Orthopedics

Feature Article Supplemental Data

Effect of Postoperative Nonsteroidal Anti-inflammatory Drugs on Nonunion Rates in Long Bone Fractures

William A. Tucker, MD; Mitchell C. Birt, MD; Archie A. Heddings, MD; Greg A. Horton, MD

Abstract

The association of nonsteroidal anti-inflammatory drugs (NSAIDs) with non-union in long bone fractures has been controversial. The purpose of this study was to evaluate whether NSAID exposure results in increased risk of non-union in operatively treated long bone fractures. The authors used International Classification of Diseases and Current Procedural Terminology codes to identify patients under a single-payer private insurance with operatively treated humeral shaft, tibial shaft, and subtrochanteric femur fractures from a large database. Patients were divided into cohorts based on NSAID use in the immediate postoperative period, and nonunion rates were compared. A cost analysis and a multivariate analysis were performed. Between 2007 and 2016, a total of 5310 tibial shaft, 3947 humeral shaft, and 8432 subtrochanteric femur fractures underwent operative fixation. Patients used NSAIDs in the first 90 days postoperatively in 900 tibial shaft, 694 humeral shaft, and 967 subtrochanteric femur fractures. In these patients, nonunion rates were 18.8%, 17.4%, and 10.4%, respectively. When no NSAIDs were used, the rates were 11.4%, 10.1%, and 4.6% for each fracture type, respectively (P<.05). Among patients taking NSAIDs, subtrochanteric femur fractures had a 2.4 times higher risk of nonunion and humeral shaft and tibial shaft fractures both had a 1.7 times higher risk of nonunion (P<.05). Multivariate analysis showed NSAID use to be an independent risk factor in all 3 types. Cost analysis showed a great increase in economic burden (P<.05). This study indicated that NSAID exposure was associated with fracture nonunion. [Orthopedics. 2020;43(x):xx–xx.]

Abstract

The association of nonsteroidal anti-inflammatory drugs (NSAIDs) with non-union in long bone fractures has been controversial. The purpose of this study was to evaluate whether NSAID exposure results in increased risk of non-union in operatively treated long bone fractures. The authors used International Classification of Diseases and Current Procedural Terminology codes to identify patients under a single-payer private insurance with operatively treated humeral shaft, tibial shaft, and subtrochanteric femur fractures from a large database. Patients were divided into cohorts based on NSAID use in the immediate postoperative period, and nonunion rates were compared. A cost analysis and a multivariate analysis were performed. Between 2007 and 2016, a total of 5310 tibial shaft, 3947 humeral shaft, and 8432 subtrochanteric femur fractures underwent operative fixation. Patients used NSAIDs in the first 90 days postoperatively in 900 tibial shaft, 694 humeral shaft, and 967 subtrochanteric femur fractures. In these patients, nonunion rates were 18.8%, 17.4%, and 10.4%, respectively. When no NSAIDs were used, the rates were 11.4%, 10.1%, and 4.6% for each fracture type, respectively (P<.05). Among patients taking NSAIDs, subtrochanteric femur fractures had a 2.4 times higher risk of nonunion and humeral shaft and tibial shaft fractures both had a 1.7 times higher risk of nonunion (P<.05). Multivariate analysis showed NSAID use to be an independent risk factor in all 3 types. Cost analysis showed a great increase in economic burden (P<.05). This study indicated that NSAID exposure was associated with fracture nonunion. [Orthopedics. 2020;43(x):xx–xx.]

Long bone nonunion results in significant burden for the treating surgeon,1 cost to the health care system,2,3 and morbidity for the patient.4 Improvements in fracture management have prompted a decline in nonunion estimates of 30% since 1993.5 This improvement is multifactorial and can be attributed to advances in implant design, surgical technique, and postoperative management.6–12 Despite these improvements, nonunion continues to occur in 1.9% to 4.9% of all fractures.6,13

The cost of managing non-healing fractures cannot be overstated. Long bone injuries account for 10% of non-fatal injuries and are the number one category of injuries requiring inpatient expenditures.14,15 When nonunion is encountered, the management becomes involved, often requiring repeat procedures.1,7 Few recent studies have evaluated the economics of fracture nonunion. In 2007, Kanakaris and Giannoudis16 reviewed the literature and found an additional cost for humeral, femoral, and tibial nonunion of (pound equivalent) $21,023, $23,245, and $22,069, respectively. Antonova et al2 found that tibial nonunion can lead to more than double the cost of primary fixation ($25,556 vs $11,686). By continuing to modify any perioperative risk factors, surgeons will help to further decrease nonunion rates.

