520695
research-article2014
FAIXXX10.1177/1071100714520695Foot & Ankle InternationalMayer et al
Article
A Comparison of Nonlocking Semitubular Plates and Precontoured Locking Plates for First Metatarsophalangeal Joint Arthrodesis
Foot & Ankle International® 2014, Vol. 35(5) 438–444 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100714520695 fai.sagepub.com
Stephanie A. Mayer, MD1, Nicole A. Zelenski, BA2, James K. DeOrio, MD1, Mark E. Easley, MD1, and James A. Nunley II, MD1 Abstract Background: First metatarsophalangeal (MTP) joint arthrodesis is a successful, commonly performed procedure for many conditions affecting this joint. The purpose of this retrospective study was to compare the clinical and radiographic outcomes between patients who had primary first MTP joint fusions with either noncontoured stainless steel semitubular plates or precontoured plates. Our hypothesis was that there would be no difference in clinical or radiographic outcomes between these groups. Methods: A search based on operative report coding was performed to detect all patients who had undergone primary first MTP joint arthrodesis with a dorsal plating construct at our institution from 2005 to 2010. A retrospective review of electronic medical records and postoperative radiographs was performed to determine time to clinical and radiographic union, pain scores, complications, MTP angle, and proximal phalanx to floor angle. A Kruskal-Wallis test was used for continuous variables, and a chi-square or Fisher exact test was used for categorical variables to determine statistical significance between the 2 groups. Results: We identified 128 patients who met our inclusion criteria. One hundred and two feet in 97 patients underwent arthrodesis with a noncontoured, stainless steel 1/3 tubular plate (group 1) and 26 feet in 21 patients with a precontoured, locking plate (group 2). The patients presented with a variety of inflammatory and noninflammatory preoperative diagnoses. The overall union rate, complication rate, time to radiographic healing, MTP angle, and patient-reported visual analog scale (VAS) score were similar between the 2 groups. Clinical time to healing was more rapid in the noncontoured group (3.7 months vs 4.8 degrees in precontoured) (P = .02). The radiographically measured proximal phalanx to floor angle was significantly decreased in noncontoured patients (2.6 degrees vs 4.6 degrees in precontoured, P = .04). In patients who had a diagnosis of an inflammatory arthropathy, the union rate fell to 84.3% (27/32) in the noncontoured group and 87.5% (7/8) in the precontoured group, and time to radiographic as well as clinical union increased in both groups. When patients with inflammatory arthropathies were compared with noninflammatory patients within groups, it was found that the time to both clinical and radiographic union was significantly longer in the noncontoured group for those with an inflammatory arthropathy (P = .0052, P = .022). For the precontoured group, these values were not statistically significant (P = .089, P = .43). Additionally, the cost of the noncontoured implants was significantly less than that of the precontoured implants. Conclusion: In this study, clinical time to healing of primary first MTP fusion with dorsal plate and screw constructs was improved with the use of noncontoured plates. However, other clinical indicators such as pain, complication rate, MTP angle, and VAS score did not differ between groups. Noncontoured plates are less expensive, and thus the decision to use precontoured plates routinely should be made cautiously. However, in the presence of inflammatory arthropathy, the use of precontoured plates may be justified given that the use of these plates in the inflammatory group resulted in improved radiographic and clinical union. Level of Evidence: Level III, comparative series. Keywords: hallux disorders, arthritis, outcome studies First metatarsophalangeal (MTP) joint arthrodesis is commonly performed for a variety of conditions affecting the great toe, including hallux valgus, hallux rigidus, hallux varus, neuromuscular conditions, failed arthrodesis, and inflammatory arthropathies (IAs).1,2,4-7,9,12,15,16 Several techniques for preparation of the bony surfaces for arthrodesis have been described; however, the most commonly accepted
1
Department of Orthopaedic Surgery, Duke University Medical Center, Durham, NC, USA 2 Duke University Medical Center, Durham, NC, USA Corresponding Author: Nicole A. Zelenski, BA, Duke University Medical Center, Department of Orthopaedic Surgery, 200 Trent Drive, Durham, NC 27710, USA. Email: [email protected]
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
439
Mayer et al technique currently is the cup and cone reaming technique.1,2,5,7,9,12 Additionally, many fixation constructs have been used, such as Kirschner wires, tension band wires, crossed screws, and dorsal plates with or without a lag screw.1-7,9-12,14-16 Dorsal plate and screw constructs have proven to be biomechanically superior to other fixation methods and therefore are commonly used.3,10,11,15,16 Both nonlocking stainless steel and titanium plates are available,1,4 and with the advances in locking plate technology, precontoured stainless steel and titanium locking plates are now available for MTP fusion.2,5-7,9,10,12 Union rates using dorsal plate and screw constructs vary from 77.2% to 100%.1,2,4-7,9,12 Clinical outcomes following this procedure are generally good to excellent. The purpose of this study was to compare the radiographic and clinical outcomes of 2 groups of patients who underwent primary first MTP joint arthrodesis with a dorsal plate and screw construct with either a precontoured locking plate or a noncontoured, nonlocking,1/3 tubular stainless steel plate. Given the large discrepancy in pricing between these 2 implants, we also wanted to make comparisons based on cost. Our hypothesis was that there would be no difference in radiographic or clinical outcomes between the groups and that the difference in cost would make the noncontoured plates more attractive.
Methods After institutional review board approval was granted, a search was performed based on operative report Current Procedural Terminology (CPT) codes for all patients who underwent first MTP joint arthrodesis by 1 of 3 orthopedic foot and ankle surgeons between January 2006 and May 2011. Patients were included if they were over 18 years old, had undergone primary MTP arthrodesis with dorsal plating techniques, and had follow-up until radiographic union occurred or a decision for revision surgery was made. Patients who underwent revision arthrodesis were excluded; however, primary arthrodesis as a salvage procedure following other operations including failed bunion correction, hallux varus correction, cheilectomy, and soft tissue arthroplasty was included. A retrospective review of electronic medical records and radiographs was performed to record diagnosis, previous procedures, demographic data, visual analog scale (VAS) scores, time to clinical healing, time to radiographic healing, final MTP angle, final proximal phalanx to floor angle, complications, and revision surgeries. We identified 128 feet in 118 patients that met our inclusion criteria: 102 feet in 97 patients underwent arthrodesis with a noncontoured, stainless steel 1/3 tubular plate (group 1) and 26 feet in 21 patients with a precontoured, locking plate (group 2). There was no notable difference in mean age between the groups—60.5 years in the noncontoured group and 61 years in the precontoured group. The patients
Table 1. Preoperative Diagnoses. Diagnosis Degenerative arthritis Rheumatoid arthritis Hallux varus Gout Neuromuscular deformity Wegener’s granulomatosis Psoriatic arthritis Scleroderma Postpolio deformity Salvage after failed arthroplasty
Noncontoured Precontoured 61 27 4 1 1 1 2 1 0 4
13 8 1 0 0 0 0 0 2 2
Table 2. Use of Lag Screws in Fixation Construct. Precontoured Noncontoured (n = 26) (n = 102) Total patients with lag screws Lag screw in IA group Lag screws in non-IA group Nonunions without lag screws
8 2/8 6/18 2/2
102 32/32 70/70 0/8
Abbreviations: IA, inflammatory arthropathy; non-IA, noninflammatory arthropathy.
