Open tibia fractures are often limb-threatening injuries due to their associated soft tissue damage. Deciding between amputation and limb salvage is complex. Both treatment options have nuanced advantages and disadvantages for the patient and the health care system in terms of pain, function, and cost.1–5 Numerous scoring systems have been developed in an attempt to clarify the decision-making process, with the goal of generating a score that would guide the physician between amputation and limb salvage. Patient and injury characteristics such as impaired plantar sensation, vascular injury, and fracture location have been proposed by some authors as predictive factors for proceeding with amputation.5–8 However, results are considerably limited by sample size. Furthermore, these scoring systems have failed to demonstrate clinical validity.9–11 Practically, the decision to proceed with amputation vs limb salvage in the setting of these limb-threatening injuries is often made on a case-by-case basis.
The burden of open tibia fracture and subsequent amputation has not yet been defined in the United States. A national database study in Sweden demonstrated a rate of open tibia fractures of approximately 2.8 to 3.4 per 100,000 person-years between 1998 and 2010.12 Similar data are not yet available for the United States. Obtaining this information would help define the societal burden posed by this problem.
The primary aim of this study was to describe the frequency of open tibia fractures along with their injury characteristics and overall amputation rates using a nationwide sample of patients. Given the previously reported association of neurovascular injury and fracture location with amputation, a secondary aim was to investigate fracture location and neurovascular injury as potential risk factors for amputation.
Materials and Methods
Design
The authors performed a retrospective analysis of all patients presenting with an open tibia fracture in the Health-care Cost and Utilization Project, Nationwide Inpatient Sample (NIS) from 2000 to 2011. The NIS is an all-payer inpatient database that provides discharge data for hospitalizations across the United States. It is the largest publicly available all-payer database. The NIS supplies inpatient diagnoses and procedures from the index hospitalization using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, as well as demographic and hospital data. Data in the NIS are provided with associated weights to generate US population numbers.13 Data provided in this analysis are in US estimates form. Institutional review board exemption was granted for this study.
Study Population
The study population was comprised of patients 18 years and older who presented with an open tibia fracture. The primary study outcome variable was whether patients underwent below knee amputation during their index hospitalization. This outcome was treated as a binary variable. Predictor variables included tibia fracture location (proximal third, middle third, and distal third), neurovascular injury (isolated tibial nerve injury, isolated arterial injury, combined neurovascular injury, and no neurovascular injury), patient age, patient sex, patient race, medical comorbidities, hospital location (rural and urban), and geographic region (Northeast, Midwest, South, and West; Table A, available in the online version of the article). The ICD-9-CM codes (Table B, available in the online version of the article) were used to identify open tibia fractures, amputation, and neurovascular injury.
Data Analysis
Patient demographics, medical comorbidities, tibia fracture location, neurovascular injury, hospital location, and geographic region were summarized using descriptive statistics and were compared between study cohorts of amputation vs no amputation using Student's t tests for continuous variables and chi-square tests for categorical variables. To examine the association between fracture location and amputation for each type of neurovascular injury, subgroup analysis was performed using chi-square tests. To study the association between amputation and injury characteristics, logistic regression was used to model amputation on fracture location, neurovascular injury (none vs any), and their interaction terms with and without covariate adjustment. All pairwise comparisons were conducted using contrast statements with Bonferroni correction for multiple testing. Confounders included age, sex, race, tobacco use, diabetes mellitus, acute renal failure, hospital location, and geographic region. Analyses were performed with SAS, version 9.4, software (SAS Institute).
Results
Using the NIS database, 175,283 open tibia fracture cases were identified for the years 2000 to 2011 in the United States. The amputation rate for all-comers during the initial hospitalization was 2.2% (n=3769). Figure 1 demonstrates the number of open tibia fractures per year with the associated amputation rates. Open tibia fracture frequency per year ranged from 12,961 to 17,601, with an associated amputation rate range of 1.6% to 2.6%. Mean age of the cohort was 41.3 years (standard error of the mean [SEM], 0.11).
