Interventional procedure overview of biodegradable subacromial spacer insertion for rotator cuff tears.
Closed for comments This consultation ended on at Request commenting lead permission
Evidence summary
Population and studies description
This interventional procedures overview is based on approximately 1500 patients from 2 RCTs, 2 systematic reviews, 1 case-control study, 1 retrospective comparative study and 3 case series. Of these patients, approximately 675 patients had the procedure. This is a rapid review of the literature, and a flow chart of the complete selection process is shown in figure 1. This overview presents 9 studies as the key evidence in table 2 and table 3, and lists other relevant studies in table 5.
Of those studies included which are not systematic reviews, 1 was from the United Kingdom, 1 from US and Canada, 1 from Italy, 1 from Greece, 1 from Israel, 1 from Turkey, 1 from Ireland and 1 from Slovenia. The mean follow-up ranged from 12 months to 5 years, and the mean age of participants ranged from 65.7 to 70.3 years. All 9 studies had inclusion and exclusion criteria, which had some small differences between studies. The majority stated that the rotator cuff tear had to be irreparable for a patient to be eligible, however this is a highly variable definition. But, 1 case series (Senekovic 2017) included patients with irreparable and reparable rotator cuff tears. Furthermore, 7 of the 9 studies stated that to be eligible the rotator cuff tear had to be defined as 'massive'. This is defined as a rotator cuff tear with retraction of the tendon to the glenoid rim or exposing two-thirds of the greater tuberosity.
Of those studies comparing spacer implantation with another group of patients, one RCT compared debridement with spacer implantation with debridement only as the control group (Metcalfe 2022). The other RCT compared InSpace implantation without repair with partial repair (Verma 2022). The case-control study and the comparative study both compared partial repair with spacer implantation with partial repair only (Malahias 2021 and Bisel 2022). Meanwhile, the systematic review by Osti et al. 2021 collated evidence from studies on patients who had been implanted with a spacer and compared outcomes to those patients who had undergone ASCR. Table 2 presents study details.
Study no. | First author, date Country | Patients (male: female) | Age | Study design | Inclusion criteria | Intervention | Follow up |
|---|---|---|---|---|---|---|---|
1 | Metcalfe et al, 2022. United Kingdom | 117 (67:50) | Mean 66.9 years | RCT | Irreparable rotator cuff tear, which had not resolved with conservative treatment and had symptoms warranting surgery. | Debridement with spacer (56 patients) versus debridement-only (61 patients). | 3, 6 and 12 months |
2 | Verma et al, 2022. United States and Canada | 184 patients (100:84) | Mean age 66.8yrs (Inspace group), 64.7yrs (partial repair group) | RCT | Patients ≥40 years of age with symptomatic, irreparable, posterosuperior, massive rotator cuff tears and an intact subscapularis who underwent failed non-operative management. Further details: 1. Male or female subject ≥40 years of age 2. Within 9 months before study enrolment, positive diagnostic imaging by MRI of the index shoulder indicating a full-thickness massive rotator cuff tear: a. Measuring ≥5 cm in diameter (Cofield classification) b. Involving ≥2 tendons 3. Functional deltoid muscle and preserved passive ROM on physical examination 4. Documented VAS pain score >30 mm 5. Underwent failed nonoperative treatment of at least 4 months' duration (time elapsed since the initial treatment) using ≥1 of the following: a. Oral analgesics b. Anti-inflammatory medication (e.g., ibuprofen, naproxen) c. Corticosteroid injection(s) d. Physical therapy e. Activity modification f. Rest (sling used) | InSpace implant insertion (with no rotator cuff repair) (93 patients) versus partial repair (suture anchor repair) (91 patients) as a primary surgical treatment for posterosuperior, massive rotator cuff tears. | 10 days, 6 weeks, 3, 6, 12 and 24 months |
