Subacromial balloon spacer – Where are we now?

Abstract

Introduction

Rotator cuff tears are a common cause of disability and pain. The ideal treatment for truly irreparable rotator cuffs is still debated, and one recent surgical advance is the development of the subacromial balloon spacer. This review aims to clarify the current evidence and indications for this device.

Methods

A comprehensive literature search was undertaken using the MeSH search terms combining “balloon spacer” and “irreparable cuff tear”. A total of 20 studies using the balloon spacer as a treatment modality in more than two patients, were analysed.

Results

A total of 513 patients were analysed, representing 83% of those initially identified as meeting the inclusion criteria. The majority of studies recommended the device, with only four suggesting it was not recommended based on their results. Notable bias was present in the studies analysed, and there were no papers providing greater than level III evidence.

Conclusion

The subacromial balloon spacer is one possible treatment option for older, low-demand patients with a full thickness rotator cuff tear involving only the supraspinatus tendon, who also have no arthritis and have preserved active elevation beyond 90°. However, the results of two large randomised prospective trials are awaited to provide satisfactory evidence regarding the use of the balloon spacer.

1. Introduction

Rotator cuff tears are a common cause of shoulder pain and can lead to disability, loss of work and a limitation in daily activities.¹ In most cases, repair of the tear is possible and has proven clinical benefit and good patient satisfaction.^2,3 However, there are some cases where the tear cannot reliably be repaired. This is most commonly seen with massive chronic rotator cuff tears.³ Many of these patients will improve with conservative measures, such as corticosteroid injections and physiotherapy.^4,5 However, there are some patients with irreparable cuff tears who do not respond to non-operative measures.⁶ In this clinical scenario, there is no simple surgical solution.

For older patients who fall into this category, a reverse polarity shoulder arthroplasty may be a good option.⁷ It can provide predictable improvement in symptoms and allow a return to activities of daily life. However, it carries its own risks, and is not an ideal choice for a younger or more active patients. For such a patient, options include:

• -debridement and bursectomy alone
• -partial tendon repairs
• -augmented or patch graft repairs
• -tendon transfers
• -superior capsular reconstruction (SCR)
• -subacromial balloon spacer

These last two options first had published reports in 2012.^8,9 Both aim to address the potential superior migration of the humeral head that can occur in the presence of a massive rotator cuff tear. Whilst the SCR is a technically demanding procedure involving anchor fixation of a graft onto both the glenoid and humerus; the balloon spacer is a relatively quick procedure where the device is inserted through an arthroscopic portal and positioned in the subacromial space.

The aims of this review are to clarify the indications for the balloon spacer, outline the surgical technique, summarise the biomechanical theory, and evaluate the current clinical results.

2. Indications

In 2012, Savarese and Romeo introduced a novel surgical technique using a new biodegradable subacromial spacer, the InSpace™ balloon⁹ (then OrthoSpace, Kfar, Israel; now Stryker, Michigan, USA). The indications were patients with a torn and irreparable supraspinatus tendon with or without a torn and irreparable infraspinatus, and an intact or repairable subscapularis tendon. The only listed contraindications were active joint infection, glenohumeral arthropathy, allergy to device material, and axillary nerve palsy.⁹

As more clinical and biomechanical studies were performed, two more factors became relevant. The need for an intact and functioning subscapularis and teres minor to restore force coupling of internal and external rotation; and the exclusion of patients with pseudoparalysis, who are likely beyond the capabilities of the balloon spacer.¹⁰

An irreparable rotator cuff is not necessarily the same as a massive rotator cuff tear, as in the acute setting a massive cuff tear may be repairable. There is no universal agreement on the definition of a massive irreparable rotator cuff tear (MIRCT), but previous studies have suggested that both size and chronicity are implicated.^11,12 Cofield suggested that a tear >5 cm is a massive cuff tear,¹³ but Gerber argued that due to differences in patient size, a tear involving two or more tendons is a more appropriate measure.³ The chronicity of the tear is reflected by the degree of tendon retraction,^12,14 degree of muscle atrophy,^11,15 and the degree of fatty infiltration.¹⁶ In addition, patients with pseudoparalysis (full passive range of motion, but <90° of active elevation), or with external rotation lag signs, are unlikely to have a successful rotator cuff repair.¹⁷ A superiorly migrated proximal humerus head is a poor prognostic indicator.^18,19

In most published studies evaluating the clinical results of the subacromial balloon spacer, patient inclusion criteria contain two or more of the above signs of chronicity and tear size, as well as a failed trial of non-operative management. The goals of this treatment, are to provide adequate pain relief through a range of motion. In order to do this, the balloon aims to restore glenohumeral joint kinematics by pushing the humeral head inferiorly so recentering the head in the glenoid.

