Arthroscopic Superior Capsule

Reconstruction Can Eliminate

Pseudoparalysis in Patients With
Irreparable Rotator Cuff Tears
Teruhisa Mihata,*yz§k MD, PhD, Thay Q. Lee,z§ PhD, Akihiko Hasegawa,y MD, PhD,
Takeshi Kawakami,y MD, PhD, Kunimoto Fukunishi,y MD, Yukitaka Fujisawa,y MD, PhD,
Yasuo Itami,y MD, Mutsumi Ohue,k MD, and Masashi Neo,y MD, PhD
Investigation performed at the Department of Orthopedic Surgery,
Osaka Medical College, Takatsuki, Japan

Background: Patients with pseudoparalysis and irreparable rotator cuff tears have very poor function. The authors developed
a superior capsule reconstruction (SCR) technique for irreparable rotator cuff tears that restores shoulder stability and muscle
balance, improving shoulder function and relieving pain.
Purpose: To evaluate whether arthroscopic SCR reversed preoperative pseudoparalysis in patients with irreparable rotator cuff
tears.
Study Design: Case series; Level of evidence, 4.
Methods: One hundred consecutive patients with irreparable rotator cuff tears underwent arthroscopic SCR with fascia lata auto-
grafts; 7 patients with deltoid weakness from cervical or axillary nerve palsy and 5 with severe presurgical shoulder stiffness were
excluded. The remaining 88 were allocated to 3 groups according to their preoperative active shoulder elevation: no pseudopar-
alysis (45 patients; mean age, 66.2 years; mean tear size, 3.5 cm), moderate pseudoparalysis (28 patients, 68.3 years, 3.5 cm),
and severe pseudoparalysis (15 patients, 62.3 years, 4.9 cm). Clinical outcome, active shoulder range of motion, acromiohumeral
distance, and healing rate were compared between patients with and without pseudoparalysis, as well as before surgery and at
final follow-up (35-110 months).
Results: American Shoulder and Elbow Surgeons score, active elevation, active external rotation, and acromiohumeral distance
increased significantly after arthroscopic SCR among all patients. Graft healing rates did not differ among the groups (P = .73):
98% (44 of 45) for no pseudoparalysis, 96% (27 of 28) for moderate pseudoparalysis, and 87% (13 of 15) for severe pseudopar-
alysis. Pseudoparalysis was reversed in 96% (27 of 28) of patients with preoperative moderate pseudoparalysis and 93% (14 of
15) with preoperative severe pseudoparalysis. Both patients with residual pseudoparalysis postoperatively (1 of 28 with preop-
erative moderate pseudoparalysis, 1 of 15 with preoperative severe pseudoparalysis) had graft tears.
Conclusion: Arthroscopic SCR restored superior glenohumeral stability and improved shoulder function among patients with or
without pseudoparalysis who had previously irreparable rotator cuff tears. In the absence of postoperative graft tear, arthroscopic
SCR reversed preoperative pseudoparalysis. Graft healing rates after arthroscopic SCR did not differ between patients with and
without pseudoparalysis.
Keywords: irreparable; pseudoparalysis; reconstruction; rotator cuff; superior capsule; tear

Rotator cuff tear is a common shoulder injury and causes This situation is similar to that in nerve injuries such as cer-
pain and dysfunction in the shoulder joint. In some patients vical palsy or axillary nerve palsy and is therefore referred
with large to massive rotator cuff tears, active shoulder ele- to as pseudoparalysis.4,6,9,34,40 Even if patients with rotator
vation is reduced to \90° mainly owing to a loss of the ‘‘force cuff tear have pseudoparalysis of the shoulder joint, com-
couple’’ that stabilizes the humeral head during elevation. plete repair of the torn tendon can restore shoulder function
and remove the pseudoparalysis (recovery rate, 95%).9,10
Some large to massive rotator cuff tears are irreparable
and severely degenerated and atrophied, making it hard to
The American Journal of Sports Medicine
expect functional recovery even after partial repair of the
1–10
DOI: 10.1177/0363546518786489 torn tendons. Reverse shoulder arthroplasty is one of
Ó 2018 The Author(s) the alternative treatments to improve active shoulder

1
2 Mihata et al The American Journal of Sports Medicine

Figure 1. Arthroscopic findings before and after superior capsule reconstruction: posterior view of the (A) subacromial space
before surgery, (B) subacromial space just after surgery, and (C) glenohumeral joint just after surgery.