presented with a variety of preoperative diagnoses, including hallux valgus, hallux rigidus, hallux varus, rheumatoid arthritis, gout, psoriatic arthritis, failed arthroplasty, Wegener’s granulomatosis, and neuromuscular hallux valgus. A summary of demographic data and preoperative diagnoses is presented in Table 1. There were 32 feet (30.4%) in the noncontoured group and 8 feet (30.7%) in the precontoured group with a diagnosis of an IA (rheumatoid arthritis, gout, or psoriatic arthritis). Lag screws were used in all patients with noncontoured plate and in 8 of the 26 patients (31%) with contoured plate. Of the lag screws used in the contoured group, there was little difference in the incidence used in patients with IAs and in those with noninflammatory arthropathies (non-IAs) (25% and 33%, respectively) (Table 2). Time to clinical healing was recorded by the attending surgeon as the time point when the patient had minimal pain, there was no motion through the fusion site clinically, wounds were healed appropriately, and the patient was deemed appropriate for full weight-bearing in regular shoe wear. Time to radiographic union was evaluated by plain films and was recorded by the attending physician when there was bony trabeculation crossing the fusion site. MTP joint and proximal phalanx to floor angles were measured by a single observer. On the standing lateral radiographs taken at the final follow-up visit, 2 measurements were taken for each angle type and the mean was used for statistical analysis. MTP joint angles were measured using the Cobb angle function on an electronic radiology system (Centricity Web
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
440
Foot & Ankle International 35(5)
Figure 1. (A) Lateral and (B) anteroposterior radiograph of a noncontoured, nonlocking 1/3 tubular plate demonstrating metatarsophalangeal angle (A: the angle between X and Y) and proximal phalanx to floor angle (A: the angle between Y and Z). (C) Lateral and (D) anteroposterior radiograph of a precontoured, locking plate.
Enterprises, GE Healthcare, Waukesha, WI) defined as the angle between a line parallel to the shaft of the metatarsal and a line parallel to the shaft of the proximal phalanx (Figure 1A). The proximal phalanx to floor angle was also measured with the Cobb angle function and was defined as the angle between a line parallel to the shaft of the proximal phalanx and a line parallel to the floor (Figure 1A). Data were collected at 6-week, 3-month, 6-month, and 12-month visits. No patients in this study had follow-up longer than 12 months because the determination of union or nonunion, the need for revision due to symptomatic nonunion, and any postoperative complications were recorded by this time point. Statistical analysis was performed using Student t test or Kruskal Wallis test (if nonparametric) for continuous variables and a chi-square or Fisher exact test (if values were less than 5) for categorical variables to determine significance between the 2 groups. A P value of .05 was used for statistical significance. The operative technique used by each surgeon was similar. Patients underwent either general anesthesia with or without a regional block or regional block with sedation. A dorsal
or dorsomedial skin incision was made. All surgeons used a cannulated cup and cone reamer system. The toe was then positioned both clinically and radiologically such that the tip of the toe rested on a flat metal surface and there was approximately 0 to 8 degrees of valgus and 20 to 30 degrees of dorsiflexion at the MTP joint level. The joint was provisionally pinned either with a retrograde K-wire or with a plantarmedial to dorsal-lateral K-wire. If a precontoured locking plate was chosen, the plate was applied and provisionally fixed to the bone. Fluoroscopy was then used to determine appropriate plate placement, and if needed the decision was made to change the bend in the plate to match the patient’s anatomy. If a noncontoured 1/3 tubular plate was chosen, the plate was also provisionally secured with K-wires and evaluated for placement and the appropriate bend needed to match the patient’s anatomy (Figure 1). The decision for the type and size of plate, and for any additional lesser toe procedures performed concomitantly, was made at the discretion of the surgeon. Postoperatively, the patients were kept non-weightbearing initially and allowed to begin progressive protected weight-bearing usually between 3 to 6 weeks.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
441
Mayer et al
Table 3. Comparison of Union Rate, Time to Clinical and Radiographic Union, and Radiographic Measurements Between Groups.
Union rate Time to clinical union Time to radiographic union Metatarsophalangeal angle Proximal phalanx to floor angle
Noncontoured
Precontoured
P Value
92.2% 3.7 months 4.0 months 24.2 degrees 2.6 degrees
92.3% 4.6 months 4.3 months 26.0 degrees 4.6 degrees
.02 .37 .09 .04
Table 4. Total Complications in Noncontoured and Precontoured Groups. Noncontoureda
Precontouredb
8 2 3 1 1 1 4 3 1 1
2 1 0 2 0 1 0 1 0 0
Nonunion Malunion Superficial infection Deep infection Delayed wound healing Wound dehiscence Hardware failure Painful hardware Painful metatarsophalangeal osteophyte Interphalangeal joint contracture a
26 complications in 19 feet in 18 patients. 7 complications in 4 feet in 4 patients.
