547082
research-article2014
FAIXXX10.1177/1071100714547082Foot & Ankle InternationalRoth et al
Article
Intraosseous Fixation Compared to Plantar Plate Fixation for First Metatarsocuneiform Arthrodesis: A Cadaveric Biomechanical Analysis
Foot & Ankle International® 2014, Vol. 35(11) 1209–1216 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100714547082 fai.sagepub.com
Klaus Edgar Roth, MD1, Jennifer Peters1, Irene Schmidtmann, PhD2, Uwe Maus, MD1, Daniel Stephan, PhD3,4, and Peter Augat, PhD3,4
Abstract Background: Metatarsocuneiform (MTC) fusion is a treatment option for management of hallux valgus. We compared the biomechanical characteristics of an internal fixation device with plantar plate fixation. Methods: Seven matched pairs of feet from human cadavers were used to compare the intramedullary (IM) device plus compression screw to plantar plate combined with a compression screw. Specimen constructs were loaded in a cyclic 4-point bending test. We obtained initial/final stiffness, maximum load, and number of cycles to failure. Bone mineral density was measured with peripheral quantitative computed tomography. Performance was compared using time to event analysis with number of cycles as time variable, and a proportional hazard model including shared frailty model fitted with treatment and bone mineral density as covariates. Results: On average the plates failed after 7517 cycles and a maximum load of 167 N, while the IM-implants failed on average after 2946 cycles and a maximum load of 69 N. In all pairs the 1 treated with IM-implant failed earlier than the 1 treated with a plate (hazard ratio for IM-implant versus plate was 79.9 (95% confidence interval [6.1, 1052.2], P = .0009). The initial stiffness was 131 N/mm for the plantar plate and 43.3 N/mm for the IM implant. Initial stiffness (r = .955) and final stiffness (r = .952) were strongly related to the number of cycles to failure. Bone mineral density had no effect on the number of cycles to failure. Conclusion: Plantar plate fixation created a stronger and stiffer construct than IM fixation. Clinical Relevance: A stiffer construct can reduce the risk of nonunion and shorten the period of non-weight-bearing. Keywords: Lapidus fusion, intramedullary implant, plantar angle stable plate, biomechanic testing Arthrodesis of the first metatarsocuneiform joint (MTC-1) is a procedure used widely for the correction of an enlarged intermetatarsal angle. Modifications of the operative technique originally described by Lapidus19 have, in general, good treatment outcomes,2,5,20,22,32 with an incidence of revision due to nonunion and a revision of malpositioning of between 1.8%30 und 13%.32 Data from the current literature indicate that the risk of revision has fallen in recent years.4,6,18,26,28,30,31,35 This can be attributed to increased understanding of the biomechanics of the MTC-1 joint as well as improvements of the hardware, such as the use of locking plates. At present, probably the most frequently used fixation technique for Lapidus fusion is crossed-screw fixation.1,13,14,26 A further biomechanical advance would seem to be the placement of osteosynthetic materials plantar to the MTC-1 joint. Marks et al21 demonstrated that this technique had a more positive influence on interfragmentary stability compared to the widely used crossed screw
osteosynthesis, while Klos et al16 have shown the advantages of a plantar positioned locking plate over an angular stable locking plate positioned dorsomedially. From a biomechanical viewpoint a plantar positioned plate is better able to cope with the bending stresses to which the construct is subjected.15 More recent prospective studies indicate that the better biomechanical results of the plantar 1
Center of Orthopedic and Trauma Surgery, University Medical Center of the Johannes Gutenberg University, Mainz, Germany 2 Institute for Medical Biometry, Epidemiology and Computer Science, Johannes Gutenberg University, Mainz, Germany 3 Institute for Biomechanics, Traumacenter, Murnau, Germany 4 Paracelsus Medical University, Salzburg, Austria Corresponding Author: Klaus Edgar Roth, MD, Center of Orthopedic and Trauma Surgery, University Medical Center of the Johannes Gutenberg University Mainz, Langenbeckstraße 1, Mainz 55131, Germany. Email: [email protected]
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1210
Foot & Ankle International 35(11)
Figure 1. IO FIX.
Figure 2. Plantar plate.
plates are reflected in clinical practice by lower complication rates.12,17 On the other hand, the plantar osteosynthesis has been associated with more access morbidity with regard to the tibialis anterior tendon, which represents a hardwarerelated impingement problem. In contrast to plate osteosynthesis an intramedullary implant has the advantage of maintaining periosteal circulation25 and a reduced risk of hardware associated impingement.9 Periosteal perfusion is, in turn, a relevant factor in the healing process. The aim of the present study was to compare the stability of an intramedullary screw and poststabilization system with those of a locked plantar plate taking into account biomechanical characteristics and bone mineral density (BMD). We hypothesized that plantar plate fixation would create a stronger and stiffer construct than intramedullary fixation.
Materials and Methods Seven matched pairs of freshly frozen lower extremity cadaver specimens were used. The mean donor age at death was 76 (range, 69 to 89) years. Two of the donors were males and 5 were females. The specimens had no evidence of prior surgery or musculoskeletal disease. For each pair of specimens, 1 was treated with the intramedullary implant IO FIX (Extremity Medical™, Parsippany, NJ, USA) and 1 with plantar locking plate (Wright Medical Technology, Inc, Arlington, TX, USA); treatment was allocated randomly. One foot of each pair was randomly assigned to have screw and post implant (Figure 1) with adjunct 4-mm-diameter compression screw (Durchbohrte Schraube, Königsee Implantate GmbH, Allendorf, Germany). For the other foot of the pair, fixation of the first MTC-joint was carried out by plantar application of a Lapidus arthrodesis plate (Figure 2) with 2 gliding holes and a 4 mm compression screw. The use of cadaveric
specimens was approved by the local ethics commission. BMD was determined in the distal tibia with use of peripheral quantitative computed tomography (pQCT-XCT 2000, Stratec, Germany). A dorsomedial incision was made along the axis of the first MTC joint. The skin, subcutaneous fat, and fascia were elevated along the capsule exposing the first cuneiform and the first metatarsal. The first MTC capsule was transected to expose the articular surface. Next, a 1-mm section was taken from each opposing joint surface using an oscillating bone saw. Both parts of the study began with the insertion of a 4-mm cannulated, self-cutting, and selftapping compression screw with a partial (distal) thread centrally into the medial cortex of the first metatarsal, at a distance of 10 to 15 mm from the joint, and exiting in the middle of the intermediate cuneiform bone. Compression was applied by finger tightening of the screw, using a 3-finger (thumb and first 2 fingers) technique. The fixation of the arthrodesis was then completed with either screw and post or with plantar plate fixation. In both groups the additional compression screw was placed in the intermediate cuneiform bone. This procedure has been recommended to provide a greater load to failure and bending moment.29,32
IO FIX The screw and post device consisted of a lag screw in combination with an anchored post. The post is designed to distribute the compression forces uniformly across the arthrodesis site through a lag effect of the screw. In this way a greater surface pressure should be achieved on the osteomie (joine line) side than is the case for a lag screw. The post was placed in the metatarsal base whereas the lag screw crossed the joint and took place in the medial cuneiform. We used the fixed configuration of the implant, which locked the implant together at a fixed angle (Figure 3).
