Orthopedics

Feature Article Free

Residual Hole Orientation After Plate Removal: Effect on the Clavicle

Jeremy James, MD; Allen Ogden, BSME; Debi Mukherjee, ScD, MBA; Todd Jaeblon, DO

Abstract

Clavicle fractures account for 2.6% to 4% of all fractures. Surgical stabilization of this type of injury is becoming more common. Anterior inferior plating and superior plating are 2 popular approaches to open reduction and internal fixation. Reports of plate removal have raised concerns about reinjury. The goal of the current study was to determine whether the orientation of screw holes in clavicles after removal of an anterior inferior plate vs a superior plate have different biomechanical effects on stiffness and load to failure. The medial and lateral ends of 28 matched pairs of fresh clavicles were potted. Pilot holes, 2.5 mm in diameter, were drilled and oriented anterior inferiorly or superiorly, simulating those left after removal of a plate for a middle-third fracture. The clavicles underwent dynamic axial compression and 3-point load to failure, replicating forces associated with reinjury. Clavicles with anterior inferior holes had a statistically significant higher median maximal load difference of 139 N compared with those with superior holes (P=.013). Anterior inferior holes showed a statistically significant median increase in stiffness of 16.3 N/mm compared with superior holes (P=.036). Clavicles with anterior inferior holes had a statistically significant increase in median maximal load to failure and an increase in median stiffness compared with those with superior holes. This finding is relevant for patients who undergo hardware removal and return to activities that put them at risk for repeat high-impact injuries to the clavicle. [Orthopedics. 2015; 38(11):e1034–e1039.]

Abstract

Clavicle fractures account for 2.6% to 4% of all fractures. Surgical stabilization of this type of injury is becoming more common. Anterior inferior plating and superior plating are 2 popular approaches to open reduction and internal fixation. Reports of plate removal have raised concerns about reinjury. The goal of the current study was to determine whether the orientation of screw holes in clavicles after removal of an anterior inferior plate vs a superior plate have different biomechanical effects on stiffness and load to failure. The medial and lateral ends of 28 matched pairs of fresh clavicles were potted. Pilot holes, 2.5 mm in diameter, were drilled and oriented anterior inferiorly or superiorly, simulating those left after removal of a plate for a middle-third fracture. The clavicles underwent dynamic axial compression and 3-point load to failure, replicating forces associated with reinjury. Clavicles with anterior inferior holes had a statistically significant higher median maximal load difference of 139 N compared with those with superior holes (P=.013). Anterior inferior holes showed a statistically significant median increase in stiffness of 16.3 N/mm compared with superior holes (P=.036). Clavicles with anterior inferior holes had a statistically significant increase in median maximal load to failure and an increase in median stiffness compared with those with superior holes. This finding is relevant for patients who undergo hardware removal and return to activities that put them at risk for repeat high-impact injuries to the clavicle. [Orthopedics. 2015; 38(11):e1034–e1039.]

Clavicle fractures account for 2.6% to 4% of all fractures1–3; 69% to 82% occur at the midshaft.4,5 Reports of favorable outcomes after operative fixation have led to increased enthusiasm for surgical stabilization.6,7 Anterior inferior plating and superior plating remain popular methods of fixation. Removal rates range from 6% to 74%,8,9 and refracture rates of as high as 17% after plate removal have been reported.10

To the authors' knowledge, the biomechanical effect of hole orientation after plate removal has not been studied. Midshaft clavicle fractures often occur in active people who, after healing, are likely to resume the same activity that led to the injury. It is not known whether plate removal increases the risk of reinjury. The primary goal of this study was to determine whether the orientation of anterior inferior or superior holes significantly affects clavicle failure during dynamic loads that simulate those of injury as well as overall stiffness.

