There remains to be substantial debate on the best treatment of the infected shoulder arthroplasty. Infection after shoulder arthroplasty is an uncommon but devastating complication with a reported incidence from 0 to 4%. The most common organism responsible for infection following rotator cuff surgery, instability surgery, ORIF proximal humerus fractures, and shoulder arthroplasty is Prop. Acnes. A thorough history is important because many patients have a history of difficulty with wound healing or drainage. Prop. Acnes typically does not start to grow until day 5, therefore it is critical to keep cultures a minimum of 10 to 14 days.

Diagnosis can be difficult, particularly among patients undergoing revision surgery. The majority of patients with a low grade infection do not have overt signs of infection such as erythema or sinus tracts. Preoperative lab values as well as intraoperative pathology have been shown to be unreliable in predicting who will have positive cultures at the time of revision surgery.

There are a number of options for treating a patient with a post-operative infection. Critical variables include the timing of infection, status of the host, the specific organism, status of implant fixation, and the status of the rotator cuff and deltoid. One of the most frequently employed options for treating the infected shoulder arthroplasty is two-stage re-implantation. However, the rate of complications with this technique as well as residual infection remains high.

There is a large and growing population of patients with shoulder arthritis that are over 70 years old. Many of these patients live alone and sling immobilization after shoulder arthroplasty is problematic. Other than improved internal rotation, there are limited benefits of anatomic shoulder arthroplasty compared to reverse arthroplasty.

Anatomic arthroplasty is associated with longer OR time, longer recovery with need for assistance to allow the subscapularis to heal, and more challenging glenoid exposure. The reverse arthroplasty is a faster operation without the need for subscapularis healing and the sphere provides a more forgiving implant position. Additional benefits of reverse arthroplasty include better ability to manage glenoid bone loss and joint subluxation.

Data from the Australian Orthopaedic Association National Joint Replacement Registry shows that within the first year of surgery the rate of revision of anatomic shoulder arthroplasty is less than reverse arthroplasty. However, after one year, the overall revision rate of reverse arthroplasty is less than anatomic shoulder arthroplasty.

Therefore, increased technical difficulty of anatomic shoulder arthroplasty together with concerns of subscapularis insufficiency, glenoid loosening, and lack of strong evidence of superiority do not warrant changing from reverse for patients over 70 years old.

Lateralization of the reverse arthroplasty may be desirable to more effectively tension the remaining rotator cuff, decrease scapular notching, improve the cosmetic appearance of the shoulder, and improve stability as well as the arc of motion prior to impingement. There are two primary options to lateralise a reverse shoulder arthroplasty: bone graft with a long post (BIO-RSA) vs. using metal. The two metal options generally include a thicker glenosphere or a thicker glenoid baseplate.

Potential benefits of a BIO-RSA include lateralization of the glenoid center of rotation but without placing the center of rotation lateral to the prosthetic-bone interface. By maintaining the position of the center of rotation, the shear forces at the prosthesis-bone interface are lessened and are converted to compressive forces which will minimise glenoid failure.

Edwards et al. performed a prospective study on a bony increased offset reverse arthroplasty. Among the 18 shoulders in the BIO-RSA group, the incidence of notching was 78% compared to controls 70%. The graft completely incorporated in 12 (67%), partially incorporated in 4 (22%), and failed to incorporate in 2 (11%).

Frankle et al. reported on the minimum 5-year follow-up of reverse arthroplasty with a central compression screw and a lateralised glenoid component. The survivorship was 94% at 5 years. There were seven (9%) cases of scapular notching and no patient had glenoid baseplate loosening or baseplate failure. The authors noted that the patients maintained their improved function and radiographic results at a minimum of five years.

In summary, lateralisation of the glenosphere is an attractive option to improve the outcome of reverse arthroplasty. Benefits of lateralisation with metal rather than bone graft include elimination of concern over bone graft healing or resorption. In addition, the procedure has the potential to be more precise with the exact offset amount known pre-operatively as well as improved efficiency of the procedure. Preparing the graft takes additional OR time and there is variable quality of the bone graft.

