- Email: [email protected]
The Cost-Effectiveness of Using Platelet-Rich Plasma During Rotator Cuff Repair: A Markov Model Analysis
The Cost-Effectiveness of Using Platelet-Rich Plasma During Rotator Cuff Repair: A Markov Model Analysis Eric M. Samuelson, M.D., Susan M. Odum, Ph.D., and James E. Fleischli, M.D.
Purpose: To perform a cost-utility analysis to determine if the use of platelet-rich plasma (PRP) products during arthroscopic rotator cuff repair (RCR) is cost-effective. Methods: A cost-utility analysis was conducted using a Markov decision model. Model inputs including health utility values, retear rates, and transition probabilities were derived from the best evidence available in the literature regarding full-thickness rotator cuff tears and their repair, as well as the augmentation of their repair with PRP. Costs were determined by examining the typical patient undergoing treatment for a full-thickness rotator cuff tear in a private orthopaedic clinic and outpatient surgery center. Results: The cost per quality-adjusted life-year ($/QALY) of RCR with and without PRP was $6,775/QALY and $6,612/QALY, respectively. In our base case, the use of PRP to augment RCR was not cost-effective because it had exactly the same “effectiveness” as RCR without PRP augmentation while being associated with a higher cost (additional $750). Sensitivity analysis showed that to achieve a willingness-to-pay threshold of $50,000/QALY, the addition of PRP would need to be associated with a 9.1% reduction in retear rates. If the cost of PRP were increased to $1,000, the retear rate would need to be reduced by 12.1% to reach this same threshold. This compared with a necessary reduction of only 6.1% if the additional cost of PRP was $500. Conclusions: This cost-utility analysis shows that, currently, the use of PRP to augment RCR is not costeffective. Sensitivity analysis showed that PRP-augmented repairs would have to show a reduced retear rate of at least 9.1% before the additional cost would be considered cost-effective. Level of Evidence: Level III, analysis of Level I, II, and III studies.
P
athology of the rotator cuff is one of the most common causes of shoulder pain; it results in 4.5 million patient visits and more than 75,000 rotator cuff repairs (RCRs) performed each year in the United From the CHI Alegent Creighton Clinic (E.M.S.), Omaha, Nebraska; OrthoCarolina Research Institute (S.M.O.), Charlotte, North Carolina; and OrthoCarolina Sports Medicine Center (J.E.F.), Charlotte, North Carolina, U.S.A. The authors report the following potential conflict of interest or source of funding: This work was funded by the OrthoCarolina Research Institute. S.M.O. receives support from Knee Society and University of North Carolina at Charlotte. J.E.F. receives support from OrthoCarolina Sports Medicine Fellowship. Presented as a poster at the American Orthopaedic Association Annual Meeting, Providence, Rhode Island, June 23-27, 2015, and on the podium at the Mid-America Orthopaedic Association Annual Meeting, Hilton Head, South Carolina, April 22-26, 2015, and the American Orthopaedic Society for Sports Medicine Annual Meeting, Orlando, Florida, July 9-12, 2015. Received February 3, 2015; accepted December 2, 2015. Address correspondence to Eric M. Samuelson, M.D., Department of Orthopaedic Surgery, CHI Health Alegent Creighton Clinic, Immanuel One Professional Center, 6829 N 72nd St, Ste 7500, Omaha, NE 68122, U.S.A. E-mail: [email protected] Ó 2016 by the Arthroscopy Association of North America 0749-8063/1593/$36.00 http://dx.doi.org/10.1016/j.arthro.2015.12.018
States alone.1-5 It is well known that there is a significant rate of recurrent rotator cuff tendon defects (11% to 94%) after RCR.1,3,6,7 Investigative focus to try to improve the rate of rotator cuff healing after repair has largely been on maximizing the biomechanical fixation of the rotator cuff.1,3,7,8 Despite these advances, however, the rate of recurrent defects after RCR remains a significant issue.1,7-9 Attention has recently turned to the biological enhancement of RCRs to reduce this significant retear rate because there seems to be a deficiency in the ability of local cellular and molecular processes to produce robust, long-lasting repair tissue.1,3,8,10-12 These biological enhancements include the application of growth factors and cytokines, use of tissue augmentation/scaffolds, use of gene therapy, and use of tissue engineering.1,10 One of the most studied of these biological factors has been platelet-rich products such as plateletrich plasma (PRP).1,3,6,8,10-17 The most basic definition of PRP is “a sample of autologous blood with concentrations of platelets above baseline values.”1 It is hypothesized that the alpha granules contained in platelets release various growth factors and cytokines in
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2016: pp 1-8
1
2
E. M. SAMUELSON ET AL.
a “physiological balance” at supraphysiological concentrations and that, when applied to a repaired rotator cuff, result in an enhanced healing potential.1,3,11,17,18 Although the benefits of PRP application to musculoskeletal tissues have shown promise in basic science and animal studies, evidence of efficacy in treating orthopaedic conditions in humans has been lacking.1,19-24 Chahal et al.,1 in a systematic review that pooled data from 5 Level I, II, and III studies, noted that there was no significant difference in functional outcome, irrespective of the outcome scoring system used, or in overall retear rates between RCRs with PRP augmentation and those without PRP augmentation. The application of PRP, as well as similar biological products, in an attempt to enhance the biological healing process is associated with a significant cost.12,18 It was estimated that in 2013, the global market for PRP products was $160 million, and this amount is expected to increase to $350 million by 2020.25 With the current concerns about escalating health care costs, physicians and providers need to be increasingly mindful of our resources to ensure that our patients are receiving the most cost-effective treatments available.2,5,18,26 This cost-utility analysis was performed using previously published health-related quality-of-life utility values regarding rotator cuff tears and their repair, as well as the average costs of a patient undergoing treatment for a full-thickness rotator cuff tear at a private orthopaedic clinic and outpatient surgery center. The purpose of this study was to perform a cost-utility analysis to determine if the use of PRP products during arthroscopic RCR is cost-effective. Our hypothesis was that the additional use of PRP to augment arthroscopic RCR would not be considered cost-effective.
Methods Model Structure Publicly available software (TreeAge Pro; TreeAge Software, Williamstown, MA) was used to construct the Markov decision tree model for this study. The base-case scenario is a 60-year-old individual with a full-thickness rotator cuff tear that meets the indications for arthroscopic RCR. There are 2 primary treatment arms in the model: RCR with PRP augmentation and RCR without PRP augmentation (Fig 1). In both treatment arms there are 3 possible postoperative outcome health states: intact repair, asymptomatic retear, and symptomatic retear (Fig 1). Patients who had an intact repair after the initial surgical intervention could continue to have an intact repair or sustain a retear of the repaired rotator cuff (Fig 1). Previous literature has suggested that the majority (85% to 100%) of rotator cuff retears occur within the first 6 months postoperatively.7,27 For this reason, we chose a cycle length of 3 months to capture this early
Fig 1. Markov model diagram showing flow of patients within Markov decision model. (PRP, platelet-rich plasma; RCR, rotator cuff repair.)
failure in healing and its potential negative effect on quality of life. Patients who entered the asymptomatic retear arm of the model could either remain asymptomatic or subsequently become symptomatic (Fig 1). Patients in the symptomatic retear state could remain in the symptomatic state or undergo revision RCR (Fig 1). In the model, patients could only undergo 1 revision arthroscopic RCR. Complications other than rotator cuff retear, such as stiffness and infection, were assumed to be identical between patients treated with PRP and those treated without PRP and were not specifically included in the model. This Markov decision tree model was developed, and subsequent analysis was performed, by 2 of the authors (E.M.S. and S.M.O.). Transition Probabilities As noted earlier, patients in each treatment arm could have an intact repair, asymptomatic retear, or symptomatic retear after arthroscopic repair. It is well known that there is a significant rate of rotator cuff retears after attempted repair (11% to 94%).1,3,6,7 In a systematic review including Level I, II, and III evidence regarding PRP-augmented RCR, Chahal et al.1 reported that there was no statistically significant difference in overall retear rate between patients treated with PRP augmentation and those treated without it. They also reported an overall pooled retear rate of 31% from the 5 included studies, with no significant difference in retear rates between the 2 groups. Therefore, in the model’s base-case scenario, it was assumed that there was an overall retear rate of 31%, with no difference in retear rates between the PRP-augmented and nonePRP-augmented groups (Table 1). Again, all of the retears in the model occurred in the first 6 months after arthroscopic repair, as suggested in follow-up imaging studies by Kluger et al.27 and Miller et al.7 The patients in the model who sustained an asymptomatic retear
3
PRP-AUGMENTED RCR COST-EFFECTIVENESS Table 1. Model Inputs Model Inputs Value
Range
Utility values* Symptomatic RCT
0.67
0.552-0.803
Intact after RCR
0.78
0.74-0.851
0.78 0.67
Assumed to be equal to intact repair Assumed to be equal to symptomatic RCT
31 5
e e
Chahal et al.1 Mather et al.28
10
e
Genuario et al.2
30,572
e
Local data
16,453 750
e e
Local data National data
Asymptomatic retear after RCR Symptomatic retear after RCR Transition probabilities, % Rotator cuff retear Rate of patients with asymptomatic retears becoming symptomatic Rate of patients with symptomatic retears electing to undergo revision RCR Costs, $ Rotator cuff repair without PRP augmentation Revision RCR Cost of PRP augmentation
Sources Vitale et al.5 Mather et al.28 Vitale et al.5 Mather et al.28 Genuario et al.2 and Mather et al.28 Genuario et al.2 and Mather et al.28
NOTE. Dashes indicate that ranges are not applicable for these factors. PRP, platelet-rich plasma; RCR, rotator cuff repair; RCT, rotator cuff tear. *Utility values (range of 0 to 1.0) represent a patient’s preference for a given health state, with a utility value of 1.0 associated with perfect health or the absence of any condition that could negatively affect health-related quality of life.