Materials and Methods

This study included 28 matched pairs of clavicles from fresh cadavers with no visible evidence of trauma or previous surgery. The specimens in the axial load group had an average age of 79.1 years (median, 78 years; range, 64–97 years). There were 11 men and 3 women. Average age of the specimens in the 3-point bending group was 75.4 years (median, 76 years; range, 51–97 years). There were 10 men and 4 women. All muscle and ligament attachments were removed. The medial and lateral ends were potted to a depth of 2 cm with polyester resin (Bondo; 3M Corporation, St Paul, Minnesota).

To simulate anterior inferior plate fixation of a midshaft clavicle fracture, a 3.5-mm, 6-hole malleable reconstruction plate template with centrally machined holes was centered on the midline of the anterior inferior surface of the specimen. Pilot holes, 2.5 mm in diameter, were drilled in the center of each hole, with 3 holes medial and 3 holes lateral to the midline. A separate template was used on the contralateral clavicle to drill holes simulating placement at the superior plate. Each of the 14 pairs was randomly assigned to either axial compression or 3-point bending.

Axial and 3-point impact destructive tests were performed at 1.5 m/s, based on previous studies by Duprey et al.11 Both axial and 3-point tests were performed on an Instron 8874 biaxial testing frame (Instron Corp, Norwood, Massachusetts), with all tests programmed with WaveMatrix (Instron). For axial impact tests, each specimen was oriented vertically, with the embedded sternal extremity rigidly secured in a vice attached to the piston and load cell of the testing frame (Figure 1). A 10-mm shim was used to establish a gap between the acromial extremity and a rigid steel collision plate. Using Instron WaveMatrix software, the piston of the testing frame was programmed to translate 20 mm at a rate of 1.5 m/s, allowing 10 mm free translation before impact against the platen and 10 mm dynamic compression. For 3-point impact, each specimen was positioned horizontally to span 2 steel columns and oriented such that a steel punch would impact its central midshaft superior surface (Figure 2). For each clavicle, the span length in millimeters was measured and recorded. Similar to axial tests, a 10-mm gap was established between the punch and the specimen. The piston was programmed to translate 20 mm at a rate of 1.5 m/s, allowing 10 mm free translation before contact with the clavicle.

The axial loading model.

Figure 1:

The axial loading model.

The 3-point bending model.

Figure 2:

The 3-point bending model.

At the conclusion of each test, the raw data files were analyzed with Excel software (Microsoft Corporation, Redmond, Washington). The maximum load, expressed in Newtons (N), for each sample was defined as the transition from a linear to a nonlinear response within a load-displacement curve generated during dynamic testing. Stiffness values, expressed in Newtons per millimeter (N/mm), were determined by graphing a scatter plot of the load vs displacement data up to the maximum load value. Within Excel, a linear trend line was generated for that data set, resulting in the sample's stiffness value. For 3-point tests, the span length was incorporated into the stiffness calculation, generating a moment expressed in Newton-meter per millimeter (N-m/mm).

Statistical Analysis

The primary goal of statistical analysis was to determine the effect of hole location (anterior inferior vs superior) on maximum load (Newtons). The secondary objective was to determine whether hole location also had an effect on stiffness (Newtons per millimeter).

To evaluate the effect of hole location, within-cadaver differences between the anterior inferior location and the superior location for maximum load and stiffness were computed for each of the 28 cadaveric pairs of clavicles. Graphic examination of within-cadaver differences showed that they were not normally distributed and were not symmetric (ie, the distribution was skewed). Thus, a nonparametric sign test was used to evaluate the effect of hole location.12 The null hypothesis (H0) for these sign tests was that the median within-cadaver difference (calculated as anterior inferior minus superior) for maximum load or stiffness was zero. In addition to the hypothesis tests, nonparametric 95% confidence intervals for the median within-cadaver difference were calculated.13

The consistency of the hole location effects for maximum load and stiffness across the 2 loading groups (axial vs 3-point bending) was evaluated with a nonparametric Wilcoxon rank-sum test. The null hypothesis (H0) for this procedure was that the median location effect (anterior inferior minus superior) for the 14 cadaver pairs in the axial loading group was equivalent to the median location effect for the 14 cadaver pairs in the 3-point bending group.

