OTA 1999 Posters

The Relationship of Socioeconomic Risk Factors to Complication Rates in External Fixation of the Tibia

Wade R. Smith, MD; Paul S. Shurnas, MD; David G. Stewart, MD; Brent A. Snee, MD; Steven J. Morgan, MD; Bruce H. Ziran, MD, Denver Health Medical Center, Denver, CO

Purpose: External fixation is an accepted means of treating high-energy fractures in the diaphysis and peri-articular regions of the tibia (1-8). However, our experience indicates that certain subgroups of patients incur high rates of major complications when using standard techniques with the external fixator placed on the tibia for either diaphyseal or peri-articular fractures. The purpose of this study was to determine the effect, if any, that potential risk factors, intravenous drug use (IVDU), alcohol use, tobacco use, poverty, and age have on the incidence of the following four major complications: nonunions, malunions, delayed unions requiring a secondary procedure, and osteomyelitis in open and closed tibia fractures treated by external fixation.

Methods: We performed a retrospective review of all patients (6000 charts with 188 adult patients and fractures identified) with diaphyseal, plateau, and pilon fractures of the tibia treated by external fixation over a five-year period at our institutions. We reviewed radiographs from the initial post- operative period, when healing was considered complete, and after the external fixator had been removed. We assessed radiographs for angulation, bone healing, and evidence of osteomyelitis, and noted the number of major complications. EpiInfo statistical analysis published by the Centers for Disease Control, chi square (multivariate), and logistic regression analysis examined the effect of potential risk factors on the incidence of these four major complications. We devised a statistically validated three-point scale (the External Fixation Risk Index- EFRI) by giving a patient one point for each statistically valid risk factor to predict those patients at increased risk for major complications when treated with external fixation of the diaphysis, plateau, or plafond.

Results: Patient age ranged from 15 to 78 years (mean 41.4 years). We studied 136 males and 52 females with 80 diaphyseal, 50 plateau, and 58 pilon fractures of the tibia. Immediate postoperative radiographs exibited no more than 8 degrees of angulation in any plane of the tibia diaphysis, no more than 2 millimeters of articular step-off, and appeared to have stable fixation based on standard fixation principles. Fifty-nine percent open injuries (111 of 188 fractures) had the following distribution: 77.5% (62 of 80) of diaphyseal fractures, 52% (26 of 50) of plateau fractures, and 39.7% (23 of 58) of pilon fractures. The most common mechanism of injury in 29.3% (55 of 188) of patients was an MVA; fall from a height was second in 25.0% (47 of 188) of patients; an auto-pedestrian accident was third most common in 20.2% (38 of 188) of patients; a motorcycle accident was fourth in 15.4% (29 of 188) of patients, and the remainder included crush, auto-bike, and gun-shot wounds. We found no significant difference in complication rate or risk factors between male and female patients. There was no significant difference in complication rates for diaphyseal, plateau, or pilon fractures or between mechanisms of injury when corrected for open versus closed injury. Of the 115 patients with polytrauma, no correlation existed between polytrauma or severity of injury and risk factors.

