Abstract
Objective:
While there has been increasing interest in minimally invasive surgery, the expenses incurred by patients undergoing this form of surgery have not been comprehensively studied. The authors compared the costs borne by patients undergoing arthroscopic rotator cuff repair with the standard mini-open repair at a tertiary hospital in an Asian population.
Patients and Methods:
This was a retrospective cohort study. The authors studied the inpatient hospital bills of patients following rotator cuff tear repair between January 2010 and October 2014 via the hospital electronic medical records system. 148 patients had arthroscopic repair and 78 had mini-open repair. The cost of implants, consumables, and the total cost of hospitalization were analyzed. Operative times and length of stay for both procedures were also studied. Constant scores and American Shoulder Elbow Scores (ASES) were recorded preoperatively and at 1 year postoperatively.
Results:
Three fellowship-trained surgeons performed arthroscopic repairs and one performed the mini-open repair. The cost of implants and consumables was significantly higher with arthroscopic repair. The duration of surgery was also significantly longer with that technique. There was no difference in length of stay between the two techniques. There was also no difference in Constant scores or ASES scores, both preoperatively and at 1 year postoperatively.
Conclusions:
The immediate costs of mini-open repair of rotator cuff tears are significantly less than that of arthroscopic repair. Most of the difference arises from the cost of implants and consumables. Equivalent functional outcomes from both techniques suggest that mini-open repair may be more cost-effective.
Introduction
Rising health care costs are a worldwide phenomenon, and keeping health care provision costs down is thus becoming increasingly important. Rotator cuff pathology accounts for a significant proportion of presentations to the orthopedic clinics, and with aging of the population becoming a global phenomenon, this number is expected to rise even further over the coming decade. Over 75,000 cuff repairs are performed annually in the United States alone, 1 with this figure likely to rise in tandem with the increase in prevalence of rotator cuff disease. Surgical repair of the rotator cuff has been shown to improve pain, restore shoulder strength, and restore preoperative range of motion 2,3 ; it has also been shown that rotator cuff repair is cost-effective, providing improved quality of life while producing net cost savings for society. 4 Technology has enabled a previously open procedure to be carried out just as well arthroscopically. 5,6 Meanwhile, the concern over deltoid failure has led to conversion of the traditional open procedure to a mini-open, deltoid-splitting approach instead. Several previous studies have shown similar outcomes between the mini-open and arthroscopic cuff repair. 5 –9 A recent meta-analysis studied 5 randomized controlled trials with a total sample size of 329 patients and found no significant difference in visual analogue scale (VAS) scores, surgery times, functional outcome scores, and range of motion between the arthroscopic and the mini-open techniques. 5 Costs associated with cuff repair include the cost of surgery and subsequent expenses incurred for follow-up clinic visits, analgesics, and physiotherapy sessions. In addition, there are the economic and social costs of time away from work. There have been few studies comparing the costs of mini-open versus arthroscopic cuff repairs. 10,11 These studies have also primarily been in Western populations.
In this study, the authors aim to compare the immediate costs incurred by patients undergoing either a mini-open repair or an arthroscopic rotator cuff repair at a tertiary hospital in an Asian population. Our null hypothesis is that there is no significant difference in the immediate costs of mini-open versus arthroscopic rotator cuff repair.
Materials and methods
The authros performed a retrospective review of hospitalization bills obtained from the finance department of our hospital, presented to patients on discharge after rotator cuff repair surgery from January 2010 to October 2014. The bill was divided into the expenses incurred for implants, consumables, and the professional fees to the surgeon and anesthetist. The duration of surgery and the length of stay were also documented. Pre- and postoperative Constant and the American Shoulder and Elbow Surgeons (ASES) scores for the two repair techniques were also studied.
The study was approved by the National Disease Specific Review Board.
Inclusion and exclusion criteria
Patients aged between 20 years and 80 years with small to massive rotator cuff tear repairs were included. Patients for whom concomitant subacromial decompression, biceps tenodesis or tenotomy or Mumford procedures were performed were included and studied separately. Revision cuff repairs and massive irreparable cuff tears were excluded. Three fellowship-trained consultant surgeons performed the arthroscopic procedure while another performed the mini-open repair.