All statistical hypothesis tests and confidence intervals were 2-sided and were performed at .05 significance. Data analysis was performed with SAS for Windows version 9 software (Microsoft Corporation).14 All calculations were performed on an Intel Centrino (Intel Corporation, Santa Clara, California) dual-processor personal computer.

Results

For both maximum load and stiffness, the null hypothesis was rejected. Anterior inferior vs superior holes had a statistically significant difference in median load to failure and stiffness. The median difference in maximum load was 139 N higher (P=.013) and the median difference in stiffness was 16.3 N/mm higher (P=.036) in the anterior inferior group.

The second null hypothesis was accepted. The effects of hole location on maximum load and stiffness across the 2 loading groups (axial vs 3-point bending) were consistent. The median location effect (anterior inferior minus superior) for the 14 cadaver groups in the axial group was equivalent to the median location effect for the 14 cadaver pairs in the 3-point bending group (maximum load, P=.733; stiffness, P=.838) (Tables 13).

Axial Loading Efficacy Data

Table 1:

Axial Loading Efficacy Data

Three-Point Bending Efficacy Data

Table 2:

Three-Point Bending Efficacy Data

Statistical Analysis of Hole Location Effect (Anterior Inferior Minus Superior)

Table 3:

Statistical Analysis of Hole Location Effect (Anterior Inferior Minus Superior)

Discussion

Reports of unsatisfactory functional outcomes after nonoperative treatment of severely displaced clavicular fractures have led to increased enthusiasm for operative stabilization.7,6,15 Anterior inferior plating and superior plating have emerged as customary forms of internal fixation for middle-third clavicle fractures, and both have purported advantages and disadvantages. Plate removal has been reported because of hardware irritation, infection, and patient request.6,16 Reported rates for anterior inferior plate removal have ranged from 6% to 38%,8,17 whereas those for superior plates have been reported as 11% to 74%.1,9 The Canadian Orthopaedic Trauma Society6 reported removal of 8 superior plates as a result of irritation or infection and cited hardware removal as a common reason for repeat intervention. Ferran et al18 performed plate removal in the active population to avoid the need to treat a more complicated periprosthetic refracture. Active patients who require implant removal frequently return to activities that may lead to reinjury. In addition, reinjury may occur through a new or former fracture site or through a residual screw hole. Refracture rates of as high as 17% have been reported after plate removal.10 Poigenfürst et al19 reported 4 refractures after plate removal, 3 that occurred through residual holes and 1 that occurred during removal of an interfragmentary screw. Chen et al9 reported 1 refracture that occurred through a residual hole.

The effect of hole orientation on reinjury may favor 1 form of fixation or 1 method of plating over another, particularly if plate removal and resumption of activity are anticipated. After fracture remodeling and plate removal in long bones, diminished load to failure becomes more a property of the residual holes and less attributable to the original fracture site, but it is affected by implant and hole size.20 The geometric complexity of the clavicle makes the effect of screw holes and their orientation less predictable. To the authors' knowledge, no previous study has explored the effect of screw hole orientation on clavicle failure after plate removal.

Dynamic loads responsible for clavicular fractures frequently occur as a result of axial loading and 3-point bending, forming the basis for the authors' test model.11,21–23 Using intact clavicles with characteristics similar to those in the current study, Duprey et al11 reported an average load to failure in axial compression of 1480±460 N. In the current study, clavicles with anterior inferior holes failed in axial compression at an average of 2175±598 N, and those with superior holes failed at 2080±411 N. Only 2 clavicles in the anterior inferior group and 1 in the superior group failed at less than the average value reported by Duprey et al.11

The current study showed that, after removal of both anterior inferior and superior plates, screw hole orientation played a significant role in both the maximal load that the clavicle could hold before failure (P=.013) and stiffness (P=.036) of the clavicle. The forces endured after resumption of high-impact contact sports or re-exposure to high-energy injury would likely pose a more serious threat of refracture than those tested here. Anterior inferior plating should be considered over superior plating in these cases.