b
Results The overall union rate was 92.2% (94/102) for the noncontoured group and 92.3% (24/26) for the precontoured group. The difference in the mean time to radiographic union between the noncontoured group (4.0 months) and precontoured group (4.3 months) did not reach statistical significance (P = .37). The difference in mean clinical healing time between the noncontoured group (3.7 months) and precontoured group (4.6 months) reached statistical significance (P = .02). There were significantly more patients clinically healed at the 3-month follow-up in the noncontoured group (74.5% vs 50%) (P = .02). Otherwise, the proportion of patients in each group who were deemed clinically or radiographically healed was not significantly different at any of the other time points measured. There was no notable difference in the patient-reported VAS score between groups at any time point. VAS scores in both groups improved at each time point. At 6 weeks, 15.7% (16/102) noncontoured versus 3.9% (1/26) precontoured were clinically healed (P = 0.19), and 12.8% (13/102) noncontoured versus 0% precontoured were radiographically healed (P = .06). At 3 months, 74.5% (76/102) noncontoured versus 50% (13/26) precontoured were clinically healed (P = .02), and 66.7% (68/102) noncontoured versus 61.5% (16/26) precontoured were radiographically healed. The mean radiographically measured MTP angle was 24.2 degrees (range, 6.0-43.7 degrees) for the noncontoured
group and 26.0 degrees (range, 14.3-46.1 degrees) for the precontoured group, which did not reach statistical significance (P = .09). The mean radiographically measured proximal phalanx to floor angle was 2.6 degrees (range, –14.1 to 30.0 degrees) for the noncontoured group and 4.6 degrees (range, –10.3 to 19.7 degrees) for the precontoured group (P = .04) (Table 3). There were 25 complications in 18 feet (24.3%) in the noncontoured group and 7 complications in 4 feet (26.9%) in the precontoured group. There were 8 nonunions in the noncontoured group (7.8%) and 2 in the precontoured group (7.7%), both of which also had deep infections. Of the 2 nonunions in the precontoured group, neither had a lag screw. Other complications reported included superficial and deep infections, delayed wound healing, wound dehiscence, hardware failure and breakage, malunion, painful hardware, persistent painful osteophyte, and interphalangeal joint contracture (Table 4). There were 9 reoperations in the noncontoured group: 3 for nonunion revision; 2 for malunion; 1 each for irrigation and debridement and hardware removal for infection, excision of persistent painful osteophytes, and removal of painful hardware; and 1 for removal of painful hardware and release of an interphalangeal joint contracture in the same patient. Reoperations in the precontoured group included irrigation and debridement and removal of hardware in 2 patients who went on to nonunion and 1 revision due to malunion and painful hardware.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
442
Foot & Ankle International 35(5)
In patients who had a diagnosis of an IA, the union rate fell to 84.3% (27/32) in the noncontoured group and 87.5% (7/8) in the precontoured group. Time to radiographic union increased to 4.8 months in the noncontoured group and 4.3 months in the precontoured group, while time to clinical union rose to 4.8 months in the noncontoured group but fell to 3.4 months in the precontoured group (P = .18, P = .32). Thirteen of 25 overall complications, including 5 of the 8 nonunions and all 4 infections in the noncontoured group, were in patients with an IA. Similarly, 3 of the 7 total complications were in this patient population. Excluding the patients with IAs from the analysis, the union rate improved to 95.7% in the noncontoured group and 94.4% in the precontoured group. Complications occurred in 8 of 71 patients in the noncontoured group and 2 of 24 patients in the precontoured group. Time to radiographic union was not statistically significant between groups. It improved to 3.7 months in the noncontoured group and 4.2 in the precontoured group (P = .28). Time to clinical union was significantly shorter at 3.3 months in the noncontoured group versus 5.0 months in the precontoured group (P = .0009). Comparing the patients who had IAs to those who did not within groups, the time to both clinical and radiographic union was significantly shorter in the noncontoured group for those without an IA (P = .0052, P = .022). For the precontoured group, these values were not statistically significant (P = .089, P = .43).
Discussion Arthrodesis of the first MTP joint is a common procedure, and as the population ages, the number of procedures per year will likely increase. As such, it would be ideal for the foot and ankle surgeon to have a reliable construct for fixation that is also affordable for the health care system. We sought to determine any clinical or radiographic difference in outcomes between the use of noncontoured, nonlocking, stainless steel 1/3 tubular plates and precontoured, locking plates for fusion. The nonlocking plates are more affordable and therefore would be more attractive if the outcomes are similar. Previous reports of MTP fusion in recent literature have cited between 77% and 100% union rates using dorsal plating techniques.1,2,4-7,9,12 A variety of dorsal plates exist, with options for noncontoured or precontoured, stainless steel or titanium, and locking or nonlocking. Several authors have compared dorsal plating constructs with or without a lag screw to other fixation methods such as crossed screws for their biomechanical properties and have found the dorsal plating technique to be stiffer.3,10,11,15,16 However, at least 1 article reports on the comparison between crossed screws and dorsal plates showing an equivalent union rate and clinical outcome.11 Because of the majority of the biomechanical
data, dorsal plating is one of the most commonly used fixation method currently. Locking plates have been used extensively for periarticular fractures, osteoporotic bone, severe comminution, and unstable metadiaphyseal fractures.8,18 The use of locking plates for first MTP fusion has been advocated by some given the low profile of the plate, ease of use, theoretical time savings in the operating room since the need to bend the plate should be obviated, and superior biomechanical properties in the above situations.2,5-7,9 Despite these potential advantages, most studies have shown that complications still occur along with a similar rate of nonunion.2,5-7,9,12 A biomechanical analysis comparing precontoured stainless steel locked and nonlocked plates for MTP fusion in a cadaver model found that precontoured locked plates are significantly stiffer and had less plantar gapping in loading tests.10 Interestingly, a recent article comparing noncontoured and precontoured locking plates found a higher rate of nonunion with locking plates, although this did not reach statistical significance.9 The authors hypothesized that the stiffness of the locking construct was not ideal for healing of the arthrodesis site since compression across the fusion site cannot occur.9 Our study found overall equivalent union rates for both type of plate. Our union rates are similar to those of other published studies for MTP fusion using dorsal plates of both types.1,2,4-7,9,12 We also found that the time to both clinical and radiographic union and the proportion of patients who were deemed clinically healed at each time point were similar between groups at all but 1 time point. Patientreported pain improved over time, and scores were comparable between the 2 constructs. The radiographic measurements for MTP angle and proximal phalanx to floor angle have been shown to correlate with the MTP angle built into precontoured plates as well as a good clinical outcome for MTP fusion.2,13 The mean MTP angle in this study was within the ideal range of 15 to 30 degrees2,5,13 for both groups and was not statistically different between groups. The proximal phalanx to floor angle was statistically different at 2.6 degrees for the noncontoured group and 4.6 degrees for the precontoured group; however, the clinical significance of this as far as functional results or patient satisfaction is unclear. There were no revisions for malunion involving a toe that did not touch the ground in either group. One potential advantage of the precontoured plates is the MTP angle built into the plate, which should decrease the need for decision making intraoperatively about the appropriate amount of bending of the plate. Despite this, the final measurements between the MTP angle and the proximal phalanx to floor angles were similar in both groups. Approximately half of the complications in each group were in patients with a diagnosis of an IA. Other reports of MTP fusion have also had a higher rate of complications in