There continues to be significant debate on the optimum treatment of the infected shoulder arthroplasty. Infection after shoulder arthroplasty is an infrequent but devastating complication with a reported incidence from 0 to 4%. The most common organism responsible for infection following rotator cuff surgery, instability surgery, ORIF proximal humerus fractures, and shoulder arthroplasty is P. acnes. A thorough history is important because many patients have a history of difficulty with wound healing or drainage. P. acnes typically does not start to grow until day 5, therefore it is critical to keep cultures a minimum of 10 to 14 days.

Diagnosis can be challenging, particularly among patients undergoing revision surgery. The majority of patients with a low grade infection do not have overt signs of infection such as erythema or sinus tracts. Pre-operative lab values as well as intra-operative pathology have been shown to be unreliable in predicting who will have positive cultures at the time of revision surgery.

There are a number of options for treating a patient with a post-operative infection. Essential variables include the timing of infection, status of the host, the specific organism, status of implant fixation, and the status of the rotator cuff and deltoid. One of the most frequently employed options for treating the infected shoulder arthroplasty is two stage re-implantation. However, the rate of complications with this technique as well as residual infection remains high.

The standard approach is through the deltopectoral interval. Among patients with prior incisions, one makes every effort to either utilise the old incision or to incorporate it into a longer incision that will allow one to approach the deltopectoral interval and retract the deltoid laterally. The deltopectoral interval is most easily developed just distal to the clavicle, where there is a natural infraclavicular triangle of fat that separates the deltoid and pectoralis major muscles even in very scarred or stiff shoulders. Typically, the deltoid is retracted laterally leaving the cephalic vein on the medial aspect of the exposure. The anterior border of the deltoid is mobilised from the clavicle to its insertion on the humerus. The anterior portion of the deltoid insertion together with the more distal periosteum of the humerus may be elevated slightly.

The next step is to identify the plane between the conjoined tendon group and the subscapularis muscle. Dissection in this area must be done very carefully due to the close proximity of the neurovascular group, the axillary nerve, and the musculocutaneous nerve. Scar is then released from around the base of the coracoid. The subacromial space is freed of scar and the shoulder is examined for range of motion. Particularly among patients with prior rotator cuff surgery, there may be severe scarring in the subacromial space. Internal rotation of the arm with dissection between the remaining rotator cuff and deltoid is critical to develop this plane.

If external rotation is less than 30 degrees, one can consider incising the subscapularis off bone rather than through its tendinous substance. For every 1 cm that the subscapularis is advanced medially, one gains approximately 20 to 30 degrees of external rotation. The rotator interval between the subscapularis and supraspinatus is then incised. This release is then continued inferiorly to incise the inferior shoulder capsule from the neck of the humerus. This is performed by proceeding from anterior to posterior with progressive external rotation of the humerus staying directly on the bone with electrocautery and great care to protect the axillary nerve.

The key for glenoid exposure as well as improvement in motion is deltoid mobilization, a large inferior capsular release, aggressive humeral head cut and osteophyte removal.

There are a variety of potential causes of shoulder arthritis in young patients including osteoarthritis, inflammatory arthritis, post-traumatic arthritis, and avascular necrosis. However, the primary etiology in my practice is related to complications of instability surgery or labral repair: thermal or anchor/suture related chondrolysis. The outcomes of arthroscopic debridement have been disappointing in patients with shoulder arthritis with worse results with increasing severity of articular cartilage changes.

Among all joint arthroplasty procedures, patients who undergo shoulder arthroplasty have the youngest average age. Results of hemiarthroplasty (HA) have been approximately 75% to 80% compared to 90% with total shoulder arthroplasty (TSA).

The largest series in the literature on shoulder arthroplasty in young patients is Schoch et al. They reviewed the results of 56 hemiarthroplasties and 19 TSA performed in patients less than 50 years old with a minimum 20-year follow-up or follow-up until reoperation. Both HA and TSA resulted in significant improvements in pain scores (p<0.001), abduction (p<0.01), and external rotation (p=0.02). Eighty-one percent of shoulders were rated much better or better than pre-operatively. Unsatisfactory ratings in HA were due to reoperations in 25 (glenoid arthrosis in 16) and limited motion, pain, or dissatisfaction in 11. Unsatisfactory ratings in TSA were due to reoperations in 6 (component loosening in 4) and limited motion in 5. Estimated 20-year survival was 75.6% (confidence interval, 65.9–86.5) for HAs and 83.2% (confidence interval, 70.5–97.8) for TSAs.

