Académique Documents
Professionnel Documents
Culture Documents
Background: There is no consensus regarding the optimal radiographic criteria for predicting the final healing of frac-
tures. The purpose of this study was to determine if the time to the radiographic appearance of cortical bridging predicted
the final healing of tibial shaft fractures, to examine the reliability of this assessment, and to determine when it is most
accurate during the postoperative period.
Methods: We retrospectively reviewed the data on 176 tibial fractures (OTA [Orthopaedic Trauma Association] 42-A, B,
and C) treated with intramedullary nailing at a level-I trauma center from 2007 through 2010. Postoperative radiographs
were assessed for varying degrees of cortical bridging, and interobserver reliability was calculated. Receiver operating
characteristic (ROC) curve and chi-square analyses determined the accuracy of cortical bridging assessments in pre-
dicting union.
Results: The nonunion rate was 7%. Any cortical bridging within four months was an excellent predictor of final healing
(accuracy = 99%, area under the curve [AUC] = 0.995, p < 0.0001) and was the most reliable criterion (kappa = 0.90). All
fractures that showed unicortical bridging eventually showed bridging of three cortices without additional intervention.
Conclusions: Assessment for any cortical bridging by four months postoperatively is a reliable, accurate predictor of
tibial shaft fracture-healing. This relatively early radiographic finding discriminates between fractures that will undergo late
union with observation alone and those destined for nonunion.
Level of Evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.
Peer Review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. It was also reviewed
by an expert in methodology and statistics. The Deputy Editor reviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication.
Final corrections and clarifications occurred during one or more exchanges between the author(s) and copyeditors.
A
ssessment of fracture-healing and the diagnosis of ever, the definition of nonunion has ranged from two to nine
nonunion are essential and at times difficult tasks in months without union in individual reports3-8.
the clinical management of tibial shaft fractures. Devel- Likewise, there is no consensus regarding the optimal
opment of nonunion is burdensome to the patient and addi- radiographic criteria to predict if a fracture will heal or at which
tional observation alone is unlikely to result in fracture union. times these criteria should be employed. Existing definitions of
Tibial nonunion is commonly defined as a failure of the frac- radiographic union vary widely, precluding direct comparison
ture to unite after six to eight months of observation1,2. How- among studies3. Although sonographic evidence of fracture
Disclosure: None of the authors received payments or services, either A commentary by Langdon A. Hartsock,
directly or indirectly (i.e., via his or her institution), from a third party in MD, is linked to the online version of this
support of any aspect of this work. None of the authors, or their institu- article at jbjs.org.
tion(s), have had any financial relationship, in the thirty-six months prior to
submission of this work, with any entity in the biomedical arena that could
be perceived to influence or have the potential to influence what is written
in this work. Also, no author has had any other relationships, or has
engaged in any other activities, that could be perceived to influence or
have the potential to influence what is written in this work. The complete
Disclosures of Potential Conflicts of Interest submitted by authors are
always provided with the online version of the article.
Fig. 3
Postoperative radiographs of a fracture requiring greater than ten months to demonstrate tricortical bridging (it occurred between the ten and thirteen-month
radiographs). *The fracture showed unicortical bridging within four months and bicortical bridging at seven months.
bridging but did not eventually have bridging of three cor- bridged cortices required (Fig. 2). Assessment for any cortical
tices did not exhibit bridging callus until after four months bridging at four months accurately predicted the healing out-
postoperatively. come of 174 fractures (99%). Assessment of additional cortical
bridging was also accurate, but required substantially longer
Bridging of Additional Cortices observation. Bicortical bridging was accurate in predicting the
Bicortical and tricortical bridging also successfully predicted
union (AUC = 0.975 and 0.990, respectively; p < 0.0001) (Fig.
