± 4.8 27.3 ± 14.5 26 ± 15.6 20.3 ± 4 22.6 ± 9 75.5 ± 4.6 73.9 ± 6.5 78.5 ± 4.2 76.3 ± 3.4 79.4 ± 4 75.6 ± 4.8 74 ± 14.2 74.9 ± 15.9 81.3 ± 3.8 78.8 ± 4.4 86.7 ± 3.5 82.5 ± 4.7 60.7 ± 27.2 70.4 ± 17.9
.89
.68 .76 .48 .64 .65 .6 .04 .04 .03
.001
.15 .61 .39 .18 .06 .02 .86 .004 .001 .15
CT, contact time; relative CT, contact time at a foot subzone as a percentage of the stance time; MTT, metatarsal; PkPt, = peak pressure timing; % ST, percentage of stance time. Values are presented as mean ± standard deviation. Statistically significant differences are presented in bold.
Previous studies showed that pain in operated limbs after ORIF indeed persisted and was comparable to the level of pain demonstrated in nonoperated limbs with similar fractures.6,11 These patients are therefore likely to spend less time on the affected limb to reduce loading that provokes pain. The fact that Bohler’s angle at latest follow-up indicated acceptable reduction of calcaneal morphology in our series (ie, range 15-35 degrees) suggest that re-creating calcaneal morphology during the operation with stable fixation cannot by itself be considered a sufficient measure to achieve symmetric gait pattern after surgery. The damaged cartilaginous articular surfaces at the subtalar joint, however, cannot be appreciated by using standard intra- or postoperative fluoroscopy or plain radiography. This intra-articular cartilage wear may provide 1 explanation for pain persistence in cases where bone alignment is satisfactory. In this regard, therapeutic alternatives that have been suggested to address cartilage wear, stimulate chondrogenesis, and improve function in affected joints including the use of mesenchymal stem cells17 and temporary distraction fixators.2 These alternatives, however, have not been implemented for cartilage wear at the subtalar joint, but may be tested in the future for this purpose. Viewing foot subzones, the most consistent finding in terms of foot loading was a lateral load shift observed at the midfoot in operated limbs compared to uninjured limbs.
Several mechanisms could have contributed to this finding. The most likely mechanism is a reduction in subtalar range of motion.10,12 Reduced normal subtalar eversion during the stance leads to relatively inverted foot positioning during foot loading, and this can result in a lateral load shift. Lateral load shift may also be related to dysfunction of the gastrocnemius, recorded in this type of injury.4 In case such weakness is secondary to Achilles shortening, it may accentuate as well an inverted foot position. A third associated factor to lateral load shift may be dysfunction of the peroneal tendons. Although we did not test the peroneal tendons strength and function specifically, we believe it is possible because the operative approach used for performing ORIF in all our cases involved a lateral L-shaped incision during which the peroneal tendons were temporarily translocated anteriorly away from their normal location around the peroneal tubercle to expose and then reduce and fix the fragments; it is possible that some dysfunction of these foot evertors persisted after surgery. The sinus tarsi approach, which avoids exposure of the peroneal tendons, is an operative alternative to the lateral approach used in our series, although it is unknown whether it can reduce the lateral load shift that is seen after surgery performed via lateral approach.14 Compartment syndrome of the foot has been recorded in a significant proportion of patients with intra-articular calcaneal fractures, leading to plantar muscle ischemia and
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
717
Hetsroni et al plantar muscle scarring.15 This scarring may result in reduced flexibility of the medial plantar arch, again contributing to a more inverted foot position. Greater surgeons’ awareness of this potential sequela, with intraoperative intracompartmental foot pressure assessment and then compartment release in case of increased pressure, may provide a solution to this problem. And last, the lateral L-shaped approach used in all our cases enabled us to assess with precision the reduction of the fragments at the lateral aspect of the calcaneus as well as subtalar joint congruency, but it may have underestimated medial calcaneal wall integrity in some cases. Thus, theoretically, an underreduced medial column may have been resulted to some extent, contributing to an inverted foot position as well. In regard to forefoot plantar pressure findings, the reductions observed in all loading variables under