3 | Osti et al, 2021 Italy | 998. Gender specified in 25 studies (756 patients, 438:318). | Mean 67.9 years | Systematic review | Studies reporting clinical and functional outcomes following the use of a subacromial spacer for massive irreparable rotator cuff tear, comparing them with ASCR. | Subacromial spacer implantation (375 patients) versus ASCR (623 patients) | Mean 27.6 months (range 4-110). For Spacer group: mean 27.0 months (range 4-60). |
4 | Johns, 2020 USA | 337 patients and 343 shoulders. Where gender specified: 158:143 | Mean 68 years | Systematic review | All studies assessing the use of implantable subacromial balloon spacers for management of massive, irreparable rotator cuff tears, reporting outcomes relating to biomechanics, clinical function, shoulder ROM, patient satisfaction, costs and complications. Published in the English language. | Insertion of implantable subacromial balloon spacer for massive, irreparable rotator cuff tear. | Mean 33 months |
5 | Malahias, 2021 Greece | 32 (13:19) | Mean group A: 65.7 years, group B 69.7 years | Retrospective case-control study | A diagnosis of symptomatic massive rotator cuff tear confirmed clinically, radiologically and intra-operatively in patients >50 years undergoing arthroscopic treatment either as combined spacer and partial repair or isolated partial repair with follow-up after 12-months post-op. | Arthroscopic partial repair with (16 patients) or without (16 patients) InSpace Balloon implanation. | 12 months |
6 | Maman and Kazum, 2022 Israel | 78 (37:41) | Mean age 70 years | Retrospective case series | Massive rotator cuff tear treated with InSpace device implantation, a minimum of 1 year follow-up, failure of at least 3 months of conservative treatment, | Balloon implantation performed arthroscopically. | Mean 56 months |
7 | Bilsel, 2022 Turkey | 32 (8:24) | Median age partial repair group: 68 years. Median age partial repair with spacer group: 68.5 years | Retrospective comparative study | Patients with a symptomatic and irreparable massive rotator cuff tear with tension retraction > stage 2, according to the Patte classification, without significant osteoarthritis and minimum 1-year follow-up | Patients who had undergone arthroscopic partial cuff repair only compared with patients who had additional implantation of a subacromial spacer | Partial repair group median follow-up: 28 months. Spacer group median follow-up: 17 months |
8 | Davey, 2021. Ireland | 45 (31: 14) | Mean age 70.3 years | Retrospective case series | Patients with a massive rotator cuff tear who underwent subacromial balloon spacer insertion alone with a minimum of 12 months follow-up. | Subacromial balloon spacer insertion | Mean 37.1 months |
9 | Senekovic, 2017. Slovenia | 24 (12:12) | Mean age 68.8 years | Prospective case series | People with persistent pain and functional disability for at least 6 months, imaging confirmation of a rotator cuff tear and failed conservative therapy. | Insertion of a biodegradable inflatable InSpace system in patients with massive reparable or irreparable rotator cuff tear. | 5 years |
Efficacy outcomes | Safety outcomes | |
|---|---|---|