4). However, ROC curve analysis demonstrated that the opti-
mum postoperative time to evaluate radiographs on the basis of
those criteria was seven months (bicortical) and twelve months
(tricortical). The times to achieve varying degrees of cortical
bridging are shown according to patient demographics and in-
jury characteristics in Table II. Among the fractures that healed,
the median time to cortical bridging was similar for all patient
and injury factors evaluated. However, high-energy fractures
demonstrated unicortical bridging early (within four months)
but required up to a year for tricortical bridging. All fractures
showing tricortical bridging later than nine months postopera-
tively (eight of 176) were related to a high-energy mechanism. All
low-energy fractures that eventually healed demonstrated tri- Fig. 4
cortical bridging within four months. Three separate receiver operating characteristic (ROC) curves demon-
strating that all degrees of cortical bridging were predictive of the final
Timing of Radiographic Assessment: Effect on Predictive healing outcome. The curves overlie each other as they all are nearly
Accuracy perfectly predictive, each with an area under the curve (AUC) approaching
The predictive accuracy of cortical bridging was highly de- 1.0. The major difference in the ROC curve analyses was the postoperative
pendent on the time of radiographic assessment and number of time required to achieve this accuracy for each criterion.
1070
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
ANY CORTICAL BRIDGING PREDICTS HEALING
V O LU M E 96-A N U M B E R 13 J U LY 2, 2 014
d d
OF TIBIA L SHAFT FRACTURES
healing outcome of 173 fractures (98%) at seven months post- and endosteal healing without bridging16. The biological pro-
operatively. Tricortical bridging was accurate for 174 fractures cess of fracture-healing requires several months for a fracture
(99%), but not until greater than twelve months postoperatively. callus in an adult to progress through the stages of fracture
Notably, two fractures (1%) were miscategorized on the basis of all hematoma, granulation tissue, cartilage intermediate, and
of the bridging criteria no matter the time of assessment. These mineralized callus before a full fracture-healing is achieved17.
were both suspected to be fibrous nonunions more than twelve Cortical bridging implies an adequate early healing response
months postoperatively on the basis of clinical signs and symp- and was predictive of union, whereas fractures that did not
toms, despite tricortical bridging radiographically. These clinical show bridging of a cortex by four months were extremely
suspicions were supported by a persistent fracture line on com- unlikely to heal. The accuracy of this assessment relies on the
puted tomography (CT) and confirmed intraoperatively during consistent timing of initial cortical bridging for fractures that
bone-grafting of the nonunion site. eventually unite without intervention (Fig. 1). Although union
rates differed according to the open-fracture type, patient and
Variation of Interobserver and Intraobserver Reliability injury characteristics did not affect the time to cortical bridging
Depending on Radiographic Criteria Studied in fractures that eventually united (Table II).
Assessment of the interobserver reliability of dichotomous The accuracy of radiographic assessment in predicting
radiographic criteria (any cortical bridging, bicortical bridging, the healing of tibial fractures has been reported to be as low as
and tricortical bridging) resulted in kappa coefficients of 0.92 50%18. In our study, some degree of cortical bridging occurred
(95% confidence interval [CI] = 0.75 to 0.99) for any cortical within four months in all fractures that eventually united, thus
bridging, 0.81 (95% CI = 0.62 to 0.92) for bicortical bridging, allowing accurate predictions to be made for nearly all fractures
and 0.85 (95% CI = 0.66 to 0.94) for tricortical bridging. As- (99%) relatively early in the postoperative period. The area
sessment of the interobserver reliability for the exact number of under the ROC curve for this criterion (0.995) is consistent with
bridged cortices (zero, one, two, three, or four) demonstrated a the criterion being an excellent predictor of the final healing
kappa coefficient of 0.58 (95% CI = 0.39 to 0.74). Intraobserver outcome19.