This study had several strengths. All specimens were fresh, so preservation did not decrease bone quality. The test model was consistent with those used in previous studies to evaluate clavicular failure in response to dynamic loads. The design minimized the effect of a healing or remodeled midshaft clavicle fracture to isolate the effects of remaining screw holes and their orientation after anterior inferior or superior plate removal.

Limitations

A limitation of this study was that the true effects of healing and remodeling of either the fracture or the area around the holes themselves cannot be predicted. An attempt was made to mimic the possibility of a direct blow to the superior clavicle with a 3-point bending model. However, a 4-point bending model would have distributed forces more evenly across the specimen, possibly leading to different results. The authors also recognize that failures occur more frequently as a result of axial loading, although it is not possible to account for multiple variations in load application. The density of bones evaluated in this study may not reflect the typical bone density of patients who undergo surgical stabilization. Bone densitometry was not performed on the study specimens. Younger patients may have an even higher load to failure than that found in the specimens used in this study, which likely were osteopenic.

The omission of tapping the holes could be considered a limitation, but its contribution at the time of screw removal is difficult to interpret. During implant removal procedures performed 3 to 6 months after surgery, small increases in endosteal and periosteal bone proliferation around the screws were noted as well as remodeling about the shaft.20,24 The prescribed 2.5-mm drill diameter exceeds the shaft diameter of a 3.5-mm screw by 0.01 mm. As a result of remodeling, at implant removal, the pilot hole actually may be smaller than the holes remaining in the specimens. The authors potted all clavicles to a depth of 2 cm to create a platform substantial enough to withstand testing while preserving the overall geometry of the specimen. This approach could have introduced some variability in the length proportion of the clavicles exposed to testing.

Limited resources did not allow the use of enough samples to achieve the desirable statistical power of 0.8. Post hoc statistical analysis of the differences between the anterior inferior and the superior hole locations showed that the anterior inferior location had a larger maximum load than the superior location in 21 (75%) of the 28 cadavers. Based on the nonparametric sign test, the statistical power to detect this hole location effect was 75%. Similarly, greater stiffness was observed in the anterior inferior location in 20 (71%) of the 28 cadavers. The statistical power to detect this hole location effect was 57%, based on the sign test. No attempt was made to compare the biomechanical differences in load to failure or stiffness of anterior inferior and superior plates.

Conclusion

Clavicles with anterior inferior holes had a statistically significant increase in median maximal load to failure and stiffness compared with superior holes. This finding is relevant after hardware removal for individuals returning to activities that put them at risk for repeat high-impact injuries to the clavicle.

