1 Introduction

Fractures of the distal radius account for roughly 18 % of adult fractures, with an average age at the time of injury of 55.5 years old.¹ Of these fractures, approximately 36 % ultimately go on to operative management while the rest are managed conservatively with cast/splint immobilization.² With the inherent bimodal age distribution of distal radius fractures (DRFs), current literature has repeatedly studied the clinical and patient reported outcomes of open reduction and internal fixation (ORIF) for DRFs based on patient age.

There are many well-accepted risk factors for postoperative complications and readmissions following ORIF of DRFs, including smoking, obesity, and diabetes.^3,4 Further, some reports have suggested that age >65 years may be a stand-alone risk factor for postoperative complications following upper extremity surgery.⁵ However, specific to DRFs, significant heterogeneity remains on the impact of age alone on short-term complications. For example, Hinds et al. reported higher rates of short-term complications (≤7 months) in patients over the age of 65 following surgical fixation of DRFs, including anesthetic complications, iatrogenic bone fracture, implant failure, loss of reduction, unspecified medical complications and death.⁶ In contrast, Mosenthal et al. found an inverse relationship between age and 12-month postoperative complication rates.⁷ Further, both Wilson et al. and Skochdopole et al. showed that while age >65 years was associated with a higher rate of 30-day postoperative complications, it was not an independent predictor after controlling for other factors.^8,9

With current literature not coming to a consensus, the objective of this study was to identify the impact of patient age on short-term outcomes following ORIF for DRFs using a sample of 32,679 patients. It was hypothesized that increasing age would be correlated with increased overall total charges, length of stay (LOS), complications, and 30- and 90-day readmission rates in individuals undergoing ORIF for DRFs.

2 Materials and methods

Data was collected from the Healthcare Cost and Utilization Project National Readmissions Database (NRD), and therefore institutional review board approval was not required. Patients included in this analysis were those who underwent an ORIF for DRFs, as identified by the International Classification of Disease Tenth Revision-Clinical Modification (ICD-10) codes 0PSH04Z (right) and 0PSJ04Z (left). Patients under the age of 18 years old or whose discharge date was after September of each year were excluded from the study; the September cutoff was implemented to avoid missed 90-day readmissions.

For each patient, demographic, injury related complications, operative complications, medical complications, and readmission information was collected. Demographic information included age at surgery, sex, insurance type, Charlson Comorbidity Index, discharge status, LOS, and overall total charges. Complications included concerns with the fracture itself, hardware, infection, nerve injury (i.e., median nerve), wound concerns, intraoperative complications, and all cause readmissions. Infection and wound related complications were further grouped together; “infection complication” included fixation infection, superficial infection, deep infection, and surgical site infection while “wound complication” referred to external/internal wound complications and unspecified wound complications. Readmissions were reported at both the 30- and 90-day marks. “Intraoperative complications” were defined as any complications of the musculoskeletal system (ICD-10 codes: M96).

Patients were stratified into five cohorts based on their age at the time of surgery, including: 18–49, 50–59, 60–69, 70–79, and ≥80 years old. A sub-analysis of those patients <65 and ≥ 65 years old was also conducted. The SciPy version 1.5.4 computer program and Statsmodels version 0.12.2 were used to conduct statistical analysis. Inferential statistics were used throughout. Categorical variables were compared using a Fisher's exact test or a X² test while continuous variables were compared using either an ANOVA or student's t-test. A multivariate logistic regression was conducted with age as the independent predictor; a second regression was run with the independent variable being ≥65 years of age. Variables controlled for in the multivariate model included age, sex, and Charlson Comorbidity Index. When less than eleven patients were identified, they were not included in the analysis in accordance with the NRD guidelines. Further, a linear regression was used for LOS. Statistical significance was set at p < 0.05.

3 Results

A total of 32,679 patients were identified. 12,766, 5,113, 5,639, 4,901, and 4260 patients were identified in the 18–49, 50–59, 60–69, 70–79, and 80+ years old cohorts, respectively, while 20,674 and 12,005 patients were <65 and ≥ 65 years old, respectively. Younger age groups were more likely to be male, have lower Charlson comorbidity index scores, experience longer LOS, incur higher total charges, and have a routine discharge (p < 0.01). These same endpoints were seen based on the age cut off of 65 years old. For example, when comparing patients <65 years old vs. ≥65 years old, they were more likely to be male (60.1 % vs. 17.6 %; p < 0.01), have longer LOS (7.6 vs. 5.6 days; p < 0.01), accrue more cost ($150,636.73 vs. $99,410.67; p < 0.01), etcetera. For a complete breakdown of demographic information with comparisons see Tables I and II.