Fracture distribution by patient age is demonstrated in the histogram shown in Figure 2, which demonstrates a bimodal distribution with a peak at age 22 years and a second, smaller peak at age 46 years. Average age difference between those who underwent amputation vs those who underwent limb salvage was 2 years (43.08 years [SEM, 0.58] vs 41.30 years [SEM, 0.11]). Most patients in the study population were men (73.9%, n=129,097). The observed rate of smoking was lower in the amputation group compared with the no amputation group (amputation: 14.9%, n=560; no amputation: 19.1%, n=32,811; P=.001), whereas the opposite was observed for diabetes mellitus (amputation: 9.3%, n=351; no amputation: 6.6%, n=11,292; P=.02) and acute renal failure (amputation: 6.1%, n=229; no amputation: 1.7%, n=2957; P<.0001). More than 85% (n=3237) of the patients in the amputation group did not have a neurovascular injury, and approximately half of the fractures in both the amputation and the no amputation group occurred in patients with midshaft fractures. The majority of amputations occurred in an urban teaching hospital (82.2%, n=3097), and no evidence suggested that amputation rates differ significantly based on hospital region (P=.63).
Amputation rates are displayed in Figure 3. Rates were 2.3% (n=129,097) in men vs 1.6% (n=45,644) in women. The amputation rate was 1.7% (n=33,371) in smokers, 3.0% (n=11,644) in patients with diabetes mellitus, and 7.2% (n=3186) in patients in acute renal failure. Amputation rates were 1.9% (n=170,887) in patients without any neurovascular injury, 20.1% (n=200) in patients with an isolated tibial nerve injury, 10.5% (n=4002) in patients with an isolated arterial injury, and 37.3% (n=195) in patients with combined neurovascular injuries. Finally, amputation rates were 1.8% (n=39,197) in proximal fractures, 2.0% (n=89,355) in midshaft fractures, and 2.7% (n=46,732) in distal fractures.
The interaction between neurovascular injury and fracture location is demonstrated in Figure 4. When evaluating patients with any neurovascular injury (isolated tibial nerve, isolated arterial, or combined injury), amputation rates were consistently higher for patients with distal fractures compared with those with midshaft fractures. However, amputation rates among patients with proximal fractures varied depending on neurovascular injury status. Chi-square tests showed that fracture location was significantly associated with amputation for patients with no neurovascular injury (P<.0001), those with an isolated arterial injury (P=.002), and those with an isolated tibial nerve injury (P=.03).
When controlling for age, sex, race, acute renal failure, diabetes mellitus, tobacco use, hospital location/teaching status, and hospital region, a significantly higher rate of amputation was observed for patients with neurovascular injury and middle third fractures (odds ratio [OR], 5.45; 95% CI, 1.73–17.19; P<.04) and distal third fractures (OR, 12.39; 95% CI, 5.52–27.83; P<.0001) compared with other groups. Although the individual effects of fracture location and neurovascular injury were associated with amputation in many pairwise comparisons in unadjusted analyses (Tables C and D, available in the online version of the article), these associations diminished after adjusting for confounding variables. However, acute renal failure was associated with an increased rate of amputation (OR, 11.69; 95% CI, 4.64–29.45; P<.0001) while controlling for the confounders listed in Table E (available in the online version of the article). In addition, when controlling for the same confounders, the odds for amputation were lower in rural (OR, 0.10; 95% CI, 0.01–0.75) and urban nonteaching hospitals (OR, 0.33; 95% CI, 0.16–0.69) compared with urban teaching hospitals, with an overall P=.001 (Table E). Although the amputation rate was significantly lower among smokers in unadjusted analyses, this association did not persist in the multivariable model.
Discussion
A projected 175,283 open tibia fractures were recorded in the United States from 2000 to 2011. In 2010, a total of 17,601 open fractures were captured by the NIS. Using US Census data for the same year, which demonstrated a US population of 234.6 million people 18 years or older, this corresponds to 7.5 new open tibia fractures per 100,000 person-years.14 In comparison, a recent study evaluating open tibia fractures and amputations in Sweden from 1998 to 2010 using the Swedish National Patient Register found open tibia fractures occurring at a rate of 2.8 to 3.4 per 100,000 person-years.12 The combined early and late amputation rate in the Swedish cohort was 3.6%.12 This is comparable to the current study's amputation rate of 2.15% for all-comers during their index hospitalization.
Mean age of patients with an open tibia fracture was 41 years (SEM, 0.11), which is comparable to other studies in the literature.10,12,15 The highest rates of open tibia fractures in the current sample were for patients in their early 20s, with a second, smaller peak for patients in their mid-40s. Although a statistically significant difference was observed in age between patients with and without amputation in unadjusted analyses, this difference did not persist in multivariable analyses. Furthermore, the magnitude of the difference in age in unadjusted analyses was small (approximately 2 years).