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
443
Mayer et al these patients.5,9,17 Of the patients with this diagnosis, the rate of complications was similar between groups but was higher than the rate of complications for patients without this diagnosis. Both nonunions in the precontoured group occurred in the setting of deep infection requiring removal of hardware, and a larger prospective study would further elucidate whether the proportion of nonunions in this group without infection would be significantly different than that in the noncontoured group. Additionally, both of the nonunions in the precontoured group lacked lag screws, which add stability to the construct. In our study, both the union rate and complication rate were equivalent between groups overall and when we compared patients with and without IAs separately. Taking into account a diagnosis of an IA when comparing time to clinical and radiographic union, there are interesting findings. Regarding clinical healing times, non-IA patients healed significantly faster in the noncontoured versus precontoured group. Unexpectedly, the IA patients had a 6-week shorter time to clinical healing than non-IA patients when using the precontoured plate; however, this did not reach statistical significance. Within the noncontoured group, the non-IA patients were clinically healed significantly faster. Regarding radiographic healing times, within the noncontoured group, the non-IA patients were healed significantly faster. In the precontoured group there was no difference between IA and non-IA patients. There was a small trend toward faster radiographic healing in IA patients in the precontoured group and a stronger trend toward faster healing in the non-IA patients in the noncontoured group. These results point to a possible advantage of noncontoured plates in non-IA patients and precontoured plates in patients with an IA. Limitations of this study include the inherent bias of a retrospective cohort study, small sample size, limited patient satisfaction outcome scores, and heterogeneous follow-up assessment and clinical judgment between 3 surgeons. There is a paucity of data for this type of precontoured locked plate for MTP arthrodesis in the literature, and therefore there is limited comparison for our outcome data. Several variables differed between the plate types, including locking versus nonlocking and the number of screws for plate fixation and lag screws. In addition, the retrospective nature of this study means that the addition of a lag screw or K-wire for fixation or prolonged protected weight-bearing was at the discretion of the surgeon. In conclusion, the clinical and radiographic outcomes of first MTP arthrodesis using noncontoured, nonlocking stainless steel plates and precontoured, locking plates overall were equivalent. The cost of the precontoured locking plate construct with screws for plate fixation and 1 lag screw is approximately 10- to 31-fold higher than the cost of a noncontoured 1/3 tubular plate at our institution. Based
on this information, the routine use of precontoured plates may not be justified, especially in patients without an IA. There was a trend toward higher union rate and shorter time to union in patients with IAs when precontoured locking plates were used instead of nonlocking plates, and perhaps the selected use of this construct is advisable in this situation. 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. Bennett GL, Kay DB, Sabetta J. First metatarsophalangeal joint arthrodesis: an evaluation of hardware failure. Foot Ankle Int. 2005;26:593-596. 2. Bennett GL, Sabetta J. First metatarsophalangeal joint arthrodesis: evaluation of plate and screw fixation. Foot Ankle Int. 2009.30:752-757. 3. Buranosky DJ, Taylor DT, Sage RA, et al. First metatarsophalangeal joint arthrodesis: quantitative mechanical testing of six hole dorsal plate versus crossed screw fixation in cadaveric specimens. J Foot Ankle Surg. 2001;40:208-213. 4. Coughlin MJ, Grebing BR, Jones CP. Arthrodesis of the first metatarsophalangeal joint for idiopathic hallux valgus: intermediate results. Foot Ankle Int. 2006;26:783-792. 5. Ellington JK, Jones CP, Cohen BE, Davis WH, Nickisch F, Anderson RB. Review of 107 hallux MTP joint arthrodesis using dome-shaped reamers and a stainless-steel dorsal plate. Foot Ankle Int. 2010;31:385-390. 6. Flavin R, Stephens MM. Arthrodesis of the first metatarsophalangeal joint using and dorsal titanium contoured plate. Foot Ankle Int. 2004;25:783-787. 7. Goucher NR, Coughlin MJ. Hallux metatarsophalangeal joint arthrodesis using dome-shaped reamers and dorsal plate fixation: a prospective study. Foot Ankle Int. 2006:27:869-876. 8. Griewe RM, Archdeacon MT. Locking plate technology: current concepts. J Knee Surg.2007;20:50-55. 9. Hunt KJ, Ellington JK, Anderson RB, Cohen BE, Davis WH, Jones CP. Locked versus nonlocked plate fixation for hallux MTP arthrodesis. Foot Ankle Int. 2011;32:704-709. 10. Hunt KJ, Barr CR, Lindsey DP, Chou LB. Locked versus nonlocked plate fixation for first metatarsophalangeal arthrodesis: a biomechanical investigation. Foot Ankle Int. 2012;33:984-990. 11. Hyer CF, Glover LP, Berlet GC, Lee TH. Cost comparison of crossed screws versus dorsal plate construct for first metatarsophalangeal joint arthrodesis. J Foot Ankle Surg. 2008;47:13-18. 12. Kumar S, Pradhan R, Rosenfeld PF. First metatarsophalangeal arthrodesis using a dorsal plate and compression screw. Foot Ankle Int. 2010;31:797-801.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
444
Foot & Ankle International 35(5)
13. Leaseburg JT, Deorio JK, Shapiro SA. Radiographic correlation of hallux MP fusion position and plate angle. Foot Ankle Int. 2009;30:873-876. 14. Molloy S, Burkhart BG, Jasper LE, Solan MC, Campbell JT, Belkoff SM. Biomechanical comparison of two fixation methods for first metatarsophalangeal joint arthrodesis. Foot Ankle Int. 2003;24:169-171. 15. Neufeld SK, Parks BG, Naseef GS, Melamed EA, Schon LC. Arthrodesis of the first metatarsophalangeal joint: a biomechanical study comparing memory compression staples,
cannulated screws and a dorsal plate. Foot Ankle Int. 2002;23:97-101. 16. Politi J, Hayes J, Njus G, Bennett GL, Kay DB. First metatarsal-phalangeal joint arthrodesis: a biomechanical assessment of stability. Foot Ankle Int. 2003;24:332-337. 17. Stevens BW, Anderson JG, Bohay DR. Hallux metatarsophalangeal joint fusion for the rheumatoid foot. Foot Ankle Clin. 2007;12:395-404. 18. Zura RD, Browne JA. Current concepts in locked plating. J Surg Orthop Adv. 2006;15:173-176.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
research-article2014
FAIXXX10.1177/1071100714520695Foot & Ankle InternationalMayer et al
Article
A Comparison of Nonlocking Semitubular Plates and Precontoured Locking Plates for First Metatarsophalangeal Joint Arthrodesis
Foot & Ankle International® 2014, Vol. 35(5) 438–444 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100714520695 fai.sagepub.com