3. What is it?

Unlike traditional bioinert implants, which stay in the body forever or until they are surgically removed, biodegradable implants degrade and are then eliminated over time. The InSpace™ balloon spacer is a biodegradable balloon film inflated in situ to separate tissues and protect them from mechanical or other stresses. It is currently also being trialled as a balloon spacer behind the prostate to protect the rectum in prostate cancer patients undergoing radiotherapy.²⁰

The Inspace™ balloon is made from poly (l-lactide-co-ε-caprolactone) (PLCL), a polymer that combines the toughness of poly-l-lactide with the elasticity of ε-caprolactone in a 70:30 ratio. It is degraded in a two-step process, which involves a well-ordered foreign body reaction leading to complete removal of the material. Histological studies of the material have shown no evidence of chronic active inflammation and no evidence of systemic toxicity.²¹ The elastic property of the PLCL can resist extreme cyclic pressure²² and it retains its tensile strength even when undergoing in vivo degradation.²³ In vivo studies have found that by week 26, the implant is reduced to 61% of its original weight and is degraded and adsorbed over this time period.²³ Small fragments of the polymer undergo spontaneous metabolisation once the degradation process starts.

The bio-balloon is folded into a cylinder and placed into a sheath in a trocar for easy arthroscopic insertion. It is deployed out of the trocar intra-operatively, expands is then filled with saline before being sealed.

Animal testing of the material and implanting a miniature version of the balloon showed that the fluid in the device reduces and is no longer detectable from week 22²¹. It was seen to degrade appropriately with no irritant properties and was associated with a well-ordered host response. There were, however, some complications: one animal developed a fibrosarcoma at the implant site, which the authors stated to be a rodent-related phenomenon, and one animal had ulceration of the skin overlying the device after it had migrated. Another animal also had migration of the device which eventually protruded through the skin and had to be removed.²¹ These were not considered to be safety concerns, however.

4. Surgical technique

The surgical technique was originally described by Savarese and Romeo,⁹ and then by Szöllösy in 2014.²⁴ Although it can be performed as a fluoroscopic technique without direct visualisation,^25,26 it is more commonly inserted under direct vision during an arthroscopic procedure, when the repairability of the rotator cuff can also be accurately assessed. Patient positioning can be either beach chair or the lateral decubitus. However, it has been suggested that the former gives a gravitational advantage and allows for a more accurate estimation of the acromiohumeral distance.¹⁰ Prior to deciding on the optimal size, debridement and bursectomy are performed as well as long head of biceps tenotomy if still present.^10,27 The subacromial space can then be measured, using the arthroscopic probe, in the anterior-posterior and medial-lateral directions. This should enable selection of the best balloon spacer size – small, medium or large.

The subacromial balloon spacer comes folded on the end of an introducer, which is inserted via the lateral portal. As Horneff III and Abboud suggest, placement through the posterior portal gives better visualisation.¹⁰ They also keep bursectomy to a minimum to avoid the complication of balloon migration. An extension tube can then be connected to the delivery system, and the balloon filled with the appropriate amount of warmed saline, between 9 ml and 25 ml depending on the size of implant.²⁷ The balloon can then be sealed and position confirmed arthroscopically.

5. Biomechanical theory

The InSpace™ balloon spacer is used to reduce friction between the humeral head and the undersurface of the acromion.²⁷ By depressing the humeral head inferiorly and restoring the kinematics of an intact shoulder, it aims to improve pain and function. Superior humeral head migration, with disruption of Moloney’s line,²⁸ is a feature of a large posterosuperior cuff tear, and is characterised radiographically by a reduced acromiohumeral interval of less than 7 mm¹⁹ (Fig. 1). Cadaveric modelling has shown that the balloon spacer can depress the humeral head by more than 2 mm inferior to its anatomic position²⁹ (Fig. 2). Improper device positioning however, can cause forces in unwanted directions and again reduce the contact area with the glenoid. In a study of 14 cadaver models with created irreparable cuff tears, the balloon spacer increased glenohumeral contact pressures by 122% at 0° abduction and by 94% at 60° abduction.³⁰ This study showed that although the balloon was able to restore contact pressure, it was not able to restore contact area. The humeral head was moved inferiorly by 6.2 mm at 0° abduction, 4.4 mm at 30° abduction, and 3 mm at 60° abduction, which gave only 40% of the intact state contact area.