In unadjusted analyses, men had higher overall rates of open tibia fractures and amputation than women, which is consistent with prior reports.12 However, the association between male sex and amputation after open tibia fracture did not persist in adjusted analyses.
Patients in the current sample with diabetes mellitus, acute renal failure, congestive heart failure, and vascular disease underwent amputation at an increased rate. However, acute renal failure was the only comorbidity significantly associated with amputation after adjusting for potential effect modifiers at an OR of 11.69 (95% CI, 4.64–29.45). It is possible that the presence of acute renal failure is a surrogate for higher local injury severity or additional visceral injury, which may preclude attempts at limb salvage.
Two-thirds of all open tibia fractures were cared for in urban teaching hospitals, and urban teaching hospitals performed more than 80% of the amputations. In the adjusted analysis, urban teaching hospitals were associated with higher odds of amputation compared with other types of hospitals. However, higher-severity injuries are often treated in tertiary academic hospitals, which may explain this association but is difficult to prove with the current data.16 Of note, patients transferred from an outside acute care hospital underwent slightly higher rates of amputation (amputation of transferred patients: 2.8%, n=130; amputation of nontransferred patients: 2.2%, n=1210). However, this difference was not statistically significant (P=.36).
Although successful limb salvage has been reported in patients with a nerve injury, the presence of a significant nerve injury has traditionally been used as criteria for amputation.6,17 The current authors observed that the overall rate of neurovascular injury in the amputation group was 14.0% (n=531), compared with 2.3% (n= 3865) in the nonamputation group. The rate of amputation within the index hospitalization in the setting of partial and complete tibial nerve transections without an associated vascular injury was 20.0% (n=40).
For arterial injuries, the literature is more developed. Moniz et al7 studied vascular injuries in the setting of lower extremity trauma and reported amputation rates of 20%, 33%, and 100% in the setting of injury 2-vessel run-off, 1-vessel run-off, and 0-vessel run-off, respectively.7 A recent review of all popliteal vascular injuries in the setting of lower extremity trauma demonstrated an amputation rate of 20% in patients with tibia fractures, although it did not characterize the open or closed nature of the fracture.18 In a study of 62 patients with blunt lower extremity trauma and vascular injury from 1995 to 2002, Rozycki et al reported an amputation rate of 18%.19 A review of 46 vascular injuries in the setting of tibia fracture demonstrated an amputation rate of 6.5% for popliteal and tibial artery and vein injuries.20 In the current series, patients with an isolated arterial injury (n=4002) underwent amputation at a rate of 10.5% (n=419). This represents all patients with evidence of dysvascularity on initial presentation regardless of degree of injury and anatomic location. When a combined neurologic and vascular injury (n=195) was observed, the amputation rate increased to 36.9% (n=72).
Fracture location was distributed approximately into a ratio of 1:2:1 for proximal, midshaft, and distal fractures. Amputation rates were highest among patients with distal third fractures at 2.7% (n=46,732), compared with 2.0% for mid-shaft (n=89,355) and 1.8% for proximal fractures (n=39,197). This may be reflective of the decreasing soft tissue envelope in the distal leg that is less tolerant of traumatic injury and more likely to be associated with soft tissue necrosis and neurovascular injury.8 Although a significant interaction was observed between the fracture site and neurovascular injury, no evidence of an association was observed between amputation and fracture site for all different types of neurovascular injury. Specifically, among patients with nerve injuries, the current authors observed the highest amputation rate for those who had proximal fractures compared with other fracture location. However, the amputation rate was highest for patients who had distal third fractures if they had arterial injuries. This is intuitive given the devastating impact of a proximally transected nerve prior to distal bifurcations. Combined neurovascular injuries were poorly tolerated by both proximal and distal third fractures compared with midshaft fractures.
In practice, considerable variability occurs in open tibia fractures and the decision to amputate is multifactorial.3,21–23 Furthermore, many patient-specific factors relating to self-efficacy and social support are being demonstrated to be of greater importance than some characteristics of injury severity.24,25 None of these factors are captured in the generated scoring systems, potentially limiting their efficacy.5,26–29
Study Limitations
The data presented here are 12-year population-level estimates regarding amputation in the setting of open tibia fractures with various injury characteristics. This information serves as a useful reference point for physicians when educating patients on initial presentation regarding outcomes and expectations. However, there were limitations to this analysis, primarily relating to the use of a large national database. This database approximates a 20% sample of US hospitals. The integrity of the analysis depends on the reliability of hospital documentation and coding of patient factors, diagnoses, and procedures. Study findings could be influenced by data that are either missing or inaccurate. In addition, it was not possible to assess fracture severity based on Orthopaedic Trauma Association classification or soft tissue injury according to the Gustilo-Anderson type because these were not available within the database. Instead, the authors used ICD-9-CM codes for nerve and vascular injury as surrogates for injury severity and evaluated these factors as clinical determinants of amputation. Finally, the authors did not have access to follow-up data in the form of patient-reported outcomes or additional surgeries.