Surgical techniques
Mini-open repair
Mini-open repair was performed under general anesthesia, using a beach-chair position. Disposable shoulder drapes were used. An anterolateral deltoid-splitting approach was used. The cuff was mobilized and the bony footprint prepared using a burr. The tear was repaired using transosseous nonabsorbable sutures. Suture anchors were not used for mini-open repairs. Subacromial decompression was not performed in most patients.
Arthroscopic repair
Arthroscopic repair was performed under general anesthesia, with the patient in a beach-chair position. Traction was applied to the arm. Disposable shoulder drapes were used. A three-portal technique was used for all patients. The tear morphology, glenohumeral joint, and subacromial space were examined. The cuff was mobilized and the bony footprint was prepared using a motorized burr. The tear was repaired using knotless suture anchors. Single or double row repairs were carried out depending on the surgeons’ preferences. Subacromial decompression was performed in most patients.
Most patients were discharged the same day or after an overnight stay with analgesics. The affected shoulder was rested in a sling postoperatively. Active motion exercises were allowed at 6 weeks postoperatively and only gentle passive motion allowed before that. Rotator cuff strengthening exercises were commenced subsequently.
Data analysis
The Mann–Whitney U test was used for statistical analysis. The 95% confidence internal interval was calculated with statistical significance set at p < 0.05.
Results
226 patients underwent rotator cuff repair between January 2010 and October 2014 (Table 1). 136 were arthroscopic and 90 were mini-open repairs; in 12 of the latter group an on-table decision was made to convert from arthroscopic to mini-open—this was performed following unsuccessful attempts at repairing the rotator cuff arthroscopically, for instance, due to difficulty in mobilizing the retracted rotator cuff or screw pull-out in osteoporotic bone. In nine of these cases, bone anchors were used at the discretion of the operating surgeon. They were included in the arthroscopic group on the basis of intention-to-treat analysis.
Demographic data of the study population.
The mean age was similar in both groups. There was an overall male preponderance. There was a higher proportion of private patients in the mini-open repair group. The majority of cuff tears were medium and small. Most of the additional procedures such as biceps tenotomy were carried out in the arthroscopic repair group.
Under the Singapore health care system, patients can undergo surgery either as subsidized or private patients. With private patients, there is a variable surcharge for the surgery performed, which can influence the overall cost of surgery. Subsidized patients receive heavy subsidies from the government on their hospitalization bills. To fully reflect difference, cost analyses were performed with patients divided into these two subgroups. The surgical details and cost analyses are provided in Table 2. Costs are presented as median values of the patient subgroups and represent the total billable amount before any government subsidy was applied.
The comparison of costs related to surgery in both arthroscopic and mini-open rotator cuff repairs.a
aA p value of <0.05 is considered significant. All costs displayed as median values in Singapore dollars (SG$)
In terms of absolute cost, arthroscopic rotator cuff repairs (SG$7543.18 and SG$11542.33) in our center were significantly more expensive compared to mini-open repairs (SG$5027.06 and SG$8688.78) in both the subsidized and private patient subgroups, respectively; this difference in costs is related primarily to the significantly higher costs of implants and consumables in the arthroscopic surgery group. These costs contribute to a significantly higher total inpatient bill. The null hypothesis is thus rejected (p < 0.0001).
Consumables in the above table include both surgical and anesthetic consumables as it was not possible to delineate the two costs within our hospital records system. However, the anesthetic equipment used in both forms of surgery are similar, and the cost differences thus largely represent the surgical consumables. The total surgery cost includes the costs of the implants, consumables, and the professional fees for both the surgeon and the anesthetist.
In addition, the median operative time for arthroscopic rotator cuff repairs was significantly higher than mini-open repairs (82 vs. 53.5 min, p < 0.0001).