could either continue to be asymptomatic or subsequently become symptomatic. Literature regarding the rate of asymptomatic rotator cuff retears becoming symptomatic after arthroscopic repair is sparse, but the rate has been estimated to be 5% (Table 1).28 In the model, patients with a symptomatic retear could either continue to be symptomatic or elect to undergo revision arthroscopic RCR. Again, there is a paucity of data regarding the percentage of symptomatic retorn RCRs that undergo revision surgery; however, this rate has been estimated to be 10% (Table 1).2 Costs The costs, in 2014 dollars, used in this study were estimated based on the average patient being seen and treated for a full-thickness rotator cuff tear at a private orthopaedic practice and undergoing surgery at a local outpatient surgery center. The specific costs included in the analysis were as follows: initial outpatient consultation fee, preoperative imaging (radiographs and magnetic resonance imaging [MRI] scan), surgeon fees, anesthesia fees, facility fees, implants (i.e., anchors), durable medical equipment, physical therapy, and the additional cost of PRP application (Table 1). Of note, all patients were assumed to undergo subacromial decompression at the time of the initial RCR. It has been reported that the cost of PRP is between $500 and $1,500 per application.29 In the model, the cost of PRP augmentation was assumed to be $750 (Table 1).26 The cost of the increased operative time needed to perform PRP augmentation, which has been estimated to be more than 10 minutes, was not specifically factored into the model.12
For patients who required revision arthroscopic RCR because they had a significantly symptomatic retear, the additional surgeon, anesthesia, and facility fees for this procedure were included. In addition, the associated postoperative MRI scan (to confirm rotator cuff retear), as well as the cost of implants used during the revision procedure, were taken into account (Table 1). Utility Values and Effectiveness Utility values represent a patient’s preference for a given health state and are used to quantify the effect of disease or injury, as well as its subsequent treatment, on a patient’s health-related quality of life. Utility values are given a range of 0 to 1.0, with a utility value of 1.0 being associated with perfect health or the absence of any condition that could negatively affect healthrelated quality of life. For patients with a condition that negatively affects quality of life, the utility value is somewhere between 0 (death) and 1.0 (perfect health). These utility values are then multiplied by the length of time, in years, an individual spends in the given health state to calculate the quality-adjusted life-year (QALY), which is the effectiveness measure in a cost-utility analysis. QALYs are then combined with the costs of various medical and surgical treatments to report their associated cost per quality-adjusted life-year ($/QALY). This information also allows one to compare numerous treatment options to choose the most cost-effective strategy available by calculating the incremental cost-effectiveness ratio (ICER). Strategies and treatments that have ICERs of less than $50,000/QALY are commonly considered cost-effective.5,28,30
4
E. M. SAMUELSON ET AL.
The utility values used in this study were derived from 2 previous reports regarding the economics and cost-effectiveness of RCR.5,28 Vitale et al.,5 when describing the utility value of patients with a rotator cuff tear, reported values of 0.803 and 0.552 when using the Health Utility Index (HUI) and European Quality of Life Measure (EuroQoL), respectively. They also reported an HUI value of 0.851 and EuroQoL of 0.763 after RCR.5 In another study reporting utility values associated with rotator cuff tears and their repair, Mather et al.28 reported a utility value of 0.66 for patients with a rotator cuff tear and 0.74 for those patients after repair when using the Short Form 12 and Short Form 6D. As one can see, there is significant variation in these utility values depending on the scoring system used. For example, the utility value for a patient with a rotator cuff tear varies from 0.552 to 0.803, and that for a patient after RCR varies from 0.74 to 0.851. In keeping with previous economic studies regarding rotator cuff tears and their repair, we assumed that patients with an asymptomatic retear after RCR had the same utility value as those with an intact repair and that patients with a symptomatic retear had a utility value equal to that of patients with an initial rotator cuff tear.2,28 Not wanting to unduly handicap those with a symptomatic retear after arthroscopic repair or discount its significant negative impact on quality of life, we chose to average these values. Therefore those patients with a symptomatic rotator cuff tear or with a symptomatic retear after RCR were assumed to have a utility value of 0.67 (Table 1). Similarly, a utility value of 0.78 was assigned to those patients with an intact RCR, as well as those with an asymptomatic retear (Table 1). Regarding the utility values of those undergoing RCR with and without PRP augmentation, numerous studies support the notion that there is no significant difference in functional outcome between the 2 groups with respect to Constant score; Simple Shoulder Test (SST) score; American Shoulder and Elbow Surgeons (ASES) score; University of California, Los Angeles (UCLA) score; Disabilities of the Arm, Shoulder and Hand score; and Single Assessment Numeric Evaluation score.3,6,8,10,13-17 Therefore we assumed no difference in utility value between the group treated with PRP augmentation and the group treated without PRP augmentation. Discounting To account for inflation, opportunity costs, and time preferences, both future costs and utility values were discounted at 3%. Threshold Sensitivity Analysis The theoretical benefit of augmenting arthroscopic RCR with PRP is increasing the rate of tendon healing and therefore improving the outcome of repair. With
Table 2. Results of Sensitivity Analysis Regarding Cost of PRP and Associated Reduction in Retear Rate Needed to Achieve Cost-Effectiveness (ICER <$50,000/QALY) Cost of PRP, $ 250 500 750 1,000 1,500
Reduction in Retear Rate Needed (31%) Compared With No PRP Augmentation, % 3.2 6.1 9.1 12.1 18.2
ICER, incremental cost-effectiveness ratio; PRP, platelet-rich plasma; QALY, quality-adjusted life-year.
respect to our base-case scenario, we performed threshold sensitivity analysis regarding the postoperative retear rate to determine at what point the additional use of PRP was considered cost-effective. This threshold was set at $50,000/QALY.5,28,30 We also varied the cost of PRP augmentation to try to account for cost differences in various health care settings and environments. Threshold sensitivity analysis was again performed with these different cost levels to determine at what point the reduction in retear rates met this $50,000/QALY mark.
Results Base Case In this cost-utility analysis, with the assumptions of our base case described earlier, RCR with PRP augmentation was associated with an increased cost of $756.73. There was no difference in effectiveness between patients with and patients without PRP augmentation because the utility values and overall retear rates were assumed to be the same, as reported in the systematic review by Chahal et al.1 Over the 10year course of the model, the $/QALY was $6,612 for arthroscopic RCR without PRP augmentation and $6,775 for RCR with PRP augmentation. We were unable to calculate an ICER for our base-case scenario because it was assumed that there was no difference in outcome (utility values, retear rates, and so on) between the 2 groups. Sensitivity Analysis Base-Case Retear Rates The retear rate for PRP-augmented RCR was varied to determine the magnitude of retear rate reduction needed from baseline to reach an ICER threshold of less than $50,000/QALY. Assuming the baseline additional cost of $750 for augmenting an RCR with PRP, we found that the retear rate would have to be reduced by 9.1% to reach this $50,000/QALY threshold (Table 2). To put it another way, assuming a baseline retear rate of 31% for RCR without PRP augmentation, we found that RCR with PRP augmentation would have to be
5
PRP-AUGMENTED RCR COST-EFFECTIVENESS Table 3. Highest-Quality Evidence in Literature Regarding Retear Rates After Rotator Cuff Repair With and Without PRP Augmentation Retear Rates, % Study Jo et al.15 (AJSM, 2015) Malavolta et al.16 (AJSM, 2014) Ruiz-Moneo et al.17 (Arthroscopy, 2013) Antuna et al.13 (AOB, 2013) Jo et al.14 (AJSM, 2013) Gumina et al.6 (JBJS Am, 2012) Weber et al.12 (AJSM, 2012) Randelli et al.8 (JSES, 2011) Castricini et al.10 (AJSM, 2011) Rodeo et al.11 (AJSM, 2012) Jo et al.3 (AJSM, 2011)
Level of Evidence I I I I I I I I I II II
With PRP 3 7.4 34.4 92.8 20 0 42.8 40 2.5 33.3 26.7
Without PRP 20 18.5 41.9 71.4 55.6 8.1 29.2 52 10.5 19.4 41.2
Statistically Significant Difference Yes, P ¼ .032 No No No Yes, P ¼ .023 Yes, P ¼ .04 No No No No No
AJSM, American Journal of Sports Medicine; AOB, Acta Orthopaedica Belgica; JBJS Am, Journal of Bone and Joint Surgery (American Volume); JSES, Journal of Shoulder and Elbow Surgery; PRP, platelet-rich plasma.