Conclusion
From 2000 to 2011, a projected 175,283 open tibia fractures occurred in the United States. The overall amputation rate for open tibia fractures was 2.2%. Amputation rates varied based on fracture location, with the highest rates in distal third fractures and the lowest rates in proximal third fractures. Combined neurovascular injuries had the highest amputation rates, followed by isolated nerve and arterial injuries.
References
- Hoogendoorn JM, van der Werken C. Grade III open tibial fractures: functional outcome and quality of life in amputees versus patients with successful reconstruction. Injury. 2001;32(4):329–334. doi:10.1016/S0020-1383(00)00250-3 [CrossRef] PMID:11325370
- Georgiadis GM, Behrens FF, Joyce MJ, Earle AS, Simmons AL. Open tibial fractures with severe soft-tissue loss: limb salvage compared with below-the-knee amputation. J Bone Joint Surg Am. 1993;75(10):1431–1441. doi:10.2106/00004623-199310000-00003 [CrossRef] PMID:8408131
- Hertel R, Strebel N, Ganz R. Amputation versus reconstruction in traumatic defects of the leg: outcome and costs. J Orthop Trauma. 1996;10(4):223–229. doi:10.1097/00005131-199605000-00001 [CrossRef] PMID:8723399
- Dagum AB, Best AK, Schemitsch EH, Mahoney JL, Mahomed MN, Blight KR. Salvage after severe lower-extremity trauma: are the outcomes worth the means?Plast Reconstr Surg.1999;103(4):1212–1220. doi:10.1097/00006534-199904010-00017 [CrossRef] PMID:10088509
- Lange RH, Bach AW, Hansen ST Jr, Johansen KH. Open tibial fractures with associated vascular injuries: prognosis for limb salvage. J Trauma. 1985;25(3):203–208. doi:10.1097/00005373-198503000-00006 [CrossRef] PMID:3981670
- Bosse MJ, McCarthy ML, Jones AL, et al. Lower Extremity Assessment Project (LEAP) Study Group. The insensate foot following severe lower extremity trauma: an indication for amputation?J Bone Joint Surg Am.2005;87(12):2601–2608. PMID:16322607
- Moniz MP, Ombrellaro MP, Stevens SL, Freeman MB, Diamond DL, Goldman MH. Concomitant orthopedic and vascular injuries as predictors for limb loss in blunt lower extremity trauma. Am Surg. 1997;63(1):24–28. PMID:8985067
- Soltanian H, Garcia RM, Hollenbeck ST. Current concepts in lower extremity reconstruction. Plast Reconstr Surg. 2015;136(6):815e–829e. doi:10.1097/PRS.0000000000001807 [CrossRef] PMID:26595037
- Ly TV, Travison TG, Castillo RC, Bosse MJ, MacKenzie EJLEAP Study Group. Ability of lower-extremity injury severity scores to predict functional outcome after limb salvage. J Bone Joint Surg Am. 2008;90(8):1738–1743. doi:10.2106/JBJS.G.00136 [CrossRef] PMID:18676906
- Higgins TF, Klatt JB, Beals TC. Lower Extremity Assessment Project (LEAP): the best available evidence on limb-threatening lower extremity trauma. Orthop Clin North Am. 2010;41(2):233–239. doi:10.1016/j.ocl.2009.12.006 [CrossRef] PMID:20399362
- Bosse MJ, MacKenzie EJ, Kellam JF, et al. A prospective evaluation of the clinical utility of the lower-extremity injury-severity scores. J Bone Joint Surg Am. 2001;83(1):3–14. doi:10.2106/00004623-200101000-00002 [CrossRef] PMID:11205855
- Tampe U, Weiss RJ, Stark B, Sommar P, Al Dabbagh Z, Jansson KA. Lower extremity soft tissue reconstruction and amputation rates in patients with open tibial fractures in Sweden during 1998–2010. BMC Surg. 2014;14(1):80. doi:10.1186/1471-2482-14-80 [CrossRef] PMID:25323662
- Healthcare Cost and Utilization Project. Overview of the National (Nationwide) Inpatient Sample (NIS). Accessed January 15, 2016. https://www.hcup-us.ahrq.gov/nisoverview.jsp
- United States Census Bureau. By decade: 2010 census. Accessed October 1, 2016. http://www.census.gov/2010census