Stephanie A. Mayer, MD1, Nicole A. Zelenski, BA2, James K. DeOrio, MD1, Mark E. Easley, MD1, and James A. Nunley II, MD1 Abstract Background: First metatarsophalangeal (MTP) joint arthrodesis is a successful, commonly performed procedure for many conditions affecting this joint. The purpose of this retrospective study was to compare the clinical and radiographic outcomes between patients who had primary first MTP joint fusions with either noncontoured stainless steel semitubular plates or precontoured plates. Our hypothesis was that there would be no difference in clinical or radiographic outcomes between these groups. Methods: A search based on operative report coding was performed to detect all patients who had undergone primary first MTP joint arthrodesis with a dorsal plating construct at our institution from 2005 to 2010. A retrospective review of electronic medical records and postoperative radiographs was performed to determine time to clinical and radiographic union, pain scores, complications, MTP angle, and proximal phalanx to floor angle. A Kruskal-Wallis test was used for continuous variables, and a chi-square or Fisher exact test was used for categorical variables to determine statistical significance between the 2 groups. Results: We identified 128 patients who met our inclusion criteria. One hundred and two feet in 97 patients underwent arthrodesis with a noncontoured, stainless steel 1/3 tubular plate (group 1) and 26 feet in 21 patients with a precontoured, locking plate (group 2). The patients presented with a variety of inflammatory and noninflammatory preoperative diagnoses. The overall union rate, complication rate, time to radiographic healing, MTP angle, and patient-reported visual analog scale (VAS) score were similar between the 2 groups. Clinical time to healing was more rapid in the noncontoured group (3.7 months vs 4.8 degrees in precontoured) (P = .02). The radiographically measured proximal phalanx to floor angle was significantly decreased in noncontoured patients (2.6 degrees vs 4.6 degrees in precontoured, P = .04). In patients who had a diagnosis of an inflammatory arthropathy, the union rate fell to 84.3% (27/32) in the noncontoured group and 87.5% (7/8) in the precontoured group, and time to radiographic as well as clinical union increased in both groups. When patients with inflammatory arthropathies were compared with noninflammatory patients within groups, it was found that the time to both clinical and radiographic union was significantly longer in the noncontoured group for those with an inflammatory arthropathy (P = .0052, P = .022). For the precontoured group, these values were not statistically significant (P = .089, P = .43). Additionally, the cost of the noncontoured implants was significantly less than that of the precontoured implants. Conclusion: In this study, clinical time to healing of primary first MTP fusion with dorsal plate and screw constructs was improved with the use of noncontoured plates. However, other clinical indicators such as pain, complication rate, MTP angle, and VAS score did not differ between groups. Noncontoured plates are less expensive, and thus the decision to use precontoured plates routinely should be made cautiously. However, in the presence of inflammatory arthropathy, the use of precontoured plates may be justified given that the use of these plates in the inflammatory group resulted in improved radiographic and clinical union. Level of Evidence: Level III, comparative series. Keywords: hallux disorders, arthritis, outcome studies First metatarsophalangeal (MTP) joint arthrodesis is commonly performed for a variety of conditions affecting the great toe, including hallux valgus, hallux rigidus, hallux varus, neuromuscular conditions, failed arthrodesis, and inflammatory arthropathies (IAs).1,2,4-7,9,12,15,16 Several techniques for preparation of the bony surfaces for arthrodesis have been described; however, the most commonly accepted
1
Department of Orthopaedic Surgery, Duke University Medical Center, Durham, NC, USA 2 Duke University Medical Center, Durham, NC, USA Corresponding Author: Nicole A. Zelenski, BA, Duke University Medical Center, Department of Orthopaedic Surgery, 200 Trent Drive, Durham, NC 27710, USA. Email: [email protected]
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
439
Mayer et al technique currently is the cup and cone reaming technique.1,2,5,7,9,12 Additionally, many fixation constructs have been used, such as Kirschner wires, tension band wires, crossed screws, and dorsal plates with or without a lag screw.1-7,9-12,14-16 Dorsal plate and screw constructs have proven to be biomechanically superior to other fixation methods and therefore are commonly used.3,10,11,15,16 Both nonlocking stainless steel and titanium plates are available,1,4 and with the advances in locking plate technology, precontoured stainless steel and titanium locking plates are now available for MTP fusion.2,5-7,9,10,12 Union rates using dorsal plate and screw constructs vary from 77.2% to 100%.1,2,4-7,9,12 Clinical outcomes following this procedure are generally good to excellent. The purpose of this study was to compare the radiographic and clinical outcomes of 2 groups of patients who underwent primary first MTP joint arthrodesis with a dorsal plate and screw construct with either a precontoured locking plate or a noncontoured, nonlocking,1/3 tubular stainless steel plate. Given the large discrepancy in pricing between these 2 implants, we also wanted to make comparisons based on cost. Our hypothesis was that there would be no difference in radiographic or clinical outcomes between the groups and that the difference in cost would make the noncontoured plates more attractive.
Methods After institutional review board approval was granted, a search was performed based on operative report Current Procedural Terminology (CPT) codes for all patients who underwent first MTP joint arthrodesis by 1 of 3 orthopedic foot and ankle surgeons between January 2006 and May 2011. Patients were included if they were over 18 years old, had undergone primary MTP arthrodesis with dorsal plating techniques, and had follow-up until radiographic union occurred or a decision for revision surgery was made. Patients who underwent revision arthrodesis were excluded; however, primary arthrodesis as a salvage procedure following other operations including failed bunion correction, hallux varus correction, cheilectomy, and soft tissue arthroplasty was included. A retrospective review of electronic medical records and radiographs was performed to record diagnosis, previous procedures, demographic data, visual analog scale (VAS) scores, time to clinical healing, time to radiographic healing, final MTP angle, final proximal phalanx to floor angle, complications, and revision surgeries. We identified 128 feet in 118 patients that met our inclusion criteria: 102 feet in 97 patients underwent arthrodesis with a noncontoured, stainless steel 1/3 tubular plate (group 1) and 26 feet in 21 patients with a precontoured, locking plate (group 2). There was no notable difference in mean age between the groups—60.5 years in the noncontoured group and 61 years in the precontoured group. The patients
Table 1. Preoperative Diagnoses. Diagnosis Degenerative arthritis Rheumatoid arthritis Hallux varus Gout Neuromuscular deformity Wegener’s granulomatosis Psoriatic arthritis Scleroderma Postpolio deformity Salvage after failed arthroplasty
Noncontoured Precontoured 61 27 4 1 1 1 2 1 0 4
13 8 1 0 0 0 0 0 2 2
Table 2. Use of Lag Screws in Fixation Construct. Precontoured Noncontoured (n = 26) (n = 102) Total patients with lag screws Lag screw in IA group Lag screws in non-IA group Nonunions without lag screws
8 2/8 6/18 2/2
102 32/32 70/70 0/8
Abbreviations: IA, inflammatory arthropathy; non-IA, noninflammatory arthropathy.
