Implant Charge Differences Between Distal Radius Fixation Constructs (CPT 25607, 25608, and 25609)

Abstract

Background:

Implants are a significant contributor to health care costs. We hypothesized that extra-articular fracture patterns would have a lower implant charge than intra-articular fractures and aimed to determine risk factors for increased cost.

Methods:

In total, 163 patients undergoing outpatient distal radius fracture fixation at 2 hospitals were retrospectively reviewed stratified by Current Procedural Terminology codes. Implants and associated charges were noted, as were sex, age, insurance status, surgeon specialty, and location. Bivariate and multivariable regression were used to determine associations.

Results:

Total implant charges were significantly lower for 25607 (extraarticular, $3,348) than 25608 (2-part intraarticular, $3,859) and 25609 (3+ part intraarticular, $3,991). In addition, intra-articular fractures had higher charges for distal screws/pegs and bone graft. Charge was lower when surgery was performed at a trauma center. There was no charge difference associated with insurance status, age, sex, hand surgery specialty, or fellow status. Substantial intersurgeon variation existed in all fracture types.

Conclusion:

Distal radius fractures may represent a good model for examining implant costs. Extra-articular fractures had lower implant charges than intra-articular fractures. These data may be used to help construct pricing for distal radius fracture bundles and potential cost savings.

Introduction

Cost is an increasingly important consideration in health care and in orthopedic surgery, and significant cost variation exists between different types of implants used for the same purpose. Orthopedic and musculoskeletal disorders accounted for an expenditure of $214 billion in 2014, or 1.2% of US gross domestic product, with these figures expected to rise significantly with the aging population. ¹ The cost of open reduction and internal fixation (ORIF) of a distal radius fracture ranged from $6577 to $8181 in a 2018 study of more than 23 000 operations. ²

Implants are typically a significant contributor to orthopedic surgery costs, although in many instances it is unclear whether more expensive constructs provide an improved outcome, such as comparing plating with percutaneous pinning in distal radius fractures or comparing locking with nonlocking plates in the treatment of tibial plateau fractures.^3,4 Many factors may drive implant cost variation for distal radius fractures, including perceived differences between construct types, opaque pricing, and lack of incentive alignment between the payer and the provider. ⁵

A number of recent studies have examined strategies to control costs, and they found that by minimizing variation and using reference pricing for particular routine fracture types, costs were lowered, whereas quality measures were either unchanged or improved.^3,6-9 United Kingdom distal radius acute fracture fixation trial (UK DRAFFT), a recent multicenter randomized-controlled study, found no clinical difference between locked plating and percutaneous Kirschner wire (K-wire) fixation, and that K-wire fixation was both cheaper and faster. However, before a surgeon can seek to control his or her costs, it is imperative that true cost be measured and over the entire surgical episode. A randomized survey found that orthopedic surgeons who knew the cost of implants decreased their costs by approximately 10% without significantly changing constructs. ⁸

To deliver better value-based care, we must better understand the charges of implants used for fixation of distal radius fractures. We hypothesized that extra-articular fracture patterns (25607) would have a lower implant charge than intra-articular fractures (25608 or 25609) and aimed to determine risk factors for increased charge.

Materials and Methods

Following institutional review board approval, our institution’s billing databases were retrospectively reviewed from January 2016 through June 2017 for patients aged ≥18 years undergoing outpatient distal radius fracture fixation (as defined by Current Procedural Terminology [CPT] codes) at our institution’s 2 primary hospitals (one a Level 1 trauma center and another a community hospital) and the 2 ambulatory surgery centers (ASCs) associated with each. These patients were subsequently stratified by CPT codes (25607 = extra-articular, 25608 = 2-piece intra-articular, 25609 = 3+ piece intra-articular); external fixation codes (20690 = uniplanar, 20691 = multiplanar) were also included. All case data were individually inspected to ensure consistency.

For proprietary and contractual reasons, charge (rather than cost) was used throughout. Implant charges (by the hospital, to the insurer) were noted and divided into the constituent components, including plates, screws/pegs, external fixators, and bone allograft. Plates and screws were further subdivided into volar plates and associated distal screws/pegs, and others. Patient demographics were collected, including sex, age, and insurance status, as were surgeon type (with or without hand fellowship training, and attending vs fellow) and location (Level 1 trauma center or community hospital).

Along with descriptive statistics, χ² and analysis of variance tests were used to determine differences between groups. For each CPT code, each surgeon’s average charge was analyzed and compared, and the variation was assessed by dividing the highest average charge by the lower. Charge-to-cost ratios were also calculated. In addition, multivariable linear regression was used to determine risk factors for increased implant charge. A value of P < .05 was considered statistically significant a priori.