Risk factors for nonunion include type of fracture, method of fixation, and patient comorbidities.6–12,17 Of these, smoking has been well established as a risk factor for nonunion in certain fracture patterns.8,9,18 Exposure to nonsteroidal anti-inflammatory drugs (NSAIDs) has been controversial in its association with subsequent nonunion. Basic science, animal studies, and human clinical trials have all provided variable results regarding NSAID use and fracture healing.19–21 Many independent studies have found NSAID exposure to be associated with fracture nonunion.22–24 In contrast, meta-analyses have been mixed and other reports within the spine literature have revealed no risk.25–29

The purpose of this study was to evaluate the association of NSAID exposure with the risk of nonunion in operatively treated long bone fractures. Secondarily, the authors sought to evaluate the impact of nonunion reimbursement with the aim of providing guidance on NSAID use in the postoperative period. To the authors' knowledge, this is the first large database study to evaluate the risk of NSAID use and subsequent nonunion in long bone fractures treated operatively.

Materials and Methods

Institutional Database

With the use of a large database of patients under the care of one private insurance provider (PearlDiver, Colorado Springs, Colorado), a retrospective cohort study was performed to assess for a relationship between the use of NSAIDs and nonunion rates in patients with operatively treated long bone fractures. The PearlDiver Humana database is a Health Insurance Portability and Accountability Act–compliant national database of more than 23.5 million orthopedic patients from 2007 to 2016 arranged by diagnosis and procedure codes. Because all patient data are deidentified, this study required no approval from an institutional review board. Patients with operatively treated humeral shaft, tibial shaft, and subtrochanteric femur fractures were identified using International Classification of Diseases, 9th and 10th revisions, and Current Procedural Terminology codes (Table A, available in the online version of the article).

Database Codes

Table A.

Database Codes

Nonunion and NSAIDs

The patients were then divided into cohorts based on the use of NSAIDs. The data on NSAIDs were derived from Humana patient claims and originated from National Drug Codes on pharmacy claims. These data were further broken down into drug categories, out of which DRUG-CAT_1012 was used for NSAIDs. Patients who proceeded to nonunion were identified using the codes in Table A. Nonunion rates in NSAID users vs controls were then found. The study also investigated the temporal relationship between NSAID use and nonunion rate. Closed and open fractures were both analyzed.

Charlson Comorbidity Index and Cost

The Charlson Comorbidity Index (CCI) is a validated metric used to identify a patient's overall level of health and mortality risk. Several patient factors, including age, diabetes, renal disease, heart disease, and many other comorbidities, are used to calculate a patient's score. The lower the score, the lower the patient's comorbidity burden.30,31 The PearlDiver software also allows for cost analysis to be performed. The cost for patients who developed a non-union was compared with the cost for those who did not in the first year of treatment.

Statistical Analysis

The relative risks and 95% confidence intervals of NSAID users compared with nonusers were calculated for association with nonunion. An unpaired t test was used to compare CCIs in the union and nonunion groups within the NSAID group and was also used to assess the cost associated with fractures with and without a diagnosis of nonunion. Finally, multivariate analysis was performed using logistic regression within the PearlDiver software to account for patient demographics (age, sex) and comorbidities (renal disease, tobacco use, diabetes, peripheral vascular disease, and obesity). Statistical significance was set at P≤.05 for all calculations.

Results

Fracture and Nonunion

Between 2007 and 2016, a total of 5310 tibial shaft fractures, 3947 humeral shaft fractures, and 8432 subtrochanteric femur fractures underwent operative fixation. Overall, operatively treated tibial shaft, humeral shaft, and subtrochanteric femur fractures had nonunion rates of 12.7%, 11.4%, and 5.3%, respectively (Table 1). Patients used NSAIDs in the first 90 days postoperatively in 900 tibial shaft, 694 humeral shaft, and 967 subtrochanteric femur fractures. In these patients, the rates increased to 18.8% in tibial shaft fractures, 17.4% in humeral shaft fractures, and 10.4% in subtrochanteric femur fractures. When no NSAIDs were used, the rates were 11.4%, 10.1%, and 4.6% for each fracture type, respectively (P<.05). In patients taking NSAIDs, subtrochanteric femur fractures had a 2.4 times higher risk of nonunion when calculating relative risk and both humeral shaft fractures and tibial shaft fractures had a 1.7 times higher risk of nonunion (P<.05).

Fracture and Nonunion Breakdown

Table 1:

Fracture and Nonunion Breakdown

Open Versus Closed Fractures

In open fractures, NSAID users had nonunions in 9 of 41 humeral shaft fractures, 52 of 216 tibial shaft fractures, and 3 of 23 subtrochanteric femur fractures. The rates in those who did not use NSAIDs were 13.5%, 19.1%, and 8.2%, respectively. These values were not statistically significant, possibly due to lower numbers of fractures (P>.05). Closed humeral shaft fractures had a nonunion rate of 17.3% in NSAID users, vs 9.9% in nonusers. This trend continued in tibial shaft (12.9% vs 3.2%) and subtrochanteric femur (10.4% vs 4.5%) fractures (P<.001). Overall, open fractures had nonunion rates of 15.1%, 20.2%, and 8.7% in humeral shaft, tibial shaft, and subtrochanteric femur fractures, respectively. These rates were much lower in closed injuries—11.2%, 4.8%, and 5.2%, respectively.