References

  1. Nordqvist A, Petersson C. The incidence of fractures of the clavicle. Clin Orthop Relat Res. 1994; 300:127–132.
  2. Nowak J, Mallmin H, Larsson S. The aetiology and epidemiology of clavicular fractures: a prospective study during a two-year period in Uppsala, Sweden. Injury. 2000; 31(5):353–358. doi:10.1016/S0020-1383(99)00312-5 [CrossRef]
  3. Postacchini F, Gumina S, De Santis P, Albo F. Epidemiology of clavicle fractures. J Shoulder Elbow Surg. 2002; 11(5):452–456. doi:10.1067/mse.2002.126613 [CrossRef]
  4. Robinson CM. Fractures of the clavicle in the adult: epidemiology and classification. J Bone Joint Surg Br. 1998; 80(3):476–484. doi:10.1302/0301-620X.80B3.8079 [CrossRef]
  5. Rowe CR. An atlas of anatomy and treatment of midclavicular fractures. Clin Orthop Relat Res. 1968; 58:29–42. doi:10.1097/00003086-196805000-00006 [CrossRef]
  6. Canadian Orthopaedic Trauma Society. Non-operative treatment compared with plate fixation of displaced midshaft clavicular fractures: a multicenter, randomized clinical trial. J Bone Joint Surg Am. 2007; 89(1):1–10. doi:10.2106/JBJS.F.00020 [CrossRef]
  7. Hill JM, McGuire MH, Crosby LA. Closed treatment of displaced middle-third fractures of the clavicle gives poor results. J Bone Joint Surg Br. 1997; 79(4):537–539. doi:10.1302/0301-620X.79B4.7529 [CrossRef]
  8. Kloen P, Werner CM, Stufkens SA, Helfet DL. Anteroinferior plating of midshaft clavicular nonunions and fractures. Oper Orthop Traumatol. 2009; 21(2):170–179. doi:10.1007/s00064-009-1705-8 [CrossRef]
  9. Chen CH, Chen JC, Wang C, Tien YC, Chang JK, Hung SH. Semitubular plates for acutely displaced midclavicular fractures: a retrospective study of 111 patients followed for 2.5 to 6 years. J Orthop Trauma. 2008; 22(7):463–466. doi:10.1097/BOT.0b013e31817996fc [CrossRef]
  10. VanBeek C, Boselli KJ, Cadet ER, Ahmad CS, Levine WN. Precontoured plating of clavicle fractures: decreased hardware-related complications?Clin Orthop Relat Res. 2011; 469(12):3337–3343. doi:10.1007/s11999-011-1868-0 [CrossRef]
  11. Duprey S, Bruyere K, Verriest JP. Influence of geometrical personalization on the simulation of clavicle fractures. J Biomech. 2008; 41(1):200–207. doi:10.1016/j.jbiomech.2007.06.020 [CrossRef]
  12. Lehmann EL. Nonparametrics: Statistical Methods Based on Ranks. San Francisco, CA: Holden-Day; 1975.
  13. Hahn GJ, Meeker WQ. Statistical Intervals: A Guide for Practitioners. New York, NY: John Wiley and Sons; 1991. doi:10.1002/9780470316771 [CrossRef]
  14. SAS Institute Inc. SAS 9.1.3 ETL Studio: User's Guide. Cary, NC: SAS Institute Inc; 2004.
  15. Nowak J, Holgersson M, Larsson S. Sequelae from clavicular fractures are common: a prospective study of 222 patients. Acta Orthop. 2005; 76(4):496–502. doi:10.1080/17453670510041475 [CrossRef]
  16. Venkatachalam S, Packer G, Sivaji C, Shipton A. Plating of fresh displaced midshaft clavicular fractures. The Internet Journal of Orthopedic Surgery. 2006; 5(1):1–4.
  17. Chen CE, Juhn RJ, Ko JY. Anterior-inferior plating of middle-third fractures of the clavicle. Arch Orthop Trauma Surg. 2010; 130(4):507–511. doi:10.1007/s00402-009-0993-7 [CrossRef]
  18. Ferran NA, Hodgson P, Vannet N, Williams R, Evans RO. Locked intramedullary fixation vs plating for displaced and shortened mid-shaft clavicle fractures: a randomized clinical trial. J Shoulder Elbow Surg. 2010; 19(6):783–789. doi:10.1016/j.jse.2010.05.002 [CrossRef]
  19. Poigenfürst J, Rappold G, Fischer W. Plating of fresh clavicular fractures: results of 122 operations. Injury. 1992; 23(4):237–241. doi:10.1016/S0020-1383(05)80006-3 [CrossRef]
  20. Tepic S, Remiger AR, Morikawa K, Predieri M, Perren SM. Strength recovery in fractured sheep tibia with a plate or an internal fixator: an experimental study with a two-year follow-up. J Orthop Trauma. 1997; 11(1):14–23. doi:10.1097/00005131-199701000-00005 [CrossRef]
  21. Kemper AR, Stitzel JD, McNally C, Gabler HC, Duma SM. Biomechanical response of the human clavicle: the effects of loading direction on bending properties. J Appl Biomech. 2009; 25(2):165–174.
  22. Renfree T, Conrad B, Wright T. Biomechanical comparison of contemporary clavicle fixation devices. J Hand Surg Am. 2010; 35(4):639–644. doi:10.1016/j.jhsa.2009.12.012 [CrossRef]
  23. Stanley D, Trowbridge EA, Norris SH. The mechanism of clavicular fracture: a clinical and biomechanical analysis. J Bone Joint Surg Br. 1988; 70(3):461–464.
  24. Schatzker J, Sanderson R, Murnaghan JP. The holding power of orthopaedic screws in vivo. Clin Orthop Relat Res. 1975; 108:115–126. doi:10.1097/00003086-197505000-00019 [CrossRef]