Infection after shoulder arthroplasty is an uncommon but devastating complication with a reported incidence from 0% to 4%. Prop. Acnes is the most common organism responsible for infection following rotator cuff surgery, instability surgery, ORIF proximal humerus fractures, and shoulder arthroplasty. A detailed history is critical because many patients have a history of difficulty with wound healing or drainage. Prop. Acnes typically does not start to grow until Day 5, therefore it is critical to keep cultures a minimum of 10 to 14 days.

Diagnosis can be difficult, particularly among patients undergoing revision surgery. The majority of patients with a low grade infection do not have overt signs of infection such as erythema or sinus tracts. Pre-operative lab values as well as intra-operative pathology have been shown to be unreliable in predicting who will have positive cultures at the time of revision surgery.

There are a number of options for treating a patient with a post-operative infection. Important variables include the timing of infection, status of the host, the specific organism, status of implant fixation, and the status of the rotator cuff and deltoid. One of the most frequently employed options for treating the infected shoulder arthroplasty is two-stage re-implantation. However, the rate of complications with this technique as well as residual infection remains high.

The humeral component has a long track record of a low rate of humeral loosening. Moreover, there are significant challenges associated with removal of a failed cemented component.

Throckmorton reviewed the results of 76 total shoulder arthroplasties for osteoarthritis with minimum two year follow-up. There were incomplete lucent lines in 5/76 stems. None of the stem were judged to be at risk for loosening. Matsen published on the outcome of 131 shoulder arthroplasties for osteoarthritis with minimum two year follow-up. In this series, there were no components with shift or tilt.

In addition to strong literature support for the use of an uncemented humeral component, revision of a cemented humeral component can be very difficult with a risk of significant destruction of the humerus.

The cortex of the humerus tends to be thin and removing the cement can be similar to trying to remove concrete from an ice cream cone.

Therefore, the extremely low rate of loosening and the challenges associated with cemented components makes the non-cemented component the ideal humeral solution.

Infection after shoulder surgery is an infrequent but devastating complication with a reported incidence from 0 to 4%. A careful history is critical because many patients have a history of a “stitch abscess” or “superficial wound infection”. Prop. Acnes is the most common organism responsible for infection following rotator cuff surgery, instability surgery, ORIF proximal humerus fractures, and shoulder arthroplasty. This organism typically does not start to grow until Day 5, therefore it is critical to keep cultures a minimum of 10 to 14 days.

The diagnosis can be challenging, principally among patients undergoing revision surgery. The majority of patients with a low grade infection do not have blatant signs of infection such as erythema or sinus tracts. Pre-operative lab values as well as intra-operative pathology have been shown to be unreliable in predicting who will have positive cultures at the time of revision surgery.

There is an assortment of options for treating a patient with a post-operative infection. Important variables include the timing of infection, status of the host, the specific organism, status of implant fixation, and the status of the rotator cuff and deltoid. One of the most frequently employed options for treating the infected shoulder arthroplasty is two stage re-implantation. However, the rate of complications with this technique as well as residual infection remains high.

A primary goal of shoulder arthroplasty is to place the components in anatomic version. However, traditional instrumentation does not accommodate glenoid wear patterns. Therefore, many investigators have attempted to use computer modeling or CT-based algorithms to create custom targeting guides to achieve this goal.

There are some recent studies investigating the use of custom guides. Iannotti et al. published in JBJS-American in 2012 on the use of patient specific instrumentation. There were 31 patients included in the study. The authors found that the planning software and patient specific instrumentation were helpful overall, but particularly of benefit in patients with retroversion in excess of 16 degrees. In this group of patients, the mean deviation was 10 degrees in the standard surgical group and 1.2 degrees in the patient specific instrumentation group.

Throckmorton presented a study at the AAOS in 2014 on 70 cadaveric shoulders. There was one high volume surgeon (>100 shoulder arthroplasties a year), two middle volume surgeons (20–50 shoulder arthroplasties a year), and two low volume surgeons (less than 20 shoulder arthroplasties per year). Overall, the custom guide was significantly more accurate than standard instrumentation. The custom guides were found to be especially more accurate among specimens with associated glenoid wear. There were no strong trends to indicate consistent differences between high, medium, and low volume surgeons. The authors concluded that custom guides have narrower standard deviation and fewer significant errors than standard instrumentation.