Adjusted mean difference debridement only versus debridement with device: OSS at 12 months: -4.2 (95% CI: -8.2 to -0.26) Constant score at 12 months: -13.8 (95%CI: -24.0 to -3.6) Abduction angle at 12 months: -34.1 (-77.1 to 8.8) Flexion angle at 12 months: -56.8 (-91.1 to -22.5) Abduction strength at 12 months: -2.3 (-3.8 to -0.8) WORC Index at 12 months: -8.4 (-16.8 to -0.1) EQ-5D-5L at 12 months: -0.056 (-0.150 to 0.035) | There were no clear differences in safety events between the two groups. Adverse events in debridement with device group: Overall: 11/56 (20%) participants had any adverse event: 6/56 exacerbation/persistence of shoulder pain or restrictive ROM, 3/56 injection into the shoulder region, 2/56 adhesive capsulitis, 1/56 persistent muscle soreness or muscle injury. 4/56 (7%) had a serious adverse event – 2 deemed related to the surgery (1 persistent pain or disability at 12 months, 1 further surgery required). Adverse events in debridement only group: Overall: 9/61 (15%) had any adverse event: 5/61 exacerbation/persistence of shoulder pain, 1/61 injection into shoulder region. 2/61 (3%) had a serious adverse event – 1 deemed related to the surgery (persistent pain or disability at 12 months). | |
Verma, 2022 | Outcomes of InSpace implant were comparable with those of partial repair at Month 24. Mean operative time: InSpace implant group 44.6 mins versus Partial repair group 71.2 mins (p<0.0001). There was earlier recovery of outcome in the InSpace group compared with partial repair group. ASES score (primary outcome): Improvement from baseline to month 24: InSpace group: 46.22±20.89, p<0.0001 versus partial repair group: 42.53±20.54, p<0.0001. Patients achieving MCID at 24 months: 83% InSpace group versus 81% partial repair (NS). Secondary outcomes:
| No device related surgical complications were noted. 4/93 (4%) re-operations required after InSpace implantation versus 3/91 (3%) re-operations after partial repair. |
Osti, 2021 | Subacromial Spacer Implantation pre-op to post-op: Constant score: Mean improved from 35.8 to 64.8. ASES: Mean increased from 45 to 84. VAS: Mean improved from 6.1 to 3.5 OSS: Mean improved from 30.8 to 33.0 ROM forward elevation: Mean increased from 94° to 150°. UCLA: Mean increased from 10.9 to 15.9 Patient satisfaction: 80.3% overall satisfaction rate. ASCR pre-op to post-op: Constant score: Mean improved from 41.8 to 70.4. ASES: Mean increased from 44 to 86. VAS: Mean improved from 5.2 to 1. OSS: Mean improved from 17.9 to 38.5. ROM forward elevation: Mean increased from 105° to 133°. UCLA: Mean increased from 9.9 to 32.4. Patient satisfaction: 76.2% overall satisfaction rate. | Subacromial Spacer Implantation: Complications reported in 25 (6.7%) patients post-op: in 3 patients the balloon migrated, 18 patients pain persisted (12 underwent reverse total shoulder arthroplasty, in 3 the balloon was reimplanted), 1 patient transient neural damage with forearm dysesthesia, 1 patient superficial wound infection, 1 patient deep wound infection treated with balloon removal, 1 patient a persistent limited motion treated with latissimus dorsi tendon transfer. ASCR: Complications reported in 92 (14.8%) patients post-op: 34 graft tears, 7 suture anchor pull-out, 6 severe shoulder contracture, 5 deep infections, 33 graft failures, 2 persistent shoulder pain, 1 anterior shoulder escape. |
Johns, 2020 | Constant score (assessed by 11 studies) All reported statistically significant improvement in Constant Score from pre-op to post-op at all timepoints. Pre-op range 22.5-41.8, post-op range 51.4-72.3. OSS (assessed by 3 studies) Pre-op range 21.3-26, post-op range 34.39-48.2 ASES score (assessed by 4 studies) All showed statistically significant improvement from pre-op to post-op. VAS pain score (assessed by 3 studies) 1 study showed statistically significant improvement at 3, 6, 12 and 24 months. 24-month result: 6.6-2.8, p=0.0019. 1 study showed statistically significant improvement at 12 and 24 months. 24-month result: 7.1-2.1, p<0.0001). 