reliability findings were similar, with kappa coefficients of 0.85 Among radiographic criteria for healing, bridging callus
(95% CI = 0.75 to 0.99) for any cortical bridging, 0.88 (95% has been found in basic-science and clinical studies to predict
CI = 0.70 to 0.97) for bicortical bridging, 0.92 (0.75 to 0.99) mechanical strength relatively reliably in comparison with
for tricortical bridging, and 0.62 (95% CI = 0.42 to 0.78) for measures such as callus area and quality18,20-24. However, there
the exact number of cortices bridged. is little evidence regarding the minimum number of bridging
cortices required for a fracture to be considered healed. Pre-
Discussion vious studies, including those satisfying U.S. Food and Drug
reported for radiographic assessments of tibial union and sur- often partially obscures cortices on standard radiographic
geons’ general impression of healing (kappa = 0.6 and 0.67, views. It is important to note that this study was performed
respectively) and is similar to that found for the RUST score with use of digital radiography. The ability to adjust the con-
(Radiographic Union Score for Tibial fractures)21,23. Agreement trast of images and ‘‘zoom in’’ on the fracture site may provide a
for the more stringent criteria of bridging of two (kappa = 0.81) distinct advantage in terms of both the predictive accuracy and
and three cortices (kappa = 0.85) remained substantial and was the reliability of cortical bridging assessments. The findings
similar to that for cortical bridging in previous reports21,28. The have potential implications for determining the indications for
reliability of the assessment of the exact number of cortices bone-grafting procedures, the optimal timing of radiographic
bridged was much lower (0.58) than that of the above dichoto- follow-up, and future studies of bone healing. n
mous criteria. Similar results were found for intraobserver reli-
ability, with higher reliability for the dichotomous measures
than for the exact number of cortices bridged. These findings
are logical given that determining whether any cortical bridging William D. Lack, MD
has occurred requires only that observers agree on whether 348A West Huron,
Chicago, IL 60654.
bridging callus is present. E-mail address: wdlack@gmail.com
Limitations of this study include the limited patient
population at a single institution. Retrospective evaluation James S. Starman, MD
prevents the degree of standardization possible in a prospective Madhav Karunakar, MD
study, and therefore patients with prolonged intervals (greater Stephen Sims, MD
than three months) between radiographs were excluded. Department of Orthopaedic Surgery-Trauma,
Carolinas Medical Center,
A strength of this study was the avoidance of testing a
1025 Morehead Medical Plaza,
priori criteria. Instead, the study was adequately powered on #300, P.O. Box 32861,
the basis of pilot data to employ an analytic method (ROC Charlotte, NC 28204
curve analysis) to determine the predictive accuracy of each
criterion and to ascertain the time at which each was most Rachel Seymour, PhD
accurate. Furthermore, it was designed to avoid previously Michael Bosse, MD
reported limitations of research on this topic by defining both Orthopaedic Clinical Research (R.S.)
and Department of Orthopaedic Surgery-Trauma (M.B.),
clinical and radiographic union and by assessing reliability29.
Carolinas Medical Center,
Future studies of interest would include external validation 1320 Scott Avenue,
through prospective evaluation of a separate patient group. This P.O. Box 32861,
would allow patients to be followed conservatively through a Charlotte, NC 28204
predetermined time point at standardized radiographic intervals.
Any cortical bridging is a simple, reliable, early radio- James Kellam, MD
graphic criterion that is an excellent predictor of final healing of Department of Orthopaedic Surgery-Trauma,
Carolinas Medical Center,
tibial shaft fractures. Early bridging of any cortex may be pre- 1000 Blythe Boulevard, Level 1,
dictive of healing of other fractures as well, although the pre- B-Wing, Room 1201,
dictive relationship that we described is likely strongest for P.O. Box 32861,
fractures treated with intramedullary fixation, as plate fixation Charlotte, NC 28204
References
1. Nicoll EA. Fractures of the tibial shaft: A survey of 705 cases. J Bone Joint Surg Br. 7. LaVelle DG. Delayed union and nonunion of fractures. In: Canale TS, editor. Camp-
1964 Aug;46:373-87. bell’s operative orthopaedics. 9th ed. St. Louis, Missouri: Mosby-Lifeline; 1998.
2. Müller ME, Allgöwer M, Schneider R, Willenegger H, editors. Manual of internal 8. Caputo AE. Healing of bone and connective tissue. In: Bronner F, Worrell RV, editors.
fixation: techniques recommended by the AO group. 2nd ed. Berlin: Springer-Verlag; Orthopaedics: principles of basic and clinical science. New York: CRC Press; 1999.