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1211
Roth et al
the construct stiffness was calculated between 10 and 40 N. Sinusiodal axial loading between lower and upper load level was performed at 1 Hz. Failure was defined as a displacement at the plantar aspect of the osteotomy of 3 mm or more and was measured by an extensometer (Instron Ltd, High Wycombe, UK).11 Cycles until failure, failure load, displacement, and plantar gapping were recorded. Stiffness of the construct was determined from the linear portion of load versus displacement curves. Construct stiffness in the last loading step before failure was determined. Construct stiffness was corrected with respect to the system stiffness. The performance of screw and post implant and plates was compared using time to event analysis with number of cycles and maximum load to failure as time variables. As pairs of specimens were considered, a shared frailty model was fitted to the time to event data. Usually in a Cox proportional hazard model the hazard for the i-th observation unit is given by λ i ( t , xi1 ,K xip ) ) = λ 0 ( t ) exp ( β1 xi1 + L + β p xi p )
Figure 3. IO FIX, X-ray.
Figure 4. Plantar plate, X-ray.
Plantar Plate At the plantar site, a plantar precontoured plate was used. Fixed-angle locked screws were inserted into the outer plate holes, while conventional nonlocked screws were used in the oval inner holes. Following the manufacturer’s recommended technique, one 3.5-mm screw was inserted into the distal inner screw hole crossing the osteotomy, to obtain additional interfragmentary compression (Figure 4). At the end of the implantation (screw and post/plantar plate) the bone segment containing the first metatarsal and first and second cuneiform was transected, eliminating the soft tissue completely. After instrumentation, the specimens were embedded in polyurethane (RENCAST©; Huntsman Advanced Materials GmbH, Switzerland). The implant and screws were coated with a modeling component to prevent contact with the embedding material. The potting was carried out using the technique of Gruber et al11 and Cohen et al.7 A 40 mm length of specimen was left exposed. Cyclic fatigue testing was performed on a servo- electric load frame (Instron E3000, Instron Ltd, High Wycombe, UK) under increasing axial load. The samples were loaded in a 4-point bending device. The upper support span was 60 mm and the lower support span was 160 mm.15 The upper load level started at 20 N and was increased by 10 N every 500 cycles. The lower load level remained at 10 N. Initially
with baseline hazard λ 0 ( t ) , covariates ( xi1 , xip ) ) for observation unit i, and regression coefficients ( β1 , , β p ) . If several observation units share certain properties, for example, specimens from 1 donor, this can be taken into account by adding a multiplicative frailty term to the model. Then λ ij ( t , xi j1 ,K xijp ) = Z i λ 0 ( t ) exp ( β1 xi j1 + L + β p xijp ) describes the hazard for bone j in donor i. Z i describes a random effect of the i-th donor, ( xi j1 , xijp ) are the covariates for bone j in donor I, and as before ( β1 , , β p ) are the regression coefficients. Here Z i is assumed to follow a Gamma distribution with expectation. Descriptive analysis included Kaplan–Meier estimates for the time to event data. Furthermore, means and standard deviations were computed for quantitative variables. Association between pairs of variables were displayed in scatter plots and described by computing the Pearson correlation coefficient. Differences between pairs of values were evaluated using the paired t test.
Results On average the plates failed after 7517 cycles and a maximum load of 167.1 N while the screw and post implants failed on average after 2946 cycles and a maximum load of 68.6 N. After 8167 cycles 50% of the plates had failed while the same failure rate was observed after 2269 cycles in the screw and post group (Figure 5). In all pairs of specimens the one treated with the screw and post implants failed earlier than the one treated with a plate. The estimated hazard ratio for the screw and post implant versus plate was 79.9 (95% confidence interval [6.1, 1052.2], P = .0009). The load was increased every 500 cycles, therefore the number of cycles to failure and the maximum load to failure essentially carry the same information. This is reflected in a
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1212
Foot & Ankle International 35(11)
Figure 5. Kaplan–Meier estimates of survival probability (probability that component has not yet failed) by treatment.
high correlation of these 2 variables (r = .998), therefore only results for cycles to failure are shown in detail. Initial stiffness (r = .955) and final stiffness (r = .952) were strongly related to number of cycles to failure (Figure 6). Thus these measures were all higher on average in the plate-treated group than in the screw and post group. Generally, there was a strong correlation between the parameters in the plate-treated and screw and post specimen of the same subject (Figure 6). Pearson correlation coefficients observed were for cycles to failure r = .94, for bone mineral density r = .98, initial stiffness r = .91, and final stiffness r = .91. In spite of the random allocation of treatment it turned out that in all but 1 pair of specimens, the 1 treated with a plate had a higher BMD than the 1 treated with screw and post implant. However, the differences were fairly small in comparison to the variability of bone densities between subjects: on average the BMD was 17.3 mg/ccm higher in the plate-treated specimens (SD of difference = 16.3 mg/ ccm), whereas BMD ranged from 137.9 mg/ccm to 428.9 (mean = 255.3, SD = 76.8). Because of this observation, we investigated the association of number of cycles to failure and BMD (Figure 7). We found that the Pearson correlation coefficient was only .010 (P = .7344). Consequently, when fitting a proportional hazard model with BMD as additional covariate, there was no effect of BMD on the number of cycles to failure, the hazard ratio being 0.997 per mg/ccm (95% confidence interval [0.963, 1.033], P = .8796). In total the
plates resisted 2.5 times the number of loading cycles before failure than did the screw and post construct. Descriptive statistics for all measured parameters can be found in Table 1.
Discussion Clinical outcomes with first MTC-joint (Lapidus) arthrodesis have been very good overall, but there is a 5% to 15% incidence of nonunion and malunion, depending on the study.3,7,26 A stable osteosynthesis is regarded as a necessary condition for rapid bone healing, reduction of the risk of nonunion and shortening of the period of non-weight-bearing.24 Currently, probably the most frequent fixation technique for Lapidus fusion is crossed-screw fixation.1,14,26 However plating in different positions is becoming increasingly popular,10 not least because of the availability of locked plating. Up until now the results of previous biomechanical studies of the stability of osteosyntheses at the MTC-1 have concerned the comparison of plates from different manufacturers and crossed-screw fixation. As, however, the plate configuration of plates intended for dorsomedial placement of different manufacturers show marked conceptual variations, statements concerning the biomechanical properties of a particular plate are only of limited relevance for other plates. The biomechanical principle of osteosyntheses on the plantar side which is under tension is markedly different compared to dorsomedial placement and suggests a beneficial
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1213
Roth et al
Figure 6. Association of cycles to failure, bone mineral density, initial stiffness and final stiffness in plate-treated and IO FIX-treated bones. Pearson correlation coefficients observed were for cycles to failure r = .94, for bone mineral density r = .98, initial stiffness r = .91, and final stiffness r = .91.