1 study showed statistically significant improvement following both partial repair with spacer and spacer alone, with no statistically significant difference between groups. UCLA shoulder score (assessed by 1 study) Improved from 10.9±3.24 pre-op to 15.9±6.87 post-op, p=0.001 Shoulder ROM (assessed by 4 studies) Statistically significant improvement of shoulder abduction (pre-op range: 70-113°, post-op range: 110-165°), shoulder flexion (pre-op range: 80-130°, post-op range: 106.5°-161°) and external rotation (pre-op range: 25-44.5°, post-op range: 35-63.7°). Patient satisfaction (assessed by 4 studies) Mean of 3.7 on the 4-point Likert satisfaction scale. 13 of 15 patients rated their satisfaction from 8-10 on a 10-point scale, with 10 representing 'very satisfied'. 1 study: 25/31 patients (80.6%) were fully or almost satisfied, 3/31 (9.6%) reported moderate satisfaction, 3/31 (9.6%) no satisfaction. 1 study: 11/24 (46%) of patient's satisfied. |
|
Malahias, 2021 | All mean post-operative clinical and functional scores of both groups statistically significantly improved in comparison to the mean pre-operative value. Patients treated with partial repair and spacer implantation had a propensity toward better functional outcomes compared to partial repair alone, but these were statistically non-significant differences. Partial repair and spacer group pre-op to post-op changes: Constant score: Increased from mean 38.8 (SD 19.9) to mean 75.8 (SD 12.1), p<0.001. ASES score: Increased from mean 47.7 (SD 19.1) to mean 89.8 (SD 10.9), p<0.001. VAS pain: Decreased from mean 53.8/100 (SD 29.4) to mean 16.9/100 (SD 23.0), p<0.001. ROM: Shoulder forward flexion: Improved from mean 128.8° (SD 56.0) to mean 175.6° (SD 7.3), p=0.02. % achieving MCID of Constant score: 93.8%. % achieving MCID of ASES score: 93.8%. Partial repair only pre-op to post-op changes: Constant score: Increased from mean 41.7 (SD 15.6) to mean 69.6 (SD 19.7), p<0.001. ASES score: Increased from mean51.0 (SD 16.5) to mean 79.8 (SD 18.8), p<0.001. VAS pain: Decreased from mean 41.3/100 (SD 30.9) to 8.7/100 (SD 15.5), p<0.001. ROM Shoulder forward flexion: Increased from mean 140.7° (SD 50.9) to 171.6° (SD 23.7), p<0.05. % achieving MCID of Constant score: 87.5% % achieving MCID of ASES score: 87.5% | Spacer only group: No re-operations or major complications. Partial repair only group: 1 patient suffered a deep infection requiring a revision shoulder arthroscopy. |
Maman and Kazum, 2022 | ROM: Abduction: Mean improvement of 14° (from a mean of 106 pre-op to 120°). External rotation: Mean improvement of 2° (from a mean of 36 pre-op to 38°). Patient report of a positive effect on their conditions: 51 (65%). Patient reported they would repeat the procedure in hindsight: 45 (58%) |
|
Bilsel, 2022 | Pre-op & post-op outcome scores. Value (range) Pre-op median ASES (range) Partial repair 30.0 (20-37.5) Partial repair + spacer 30.8 (20-42) p-value 0.4 Post-op median ASES (range) Partial repair 55.0 (37.5-65) Partial repair + spacer 75.5 (55-88.3) P-value <0.001 Δ median ASES Partial repair 28.0 (7-40) Partial repair + spacer 40.2 (26.7-63.3) P-value <0.001 % achieving MCID ASES Partial repair 70 Partial repair + spacer 100 P-value 0.04 Pre-op median Constant score (range) Partial repair 26.0 (20-38) Partial repair + spacer 28.5 (20-40) P-value 0.6 Post-op median Constant score (range) Partial repair 55.0 (31-79) Partial repair + spacer 40.0 (43-79) P-value 0.01 Δ median Constant score (range) Partial repair 29.0 (8-53) Partial repair + spacer 39.0 (23-53) P-value 0.01 % achieving MCID constant score Partial repair 95 Partial repair + spacer 100 P-value 0.6 Pre-op median VAS (range) Partial repair 8.0 (7-9) Partial repair + spacer 7.5 (6-9) P-value 0.6 Post-op median VAS (range) Partial repair 2.0 (0-4) Partial repair + spacer 1.0 (0-3) P-value 0.04 Δ median VAS (range) Partial repair 5.5 (3-8) Partial repair + spacer P-value 0.1 % achieving MCID VAS Partial repair 100 Partial repair + spacer 100 P-value n/a