Major complications occurred in 117 of 188 patients (62.2%). We identified 160 major complications in 117 patients as follows: nonunion in 37.2% (70 of 188 fractures), malunion in 16.5% (31 of 188 fractures), delayed union in 15.4% (29 of 188 fractures), osteomyelitis in 16.0% (30 of 188 fractures). Complication rates for the 63 patients with no risk factors were the following: 18 nonunions (28.6%), 8 delayed unions (12.7%), 7 malunions (11.1%), 9 osteomyelitis (14.3%), and overall complication rate of 50.7% (32 of 63 patients). Complication rates for the 17 patients with a risk factor of alcohol only were the following: 6 nonunions (33.3%), 4 delayed unions (22.2%), 4 malunions (22.2%), 4 osteomyelitis (22.2%), and an overall complication rate of 70.6% (12 of 17 patients). Complication rates for the 61 patients with alcohol and tobacco as risk factors were the following: 31 nonunions (50.8%), 12 delayed unions (19.7%), 14 malunions (23.0%), 9 osteomyelitis (14.8%), and an overall complication rate of 77.1% (47 of 61 patients). Complication rates in the 37 patients with alcohol and poverty as risk factors were the following: 19 nonunions (51.4%), 9 delayed unions (24.3%), 11 malunions (29.7%), 3 osteomyelitis (8.1%), and overall complication rate of 78.4% (29 of 37 patients). Complication rates for the 10 patients with tobacco only as a risk factor were the following: 4 nonunions (30.8%), 3 delayed unions (23.1%), no malunions, 2 osteomyelitis (15.4%), and an overall complication rate of 60.0% (6 of 10 patients). Complications rates for the 40 patients with tobacco and poverty as risk factors were the following: 20 nonunions (50.0%), 10 delayed unions (25.0%), 12 malunions (30.0%), 7 osteomyelitis (17.5%), and an overall complication rate of 85.0% (34 of 40 patients). Complication rates for the 9 patients with poverty alone as a risk factor were the following: 6 nonunions (42.9%), 1 delayed union (7.1%), 3 malunions (21.4%), 2 osteomyelitis (14.3%), and an overall complication rate of 55.6% (5 of 9 patients). Complication rates for the 29 patients with alcohol, tobacco, and poverty as risk factors were the following: 17 nonunions (58.6%), 9 delayed unions (31.0%), 10 malunions (34.5%), 3 osteomyelitis (10.3%), and an overall complication rate of 93.1% ( 27 of 29 patients). Minor complications included pin tract infection in 33.0% (62 of 188), and failure of external fixation in 4.8% (9 of 188). Pin tract infection had no correlation with the development of osteomyelitis, and the rate was not significantly different in open versus closed injury. Only 12.2% (23 of 188) of patients did not require a secondary procedure such as IM rod, ORIF, bone graft, arthrodesis, or amputation; bone graft/reconstruction for severe bone loss was not included in the study.

Alcohol, tobacco, and poverty are the statistically significant risk factors independently and in combination that equally contribute to increase the incidence of major complications. IVDU and age are not significant risk factors with the number of patients available for study. Based on multivariate analysis, the odds ratio for determining the risk of major complications from alcohol, tobacco, and poverty alone (FFRI of 1) versus no risk factors is 2.1 (p= .02), 2.1 (p= .02), and 2.0 (p= .05) respectively. The odds ratio for alcohol plus tobacco, tobacco plus poverty, alcohol plus poverty (EFRI of 2), and from all three risk factors (EFRI of 3) versus no risk factors is 2.7 (p= .006), 3.6 (p=.005), 3.5 (p= .005), and 10.4 (p= .0004) respectively.

Logistic regression verifies that given any combination of risk factors the EFRI model is predictive for an EFRI of 3 with an odds ratio of 9.1 (p= .003), and for an EFRI of 2 with an odds ratio of 3.1 (p= .01). However, the model is not predictive for all risk factor combinations of alcohol, tobacco, and poverty with the numbers available for study. Power analysis indicates that an N= 200 (power = .8) for each risk factor would be necessary to reach statistical significance for all random combinations. For any significant risk factor alone or in combination the EFRI shows the complication rate increases over patients without risk factors. This holds true for external fixation of diaphyseal, plateau, or pilon fractures, and the likelihood that this is a chance occurrence is low.

Discussion: We used standard external fixation techniques in the treatment of high-energy diaphyseal and peri-articular fractures of the tibia. Our experience with this type of treatment indicated that our complication rates were unacceptable when compared to reported rates in the literature (9-12). We were unable to explain the increased incidence of major complications based on polytrauma, injury severity, length of hospitalization, open injury, mechanism of injury, or by fracture site on the tibia. The results of this study indicate that the increased incidence is most likely correlated with socioeconomic risk factors in our patient populations. Although the mechanisms by which alcohol, tobacco, and poverty increase the incidence of major complications is not known, the External Fixation Risk Index predicts preoperative complication risk. This has changed our treatment approach for high-risk patients.

Conclusion: Patients with a risk factor of poverty, alcohol, and tobacco alone or in combination have a significantly increased incidence of major complications when treated with external fixation of the tibial diaphysis, plateau, and plafond over those patients without such risk factors.