biomechanical effect on the interfragmentary stability.15 Marks et al21 and Scranton et al34 reported that a nonlocking plantar plate has a higher load to failure and a lower initial displacement than screw fixation. Klos et al16 stated biomechanical advantages over the dorsomedially applied plate (both with an added compression screw) and Campbell et al3 have shown in a biomechanical study that plantar plate
fixation for a first MTC-osteotomy had better stability than screw fixation. Thus the aim of our study was to compare plantar plate treatment with intramedullary fixation, which is considered by many surgeons as the gold standard in joint arthrodesis but has rarely been considered in foot surgery. In comparison to plate osteosyntheses, which must be placed firmly against the cortical wall of the bone,
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1214
Foot & Ankle International 35(11)
Figure 7. Association of cycles to failure to maximum load to failure, bone mineral density, initial stiffness, and final stiffness, with Pearson correlation coefficients of .998, .01, .96, .95, respectively.
intramedullary fixation allows limited stripping of periosteum with preservation of blood supply and limited incision exposure.25 Another advantage of intramedullary osteosyntheses is the avoidance of soft tissue irritation caused by protruding osteosyntheses material8,27,36 avoiding the potential need for additional surgery to remove hardware. Over a 3- to 9-year duration of follow-up, DeVries et al9 reported the incidence of hardware removal following Lapidus fusion to be nearly identical for dorsomedial plate fixation (13%, 6 of 47 fusions) and screw fixation (15%, 14 of 96 fusions). Saxena et al33 described the incidence of hardware removal following Lapidus fusion with a dorsomedial plate to be 7.5% (3 of 40 fusions), over a 1- to 6-year duration of follow-up. There appears to be no data in the literature concerning the plantar plate, but it would seem plausible that the tibialis anterior tendon in particular, which some surgeons trim during plantar placement of the plate, might be damaged. Although the access to the site in plantar plating is comparatively large, the widely proximal location of the osteotomy can achieve a significant degree of correction. The recently introduced intraosseus screw and post implant should, according to the manufacturers, combine the advantages of soft tissue damage prevention over plates with high biomechanical stability. The current data show that plantar plate fixation strength was superior to the screw and post implant for MTC-joint arthrodesis. It would appear that here the plate demonstrates the biomechanical advantage of its plantar position, which the more dorsal positioning of the intramedullary implant is unable to match. Alongside the advantages the plantar plate obtains from its plantar positioning, the stiffness of the osteosyntheses would also seem to play a major role. This was related strongly to the number of cycles before the failure of the construct. Individual variations in BMD were eliminated as a possible cause of this observation. The study presented here is unable to determine the extent to which
the degree of interfragmentary compression can be held responsible for the variable results. However, the IO FIX fixed angle construct had limited compression because of a locking Morse taper. To balance out this disadvantage we also used a compression screw in addition to the plate construct for the screw and post group, which was not originally foreseen by the manufacturer. It was thus hoped to combine the positive biomechanical properties of the crossed arthrodesis screw with the greater plate pressure of the screw and post implant. Finally it is impossible to exclude that this screw has a negative influence on the biomechanical properties of the screw and post implant, particularly when it leads to an unintended fracture of the cortical shell at the entry point of the compression screw. This failure mechanism23 may be the cause of the premature failure of 2 implants in the screw and post group (Table 1). It is interesting that the number of cycles before failure and the load to failure of the corresponding matching partners (plate group) of these early failures also lay below the remaining specimens, although there were no great differences in the BMD compared to the other tested pairs. It would appear that factors such as individual variability of specimens might play a role here. In common with all biomechanical investigations, our study had certain limitations. In particular bone consolidation and the influence of soft-tissue preparation were obviously not observed. To reflect clinical practice we removed the articular cartilage. Ray et al29 have shown that differences in the preparation of the first MTC joint affect the biomechanical behavior of the specimens. The sample size in this study was quite small (n = 14). Nevertheless, the results are so striking that it is obvious that plate treatment resulted in higher durability than treatment with screw and post implant. There was some indication that the proportional hazard assumption was not met. However, other analysis strategies considered in sensitivity analyses have also
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1215
Roth et al Table 1. Descriptive Statistics for Parameters describing durability and bone properties. Group Plate IO FIX
Parameter Bone mineral density (mg/ccm) Number of cycles to failure Initial Stiffness (N/mm) Maximum load to failure (N) Final Stiffness (N/mm) Bone mineral density (mg/ccm) Number of cycles to failure Initial Stiffness (N/mm) Maximum load to failure (N) Final Stiffness (N/mm) Difference between Plate and IO FIX Difference in bone mineral density (mg/ccm) Difference in number of cycles to failure Difference in initial stiffness (N/mm) Difference in maximum load to failure (N) Difference in final stiffness (N/mm)
Minimum
Lower Quartile
Median
Upper Quartile
78.8 3798.2 82.5 75.2 103.9 79.8 2699.2 32.9 58.7 52.0
180.5 2528.0 22.0 70.0 55.0 137.9 1.0 5.2 0.0 0.0
216.7 3162.0 57.1 80.0 75.3 220.4 1.0 5.8 0.0 0.0
253.0 8167.0 153.7 180.0 241.3 227.0 2269.0 62.3 60.0 80.1
273.4 11581.0 195.0 250.0 263.2 268.4 5383.0 64.4 120.0 108.1
428.9 12183.0 243.1 260.0 308.6 402.2 6398.0 85.6 140.0 122.7
17.3
16.0
–4.3
5.0
18.5
26.7
42.6
7
4571.1
1578.4
2527.0
3161.0
4780.0
5898.0
6920.0
7
87.7
54.4
16.8
50.1
89.3
116.3
180.8
7
98.6
26.7
70.0
70.0
100.0
120.0
140.0
7
124.7
60.3
55.0
75.3
133.2
162.3
228.5
N
Mean
SD
7 7 7 7 7 7 7 7 7 7
263.9 7517.1 131.0 167.1 188.1 246.7 2946.0 43.3 68.6 63.4
7
demonstrated that the number of cycles to failure is significantly higher if plates were used rather than screw and post implant and that the difference in BMD did not account for this effect. Characteristics of the implant such as rigidity, stiffness and strength were interpreted generally as important characteristics of a successful osteosynthesis. This hypothesis would appear to be extendable beyond the biomechanical study to clinical routine. It remains unclear, however, to what extent they are actually required to provide sufficient stability for healing, The results of the present study thus cannot be directly translated into the clinical setting.
Authors’ Note
Conclusions
References
Plantar plate fixation of the first MTC fusion created a stronger and stiffer construct than intraosseous fixation. This was likely due to the plantar implant position. It is hoped that a stiffer construct will reduce the risk of nonunion and shorten the period of non-weight-bearing. Clinical studies will be required to show whether this translates into earlier resumption of weight-bearing and lower rates of nonunion. Acknowledgments The implants used in this study were donated by Wright Medical Technology, Inc, Arlington, TX, USA and Extremity Medical, Parsippany, NJ, USA.
Maximum
The manuscript presented here contains part of a doctoral thesis (JP).
Declaration of Conflicting Interests The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The implants used were donated by the manufacturer.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
1. Baravarian B, Briskin GB, Burns P. Lapidus bunionectomy: arthrodesis of the first metatarsocunieform joint. Clin Podiatr Med Surg. 2004;21(1):97-111. 2. Butson AR. A modification of the Lapidus operation for hallux valgus. J Bone Joint Surg Br. 1980;62(3):350-352. 3. Campbell JT, Schon LC, Parks BG, Wang Y, Berger BI. Mechanical comparison of biplanar proximal closing wedge osteotomy with plantar plate fixation versus crescentic osteotomy with screw fixation for the correction of metatarsus primus varus. Foot Ankle Int. 1998;19(5):293-299. 4. Catanzariti AR, Mendicino RW, Lee MS, Gallina MR. The modified Lapidus arthrodesis: a retrospective analysis. J Foot Ankle Surg. 1999;38(5):322-332.