- Jaña Neto FC, de Paula Canal M, Alves BA, Ferreira PM, Ayres JC, Alves R. Analysis of the characteristics of patients with open tibial fractures of Gustilo and Anderson type III. Rev Bras Ortop. 2016;51(2):143–149. doi:10.1016/j.rboe.2016.01.002 [CrossRef] PMID:27069881
- Rogers FB, Osler TM, Shackford SR, Cohen M, Camp L, Lesage M. Study of the outcome of patients transferred to a level I hospital after stabilization at an outlying hospital in a rural setting. J Trauma. 1999;46(2):328–333. doi:10.1097/00005373-199902000-00022 [CrossRef] PMID:10029042
- Momoh AO, Kumaran S, Lyons D, et al. An argument for salvage in severe lower extremity trauma with posterior tibial nerve injury: the Ganga Hospital experience. Plast Reconstr Surg. 2015;136(6):1337–1352. doi:10.1097/PRS.0000000000001814 [CrossRef] PMID:26270902
- Mullenix PS, Steele SR, Andersen CA, Starnes BW, Salim A, Martin MJ. Limb salvage and outcomes among patients with traumatic popliteal vascular injury: an analysis of the National Trauma Data Bank. J Vasc Surg. 2006;44(1):94–100. doi:10.1016/j.jvs.2006.02.052 [CrossRef] PMID:16828431
- Rozycki GS, Tremblay LN, Feliciano DV, McClelland WB. Blunt vascular trauma in the extremity: diagnosis, management, and outcome. J Trauma. 2003;55(5):814–824. doi:10.1097/01.TA.0000087807.44105.AE [CrossRef] PMID:14608150
- al-Salman MM, al-Khawashki H, Sindigki A, Rabee H, al-Saif A, al-Salman Fachartz F. Vascular injuries associated with limb fractures. Injury. 1997;28(2):103–107. doi:10.1016/S0020-1383(96)00169-6 [CrossRef] PMID:9205574
- MacKenzie EJ, Bosse MJ. Factors influencing outcome following limb-threatening lower limb trauma: lessons learned from the Lower Extremity Assessment Project (LEAP). J Am Acad Orthop Surg. 2006;14(10):S205–S210. doi:10.5435/00124635-200600001-00044 [CrossRef] PMID:17003200
- Chung KC, Shauver MJ, Saddawi-Konefka D, Haase SC. A decision analysis of amputation versus reconstruction for severe open tibial fracture from the physician and patient perspectives. Ann Plast Surg. 2011;66(2):185–191. doi:10.1097/SAP.0b013e3181cbfcce [CrossRef] PMID:20842003
- Huh J, Stinner DJ, Burns TC, Hsu JRLate Amputation Study Team. Infectious complications and soft tissue injury contribute to late amputation after severe lower extremity trauma. J Trauma. 2011;71(1)(suppl):S47–S51. doi:10.1097/TA.0b013e318221181d [CrossRef] PMID:21795878
- O'Toole RV, Castillo RC, Pollak AN, MacKenzie EJ, Bosse MJLEAP Study Group. Determinants of patient satisfaction after severe lower-extremity injuries. J Bone Joint Surg Am. 2008;90(6):1206–1211. doi:10.2106/JBJS.G.00492 [CrossRef] PMID:18519312
- Cannada LK, Jones AL. Demographic, social and economic variables that affect lower extremity injury outcomes. Injury. 2006;37 (12):1109–1116. doi:10.1016/j.injury.2006.07.016 [CrossRef] PMID:17083942
- Gregory RT, Gould RJ, Peclet M, et al. The mangled extremity syndrome (M.E.S.): a severity grading system for multisystem injury of the extremity. J Trauma. 1985;25(12):1147–1150. doi:10.1097/00005373-198512000-00005 [CrossRef] PMID:3934398
- Howe HR Jr, Poole GV Jr, Hansen KJ, et al. Salvage of lower extremities following combined orthopedic and vascular trauma: a predictive salvage index. Am Surg. 1987;53(4):205–208. PMID:3579025
- Johansen K, Daines M, Howey T, Helfet D, Hansen ST Jr, . Objective criteria accurately predict amputation following lower extremity trauma. J Trauma. 1990;30(5):568–572. doi:10.1097/00005373-199005000-00007 [CrossRef] PMID:2342140
- Rajasekaran S. Ganga Hospital open injury severity score: a score to prognosticate limb salvage and outcome measures in type IIIb open tibial fractures. Indian J Orthop. 2005;39(1):4–13.