Custom guides continue to evolve for use in shoulder arthroplasty including some guides that allow the surgeon to decide intra-operatively between anatomic shoulder arthroplasty and reverse arthroplasty. Additional studies will be necessary to further define the role of patient specific instrumentation in practice.

Removal of a well-fixed humeral component during revision shoulder arthroplasty presents a challenging problem. If the humeral component cannot be extracted simply from above, an alternate approach must be taken that may include compromising bone architecture to remove the implant. Two potential solutions to this problem that allow removal of the well-fixed prosthesis are making a humeral window or creating a longitudinal split in the humerus.

A retrospective review was performed at the Mayo Clinic to determine the complications associated with performing humeral windows and longitudinal splits during the course of revision shoulder arthroplasty. This study included 427 patients from 1994–2010 at Mayo Clinic undergoing revision shoulder arthroplasty. From this cohort, those who required a humeral window or a longitudinal split to assist removal of a well-fixed humeral component were identified. Twenty-seven patients had a humeral window produced to remove a well-fixed humeral component. Six intra-operative fractures were reported from this group: 5 were in the greater tuberosity and 1 was in the distal humeral shaft. At the latest radiographic follow-up, 24 of 27 windows healed, 2 patients had limited inconclusive radiographic follow-up (1 and 2 months), and 1 did not have follow-up at our institution. Twenty-four patients underwent longitudinal osteotomy to extract a well-fixed humeral component. From this group, 1 had intra-operative fracture in the greater tuberosity. At most recent radiographic follow-up, 22 of 24 longitudinal splits healed, 1 had short follow-up (1 ½ months) with demonstrated signs of healing, and 1 did not have follow-up at our institution.

In both groups, there were no cases of window malunion and no components have developed clinical loosening. Data from this study suggests humeral windows and longitudinal splits can assist with controlled removal of well-fixed humeral components with a high rate of union and a low rate of intra-operative and post-operative sequelae.

As reverse total shoulder arthroplasty (RTSA) systems expand with longer durations in vivo, so does the concern and potential complications of wear, debris and osteolysis. Despite some other profound attempts, no wear testing method has stood out to compare implants across systems and labs. The main reasons may have been the diverse sources of forces and motions used in testing, widely different wear amounts which resulted and the general lack of dedicated shoulder simulators. To add a dedicated shoulder simulator to hip and knee simulators would burden the resources of any testing lab. In this study we propose a shoulder wear test method which addresses the above.

Harnessing the wealth of force-motion data from telemetrized shoulder implants from the Bergman's group in Berlin, we synthesized their results to devise a wholistic multi-axes simulation regime for reverse shoulders. The alignment and motions of the humeral cup and the glenosphere were kept anatomically correct (relative to each other) and yielded a physiologically realistic wear-inducing articulation. However, we opted for a very unusual installation/orientation of the whole implant system to allow a twelve station AMTI (hip) simulator to be adapted for this study. The shoulder constructs were aligned with novel fixtures such that the machine's vertical compressive force mimicked the average forces of the shoulder found from the in vivo telemetry data in magnitude and nominal direction. Aligned thus, a patient with a shoulder installed would neither stand, nor lie down, but be oriented in a composite angle relative the simulator original axes. Each anatomic shoulder motion would be achieved by unique computed combinations of the three simulator motion actuators, none of which would be aligned anatomically for the shoulder on its own.

The maximum ranges of cyclic shoulder motion achieved with the constraints of the simulator were 38°–79° of forward elevation repeated in two separate (15°and 45°) elevation planes. The change of elevation plane inherently involved abduction-adduction motion, and simultaneously also involved variation of internal-external rotation within a 57° range. Each elevation rise (twice per cycle) was also accompanied by a sinusoidally rising and falling compressive load in the range 50N–1700N.

The test method was tested (!) by simulating for 2.5 million of the above (double-elevation) cycles and gravimetrically measuring wear of twelve 36 mm size RTSA systems. We compared six systems having vitamin E-infused highly cross-linked polyethylene bearings (100 kGy radiation) to six controls with a medium cross-linked polyethylene of half the radiation dose. Significant wear resulted for the control bearing material (average 17.9 ± 0.851 mg/MC) which was no less than many hips and knees. Multiply (and statistically significantly, p < 0.001) less average wear (3.42 ± 0.22 mg/MC) resulted for the highly cross linked bearings.