presented with a variety of preoperative diagnoses, including hallux valgus, hallux rigidus, hallux varus, rheumatoid arthritis, gout, psoriatic arthritis, failed arthroplasty, Wegener’s granulomatosis, and neuromuscular hallux valgus. A summary of demographic data and preoperative diagnoses is presented in Table 1. There were 32 feet (30.4%) in the noncontoured group and 8 feet (30.7%) in the precontoured group with a diagnosis of an IA (rheumatoid arthritis, gout, or psoriatic arthritis). Lag screws were used in all patients with noncontoured plate and in 8 of the 26 patients (31%) with contoured plate. Of the lag screws used in the contoured group, there was little difference in the incidence used in patients with IAs and in those with noninflammatory arthropathies (non-IAs) (25% and 33%, respectively) (Table 2). Time to clinical healing was recorded by the attending surgeon as the time point when the patient had minimal pain, there was no motion through the fusion site clinically, wounds were healed appropriately, and the patient was deemed appropriate for full weight-bearing in regular shoe wear. Time to radiographic union was evaluated by plain films and was recorded by the attending physician when there was bony trabeculation crossing the fusion site. MTP joint and proximal phalanx to floor angles were measured by a single observer. On the standing lateral radiographs taken at the final follow-up visit, 2 measurements were taken for each angle type and the mean was used for statistical analysis. MTP joint angles were measured using the Cobb angle function on an electronic radiology system (Centricity Web
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
440
Foot & Ankle International 35(5)
Figure 1. (A) Lateral and (B) anteroposterior radiograph of a noncontoured, nonlocking 1/3 tubular plate demonstrating metatarsophalangeal angle (A: the angle between X and Y) and proximal phalanx to floor angle (A: the angle between Y and Z). (C) Lateral and (D) anteroposterior radiograph of a precontoured, locking plate.
Enterprises, GE Healthcare, Waukesha, WI) defined as the angle between a line parallel to the shaft of the metatarsal and a line parallel to the shaft of the proximal phalanx (Figure 1A). The proximal phalanx to floor angle was also measured with the Cobb angle function and was defined as the angle between a line parallel to the shaft of the proximal phalanx and a line parallel to the floor (Figure 1A). Data were collected at 6-week, 3-month, 6-month, and 12-month visits. No patients in this study had follow-up longer than 12 months because the determination of union or nonunion, the need for revision due to symptomatic nonunion, and any postoperative complications were recorded by this time point. Statistical analysis was performed using Student t test or Kruskal Wallis test (if nonparametric) for continuous variables and a chi-square or Fisher exact test (if values were less than 5) for categorical variables to determine significance between the 2 groups. A P value of .05 was used for statistical significance. The operative technique used by each surgeon was similar. Patients underwent either general anesthesia with or without a regional block or regional block with sedation. A dorsal
or dorsomedial skin incision was made. All surgeons used a cannulated cup and cone reamer system. The toe was then positioned both clinically and radiologically such that the tip of the toe rested on a flat metal surface and there was approximately 0 to 8 degrees of valgus and 20 to 30 degrees of dorsiflexion at the MTP joint level. The joint was provisionally pinned either with a retrograde K-wire or with a plantarmedial to dorsal-lateral K-wire. If a precontoured locking plate was chosen, the plate was applied and provisionally fixed to the bone. Fluoroscopy was then used to determine appropriate plate placement, and if needed the decision was made to change the bend in the plate to match the patient’s anatomy. If a noncontoured 1/3 tubular plate was chosen, the plate was also provisionally secured with K-wires and evaluated for placement and the appropriate bend needed to match the patient’s anatomy (Figure 1). The decision for the type and size of plate, and for any additional lesser toe procedures performed concomitantly, was made at the discretion of the surgeon. Postoperatively, the patients were kept non-weightbearing initially and allowed to begin progressive protected weight-bearing usually between 3 to 6 weeks.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
441
Mayer et al
Table 3. Comparison of Union Rate, Time to Clinical and Radiographic Union, and Radiographic Measurements Between Groups.
Union rate Time to clinical union Time to radiographic union Metatarsophalangeal angle Proximal phalanx to floor angle
Noncontoured
Precontoured
P Value
92.2% 3.7 months 4.0 months 24.2 degrees 2.6 degrees
92.3% 4.6 months 4.3 months 26.0 degrees 4.6 degrees
.02 .37 .09 .04
Table 4. Total Complications in Noncontoured and Precontoured Groups. Noncontoureda
Precontouredb
8 2 3 1 1 1 4 3 1 1
2 1 0 2 0 1 0 1 0 0
Nonunion Malunion Superficial infection Deep infection Delayed wound healing Wound dehiscence Hardware failure Painful hardware Painful metatarsophalangeal osteophyte Interphalangeal joint contracture a
26 complications in 19 feet in 18 patients. 7 complications in 4 feet in 4 patients.