Fig. 1.

A line drawing showing superior hhumeral head migration and disruption of Moloney’s line in a torn rotator cuff tendon.

Fig. 2.

A graphic representation of the balloon spacer acting to restore the humeral head position in realtion to the glenoid, and thereby restore Moloney’s line.

As a previous study showed, the abduction angles where the rotator cuff tendons are closest to the under surface of the acromion are between 60 and 120°.³¹ Another cadaveric study looked at peak subacromial pressures, which were noted to significantly reduce after introduction of the balloon.³² In her thesis, Singh aimed to compare the effects of the SCR and the subacromial balloon spacer, hypothesising that the balloon would function better than SCR at depressing the humeral head at higher abduction angles.³³ This work found that both techniques restore functional abduction in cadaveric simulation,³⁴ but that a fill volume of 10 ml in the balloon was ineffective and required the manufacturer-recommended 25 ml to depress the humeral head at all abduction angles.³⁵

6. Search methodology

A literature search using the MEDLINE, SciELO, Web of Science Core collection and BIOSIS databases was conducted using the search terms “balloon spacer” OR “interposition balloon” OR “interpositional balloon” OR “subacromial spacer” OR “subacromial balloon” OR “InSpace” OR “balloon arthroplasty” OR “biodegradable spacer” AND “shoulder” OR “rotator cuff” OR “cuff tear” OR “massive cuff” OR “irreparable cuff” OR “tendinopathy” OR “impingement” OR “subacromial”, and limited to the English language. This yielded 75 articles published between 1900 and 2021. A Cochrane Library search was also conducted using the same search criteria and yielded 14 results. The articles were then screened by the authors and 24 articles were identified as completed non-review articles in English giving clinical or biomechanical outcomes of the subacromial balloon spacer. Two studies were excluded as duplicate patient cohorts. 20 studies included more than two patients. Fig. 3 is a flowchart showing the search methodology. A further ten studies were lab-based animal or cadaveric studies. One study provided a cost-analysis, 14 were either review articles or editorials, and a further three were surgical technique articles. The remaining 23 articles were either not deemed relevant, or were articles with as yet unpublished outcomes.

Fig. 3.

A flowchart to show the search methodology undertaken and number of studies included.

7. Clinical results

The in vitro and cadaveric testing suggests that the subacromial balloon spacer will act to depress the humeral head and restore function. Clinical results of the device have been published since 2012 and have looked into various outcomes.

Senekovic published the first clinical results of 20 patients in 2012, of these, three were deemed as lost to follow-up, but two of these had withdrawn as they were dissatisfied. Of the 17 patients analysed, they reported no adverse events, and 55.5% of patients improved by the minimally important clinical difference for the Constant-Murley score (CS).³⁶ They also published their 5-year outcomes for 24 patients, which included the above cohort. Of these 24, 9 were lost to follow-up (37.5%).

Following this, another Israeli study showed that the device could successfully be deployed under local anaesthetic and fluoroscopically-guided into the correct position.^25,26 This had the advantage of confirming depression of the humeral head after balloon inflation, but does not allow for intra-operative assessment of the rotator cuff tear itself. It also does not allow for any bursectomy or debridement of the cuff which is necessary for accurate sizing as suggested in the surgical technique guide. The 15 patients in this study had an average age of 74 and a third had failed previous surgical intervention. The authors reported that all had improvement in their American Shoulder and Elbow Score (ASES) and CS, although one patient was referred for reverse shoulder arthroplasty (rTSA) at six weeks post-intervention, suggesting no significant improvement in their outcome.