elevation (ie, above shoulder level) with pain relief, improv- according to their preoperative active shoulder elevation
ing daily activity3,36,41,42 and permitting a return to low- (Table 1):
intensity activity13; however, various postoperative compli-
No pseudoparalysis: 90° of active shoulder elevation
cations have been reported, including scapular notching,
(45 patients, all primary cases)
dislocation, infection, nerve palsy, glenoid loosening,
Moderate pseudoparalysis: no shoulder stiffness, \90°
humeral loosening, fracture of acromion and scapular
of active shoulder elevation; patients maintained
neck, and polyethylene wear.5,12,35,36 Farshad and Gerber12
.90° elevation once the shoulder was elevated pas-
concluded that reverse shoulder arthroplasty is associated
sively (28 patients: 26 primary cases and 2 revision
with a high rate of complications. Furthermore, in their
cases after a failed rotator cuff repair)
study, all complications requiring removal of the implant
Severe pseudoparalysis: no shoulder stiffness, \90° of
left the patient with very poor function. Reverse shoulder
active shoulder elevation; patients had a positive
arthroplasty for the treatment of primary glenohumeral
drop-arm sign (15 patients: 13 primary cases and 2
osteoarthritis with intact rotator cuff can result in excellent
revision cases after a failed rotator cuff repair)
clinical outcomes.24,33
Superior capsule reconstruction (SCR) was more recently While pain and some muscle weakness cause limited
developed for the treatment of irreparable rotator cuff elevation in moderate pseudoparalysis, severe muscle
tears.25,27-31 Restoration of superior stability by SCR is weakness is the main cause of limited elevation in severe
reported to improve functional outcomes.19,27 We investigated pseudoparalysis. Patients without pseudoparalysis were
whether arthroscopic SCR reversed preoperative pseudoparal- the control group of this study. There was no significant
ysis among patients with irreparable rotator cuff tears. Our difference in age among the 3 groups. Tear size in the
hypothesis was that SCR would restore shoulder function severe pseudoparalysis group was significantly greater
for patients with or without preoperative pseudoparalysis. than that in the other 2 groups (P \ .05) (Table 1). Each
patient signed an informed consent form approved by the
Institutional Review Board of our university (Osaka Medi-
METHODS cal College, No. 1854). The current study included updated
data from the previous study.27 Irreparable rotator cuff tear
A series of 100 consecutive patients with irreparable rota- was evaluated during shoulder arthroscopy. When the torn
tor cuff tears that had failed nonoperative treatment (for at tendon could not reach the original footprint, the rotator
least 3 months) underwent arthroscopic SCR with fascia cuff tear was defined as an irreparable tear.
lata autografts (Figure 1) by a single surgeon between The stage of osteoarthritis before surgery is classified by
2007 and 2014. Seven patients with deltoid weakness the Hamada grade.17 In this system, grade 1 is associated
attributed to cervical or axillary nerve palsy and 5 patients with minimal radiographic changes; grade 2, narrowing of
with severe shoulder stiffness (passive shoulder elevation the subacromial space to \6 mm; grade 3, erosion and so-
\90°) before surgery were excluded from the study popula- called acetabulization of the acromion caused by superior
tion. The remaining 88 patients were allocated to 3 groups migration of the humeral head; grade 4, glenohumeral

*Address correspondence to Teruhisa Mihata, MD, PhD, Department of Orthopedic Surgery, Osaka Medical College, 2-7 Daigaku-machi, Takatsuki,
Osaka 569-8686, Japan (emails: [email protected], [email protected]).
y
Department of Orthopedic Surgery, Osaka Medical College, Takatsuki, Japan.
z
Orthopaedic Biomechanics Laboratory, VA Healthcare System, Long Beach, California, USA.
§
Department of Orthopaedic Surgery, University of California, Irvine, California, USA.
k
Katsuragi Hospital, Kishiwada, Japan.
Presented at the interim meeting of the AOSSM, San Diego, California, USA, March 2017.
The authors declared that they have no conflicts of interest in the authorship and publication of this contribution.
AJSM Vol. XX, No. X, XXXX ASCR Eliminates Pseudoparalysis for RCT 3

TABLE 1
Patient Age and Severity of Rotator Cuff Teara

Pseudoparalysis, Mean (Range) or n

No (n = 45) Moderate (n = 28) Severe (n = 15)

Age, y 66.2 (43-78) 68.3 (45-82) 62.3 (45-80)

Tear size in anterior-posterior direction, cm 3.5 (2-7) 3.5 (2-6) 4.9b (3-7 )
Acromiohumeral distance, mm
Before surgery 4.4 (0.6-8.7) 4.7 (1.2-9.0) 4.1 (1.7-6.8)
At the final follow-up after SCR 9.7 (2.3-13.8) 9.8 (3.5-15.9) 8.1 (2.4-12.5)
Torn tendons
2: supraspinatus and infraspinatus 25 19 7
3: supraspinatus, infraspinatus, subscapularis 18 8 8
3: supraspinatus, infraspinatus, teres minor 2 0 0
4 0 1 0
Hamada grade
1 7 7 3
2 26 17 10
3 11 3 2
4a 0 1 0
4b 1 0 0
5 0 0 0
Goutallier stage
0/1 0 0 0
2 3 1 1
3 19 11 4
4 23 16 10

a
SCR, superior capsule reconstruction.
b
Tear size in the severe pseudoparalysis group was significantly greater than that in the no pseudoparalysis group and the moderate pseu-
doparalysis group (P \ .05).