the second metatarsal and in FTI under the first metatarsal, again support a lateral load shift pattern and are consistent in this respect with findings of previous investigators.20 Reduced loading was observed under all toes in our series as well, and this again may represent toe clawing, resulting from compartment syndrome of the foot, plantar muscle ischemia and eventual scarring.15 As to heel loading pattern, we did not identify side-toside mismatch between the operated and the uninjured limb in any of the loading variables, which is in contrast to other investigators that showed decreased pressure at the medial heel.3,20 This may have been related to the fact Bohler’s angle was re-created in our series and patients were examined at a relatively longer follow-up, potentially enabling longer recovery and adaptations to occur, compared to previous investigations. Plantar pressure distribution was previously measured on normal healthy individuals of similar age-range as ours.18 No significant side-to-side differences in any of the foot loading variables were demonstrated.18 The asymmetric foot loading observed between the operated and the uninjured limb in our operated population should be therefore viewed as representing an abnormal loading pattern. Furthermore, assessment of the relationships between the clinical outcome (ie, AOFAS score) and the abnormal foot loadings showed in this study significant correlations. More pronounced pressure lateralization at the midfoot and reduction in toe pressures were correlated with poorer clinical outcomes (ie, lower AOFAS scores). Although this correlation does not necessarily imply causative relationships, it may strengthen the notion that modifications in treatment which contribute to improvements in loading pattern may as well have positive effects on the clinical outcome. In terms of fabricating orthotic support, which is one alternative to address this issue, reducing the elevated pressures at the lateral midfoot during midfoot support and redistributing pressures at the forefoot to improve the effectiveness of the push-off phase, possibly by introducing hard materials
at the transition zones between the midfoot and forefoot, may be desirable in this population. This requires differential use of materials with specific mechanical properties at the distinct foot areas with respect to both latitudinal and longitudinal distribution. Limitations of this study include the fact only fluoroscopy and plain radiographs were used to assess calcaneal morphology during the operation and at follow-up. While these are acceptable imaging modalities to assess fracture reduction during and after surgery in these cases, CT scans, and potentially 3D CT reconstructions, could have delineated with greater accuracy calcaneal morphology and subtalar joint congruency and indicated whether these cases should have been indeed referred to as postoperatively adequately aligned cases of calcaneal fractures. As CT exposes patients to additional irradiation, we felt it was not justified to perform follow-up CT imaging for the purpose of this study. In conclusion, side-to-side plantar pressure mismatch persisted at more than 2 years after ORIF of high-grade calcaneal fractures performed via lateral approach, despite recreation of Bohler’s angle. This was characterized by shortened stance phase, delayed timing of peak of pressure under the hallux and second toe, lateral load shift at the midfoot, and decreased toe pressures in operated limbs. Since loading abnormalities were correlated with the clinical outcome, modifications in treatment strategy that can improve foot loading may be desirable in these cases. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
References 1. Bajammal S, Tornetta P III, Sanders D, Bhandari M. Displaced intra-articular calcaneal fractures. J Orthop Trauma. 2005;19:360-364. 2. Barg A, Amendola A, Beaman DN, Saltzman CL. Ankle joint distraction arthroplasty: why and how? Foot Ankle Clin. 2013;18:459-470. 3. Besch L, Radke B, Mueller M, et al. Dynamic and functional gait analysis of severely displaced intra-articular calcaneus fractures treated with a hinged external fixator or internal stabilization. J Foot Ankle Surg. 2008;47:19-25. 4. Bozkurt M, Kentel BB, Yavuzer G, et al. Functional evaluation of intraarticular severely comminuted fractures of the calcaneus with gait analysis. J Foot Ankle Surg. 2004;43:374-379. 5. Bridgman S, Dunn K, McBride D, Richards P. Withdrawn: interventions for treating calcaneal fractures. Cochrane Database Syst Rev. 2008;4:CD001161.