b
Results The overall union rate was 92.2% (94/102) for the noncontoured group and 92.3% (24/26) for the precontoured group. The difference in the mean time to radiographic union between the noncontoured group (4.0 months) and precontoured group (4.3 months) did not reach statistical significance (P = .37). The difference in mean clinical healing time between the noncontoured group (3.7 months) and precontoured group (4.6 months) reached statistical significance (P = .02). There were significantly more patients clinically healed at the 3-month follow-up in the noncontoured group (74.5% vs 50%) (P = .02). Otherwise, the proportion of patients in each group who were deemed clinically or radiographically healed was not significantly different at any of the other time points measured. There was no notable difference in the patient-reported VAS score between groups at any time point. VAS scores in both groups improved at each time point. At 6 weeks, 15.7% (16/102) noncontoured versus 3.9% (1/26) precontoured were clinically healed (P = 0.19), and 12.8% (13/102) noncontoured versus 0% precontoured were radiographically healed (P = .06). At 3 months, 74.5% (76/102) noncontoured versus 50% (13/26) precontoured were clinically healed (P = .02), and 66.7% (68/102) noncontoured versus 61.5% (16/26) precontoured were radiographically healed. The mean radiographically measured MTP angle was 24.2 degrees (range, 6.0-43.7 degrees) for the noncontoured
group and 26.0 degrees (range, 14.3-46.1 degrees) for the precontoured group, which did not reach statistical significance (P = .09). The mean radiographically measured proximal phalanx to floor angle was 2.6 degrees (range, –14.1 to 30.0 degrees) for the noncontoured group and 4.6 degrees (range, –10.3 to 19.7 degrees) for the precontoured group (P = .04) (Table 3). There were 25 complications in 18 feet (24.3%) in the noncontoured group and 7 complications in 4 feet (26.9%) in the precontoured group. There were 8 nonunions in the noncontoured group (7.8%) and 2 in the precontoured group (7.7%), both of which also had deep infections. Of the 2 nonunions in the precontoured group, neither had a lag screw. Other complications reported included superficial and deep infections, delayed wound healing, wound dehiscence, hardware failure and breakage, malunion, painful hardware, persistent painful osteophyte, and interphalangeal joint contracture (Table 4). There were 9 reoperations in the noncontoured group: 3 for nonunion revision; 2 for malunion; 1 each for irrigation and debridement and hardware removal for infection, excision of persistent painful osteophytes, and removal of painful hardware; and 1 for removal of painful hardware and release of an interphalangeal joint contracture in the same patient. Reoperations in the precontoured group included irrigation and debridement and removal of hardware in 2 patients who went on to nonunion and 1 revision due to malunion and painful hardware.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
442
Foot & Ankle International 35(5)
In patients who had a diagnosis of an IA, the union rate fell to 84.3% (27/32) in the noncontoured group and 87.5% (7/8) in the precontoured group. Time to radiographic union increased to 4.8 months in the noncontoured group and 4.3 months in the precontoured group, while time to clinical union rose to 4.8 months in the noncontoured group but fell to 3.4 months in the precontoured group (P = .18, P = .32). Thirteen of 25 overall complications, including 5 of the 8 nonunions and all 4 infections in the noncontoured group, were in patients with an IA. Similarly, 3 of the 7 total complications were in this patient population. Excluding the patients with IAs from the analysis, the union rate improved to 95.7% in the noncontoured group and 94.4% in the precontoured group. Complications occurred in 8 of 71 patients in the noncontoured group and 2 of 24 patients in the precontoured group. Time to radiographic union was not statistically significant between groups. It improved to 3.7 months in the noncontoured group and 4.2 in the precontoured group (P = .28). Time to clinical union was significantly shorter at 3.3 months in the noncontoured group versus 5.0 months in the precontoured group (P = .0009). Comparing the patients who had IAs to those who did not within groups, the time to both clinical and radiographic union was significantly shorter in the noncontoured group for those without an IA (P = .0052, P = .022). For the precontoured group, these values were not statistically significant (P = .089, P = .43).
Discussion Arthrodesis of the first MTP joint is a common procedure, and as the population ages, the number of procedures per year will likely increase. As such, it would be ideal for the foot and ankle surgeon to have a reliable construct for fixation that is also affordable for the health care system. We sought to determine any clinical or radiographic difference in outcomes between the use of noncontoured, nonlocking, stainless steel 1/3 tubular plates and precontoured, locking plates for fusion. The nonlocking plates are more affordable and therefore would be more attractive if the outcomes are similar. Previous reports of MTP fusion in recent literature have cited between 77% and 100% union rates using dorsal plating techniques.1,2,4-7,9,12 A variety of dorsal plates exist, with options for noncontoured or precontoured, stainless steel or titanium, and locking or nonlocking. Several authors have compared dorsal plating constructs with or without a lag screw to other fixation methods such as crossed screws for their biomechanical properties and have found the dorsal plating technique to be stiffer.3,10,11,15,16 However, at least 1 article reports on the comparison between crossed screws and dorsal plates showing an equivalent union rate and clinical outcome.11 Because of the majority of the biomechanical
data, dorsal plating is one of the most commonly used fixation method currently. Locking plates have been used extensively for periarticular fractures, osteoporotic bone, severe comminution, and unstable metadiaphyseal fractures.8,18 The use of locking plates for first MTP fusion has been advocated by some given the low profile of the plate, ease of use, theoretical time savings in the operating room since the need to bend the plate should be obviated, and superior biomechanical properties in the above situations.2,5-7,9 Despite these potential advantages, most studies have shown that complications still occur along with a similar rate of nonunion.2,5-7,9,12 A biomechanical analysis comparing precontoured stainless steel locked and nonlocked plates for MTP fusion in a cadaver model found that precontoured locked plates are significantly stiffer and had less plantar gapping in loading tests.10 Interestingly, a recent article comparing noncontoured and precontoured locking plates found a higher rate of nonunion with locking plates, although this did not reach statistical significance.9 The authors hypothesized that the stiffness of the locking construct was not ideal for healing of the arthrodesis site since compression across the fusion site cannot occur.9 Our study found overall equivalent union rates for both type of plate. Our union rates are similar to those of other published studies for MTP fusion using dorsal plates of both types.1,2,4-7,9,12 We also found that the time to both clinical and radiographic union and the proportion of patients who were deemed clinically healed at each time point were similar between groups at all but 1 time point. Patientreported pain improved over time, and scores were comparable between the 2 constructs. The radiographic measurements for MTP angle and proximal phalanx to floor angle have been shown to correlate with the MTP angle built into precontoured plates as well as a good clinical outcome for MTP fusion.2,13 The mean MTP angle in this study was within the ideal range of 15 to 30 degrees2,5,13 for both groups and was not statistically different between groups. The proximal phalanx to floor angle was statistically different at 2.6 degrees for the noncontoured group and 4.6 degrees for the precontoured group; however, the clinical significance of this as far as functional results or patient satisfaction is unclear. There were no revisions for malunion involving a toe that did not touch the ground in either group. One potential advantage of the precontoured plates is the MTP angle built into the plate, which should decrease the need for decision making intraoperatively about the appropriate amount of bending of the plate. Despite this, the final measurements between the MTP angle and the proximal phalanx to floor angles were similar in both groups. Approximately half of the complications in each group were in patients with a diagnosis of an IA. Other reports of MTP fusion have also had a higher rate of complications in