Most studies included patients with symptomatic irreparable cuff tears who had failed non-operative treatment for a variable minimum period. Five studies had defined radiological inclusion criteria such as tendon retraction of Patte stage 3, fatty infiltration according to Goutallier’s classification of stage 3 or 4 ^37-41, and one more used MRI criteria that was not described.⁴² 15 studies stated glenohumeral joint osteoarthritis (OA) as an exclusion criterion. Five studies regarded pseudoparalysis in their cohort; 2 of these excluded patients with pseudoparalysis,^43,44 and 3 included them.^40,45,46 (See Table 1 for study information). It has been suggested that pseudoparalytic patients would not benefit from subacromial balloon spacer,^10,43,46 but one study found good results even in this cohort of patients.⁴⁵ Where described, set-up was mainly in the beach chair position, but three studies included those with lateral decubitus positioning.^39,47,48 Most surgical techniques included a bursectomy and long head of biceps tenotomy, but significant variation existed between the studies regarding what additional surgical procedures were performed. Some studies included those with concomitant partial cuff repairs^43,45,46,48, 49, 50; whereas others excluded those patients who had partial repairs.^36,51 Malahias et al. noticed that in their initial study, where they added a partial cuff repair in association with balloon spacer intervention, this provided questionable additional benefit.⁴³ They subsequently undertook a 2-arm intervention looking at partial cuff repairs with and without the use of the balloon spacer and found no significant differences, and therefore could not recommend the use of the device.⁴⁹ Outcome measures were also variable, with some studies purely analysing the range of movement, and others including validated patient-reported outcome measures.

Table 1.

Information for all clinical studies analysed. [MIRCT = massive irreparable rotator cuff tear; GHJ OA = glenohumeral joint osteoarthritis; AHI = acromiohumeral interval; rTSA = reverse polarity total shoulder arthroplasty].