Axial Loading Efficacy Data

Clavicle No.Maximum Axial Load, NLocation Effect (Anterior Inferior-Superior), NStiffness, N/mmLocation Effect (Anterior Inferior-Superior), N/mm


Anterior InferiorSuperiorAnterior InferiorSuperior
12652.842496.59156.25890.94311.94579.00
22352.462122.98229.48587.21581.006.21
31938.231950.51−12.28302.22367.06−64.84
41978.641819.46159.18417.14415.731.41
53118.602243.34875.26672.34636.9935.35
62942.122590.60351.53436.32351.1885.14
73056.132886.26169.87907.79356.54551.25
82234.171845.67388.49603.08414.69188.39
92300.252230.7869.46644.76684.82−40.06
101442.862124.06−681.20606.65590.0216.63
111986.151496.18489.97645.30311.70333.60
121545.402193.31−647.91394.04434.61−40.57
131376.281656.26−279.98376.22327.3548.87
141526.561461.7164.86245.60405.94−160.34

Three-Point Bending Efficacy Data

Clavicle No.Maximum 3-Point Load, NLocation Effect (Anterior Inferior-Superior), NStiffness, N/mmLocation Effect (Anterior Inferior-Superior), N/mm


Anterior InferiorSuperiorAnterior InferiorSuperior
1368.05234.9483133.1198.1780.68617.49
21015.32723.5229291.80241.52233.328.20
3878.51537.41341.09321.00188.98132.02
4409.68287.9028121.77128.21117.6510.56
5395.50374.45921.04144.53128.5815.95
6290.79631.7955−341.01115.33153.95−38.62
7718.52567.67150.85151.91166.04−14.13
8566.26453.50112.76170.03150.2419.79
9592.98683.53−90.55232.71152.4380.28
10700.04555.13144.91157.95163.32−5.37
111138.25780.92357.33400.11257.57142.54
12530.08506.8623.22179.66167.4112.25
131163.40638.38525.02240.74150.3390.41
14305.40432.93−127.5498.42132.96−34.55

Statistical Analysis of Hole Location Effect (Anterior Inferior Minus Superior)

VariableNo. of ClaviclesMean25th PercentileMedian75th PercentilePa95% Confidence Interval for Median
Maximum load28107.04.4139316.4.012523.2–229.5
Stiffness2870.6−9.716.387.8.03571.4–80.3
Authors

The authors are from the Department of Orthopaedic Surgery, Louisiana State University Health Sciences Center, Shreveport, Louisiana.

Dr James, Mr Ogden, and Dr Mukherjee have no relevant financial relationships to disclose. Dr Jaeblon is on the speaker's bureau of DePuy Synthes and Smith & Nephew and receives travel reimbursement from Smith & Nephew and Stryker, and his institution receives grants from DePuy Synthes.

Correspondence should be addressed to: Todd Jaeblon, DO, Department of Orthopaedic Surgery, Louisiana State University Health Sciences Center, 1501 Kings Hwy, Shreveport, LA 71130 ( tjaebl@lsuhsc.edu).

Received: November 22, 2014
Accepted: March 23, 2015

10.3928/01477447-20151020-13

Sign up to receive

Journal E-contents