1 Introduction

The cost of healthcare continues to increase in the United States (US) each year. In 2014, $214 billion dollars were spent on orthopaedic conditions alone (Bone and Joint Initiative USA). Distal radius fractures (DRFs) are one of the most common orthopaedic injuries in the US with approximately 634,000 occurring annually.¹ They have been estimated to generate Medicare expenditures of $385 to $535 million annually in just the geriatric population.^2,3 As the population continues to age, the amount of resources spent on fractures is anticipated to increase. Traditionally, the treatment of DRFs was relatively low-cost, consisting of nonoperative care or percutaneous fixation; however, with the introduction of the volar locking plate (VLP), the incidence of surgical treatment and cost of treatment for DRFs has greatly increased.^4,5

Although the optimal method of treatment for DRFs is still a topic of controversy, implants have been shown to be one of the greatest contributors (32%) to the total cost of the operative encounter for surgically-treated DRFs.⁶ Implants thus likely contribute to value, defined as the change in patient-reported outcome score over total cost of care, in DRF treatment. This affords surgeons the ability to provide cost-effective care to their patients by growing cognizant of implant costs while maintaining good patient outcomes. However, it has been shown that most orthopaedic surgeons lack awareness of implant costs.⁷

Studies have revealed construct cost variability in other fracture types and have shown that factors such as locking plates, cannulated screws, and fellowship training function as key cost drivers.^8–10 In DRF specifically, factors such as fellowship training, implant manufacturer, and number of screws used have been found to significantly influence cost.^6,11 There is a paucity of literature in this area, and as the US moves towards value-based healthcare, there is a need to determine cost variation and key cost drivers in the treatment of DRFs. The aims of this study were to evaluate for cost variation in DRFs treated with a VLP and to identify key factors that affect the total construct cost. It was hypothesized that increased articular involvement and locking screws would lead to higher construct costs.

2 Materials and methods

After receiving approval by the Institutional Review Board, electronic medical records (EMR) for a single metropolitan healthcare system were queried for patients with a surgically-treated DRF from May 2014 to December 2021. There were 189 patients identified who sustained an isolated DRF treated with open reduction and internal fixation (ORIF). Exclusion criteria consisted of skeletal immaturity, polytrauma (any fracture other than a DRF), or open fracture (due to minimal cases). Any patients treated nonoperatively or with any surgical fixation besides a VLP were also excluded.

Variables collected for the study included demographics, comorbidities, hand dominance, AO/OTA classification, surgeon specialty, and facility of treatment. Implant variables included implant vendor, as well as number and type of plate, pegs, non-locking screws, variable angle screws, and locking screws used. The use of specific implants for each patient was determined by surgeon preference. Implant model numbers for each patient were cross-referenced with the healthcare system's charge-master database to calculate implant cost.

SPSS version 28 (Chicago, IL) was used to analyze the collected data. The distribution of total implant costs is reported using descriptive statistics (i.e., mean, standard deviation, sample size, percentages). Two multiple regressions were performed. The first multiple regressions analysis was run to examine the potential predictive relationship between the dependent variable of construct cost and independent variables of AO/OTA fracture classification, facility acuity level, implant vendor, and fellowship training. The second examined whether potentially predictive relationship existed between the dependent variable of total implant cost and independent variables of varying implant types: multidirectional threaded peg (MDTP), locking peg, non-locking screw, locking screw, and variable angle locking screw. Both regressions were assessed for linearity using partial regression plots and studentized residuals against predicted values. Both tests were also assessed for normality using Q-Q plots.

3 Results

A total of 140 patients who underwent ORIF of the distal radius met the study criteria. The majority of patients were female (n = 120, 85.7%) and had an average age of 59 ± 13.7 years old. Approximately 28% (n = 54) of all patients had a pre-existing mental health diagnosis, and most were former smokers or non-smokers (n = 106, 75.8%). Patients more commonly injured their dominant hand (n = 75, 53.6%) and typically presented with an AO/OTA 23C fracture (n = 93, 66.4%). There were a total of 22 surgeons included in the study representing three subspecialties (orthopaedic-trauma, orthopaedic-hand, and plastic surgery-hand). Eighteen surgeons graduated from an orthopaedic residency program with seven completing a trauma fellowship and eleven completing a hand fellowship. Four surgeons graduated from a plastic surgery residency and completed a hand fellowship. The majority of patients were treated by an orthopedic hand surgeon (n = 116, 82.9%). Orthopaedic hand surgeons also treated the highest proportion of 23C fractures (69.8%). There were 90 patients (64.3%) treated at an ambulatory surgery center (Table 1).