Acromioplasties were performed in 72.3% and 12.9% of arthroscopic and mini-open repairs, respectively (Table 1). Median operative times for all rotator cuff repairs were significantly longer in the arthroscopic group than the mini-open group (82 vs. 53.5 min, p < 0.0001). To account for time spent performing acromioplasties and additional procedures, subgroup analyses were performed to compare operative times for patients in both groups. The significantly longer median operative times for arthroscopic repairs were again seen in both rotator cuff repair with acromioplasty only (78 vs. 54 min, p < 0.0001) and rotator cuff repair without acromioplasty (85 vs. 53 min; p < 0.0001). Similarly, in subgroups where rotator cuff repairs were performed exclusively (i.e. no acromioplasty, biceps tenodesis, tenotomy, or Mumford procedure performed), this significant difference remained (79 vs. 52 min, p < 0.0001).
There was no significant difference in the median lengths of hospital stay between arthroscopic and mini-open repair patients in both the private and subsidized subgroups (Table 2).
At 1 year postoperatively, both groups showed a significant improvement from the preoperative Constant and ASES scores (Table 3). There was no significant difference in preoperative Constant or ASES scores between the arthroscopic and mini-open repair groups. There was also no significant difference in postoperative Constant or ASES scores between the arthroscopic and mini-open repair groups.
Functional outcome scores (ASES scores) in both arthroscopic and mini-open groups preoperatively and 1-year postoperatively.a
ASES: American Shoulder Elbow Scores.
aA p value of <0.05 is considered significant.
Discussion
Previous studies have pointed out the similarly excellent outcomes between arthroscopic and mini-open rotator cuff repairs. 5 –9 Few studies have analyzed the costs incurred by patients undergoing these procedures. Adla et al. 10 showed that although the clinical results were equivalent between the two techniques, the mini-open cuff repair was more cost-effective. The study involved 30 patients in the United Kingdom. Churchill and Ghorai 11 studied 5224 patients in Milwaukee, Wisconsin, USA, and showed that the mini-open cuff repair required significantly less operative time and was also significantly less expensive.
This study was carried out in a tertiary hospital in Singapore. This study analyzed only the costs incurred for the episode of hospitalization during which surgery was performed. A previous study by Vitale et al. 1 has shown that most costs are incurred during this period. The authors avoided analyzing expenses incurred during subsequent outpatient visits, radiological investigations, and physiotherapy as there were likely to be too many variables. Indirect costs from loss of income from time off work were also not studied as the employment statuses of patients varied.
The findings reinforce those from previous studies that mini-open rotator cuff repairs cost significantly less than arthroscopic repairs. 10,11 To date, this study and that by Adla et al. 9 are the only detailed analyses of the costs incurred by patients during rotator cuff repair surgery.
Suture anchors were not used for the mini-open repairs in this study. Instead, nonabsorbable sutures passed through drill holes in the greater tuberosity were used. This resulted in significant cost savings (Table 2).
Consumables also add to the cost of rotator cuff surgery, especially when performed arthroscopically. 11 Besides the standard drapes, arthroscopic burrs and radiofrequency wands used to clear bursal tissue add to the cost. These contributed for one-third of the costs in the study by Adla et al. In this study, disposable burrs and wands contributed a significant portion of the bill in the arthroscopic procedure. A reusable burr was used for acromioplasty and to prepare the rotator cuff footprint in this mini-open approach. The cost of consumables was thus lower with the mini-open approach (Table 2).
Arthroscopic repairs took significantly longer to perform compared to mini-open repairs. In our center, this occurred despite the fact that all three surgeons were performing these arthroscopic repairs almost exclusively and would thus be experienced with the procedure. The shorter operative time with the mini-open technique, however, did not translate into cost savings in this study. This was because the cost of surgery was based on the table code (which was the same for both techniques in our center) and not the operative duration. Difference in operative times may lead to different costs in centers where patients are billed per unit time of operating theatre usage.
In this study, it was not possible to compare the operative times for the two techniques performed by the same surgeon, again due to the fact that each surgeon performed either open or arthroscopic repairs almost exclusively. The times for both open and arthroscopic repairs in our center were however comparable to those found in the current literature. Adla et al. 10 reported mean times of 67.7 and 81.7 min for open and arthroscopic repairs, respectively. Similarly, the median times for high-volume centers in New York were 77.0 and 103.5 min for open and arthroscopic repairs, respectively. 11 The authors thus believe the difference in operative times reflects a genuine difference between the two techniques rather than simply a comparison of surgeon skill.