Characteristics between patients who had amputation and those who had no amputation for open tibia fractures
| Amputation (N=3769) | No amputation (N=171515) | Total (N=175283) | p-value |
|---|
| Age | | | | 0.0058 |
| Mean (SE of Mean) | 43.08 (0.58) | 41.30 (0.11) | 41.34 (0.11) | |
| Age (>50) | | | | 0.0116 |
| No | 2509 (66.6%) | 122051 (71.2%) | 124560 (71.1%) | |
| Yes | 1259 (33.4%) | 49464 (28.8%) | 50723 (28.9%) | |
| Age (>80) | | | | 0.1866 |
| No | 3630 (96.3%) | 166875 (97.3%) | 170505 (97.3%) | |
| Yes | 138 (3.7%) | 4640 (2.7%) | 4778 (2.7%) | |
| Race | | | | 0.0081 |
| Missing | 753 | 38881 | 39634 | |
| White | 2149 (71.3%) | 86024 (64.9%) | 88173 (65.0%) | |
| Black | 366 (12.1%) | 22176 (16.7%) | 22542 (16.6%) | |
| Hispanic | 343 (11.4%) | 17448 (13.2%) | 17792 (13.1%) | |
| Asian | 47 (1.5%) | 1634 (1.2%) | 1681 (1.2%) | |
| Other | 110 (2.9%) | 5351 (3.1%) | 5461 (3.1%) | |
| Sex | | | | <.0001 |
| Missing | 0 | 543 | 543 | |
| Male | 3041 (80.7%) | 126056 (73.7%) | 129097 (73.9%) | |
| Female | 727 (19.3%) | 44917 (26.3%) | 45644 (26.1%) | |
| Medical Co-Morbidities | | | | |
| Acute renal failure | 229 (6.1%) | 2957 (1.7%) | 3186 (1.8%) | <.0001 |
| Alcohol-related disorders | 137 (3.6%) | 8721 (5.1%) | 8858 (5.1%) | 0.0341 |
| Chronic kidney disease | 30 (0.8%) | 1529 (0.9%) | 1559 (0.9%) | 0.7964 |
| Congestive heart failure | 106 (2.8%) | 2650 (1.6%) | 2756 (1.6%) | 0.0449 |
| Diabetes | 351 (9.3%) | 11292 (6.6%) | 11644 (6.6%) | 0.0205 |
| Drug-related disorders | 110 (2.9%) | 5238 (3.1%) | 5348 (3.1%) | 0.8480 |
| HIV | 10 (0.3%) | 655 (0.4%) | 665 (0.4%) | 0.4841 |
| Hypertension | 641 (17.0%) | 27186 (15.9%) | 27827 (15.9%) | 0.4448 |
| Peripheral and visceral atherosclerosis | 99 (2.6%) | 1390 (0.8%) | 1489 (0.9%) | 0.0013 |
| Psychotic disorders | 35 (0.9%) | 902 (0.5%) | 937 (0.5%) | 0.2449 |
| Smoking status | 560 (14.9%) | 32811 (19.1%) | 33371 (19.0%) | 0.0014 |
| Transfer in (from acute care hospital) | | | | 0.3566 |
| No | 1210 (90.3%) | 53109 (92.3%) | 54319 (92.2%) | |
| Yes | 130 (9.7%) | 4443 (7.7%) | 4573 (7.8%) | |
| Fracture location | | | | 0.0003 |
| Proximal | 725 (19.3%) | 38471 (22.4%) | 39197 (22.4%) | |
| Middle | 1763 (46.8%) | 87592 (51.1%) | 89355 (51.0%) | |
| Distal | 1280 (34.0%) | 45451 (26.5%) | 46732 (26.7%) | |
| Neurovascular Injury | | | | <.0001 |
| None | 3237 (85.9%) | 167650 (97.8%) | 170887 (97.5%) | |
| Isolated Tibial Nerve Injury | 40 (1.1%) | 159.85 (0.1%) | 200 (0.1%) | |
| Isolated Arterial Injury | 419 (11.1%) | 3583 (2.1%) | 4002 (2.3%) | |
| Combined Neurovascular Injury | 72 (1.9%) | 122 (0.1%) | 195 (0.1%) | |