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1216
Foot & Ankle International 35(11)
5. Clark HR, Veith RG, Hansen ST Jr. Adolescent bunions treated by the modified Lapidus procedure. Bull Hosp Joint Dis Orthop Inst. 1987;47(2):109-122. 6. Coetzee JC, Resig SG, Kuskowski M, Saleh KJ. The Lapidus procedure as salvage after failed surgical treatment of hallux valgus: a prospective cohort study. J Bone Joint Surg Am. 2003;85-A(1):60-65. 7. Cohen DA, Parks BG, Schon LC. Screw fixation compared to H-locking plate fixation for first metatarsocuneiform arthrodesis: a biomechanical study. Foot Ankle Int. 2005;26(11): 984-989. 8. Coughlin MJ, Grebing BR, Jones CP. Arthrodesis of the first metatarsophalangeal joint for idiopathic hallux valgus: intermediate results. Foot Ankle Int. 2005;26(10):783-792. 9. DeVries JG, Granata JD, Hyer CF. Fixation of first tarsometatarsal arthrodesis: a retrospective comparative cohort of two techniques. Foot Ankle Int. 2011;32(2):158-162. 10. Fuhrmann RA. Arthrodesis of the first tarsometatarsal joint for correction of the advanced splayfoot accompanied by a hallux valgus. Operative Orthopadie und Traumatologie. 2005;17(2):195-210. 11. Gruber F, Sinkov VS, Bae SY, Parks BG, Schon LC. Crossed screws versus dorsomedial locking plate with compression screw for first metatarsocuneiform arthrodesis: a cadaver study. Foot Ankle Int. 2008;29(9):927-930. 12. Gutteck N, Wohlrab D, Zeh A, Radetzki F, Delank KS, Lebek S. Comparative study of Lapidus bunionectomy using different osteosynthesis methods. Foot Ankle Surg. 2013;19(4): 218-221. 13. Klaue K. [Hallux valgus and hypermobility of the first ray—causal treatment using tarso-metatarsal reorientation arthrodesis]. Therapeutische Umschau Revue therapeutique. 1991;48(12):817-823. 14. Klaue K, Hansen ST, Masquelet AC. Clinical, quantitative assessment of first tarsometatarsal mobility in the sagittal plane and its relation to hallux valgus deformity. Foot Ankle Int. 1994;15(1):9-13. 15. Klos K, Gueorguiev B, Muckley T, et al. Stability of medial locking plate and compression screw versus two crossed screws for lapidus arthrodesis. Foot Ankle Int. 2010;31(2):158163. 16. Klos K, Simons P, Hajduk AS, et al. Plantar versus dorsomedial locked plating for Lapidus arthrodesis: a biomechanical comparison. Foot Ankle Int. 2011;32(11):1081-1085. 17. Klos K, Wilde CH, Lange A, et al. Modified Lapidus arthrodesis with plantar plate and compression screw for treatment of hallux valgus with hypermobility of the first ray: a preliminary report. Foot Ankle Surg. 2013;19(4):239-244. 18. Kopp FJ, Patel MM, Levine DS, Deland JT. The modified Lapidus procedure for hallux valgus: a clinical and radiographic analysis. Foot Ankle Int. 2005;26(11):913-917. 19. Lapidus PW. Operative correction of the metarasus primus varus in hallux valgus. Surg Gynecol Ostet. 1934;58:183-191.
20. Manoli A II, Hansen ST Jr. Screw hole preparation in foot surgery. Foot Ankle. 1990;11(2):105-106. 21. Marks RM, Parks BG, Schon LC. Midfoot fusion technique for neuroarthropathic feet: biomechanical analysis and rationale. Foot Ankle Int. 1998;19(8):507-510. 22. Mauldin DM, Sanders M, Whitmer WW. Correction of hallux valgus with metatarsocuneiform stabilization. Foot Ankle. 1990;11(2):59-66. 23. Messer TM, Nagle DJ, Martinez AG. Thumb metacarpophalangeal joint arthrodesis using the AO 3.0-mm cannulated screw: surgical technique. J Hand Surg. 2002;27(5):910-912. 24. Muckley T, Hoffmeier K, Klos K, Petrovitch A, von Oldenburg G, Hofmann GO. Angle-stable and compressed angle-stable locking for tibiotalocalcaneal arthrodesis with retrograde intramedullary nails. Biomechanical evaluation. J Bone Joint Surg Am. 2008;90(3):620-627. 25. Mueller ME, Allgoewer M, Schneider R, Willenegger H. Manual of Internal Fixation. Berlin, Germany: SpringerVerlag; 1991. 26. Myerson M, Allon S, McGarvey W. Metatarsocuneiform arthrodesis for management of hallux valgus and metatarsus primus varus. Foot Ankle. 1992;13(3):107-115. 27. O’Sullivan ME, Chao EY, Kelly PJ. The effects of fixation on fracture-healing. J Bone Joint Surg Am. 1989;71(2): 306-310. 28. Patel S, Ford LA, Etcheverry J, Rush SM, Hamilton GA. Modified lapidus arthrodesis: rate of nonunion in 227 cases. J Foot Ankle Surg. 2004;43(1):37-42. 29. Ray RG, Ching RP, Christensen JC, Hansen ST Jr. Biomechanical analysis of the first metatarsocuneiform arthrodesis. J Foot Ankle Surg. 1998;37(5):376-385. 30. Rink-Brune O. Lapidus arthrodesis for management of hallux valgus—a retrospective review of 106 cases. J Foot Ankle Surg. 2004;43(5):290-295. 31. Saffo G, Wooster MF, Stevens M, Desnoyers R, Catanzariti AR. First metatarsocuneiform joint arthrodesis: a five-year retrospective analysis. J Foot Surg. 1989;28(5):459-465. 32. Sangeorzan BJ, Hansen ST Jr. Modified Lapidus procedure for hallux valgus. Foot Ankle. 1989;9(6):262-266. 33. Saxena A, Nguyen A, Nelsen E. Lapidus bunionectomy: early evaluation of crossed lag screws versus locking plate with plantar lag screw. J Foot Ankle Surg. 2009;48(2):170-179. 34. Scranton PE, Coetzee JC, Carreira D. Arthrodesis of the first metatarsocuneiform joint: a comparative study of fixation methods. Foot Ankle Int. 2009;30(4):341-345. 35. Thompson IM, Bohay DR, Anderson JG. Fusion rate of first tarsometatarsal arthrodesis in the modified Lapidus procedure and flatfoot reconstruction. Foot Ankle Int. 2005;26(9): 698-703. 36. Ungersboek A, Geret V, Pohler O, Schuetz M, Wuest W. Tissue reaction to bone plates made of pure titanium: a prospective, quantitative clinical study. J Materials Sci. 1995;6:223-229
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
research-article2014
FAIXXX10.1177/1071100714547082Foot & Ankle InternationalRoth et al
Article
Intraosseous Fixation Compared to Plantar Plate Fixation for First Metatarsocuneiform Arthrodesis: A Cadaveric Biomechanical Analysis
Foot & Ankle International® 2014, Vol. 35(11) 1209–1216 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100714547082 fai.sagepub.com
Klaus Edgar Roth, MD1, Jennifer Peters1, Irene Schmidtmann, PhD2, Uwe Maus, MD1, Daniel Stephan, PhD3,4, and Peter Augat, PhD3,4