At each location, surgeons can freely choose among the implant sets that are approved by the health system administration, each of which is chosen through a periodic request-for-proposal process, with prices set. Systems from several manufacturers are readily available, including Hand Innovations/Zimmer Biomet (Anatomic and CrossLock DVR; Zimmer Biomet, Warsaw, Indiana), Skeletal Dynamics (Geminus, Miami, Florida), Synthes (LCP, Warsaw, Indiana), and Medartis (Aptus, Zurich, Switzerland), as well as standard small- and mini-fragment sets. In addition, sets from other manufacturers may be brought in on a case-by-case basis. Kirschner wires were not analyzed separately, but were included in the costs of overall constructs. Each ambulatory surgery center is associated with a different hospital; both are wholly owned by the health system, and there are no cost-sharing arrangements or physician financial incentives in place.

Results

Over the 18-month period from January 2016 through June 2017, 163 patients underwent outpatient distal radius ORIF (CPT 25607: 47 [28.8%], CPT 25608: 60 [36.8%], and CPT 25609: 56 [34.4%]). Demographic information was generally similar among the groups, except for location of surgery (Table 1). One patient underwent external fixation (CPT 20690). The cost-to-charge ratio averaged 2.6, with a standard deviation of 0.7.

Table 1.

Demographics.

Demographics	25607	25608	25609	Total	P value
Patients (n)	47	60	56	163
Patients, %	28.8	36.8	34.4	100.0
Age (mean), y	53.1	55.7	54.6	54.5	.646
Age (SD), y	13.7	14.4	14.0	14.2
Male (n)	11	17	13	41	.775
Male, %	23.4	28.3	23.2	25.2
Level 1 trauma (n)	26	49	26	101	<.0001
Level 1 trauma, %	55.3	81.7	46.4	62.0

Note. 25607 = Current Procedural Terminology (CPT) code for extra-articular distal radius fracture; 25608 = CPT code for 2-part intra-articular fracture; 25609 = CPT code for 3+ part intra-articular fracture.

Most patients with each fracture type were managed with volar locked plating (VLP; Table 2). Among these, patients with extra-articular or 2-part intra-articular fractures had fewer points of distal screw/peg fixation than those with 3+ part intra-articular fractures (5.2 and 5.2 vs 5.7), although this did not reach statistical significance. Charges for VLP alone varied from $1613 to $3465. Dorsal bridge plating was used in three 25608 patients (5%, with additional VLP in 2) and in two 25609 patients (3.6%, with additional VLP in 1), with no usage in 25607 patients. There was no difference in the incidence of allograft usage between groups.

Table 2.

Fixation Outcomes, by Current Procedural Terminology (CPT) Code.

Fixation outcomes	25607	25608	25609	P value
No. of plates (average)	1.00	1.03	1.00	.785
No. treated with VLP	46	57	51	.328
% treated with VLP	97.9	95.0	91.1
No. of distal fixation points (for VLP)	5.2	5.18	5.71	.154
No. treated with bone allograft	7	6	13	.148
% treated with bone allograft	14.9	10.0	23.2

Note. 25607 = CPT code for extra-articular distal radius fracture; 25608 = CPT code for 2-part intra-articular fracture; 25609 = CPT code for 3+ part intra-articular fracture. VLP = volar locked plating.

There were significant differences in the case mix between 8 surgeons with hand fellowship training and 16 surgeons without; hand-fellowship-trained surgeons treated a significantly higher proportion of 25607 (74.5%) and 25609 patients (76.8%) than 25608 patients (51.7%). Trainees were involved in most cases.

Implant charges were significantly lower on univariate analysis for 25607 ($3348) than 25608 ($3859) and 25609 ($3991; Figure 1 and Table 3). Multivariable regression analysis demonstrated an increased implant charge associated with fracture pattern (25608 by $901, P = .001, and 25609 by $539, P = .041, compared with 25607), whereas cases performed at the Level 1 trauma center (and affiliated ASC) were associated with a lower implant charge, by $643 (P = .009). There was no charge difference associated with insurance status, age, sex, hand fellowship training, or fellow status.

Figure 1.

Average implant charges, by surgeon and CPT code. Error bars denote standard deviation: (a) 25607 (extra-articular fractures), (b) 25608 (2-part intra-articular fractures), and (c) 25609 (3+ part intra-articular fractures).

Note: Surgeon designations change between groups for anonymization. CPT = Current Procedural Terminology.