associated with an overall retear rate of 21.9% to be considered cost-effective. Cost of PRP Augmentation and Retear Rates When the cost of PRP augmentation was increased, the magnitude of retear rate reduction needed to be considered cost-effective (ICER <$50,000/QALY threshold) also increased. If the additional cost of PRP augmentation was assumed to be $1,000, a 12.1% reduction in retear rate was necessary to reach the $50,000/QALY threshold (Table 2). Similarly, an 18.2% reduction was needed if the cost of PRP augmentation was $1,500 (Table 2). Likewise, the difference in retear rates needed to be considered cost-effective decreased when the cost of PRP augmentation decreased. If the cost of PRP application were reduced to $500 and to $250, the reduction in retear rates needed to reach the $50,000/QALY threshold would be 6.1% and 3.2%, respectively (Table 2).
Discussion This cost-utility study, using the most current literature available regarding the use of PRP in arthroscopic RCR, shows that the augmentation of these repairs with PRP is not cost-effective. In fact, in our base-case scenario (in which the cost of PRP application is assumed to be $750), the additional use of PRP in arthroscopic RCR would need to be associated with a reduction in the retear rate of approximately 9.1%, as compared with baseline, before its use would be considered costeffective (ICER <$50,000/QALY). The systematic review by Chahal et al.,1 which pooled the results of 5 Level I, II, and III studies regarding the use of PRP to augment RCR, reported a pooled overall retear rate of 31%, with no significant difference between those treated with PRP augmentation and those treated without it. When the authors broke these findings down into groups based on the size of the rotator cuff tear, they found a statistically significant
difference in retear rates in those patients with small- to medium-sized rotator cuff tears. Specifically, they found a 7.9% retear rate in the PRP-augmented group in those patients with small- to medium-size rotator cuff tears compared with a 26.8% retear rate in the nonePRPaugmented group (P ¼ .006). They did not find a statistically significant difference between the groups when looking at large- to massive-sized rotator cuff tears. However, their systematic review, which again included 5 studies with Level I, II, and III evidence, was published in 2012, and there have been several additional Level I and II studies published since regarding PRP-augmented RCRs (Table 3).3,6,8,10-14,15-17 To our knowledge, there are 11 Level I and II reports regarding PRP-augmented RCR published in the literature (Table 3).3,6,8,10-17 Of these Level I and II studies, only 3 have shown a significant reduction in retear rates between the groups.6,14,15 At greater than 1 year of follow-up, MRI scan showed a statistically lower retear rate in the PRPaugmented groups in studies performed by Gumina et al.6 and Jo et al.14,15 (Table 3). In a study examining tears greater than 3 cm, Jo et al.14 reported retear rates of 20% in the PRP-augmented group and 55.6% in the nonePRP-augmented group (P ¼ .023) (Table 3). In another report these same authors found a significant reduction in retear rates in the group with PRP augmentation (20% v 3%, P ¼ .032) when used in tears greater than 1 cm and less than 5 cm.15 The authors of these studies used a “fully automated plateletpheresis system” and standardized the concentration of platelets in the final PRP product (1,000 103 platelets/mL).14 The authors also noted that they used 3 PRP gels per patient and noted the “relative inconvenience and high expense” of their technique.14,15 The study by Gumina et al. reported a retear rate of 0% after the repair of large rotator cuff tears (2 to 4 cm) augmented with PRP compared with an 8.1% retear rate in the nonePRPaugmented group (P ¼ .04, statistically significant) (Table 3). The remaining 8 studies did not show a
6
E. M. SAMUELSON ET AL.
significant difference in retear rates between the groups.3,8,10-13,16,17 In addition, since the initiation of our study, there have been 3 meta-analyses published examining these Level I and II studies.31-33 All of these systematic reviews came to the same conclusion that there are “no statistically significant differences in overall gain in outcome scores or retear rates between treatment groups.”31-33 It should be noted that although the studies by Li et al.31 and Zhao et al.33 did not examine subgroups based on size, the study by Warth et al.32 did find that there was a significant reduction in retear rates in PRP-augmented double-row repairs of rotator cuff tears greater than 3 cm (P ¼ .046). Given the results of these systematic reviews, as well as the conflicting evidence as to whether there is increased benefit for smaller or larger rotator cuff tears, we believe that the use of a 31% retear rate after RCR, as reported in the systematic review by Chahal et al., is reasonable and that, at least at this time, it is also reasonable to assume an equal retear rate between these groups.3 The utility values used in this study were derived from 2 previous reports regarding the cost-effectiveness and economics of RCR.5,28 In these studies, however, there was quite a range in utility values depending on the system used. As one can imagine, these healthrelated quality-of-life scoring systems also had wide variations regarding the range of the utility value difference between patients with a symptomatic rotator cuff tear and those with a healed RCR. When the HUI was used, there was only a 0.048 difference in utility between those patients with a symptomatic tear and those after successful repair.5 This difference was much larger, 0.2, when the EuroQoL scoring system was used.5 We believed that if we chose a system with a large difference in utility values between patients with a symptomatic tear or retear and those with a healed or intact repair, such as the EuroQoL, we would be unduly handicapping those with a symptomatic retear and possibly favoring the PRP augmentation group. In a similar fashion, we did not want to choose a system, such as the HUI, that had a small difference between the 2 groups because this would possibly discount the significant negative impact on quality of life for patients with a symptomatic retear and favor the nonePRPaugmented group. For this reason, we chose to average the utility values from these studies and used a utility value of 0.67 for those patients with a symptomatic rotator cuff tear or symptomatic retear and a value of 0.78 for those patients with a healed or intact RCR or an asymptomatic retear. Most studies have failed to identify significant differences in functional outcomes between patients undergoing RCR with PRP augmentation and those without it.3,6,8,10,13-17 As noted previously, analysis of the pooled data from the systematic review by Chahal
et al.1 showed no significant difference in functional outcome regarding Constant, SST, ASES, UCLA, and Single Assessment Numeric Evaluation scores.3 This article, however, was published in 2012, and there have been several high-quality studies (Level I and II evidence) published since.6,11-17 In a Level II study by Rodeo et al.,11 there was a trend (P ¼ .054) toward better ASES scores in the group that underwent RCR without PRP augmentation at 12 months postoperatively. The authors, however, did not find a significant difference regarding L’Insalata scores between the groups. Weber et al.,12 in a Level I prospective, randomized trial, found a significantly higher UCLA score (P < .046) at 12 months in the nonePRPaugmented RCR group. There was no significant difference in ASES or SST scores between the groups at 1 year of follow-up in their study. The remaining highquality studies also showed no statistically significant differences in functional outcomes at final follow-up between patients undergoing RCR with PRP augmentation and those without it.6,13-17 In addition, 3 recent meta-analyses of these Level I and II studies also showed no significant differences in most of the functional outcome scores measured.31-33 We believe that, overall, these data show no significant difference in outcome between these 2 groups and support the use of equal utility values between groups. There are numerous basic science and animal studies that support the notion that PRP has a beneficial effect on tendon healing.19-24 The application of PRP in these studies has been associated with increased gene expression and production of type I collagen and growth factors (vascular endothelial growth factor, hepatocyte growth factor, and so on), as well as increased cell migration and proliferation.19,21,23,24 With these positive results, it is easy to understand why there has been so much enthusiasm surrounding these products and their potential applications to human musculoskeletal conditions. Unfortunately, it has not yet conclusively been shown that the use of PRP in humans, specifically the use of PRP to augment arthroscopic RCR, has significant benefit.1,31-33 Some of the reasons that PRP products may not have shown a clinical difference are that the concentration and capabilities of the components themselves (platelets, leukocytes, activating factors, patient age, and so on) vary widely between various formulations and products.14,18,22 It is clear that, as suggested by Jo et al.,14 further research is needed regarding the use of PRP to augment RCR and that these future studies should fully characterize the PRP product used regarding the aforementioned factors. These future studies also need to show a significant and substantial benefit to the use of these PRP products to “justify the increases in cost and surgical time associated with this technology.”26
PRP-AUGMENTED RCR COST-EFFECTIVENESS
Limitations There are limitations to our study. First, the assumptions regarding utility values and transition probabilities were derived from the literature. Assumptions regarding the functional outcomes and retear rates of patients undergoing RCR with and without PRP augmentation were derived from the systematic review by Chahal et al.1 Since the initiation of our study, there have been 3 additional metaanalyses published evaluating the Level I and II data regarding PRP-augmented RCR.31-33 We believe that these meta-analyses further confirm the assumptions of our study model.31-33 Although actual retear rates may be higher than the retear rate noted in our study model, possibly up to 20%, we believe that the preponderance of evidence, as noted in these 4 meta-analyses regarding PRP-augmented RCR, suggests that there is no significant difference in overall retear rates between groups. Therefore, whereas the specific $/QALY for these 2 different techniques may vary slightly, the overall conclusion of our analysis that PRP-augmented RCR is less cost-effective than RCR without PRP augmentation would hold true because there is no difference in retear rates between groups. Finally, the costs were estimated by examining the costs for patients undergoing treatment for fullthickness rotator cuff tears in a private orthopaedic clinic and local outpatient surgery center. These costs may not be representative and generalizable to the population as a whole depending on locale, resources, insurance status of the patient, and so on. In addition, the cost of PRP varies substantially ($500 to $1,500 per application).29 We chose a relatively conservative value of $750 in our base-case scenario but performed a sensitivity analysis regarding this value to determine at what point the use of PRP to augment RCR would be considered cost-effective.