Abstract Background: Metatarsocuneiform (MTC) fusion is a treatment option for management of hallux valgus. We compared the biomechanical characteristics of an internal fixation device with plantar plate fixation. Methods: Seven matched pairs of feet from human cadavers were used to compare the intramedullary (IM) device plus compression screw to plantar plate combined with a compression screw. Specimen constructs were loaded in a cyclic 4-point bending test. We obtained initial/final stiffness, maximum load, and number of cycles to failure. Bone mineral density was measured with peripheral quantitative computed tomography. Performance was compared using time to event analysis with number of cycles as time variable, and a proportional hazard model including shared frailty model fitted with treatment and bone mineral density as covariates. Results: On average the plates failed after 7517 cycles and a maximum load of 167 N, while the IM-implants failed on average after 2946 cycles and a maximum load of 69 N. In all pairs the 1 treated with IM-implant failed earlier than the 1 treated with a plate (hazard ratio for IM-implant versus plate was 79.9 (95% confidence interval [6.1, 1052.2], P = .0009). The initial stiffness was 131 N/mm for the plantar plate and 43.3 N/mm for the IM implant. Initial stiffness (r = .955) and final stiffness (r = .952) were strongly related to the number of cycles to failure. Bone mineral density had no effect on the number of cycles to failure. Conclusion: Plantar plate fixation created a stronger and stiffer construct than IM fixation. Clinical Relevance: A stiffer construct can reduce the risk of nonunion and shorten the period of non-weight-bearing. Keywords: Lapidus fusion, intramedullary implant, plantar angle stable plate, biomechanic testing Arthrodesis of the first metatarsocuneiform joint (MTC-1) is a procedure used widely for the correction of an enlarged intermetatarsal angle. Modifications of the operative technique originally described by Lapidus19 have, in general, good treatment outcomes,2,5,20,22,32 with an incidence of revision due to nonunion and a revision of malpositioning of between 1.8%30 und 13%.32 Data from the current literature indicate that the risk of revision has fallen in recent years.4,6,18,26,28,30,31,35 This can be attributed to increased understanding of the biomechanics of the MTC-1 joint as well as improvements of the hardware, such as the use of locking plates. At present, probably the most frequently used fixation technique for Lapidus fusion is crossed-screw fixation.1,13,14,26 A further biomechanical advance would seem to be the placement of osteosynthetic materials plantar to the MTC-1 joint. Marks et al21 demonstrated that this technique had a more positive influence on interfragmentary stability compared to the widely used crossed screw
osteosynthesis, while Klos et al16 have shown the advantages of a plantar positioned locking plate over an angular stable locking plate positioned dorsomedially. From a biomechanical viewpoint a plantar positioned plate is better able to cope with the bending stresses to which the construct is subjected.15 More recent prospective studies indicate that the better biomechanical results of the plantar 1
Center of Orthopedic and Trauma Surgery, University Medical Center of the Johannes Gutenberg University, Mainz, Germany 2 Institute for Medical Biometry, Epidemiology and Computer Science, Johannes Gutenberg University, Mainz, Germany 3 Institute for Biomechanics, Traumacenter, Murnau, Germany 4 Paracelsus Medical University, Salzburg, Austria Corresponding Author: Klaus Edgar Roth, MD, Center of Orthopedic and Trauma Surgery, University Medical Center of the Johannes Gutenberg University Mainz, Langenbeckstraße 1, Mainz 55131, Germany. Email: [email protected]
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1210
Foot & Ankle International 35(11)
Figure 1. IO FIX.
Figure 2. Plantar plate.
plates are reflected in clinical practice by lower complication rates.12,17 On the other hand, the plantar osteosynthesis has been associated with more access morbidity with regard to the tibialis anterior tendon, which represents a hardwarerelated impingement problem. In contrast to plate osteosynthesis an intramedullary implant has the advantage of maintaining periosteal circulation25 and a reduced risk of hardware associated impingement.9 Periosteal perfusion is, in turn, a relevant factor in the healing process. The aim of the present study was to compare the stability of an intramedullary screw and poststabilization system with those of a locked plantar plate taking into account biomechanical characteristics and bone mineral density (BMD). We hypothesized that plantar plate fixation would create a stronger and stiffer construct than intramedullary fixation.
Materials and Methods Seven matched pairs of freshly frozen lower extremity cadaver specimens were used. The mean donor age at death was 76 (range, 69 to 89) years. Two of the donors were males and 5 were females. The specimens had no evidence of prior surgery or musculoskeletal disease. For each pair of specimens, 1 was treated with the intramedullary implant IO FIX (Extremity Medical™, Parsippany, NJ, USA) and 1 with plantar locking plate (Wright Medical Technology, Inc, Arlington, TX, USA); treatment was allocated randomly. One foot of each pair was randomly assigned to have screw and post implant (Figure 1) with adjunct 4-mm-diameter compression screw (Durchbohrte Schraube, Königsee Implantate GmbH, Allendorf, Germany). For the other foot of the pair, fixation of the first MTC-joint was carried out by plantar application of a Lapidus arthrodesis plate (Figure 2) with 2 gliding holes and a 4 mm compression screw. The use of cadaveric
specimens was approved by the local ethics commission. BMD was determined in the distal tibia with use of peripheral quantitative computed tomography (pQCT-XCT 2000, Stratec, Germany). A dorsomedial incision was made along the axis of the first MTC joint. The skin, subcutaneous fat, and fascia were elevated along the capsule exposing the first cuneiform and the first metatarsal. The first MTC capsule was transected to expose the articular surface. Next, a 1-mm section was taken from each opposing joint surface using an oscillating bone saw. Both parts of the study began with the insertion of a 4-mm cannulated, self-cutting, and selftapping compression screw with a partial (distal) thread centrally into the medial cortex of the first metatarsal, at a distance of 10 to 15 mm from the joint, and exiting in the middle of the intermediate cuneiform bone. Compression was applied by finger tightening of the screw, using a 3-finger (thumb and first 2 fingers) technique. The fixation of the arthrodesis was then completed with either screw and post or with plantar plate fixation. In both groups the additional compression screw was placed in the intermediate cuneiform bone. This procedure has been recommended to provide a greater load to failure and bending moment.29,32
IO FIX The screw and post device consisted of a lag screw in combination with an anchored post. The post is designed to distribute the compression forces uniformly across the arthrodesis site through a lag effect of the screw. In this way a greater surface pressure should be achieved on the osteomie (joine line) side than is the case for a lag screw. The post was placed in the metatarsal base whereas the lag screw crossed the joint and took place in the medial cuneiform. We used the fixed configuration of the implant, which locked the implant together at a fixed angle (Figure 3).