1st Author (Year)	Inclusion criteria	Exclusions	Patients meeting inclusion criteria	Patients analysed	Av age (youngest)	Other surgical procedure	Av op time	Minimum Follow-up (Average)	Study design	Conflict of Interest	Revisions	Complications	Recommend use	Comments
Bakti (2018)46	MRI proven MIRCT; failed conservative management	GHJ OA; subscapularis tear; significant medical comorbidities	54	14	67 (49)	Bursectomy; LHB tenotomy	38min	5y	Retrospective case review	No	3 (all rTSA)	No	Yes	Done in lat decubitus or beach chair
Basat (2017)39	MIRCT (defined as Patte stage 3; Goutallier 3 or 4; AHI <7 mm; irreparability confirmed arthroscopically); failed 6 months conservative management; age >60	Cuff tear arthropathy; neurologic deficit (except pseudoparalysis); infection)	12	12	64.3 (60)	Debridement; LHB tenotomy	33min	24 months (38.3 months)	Retrospective case review	No	0	No	Yes	Included pseudoparalysis patients
Deranlot (2017)40	MIRCT (defined as Patte stage 3; Goutallier 3 or 4; irreparability confirmed arthroscopically); no OA (<grade 3 Hamada)	OA grade >3 Hamada; subscapularis tear	39	39	69.8 (53)	Debridement; LHB tenotomy	Not stated	12 months	Retrospective case review	No	1 (for spacer migration)	1 Spacer migration	Yes	15% had progression of OA; minimal reduction in acromiohumeral distance
Familiari (2021)43	Large posterosuperior cuff tears on MRI >3 cm); failed conservative management >6 months	Pseudoparalytic; subscapularis tear; neurologic deficit; glenohumeral joint OA; previous surgery same joint	61	51	63 (50)	Bursectomy; LHB tenotomy	Not stated	12 months	Retrospective case review	No	6 (5 rTSA; 1 for progression of cuff arthropathy)	None	Yes	Included majority of patients with Goutallier grade 2 fatty infiltration; included those with stage 2 (Patte)tendon retraction; unclear whether those not satisfied included those revised
Gervasi (2016)25	MIRCT; failure of conservative management >4 months; age >50	GHJ OA; infection; instability; major joint trauma	15	15	74.6	N/A	10min	12 months	Prospective single arm	Yes	1 (1rTSA)	1 Persistent pain	Yes	Fluoroscopic insertion
Holschen (2017)45	MIRCT; loss of shoulder function	Cuff tear arthropathy; >grade 2 Hamada	25	23	63.5	Partial cuff repair; LHB tenotomy	Not stated	22.3 months (average)	Retrospective case-control – 2-arm	No	2 (both rTSA)	No	Yes	Excluded the 2 patients needing revision from analysed data; pseudoparalytic shoulder improved very little
Ibán (2018)38	MIRCT (defined as MRI confirmed massive cuff tear; retraction >3 cm; Goutallier 3 or 4; irreparability confirmed arthroscopically); failed conservative management >3 months; symptoms >6 months; age >50	OA or degenerative changes on radiology	16	15	Minimum 50y	Bursectomy; LHB tenotomy	27.3min	2y	Prospective longitudinal study	Yes	5 (all rTSA)	No	No	Done in lateral decubitus position
Malahias (2019)42	MIRCT radiologically and confirmed as irreparable arthroscopically; age >50	Pseudoparalysis; GHJ OA; systemic medical condition; irreparable subscapularis tear; previous surgery same joint	31	31	65.2	Bursectomy; partial repair	Not stated	6 months	Retrospective case cohort – 2-arm	Yes	2	No	Not conclusive	Included majority of patients with Goutallier grade 2 fatty infiltration; included acute tears as well as chronic
Malahias (2020)48	MIRCT (defined as >2 tendons torn on MRI, with tear size >3 cm; clinically massive cuff tear); age >50	Irreparable subscapularis tear; previous surgery same joint; infection; other systemic medical comorbidity	32	32	67.7	Partial cuff repair; LHB tenotomy	Not stated	12 months	Retrospective case-control – 2-arm	No	1 (revision arthroscopy in non-balloon arm)	No	No	Included majority of patients with Fuch’s grade 1or 2 fatty infiltration; included acute tears as well as chronic; 16 had balloon
Maman (2017)59	MRI or CT proven cuff tear; failure of conservative management >4months	OA; instability; previous shoulder surgery; infection; uncontrolled diabetes mellitus, coagulopathy and immunosuppression	58	42	Not stated	Debridement only; LHB tenotomy in one arm	Not stated	12 months	Prospective multicentre – 2-arm	Yes	0	No	Yes	10 patients excluded as fluoroscopic; a further 6 excluded for partial repair or not following protocol
Naggar (2018)44	MIRCT	Cuff tear arthropathy; insufficiency of external rotators	22	22	69.3 (52)	Partial cuff repairs	Not stated	52.5 months (average)	Prospective case series	Not declared	0	No	Yes	Included pseudoparalysis patients with good results; published abstract only
Oh (2019)49	MIRCT confirmed arthroscopically	Not stated	68	52	64.2	Partial cuff repair; patch graft; LHB augmentation; bursectomy; acromioplasty	80.3min	2 year	Retrospective cohort	No	0	5 (3 retears of partial cuff repairs; 2 new onset pseudoparalysis in those with only balloon spacer without subscapularis repair)	Yes	19 had balloon and 9 of these had full thickness subscapularis tears; patients with partial repairs and balloon fared better
Piekaar (2019)47	MIRCT on radiology and confirmed arthroscopically; failed conservative management or failed previous surgery	Severe OA; rupture of subscapularis; infection; allergy to device	46	39	65 (45)	Partial cuff repair; bursectomy; LHB tenotomy; distal clavicle excision	Not stated	34 months	Prospective single arm	No	2 (both rTSA)	1 superficial wound infection - but patient excluded from analysis	Yes	Lateral decubitus position; revision patients and complication excluded from analysis
Prasad (2018)57	MIRCT on MRI and confirmed arthroscopically	Not stated	25	17	Not stated	Not stated	Not stated	12 months	Prospective pilot study	Not declared	2 (both rTSA)	No	Not conclusive	Only abstract published
Prat (2018)51	MIRCT with failed non-op treatment	Inflammatory arthropathy; OA	26	24	70.7	Bursectomy, LHB tenotomy	Not stated	4 months (14 .4)	Retrospective case review	No	2 (unspecified)	4 (1 balloon migration; 1 wound infection; 1 lat cut n dysaesthesia; 1 deep infection)	No	Recorded UMI - no change pre- to post-op
Ricci (2017)37	MIRCT (defined clinically and on MRI with Goutallier 3 or 4) pain at least 6 months; failure of conservative management;	GHJ OA; previous shoulder surgery; infection	30	30	65.7	Acromioplasty, bursectomy and LHB tenotomy	Not stated	3 months	Retrospective case review	Not declared	0	0	Yes	Varying follow-up lengths
Rosa (2013)41	MIRCT confirmed arthroscopically	Not stated	10	10	69	Bursectomy	Not stated	Not stated	Prospective case series	Not declared	Not stated	Not stated	Yes	All patients satisfied
Senekovic (2017)50	MICRT on CT or MRI >5 cm and confirmed irreparable arthroscopically; failure of conservative management	GHJ OA; instability; active shoulder infection; previous shoulder surgery; diabetes mellitus, immunosuppression, coagulopathy.	24	20	70.5 (54)	Bursectomy; LHB tenotomy; partial repair - but these excluded from analysis	2–20min	3 years	Prospective case series	Yes	1 (rTSA)	2 had synovitis with deterioration in function	Yes	37.5% dropout rate, 1 withdrew as unsatisfied; analysis performed of the 15 who completed 5y f/u with balloon still in situ; 4 with partial repairs excluded from analysis
Yallpragada (2018)36	MIRCT (defined as Patte stage 3; Goutallier 3 or 4; irreparability confirmed arthroscopically); muscle atrophy; no OA (<grade 3 Hamada); failed conservative management >6 months	GHJ OA - Hamada grade 3; loss of passive ROM	14	14	76.2 (70)	Bursectomy, LHB tenotomy	Not stated	8 months (10.9 months)	Prospective case series	Not declared	0	1 spacer migration	Yes
Yamak (2019)52	MIRCT with failed conservative management; age >65	Not stated	11	11	69.1 (65)	Not stated	Not stated	37 months (45.1 months)	Retrospective case review	Not declared	Not stated	Not stated	No	Only abstract published; AHI measured on plain film radiographs; no difference in pre-op to post-op QuickDASH or VAS