Conclusions This cost-utility analysis shows that, currently, the use of PRP to augment RCR is not cost-effective. Sensitivity analysis showed that PRP-augmented repairs would have to show a reduced retear rate of at least 9.1% before the additional cost would be considered costeffective.
References 1. Chahal J, Van Thiel GS, Mall N, et al. The role of plateletrich plasma in arthroscopic rotator cuff repair: A systematic review with quantitative synthesis. Arthroscopy 2012;28:1718-1727. 2. Genuario JW, Donegan RP, Hamman D, et al. The costeffectiveness of single-row compared with double-row arthroscopic rotator cuff repair. J Bone Joint Surg Am 2012;94:1369-1377.
7
3. Jo CH, Kim JE, Yoon KS, et al. Does platelet-rich plasma accelerate recovery after rotator cuff repair? A prospective cohort study. Am J Sports Med 2011;39:2082-2090. 4. Yamaguchi K, Ditsios K, Middleton WD, Hildebolt CF, Galatz LM, Teefey SA. The demographic and morphological features of rotator cuff disease: A comparison of asymptomatic and symptomatic shoulders. J Bone Joint Surg Am 2006;88:1699-1704. 5. Vitale MA, Vitale MG, Zivin JG, Braman JP, Bigliani LU, Flatow EL. Rotator cuff repair: An analysis of utility scores and cost-effectiveness. J Shoulder Elbow Surg 2007;16:181-187. 6. Gumina S, Campagna V, Ferrazza G, et al. Use of plateletleukocyte membrane in arthroscopic repair of large rotator cuff tears: A prospective randomized study. J Bone Joint Surg Am 2012;94:1345-1352. 7. Miller BS, Downie BK, Kohen RB, et al. When do rotator cuff repairs fail? Serial ultrasound examination after arthroscopic repair of large and massive rotator cuff tears. Am J Sports Med 2011;39:2064-2070. 8. Randelli P, Arrigoni P, Ragone V, Aliprandi A, Cabitza P. Platelet rich plasma in arthroscopic rotator cuff repair: A prospective RCT study, 2-year follow-up. J Shoulder Elbow Surg 2011;20:518-528. 9. Dehaan AM, Axelrod TW, Kaye E, Silvestri L, Puskas B, Foster TE. Does double-row rotator cuff repair improve functional outcome of patients compared with single-row technique? A systematic review. Am J Sports Med 2012;40: 1176-1185. 10. Castricini R, Longo UG, Benedetto MD, et al. Platelet-rich plasma augmentation for arthroscopic rotator cuff repair: A randomized controlled trial. Am J Sports Med 2011;39: 258-265. 11. Rodeo SA, Delos D, Williams RJ, Adler RS, Pearle A, Warren RF. The effect of platelet-rich fibrin matrix on rotator cuff tendon healing: A prospective, randomized clinical study. Am J Sports Med 2012;40:1234-1241. 12. Weber SC, Kauffman JI, Parise C, Weber SJ, Katz SD. Platelet-rich fibrin matrix in the management of arthroscopic repair of the rotator cuff: A prospective, randomized, double-blinded study. Am J Sports Med 2012;41: 263-270. 13. Antuna S, Barco R, Martinez-Diez JM, SanchezMarquez JM. Platelet-rich fibrin in arthroscopic repair of massive rotator cuff tears: A prospective randomized pilot clinical trial. Acta Orthop Belg 2013;79:25-30. 14. Jo CH, Shin JS, Lee YG, et al. Platelet-rich plasma for arthroscopic repair of large to massive rotator cuff tears: A randomized, single-blind, parallel-group trial. Am J Sports Med 2013;41:2240-2248. 15. Jo CH, Shin JS, Shin WH, Lee SY, Yoon KS, Shin S. Platelet-rich plasma for arthroscopic repair of medium to large rotator cuff tears: A randomized controlled trial. Am J Sports Med 2015;43:2102-2110. 16. Malavolta EA, Gracitelli MEC, Neto AAF, Assuncao JH, Bordalo-Rodreigues M, de Camargo OP. Platelet-rich plasma in rotator cuff repair: A prospective randomized study. Am J Sports Med 2014;42:2446-2454. 17. Ruiz-Moneo P, Molano-Munoz J, Prieto E, Algorta J. Plasma rich in growth factors in arthroscopic rotator cuff repair: A randomized double-blind, controlled clinic trial. Arthroscopy 2013;29:2-9.
8
E. M. SAMUELSON ET AL.
18. Hsu WK, Mishra A, Rodeo S, et al. Platelet-rich plasma in orthopaedic applications: Evidence-based recommendations for treatment. J Am Acad Orthop Surg 2013;21: 739-748. 19. Anitua E, Andia I, Sanchez M, et al. Autologous preparations rich in growth factors promote proliferation and induce VEGF and HGF production by human tendon cells in culture. J Orthop Res 2005;23:281-286. 20. Aspenberg P, Virchenko O. Platelet concentrate injection improves Achilles tendon repair in rats. Acta Orthop Scand 2004;75:93-99. 21. de Mos M, van der Windt A, Holger J, et al. Can platelet rich plasma enhance tendon repair: A cell culture study. Am J Sports Med 2008;36:1171-1178. 22. Hall MP, Band PA, Meislin RJ, Jazrawi LM, Cardone DA. Platelet-rich plasma: Current concepts and application in sports medicine. J Am Acad Orthop Surg 2009;17:602-608. 23. Jo CH, Kim JE, Yoon KS, Shin S. Platelet-rich plasma stimulates cell proliferation and enhances matrix gene expression and synthesis in tenocytes from human rotator cuff tendons with degenerative tears. Am J Sports Med 2012;40:1035-1045. 24. Kajikawa Y, Morihara T, Sakamoto H, et al. Platelet-rich plasma enhances the initial mobilization of circulation derived cells for tendon healing. J Cell Physiol 2008;215: 837-845. 25. Platelet rich plasma market: Asia Pacific to show robust growth between 2014 and 2020. Available from: http://www.transparencymarketresearch.com/pressrelease/ platelet-rich-plasma-market.htm. Accessed June 25, 2015.
26. Lee DH. Platelet-rich plasma: Is it ready for prime time? Commentary and perspective. J Bone Joint Surg Am 2012;94:e114. 27. Kluger R, Bock P, Mittlbock M, Krampla W, Engel A. Long-term survivorship of rotator cuff repairs using ultrasound and magnetic resonance imaging analysis. Am J Sports Med 2011;39:2071-2081. 28. Mather RC III, Koenig L, Acevedo D, et al. The societal and economic value of rotator cuff repair. J Bone Joint Surg Am 2013;95:1993-2000. 29. Platelet rich plasma (PRP) Therapy reimbursement overview. Available from: http://www.rsmedical.com/ documents/331-0034-00-31reva.pdf. Accessed September 23, 2014. 30. Mather RC III, Hettrich CM, Dunn WR, et al. Cost-effectiveness analysis of early reconstruction versus rehabilitation and delayed reconstruction for anterior cruciate ligament tears. Am J Sports Med 2014;42:1583-1591. 31. Li X, Xu CP, Hou YL, Song JQ, Cui Z, Yu B. Are platelet concentrates an ideal biomaterial for arthroscopic rotator cuff repair? A meta-analysis of randomized controlled trials. Arthroscopy 2014;30:1483-1490. 32. Warth RJ, Dornan GJ, James EW, Horan MP, Millett PJ. Clinical and structural outcomes after arthroscopic repair of full-thickness rotator cuff tears with and without platelet-rich product supplementation: A meta-analysis and meta-regression. Arthroscopy 2015;31:306-320. 33. Zhao JG, Zhao L, Jiang YX, Wang ZL, Wang J, Zhang P. Platelet-rich plasma in arthroscopic rotator cuff repair: A meta-analysis of randomized controlled trials. Arthroscopy 2015;31:125-135.