Table 3.

Cost Outcomes, by Current Procedural Terminology (CPT) code.

Charge outcomes	25607	25608	25609	P value
Plates (all types)	$1740	$1997	$1942	.165
Distal fixation (for VLP)	$1089	$1129	$1332	.004
Bone allograft	$340	$399	$495	.090
Total implant	$3348	$3859	$3991	.014
SD	$698	$1272	$1266
Median	$3434	$3600	$3765

Note. 25607 = CPT code for extra-articular distal radius fracture; 25608 = CPT code for 2-part intra-articular fracture; 25609 = CPT code for 3+ part intra-articular fracture. VLP = volar locked plating

There was significant variation in each surgeon’s average implant charge (Figure 1). For 25607 (Figure 1a), the highest average was 160% higher of the lowest, with a spread of 130% for 25608 (Figure 1b) and 94% for 25609 (Figure 1c).

Discussion

In this investigation of outpatient distal radius fracture fixation, we found significant variation in charges for implants used to fix distal radius fractures. After multivariable regression analysis, these charges varied both by fracture type (implants used in intra-articular fractures, CPT 25608 and 25609, had significantly higher charges than those used in extra-articular fractures, CPT 25607), and by location of treatment. We found that more complex fractures patterns required more points of distal fixation than simpler ones (although this was not statistically significant) and that fracture pattern was a significant driver of variation in overall construct charge. There was also substantial variability in average implant charges between surgeons treating each fracture pattern.

This charge variation is not specific to distal radius fracture implants, but is more likely ubiquitous across other orthopedic implants. Several studies show wide variation in costs between different hospitals for the same operation. Robinson et al ¹⁰ examined a database of more than 15 000 knee and hip arthroplasties over 61 hospitals and found a range of pricing for the implant cost for total knee replacements to be $1797 to $12 093 and for total hip replacements to be $2392 to $12 651. A significant portion of the variation (36.5% for knee replacement, 59.5% for hip replacement) was not attributable to either hospital or patient characteristics; these may warrant further investigation and may be an opportunity for cost reduction (ie, using similar but less expensive implants). The cost differences for the same procedures show this field and this research to be a potential for health care cost savings, as a more homogeneous market and transparent pricing may allow surgeons to choose cheaper implants without sacrificing patient outcomes. This is similar to our findings, in which the only explanatory variables for charge included fracture morphology and treatment site, but could not account for the full spectrum of variability. Treatment site as a predictor of charge may be reflective of the fact that all surgeons did not operate equally at both sites, and therefore this may be more indicative of between-surgeon differences.

As new implant designs come to market, another area of research is determining whether newer (and frequently higher cost) implants change outcomes. The UK DRAFFT study found no difference in outcomes between patients with dorsally displaced distal radius fractures randomized to percutaneous pin fixation and volar plating. ⁶ Cavallero et al ⁴ investigated locked versus nonlocked fixation for treatment of bicondylar tibial plateau fractures and found no difference in clinical or radiographic outcome, but with nearly double the cost in the locked implant group ($4453 vs $2569, P < .01). Wetzel et al ¹¹ found significant variation in the cost of fixation for bimalleolar ankle fractures and bicondylar tibial plateau fractures, with the most expensive surgeon’s costs being more than 300% the least expensive surgeon’s cost for bimalleolar ankle fractures ($2243 vs $613) and 200% for bicondylar tibial plateau fractures ($1839 vs $4088). Finally, Heare et al ¹² retrospectively reviewed their treatment of pediatric diaphyseal forearm fractures and found equivalent outcomes, surgical time, and complication rates between titanium elastic nails and K-wires, with a cost difference of $639 versus $24 (P < .001). Similarly, we found substantial variation in the average cost of treating similar fracture patterns between different surgeons.

The source of this variation is multifactorial, and several studies, including ours, have investigated factors influencing this wide pricing range. Our study found significant explanatory variables to include location of treatment and intra-articular versus extra-articular fracture morphology. We found that patients treated at the Level 1 trauma center (and associated ASC) had significantly lower charges than those treated at the Level 2 center (and associated ASC). This may be due to several factors, including surgeon preference (most surgeons operate mostly at one institution or the other), implant contract disparities, or other factors that were not adequately captured in the data. Virkus et al investigated fellowship training as a cause of cost variation, examining 208 ankle fractures treated by trauma-trained orthopedic surgeons versus non–trauma-trained orthopedic surgeons. The adequacy of reduction on radiographs was similar at final follow-up between the 2 groups, but the nonfellowship group average costs were $2904 versus $1233 in the trauma-trained group (P < .001). ⁵ Our study did not find any differences in implant charges between those with hand fellowship training and those without. In an examination of American Board of Orthopaedic Surgery candidate data, Childs et al found that hand-fellowship-trained surgeons had a different profile for distal radius ORIF treatment, with more complex fractures treated; our study found this to be the case as well, but also found that surgeons with hand fellowship training were also responsible for a higher proportion of 3+ intra-articular fractures (CPT 25609). Years of career experience were not specifically assessed.