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1211
Roth et al
the construct stiffness was calculated between 10 and 40 N. Sinusiodal axial loading between lower and upper load level was performed at 1 Hz. Failure was defined as a displacement at the plantar aspect of the osteotomy of 3 mm or more and was measured by an extensometer (Instron Ltd, High Wycombe, UK).11 Cycles until failure, failure load, displacement, and plantar gapping were recorded. Stiffness of the construct was determined from the linear portion of load versus displacement curves. Construct stiffness in the last loading step before failure was determined. Construct stiffness was corrected with respect to the system stiffness. The performance of screw and post implant and plates was compared using time to event analysis with number of cycles and maximum load to failure as time variables. As pairs of specimens were considered, a shared frailty model was fitted to the time to event data. Usually in a Cox proportional hazard model the hazard for the i-th observation unit is given by λ i ( t , xi1 ,K xip ) ) = λ 0 ( t ) exp ( β1 xi1 + L + β p xi p )
Figure 3. IO FIX, X-ray.
Figure 4. Plantar plate, X-ray.
Plantar Plate At the plantar site, a plantar precontoured plate was used. Fixed-angle locked screws were inserted into the outer plate holes, while conventional nonlocked screws were used in the oval inner holes. Following the manufacturer’s recommended technique, one 3.5-mm screw was inserted into the distal inner screw hole crossing the osteotomy, to obtain additional interfragmentary compression (Figure 4). At the end of the implantation (screw and post/plantar plate) the bone segment containing the first metatarsal and first and second cuneiform was transected, eliminating the soft tissue completely. After instrumentation, the specimens were embedded in polyurethane (RENCAST©; Huntsman Advanced Materials GmbH, Switzerland). The implant and screws were coated with a modeling component to prevent contact with the embedding material. The potting was carried out using the technique of Gruber et al11 and Cohen et al.7 A 40 mm length of specimen was left exposed. Cyclic fatigue testing was performed on a servo- electric load frame (Instron E3000, Instron Ltd, High Wycombe, UK) under increasing axial load. The samples were loaded in a 4-point bending device. The upper support span was 60 mm and the lower support span was 160 mm.15 The upper load level started at 20 N and was increased by 10 N every 500 cycles. The lower load level remained at 10 N. Initially
with baseline hazard λ 0 ( t ) , covariates ( xi1 , xip ) ) for observation unit i, and regression coefficients ( β1 , , β p ) . If several observation units share certain properties, for example, specimens from 1 donor, this can be taken into account by adding a multiplicative frailty term to the model. Then λ ij ( t , xi j1 ,K xijp ) = Z i λ 0 ( t ) exp ( β1 xi j1 + L + β p xijp ) describes the hazard for bone j in donor i. Z i describes a random effect of the i-th donor, ( xi j1 , xijp ) are the covariates for bone j in donor I, and as before ( β1 , , β p ) are the regression coefficients. Here Z i is assumed to follow a Gamma distribution with expectation. Descriptive analysis included Kaplan–Meier estimates for the time to event data. Furthermore, means and standard deviations were computed for quantitative variables. Association between pairs of variables were displayed in scatter plots and described by computing the Pearson correlation coefficient. Differences between pairs of values were evaluated using the paired t test.
Results On average the plates failed after 7517 cycles and a maximum load of 167.1 N while the screw and post implants failed on average after 2946 cycles and a maximum load of 68.6 N. After 8167 cycles 50% of the plates had failed while the same failure rate was observed after 2269 cycles in the screw and post group (Figure 5). In all pairs of specimens the one treated with the screw and post implants failed earlier than the one treated with a plate. The estimated hazard ratio for the screw and post implant versus plate was 79.9 (95% confidence interval [6.1, 1052.2], P = .0009). The load was increased every 500 cycles, therefore the number of cycles to failure and the maximum load to failure essentially carry the same information. This is reflected in a
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1212
Foot & Ankle International 35(11)
Figure 5. Kaplan–Meier estimates of survival probability (probability that component has not yet failed) by treatment.
high correlation of these 2 variables (r = .998), therefore only results for cycles to failure are shown in detail. Initial stiffness (r = .955) and final stiffness (r = .952) were strongly related to number of cycles to failure (Figure 6). Thus these measures were all higher on average in the plate-treated group than in the screw and post group. Generally, there was a strong correlation between the parameters in the plate-treated and screw and post specimen of the same subject (Figure 6). Pearson correlation coefficients observed were for cycles to failure r = .94, for bone mineral density r = .98, initial stiffness r = .91, and final stiffness r = .91. In spite of the random allocation of treatment it turned out that in all but 1 pair of specimens, the 1 treated with a plate had a higher BMD than the 1 treated with screw and post implant. However, the differences were fairly small in comparison to the variability of bone densities between subjects: on average the BMD was 17.3 mg/ccm higher in the plate-treated specimens (SD of difference = 16.3 mg/ ccm), whereas BMD ranged from 137.9 mg/ccm to 428.9 (mean = 255.3, SD = 76.8). Because of this observation, we investigated the association of number of cycles to failure and BMD (Figure 7). We found that the Pearson correlation coefficient was only .010 (P = .7344). Consequently, when fitting a proportional hazard model with BMD as additional covariate, there was no effect of BMD on the number of cycles to failure, the hazard ratio being 0.997 per mg/ccm (95% confidence interval [0.963, 1.033], P = .8796). In total the
plates resisted 2.5 times the number of loading cycles before failure than did the screw and post construct. Descriptive statistics for all measured parameters can be found in Table 1.
Discussion Clinical outcomes with first MTC-joint (Lapidus) arthrodesis have been very good overall, but there is a 5% to 15% incidence of nonunion and malunion, depending on the study.3,7,26 A stable osteosynthesis is regarded as a necessary condition for rapid bone healing, reduction of the risk of nonunion and shortening of the period of non-weight-bearing.24 Currently, probably the most frequent fixation technique for Lapidus fusion is crossed-screw fixation.1,14,26 However plating in different positions is becoming increasingly popular,10 not least because of the availability of locked plating. Up until now the results of previous biomechanical studies of the stability of osteosyntheses at the MTC-1 have concerned the comparison of plates from different manufacturers and crossed-screw fixation. As, however, the plate configuration of plates intended for dorsomedial placement of different manufacturers show marked conceptual variations, statements concerning the biomechanical properties of a particular plate are only of limited relevance for other plates. The biomechanical principle of osteosyntheses on the plantar side which is under tension is markedly different compared to dorsomedial placement and suggests a beneficial
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1213
Roth et al
Figure 6. Association of cycles to failure, bone mineral density, initial stiffness and final stiffness in plate-treated and IO FIX-treated bones. Pearson correlation coefficients observed were for cycles to failure r = .94, for bone mineral density r = .98, initial stiffness r = .91, and final stiffness r = .91.