Pre-op median forward flexion (range) Partial repair 100.0° (75-120°) Partial repair + spacer 105.0° (75-120°) P-value 0.5 Post-op median forward flexion (range) Partial repair 120.0° (80-153°) Partial repair + spacer 140.0 (90-150°) P-value 0.01 Δ median forward flexion (range) Partial repair 17.5° (-10, 33°) Partial repair + spacer 30.0 (15-40°) P-value <0.001 Pre-op median abduction (range) Partial repair 80.0° (60-100°) Partial repair + spacer 85.0 (60-100°) P-value 0.5 Post-op median abduction (range) Partial repair 90.0° (70-110°) Partial repair + spacer 100.0° (70-130°) P-value 0.03 Δ median abduction (range) Partial repair 10.0° (-10-30°) Partial repair + spacer 20.0° (0-40°) P-value 0.05 Pre-op median external rotation Partial repair 3.0° (2-3°) Partial repair + spacer 3.0° (2-3°) P-value 0.9 Post-op median external rotation Partial repair 3.0° (2-4°) Partial repair + spacer 3.0° (2-5°) P-value 0.4 Δ median external rotation Partial repair 0.0° (-1-2°) Partial repair + spacer 1.0° (-1-2°) P-value 0.5 | Not assessed |
Davey, 2021 | Final follow-up: (NB no pre-op values measured) Mean ASES score: 73.4 ± 21.8 Mean Subjective Shoulder Value (SSV): 76.4 ± 16.0 Patient satisfaction: % relatively satisfied: 40 (88.9%) % very satisfied: 37 (82.2%) % that would opt to have procedure again: 40 (88.9%) |
|
Senekovic, 2017 | Change between pre-op and 3, 4 and 5-year follow-up: Total Constant Score: 3 years: +23.28 (19.42), p<0.0001 4 years: +26.55 (19.51), p<0.0001 5 years: +28.56 (17.65), p<0.0001 At 5 year follow-up, 84.6% showed improvement of 15 points, 61.5% showed improvement of 25 points. |
|
Procedure technique
All 9 studies detailed the procedure technique and devices used. All used the InSpace implant (Stryker, US) as the surgical device for insertion.
As outlined previously, there have been some differences in the surgical technique used whilst inserting the InSpace device. Of the studies comparing spacer implantation with another group of patients, 1 RCT compared debridement with spacer implantation with debridement only as the control group (Metcalfe 2022). It has been proposed by a recent review that extensive debridement in addition to spacer implantation may theoretically lead to balloon migration and therefore inferior outcomes (Mease 2023). The other RCT compared InSpace implantation without repair with partial repair (Verma 2022). The case-control study and the comparative study both compared partial repair with spacer implantation with partial repair only (Malahias 2021 and Bisel 2022). Meanwhile, the systematic review by Osti et al. collated evidence from studies on patients who had a spacer implanted and compared outcomes to those patients who had ASCR.
Efficacy
Oxford Shoulder Score
The OSS was assessed by 1 RCT and 2 systematic reviews. The RCT found a statistically significantly higher OSS in the control group (debridement only) compared with the intervention group (debridement with spacer) at 12 months follow up (OSS of 34.3 [SD 11.1] in the debridement group versus 30.3 [10.9] in the debridement plus device group, mean difference -4.2 [95% CI -8.2 to -0.26; Metcalfe 2022]. A systematic review comparing spacer implantation with ASCR found a higher post-operative OSS in patients having ASCR compared with those treated with spacer implantation (mean increase in OSS from pre-operation to post-operation of 30.8 to 33.0 in the spacer implantation group versus 17.9 to 38.5 in the ASCR group). The spacer implantation group consisted of 128 patients from 4 studies while the ASCR group consisted of 25 patients from 1 study (Osti, 2021). Another systematic review found a statistically significant increase in OSS in patients having spacer implantation at a mean follow up of 33 months (pre-operative range 21.3 to 26; post-operative range 34.4 to 48.2; Johns 2020).