Purpose: To perform a cost-utility analysis to determine if the use of platelet-rich plasma (PRP) products during arthroscopic rotator cuff repair (RCR) is cost-effective. Methods: A cost-utility analysis was conducted using a Markov decision model. Model inputs including health utility values, retear rates, and transition probabilities were derived from the best evidence available in the literature regarding full-thickness rotator cuff tears and their repair, as well as the augmentation of their repair with PRP. Costs were determined by examining the typical patient undergoing treatment for a full-thickness rotator cuff tear in a private orthopaedic clinic and outpatient surgery center. Results: The cost per quality-adjusted life-year ($/QALY) of RCR with and without PRP was $6,775/QALY and $6,612/QALY, respectively. In our base case, the use of PRP to augment RCR was not cost-effective because it had exactly the same “effectiveness” as RCR without PRP augmentation while being associated with a higher cost (additional $750). Sensitivity analysis showed that to achieve a willingness-to-pay threshold of $50,000/QALY, the addition of PRP would need to be associated with a 9.1% reduction in retear rates. If the cost of PRP were increased to $1,000, the retear rate would need to be reduced by 12.1% to reach this same threshold. This compared with a necessary reduction of only 6.1% if the additional cost of PRP was $500. Conclusions: This cost-utility analysis shows that, currently, the use of PRP to augment RCR is not costeffective. Sensitivity analysis showed that PRP-augmented repairs would have to show a reduced retear rate of at least 9.1% before the additional cost would be considered cost-effective. Level of Evidence: Level III, analysis of Level I, II, and III studies.
P
athology of the rotator cuff is one of the most common causes of shoulder pain; it results in 4.5 million patient visits and more than 75,000 rotator cuff repairs (RCRs) performed each year in the United From the CHI Alegent Creighton Clinic (E.M.S.), Omaha, Nebraska; OrthoCarolina Research Institute (S.M.O.), Charlotte, North Carolina; and OrthoCarolina Sports Medicine Center (J.E.F.), Charlotte, North Carolina, U.S.A. The authors report the following potential conflict of interest or source of funding: This work was funded by the OrthoCarolina Research Institute. S.M.O. receives support from Knee Society and University of North Carolina at Charlotte. J.E.F. receives support from OrthoCarolina Sports Medicine Fellowship. Presented as a poster at the American Orthopaedic Association Annual Meeting, Providence, Rhode Island, June 23-27, 2015, and on the podium at the Mid-America Orthopaedic Association Annual Meeting, Hilton Head, South Carolina, April 22-26, 2015, and the American Orthopaedic Society for Sports Medicine Annual Meeting, Orlando, Florida, July 9-12, 2015. Received February 3, 2015; accepted December 2, 2015. Address correspondence to Eric M. Samuelson, M.D., Department of Orthopaedic Surgery, CHI Health Alegent Creighton Clinic, Immanuel One Professional Center, 6829 N 72nd St, Ste 7500, Omaha, NE 68122, U.S.A. E-mail: [email protected] Ó 2016 by the Arthroscopy Association of North America 0749-8063/1593/$36.00 http://dx.doi.org/10.1016/j.arthro.2015.12.018
States alone.1-5 It is well known that there is a significant rate of recurrent rotator cuff tendon defects (11% to 94%) after RCR.1,3,6,7 Investigative focus to try to improve the rate of rotator cuff healing after repair has largely been on maximizing the biomechanical fixation of the rotator cuff.1,3,7,8 Despite these advances, however, the rate of recurrent defects after RCR remains a significant issue.1,7-9 Attention has recently turned to the biological enhancement of RCRs to reduce this significant retear rate because there seems to be a deficiency in the ability of local cellular and molecular processes to produce robust, long-lasting repair tissue.1,3,8,10-12 These biological enhancements include the application of growth factors and cytokines, use of tissue augmentation/scaffolds, use of gene therapy, and use of tissue engineering.1,10 One of the most studied of these biological factors has been platelet-rich products such as plateletrich plasma (PRP).1,3,6,8,10-17 The most basic definition of PRP is “a sample of autologous blood with concentrations of platelets above baseline values.”1 It is hypothesized that the alpha granules contained in platelets release various growth factors and cytokines in
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2016: pp 1-8
1
2
E. M. SAMUELSON ET AL.
a “physiological balance” at supraphysiological concentrations and that, when applied to a repaired rotator cuff, result in an enhanced healing potential.1,3,11,17,18 Although the benefits of PRP application to musculoskeletal tissues have shown promise in basic science and animal studies, evidence of efficacy in treating orthopaedic conditions in humans has been lacking.1,19-24 Chahal et al.,1 in a systematic review that pooled data from 5 Level I, II, and III studies, noted that there was no significant difference in functional outcome, irrespective of the outcome scoring system used, or in overall retear rates between RCRs with PRP augmentation and those without PRP augmentation. The application of PRP, as well as similar biological products, in an attempt to enhance the biological healing process is associated with a significant cost.12,18 It was estimated that in 2013, the global market for PRP products was $160 million, and this amount is expected to increase to $350 million by 2020.25 With the current concerns about escalating health care costs, physicians and providers need to be increasingly mindful of our resources to ensure that our patients are receiving the most cost-effective treatments available.2,5,18,26 This cost-utility analysis was performed using previously published health-related quality-of-life utility values regarding rotator cuff tears and their repair, as well as the average costs of a patient undergoing treatment for a full-thickness rotator cuff tear at a private orthopaedic clinic and outpatient surgery center. The purpose of this study was to perform a cost-utility analysis to determine if the use of PRP products during arthroscopic RCR is cost-effective. Our hypothesis was that the additional use of PRP to augment arthroscopic RCR would not be considered cost-effective.
Methods Model Structure Publicly available software (TreeAge Pro; TreeAge Software, Williamstown, MA) was used to construct the Markov decision tree model for this study. The base-case scenario is a 60-year-old individual with a full-thickness rotator cuff tear that meets the indications for arthroscopic RCR. There are 2 primary treatment arms in the model: RCR with PRP augmentation and RCR without PRP augmentation (Fig 1). In both treatment arms there are 3 possible postoperative outcome health states: intact repair, asymptomatic retear, and symptomatic retear (Fig 1). Patients who had an intact repair after the initial surgical intervention could continue to have an intact repair or sustain a retear of the repaired rotator cuff (Fig 1). Previous literature has suggested that the majority (85% to 100%) of rotator cuff retears occur within the first 6 months postoperatively.7,27 For this reason, we chose a cycle length of 3 months to capture this early
Fig 1. Markov model diagram showing flow of patients within Markov decision model. (PRP, platelet-rich plasma; RCR, rotator cuff repair.)
failure in healing and its potential negative effect on quality of life. Patients who entered the asymptomatic retear arm of the model could either remain asymptomatic or subsequently become symptomatic (Fig 1). Patients in the symptomatic retear state could remain in the symptomatic state or undergo revision RCR (Fig 1). In the model, patients could only undergo 1 revision arthroscopic RCR. Complications other than rotator cuff retear, such as stiffness and infection, were assumed to be identical between patients treated with PRP and those treated without PRP and were not specifically included in the model. This Markov decision tree model was developed, and subsequent analysis was performed, by 2 of the authors (E.M.S. and S.M.O.). Transition Probabilities As noted earlier, patients in each treatment arm could have an intact repair, asymptomatic retear, or symptomatic retear after arthroscopic repair. It is well known that there is a significant rate of rotator cuff retears after attempted repair (11% to 94%).1,3,6,7 In a systematic review including Level I, II, and III evidence regarding PRP-augmented RCR, Chahal et al.1 reported that there was no statistically significant difference in overall retear rate between patients treated with PRP augmentation and those treated without it. They also reported an overall pooled retear rate of 31% from the 5 included studies, with no significant difference in retear rates between the 2 groups. Therefore, in the model’s base-case scenario, it was assumed that there was an overall retear rate of 31%, with no difference in retear rates between the PRP-augmented and nonePRP-augmented groups (Table 1). Again, all of the retears in the model occurred in the first 6 months after arthroscopic repair, as suggested in follow-up imaging studies by Kluger et al.27 and Miller et al.7 The patients in the model who sustained an asymptomatic retear
3
PRP-AUGMENTED RCR COST-EFFECTIVENESS Table 1. Model Inputs Model Inputs Value
Range
Utility values* Symptomatic RCT
0.67
0.552-0.803
Intact after RCR
0.78
0.74-0.851
0.78 0.67
Assumed to be equal to intact repair Assumed to be equal to symptomatic RCT
31 5
e e
Chahal et al.1 Mather et al.28
10
e
Genuario et al.2
30,572
e
Local data
16,453 750
e e
Local data National data
Asymptomatic retear after RCR Symptomatic retear after RCR Transition probabilities, % Rotator cuff retear Rate of patients with asymptomatic retears becoming symptomatic Rate of patients with symptomatic retears electing to undergo revision RCR Costs, $ Rotator cuff repair without PRP augmentation Revision RCR Cost of PRP augmentation
Sources Vitale et al.5 Mather et al.28 Vitale et al.5 Mather et al.28 Genuario et al.2 and Mather et al.28 Genuario et al.2 and Mather et al.28
NOTE. Dashes indicate that ranges are not applicable for these factors. PRP, platelet-rich plasma; RCR, rotator cuff repair; RCT, rotator cuff tear. *Utility values (range of 0 to 1.0) represent a patient’s preference for a given health state, with a utility value of 1.0 associated with perfect health or the absence of any condition that could negatively affect health-related quality of life.