biomechanical effect on the interfragmentary stability.15 Marks et al21 and Scranton et al34 reported that a nonlocking plantar plate has a higher load to failure and a lower initial displacement than screw fixation. Klos et al16 stated biomechanical advantages over the dorsomedially applied plate (both with an added compression screw) and Campbell et al3 have shown in a biomechanical study that plantar plate
fixation for a first MTC-osteotomy had better stability than screw fixation. Thus the aim of our study was to compare plantar plate treatment with intramedullary fixation, which is considered by many surgeons as the gold standard in joint arthrodesis but has rarely been considered in foot surgery. In comparison to plate osteosyntheses, which must be placed firmly against the cortical wall of the bone,
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1214
Foot & Ankle International 35(11)
Figure 7. Association of cycles to failure to maximum load to failure, bone mineral density, initial stiffness, and final stiffness, with Pearson correlation coefficients of .998, .01, .96, .95, respectively.
intramedullary fixation allows limited stripping of periosteum with preservation of blood supply and limited incision exposure.25 Another advantage of intramedullary osteosyntheses is the avoidance of soft tissue irritation caused by protruding osteosyntheses material8,27,36 avoiding the potential need for additional surgery to remove hardware. Over a 3- to 9-year duration of follow-up, DeVries et al9 reported the incidence of hardware removal following Lapidus fusion to be nearly identical for dorsomedial plate fixation (13%, 6 of 47 fusions) and screw fixation (15%, 14 of 96 fusions). Saxena et al33 described the incidence of hardware removal following Lapidus fusion with a dorsomedial plate to be 7.5% (3 of 40 fusions), over a 1- to 6-year duration of follow-up. There appears to be no data in the literature concerning the plantar plate, but it would seem plausible that the tibialis anterior tendon in particular, which some surgeons trim during plantar placement of the plate, might be damaged. Although the access to the site in plantar plating is comparatively large, the widely proximal location of the osteotomy can achieve a significant degree of correction. The recently introduced intraosseus screw and post implant should, according to the manufacturers, combine the advantages of soft tissue damage prevention over plates with high biomechanical stability. The current data show that plantar plate fixation strength was superior to the screw and post implant for MTC-joint arthrodesis. It would appear that here the plate demonstrates the biomechanical advantage of its plantar position, which the more dorsal positioning of the intramedullary implant is unable to match. Alongside the advantages the plantar plate obtains from its plantar positioning, the stiffness of the osteosyntheses would also seem to play a major role. This was related strongly to the number of cycles before the failure of the construct. Individual variations in BMD were eliminated as a possible cause of this observation. The study presented here is unable to determine the extent to which
the degree of interfragmentary compression can be held responsible for the variable results. However, the IO FIX fixed angle construct had limited compression because of a locking Morse taper. To balance out this disadvantage we also used a compression screw in addition to the plate construct for the screw and post group, which was not originally foreseen by the manufacturer. It was thus hoped to combine the positive biomechanical properties of the crossed arthrodesis screw with the greater plate pressure of the screw and post implant. Finally it is impossible to exclude that this screw has a negative influence on the biomechanical properties of the screw and post implant, particularly when it leads to an unintended fracture of the cortical shell at the entry point of the compression screw. This failure mechanism23 may be the cause of the premature failure of 2 implants in the screw and post group (Table 1). It is interesting that the number of cycles before failure and the load to failure of the corresponding matching partners (plate group) of these early failures also lay below the remaining specimens, although there were no great differences in the BMD compared to the other tested pairs. It would appear that factors such as individual variability of specimens might play a role here. In common with all biomechanical investigations, our study had certain limitations. In particular bone consolidation and the influence of soft-tissue preparation were obviously not observed. To reflect clinical practice we removed the articular cartilage. Ray et al29 have shown that differences in the preparation of the first MTC joint affect the biomechanical behavior of the specimens. The sample size in this study was quite small (n = 14). Nevertheless, the results are so striking that it is obvious that plate treatment resulted in higher durability than treatment with screw and post implant. There was some indication that the proportional hazard assumption was not met. However, other analysis strategies considered in sensitivity analyses have also
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1215
Roth et al Table 1. Descriptive Statistics for Parameters describing durability and bone properties. Group Plate IO FIX
Parameter Bone mineral density (mg/ccm) Number of cycles to failure Initial Stiffness (N/mm) Maximum load to failure (N) Final Stiffness (N/mm) Bone mineral density (mg/ccm) Number of cycles to failure Initial Stiffness (N/mm) Maximum load to failure (N) Final Stiffness (N/mm) Difference between Plate and IO FIX Difference in bone mineral density (mg/ccm) Difference in number of cycles to failure Difference in initial stiffness (N/mm) Difference in maximum load to failure (N) Difference in final stiffness (N/mm)
Minimum
Lower Quartile
Median
Upper Quartile
78.8 3798.2 82.5 75.2 103.9 79.8 2699.2 32.9 58.7 52.0
180.5 2528.0 22.0 70.0 55.0 137.9 1.0 5.2 0.0 0.0
216.7 3162.0 57.1 80.0 75.3 220.4 1.0 5.8 0.0 0.0
253.0 8167.0 153.7 180.0 241.3 227.0 2269.0 62.3 60.0 80.1
273.4 11581.0 195.0 250.0 263.2 268.4 5383.0 64.4 120.0 108.1
428.9 12183.0 243.1 260.0 308.6 402.2 6398.0 85.6 140.0 122.7
17.3
16.0
–4.3
5.0
18.5
26.7
42.6
7
4571.1
1578.4
2527.0
3161.0
4780.0
5898.0
6920.0
7
87.7
54.4
16.8
50.1
89.3
116.3
180.8
7
98.6
26.7
70.0
70.0
100.0
120.0
140.0
7
124.7
60.3
55.0
75.3
133.2
162.3
228.5
N
Mean
SD
7 7 7 7 7 7 7 7 7 7
263.9 7517.1 131.0 167.1 188.1 246.7 2946.0 43.3 68.6 63.4
7
demonstrated that the number of cycles to failure is significantly higher if plates were used rather than screw and post implant and that the difference in BMD did not account for this effect. Characteristics of the implant such as rigidity, stiffness and strength were interpreted generally as important characteristics of a successful osteosynthesis. This hypothesis would appear to be extendable beyond the biomechanical study to clinical routine. It remains unclear, however, to what extent they are actually required to provide sufficient stability for healing, The results of the present study thus cannot be directly translated into the clinical setting.
Authors’ Note
Conclusions
References
Plantar plate fixation of the first MTC fusion created a stronger and stiffer construct than intraosseous fixation. This was likely due to the plantar implant position. It is hoped that a stiffer construct will reduce the risk of nonunion and shorten the period of non-weight-bearing. Clinical studies will be required to show whether this translates into earlier resumption of weight-bearing and lower rates of nonunion. Acknowledgments The implants used in this study were donated by Wright Medical Technology, Inc, Arlington, TX, USA and Extremity Medical, Parsippany, NJ, USA.
Maximum
The manuscript presented here contains part of a doctoral thesis (JP).
Declaration of Conflicting Interests The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The implants used were donated by the manufacturer.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
1. Baravarian B, Briskin GB, Burns P. Lapidus bunionectomy: arthrodesis of the first metatarsocunieform joint. Clin Podiatr Med Surg. 2004;21(1):97-111. 2. Butson AR. A modification of the Lapidus operation for hallux valgus. J Bone Joint Surg Br. 1980;62(3):350-352. 3. Campbell JT, Schon LC, Parks BG, Wang Y, Berger BI. Mechanical comparison of biplanar proximal closing wedge osteotomy with plantar plate fixation versus crescentic osteotomy with screw fixation for the correction of metatarsus primus varus. Foot Ankle Int. 1998;19(5):293-299. 4. Catanzariti AR, Mendicino RW, Lee MS, Gallina MR. The modified Lapidus arthrodesis: a retrospective analysis. J Foot Ankle Surg. 1999;38(5):322-332.