Intramedullary Nail Versus Open Reduction and Internal Fixation

To control for treatment strategy, fixations of tibia fractures were compared. A total of 3416 tibia fractures were treated with an intramedullary nail, and 1306 underwent open reduction and internal fixation. The intramedullary nail group had an 11.4% nonunion rate, vs 15.39% for open reduction and internal fixation (P<.001). When NSAIDs were taken into account, intramedullary nail patients who took NSAIDs had a nonunion rate of 18.7%, vs 9.9% in those who did not take NSAIDs (P<.001). Similar findings were encountered in the open reduction and internal fixation group. The NSAID users had a nonunion rate of 14.3%, vs 20.6% for nonusers (P<.001).

Effect of Timing and NSAID Use

Additionally, patients were placed into groups depending on when they took NSAIDs postoperatively. Patients who did so within 30 days of surgery were at the highest risk of nonunion when compared those who did so between 30 and 60 days and between 60 and 90 days. All groups were found to have a statistically significant increased risk of nonunion except for patients with humeral shaft fractures who took NSAIDs between 60 and 90 days after surgical fixation (Table 2).

Nonsteroidal Anti-inflammatory Drug Use and Nonunion Ratea

Table 2:

Nonsteroidal Anti-inflammatory Drug Use and Nonunion Rate

Charlson Comorbidity Index

The CCI was used to assess overall comorbidity burden within the NSAID groups. For patients who used NSAIDs, no statistical significance was found in any of the fracture groups, meaning that the patients likely had a similar comorbidity burden. In all fracture groups for NSAID-naïve patients, the CCI was either the same as or even lower (humeral shaft nonunion) than that of the patients who went on to nonunion. Therefore, their comorbidity burden was similar to or even greater than that of patients with nonunions (Table 3).

Charlson Comorbidity Index in Union Versus Nonunion

Table 3:

Charlson Comorbidity Index in Union Versus Nonunion

Multivariate Analysis

In a further effort to ensure that the difference in fracture nonunion was not due to a confounding variable, multivariate analysis was performed using a logistic regression to control for comorbidities and demographics while assessing the impact of NSAID use. Odds ratios showed NSAIDs to be an independent risk factor for nonunion in all 3 groups, even when accounting for these other factors. Subtrochanteric femur fractures had an odds ratio of 1.50. Additionally, tibial shaft fractures were at increased risk of nonunion with NSAIDs (odds ratio, 1.42). Humeral shaft fractures had the lowest risk (odds ratio, 1.20). Infection was evaluated as well and had almost a 2-fold increased risk in all 3 fractures. Of note, tobacco use caused increased risk of nonunion in humeral shaft and tibial shaft fractures but not subtrochanteric femur fractures. Diabetes mellitus did not cause an increased risk in any of the 3 (Table 4).

Multivariate Logistic Regression Analysis of Risk Factors for Fracture Nonunion

Table 4:

Multivariate Logistic Regression Analysis of Risk Factors for Fracture Nonunion

Cost Analysis

Results showed a large difference in cost for patients who went on to nonunion compared with those who did not. As seen in Figure 1, the average 1-year reimbursement associated with operatively treated subtrochanteric femur fractures was $22,435.05. In comparison, the reimbursement for patients with a diagnosis of nonunion with the same fracture was $35,968.19 (P<.001). Similar results were found for tibia fractures. Nonunion resulted in a reimbursement of $49,995.25, vs $28,077.04 for those without that diagnosis (P<.001). Finally, humeral shaft nonunions cost $31,746.91 at 1 year, vs $21,240.30 for humeral shaft fractures without nonunion (P<.001).

Cost (in dollars) of union vs nonunion.

Figure 1:

Cost (in dollars) of union vs nonunion.

Discussion

In the current health care environment and with the goal of minimizing opiate analgesia during fracture management, NSAIDs play a key role in the control of patients' pain.32,33 However, these medications must not be used at the expense of fracture healing. Using the PearlDiver database, the authors identified 5310 tibial shaft fractures, 3947 humeral shaft fractures, and 8432 subtrochanteric femur fractures that underwent operative fixation. They chose fractures of the humerus, tibia, and subtrochanteric femur for analysis due to their increased incidence of nonunion and evaluation in prior literature.12,17,34

These data indicate that exposure to NSAIDs at any point within 90 days postoperatively is associated with subsequent fracture nonunion. Most importantly, the current study found an association between NSAID exposure during the immediate postoperative period (0 to 30 days) and subsequent nonunion. Significant risk of nonunion developed in all analyzed fractures: humeral shaft (19.5% vs 10.1%), subtrochanteric femur (11.2% vs 4.6%), and tibial shaft (20.0% vs 11.4%). Overall, NSAID exposure carried a relative risk of nonunion of 2.4 for subtrochanteric femur fractures, 1.7 for tibial shaft fractures, and 1.7 for humeral shaft fractures. The relative risk was highest 0 to 30 days following surgery—2.4, 2.0, and 1.9 for femur, tibia, and humerus fractures, respectively.