With regards to complications, age cohorts differed significantly in fixation infection, median nerve injury, 30-day readmissions, and 90-day readmissions (p < 0.01). There was a trend towards unspecified wound complications also differing that did not reach significance (p = 0.063). A breakdown of the comparisons can be seen in Table III. These same findings held true in the 65 years cut off cohorts. Here, those <65 years of age at the time of surgery saw higher fixation infection rates (0.3 % vs. 0.1 %; p < 0.01) and nerve injury rates (5.1 % v. 3.8 %; p < 0.01). While the trend for unspecified wound complications was lost, there existed one for external wound complications (p = 0.097). Patients <65 years old had higher rates of internal wound complications (0.2 % vs. 0.1 %; p = 0.049); however, given that the sample size was <11 for the ≥65 years old cohort, conclusions cannot be made based on this data. Readmission rates were higher for those ≥65 years old at both 30-days (5.2 % vs. 7.0 %; p < 0.01) and 90-days (8.6 % vs. 13.3 %; p < 0.01). For complete results using the 65 years old cut off see Table IV.

In the multivariate logistic regression, age was found to be an independent predictor of multiple complications, both alone and with age ≥65 years old as the independent variable. For age ≥65 years old, these included fixation infections (OR: 0.52, CI: 0.29–0.93; p = 0.028), median nerve injury (OR: 0.77, CI: 0.68–0.87; p < 0.01), 30-day readmissions (OR: 1.19, CI: 1.07–1.33; p < 0.01), and 90-day readmissions (OR: 1.32, CI: 1.21–1.43; p < 0.01). Additionally, age <65 years old was an independent predictor of the pulled variables infection and wound complications but not intraoperative complications. See Table V for the complete multivariate results. For the multivariate logistic regression results with age as the independent variable please also see Table V. With regards to LOS, the OR was −1.59 [CI: 1.85 to −1.31] (p < 0.01).

4 Discussion

ORIF of the distal radius is a common procedure with generally good outcomes and relatively low complication rates. Unplanned 30-day readmission rates have been reported as low as 1.8 % using the National Surgical Quality Improvement Program (NSQIP) database.³ However, with an intrinsic bimodal age distribution, it is important to characterize risk factors associated with complications following ORIF for DRFs based on age. This point becomes increasingly prudent when considering the current literary landscape, where there have been inconsistencies in the data regarding the impact age has on outcomes for surgically corrected DRFs. Moreover, randomized clinical trials have even suggested that patient reported and pain outcomes are comparable in DRFs that are managed operatively and non-operatively in patients ≥60 years old.¹⁰ With 32,679 patients identified here, the present study represents one of the largest and most up-to-date database analyses of short-term complications and readmissions following ORIF for DRFs, and is the first to our knowledge to utilize the NRD.

The aforementioned bimodal age distribution was displayed appropriately in this study, with roughly 55 % of patients being <60 years old and 45 % being >60 years old, which is in accordance with the established literature.^11,12 Moreover, younger patients were more likely to be male while older patients were likely to be female. This finding agrees with the expected high energy mechanisms of injury seen in younger patients, who are more likely to be males, and fragility fractures in older patients, most often osteoporotic females. Thus, the cohorts analyzed here act as a good representation of expected population distributions for those who sustain DRFs.

Previous literature has demonstrated that higher energy injuries are associated with increased total hospital charges, related to factors such as polytrauma and open fractures.^13,14 Assuming that the younger male-predominant population in this study presented primarily with trauma-related injuries likely explains, at least in part, why they incurred higher total charges compared to the older cohorts. It is therefore probable that increased total charges are related to the LOS being two days longer on average for patients <65 years compared to those ≥65 years old. Moreover, older patients typically experience simpler extra articular fracture patterns and therefore shorter operative times, a variable that has been related to decreased LOS and wound complications.¹⁵

