Primary versus revision total shoulder arthroplasty: comparing relative value and reimbursement trends.

BACKGROUND: Total shoulder arthroplasty (TSA) has been demonstrated to be an effective treatment for multiple shoulder pathologies. The purpose of our study was to compare the relative value units (RVUs) per minute of surgical time for primary and revision TSA.
METHODS: The American College of Surgeons National Surgical Quality Improvement Program database was queried to identify patients that underwent primary TSA, one-component revision TSA, and two-component revision TSA between January 1, 2015 and December 31, 2017 using current procedure terminology codes. RVUs were divided by mean operative time for each procedure to determine the amount of revenue generated per minute. Rates were compared between the groups using a one-way analysis of variance with post-hoc Tukey test. Statistical significance was set at p<0.05.
RESULTS: When dividing compensation by surgical time, we found that two-component revision generated more compensation per minute compared to primary TSA (0.284±0.114 vs. 0.239±0.278 RVU per minute or $10.25±$4.11 vs. $8.64±$10.05 per minute, respectively; p=0.001).
CONCLUSIONS: The relative value of revision TSA procedures is weighted to account for the increased technical challenges and time associated with these procedures. This study confirms that reimbursement is higher for revision TSA compared to primary TSA.

INTRODUCTION

Total shoulder arthroplasty (TSA) is an effective treatment for multiple shoulder pathologies [1-7]. Studies have shown that the use of TSA is increasing every year in the United Stated at a rate higher than that of total hip arthroplasty (THA) or total knee arthroplasty (TKA) [8-11]. Wagner et al. [10] found that between 2011 and 2017 the incidence of primary TSA performed per year increased 103.7%, while the incidence of primary THA and TKA increased by 29.1% and 17.8%, respectively. As the number of TSAs increases, the number of revision surgeries will likely increase as well.

Although arthroplasty provides patients with excellent outcomes for various glenohumeral disorders, the procedure is not without risk [12-14]. Implant failure (most commonly of the glenoid component), periprosthetic fracture, and infection are all complications of TSA that can necessitate a revision surgery, which can be more technically difficult than primary TSA with a significantly higher complication rate [15-20].

Although revision surgery is technically more challenging than primary surgery, the relative value of revision surgery does not necessarily reflect these differences. In the United States (US), the value and reimbursement for many procedures is estimated by the relative value unit (RVU) model. In this model, compensation is determined by a government agency known as the Centers for Medicare and Medicaid Services (CMS), which assigns a certain number of RVUs to a surgical procedure based on the agency’s evaluation of the procedure’s operative time, complexity, and workload. One RVU is equivalent to a certain amount of US dollars of compensation [21]. In the field of upper extremity surgery, the literature demonstrates that the CMS estimates of complexity, operative time, and workload used to determine procedures’ RVUs poorly correlate with reported values [22]. In the total hip and knee arthroplasty literature, several studies have demonstrated that revision procedures may be undervalued [23-25].

The relative valuation of different arthroplasty procedures is further complicated by the fact that revision arthroplasty procedures are weighted differently depending on the joint of interest, which has been shown to yield significant differences in the relative value of primary and revision procedures [23,25,26]. Although these differences have been reported in the total hip and knee arthroplasty literature, it has not been reported for TSA. A further consideration is that one-component and two-component revision procedures for TSA are valued differently. Thus, the purpose of our study was to determine if the relative value and reimbursement in the current RVU model properly account for the increased complexity of different revision TSA procedures.

METHODS

Institutional Review Board approval and informed consent were not required for this study.