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
718
Foot & Ankle International 35(7)
6. Bruce J, Sutherland A. Surgical versus conservative interventions for displaced intra-articular calcaneal fractures. Cochrane Database Syst Rev. 2013;1:CD008628. 7. Buckley R, Tough S, McCormack R, et al. Operative compared with nonoperative treatment of displaced intra-articular calcaneal fractures: a prospective, randomized, controlled multicenter trial. J Bone Joint Surg Am. 2002;84:1733-1744. 8. Davies MB, Betts RP, Scott IR. Optical plantar pressure analysis following internal fixation for displaced intra-articular os calcis fractures. Foot Ankle Int. 2003;24:851-856. 9. Hetsroni I, Ayalon M, Mann G, Meyer G, Nyska M. Walking and running plantar pressure analysis before and after resection of tarsal coalition. Foot Ankle Int. 2007;28:575-580. 10. Hirschmuller A, Konstantinidis L, Baur H, et al. Do changes in dynamic plantar pressure distribution, strength capacity and postural control after intra-articular calcaneal fracture correlate with clinical and radiological outcome? Injury. 2011;42:1135-1143. 11. Jiang N, Lin QR, Diao XC, Wu L, Yu B. Surgical versus nonsurgical treatment of displaced intra-articular calcaneal fracture: a meta-analysis of current evidence base. Int Orthop. 2012;36:1615-1622. 12. Kinner BJ, Best R, Falk K, Thon KP. Is there a reliable outcome measurement for displaced intra-articular calcaneal fractures? J Trauma. 2002;53:1094-1101, discussion 1102. 13. Kitoaka HB, Alexander IJ, Adelaar RS, et al. Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int. 1994;15:349-353. 14. Kline AJ, Anderson RB, Davis WH, Jones CP, Cohen BE. Minimally invasive technique versus an extensile lateral
approach for intra-articular calcaneal fractures. Foot Ankle Int. 2013;34:773-780. 15. Mittlmeier T, Machler G, Lob G, et al. Compartment syndrome of the foot after intraarticular calcaneal fracture. Clin Orthop Relat Res. 1991;269:241-248. 16. Mittlmeier T, Morlock MM, Hertlein H, et al. Analysis of morphology and gait function after intraarticular calcaneal fracture. J Orthop Trauma. 1993;7:303-310. 17. Orozco L, Munar A, Soler R, et al. Treatment of knee osteoarthritis with autologous mesenchymal stem cells: a pilot study. Transplantation. 2013;95:1535-1541. 18. Putti AB, Arnold GP, Cochrane L, Abboud RJ. The pedar inshoe system: repeatability and normal pressure values. Gait Posture. 2007;25:401-405. 19. Randle JA, Kreder HJ, Stephen D, et al. Should calcaneal fractures be treated surgically? A meta-analysis. Clin Orthop Relat Res. 2000;377:217-227. 20. Rosenbaum D, Lubke B, Bauer G, Claes L. Long-term effects of hindfoot fractures evaluated by means of plantar pressure analyses. Clin Biomech. 1995;10:345-351. 21. Sanders R. Displaced intra-articular fractures of the calcaneus. J Bone Joint Surg Am. 2000;82:225-250. 22. Sanders R, Fortin P, DiPasquale T, Walling A. Operative treatment in 120 displaced intraarticular clacaneal fractures. Results using a prognostic computed tomography scan classification. Clin Orthop Relat Res. 1993;290: 87-95. 23. Tsung BY, Zhang M, Mak AF, Wong MW. Effectiveness of insoles plantar pressure redistribution. J Rehabil Res Dev. 2004;41:767-774.
Downloaded from fai.sagepub.com at University of British Columbia Library on June 18, 2015