The standard locking plate (three holes proximally and seven holes distally) was the most common size used (n = 98, 70%). The average number of non-locking screws was 3.1(range: 0–8). A total of 6.1 (range: 0–9) pegs (locking and MDTP) or 7.1 (range: 0–9) locking screws (locking and variable angle) were used on per construct. The average construct cost per patient was $1289.67 ± $215.32 (range: $857.83-$2156.95), representing a coefficient of variance of 20.1% (Fig. 1). The DRFs in this study were most commonly treated with implants from Zimmer-Biomet (n = 90, 64.3%). The most expensive fixation constructs were those that involved a variable angle locking screw ($1316.75 ± $264.99) or a (MDTP) ($1321.67 ± $192.94). Fracture classification, treatment facility, implant vendor, and surgeon fellowship training did not statistically significantly predict total construct cost (F = 0.60, p = 0.66, adjusted R² = 0.02) (Figs. 2–3). None of the four variables added statistically significantly to the prediction (all p-values >0.17). Specific implants similarly did not statistically significantly predict cost (F = 0.56, p = 0.77, adjusted R² = 0.02), with none of the implant categories adding statistically significant predictive ability (all p-values >0.27) (Table 2).

Close

Fig. 1 Distal Radius Implant Cost Variation. A scatterplot demonstrating the total construct cost between 2014 and 2021 (n = 140). Construct cost is organized by date of procedure.

Export to PPT

Close

Fig. 2 Total Implant Cost by AO/OTA Classification. Box and whisker plots demonstrating the total construct cost by AO/OTA Classification between 2014 and 2021 (n = 140). Construct cost is organized by increasing articular involvement.

Export to PPT

Close

Fig. 3 Total Implant Cost by Surgeon Training. Box and whisker plots demonstrating the total construct cost by residency and fellowship training between 2014 and 2021 (n = 140).

Export to PPT

4 Discussion

Fixation construct costs can be influenced by a variety of factors such as patient's age, articular involvement, and the number and type of implants used. The purpose of this study was to describe implant cost variability and identify key factors that influence total implant costs in the setting of an isolated DRF. In contrast to previous literature this study found limited variability of the total construct cost with an average of $1289.67 ± $215.32^{8−10, 12}. Additionally, articular involvement as well as implant type and vendor were not found to significantly influence implant costs (p-values >0.27). These results suggest that when a VLP is chosen as the method of fixation for a DRF, regardless of the AO/OTA classification of fracture, there is minimal cost variation. Surgeon subspecialty and the type of treatment facility did not correlate with cost (p = 0.81 and p = 0.90, respectively), suggesting that surgeon training and where they perform surgery does not impact the overall construct cost.

This study found limited variation in total implant cost compared to previous literature, highlighting the importance of cost containment through contract negotiation. One study found that administrators and surgeons were unable to accurately identify implant cost, although surgeons’ estimates were closer to the actual value.¹³ Wasterlain et al. presented upper extremity fracture cases to two groups of surgeons, with one of the groups blinded to the cost of implants available to treat the fractures. They demonstrated in the price-aware group that implant costs were reduced by 9–11%.¹⁴ When given the opportunity, surgeons chose the lower cost implants. If administrators and surgeons are allowed to collaborate together on contract negotiations it can lead to decreases in cost.¹⁵ One of example was demonstrated at an institution where a single price was mandated, and all vendors were allowed to bring implants in at that price point or not.¹⁵ Their method led to reported cost savings of $2 million in the first year alone. Another potential negotiation strategy is to use a single-vendor model. Providing one vendor the opportunity to exclusively provide implants for an institution can led to significant discounts (up to 45% at one institution).¹⁶ Institutions should take into account whether the vendor quotes consist of construct prices (single price for a particular construct) or vendor prices (price per item). With construct prices, there are no additional savings for reducing the number of implants used which may be more beneficial in constructs with less variability as has been shown in this study with VLPs.

The different types of implants used in surgical management present an opportunity to decrease the total cost of implants. The responsibility of making the active decision to use more cost-effective implants is up to surgeon discretion. Although this study did not find a significant difference between the types of screws used or implant company, this may be due to a lack of variability in construct inherent to VLPs. Typically, the proximal holes are filled with non-locking screws, and the distal holes are filled with locking pegs or screws. In a study comparing the biomechanical properties of locking vs. non-locking screws in distal humerus fractures, it was shown that while there was no difference in biomechanical advantage between the two groups, the cost of a non-locking screw was substantially lower than that of a locking screw ($16.00 vs $93.50).¹⁷ Okelana et al. demonstrated in ankle fractures that constructs with locking plates and cannulated screws were the most costly.⁹ Variable angle and multidirectional pegs and screws were the most expensive implants in this study, but these were not shown to significantly impact costs possibly due to sample size. Choosing to reduce the number of screws and especially the more expensive types of screws may help patients decrease overall cost. Considering the mean number of non-locking screws of 3.1(range: 0–8) and the mean number of locking pegs or locking screws to be 6.1 (range: 0–9) and 7.1 (range: 0–9), respectively, these results suggest that the surgeons in this study fill all or most screw holes in the plate regardless of amount of distal comminution, leading to the similar cost among fracture patterns. Tailoring screw choice according to fracture comminution could potentially lead to cost containment while maintaining patient outcomes. A previous study looked at outcomes in constructs with a high-density of screws compared to a low density of screws in adolescent scoliosis cases.¹⁸ The authors found no differences in radiographic, clinical, or complication related outcomes, but the total construct cost was significantly different. Biomechanical studies in DRF literature have demonstrated that using four of the VLP's distal screw holes in an alternating pattern may provide sufficient stability in extra-articular and intra-articular fractures.^19–21 Future studies should analyze the relationship between number of screws used in VLPs and patient outcomes to see how many are actually necessary for patients to achieve maximum benefit.