ASES
ASES was examined by 1 RCT, 2 systematic reviews, 1 case-control study, 1 retrospective comparative study and 1 case series. The RCT found statistically significant and comparable improvements in ASES from baseline to month 24 in both the spacer and partial repair groups. (The InSpace group improvement from baseline to month 24 was 46.2 [SD 20.9], p<0.001 compared with 42.5 [SD 20.5], p<0.001.) There was no statistically significant difference in the percentage of patients achieving MCID in ASES at 24 months (83% for the InSpace group versus 81% for the partial repair group; Verma 2022). In contrast, a comparative study found the percentage achieving MCID in ASES was statistically significantly higher in the partial repair with spacer group compared with the partial repair only group (100% compared with 70%, p=0.04; Bilsel, 2022). A systematic review found a slightly greater increase in ASES in the ASCR group compared with the spacer group (mean increase in the spacer group of 45 to 84 compared with a mean increase from 44 to 86 in the ASCR group; Osti 2021). A case-control study found statistically significant improvements in ASES in both the partial repair with spacer and partial repair only group at 12 months follow up (mean improved from 47.7 [SD 19.1] to 89.8 [SD 10.9], p<0.001 in the partial repair plus spacer group compared with mean improvement from 51.0 [SD 16.5] to 79.8 [SD 18.8] in the partial repair group, p<0.001; Malahias 2021). A systematic review with mean follow up of 33 months, which included 4 studies examining ASES found a statistically significant improvement in ASES score after spacer insertion, with a pre-operative range of 24.5 to 59.1 and a post-operative range of 72.5 to 85.7.
Constant Score
Constant Score was examined by 2 RCTs, 2 systematic reviews, 1 case-control study, 1 comparative study and 1 case series. The 2 RCTs had conflicting results, with one finding a mean difference at 12 months follow up in Constant Score between the debridement only and debridement with device groups of ‑13.8 (95% CI -24.0 to -3.6 favouring the debridement only group; Metcalfe 2022). In contrast, another RCT found a statistically significant difference in the improvement in Constant Score between the InSpace and partial repair groups at both the 6‑week and 24‑month follow-up points, favouring the InSpace group (no figures available; Verma 2022). The comparative study also found a statistically significant difference between the spacer with partial repair and partial repair only groups, favouring the spacer group (change in median Constant Score of 29.0 for partial repair compared with 39.0 for partial repair with spacer, p=0.01; Bilsel 2022). All 11 studies assessing Constant Score in the systematic review by Johns et al. reported a statistically significant improvement in Constant Score after spacer insertion (pre-operative range 22.5 to 41.8 and post-operative range 51.4 to 72.3). One case series examined change in Constant Score between pre-operation and post-operation at 3, 4 and 5 years follow up. At all follow-up points, there continued to be a statistically significant improvement compared with baseline. At 5 years follow up, the mean improvement in Constant Score was 28.6 (SD 17.7), p<0.0001.
WORC score
WORC score was assessed by the 2 RCTs. One RCT found a mean difference in WORC score between the debridement only compared with the debridement with device group which favoured debridement only (mean difference -8.4 [95% CI -16.8 to -0.1] Metcalfe 2022). The other RCT showed no statistically significant difference in improvement between baseline and all follow-up points between the InSpace and partial repair groups, apart from at day 10 which favoured the InSpace group (figures not available).