Data Extraction and Inclusion Criteria

Data was obtained through the American College of Surgeons National Surgical Quality Improvement Program database (ACS NSQIP). These data were collected by a group of trained surgical clinical reviewers who record perioperative data at over 700 hospitals across the United States, including International Classification of Disease 9th and/or 10th revisions (ICD-9, ICD-10) codes, current procedural terminology (CPT) codes, and data related to operations, discharge disposition, reoperations, readmissions, and mortality through 30 days postoperatively [27]. This database is among the most precise and accurate databases available for measuring patient outcomes after surgery given its high rate of complete data sets, operative data points, validation, and inter-rater reliability [28,29]. The database was queried to identify patients that underwent primary TSA, one-component revision TSA, and two-component revision TSA between January 1, 2015 and December 31, 2017 using the CPT codes 23472, 23473, and 23474, respectively. Any surgical cases with concurrent procedures performed were excluded from final analysis.

Variables of Interest

Patient variables were extracted and compared between treatment groups, including sex, age, body mass index (BMI), American Society of Anesthesiologists (ASA) class, and medical comorbidities—diabetes, smoking, chronic obstructive pulmonary disease (COPD), ascites, congestive heart failure, hypertension, renal failure, dialysis, and chronic steroid use. The total operative time in minutes and RVUs for each case were collected. The RVU per minute was calculated for each case by dividing the work RVU by the total operative time. Cases with concurrent procedures (n=2,898) and their RVU values were excluded from the analysis. Within the 3-year period from which data were extracted, there were no changes in the work RVUs for primary TSA, one-component revision TSA, or two-component revision TSA. A Medicare conversion factor of 36.0896 dollars per RVU was used to determine the dollar reimbursement for each case [21]. The average reimbursement per case was determined by dividing the reimbursement for each case by the operative time.

Statistical Analysis

A Kruskal-Wallis test with post-hoc Bonferroni correction was used to compare differences in demographic and patient-specific variables between the primary TSA, one-component revision TSA, and two-component revision TSA groups. Variables of interest included sex, age, BMI, ASA class, and a variety of medical comorbidities—diabetes, smoking, COPD, ascites, congestive heart failure, hypertension, renal failure, dialysis, and chronic steroid use.

One-way analysis of variance with post-hoc Tukey test was used to compare differences in RVU, case length, RVU per minute, reimbursement per case, and reimbursement per minute between the primary TSA, one-component revision TSA, and two-component revision TSA groups. Statistical significance was set at p<0.05.

RESULTS

A total of 9,855 procedures were included in the study. Of these, 9,251 patients underwent a primary TSA, 229 underwent a one-component revision TSA, and 375 underwent a two-component revision TSA. The primary TSA group was older compared to the one- and two-component revision groups (under age 60, 15% vs. 21.8% and 17.1%, respectively; p=0.01), had lower rates of class 1 obesity compared to one-component, though higher than two-component (26% vs. 33.2% and 22.9%, respectively; p=0.019), and lower rates of diabetes compared to both one- and two-component revision (17.9% vs. 22.7% and 21.3%, respectively; p=0.049). There were otherwise no statistically significant differences in demographic variables between the groups. The findings are summarized in Table 1.

Table 1.