These results support prior literature, which found a risk with NSAID use. Jeffcoach et al34 retrospectively reviewed 1901 patients with long bone fractures at a single institution. Of these, 60 patients developed long bone nonunion. Postoperative use of NSAIDs resulted in an odds ratio of 2.17 for nonunion, and smokers were 3.19 times as likely to develop nonunion. However, this is in contrast to a 2005 database study by Bhattacharyya et al35 evaluating humeral shaft fracture. They found that opioid and NSAID exposure was associated with nonunion only during the 61 to 90 days period. They concluded that this was more likely associated with pain control of non-union than a causal relationship resulting in a protopathic bias.

As discussed previously, studies regarding NSAID use in spinal fusion procedures have tended to report no inhibitory effect and this may be the case, at least in the pediatric population.25,26 A 2010 meta-analysis by Dodwell et al22 analyzing 11 studies demonstrated an association between NSAID use and nonunion (odds ratio, 3.0). When only higher level studies were used, 7 spine fusion studies found no association.

Multivariate analysis was useful for assessing the effects of NSAIDs in fracture healing and taking into account possible confounders. In both subtrochanteric femur and tibia fractures, NSAIDs had a higher risk for nonunion than did renal disease, vascular disease, tobacco use, obesity, or diabetes. Only tobacco use and renal disease represented higher risks in humeral shaft fractures. Tobacco use did still pose significant risk for nonunion in humeral and tibia fractures. However, tobacco use was not found to be a risk factor in subtrochanteric fractures, which is consistent with previous literature.36 Interestingly, diabetes was not found to be a significant risk factor for any of the 3 fracture patterns, which does contrast with prior literature (Table 4).6

The nonunion rates found in the current analysis are consistent with those reported in prior studies. These data indicated total tibial shaft fracture nonunion of 12.7%, humeral shaft fracture nonunion of 11.4%, and subtrochanteric femur fracture nonunion of 5.3%. Rates were higher in open fractures than in closed fractures. Although the lack of a universally accepted definition of nonunion can affect outcome reporting,10,37 the current results are in accordance with prior literature.12 An evidence review by Tzioupis and Giannoudis12 found nonunion rates ranging from 2% to 13% for nonoperative humerus fractures and 2% to 7% for those treated with plate fixation. Intramedullary nailing had higher nonunion rates, ranging from 0% to 33%. Subtrochanteric fractures trended toward nonunion rates up to 5%. The nonunion rate for tibial fractures was 5% to 6.5% when using reamed nailing.12

In the subanalysis of cost, the data demonstrated that cost was substantially increased for all nonunion management, as expected and in accordance with prior studies.2,16 The current authors found that non-union of tibial shaft fractures had the greatest impact on cost, with standard healing having a reimbursement of $28,077.04, vs $49,995.25 with an associated nonunion. Significant cost increases were found in each of the other fractures as well.

This study had some limitations. First, this was a retrospective review of a database, which has inherent weaknesses and is dependent on International Classification of Diseases and Current Procedural Terminology coding. Studies have shown that databases, when used appropriately, can contribute to orthopedic research.38–40 The PearlDiver database has been used in numerous studies published in high impact journals.41–44 Databases do have a tendency to underreport complications.45 Individual patient factors could not be isolated and compared. Additionally, this study could have underestimated the cost of nonunion because the treating surgeon may have miscoded subsequent treatment. Although the period of NSAID use could be isolated, no determination was made regarding the specific type of NSAID or the exact duration, frequency, or dosages to establish a dose-dependent relationship. Finally, as data on NSAIDs were derived from Humana patient claims and originated from coding on pharmacy claims, over-the-counter NSAID use was not reported within the database.

Strengths of this study included its longitudinal nature, which allowed for patients to be followed over time and over a large number of providers. Additionally, the volume of patients improved the power of the study. The PearlDiver database has the ability to perform multivariate regression analysis, which can isolate, for instance, the impact of NSAIDs on nonunion when compared with many other risk factors. The use of the CCI also helped account for comorbidity burden.

Conclusion

This study indicated that NSAID exposure in the postoperative period was associated with fracture nonunion, given cross-sectional analysis of long bone fracture and nonunion across the United States. It provided a cost analysis comparing the 1-year reimbursement of the tested fractures with and without subsequent nonunion.