The disparity also cannot be explained solely by the higher proportion of acromioplasties performed during arthroscopic surgery as subgroup analysis of patients in both surgical groups who did not undergo acromioplasties revealed similarly shorter operative times in the mini-open group. The authors did not perform a separate analysis of the time taken for other additional procedures such as the Mumford procedure, biceps tenotomy, and tenodesis. The numbers were small in both groups (Table 1), except for biceps tenotomy in the arthroscopic group. It is unlikely that biceps tenotomy would lead to a significant increase in operative time.
The authors attempted to test the theory that increased operative times for arthroscopic rotator cuff repair was due to the steeper learning curve involved when training orthopaedic residents. They thus compared the operative times of only the private patients who did not undergo acromioplasty and again found that the significant difference in operative times remained. The surgeries on private patients would have been performed almost exclusively by the consultant-in-charge and theoretically eliminate any influence of resident teaching on the total operative time.
There was no significant difference in the length of stay between the two techniques. Of note, Baker and Liu 7 reported the arthroscopically repaired group to be hospitalized 1.2 days fewer than the mini-open group. Vitale et al. 1 reported a mean of 1.5 days of hospitalization for the mini-open approach in contrast to the 1 day reported in this study. As there was a marked difference in the daily hospitalization rates depending on the type of ward the patients were admitted to, the absolute cost difference related to the length of stay was not calculated in this study.
In light of findings from this study and others, 10,11 and evidence suggesting equivalent outcomes between the two techniques, 5 –9 the role of the arthroscopic rotator cuff repair needs review. Should the technique be offered to all patients uniformly, especially in heavily subsidized health care organizations in view of its apparently lower cost-effectiveness? Second, should the mini-open repair technique be reemphasized in residency training programs where minimally invasive surgical techniques are gaining favor? A reappraisal may perhaps be required in countries where health care resources are more limited.
A debate surrounding the relative merits of transosseous suture repair and the more recently introduced suture anchors has been ongoing for the past two decades, with no clear superiority demonstrated for either technique. 12 –14 The usual practice in our institution is not to use suture anchors when performing mini-open repairs. The authors are however aware of the fact that other centers may still utilize suture anchors even in mini-open repairs. It is also not uncommon to perform a diagnostic arthroscopy followed by a mini-open repair in cases of diagnostic uncertainty. In these instances, they believe there will still be a cost benefit, both due to the use of less expensive equipment and the generally shorter operative time with mini-open repairs. Further work needs to be performed to elucidate the extent of these differences under the above settings.
This study has several limitations, including the fact that it is retrospective. Also, functional outcomes were only analyzed at 1 year postoperatively. Analysis of 2-year functional outcomes would have been preferable but the available data was inadequate. Of note, functional outcome scores for just 39 patients in the arthroscopic group and 12 patients in the mini-open group were available for review at 1 year postoperatively. A probable explanation for this is that patients frequently fail to keep follow-up visits if they are well and have no complaints following surgery. However, previous publications with superior patient follow-up data have shown similar functional outcomes between the two techniques. 5 –9 The findings of this study add to this existing body of evidence.
The cost of suture anchors, arthroscopic burrs, and radiofrequency wands was assumed to remain constant. With wider use, the cost of these devices would likely decrease in the future, decreasing the overall cost difference between the two techniques. With greater experience and wider use of the arthroscopic technique in residency training programs, the times for the surgical procedure are also likely to decrease to a level comparable to the mini-open procedure.
In summary, the immediate costs of mini-open rotator cuff repairs are significantly less than those of arthroscopic repair. Most of the difference arises from the cost of implants and consumables. Functional outcomes seem equivalent for both repair techniques at short- to medium-term follow-up. These findings suggest that mini-open repair may be more cost-effective than arthroscopic repair.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.