1979. 9. Moed BR, Subramanian S, van Holsbeeck M, Watson JT, Cramer KE, Karges DE,
3. Bhandari M, Guyatt GH, Tong D, Adili A, Shaughnessy SG. Reamed versus non- Craig JG, Bouffard JA. Ultrasound for the early diagnosis of tibial fracture healing
reamed intramedullary nailing of lower extremity long bone fractures: a systematic after static interlocked nailing without reaming: clinical results. J Orthop Trauma.
overview and meta-analysis. J Orthop Trauma. 2000 Jan;14(1):2-9. 1998 Mar-Apr;12(3):206-13.
4. Centers for Medicaid and Medicare Services. Amend coverage issues manual 10. Sarmiento A. On the behavior of closed tibial fractures: clinical/radiological
section 35-48. 1999 Nov 9. http://www.cms.gov/medicare-coverage-database/ correlations. J Orthop Trauma. 2000 Mar-Apr;14(3):199-205.
details/nca-decision-memo.aspx?NCAId=24&NCDId=65&ncdver=2&NcaName= 11. Oni OO, Hui A, Gregg PJ. The healing of closed tibial shaft fractures. The natural
Electrical1Stimulation1for1Fracture1Healing&IsPopup=y&bc=AAAAAAAAEAAA&. history of union with closed treatment. J Bone Joint Surg Br. 1988 Nov;70(5):787-90.
Accessed 2012 Nov 2. 12. Clancey GJ, Winquist RA, Hansen ST Jr. Nonunion of the tibia treated with
5. Bishop JA, Palanca AA, Bellino MJ, Lowenberg DW. Assessment of compromised Küntscher intramedullary nailing. Clin Orthop Relat Res. 1982 Jul;(167):191-6.
fracture healing. J Am Acad Orthop Surg. 2012 May;20(5):273-82. 13. Edwards CC, Jaworski MF. Hoffman external fixation in open tibial fractures with
6. Schemitsch EH, Bhandari M, Guyatt G, Sanders DW, Swiontkowski M, Tornetta P, tissue loss. Orthop Trans. 1979;3:261-2.
Walter SD, Zdero R, Goslings JC, Teague D, Jeray K, McKee MD; Study to Pro- 14. Govender S, Csimma C, Genant HK, Valentin-Opran A, Amit Y, Arbel R, Aro H,
spectively Evaluate Reamed Intramedullary Nails in Patients with Tibial Fractures Atar D, Bishay M, Börner MG, Chiron P, Choong P, Cinats J, Courtenay B, Feibel R,
(SPRINT) Investigators. Prognostic factors for predicting outcomes after intramed- Geulette B, Gravel C, Haas N, Raschke M, Hammacher E, van der Velde D, Hardy P,
ullary nailing of the tibia. J Bone Joint Surg Am. 2012 Oct 3;94(19):1786-93. Holt M, Josten C, Ketterl RL, Lindeque B, Lob G, Mathevon H, McCoy G, Marsh D,
1072
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
ANY CORTICAL BRIDGING PREDICTS HEALING
V O LU M E 96-A N U M B E R 13 J U LY 2, 2 014
d d
OF TIBIA L SHAFT FRACTURES
Miller R, Munting E, Oevre S, Nordsletten L, Patel A, Pohl A, Rennie W, Reynders P, 22. McClelland D, Thomas PB, Bancroft G, Moorcraft CI. Fracture healing assess-
Rommens PM, Rondia J, Rossouw WC, Daneel PJ, Ruff S, Rüter A, Santavirta S, ment comparing stiffness measurements using radiographs. Clin Orthop Relat Res.