References

  1. Egol KA, Bechtel C, Spitzer AB, Rybak L, Walsh M, Davidovitch R. Treatment of long bone nonunions: factors affecting healing. Bull NYU Hosp Jt Dis. 2012;70(4):224–231. PMID:23267445
  2. Antonova E, Le TK, Burge R, Mershon J. Tibia shaft fractures: costly burden of nonunions. BMC Musculoskelet Disord. 2013;14(1):42. doi:10.1186/1471-2474-14-42 [CrossRef] PMID:23351958
  3. Dahabreh Z, Calori GM, Kanakaris NK, Nikolaou VS, Giannoudis PV. A cost analysis of treatment of tibial fracture nonunion by bone grafting or bone morphogenetic protein-7. Int Orthop. 2009;33(5):1407–1414. doi:10.1007/s00264-008-0709-6 [CrossRef] PMID:19052743
  4. Babhulkar S, Pande K, Babhulkar S. Non-union of the diaphysis of long bones. Clin Orthop Relat Res. 2005;(431):50–56. doi:10.1097/01.blo.0000152369.99312.c5 [CrossRef] PMID:15685055
  5. Wang Z, Bhattacharyya T. Trends of non-union and prescriptions for non-steroidal anti-inflammatory drugs in the United States, 1993–2012. Acta Orthop. 2015;86(5):632–637. doi:10.3109/17453674.2015.1028860 [CrossRef] PMID:25761793
  6. Zura R, Xiong Z, Einhorn T, et al. Epidemiology of fracture nonunion in 18 human bones. JAMA Surg. 2016;151(11):e162775. doi:10.1001/jamasurg.2016.2775 [CrossRef] PMID:27603155
  7. de Vries JS, Kloen P, Borens O, Marti RK, Helfet DL. Treatment of subtrochanteric non-unions. Injury. 2006;37(2):203–211. doi:10.1016/j.injury.2005.09.017 [CrossRef] PMID:16417905
  8. Castillo RC, Bosse MJ, MacKenzie EJ, Patterson BMLEAP Study Group. Impact of smoking on fracture healing and risk of complications in limb-threatening open tibia fractures. J Orthop Trauma. 2005;19(3):151–157. doi:10.1097/00005131-200503000-00001 [CrossRef] PMID:15758667
  9. Harvey EJ, Agel J, Selznick HS, Chapman JR, Henley MB. Deleterious effect of smoking on healing of open tibia-shaft fractures. Am J Orthop (Belle Mead NJ).2002;31(9):518–521. PMID:12650537
  10. Schemitsch EH, Bhandari M, Guyatt G, et al. Study to Prospectively Evaluate Reamed Intramedullary Nails in Patients with Tibial Fractures (SPRINT) Investigators. Prognostic factors for predicting outcomes after intramedullary nailing of the tibia. J Bone Joint Surg Am. 2012;94(19):1786–1793. doi:10.2106/JBJS.J.01418 [CrossRef] PMID:23032589
  11. Calori GM, Albisetti W, Agus A, Iori S, Tagliabue L. Risk factors contributing to fracture non-unions. Injury. 2007;38(suppl 2):S11–S18. doi:10.1016/S0020-1383(07)80004-0 [CrossRef] PMID:17920412
  12. Tzioupis C, Giannoudis PV. Prevalence of long-bone non-unions. Injury. 2007;38 (suppl 2):S3–S9. doi:10.1016/S0020-1383(07)80003-9 [CrossRef] PMID:17920415
  13. Mills LA, Aitken SA, Simpson AHRW. The risk of non-union per fracture: current myths and revised figures from a population of over 4 million adults. Acta Orthop. 2017;88(4):434–439. doi:10.1080/17453674.2017.1321351 [CrossRef] PMID:28508682
  14. Miller TR, Lestina DC. Patterns in US medical expenditures and utilization for injury, 1987. Am J Public Health. 1996;86(1):89–93. doi:10.2105/AJPH.86.1.89 [CrossRef] PMID:8561252
  15. Vyrostek SB, Annest JL, Ryan GW. Surveillance for fatal and nonfatal injuries: United States, 2001. MMWR Surveill Summ. 2004;53(7):1–57. PMID:15343143
  16. Kanakaris NK, Giannoudis PV. The health economics of the treatment of long-bone non-unions. Injury. 2007;38(suppl 2):S77–S84. doi:10.1016/S0020-1383(07)80012-X [CrossRef] PMID:17920421
  17. Ring D, Barrick WT, Jupiter JB. Recalcitrant nonunion. Clin Orthop Relat Res. 1997;340:181–189. doi:10.1097/00003086-199707000-00023 [CrossRef] PMID:9224254
  18. Schmitz MA, Finnegan M, Natarajan R, Champine J. Effect of smoking on tibial shaft fracture healing. Clin Orthop Relat Res. 1999;365:184–200. doi:10.1097/00003086-199908000-00024 [CrossRef] PMID:10627703
  19. Ho ML, Chang JK, Chuang LY, Hsu HK, Wang GJ. Effects of nonsteroidal anti-inflammatory drugs and prostaglandins on osteoblastic functions. Biochem Pharmacol. 1999;58(6):983–990. doi:10.1016/S0006-2952(99)00186-0 [CrossRef] PMID:10509750
  20. Evans CE, Butcher C. The influence on human osteoblasts in vitro of non-steroidal anti-inflammatory drugs which act on different cyclooxygenase enzymes. J Bone Joint Surg Br. 2004;86(3):444–449. doi:10.1302/0301-620X.86B3.14592 [CrossRef] PMID:15125136
  21. Pountos I, Georgouli T, Calori GM, Giannoudis PV. Do nonsteroidal anti-inflammatory drugs affect bone healing? A critical analysis. Scientific World Journal. 2012;2012:606404. doi:10.1100/2012/606404 [CrossRef] PMID:22272177
  22. Dodwell ER, Latorre JG, Parisini E, et al. NSAID exposure and risk of nonunion: a meta-analysis of case-control and cohort studies. Calcif Tissue Int. 2010;87(3):193–202. doi:10.1007/s00223-010-9379-7 [CrossRef] PMID:20552333
  23. Giannoudis PV, MacDonald DA, Matthews SJ, Smith RM, Furlong AJ, De Boer P. Nonunion of the femoral diaphysis: the influence of reaming and non-steroidal anti-inflammatory drugs. J Bone Joint Surg Br. 2000;82(5):655–658. doi:10.1302/0301-620X.82B5.0820655 [CrossRef] PMID:10963160