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
1216
Foot & Ankle International 35(11)
5. Clark HR, Veith RG, Hansen ST Jr. Adolescent bunions treated by the modified Lapidus procedure. Bull Hosp Joint Dis Orthop Inst. 1987;47(2):109-122. 6. Coetzee JC, Resig SG, Kuskowski M, Saleh KJ. The Lapidus procedure as salvage after failed surgical treatment of hallux valgus: a prospective cohort study. J Bone Joint Surg Am. 2003;85-A(1):60-65. 7. Cohen DA, Parks BG, Schon LC. Screw fixation compared to H-locking plate fixation for first metatarsocuneiform arthrodesis: a biomechanical study. Foot Ankle Int. 2005;26(11): 984-989. 8. Coughlin MJ, Grebing BR, Jones CP. Arthrodesis of the first metatarsophalangeal joint for idiopathic hallux valgus: intermediate results. Foot Ankle Int. 2005;26(10):783-792. 9. DeVries JG, Granata JD, Hyer CF. Fixation of first tarsometatarsal arthrodesis: a retrospective comparative cohort of two techniques. Foot Ankle Int. 2011;32(2):158-162. 10. Fuhrmann RA. Arthrodesis of the first tarsometatarsal joint for correction of the advanced splayfoot accompanied by a hallux valgus. Operative Orthopadie und Traumatologie. 2005;17(2):195-210. 11. Gruber F, Sinkov VS, Bae SY, Parks BG, Schon LC. Crossed screws versus dorsomedial locking plate with compression screw for first metatarsocuneiform arthrodesis: a cadaver study. Foot Ankle Int. 2008;29(9):927-930. 12. Gutteck N, Wohlrab D, Zeh A, Radetzki F, Delank KS, Lebek S. Comparative study of Lapidus bunionectomy using different osteosynthesis methods. Foot Ankle Surg. 2013;19(4): 218-221. 13. Klaue K. [Hallux valgus and hypermobility of the first ray—causal treatment using tarso-metatarsal reorientation arthrodesis]. Therapeutische Umschau Revue therapeutique. 1991;48(12):817-823. 14. Klaue K, Hansen ST, Masquelet AC. Clinical, quantitative assessment of first tarsometatarsal mobility in the sagittal plane and its relation to hallux valgus deformity. Foot Ankle Int. 1994;15(1):9-13. 15. Klos K, Gueorguiev B, Muckley T, et al. Stability of medial locking plate and compression screw versus two crossed screws for lapidus arthrodesis. Foot Ankle Int. 2010;31(2):158163. 16. Klos K, Simons P, Hajduk AS, et al. Plantar versus dorsomedial locked plating for Lapidus arthrodesis: a biomechanical comparison. Foot Ankle Int. 2011;32(11):1081-1085. 17. Klos K, Wilde CH, Lange A, et al. Modified Lapidus arthrodesis with plantar plate and compression screw for treatment of hallux valgus with hypermobility of the first ray: a preliminary report. Foot Ankle Surg. 2013;19(4):239-244. 18. Kopp FJ, Patel MM, Levine DS, Deland JT. The modified Lapidus procedure for hallux valgus: a clinical and radiographic analysis. Foot Ankle Int. 2005;26(11):913-917. 19. Lapidus PW. Operative correction of the metarasus primus varus in hallux valgus. Surg Gynecol Ostet. 1934;58:183-191.
20. Manoli A II, Hansen ST Jr. Screw hole preparation in foot surgery. Foot Ankle. 1990;11(2):105-106. 21. Marks RM, Parks BG, Schon LC. Midfoot fusion technique for neuroarthropathic feet: biomechanical analysis and rationale. Foot Ankle Int. 1998;19(8):507-510. 22. Mauldin DM, Sanders M, Whitmer WW. Correction of hallux valgus with metatarsocuneiform stabilization. Foot Ankle. 1990;11(2):59-66. 23. Messer TM, Nagle DJ, Martinez AG. Thumb metacarpophalangeal joint arthrodesis using the AO 3.0-mm cannulated screw: surgical technique. J Hand Surg. 2002;27(5):910-912. 24. Muckley T, Hoffmeier K, Klos K, Petrovitch A, von Oldenburg G, Hofmann GO. Angle-stable and compressed angle-stable locking for tibiotalocalcaneal arthrodesis with retrograde intramedullary nails. Biomechanical evaluation. J Bone Joint Surg Am. 2008;90(3):620-627. 25. Mueller ME, Allgoewer M, Schneider R, Willenegger H. Manual of Internal Fixation. Berlin, Germany: SpringerVerlag; 1991. 26. Myerson M, Allon S, McGarvey W. Metatarsocuneiform arthrodesis for management of hallux valgus and metatarsus primus varus. Foot Ankle. 1992;13(3):107-115. 27. O’Sullivan ME, Chao EY, Kelly PJ. The effects of fixation on fracture-healing. J Bone Joint Surg Am. 1989;71(2): 306-310. 28. Patel S, Ford LA, Etcheverry J, Rush SM, Hamilton GA. Modified lapidus arthrodesis: rate of nonunion in 227 cases. J Foot Ankle Surg. 2004;43(1):37-42. 29. Ray RG, Ching RP, Christensen JC, Hansen ST Jr. Biomechanical analysis of the first metatarsocuneiform arthrodesis. J Foot Ankle Surg. 1998;37(5):376-385. 30. Rink-Brune O. Lapidus arthrodesis for management of hallux valgus—a retrospective review of 106 cases. J Foot Ankle Surg. 2004;43(5):290-295. 31. Saffo G, Wooster MF, Stevens M, Desnoyers R, Catanzariti AR. First metatarsocuneiform joint arthrodesis: a five-year retrospective analysis. J Foot Surg. 1989;28(5):459-465. 32. Sangeorzan BJ, Hansen ST Jr. Modified Lapidus procedure for hallux valgus. Foot Ankle. 1989;9(6):262-266. 33. Saxena A, Nguyen A, Nelsen E. Lapidus bunionectomy: early evaluation of crossed lag screws versus locking plate with plantar lag screw. J Foot Ankle Surg. 2009;48(2):170-179. 34. Scranton PE, Coetzee JC, Carreira D. Arthrodesis of the first metatarsocuneiform joint: a comparative study of fixation methods. Foot Ankle Int. 2009;30(4):341-345. 35. Thompson IM, Bohay DR, Anderson JG. Fusion rate of first tarsometatarsal arthrodesis in the modified Lapidus procedure and flatfoot reconstruction. Foot Ankle Int. 2005;26(9): 698-703. 36. Ungersboek A, Geret V, Pohler O, Schuetz M, Wuest W. Tissue reaction to bone plates made of pure titanium: a prospective, quantitative clinical study. J Materials Sci. 1995;6:223-229
Downloaded from fai.sagepub.com at GEORGETOWN UNIV MED CTR on May 23, 2015