VAS pain score
VAS pain score was assessed by 1 RCT, 2 systematic reviews, 1 case-control study and 1 comparative study. The RCT found no statistically significant difference between the InSpace and partial repair groups in terms of improvement in VAS at any follow-up time point (Verma 2022). A systematic review found a mean improvement in VAS after spacer implantation to be 6.1 to 3.5 (98 patients) compared with 5.2 to 1 in ASCR (340 patients; Osti 2021). Another systematic review included 3 studies assessing improvement in VAS after spacer implantation, all 3 of which showed statistically significant improvements compared with the pre-operation VAS pain value. The comparative study found that 100% of patients in both the partial repair and partial repair plus spacer groups had an MCID in VAS pain score after their procedure (Bilsel 2022). The case-control study showed statistically significant improvements in VAS in both the partial with spacer and partial only groups (reduction in mean VAS of 53.8 of 100 [SD 29.4] to 16.9 of 100 [23.0] p<0.001 in the partial and spacer group compared with mean reduction from 41.3 of 100 [30.9] to 8.7 of 100 [15.5], p<0.001 in the partial only group; Malahias 2021).
EQ-5D-5L
EQ-5D-5L was investigated by the 2 RCTs. Both found no statistically significant difference between the spacer and control groups in terms of improvement in EQ-5D-5L between baseline and any follow-up time point. One found a mean difference between the debridement only and debridement plus device group of ‑0.056 (95% CI -0.150 to 0.035).
ROM
Active ROM was assessed by 2 RCTs, 2 systematic reviews, 1 case series and 1 comparative study. It is important to note that not all studies examined the same movements. 1 RCT and 1 comparative study had conflicting results in terms of forward flexion. The RCT found that at 12 months follow up, the control group had a statistically significantly greater increase in flexion compared with the spacer group (mean difference -56.8, 95% CI -91.1 to -22.5; Metcalfe 2022). In contrast, the comparative study found a statistically significantly greater median change in forward flexion angle in the spacer group compared with the partial repair group (median increase of 17.5 degrees versus 30.0 degrees, p<0.001; Bilsel 2022).
There was a statistically significantly greater improvement in forward elevation from baseline to all follow-up time points (day 10, week 6, month 12 and month 24) in the spacer group compared with the partial repair group in 1 RCT (Verma 2022). A systematic review found that mean forward elevation in 288 patients with a spacer implant increased from 94 degrees pre-operation to 150 degrees post-operation, compared with the mean increase in the ASCR group which was from 105 degrees to 133 degrees.
In the 4 studies which examined abduction in a systematic review, all showed statistically significant improvement in abduction between pre-operation and post-operation (pre-operation range 70 to 113 degrees compared with post-operation range 80 to 130 degrees; Osti 2021). But the comparative study found no statistically significant difference in the change in abduction angle between the partial repair only and InSpace implantation groups (Bilsel 2022).
Patient satisfaction
Patient satisfaction was assessed in 2 systematic reviews and 1 case series. One systematic review found an overall satisfaction rate of spacer implantation to be 80% compared with 76% for those who had ASCR (Osti 2021). Another systematic review found that in 1 study, 13 of 15 patients rated their satisfaction after spacer implantation between 8 and 10 on a 10‑point scale, with 10 representing very satisfied while another study found that 81% were fully or almost satisfied, 10% reported moderate satisfaction and 10% no satisfaction. However, a further study included in the systematic review found that only 46% were satisfied (Johns 2020). In the case series, 82% were very satisfied after spacer implantation.
Safety
Deep wound infection
One out of 350 patients in a systematic review had a deep wound infection which needed a 1‑week course of intravenous antibiotics followed by 2 weeks of oral antibiotics.
Persistence or exacerbation of shoulder pain, or persistent limited motion
Three studies (1 systematic review, 1 RCT and 1 case series) reported patients who had persistent worsening of their shoulder pain or continued limited ROM. The systematic review by Osti et al. noted that 19 of 373 (5%) patients experienced this after balloon implantation. The RCT found that 6 of 56 (11%) patients were experiencing these symptoms by 12 months follow up, with 1 further patient experiencing persistent muscle soreness or muscle injury (Metcalfe 2022). In the case series, 1 out of 45 patients experienced persistent symptoms and required removal of suture anchor for residual pain within the 37 month follow-up period (Davey 2021).