Recent studies have more closely assessed the cost of treating distal radius fractures.^7,13-16 A cost minimization analysis in the elderly by Pang et al, ¹³ which holds clinical outcomes equal, found that closed reduction may be a higher value intervention in patients aged ≥65 years, with a lower total cost of treatment, even when complications were included. Alternatively, the DRAFFT investigators found that K-wire fixation was clinically equivalent to ORIF and significantly cheaper. ¹⁴ An administrative database analysis found that surgical cost was the most important driver of potential distal radius surgery bundle cost, underscoring the need to appreciate and control surgical costs; there was also substantial variability in the cost of ORIF. ⁷ Similarly, Kazmers et al ¹⁵ determined that implants were the larger drive in cost variation for distal radius fractures, representing 48% of the direct costs and 32% of the complete costs. These studies, along with other similar investigations, point to the critical nature of understanding cost drivers in orthopedic surgery and working to minimize them without compromising patient care.^11,17-19

As cost becomes an ever-increasing focus in health care and insurance reimbursement, there will be a significant drive toward cost-effective management and fixation of orthopedic injuries without sacrificing quality. A recent study by Okike et al ²⁰ showed that only 21% of orthopedic surgeons were able to estimate the cost of implants to within 20% of their actual cost. Orthopedic surgeons tended to overestimate the cost of inexpensive implants and underestimate the cost of expensive ones, showing to many surgeons that they view cost to be more homogeneous across implant choices than it truly is. Streit et al ²¹ surveyed 97 residents and attending surgeons, and found underestimation of implant cost to be twice as common as overestimation, with a mean percentage error of 73% for residents and 59% for attendings. Another study showed that in a survey of 57 surgical residents and attendings in New Zealand, the estimated prices of 14 commonly used items and instruments were off by at least 25% the actual value 87% of the time, most commonly due to underestimates. ²² At our institutions, there is minimal pricing transparency among the hospital system, the surgeons, and the implant representatives.

Limitations

There are several limitations to this study. Most importantly, this is a retrospective, nonrandomized review of cases performed at 2 surgery centers within 1 health system, and the results may not be widely generalizable. However, the small size of the database is also a strength, allowing for clinical, radiographic, and financial investigation of the implants used in each case. This study also does not consider other pertinent costs, such as operating room time, disposable items, and waste. While these are important, focusing on implants may help the generalizability of the data. There is no clinical follow-up, and so any potential differential effects between constructs of different costs cannot be analyzed. Grouping the fractures by CPT is an imprecise method, as simple classification systems can belie the severity of certain fracture patterns and may group dissimilar fractures that may benefit from different treatments (and different implants). Nevertheless, using CPT codes is directly applicable to the billing aspect of orthopedic surgery, and reimbursement (including for implants) may be directly tied to this classification scheme. There is also a limited surgeon appreciation for cost (both at our institution and elsewhere), which may make the results found here less actionable. In addition, implant charges often vary substantially from one institution to another, and the financial contracts are frequently convoluted and opaque, including rebates, volume discounts, single-use instruments, and other measures to obfuscate the true cost to the system. Finally, due to nondisclosure agreements and hospital policy, data could only be presented as charges, rather than actual costs. To help clarify the true cost, cost-to-charge ratios were given.

Conclusion

We found that fixation of extra-articular distal radius fractures required fewer points of distal fixation and was associated with lower implant charges than fixation of intra-articular fractures. This is consistent with and extends previous literature in this area, but is based on more granular clinical data than most database studies, which generally lack imaging and therefore information about the fracture pattern. In addition, we found no difference in implant charges between hand-fellowship-trained and non–hand-fellowship-trained surgeons.

Distal radius fractures may represent a good model for examining implant costs. Volar locking plates and screws are common and tend be nearly a commodity implant, demonstrating significant variation in charges without strong evidence to show improved clinical outcomes. There was a substantial spread in the average charges for each of the different CPT codes, ranging from 170% to 260%. This value proposition varies significantly with price, and these data may be used to help construct pricing for distal radius fracture bundles and potential cost savings.