Schildhauer TA, Gekle C, Schnettler R, Segal D, Seiler H, Snowdowne RB, Stapert J, 2007 Apr;(457):214-9.
Taglang G, Verdonk R, Vogels L, Weckbach A, Wentzensen A, Wisniewski T; BMP-2 23. Whelan DB, Bhandari M, Stephen D, Kreder H, McKee MD, Zdero R, Schemitsch
Evaluation in Surgery for Tibial Trauma (BESTT) Study Group. Recombinant human EH. Development of the radiographic union score for tibial fractures for the as-
bone morphogenetic protein-2 for treatment of open tibial fractures: a prospective, sessment of tibial fracture healing after intramedullary fixation. J Trauma. 2010
controlled, randomized study of four hundred and fifty patients. J Bone Joint Surg Am. Mar;68(3):629-32.
2002 Dec;84(12):2123-34. 24. Kooistra BW, Dijkman BG, Busse JW, Sprague S, Schemitsch EH, Bhandari M.
15. Sanders R, Jersinovich I, Anglen J, DiPasquale T, Herscovici D Jr. The treatment The radiographic union scale in tibial fractures: reliability and validity. J Orthop
of open tibial shaft fractures using an interlocked intramedullary nail without ream- Trauma. 2010 Mar;24(Suppl 1):S81-6.
ing. J Orthop Trauma. 1994 Dec;8(6):504-10. 25. Hernigou P, Poignard A, Beaujean F, Rouard H. Percutaneous autologous bone-
16. Marsh D. Concepts of fracture union, delayed union and nonunion. Clin Orthop marrow grafting for nonunions. Influence of the number and concentration of pro-
Relat Res. 1998 Oct;(355)(Suppl):S22-30. genitor cells. J Bone Joint Surg Am. 2005 Jul;87(7):1430-7.
17. McKinley DW, Chambliss ML. Follow-up radiographs to detect callus formation 26. Keating JF, O’Brien PJ, Blachut PA, Meek RN, Broekhuyse HM. Locking
after fractures. Arch Fam Med. 2000 Apr;9(4):373-4. intramedullary nailing with and without reaming for open fractures of the tibial
18. Hammer RR, Hammerby S, Lindholm B. Accuracy of radiologic assessment of shaft. A prospective, randomized study. J Bone Joint Surg Am. 1997 Mar;79(3):
tibial shaft fracture union in humans. Clin Orthop Relat Res. 1985 Oct;(199):233-8. 334-41.
19. Tape TG. The Area Under an ROC Curve. Intrepreting diagnostic tests. http:// 27. Landis JR, Koch GG. The measurement of observer agreement for categorical
gim.unmc.edu/dxtests/roc3.htm. Accessed 2012 Aug 23. data. Biometrics. 1977 Mar;33(1):159-74.
20. Panjabi MM, Walter SD, Karuda M, White AA, Lawson JP. Correlations of ra- 28. Lack WD, Fredericks D, Petersen E, Donovan M, George M, Nepola J, Smucker J,
diographic analysis of healing fractures with strength: a statistical analysis of Femino JE. Effect of aspirin on bone healing in a rabbit ulnar osteotomy model [Epub
experimental osteotomies. J Orthop Res. 1985;3(2):212-8. ahead of print]. J Bone Joint Surg Am. 2013 Mar 20;95(6):488-96.
21. Whelan DB, Bhandari M, McKee MD, Guyatt GH, Kreder HJ, Stephen D, Schemitsch 29. Corrales LA, Morshed S, Bhandari M, Miclau T 3rd. Variability in the assessment
EH. Interobserver and intraobserver variation in the assessment of the healing of tibial of fracture-healing in orthopaedic trauma studies. J Bone Joint Surg Am. 2008
fractures after intramedullary fixation. J Bone Joint Surg Br. 2002 Jan;84(1):15-8. Sep;90(9):1862-8.