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
443
Mayer et al these patients.5,9,17 Of the patients with this diagnosis, the rate of complications was similar between groups but was higher than the rate of complications for patients without this diagnosis. Both nonunions in the precontoured group occurred in the setting of deep infection requiring removal of hardware, and a larger prospective study would further elucidate whether the proportion of nonunions in this group without infection would be significantly different than that in the noncontoured group. Additionally, both of the nonunions in the precontoured group lacked lag screws, which add stability to the construct. In our study, both the union rate and complication rate were equivalent between groups overall and when we compared patients with and without IAs separately. Taking into account a diagnosis of an IA when comparing time to clinical and radiographic union, there are interesting findings. Regarding clinical healing times, non-IA patients healed significantly faster in the noncontoured versus precontoured group. Unexpectedly, the IA patients had a 6-week shorter time to clinical healing than non-IA patients when using the precontoured plate; however, this did not reach statistical significance. Within the noncontoured group, the non-IA patients were clinically healed significantly faster. Regarding radiographic healing times, within the noncontoured group, the non-IA patients were healed significantly faster. In the precontoured group there was no difference between IA and non-IA patients. There was a small trend toward faster radiographic healing in IA patients in the precontoured group and a stronger trend toward faster healing in the non-IA patients in the noncontoured group. These results point to a possible advantage of noncontoured plates in non-IA patients and precontoured plates in patients with an IA. Limitations of this study include the inherent bias of a retrospective cohort study, small sample size, limited patient satisfaction outcome scores, and heterogeneous follow-up assessment and clinical judgment between 3 surgeons. There is a paucity of data for this type of precontoured locked plate for MTP arthrodesis in the literature, and therefore there is limited comparison for our outcome data. Several variables differed between the plate types, including locking versus nonlocking and the number of screws for plate fixation and lag screws. In addition, the retrospective nature of this study means that the addition of a lag screw or K-wire for fixation or prolonged protected weight-bearing was at the discretion of the surgeon. In conclusion, the clinical and radiographic outcomes of first MTP arthrodesis using noncontoured, nonlocking stainless steel plates and precontoured, locking plates overall were equivalent. The cost of the precontoured locking plate construct with screws for plate fixation and 1 lag screw is approximately 10- to 31-fold higher than the cost of a noncontoured 1/3 tubular plate at our institution. Based
on this information, the routine use of precontoured plates may not be justified, especially in patients without an IA. There was a trend toward higher union rate and shorter time to union in patients with IAs when precontoured locking plates were used instead of nonlocking plates, and perhaps the selected use of this construct is advisable in this situation. 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. Bennett GL, Kay DB, Sabetta J. First metatarsophalangeal joint arthrodesis: an evaluation of hardware failure. Foot Ankle Int. 2005;26:593-596. 2. Bennett GL, Sabetta J. First metatarsophalangeal joint arthrodesis: evaluation of plate and screw fixation. Foot Ankle Int. 2009.30:752-757. 3. Buranosky DJ, Taylor DT, Sage RA, et al. First metatarsophalangeal joint arthrodesis: quantitative mechanical testing of six hole dorsal plate versus crossed screw fixation in cadaveric specimens. J Foot Ankle Surg. 2001;40:208-213. 4. Coughlin MJ, Grebing BR, Jones CP. Arthrodesis of the first metatarsophalangeal joint for idiopathic hallux valgus: intermediate results. Foot Ankle Int. 2006;26:783-792. 5. Ellington JK, Jones CP, Cohen BE, Davis WH, Nickisch F, Anderson RB. Review of 107 hallux MTP joint arthrodesis using dome-shaped reamers and a stainless-steel dorsal plate. Foot Ankle Int. 2010;31:385-390. 6. Flavin R, Stephens MM. Arthrodesis of the first metatarsophalangeal joint using and dorsal titanium contoured plate. Foot Ankle Int. 2004;25:783-787. 7. Goucher NR, Coughlin MJ. Hallux metatarsophalangeal joint arthrodesis using dome-shaped reamers and dorsal plate fixation: a prospective study. Foot Ankle Int. 2006:27:869-876. 8. Griewe RM, Archdeacon MT. Locking plate technology: current concepts. J Knee Surg.2007;20:50-55. 9. Hunt KJ, Ellington JK, Anderson RB, Cohen BE, Davis WH, Jones CP. Locked versus nonlocked plate fixation for hallux MTP arthrodesis. Foot Ankle Int. 2011;32:704-709. 10. Hunt KJ, Barr CR, Lindsey DP, Chou LB. Locked versus nonlocked plate fixation for first metatarsophalangeal arthrodesis: a biomechanical investigation. Foot Ankle Int. 2012;33:984-990. 11. Hyer CF, Glover LP, Berlet GC, Lee TH. Cost comparison of crossed screws versus dorsal plate construct for first metatarsophalangeal joint arthrodesis. J Foot Ankle Surg. 2008;47:13-18. 12. Kumar S, Pradhan R, Rosenfeld PF. First metatarsophalangeal arthrodesis using a dorsal plate and compression screw. Foot Ankle Int. 2010;31:797-801.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
444
Foot & Ankle International 35(5)
13. Leaseburg JT, Deorio JK, Shapiro SA. Radiographic correlation of hallux MP fusion position and plate angle. Foot Ankle Int. 2009;30:873-876. 14. Molloy S, Burkhart BG, Jasper LE, Solan MC, Campbell JT, Belkoff SM. Biomechanical comparison of two fixation methods for first metatarsophalangeal joint arthrodesis. Foot Ankle Int. 2003;24:169-171. 15. Neufeld SK, Parks BG, Naseef GS, Melamed EA, Schon LC. Arthrodesis of the first metatarsophalangeal joint: a biomechanical study comparing memory compression staples,
cannulated screws and a dorsal plate. Foot Ankle Int. 2002;23:97-101. 16. Politi J, Hayes J, Njus G, Bennett GL, Kay DB. First metatarsal-phalangeal joint arthrodesis: a biomechanical assessment of stability. Foot Ankle Int. 2003;24:332-337. 17. Stevens BW, Anderson JG, Bohay DR. Hallux metatarsophalangeal joint fusion for the rheumatoid foot. Foot Ankle Clin. 2007;12:395-404. 18. Zura RD, Browne JA. Current concepts in locked plating. J Surg Orthop Adv. 2006;15:173-176.
Downloaded from fai.sagepub.com at UNIV OF NORTH DAKOTA on June 2, 2015