Re-operations
Several studies noted the risk of a further operation being needed. In an RCT, 4 of 93 (4%) patients required a re-operation by 24 months follow up (1 arthroscopy for persistent pain, 2 conversions to RTSA for failure, 1 conversion to RTSA for fracture non-union after a fall). A systematic review found that at mean follow up of 33 months, re-operation was required in 11 of 350 (3%) patients, including 5 of 350 (1%) for balloon migration, 1 of 350 for synovitis and 6 of 350 (2%) underwent RTSA because of absence of or worsening symptoms (Johns 2020). In a case series, 3 (7%) needed a subsequent procedure, including 2 (4%) for removal of the balloon (Davey 2021).
Superficial wound infection
Superficial wound infection was documented in 1 of 350 patients in a systematic review. This resolved after a course of antibiotics. Furthermore, 2 of 78 (3%) in a case-control study experienced a superficial wound infection.
Synovitis
Synovitis was documented in 4 patients included in a systematic review (Johns 2020). This was found on MRI at 3 years post implantation of the spacer device. Two of 24 patients in a case series presented within the 5-year follow-up period with synovitis (Senekovic 2017). But because there was no pre-operative imaging available, it was unclear if this was related to the device.
Recurrent rotator cuff tear
Recurrent rotator cuff tear was experienced by 1 of 24 patients in a case series with 5-years follow up (Senekovic 2017).
Transient neural damage with forearm dysesthesia
One of 373 patients experienced transient neural damage with forearm dysesthesia in a systematic review with mean follow up of 27 months (Osti 2021).
Shoulder dislocation
One of 350 patients experienced shoulder dislocation in a systematic review with mean follow up 33 months. It was documented as being secondary to acute trauma. (Johns 2020).
Remnants of deflated balloon transforming into scar tissue
Remnants of deflated balloon transforming into scar tissue was documented in 1 of 350 patients in a systematic review with mean follow-up time of 33 months (Johns 2020).
Anecdotal and theoretical adverse events
Expert advice was sought from consultants who have been nominated or ratified by their professional society or royal college. They were asked if they knew of any other adverse events for this procedure that they had heard about (anecdotal), which were not reported in the literature. They were also asked if they thought there were other adverse events that might possibly occur, even if they had never happened (theoretical).
They listed the following anecdotal adverse events:
anterior escape of the balloon in the shoulder leading to pain
inserting a balloon which is too large and overfilling the device
failure to ensure that the device is appropriately sited
balloon bursting.
Four professional expert questionnaires for this procedure were submitted. Find full details of what the professional experts said about the procedure in the specialist advice questionnaires for this procedure.
Validity and generalisability
Overall, results from studies which have simply compared shoulder functioning and pain before and after insertion of a biodegradable spacer have mostly shown improved shoulder functioning and reduced pain after biodegradable subacromial spacer insertion. The procedure also appears to have a low rate of complications. But the 2 recent RCTs have conflicting findings, with 1 showing non-inferiority to partial rotator cuff repair and the other finding inferiority to debridement alone.
Pre-operative patient selection may have contributed to these conflicting results. The 2 RCTs had a difference in their study population's pre-operative active forward flexion (74.1 in the UK study by Metcalfe et al. and 115 in the US/Canada study by Verma et al.). It has been proposed that pre-operative ROM may influence final outcomes (Mease 2023). In the Verma et al. study, participants were screened by MRI to identify tears ≥ 5 cm and involving greater than or equal to 2 tendons. In comparison, in the Metcalfe et al. study there was no cut-off for tear size (Mease 2023). Furthermore, the 2 RCTs had differing post-operative rehabilitation protocols. An important finding from both RCTs was that male participants performed better than females (although subgroup comparison was based on small numbers). Finally, it is important to note that the study by Verma et al. was funded by OrthoSpace (now Stryker), the manufacturer of the InSpace device, and 2 of the authors are Stryker or OrthoSpace employees.
The systematic reviews on this topic are limited by most of the studies being case series with small numbers of patients and relatively short follow up.
How are you taking part in this consultation?
You will not be able to change how you comment later.
You must be signed in to answer questions