Surgeon training has been shown to be associated with implant cost variation. In ankle fractures and pilon fractures, trauma-trained orthopaedic surgeons have been shown to have decreased construct costs compared to non-trauma trained surgeons.^9,10,22 In contrast, Goodman et al. found no difference in implant cost based on surgeon training for DRFs.¹² This study similarly did not find any difference in the total implant costs based on the type of training of each surgeon, indicating that the use and cost of a VLP seems ubiquitous across specialties. One confounding factor to these results may be the make-up of patients seen by each type of surgeon. Literature has shown the patient profile that hand fellowship-trained surgeons see typically presents with more complex fractures compared to other specialties,^12,23 which is also similar to the results we found in this study as a higher proportion of 23C fractures were treated by the orthopaedic hand-trained surgeons.

As measured by the AO/OTA classification, type C “intra-articular” fracture patterns have been shown to be associated with higher costs in ankle fractures.⁹ Recently, Goodman et al. found that surgically-treated DRFs that were intra-articular had significantly higher charges associated with them compared to extra-articular DRFs.¹² Dzaja et al. compared percutaneous fixation to VLPs in simple extra-articular and simple intra-articular DRFs and found that, despite comparable patient outcomes between groups, the VLP group did have higher initial costs associated with surgery ($1637.27 vs. $733.91).²⁴ The present study did not account for other implant types but allowed for variability in number and type of VLPs, screws, and vendors used. Total implant cost was not shown to be affected by fracture classification, which may suggest that fracture pattern may be a lower priority to surgeons when using a VLP for treatment.

This study had a number of strengths and weaknesses. One strength was that patients were only treated with VLPs, limiting the variability of implants between patients and allowing for a more direct comparison of cost. Implants used in the study are from some of the more common medical device companies in the US, allowing for generalizability to other centers. This study also leveraged the resources of a large health system across multiple providers and facilities. An important limitation was that this was a retrospective review, thus rendering implant numbers and cost dependent on the accuracy of the EMR. The study excluded 49 patients that were ineligible for analysis, limiting the results and conclusions. There is a possibility of selection bias in this study due to the exclusion of patients and retrospective study design. Implant prices have large variability between healthcare systems and treatment facilities because of contract negotiations with medical device companies. This limits the ability to directly compare the prices used in this study to other studies or healthcare systems. Disposable implants also were not included in the total construct cost to allow for equal comparison among vendors, but this decision also made this analysis incomplete. The construct cost was also determined on the day of surgery and not adjusted for inflation which may impact the results. Future research should evaluate the total cost with the inclusion of disposable implants as vendors have demonstrated varying prices for these types of implants. Finally, years of experience were not assessed between the surgeons possibly limiting the results regarding fellowship training.

5 Conclusion

Surgically treated DRFs with a VLP demonstrated similar implant costs regardless of fracture pattern, surgeon specialty, or treatment facility. Contrary to previous literature, VLPs showed minimal cost variation, although some surgeons were able to decrease the overall cost by reducing the number of screws used. Further investigation should be done to evaluate the relationship between the number of screws used in fixation and patient-reported outcomes to provide high-value care to patients. Given the current healthcare climate and implants being one of the primary drivers of cost, surgeons should endeavor to reduce the amount of costlier implants (i.e., variable angle locking screws) in their fixation constructs.

Funding & sponsorship

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Patient consent

Patients consent was not considered necessary for this study as this was a retrospective review.

Institutional Ethical Committee Approval.

All procedures were conducted in accordance with the approval of the author's IRB. This study was assigned approval number A17-376.

Authors contribution

Stephen A. Doxey DO: Conceptualization, Methodology, Visualization, Investigation, Writing-Original Draft Prepration. Fernando A. Huyke-Hernández BS: Conceptualization, Data Curation, Writing-Reviewing Editing. Jennifer L. Robb PhD: Validation, Statistical Analysis. Deborah C. Bohn MD: Writing-Reviewing Editing, Supervision. Brian P. Cunningham MD: Conceptualization, Writing-Reviewing and Editing, Supervision.