  24. Burd TA, Hughes MS, Anglen JO. Heterotopic ossification prophylaxis with indomethacin increases the risk of long-bone nonunion. J Bone Joint Surg Br. 2003;85(5):700–705. doi:10.1302/0301-620X.85B5.13970 [CrossRef] PMID:12892193
  25. Sucato DJ, Lovejoy JF, Agrawal S, Elerson E, Nelson T, McClung A. Postoperative ketorolac does not predispose to pseudoarthrosis following posterior spinal fusion and instrumentation for adolescent idiopathic scoliosis. Spine. 2008;33(10):1119–1124. doi:10.1097/BRS.0b013e31816f6a2a [CrossRef] PMID:18449047
  26. Vitale MG, Choe JC, Hwang MW, et al. Use of ketorolac tromethamine in children undergoing scoliosis surgery: an analysis of complications. Spine J. 2003;3(1):55–62. doi:10.1016/S1529-9430(02)00446-1 [CrossRef] PMID:14589246
  27. Kay RM, Directo MP, Leathers M, Myung K, Skaggs DL. Complications of ketorolac use in children undergoing operative fracture care. J Pediatr Orthop. 2010;30(7):655–658. doi:10.1097/BPO.0b013e3181efb8b4 [CrossRef] PMID:20864848
  28. Horn PL, Wrona S, Beebe AC, Klamar JE. A retrospective quality improvement study of ketorolac use following spinal fusion in pediatric patients. Orthop Nurs. 2010;29(5):342–343. doi:10.1097/NOR.0b013e3181edd876 [CrossRef] PMID:20856090
  29. Chen MR, Dragoo JL. The effect of nonsteroidal anti-inflammatory drugs on tissue healing. Knee Surg Sports Traumatol Arthrosc. 2013;21(3):540–549. doi:10.1007/s00167-012-2095-2 [CrossRef] PMID:22744434
  30. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245–1251. doi:10.1016/0895-4356(94)90129-5 [CrossRef] PMID:7722560
  31. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383. doi:10.1016/0021-9681(87)90171-8 [CrossRef] PMID:3558716
  32. Drendel AL, Gorelick MH, Weisman SJ, Lyon R, Brousseau DC, Kim MK. A randomized clinical trial of ibuprofen versus acetaminophen with codeine for acute pediatric arm fracture pain. Ann Emerg Med. 2009;54(4):553–560. doi:10.1016/j.annemergmed.2009.06.005 [CrossRef] PMID:19692147
  33. Ong CK, Lirk P, Tan CH, Seymour RA. An evidence-based update on nonsteroidal anti-inflammatory drugs. Clin Med Res. 2007;5(1):19–34. doi:10.3121/cmr.2007.698 [CrossRef] PMID:17456832
  34. Jeffcoach DR, Sams VG, Lawson CM, et al. University of Tennessee Medical Center, Department of Surgery. Nonsteroidal anti-inflammatory drugs' impact on nonunion and infection rates in long-bone fractures. J Trauma Acute Care Surg. 2014;76(3):779–783. doi:10.1097/TA.0b013e3182aafe0d [CrossRef] PMID:24553548
  35. Bhattacharyya T, Levin R, Vrahas MS, Solomon DH. Nonsteroidal antiinflammatory drugs and nonunion of humeral shaft fractures. Arthritis Rheum. 2005;53(3): 364–367. doi:10.1002/art.21170 [CrossRef] PMID:15934108
  36. Park SH, Kong GM, Ha BH, Park JH, Kim KH. Nonunion of subtrochanteric fractures: comminution or malreduction. Pak J Med Sci. 2016;32(3):591–594. doi:10.12669/pjms.323.9897 [CrossRef] PMID:27375695
  37. Morshed S. Current options for determining fracture union. Adv Med. 2014;2014:708574. doi:10.1155/2014/708574 [CrossRef] PMID:26556422
  38. Patel AA, Singh K, Nunley RM, Minhas SV. Administrative databases in orthopaedic research: pearls and pitfalls of big data. J Am Acad Orthop Surg. 2016;24(3):172–179. doi:10.5435/JAAOS-D-13-00009 [CrossRef] PMID:26836377
  39. Pugely AJ, Martin CT, Harwood J, Ong KL, Bozic KJ, Callaghan JJ. Database and registry research in orthopaedic surgery: Part I. Claims-based data. J Bone Joint Surg Am. 2015;97(15):1278–1287. doi:10.2106/JBJS.N.01260 [CrossRef] PMID:26246263
  40. Alluri RK, Leland H, Heckmann N. Surgical research using national databases. Ann Transl Med. 2016;4(20):393. doi:10.21037/atm.2016.10.49 [CrossRef] PMID:27867945
  41. Virk SS, Phillips FM, Khan SN. Reimbursement related to a 90-day episode of care for a one or two-level anterior cervical discectomy and fusion. J Bone Joint Surg Am. 2016;98(16):1378–1384. doi:10.2106/JBJS.15.01169 [CrossRef] PMID:27535440
  42. Rosas S, Sabeh KG, Buller LT, Law TY, Kalandiak SP, Levy JC. Comorbidity effects on shoulder arthroplasty costs analysis of a nationwide private payer insurance data set. J Shoulder Elbow Surg. 2017;26(7):e216–e221. doi:10.1016/j.jse.2016.11.044 [CrossRef] PMID:28139384
  43. Qureshi R, Puvanesarajah V, Jain A, Shimer AL, Shen FH, Hassanzadeh H. A comparison of anterior and posterior lumbar interbody fusions: complications, re-admissions, discharge dispositions, and costs. Spine. 2017;42(24):1865–1870. doi:10.1097/BRS.0000000000002248 [CrossRef] PMID:28549000
  44. Erickson BJ, Nwachukwu BU, Rosas S, et al. Trends in medial ulnar collateral ligament reconstruction in the United States: a retrospective review of a large private-payer database from 2007 to 2011. Am J Sports Med. 2015;43(7):1770–1774. doi:10.1177/0363546515580304 [CrossRef] PMID:26129959
  45. Bozic KJ, Bashyal RK, Anthony SG, Chiu V, Shulman B, Rubash HE. Is administratively coded comorbidity and complication data in total joint arthroplasty valid?Clin Orthop Relat Res.2013;471(1):201–205. doi:10.1007/s11999-012-2352-1 [CrossRef] PMID:22528384