The above demonstrated the effectiveness of the test method. Significant wear resulted under physiologically realistic cyclic motion and forces with strong discrimination between two systems whose bearing materials were known to be different in resilience to wear. Using novel fixtures and unusual orientation to utilize a standard commercially available joint simulator promises efficacy of the test method and utility across different labs.

Purpose: The indication to perform a total shoulder arthroplasty (TSA) versus a hemiarthroplasty is guided by a patient’s intended level of activity after surgery. It is unclear what activities patients actually perform following shoulder replacement, therefore, the purpose of this study was to compare the self-reported activities of patients following either a TSA or hemiarthroplasty.

Method: Two groups of 75 patients each, following TSA or hemiarthroplasty, were matched for a variety of demographic variables. A mailed activity questionnaire asked patients to report their level of pain, motion, strength, and a choice of 70 different activities. Reported activities were classified as high (i.e. tennis) or low (i.e. fishing) demand, and categorized as household (i.e. cooking), yard work (i.e. gardening), sporting (i.e. golf), or musical (i.e. piano).

Results: Ninety-six (64%) patients completed the survey, 50 in the TSA group (27F:19M, avg. 53.2 yrs), and 46 in the HA group (29F:21M, avg. 53.5 yrs). Pain was not different between groups (3.6/10 TSA: 3.9/10 HA), yet a significant difference was reported in forward flexion (145° TSA: 120° HA, P< .002) and strength (6.3/10 TSA: 5.3/10 HA, P< .01). Across all categories whether it be high or low demand, the TSA group (10.4 activities/person) reported more activities compared with the hemiarthroplasty group (8.6 activities/person).

Conclusion: The conventional understanding that a hemiarthroplasty provides the possibility for more activity following surgery is not supported by our data. Patients following a TSA reported better motion and strength and were more active than the hemiarthroplasty group.

Purpose: We conducted a retrospective review to evaluate outcomes, complications and implant survival after shoulder reconstruction for bone neoplasm using anatomic shoulder prostheses.

Method: Thirty-four anatomic shoulder prostheses were reviewed at an average follow-up of 51 (range, 6 to 143) months. The mean age at the time of surgery was 47 (range, 15 to 74) years. Twenty-five patients (74%) had reconstruction with an allograft-prosthetic composite (APC). The average amount of humerus resected in patients with an APC was 13 cm (range, 5 to 36 cm).

Results: At most recent follow-up, 82% of patients had no pain (11), slight pain (12) or moderate pain with strenuous activities (5). Twenty-eight patients (82%) were subjectively satisfied despite only 29% being satisfactory according to Neer rating. Sixteen complications occurred in 13 patients, including instability (5), host/graft non-union (3), aseptic loosening (3), arthrofibrosis (1), tumor recurrence (2) and superficial infection (1). Host/graft nonunion occurred in 3/9 patients with an APC using press-fit or plate fixation for distal humeral fixation and 0/16 shoulders with cemented distal humeral fixation (p=0.02). Three implants required revision for aseptic loosening (1), host/graft nonunion (1) and instability (1). Kaplan-Meier survivorship at 10 years was 88% for implant revision and 80% for mechanical failure.

Conclusion: Reconstruction of the proximal humerus using an anatomic prosthesis after resection of bone neoplasms is associated with a low rate of mechanical failure but a moderate rate of shoulder instability. Cement fixation into the native distal humerus should be considered when reconstruction using an APC is required.

Purpose: Loosening of the humeral component is rarely a cause for revision shoulder surgery. Most long-term series are not large enough to stratify the many risk factors that might influence the survivorship of humeral component designs. The purpose of this study was to determine long-term survivorship of the Neer and Cofield humeral components and to define the risk factors associated with humeral component removal or revision.

Method: 1584 primary Neer and Cofield shoulder arthroplasties (1423 patients) were performed at our institution from 1984 to 2004. There were 619 men (694 shoulders) and 804 women (890 shoulders), with a mean age at arthroplasty of 65.6 years (range, 16–94 years). Kaplan-Meier survivorship analysis was used to determine the effect of etiology of the disease, gender, age, surgery type (hemi versus total), fixation type (cemented versus noncemented), and the humeral component type (Neer II, Cofield I or II) on the estimated survival free of humeral component revision or removal.