could either continue to be asymptomatic or subsequently become symptomatic. Literature regarding the rate of asymptomatic rotator cuff retears becoming symptomatic after arthroscopic repair is sparse, but the rate has been estimated to be 5% (Table 1).28 In the model, patients with a symptomatic retear could either continue to be symptomatic or elect to undergo revision arthroscopic RCR. Again, there is a paucity of data regarding the percentage of symptomatic retorn RCRs that undergo revision surgery; however, this rate has been estimated to be 10% (Table 1).2 Costs The costs, in 2014 dollars, used in this study were estimated based on the average patient being seen and treated for a full-thickness rotator cuff tear at a private orthopaedic practice and undergoing surgery at a local outpatient surgery center. The specific costs included in the analysis were as follows: initial outpatient consultation fee, preoperative imaging (radiographs and magnetic resonance imaging [MRI] scan), surgeon fees, anesthesia fees, facility fees, implants (i.e., anchors), durable medical equipment, physical therapy, and the additional cost of PRP application (Table 1). Of note, all patients were assumed to undergo subacromial decompression at the time of the initial RCR. It has been reported that the cost of PRP is between $500 and $1,500 per application.29 In the model, the cost of PRP augmentation was assumed to be $750 (Table 1).26 The cost of the increased operative time needed to perform PRP augmentation, which has been estimated to be more than 10 minutes, was not specifically factored into the model.12
For patients who required revision arthroscopic RCR because they had a significantly symptomatic retear, the additional surgeon, anesthesia, and facility fees for this procedure were included. In addition, the associated postoperative MRI scan (to confirm rotator cuff retear), as well as the cost of implants used during the revision procedure, were taken into account (Table 1). Utility Values and Effectiveness Utility values represent a patient’s preference for a given health state and are used to quantify the effect of disease or injury, as well as its subsequent treatment, on a patient’s health-related quality of life. Utility values are given a range of 0 to 1.0, with a utility value of 1.0 being associated with perfect health or the absence of any condition that could negatively affect healthrelated quality of life. For patients with a condition that negatively affects quality of life, the utility value is somewhere between 0 (death) and 1.0 (perfect health). These utility values are then multiplied by the length of time, in years, an individual spends in the given health state to calculate the quality-adjusted life-year (QALY), which is the effectiveness measure in a cost-utility analysis. QALYs are then combined with the costs of various medical and surgical treatments to report their associated cost per quality-adjusted life-year ($/QALY). This information also allows one to compare numerous treatment options to choose the most cost-effective strategy available by calculating the incremental cost-effectiveness ratio (ICER). Strategies and treatments that have ICERs of less than $50,000/QALY are commonly considered cost-effective.5,28,30
4
E. M. SAMUELSON ET AL.
The utility values used in this study were derived from 2 previous reports regarding the economics and cost-effectiveness of RCR.5,28 Vitale et al.,5 when describing the utility value of patients with a rotator cuff tear, reported values of 0.803 and 0.552 when using the Health Utility Index (HUI) and European Quality of Life Measure (EuroQoL), respectively. They also reported an HUI value of 0.851 and EuroQoL of 0.763 after RCR.5 In another study reporting utility values associated with rotator cuff tears and their repair, Mather et al.28 reported a utility value of 0.66 for patients with a rotator cuff tear and 0.74 for those patients after repair when using the Short Form 12 and Short Form 6D. As one can see, there is significant variation in these utility values depending on the scoring system used. For example, the utility value for a patient with a rotator cuff tear varies from 0.552 to 0.803, and that for a patient after RCR varies from 0.74 to 0.851. In keeping with previous economic studies regarding rotator cuff tears and their repair, we assumed that patients with an asymptomatic retear after RCR had the same utility value as those with an intact repair and that patients with a symptomatic retear had a utility value equal to that of patients with an initial rotator cuff tear.2,28 Not wanting to unduly handicap those with a symptomatic retear after arthroscopic repair or discount its significant negative impact on quality of life, we chose to average these values. Therefore those patients with a symptomatic rotator cuff tear or with a symptomatic retear after RCR were assumed to have a utility value of 0.67 (Table 1). Similarly, a utility value of 0.78 was assigned to those patients with an intact RCR, as well as those with an asymptomatic retear (Table 1). Regarding the utility values of those undergoing RCR with and without PRP augmentation, numerous studies support the notion that there is no significant difference in functional outcome between the 2 groups with respect to Constant score; Simple Shoulder Test (SST) score; American Shoulder and Elbow Surgeons (ASES) score; University of California, Los Angeles (UCLA) score; Disabilities of the Arm, Shoulder and Hand score; and Single Assessment Numeric Evaluation score.3,6,8,10,13-17 Therefore we assumed no difference in utility value between the group treated with PRP augmentation and the group treated without PRP augmentation. Discounting To account for inflation, opportunity costs, and time preferences, both future costs and utility values were discounted at 3%. Threshold Sensitivity Analysis The theoretical benefit of augmenting arthroscopic RCR with PRP is increasing the rate of tendon healing and therefore improving the outcome of repair. With
Table 2. Results of Sensitivity Analysis Regarding Cost of PRP and Associated Reduction in Retear Rate Needed to Achieve Cost-Effectiveness (ICER <$50,000/QALY) Cost of PRP, $ 250 500 750 1,000 1,500
Reduction in Retear Rate Needed (31%) Compared With No PRP Augmentation, % 3.2 6.1 9.1 12.1 18.2
ICER, incremental cost-effectiveness ratio; PRP, platelet-rich plasma; QALY, quality-adjusted life-year.
respect to our base-case scenario, we performed threshold sensitivity analysis regarding the postoperative retear rate to determine at what point the additional use of PRP was considered cost-effective. This threshold was set at $50,000/QALY.5,28,30 We also varied the cost of PRP augmentation to try to account for cost differences in various health care settings and environments. Threshold sensitivity analysis was again performed with these different cost levels to determine at what point the reduction in retear rates met this $50,000/QALY mark.
Results Base Case In this cost-utility analysis, with the assumptions of our base case described earlier, RCR with PRP augmentation was associated with an increased cost of $756.73. There was no difference in effectiveness between patients with and patients without PRP augmentation because the utility values and overall retear rates were assumed to be the same, as reported in the systematic review by Chahal et al.1 Over the 10year course of the model, the $/QALY was $6,612 for arthroscopic RCR without PRP augmentation and $6,775 for RCR with PRP augmentation. We were unable to calculate an ICER for our base-case scenario because it was assumed that there was no difference in outcome (utility values, retear rates, and so on) between the 2 groups. Sensitivity Analysis Base-Case Retear Rates The retear rate for PRP-augmented RCR was varied to determine the magnitude of retear rate reduction needed from baseline to reach an ICER threshold of less than $50,000/QALY. Assuming the baseline additional cost of $750 for augmenting an RCR with PRP, we found that the retear rate would have to be reduced by 9.1% to reach this $50,000/QALY threshold (Table 2). To put it another way, assuming a baseline retear rate of 31% for RCR without PRP augmentation, we found that RCR with PRP augmentation would have to be
5
PRP-AUGMENTED RCR COST-EFFECTIVENESS Table 3. Highest-Quality Evidence in Literature Regarding Retear Rates After Rotator Cuff Repair With and Without PRP Augmentation Retear Rates, % Study Jo et al.15 (AJSM, 2015) Malavolta et al.16 (AJSM, 2014) Ruiz-Moneo et al.17 (Arthroscopy, 2013) Antuna et al.13 (AOB, 2013) Jo et al.14 (AJSM, 2013) Gumina et al.6 (JBJS Am, 2012) Weber et al.12 (AJSM, 2012) Randelli et al.8 (JSES, 2011) Castricini et al.10 (AJSM, 2011) Rodeo et al.11 (AJSM, 2012) Jo et al.3 (AJSM, 2011)
Level of Evidence I I I I I I I I I II II
With PRP 3 7.4 34.4 92.8 20 0 42.8 40 2.5 33.3 26.7
Without PRP 20 18.5 41.9 71.4 55.6 8.1 29.2 52 10.5 19.4 41.2
Statistically Significant Difference Yes, P ¼ .032 No No No Yes, P ¼ .023 Yes, P ¼ .04 No No No No No
AJSM, American Journal of Sports Medicine; AOB, Acta Orthopaedica Belgica; JBJS Am, Journal of Bone and Joint Surgery (American Volume); JSES, Journal of Shoulder and Elbow Surgery; PRP, platelet-rich plasma.
associated with an overall retear rate of 21.9% to be considered cost-effective. Cost of PRP Augmentation and Retear Rates When the cost of PRP augmentation was increased, the magnitude of retear rate reduction needed to be considered cost-effective (ICER <$50,000/QALY threshold) also increased. If the additional cost of PRP augmentation was assumed to be $1,000, a 12.1% reduction in retear rate was necessary to reach the $50,000/QALY threshold (Table 2). Similarly, an 18.2% reduction was needed if the cost of PRP augmentation was $1,500 (Table 2). Likewise, the difference in retear rates needed to be considered cost-effective decreased when the cost of PRP augmentation decreased. If the cost of PRP application were reduced to $500 and to $250, the reduction in retear rates needed to reach the $50,000/QALY threshold would be 6.1% and 3.2%, respectively (Table 2).