Overall, 14 studies concluded by recommending the subacromial balloon spacer or suggesting that it would be a viable option. Only four studies did not recommend using the spacer based on their results.^39,49,52,53 In total, four studies performed a double-arm designed study, 3 of these comparing partial cuff repair in conjunction with balloon spacer with balloon spacer alone (Table 2). A further article reported 2 cases of augmentation of a massive rotator cuff repair with the subacromial balloon spacer, with a post-op MRI confirming degradation of the balloon and a film of bursa-like tissue covering the repair.⁵⁴ Another study also concluded that patients with a partial repair and the balloon fared better than those with balloon only intervention.⁵⁰

Table 2.

Showing the 2-arm studies with comments on those excluded from analysis.

1st Author (Year)	Inclusion criteria	Exclusions	Patients meeting inclusion criteria	Patients analysed	Average age (youngest)	2 arms	Follow-up	Study design	CoI	Revisions	Complications	Recommend use	Comments
Holschen (2017)45	MIRCT; loss of shoulder function	Cuff tear arthropathy; >grade 2 Hamada	25	23	63.5	Partial cuff repair; LHB tenotomy + partial cuff repair + spacer	22.3 months (average)	Retrospective case-control	No	2 (both to rTSA)	No	Yes	Excluded the 2 patients needing revision from analysed data
Malahias (2019)42	MIRCT radiologically and confirmed as irreparable arthroscopically; age >50	Pseudoparalysis; GHJ OA; systemic medical condition; irreparable subscapularis tear; previous surgery same joint	31	31	65.2	Partial cuff repair; LHB tenotomy and spacer	6 months	Retrospective case review	Yes	2 (not clarified)	No	Not conclusive	Included majority of patients with Goutallier grade 2 fatty infiltration; included acute tears as well as chronic
Malahias (2020)48	MIRCT (defined as >2 tendons torn on MRI, with tear size >3 cm; clinically massive cuff tear); age >50	Irreparable subscapularis tear; previous surgery same joint; infection; other systemic medical comorbidity	32	32	67.7	Partial cuff repair and spacer; LHB tenotomy and spacer	12 months	Retrospective case-control	No	1 (required revision arthroscopy)	No	No	Included majority of patients with Fuch’s grade 1or 2 fatty infiltration; included acute tears as well as chronic; 16 had balloon
Maman (2017)59	MRI or CT proven cuff tear; failure of conservative management >4months	OA; instability; previous shoulder surgery; infection; uncontrolled diabetes mellitus, coagulopathy and immunosuppression	58	42	Not stated	Debridement only; LHB tenotomy and spacer	12 months	Prospective multicentre	Yes	0	No	Yes	10 patients excluded as fluoroscopic; a further 6 excluded for partial repair or not following protocol