Results: There were 108 revisions and 17 removals of the humeral component. The overall rate of removal or revision of the humeral component was 7.9% with an average followup of 8.1 years. The rates of survivorship free of revision or removal of the humeral component for any reason was 94.8% at 5 years, 92% at 10 years, 86.7% at 15 years and 82.8% at 20 years. Seventy-one of 632 shoulders (11.2%) in patients younger than 65 years required humeral component revision or removal, whereas only 54 of 952 shoulders (5.7%) in patients 65 years and older required humeral component revision or removal (Odds ratio=2.1; 95% confidence interval, 1.5–3, p=0.001). Patients with posttraumatic arthritis had a higher risk of needing revision or removal of the humeral component (Odds ratio=2.1, 95% confidence interval 1.3–3.3) compared to osteoarthritis. Eighty-four of 526 shoulders (16.0%) with metal-back glenoid components required humeral component revision or removal, whereas only 41 of 1058 shoulders (3.9%) with non metal-backed glenoid components required humeral component revision or removal (Odds ratio=4.7; 95% confidence interval, 3.2–7, p=0.001).

Conclusion: Younger age, replacement due to post-traumatic arthritis and presence of a metal-backed glenoid increased the likelihood of humeral component failure. Similar short-term survival can be achieved with Cofield II and Neer II humeral components.

Purpose: Currently, there is little information available concerning the outcome of patients with infection following rotator cuff repair. Therefore, the purpose of this study was to review the incidence, management and outcomes of patients with rotator cuff repair complicated with deep infection.

Methods: Between 1975 and 2003, 39 cases of deep infection following rotator cuff repair were identified. The medical records and radiographs were retrospectively examined. At a mean follow-up of 8.2 years, seven patients had died and two patients had been lost to follow-up leaving 30 cases for outcome evaluation.

Results: The incidence of deep infection after rotator cuff repairs that were performed at our institution was 0.43% (21 of 4886 cases). The mean interval from rotator cuff repair to the time of infection diagnosis was 49 days. ESR and C-reactive protein were elevated in only 60% and 50% of patients, respectively. Propionibacterium was the most common organism isolated, infecting 51% of cases. A mean of 3.3 surgical debridements were necessary for the eradication of infection. At final follow-up, mean active abduction was 121° and mean external rotation was 44°. The ASES score averaged 67 points and the Simple Shoulder Test score averaged 7.3 points.

Conclusions: The data from this study suggests that the eradication of deep infection following rotator cuff repair is possible, however, substantial functional limitations are not unusual. Additionally, the treating surgeon should be aware of the high incidence of Propionibacterium and the importance of allowing a minimum of seven days of culture to identify this organism.

Between January 1, 1976 and December 31, 1991, 195 total shoulder arthroplasties and 108 hemiarthroplasties were performed by the senior author in patients with rheumatoid arthritis. One hundred eighty-seven total shoulder arthroplasties and 95 hemiarthroplasties with complete preoperative evaluation, operative records, and minimum 2-year follow-up (mean 11.6 years) or follow-up until revision were included in the clinical analysis. Twenty patients died and one was lost to follow-up. All 303 shoulders were included in the survival analysis.

There was significant long term pain relief (P< .0001), improvement in active abduction (P< .0001), and external rotation (P< .0001) with both, hemiarthroplasty and total shoulder arthroplasty. There was not a significant difference in improvement in pain and motion comparing hemiarthroplasty and total shoulder arthroplasty for patients with a thin or torn rotator cuff. However, among patients with an intact rotator cuff, improvement in pain and abduction were significantly greater with total shoulder arthroplasty. Additionally, among patients with an intact rotator cuff, the risk for revision was significantly lower for total shoulder arthroplasty (p=0.04).

Radiographs were available for 152 total shoulder arthroplasties and 63 hemiarthroplasties with a minimum 2 year follow-up. Glenoid erosion was present in 62 of 63 hemiarthroplasties (98%). Glenoid periprosthetic lucency was present in 110 of 152 total shoulder arthroplasties (72%).

The data from this study indicate there is marked long term pain relief and improvement in motion with shoulder arthroplasty. Among patients with an intact rotator cuff, total shoulder arthroplasty appears to be the preferred procedure for pain relief, improvement in abduction, and lower risk of revision surgery.