Discussion This cost-utility study, using the most current literature available regarding the use of PRP in arthroscopic RCR, shows that the augmentation of these repairs with PRP is not cost-effective. In fact, in our base-case scenario (in which the cost of PRP application is assumed to be $750), the additional use of PRP in arthroscopic RCR would need to be associated with a reduction in the retear rate of approximately 9.1%, as compared with baseline, before its use would be considered costeffective (ICER <$50,000/QALY). The systematic review by Chahal et al.,1 which pooled the results of 5 Level I, II, and III studies regarding the use of PRP to augment RCR, reported a pooled overall retear rate of 31%, with no significant difference between those treated with PRP augmentation and those treated without it. When the authors broke these findings down into groups based on the size of the rotator cuff tear, they found a statistically significant
difference in retear rates in those patients with small- to medium-sized rotator cuff tears. Specifically, they found a 7.9% retear rate in the PRP-augmented group in those patients with small- to medium-size rotator cuff tears compared with a 26.8% retear rate in the nonePRPaugmented group (P ¼ .006). They did not find a statistically significant difference between the groups when looking at large- to massive-sized rotator cuff tears. However, their systematic review, which again included 5 studies with Level I, II, and III evidence, was published in 2012, and there have been several additional Level I and II studies published since regarding PRP-augmented RCRs (Table 3).3,6,8,10-14,15-17 To our knowledge, there are 11 Level I and II reports regarding PRP-augmented RCR published in the literature (Table 3).3,6,8,10-17 Of these Level I and II studies, only 3 have shown a significant reduction in retear rates between the groups.6,14,15 At greater than 1 year of follow-up, MRI scan showed a statistically lower retear rate in the PRPaugmented groups in studies performed by Gumina et al.6 and Jo et al.14,15 (Table 3). In a study examining tears greater than 3 cm, Jo et al.14 reported retear rates of 20% in the PRP-augmented group and 55.6% in the nonePRP-augmented group (P ¼ .023) (Table 3). In another report these same authors found a significant reduction in retear rates in the group with PRP augmentation (20% v 3%, P ¼ .032) when used in tears greater than 1 cm and less than 5 cm.15 The authors of these studies used a “fully automated plateletpheresis system” and standardized the concentration of platelets in the final PRP product (1,000 103 platelets/mL).14 The authors also noted that they used 3 PRP gels per patient and noted the “relative inconvenience and high expense” of their technique.14,15 The study by Gumina et al. reported a retear rate of 0% after the repair of large rotator cuff tears (2 to 4 cm) augmented with PRP compared with an 8.1% retear rate in the nonePRPaugmented group (P ¼ .04, statistically significant) (Table 3). The remaining 8 studies did not show a
6
E. M. SAMUELSON ET AL.
significant difference in retear rates between the groups.3,8,10-13,16,17 In addition, since the initiation of our study, there have been 3 meta-analyses published examining these Level I and II studies.31-33 All of these systematic reviews came to the same conclusion that there are “no statistically significant differences in overall gain in outcome scores or retear rates between treatment groups.”31-33 It should be noted that although the studies by Li et al.31 and Zhao et al.33 did not examine subgroups based on size, the study by Warth et al.32 did find that there was a significant reduction in retear rates in PRP-augmented double-row repairs of rotator cuff tears greater than 3 cm (P ¼ .046). Given the results of these systematic reviews, as well as the conflicting evidence as to whether there is increased benefit for smaller or larger rotator cuff tears, we believe that the use of a 31% retear rate after RCR, as reported in the systematic review by Chahal et al., is reasonable and that, at least at this time, it is also reasonable to assume an equal retear rate between these groups.3 The utility values used in this study were derived from 2 previous reports regarding the cost-effectiveness and economics of RCR.5,28 In these studies, however, there was quite a range in utility values depending on the system used. As one can imagine, these healthrelated quality-of-life scoring systems also had wide variations regarding the range of the utility value difference between patients with a symptomatic rotator cuff tear and those with a healed RCR. When the HUI was used, there was only a 0.048 difference in utility between those patients with a symptomatic tear and those after successful repair.5 This difference was much larger, 0.2, when the EuroQoL scoring system was used.5 We believed that if we chose a system with a large difference in utility values between patients with a symptomatic tear or retear and those with a healed or intact repair, such as the EuroQoL, we would be unduly handicapping those with a symptomatic retear and possibly favoring the PRP augmentation group. In a similar fashion, we did not want to choose a system, such as the HUI, that had a small difference between the 2 groups because this would possibly discount the significant negative impact on quality of life for patients with a symptomatic retear and favor the nonePRPaugmented group. For this reason, we chose to average the utility values from these studies and used a utility value of 0.67 for those patients with a symptomatic rotator cuff tear or symptomatic retear and a value of 0.78 for those patients with a healed or intact RCR or an asymptomatic retear. Most studies have failed to identify significant differences in functional outcomes between patients undergoing RCR with PRP augmentation and those without it.3,6,8,10,13-17 As noted previously, analysis of the pooled data from the systematic review by Chahal
et al.1 showed no significant difference in functional outcome regarding Constant, SST, ASES, UCLA, and Single Assessment Numeric Evaluation scores.3 This article, however, was published in 2012, and there have been several high-quality studies (Level I and II evidence) published since.6,11-17 In a Level II study by Rodeo et al.,11 there was a trend (P ¼ .054) toward better ASES scores in the group that underwent RCR without PRP augmentation at 12 months postoperatively. The authors, however, did not find a significant difference regarding L’Insalata scores between the groups. Weber et al.,12 in a Level I prospective, randomized trial, found a significantly higher UCLA score (P < .046) at 12 months in the nonePRPaugmented RCR group. There was no significant difference in ASES or SST scores between the groups at 1 year of follow-up in their study. The remaining highquality studies also showed no statistically significant differences in functional outcomes at final follow-up between patients undergoing RCR with PRP augmentation and those without it.6,13-17 In addition, 3 recent meta-analyses of these Level I and II studies also showed no significant differences in most of the functional outcome scores measured.31-33 We believe that, overall, these data show no significant difference in outcome between these 2 groups and support the use of equal utility values between groups. There are numerous basic science and animal studies that support the notion that PRP has a beneficial effect on tendon healing.19-24 The application of PRP in these studies has been associated with increased gene expression and production of type I collagen and growth factors (vascular endothelial growth factor, hepatocyte growth factor, and so on), as well as increased cell migration and proliferation.19,21,23,24 With these positive results, it is easy to understand why there has been so much enthusiasm surrounding these products and their potential applications to human musculoskeletal conditions. Unfortunately, it has not yet conclusively been shown that the use of PRP in humans, specifically the use of PRP to augment arthroscopic RCR, has significant benefit.1,31-33 Some of the reasons that PRP products may not have shown a clinical difference are that the concentration and capabilities of the components themselves (platelets, leukocytes, activating factors, patient age, and so on) vary widely between various formulations and products.14,18,22 It is clear that, as suggested by Jo et al.,14 further research is needed regarding the use of PRP to augment RCR and that these future studies should fully characterize the PRP product used regarding the aforementioned factors. These future studies also need to show a significant and substantial benefit to the use of these PRP products to “justify the increases in cost and surgical time associated with this technology.”26
PRP-AUGMENTED RCR COST-EFFECTIVENESS
Limitations There are limitations to our study. First, the assumptions regarding utility values and transition probabilities were derived from the literature. Assumptions regarding the functional outcomes and retear rates of patients undergoing RCR with and without PRP augmentation were derived from the systematic review by Chahal et al.1 Since the initiation of our study, there have been 3 additional metaanalyses published evaluating the Level I and II data regarding PRP-augmented RCR.31-33 We believe that these meta-analyses further confirm the assumptions of our study model.31-33 Although actual retear rates may be higher than the retear rate noted in our study model, possibly up to 20%, we believe that the preponderance of evidence, as noted in these 4 meta-analyses regarding PRP-augmented RCR, suggests that there is no significant difference in overall retear rates between groups. Therefore, whereas the specific $/QALY for these 2 different techniques may vary slightly, the overall conclusion of our analysis that PRP-augmented RCR is less cost-effective than RCR without PRP augmentation would hold true because there is no difference in retear rates between groups. Finally, the costs were estimated by examining the costs for patients undergoing treatment for fullthickness rotator cuff tears in a private orthopaedic clinic and local outpatient surgery center. These costs may not be representative and generalizable to the population as a whole depending on locale, resources, insurance status of the patient, and so on. In addition, the cost of PRP varies substantially ($500 to $1,500 per application).29 We chose a relatively conservative value of $750 in our base-case scenario but performed a sensitivity analysis regarding this value to determine at what point the use of PRP to augment RCR would be considered cost-effective.