Summary of patient demographics

Variable	Primary TSA (n=9,251)	One-component revision TSA (n=229)	Two-component revision TSA (n=375)	p-value
Sex
Male	4,094 (44.3)	101 (44.1)	172 (45.9)	0.825
Female	5,157 (54.7)	128 (55.9)	203 (54.1)	NA
Age (yr)
<60	1,388 (15.0)	50 (21.8)	64 (17.1)	0.011
60–69	3,168 (34.2)	76 (33.2)	145 (38.7)	0.195
70–79	3,444 (37.2)	78 (34.1)	120 (32.0)	0.079
80–89	1,190 (12.9)	25 (10.9)	45 (12.0)	0.614
≥90	61 (0.7)	0	1 (0.3)	0.305
Body mass index (kg/m2)
Underweight	126 (1.4)	3 (1.3)	6 (1.6)	0.924
Normal	1,487 (16.1)	36 (15.7)	59 (15.7)	0.975
Overweight	2,964 (32.0)	68 (29.7)	140 (37.3)	0.071
Obese class I	2,403 (26.0)	76 (33.2)	86 (22.9)	0.019
Obese class II	1,299 (14.0)	30 (13.1)	46 (12.3)	0.581
Obese class III	972 (10.5)	16 (7.0)	38 (10.1)	0.223
Comorbidity
Diabetes	1,660 (17.9)	52 (22.7)	80 (21.3)	0.049
Smoking	1,041 (11.3)	33 (14.4)	43 (11.5)	0.329
COPD	672 (7.3)	20 (8.7)	19 (5.1)	0.182
Ascites	2 (0.0)	0	0	0.937
Congestive heart failure	60 (0.6)	1 (0.4)	1 (0.3)	0.613
Hypertension	6,202 (67.0)	168 (73.4)	256 (68.3)	0.120
Renal failure	5 (0.1)	0	1 (0.3)	0.244
Dialysis	33 (0.4)	0	0	0.339
Chronic steroid use	452 (4.9)	13 (5.7)	24 (6.4)	0.367
ASA class
Class 1 (no disturbance)	133 (1.4)	1 (0.4)	5 (1.3)	0.443
Class 2 (mild disturbance)	3,778 (40.8)	97 (42.4)	147 (39.2)	0.729
Class 3 (severe disturbance)	5,079 (54.9)	123 (53.7)	212 (56.5)	0.767
Class 4+ (life threatening)	248 (2.7)	7 (3.1)	10 (2.7)	0.941

Values are presented as number (%).

TSA: total shoulder arthroplasty, NA: not available, COPD: chronic obstructive pulmonary disease, ASA: American Society of Anesthesiologists.

During the study period, primary TSA generated 22.1 RVUs or $798.66, one-component revision TSA generated 25.0 RVUs or $902.24, and two-component revision generated 27.2 RVUs or $982.00. There was no statistically significant difference in surgical time between the three groups. When dividing compensation by surgical time, we found that two-component revision generated significantly more compensation per minute compared to primary TSA (0.284±0.114 vs. 0.239±0.278 RVU per minute or $10.25±$4.11 vs. $8.64±$10.05 per minute, respectively; p=0.001). The findings are summarized in Tables 2-5.

Table 2.

Comparisons of average RVU, case length, and RVU per minute for each procedure

Variable	Primary TSA	One-component revision TSA	Two-component revision TSA	p-value
RVU	22.1±0.0	25.0±0.0	27.2±0.0	-
Case length (min)	109.4±44.7	105.3±41.6	108.8±38.0	0.388
RVU per minute	0.239±0.278	0.278±0.116	0.284±0.114	0.001

Values are presented as mean±standard deviation.

RVU: relative value unit, TSA: total shoulder arthroplasty.

Table 3.

Comparisons of average reimbursement for each procedure

Variable	Primary TSA	One-component revision TSA	Two-component revision TSA	p-value
Reimbursement per case ($)	798.66±0.00	902.24±0.00	982.00±0.00	-
Reimbursement per minute ($/min)	8.64±10.05	10.02±4.19	10.25±4.11	0.001

Values are presented as mean±standard deviation.

TSA: total shoulder arthroplasty.

Table 4.

Comparisons of average RVU, case length, and RVU per minute for each procedure

Variable	p-value
	Primary vs. one-component revision TSA	Primary vs. two-component revision TSA	One-component vs. two-component revision TSA
RVU	<0.001	<0.001	<0.001
Case length	0.361	0.970	0.615
RVU per minute	0.087	0.005	0.958

RVU: relative value unit, TSA: total shoulder arthroplasty.

Table 5.

Comparisons of average reimbursement for each procedure

Variable	p-value
	Primary vs. one-component revision TSA	Primary vs. two-component revision TSA	One-component vs. two-component revision TSA
Reimbursement per case	<0.001	<0.001	<0.001
Reimbursement per minute	0.087	0.005	0.958

TSA: total shoulder arthroplasty.