Fracture and Nonunion Breakdown

CharacteristicSFFTSFHSF



No. (N=8432)No. of Nonunions (N=448)Nonunion Percent (5.3%)No. (N=5310)No. of Nonunions (N=674)Nonunion Percent (12.7%)No. (N=3947)No. of Nonunions (N=452)Nonunion Percent (11.4%)
Year
  2007416174.1333319.3201189.0
  2008597233.94244911.62613111.9
  2009608304.93976817.12933913.3
  2010760486.34405813.23763910.4
  2011859485.65056713.33984411.1
  2012920495.35218516.34174611.0
  20131022605.96107312.04355111.7
  20141300836.476310013.16188012.9
  20151272745.889711212.56436910.7
  2016678162.4420317.43053511.5
Age, y
  <658309611.6286339213.7107315614.5
  ≥6576023524.6244728211.5287429610.3
Region
  Midwest21291245.8132414711.193010511.3
  Northeast243124.989910.190910.0
  South52282584.9338044713.2257130311.8
  West832546.55177113.7356359.8
Sex
  Female59073135.3269929310.9277331611.4
  Male25251355.3261138114.6117413611.6

Nonsteroidal Anti-inflammatory Drug Use and Nonunion Ratea

NSAID Use PODSFFTSFHSF



Nonunion PercentRelative Risk (CI)Nonunion PercentRelative Risk (CI)Nonunion PercentRelative Risk (CI)
0–3011.22.4 (1.8–3.3)20.02.0 (1.6–2.4)19.51.9 (1.5–2.4)
30–609.92.1 (1.5–3.0)18.51.8 (1.4–2.4)16.01.6 (1.1–2.2)
60–908.61.9 (1.2–2.8)15.91.6 (1.1–2.2)13.81.4 (0.9–2.2)b
0–9010.42.4 (1.9–2.9)18.81.6 (1.4–1.9)17.41.7 (1.4–2.1)

Charlson Comorbidity Index in Union Versus Nonunion

Fracture Type/NSAID UseCharlson Comorbidity IndexPa

NonunionUnion
Subtrochanteric femur
  NSAID use3.23.7.16
  No NSAID use3.43.7.06
Tibial shaft
  NSAID use2.42.3.59
  No NSAID use2.12.2.51
Humeral shaft
  NSAID use2.73.1.11
  No NSAID use2.83.5.01

Multivariate Logistic Regression Analysis of Risk Factors for Fracture Nonunion

Risk FactorSFF Nonunion RiskTSF Nonunion RiskHSF Nonunion Risk



OR (95% CI)PaOR (95% CI)PaOR (95% CI)Pa
NSAIDs1.50 (1.24–1.84)<.0011.42 (1.19–1.69)<.0011.20 (1.00–1.46).05
ESRD1.14 (0.88–1.45)>.10.90 (0.68–1.18)>.11.52 (1.16–2.00).01
Tobacco1.04 (0.86–1.26)>.11.45 (1.23–1.70)<.0011.37 (1.14–1.65)<.001
PVD1.33 (1.10–1.62)<.011.37 (1.12–1.67)<.0010.97 (0.78–1.19)>.1
Obesity1.29 (1.05–1.58)<.051.22 (1.02–1.45)<.0011.28 (1.05–1.55).05
DM1.00 (0.82–1.21)>.10.89 (0.74–1.07)>.10.85 (0.70–1.03).1
Infection2.09 (1.63–2.68)<.0012.67 (2.20–3.23)<.0011.93 (1.45–2.54)<.001