In this study, younger patients also had higher rates of median nerve injuries, infections, and wound complications. Median nerve neuropathies after DRF have previously been reported to occur in 3 %–17 % of patients.¹⁶ Previous literature has shown that median nerve injuries are not necessarily more prevalent in the younger population.¹⁶ Yet, in this study, those ≥65 years old had 23 % lower odds of experiencing median nerve injuries as compared to their younger counterparts. Thus, it is imperative to discuss this risk with patients and to monitor for signs of median nerve complications, especially in younger populations. Interestingly, the present study identified increasing age to be an independent predictor for less infectious complications, including fixation related infections. This finding is likely related to the presumed traumatic nature of younger DRF patients. Using the PearlDiver dataset, Constantine et al. demonstrated a similar infectious odd (OR: 0.62) for DRF fixation patients >65 years old, a finding they attributed to lower energy mechanisms of injury.¹⁷ Albright et al. also demonstrated that young adult populations have increased rates of postoperative infections following DFR fixation.¹⁸ Thus, these populations need to be monitored closely in order to avoid potential devastating complications such as functional impairments, nonunion, osteomyelitis, or even revision surgery.

The increased complication rate in the young adult population is an important finding when considering that infectious complications are amongst the most common reasons for early patient readmissions following DRFs treated with ORIF. Further, infectious complications are associated with increased overall hospital charges, reported up to $23,355 for patients with infections compared to $6383 for patients without infection complications.¹⁷ The burden of this nearly four-fold increased cost demonstrates the importance of identifying patient factors that increase the risk of infection to aid in careful selection of operative intervention. Moreover, intraoperative complications were not found to be impacted by age here, however patient comorbidities still need to be considered strongly before proceeding to surgery, especially in the elderly population.

Some additional reasons for unplanned health care contact following ORIF for DRFs include inadequate pain control, cast/splint problems, and swelling with most reasons for 30-day readmission being unrelated to the fracture.¹⁹ This is consistent with the finding of increased rates of early readmissions in the elderly population, who often have more significant medical comorbidities and likely seek medical care for non-orthopedic related concerns postoperatively. Highlighted in this study, younger patients were more likely to have routine dischargers. Moreover, roughly three and five more patients per 100 patients were readmitted within 30 and 90 days, respectively, when comparing patients at the 65 years old cutoff. However, this association of age and readmission was relatively weak in the multivariate analysis (OR: 1.01; p < 0.01), presumably secondary to controlling for patient comorbidities. These findings were more robust with ≥65 years old as the independent variable (OR - 30-days: 1.19, 90-days: 1.32; p < 0.01). Nevertheless, it remains important for patients and physicians to be aware of the increased risk for readmission with increased age. This becomes particularly true given that previous literature has shown non-inferiority for conservative versus operative intervention in the elderly population who sustain DRFs.^3,20 Thus, the potential for short-term hospital readmission must also be considered strongly when indicating older and sicker patients for ORIF for DRFs.

The present study is not without its limitations. First, the data presented was from a large national database, lending the study to a retrospective design. Utilizing the NRD meant that clinical data such as patient reported outcomes and range of motion could not be recorded, nor could the mechanism of injury, surgical approach, or hardware type. Further, cost analysis can only be represented as raw total charges which may not always be related to the procedure and malady of interest. Moreover, with a large sample size statistical significance may not directly correlate to clinical significance.

5 Conclusion

Age can have a variable impact on short-term complications following ORIF for DRFs. In the present analysis of 32,679 patients, increased patient age was found to be an independent predictor of increased 30-day and 90-day readmissions, however older patients experienced decreased rates of infectious complications, median nerve injuries, and prolonged length of stay. Moreover, there was a stepwise decrease in total charges associated with increased age. Ultimately, this study acts to identify crucial short-term complications associated with patient age and should be used alongside clinical outcomes literature to appropriately counsel patients who sustain operative DRFs.

CRediT authorship contribution statement

Christopher A. White: Conceptualization, Data curation, Project administration, Investigation, Methodology, Writing – original draft, Writing – review & editing. Kira K. Tanghe: Conceptualization, Data curation, Investigation, Methodology, Writing – original draft. Akiro Duey: Data curation, Investigation, Methodology, Writing – review & editing. Claudia Siniakowicz: Data curation, Investigation, Methodology, Writing – review & editing. Thomas Evashwick-Rogler: Data curation, Investigation, Methodology, Writing – review & editing. Jaehon Kim: Conceptualization, Investigation, Methodology, Project administration, Resources, Writing – review & editing, Validation, Visualization.

6 Disclosures

Jaehon M. Kim.

Consulting Fees: Depuy Synthes.

Ethical statement

N/a.

Cosent statement

The following project utilized publicly available information and therefore guardian/patient consent were not required.

Funding statement

Let this document serve to indicate that there was no funding used for the completion of this research.