DISCUSSION

Our study demonstrates that when accounting for operative time, two-component revision TSA generates significantly more revenue compared to primary TSA. One-component revision does not generate any more or less revenue than primary TSA or two-component revision TSA.

Our findings are in contrast to similar studies that compared compensation for primary versus revision total hip and knee arthroplasty. Sodhi et al. [23] found that revision THA generated significantly less RVUs per minute compared to primary arthroplasty. This finding was confirmed in a more recent study by Feng et al. [30] which assessed both one- and two-component revision hip arthroplasty and found that both forms of revision arthroplasty generated significantly less revenue compared to primary arthroplasty. A study by Peterson et al. [25] also found that revision TKA had a lower reimbursement rate per minute when compared to primary TKA. One possible explanation for the difference in results between our study and the aforementioned studies may be that in our study there was no statistically significant difference in operative time between primary and revision procedures. In contrast, studies in both hip and knee arthroplasty found that revision cases took significantly longer to perform than primary cases. Operative time may not have been significantly different between the three groups in our study due to the fact that the revision arthroplasty group in our study was significantly younger than the primary group. This was a surprising finding given it would be expected that older patients would be more likely to undergo revision surgery. Regardless, our younger revision group likely had better bone and soft tissue quality, which could allow for a quicker revision surgery.

Outside of the arthroplasty literature, a study by Orr et al. [24] compared the RVUs per minute generated in spinal surgery based on the number of levels operated on, and they found that as more segments were fused, the revenue generated per minute decreased, concluding that shorter cases with fewer levels of fusion had a higher value. Lastly, a study by Schwartz et al. [31] comparing elective versus emergent general surgery procedures found that the two types of procedures were assigned the same RVUs, despite emergent procedures having higher rates of complication and longer length of stay.

In agreement with our findings, a different study by Sodhi et al. [26] comparing primary to revision total ankle arthroplasty found no differences in operative time between the two procedures, but a higher RVU assignment for revision cases. Our study similarly found no difference in operating time between primary and revision cases, but more RVUs generated in revision cases. Another study in agreement with ours was a general surgery study by Doval et al. [32] which found that although operating room times were longer for revision hernia repair cases, they generated both higher total RVUs as well as higher RVUs per minute compared to primary inguinal hernia repair. The differences in results between our findings as well as those of the two aforementioned studies with the rest of the literature may be explained by the poor correlation of CMS estimates of operative time compared to other reported values [22]. When the CMS estimated times are consistent with or longer than that of actual times, which may be the case with revision ankle and shoulder arthroplasty, surgeons are reimbursed for an equal or greater amount than primary arthroplasty. Conversely, revision estimate times below that of actual times will generate less than primary arthroplasty cases, which may be the case for hip and knee arthroplasty.

This study has several weaknesses that should be considered when interpreting our findings. First, this study utilized a database which can be subject to input errors and can only provide a limited number of metrics regarding a patient’s medical care. Furthermore, the NSQIP database does not provide specific information regarding severity of comorbidities or specific outcomes related to a given procedure. Access to the database is limited to contributing institutions, and the high cost of participation has led to a disproportionate contribution from large teaching hospitals [28,33]. Second, there are not separate CPT codes for anatomic TSA or reverse TSA, so we were not able to identify if a difference in reimbursement exists between the two procedures. Lastly, the database does not contain long-term follow-up information, and thus the effect of factors that may further impact the overall revenue from a procedure, such as management of complications, was not included in our study.

This study is the first to compare reimbursement rates between primary and revision TSA as it relates to operative time. This data improves the understanding of weighted differences in the relative value of primary versus revision TSA, with revision surgery being worth more RVUs, and it adds to the general body of literature regarding the relative value of primary and revision procedures. Our findings can assist providers in understanding reimbursement trends for revision TSA cases.

The relative value of revision TSA procedures is reasonably weighted to account for the increased technical challenges and time associated with these procedures. Surgeons need not be deterred from performing revision TSAs on the historical basis that revision arthroplasty cases in general receive lower compensation than primary cases.