Database Codes

Subtrochanteric Femur FractureICD-9 CODES: 820.22, 820.32 ICD-10 CODES: S72.21XA, S72.21XC, S72.22XA, S72.22XB, S72.23XA, S72.23XB, S72.23XC, S72.24XA, S72.25XA, S72.26XA
Tibial Shaft FractureICD-9 CODES: 823.20, 823.22, 823.30, 823.32 ICD-10 CODES: S82.201A, S82.201B, S82.201C, S82.202A, S82.202B, S82.202C, S82.209A, S82.209B, S82.209C, S82.221A, S82.221B, S82.221C, S82.222A, S82.222B, S82.222C, S82.223A, S82.224A, S82.225A, S82.231A, S82.231B, S82.231C, S82.232A, S82.232B, S82.233A, S82.234A, S82.235A, S82.241A, S82.241B, S82.242A, S82.242B, S82.242C, S82.243A, S82.244A, S82.245A, S82.245B, S82.246A, S82.251A, S82.251B, S82.251C, S82.252A, S82.252B, S82.252C, S82.253A, S82.254A, S82.254B, S82.255A, S82.261A, S82.261B, S82.261C, S82.262A, S82.262B, S82.264A, S82.265A, S82.291A, S82.291B, S82.291C, S82.292A, S82.292B, S82.299A
Humeral shaft FractureICD-9 CODES: 812.21, 812.31 ICD-10 CODES: S42.301A, S42.301B, S42.302A, S42.302B, S42.309A, S42.309B, S42.311A, S42.312A, S42.321A, S42.321B, S42.322A, S42.323A, S42.324A, S42.325A, S42.331A, S42.332A, S42.333A, S42.334A, S42.335A, S42.341A, S42.342A, S42.343A, S42.344A, S42.345A, S42.351A, S42.351B, S42.352A, S42.352B, S42.353A, S42.354A, S42.355A, S42.355B, S42.356A, S42.361A, S42.362A, S42.363A, S42.364A, S42.365A, S42.366A, S42.391A, S42.391B, S42.392A, S42.392B, S42.399A, S42.399B
Subtrochanteric Femur Fracture FixationICD-9-CODES: 79.15, 79.35 ICD-10-CODES: 0QH606Z, 0QH636Z, 0QH646Z, 0QH706Z, 0QH736Z, 0QH746Z, 0QH634Z, 0QH644Z, 0QH704Z, 0QH734Z, 0QH744Z CPT CODES: 27244, 27245
Tibial Shaft Fracture FixationICD-9-CODES: 79.16, 79.36 ICD-10-CODES: 0QHG04Z, 0QHG06Z, 0QHG34Z, 0QHG36Z, 0QHH04Z, 0QHH06Z, 0QHH34Z, 0QHH36Z, 0QHH46Z, CPT-27756 CPT CODES: 27758, 27759
Humeral Shaft Fracture FixationICD-9-CODES: 79.11, 79.31 ICD-10-CODES: 0PHF04Z, 0PHF06Z, 0PHF34Z, 0PHF36Z, 0PHF46Z, 0PHG04Z, 0PHG06Z, 0PHG34Z, 0PHG36Z CPT CODES: 24515, 24516
Subtrochanteric Femur NonunionICD-9 CODES: 73382 ICD-10 CODES: S72.21XK, S72.22XK, S72.22XM, S72.23XK, S72.24XK, S72.25XK
Tibial Shaft NonunionICD-9 CODES: 73382 ICD-10 CODES: S82.201K, S82.201M, S82.201N, S82.202K, S82.202M, S82.202N, S82.209K, S82.221K, S82.221M, S82.222K, S82.222M, S82.224K, S82.225K, S82.231K, S82.231M, S82.231N, S82.232K, S82.235K, S82.241K, S82.242K, S82.245K, S82.251K, S82.251M, S82.251N, S82.252K, S82.261K, S82.261M, S82.261N, S82.264K, S82.291K, S82.291M, S82.292K
Humeral Shaft NonunionICD-9 CODES: 73382 ICD-10 CODES: S42.301K, S42.302K, S42.309K, S42.321K, S42.322K, S42.325K, S42.331K, S42.332K, S42.341K, S42.342K, S42.351K, S42.352K, S42.361K, S42.364K, S42.391K, S42.392K, S42.399K
Authors

The authors are from the Department of Orthopedic Surgery, University of Kansas Medical Center, Kansas City, Kansas.

The authors have no relevant financial relationships to disclose.

Correspondence should be addressed to: William A. Tucker, MD, 621 E 71st Terrace, Kansas City, MO 64131 ( williamatucker88@gmail.com).

Received: November 17, 2019
Accepted: March 03, 2020
Posted Online: May 05, 2020

10.3928/01477447-20200428-06

Sign up to receive

Journal E-contents