| Hospital Location/Teaching | | | | <.0001 |
| Rural | 124 (3.3%) | 13751 (8.0%) | 13875 (7.9%) | |
| Urban nonteaching | 548 (14.5%) | 42208 (24.6%) | 42756 (24.4%) | |
| Urban teaching | 3097 (82.2%) | 115556 (67.4%) | 118652 (67.7%) | |
| Hospital Region | | | | 0.1827 |
| Northeast | 634 (16.8%) | 27871 (16.3%) | 28505 (16.3%) | |
| Midwest | 798 (21.2%) | 36412 (21.2%) | 37210 (21.2%) | |
| South | 1698 (45.1%) | 72313 (42.2%) | 74011 (42.2%) | |
| West | 639 (17.0%) | 34919 (20.4%) | 35558 (20.3%) | |
Variables of interest
| Diagnosis | ICD-9 Code |
|---|
| Open Tibia Fracture | |
| Proximal | 823.10, 823.12 |
| Mid-shaft | 823.30, 823.32 |
| Distal | 823.90, 823.92 |
| Amputation | 84.10, 84.13, 84.14, 84.15 |
| Vascular Injury | 904.41, 904.50, 904.51, 904.53 |
| Nerve Injury | 956.2 |
Amputation rate by fracture location in patients with no neurovascular injury
| Unadjusted Analysis | Adjusted Analysis |
|---|
|
|
|---|
| OR (95% CI) | adj p-value | OR (95% CI) | p-value | adj p-value |
|---|
| Middle vs. Distal | 0.68 (0.57, 0.82) | <0.0001 | 0.66 (0.37, 1.18) | 0.16 | 1 |
| Proximal vs. Distal | 0.61 (0.48, 0.78) | <0.0001 | 0.43 (0.20, 0.94) | 0.03 | 0.27 |
| Proximal vs. Middle | 0.90 (0.71, 1.14) | 1 | 0.65 (0.29, 1.49) | 0.31 | 1 |
Amputation rate by fracture location with neurovascular injury in comparison to no neurovascular injury
| Unadjusted Analysis | Adjusted Analysis |
|---|
|
|
|---|
| OR (95% CI) | p-value | adj p-value | OR (95% CI) | p-value | adj p-value |
|---|
| Distal | 7.30 (4.69, 11.35) | <0.0001 | <0.0001 | 5.45 (1.73, 17.19) | 0.004 | 0.04 |
| | <0.0001 | | | <0.000 | |
| Middle | 8.52 (6.14, 11.82) | | <0.0001 | 12.39 (5.52, 27.83) | 1 | <0.0001 |
| Proximal | 6.26 (3.81, 10.29) | <0.0001 | <0.0001 | 1.93 (0.19, 19.44) | 0.58 | 1 |
Regression results for covariates included in regression model
| OR (95% CI) | p-value |
|---|
| Age | 1.00 (0.99, 1.02) | 0.86 |
| Gender | | |
| Female | 0.78 (0.45, 1.35) | 0.37 |
| Male | Reference | |
| Race | | 0.86 |
| White | Reference | |
| Asian | 1.44 (0.19, 11.12) | |
| Black | 0.78 (0.42, 1.44) | |
| Hispanic | 0.70 (0.29, 1.69) | |
| Other | 0.92 (0.25, 3.40) | |
| Medical Co-Morbidities | | |
| Acute renal failure | 11.69 (4.64, 29.45) | <0.0001 |
| Diabetes | 0.57 (0.13, 2.51) | 0.46 |
| Smoking status | 1.32 (0.61, 2.86) | 0.48 |
| Hospital Location/Teaching | | 0.001 |
| Rural | 0.10 (0.01, 0.75) | |
| Urban nonteaching | 0.33 (0.16, 0.69) | |
| Urban teaching | Reference | |
| Hospital Region | | 0.63 |
| Northeast | 0.86 (0.46, 1.58) | |
| Midwest | 1.33 (0.70, 2.52) | |
| South | Reference | |
| West | 0.88 (0.40, 1.91) | |