Conclusions This cost-utility analysis shows that, currently, the use of PRP to augment RCR is not cost-effective. Sensitivity analysis showed that PRP-augmented repairs would have to show a reduced retear rate of at least 9.1% before the additional cost would be considered costeffective.
References 1. Chahal J, Van Thiel GS, Mall N, et al. The role of plateletrich plasma in arthroscopic rotator cuff repair: A systematic review with quantitative synthesis. Arthroscopy 2012;28:1718-1727. 2. Genuario JW, Donegan RP, Hamman D, et al. The costeffectiveness of single-row compared with double-row arthroscopic rotator cuff repair. J Bone Joint Surg Am 2012;94:1369-1377.
7
3. Jo CH, Kim JE, Yoon KS, et al. Does platelet-rich plasma accelerate recovery after rotator cuff repair? A prospective cohort study. Am J Sports Med 2011;39:2082-2090. 4. Yamaguchi K, Ditsios K, Middleton WD, Hildebolt CF, Galatz LM, Teefey SA. The demographic and morphological features of rotator cuff disease: A comparison of asymptomatic and symptomatic shoulders. J Bone Joint Surg Am 2006;88:1699-1704. 5. Vitale MA, Vitale MG, Zivin JG, Braman JP, Bigliani LU, Flatow EL. Rotator cuff repair: An analysis of utility scores and cost-effectiveness. J Shoulder Elbow Surg 2007;16:181-187. 6. Gumina S, Campagna V, Ferrazza G, et al. Use of plateletleukocyte membrane in arthroscopic repair of large rotator cuff tears: A prospective randomized study. J Bone Joint Surg Am 2012;94:1345-1352. 7. Miller BS, Downie BK, Kohen RB, et al. When do rotator cuff repairs fail? Serial ultrasound examination after arthroscopic repair of large and massive rotator cuff tears. Am J Sports Med 2011;39:2064-2070. 8. Randelli P, Arrigoni P, Ragone V, Aliprandi A, Cabitza P. Platelet rich plasma in arthroscopic rotator cuff repair: A prospective RCT study, 2-year follow-up. J Shoulder Elbow Surg 2011;20:518-528. 9. Dehaan AM, Axelrod TW, Kaye E, Silvestri L, Puskas B, Foster TE. Does double-row rotator cuff repair improve functional outcome of patients compared with single-row technique? A systematic review. Am J Sports Med 2012;40: 1176-1185. 10. Castricini R, Longo UG, Benedetto MD, et al. Platelet-rich plasma augmentation for arthroscopic rotator cuff repair: A randomized controlled trial. Am J Sports Med 2011;39: 258-265. 11. Rodeo SA, Delos D, Williams RJ, Adler RS, Pearle A, Warren RF. The effect of platelet-rich fibrin matrix on rotator cuff tendon healing: A prospective, randomized clinical study. Am J Sports Med 2012;40:1234-1241. 12. Weber SC, Kauffman JI, Parise C, Weber SJ, Katz SD. Platelet-rich fibrin matrix in the management of arthroscopic repair of the rotator cuff: A prospective, randomized, double-blinded study. Am J Sports Med 2012;41: 263-270. 13. Antuna S, Barco R, Martinez-Diez JM, SanchezMarquez JM. Platelet-rich fibrin in arthroscopic repair of massive rotator cuff tears: A prospective randomized pilot clinical trial. Acta Orthop Belg 2013;79:25-30. 14. Jo CH, Shin JS, Lee YG, et al. Platelet-rich plasma for arthroscopic repair of large to massive rotator cuff tears: A randomized, single-blind, parallel-group trial. Am J Sports Med 2013;41:2240-2248. 15. Jo CH, Shin JS, Shin WH, Lee SY, Yoon KS, Shin S. Platelet-rich plasma for arthroscopic repair of medium to large rotator cuff tears: A randomized controlled trial. Am J Sports Med 2015;43:2102-2110. 16. Malavolta EA, Gracitelli MEC, Neto AAF, Assuncao JH, Bordalo-Rodreigues M, de Camargo OP. Platelet-rich plasma in rotator cuff repair: A prospective randomized study. Am J Sports Med 2014;42:2446-2454. 17. Ruiz-Moneo P, Molano-Munoz J, Prieto E, Algorta J. Plasma rich in growth factors in arthroscopic rotator cuff repair: A randomized double-blind, controlled clinic trial. Arthroscopy 2013;29:2-9.
8
E. M. SAMUELSON ET AL.
18. Hsu WK, Mishra A, Rodeo S, et al. Platelet-rich plasma in orthopaedic applications: Evidence-based recommendations for treatment. J Am Acad Orthop Surg 2013;21: 739-748. 19. Anitua E, Andia I, Sanchez M, et al. Autologous preparations rich in growth factors promote proliferation and induce VEGF and HGF production by human tendon cells in culture. J Orthop Res 2005;23:281-286. 20. Aspenberg P, Virchenko O. Platelet concentrate injection improves Achilles tendon repair in rats. Acta Orthop Scand 2004;75:93-99. 21. de Mos M, van der Windt A, Holger J, et al. Can platelet rich plasma enhance tendon repair: A cell culture study. Am J Sports Med 2008;36:1171-1178. 22. Hall MP, Band PA, Meislin RJ, Jazrawi LM, Cardone DA. Platelet-rich plasma: Current concepts and application in sports medicine. J Am Acad Orthop Surg 2009;17:602-608. 23. Jo CH, Kim JE, Yoon KS, Shin S. Platelet-rich plasma stimulates cell proliferation and enhances matrix gene expression and synthesis in tenocytes from human rotator cuff tendons with degenerative tears. Am J Sports Med 2012;40:1035-1045. 24. Kajikawa Y, Morihara T, Sakamoto H, et al. Platelet-rich plasma enhances the initial mobilization of circulation derived cells for tendon healing. J Cell Physiol 2008;215: 837-845. 25. Platelet rich plasma market: Asia Pacific to show robust growth between 2014 and 2020. Available from: http://www.transparencymarketresearch.com/pressrelease/ platelet-rich-plasma-market.htm. Accessed June 25, 2015.
26. Lee DH. Platelet-rich plasma: Is it ready for prime time? Commentary and perspective. J Bone Joint Surg Am 2012;94:e114. 27. Kluger R, Bock P, Mittlbock M, Krampla W, Engel A. Long-term survivorship of rotator cuff repairs using ultrasound and magnetic resonance imaging analysis. Am J Sports Med 2011;39:2071-2081. 28. Mather RC III, Koenig L, Acevedo D, et al. The societal and economic value of rotator cuff repair. J Bone Joint Surg Am 2013;95:1993-2000. 29. Platelet rich plasma (PRP) Therapy reimbursement overview. Available from: http://www.rsmedical.com/ documents/331-0034-00-31reva.pdf. Accessed September 23, 2014. 30. Mather RC III, Hettrich CM, Dunn WR, et al. Cost-effectiveness analysis of early reconstruction versus rehabilitation and delayed reconstruction for anterior cruciate ligament tears. Am J Sports Med 2014;42:1583-1591. 31. Li X, Xu CP, Hou YL, Song JQ, Cui Z, Yu B. Are platelet concentrates an ideal biomaterial for arthroscopic rotator cuff repair? A meta-analysis of randomized controlled trials. Arthroscopy 2014;30:1483-1490. 32. Warth RJ, Dornan GJ, James EW, Horan MP, Millett PJ. Clinical and structural outcomes after arthroscopic repair of full-thickness rotator cuff tears with and without platelet-rich product supplementation: A meta-analysis and meta-regression. Arthroscopy 2015;31:306-320. 33. Zhao JG, Zhao L, Jiang YX, Wang ZL, Wang J, Zhang P. Platelet-rich plasma in arthroscopic rotator cuff repair: A meta-analysis of randomized controlled trials. Arthroscopy 2015;31:125-135.