In total 513 patients have been analysed as part of clinical studies investigating the outcome of the subacromial balloon spacer; however, there were 619 patients recruited and included in the studies. Only 83% of patients initially meeting the inclusion criteria were analysed, the majority of the rest were either lost to follow-up or withdrew. These patients (17%) are not spread evenly among the studies, but mainly confined to 8 of the 20 studies. These eight studies initially included a total patient cohort of 361, and analysed the outcomes of 258 – giving a loss to follow-up of 29%. All but one of these 8 studies conclude by recommending the device.

8. Complications

Many complications experienced by patients are stated in the discussion sections, but often have been excluded from the analysis. There were three stated spacer migrations and 28 revision procedures, including 23 revisions to reverse total shoulder arthroplasty. In another case report, a further spacer migration associated with severe pain is reported.⁵⁵ The symptoms, in this case, resolved as soon as the balloon spacer was removed. Further complications include new-onset pseudoparalysis in two patients following balloon insertion without addressing the associated subscapularis tear.⁵⁰

Although the whole biomechanical theory of the InSpace™ balloon is to depress the humeral head inferiorly, thereby recreating native glenohumeral joint kinematics, few studies have analysed the radiographic evidence of this. One of the studies that used plain film radiographs post-operatively concluded that there was no significant effect of the balloon spacer on the upper migration index (UMI) with a pre-op UMI of 1.15 and a post-op of 1.16⁵² (UMI < 1.25 is suggestive of severe proximal migration of the humeral head). In this study, only 46% of patients were satisfied with their outcome following balloon spacer interposition.

9. A decade on….where are we?

The InSpace™ balloon was brought to the clinical market in 2010, and now a decade later, there is still no definitive evidence of whether it has a role in the treatment of massive, irreparable rotator cuff tears. As suggested by some recent reviews, there is too much heterogeneity in study design, patient inclusion and concomitant surgery to draw any conclusions.56, 57, 58 Some studies include tears that seem to have potential for repair, with grade 1 or 2 fatty infiltration diagnosed pre-operatively,^43,44,49 and partial repair possible on table.^43,49,50 Almost all studies to date provide level III or IV evidence, and many end by suggesting a randomised protocol study to enable firm conclusions to be drawn. In Burnier’s review, a treatment algorithm suggests that the subacromial balloon spacer is probably an option in those low demand patients without cartilage degeneration, without pseudoparalysis and with failed non-operative treatment.⁵⁹

In addition, there has been some wariness about studies that have a clear stated conflict of interest. Studies by Senekovic,^36,51 Malahias,⁴³ Lobao,³⁰ Zada^22,23 and Maman⁶⁰ have all been funded by OrthoSpace or has an author who works as a consultant for the company and bias in their conclusions cannot be excluded. It should also be noted that these and other studies have excluded patients requiring revision or with complications from the final analysis of patient outcome and satisfaction. With the evidence available at present, it would seem that the surgical indication for balloon spacer insertion is the older, low-demand patient, with a full thickness rotator cuff tear, but intact subscapularis and intact infraspinatus, who has failed non-surgical treatment and who has neither glenohumeral arthritis nor pseudoparalysis.

For the reasons stated above, in the UK, the National Institute for Health and Care Excellence (NICE) has limited the use of the InSpace™ balloon to research purposes only. Currently, two prospective randomised controlled trials are being performed. One in the United States, which is a company (OrthoSpace) funded trial, is a single-blinded 2-arm study aiming to compare the use of the balloon spacer with partial rotator cuff repairs in massive, irreparable rotator cuff tears.⁶¹ The other is a UK-based, NIHR funded double-blinded 2-arm trial to compare the results of subacromial balloon spacer intervention to subacromial decompression alone.⁶² These two studies represent the only prospective, randomised trials, with a power study performed for a clear outcome. The UK-based study has a novel design intended to provide early outcomes following an adaptive design methodology.⁶³ With the results of this study eagerly awaited, there will soon be a high level of evidence that will allow more reliable conclusions to be drawn and direct our use of this orthopaedic biodegradable implant.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.