arthritis; and grade 5, the presence of humeral head 41), and grade 5 was divided into 2 grades (5– and 5).21
osteonecrosis. We recorded the muscle strength of shoulder abduction
We also evaluated fatty degeneration of the supraspina- with the thumb up, which is known as the full can posi-
tus muscle with preoperative magnetic resonance imaging tion.22,23 External rotation strength was measured with
(MRI) using the grading system of Goutallier et al.16 The the arm at the side.8 We assessed strength in lifting the
grading classifies fatty degeneration into 5 stages: stage hand off the back to assess the internal rotation strength.14
0, completely normal muscle without any fatty streak; The hornblower sign38 and external and internal rotation
stage 1, muscle with some fatty streaks; stage 2, increased lag signs18 were assessed before surgery and at final
fatty infiltration but still more muscle than fat; stage 3, as follow-up after surgery. All patients were assessed preop-
much fat as muscle; and stage 4, more fat than muscle. eratively with the scoring systems of the shoulder index
of the American Shoulder and Elbow Surgeons (ASES; a
100-point scoring system) and the Japanese Orthopaedic
Patient Assessment Association (JOA; a 100-point scoring system) and were
reassessed at the time of the final follow-up. The postoper-
The patients provided a standard history. Shoulder eleva- ative complication rate was also recorded. MRI was per-
tion, external rotation at side, and internal rotation were formed with a 1.5-T closed-type scanner (MRT-2000/V2;
measured actively with a goniometer by the primary sur- Toshiba) before surgery and at final follow-up after sur-
geon before surgery and at the final follow-up. Internal gery. Oblique coronal, oblique sagittal, and axial T2-
rotation was measured as the highest vertebral body that weighted MRI scans were acquired for structural and qual-
the patient was able to reach with the thumb of the itative assessment of the rotator cuff and repair integrity
affected arm. We evaluated muscle strength by manual after surgery (Figure 2). The mean time to final follow-up
muscle testing (MMT) on a scale of 0 to 5: 5, normal was 60 months (range, 35-110 months).
amount of resistance to applied force; 4, resistance between
5 and 3; 3, ability to move the segment (the arm) through
its range of motion against gravity; 2, ability to move the Surgical Technique
segment through its range of motion but not against grav-
ity; 1, presence of contraction in the muscle without joint Preparation. All procedures were performed with the
motion; 0, no muscle contraction.8 Grades 3 and 4 were fur- patient under general anesthesia in the lateral decubitus
ther divided into 3 grades (3–, 3, and 31 and 4–, 4, and position. Normal arthroscopic pump pressure was set
4 Mihata et al The American Journal of Sports Medicine

Measurement of Defect Size. The defect size was mea-

sured in 2 directions: mediolateral (from the superior edge
of the glenoid to the lateral edge of the greater tuberosity:
no pseudoparalysis, mean = 3.9 cm; moderate, 3.8 cm;
severe, 4.1 cm) and anteroposterior (from the anterior
edge to the posterior edge of the torn tendon: no pseudopa-
ralysis, mean = 3.5 cm; moderate, 3.5 cm; severe, 4.9 cm).
Length in the anteroposterior direction was measured with-
out partial repair of the infraspinatus tendon.
Choosing the Graft Size and Harvesting the Fascia
Lata. Appropriate graft size was the most important point
in this surgery. If the graft was torn after surgery, the clin-
ical results were likely to be poor27: partial graft tear might
be acceptable, but complete graft tear gave poor results.
We found that the optimal graft length in the anteroposte-
rior direction was exactly the same as the length of the
defect without partial repair of the torn infraspinatus ten-
don. The graft length in the mediolateral direction (no
pseudoparalysis, mean = 5.4 cm; moderate, 5.3 cm; severe,
5.6 cm) was 15 mm longer than the distance from the supe-
rior edge of the glenoid to the lateral edge of the greater
tuberosity to give a 10-mm footprint on the superior glen-
Figure 2. Magnetic resonance image findings before and oid and a leeway of 5 mm to regulate graft size.
after arthroscopic superior capsule reconstruction. (A) Coro- A vertical skin incision was made over the lateral thigh,
nal view before surgery. The torn supraspinatus tendon is beginning 1 to 2 cm distal to the greater trochanter of the
severely retracted, and the supraspinatus muscle is severely femur, with care to include the posterior thicker tissue (Fig-
atrophied and infiltrated with fat. (B) Sagittal view before sur- ure 3, A-C). The average thickness of a single layer of autol-
gery. The supraspinatus, infraspinatus, and teres minor are ogous fascia lata was 2 to 4 mm. Therefore, a graft thickness
torn. (C) Coronal view 2 years after surgery. The graft has of 6 to 8 mm28 was achieved by folding the fascia lata 2 or 3
healed. White arrows represent the healed graft. (D) Sagittal times. Also, the fascia lata includes an intermuscular septum
view 2 years after surgery. The graft has connected with sub- that consists of the tissues of 2 tendons and connects the fas-
scapularis anteriorly and teres minor posteriorly, suggesting cia lata to the femur (Figure 3C). To make a thicker graft,
that force coupling in the anteroposterior direction has this intermuscular septum should be included in it (Figure
been restored. White arrows represent the healed graft. 3, D and E). The fascia lata was mostly thinner at its anterior
aspect than posteriorly; therefore, to make a flat graft of even
thickness, the intermuscular septum was usually sutured to
between 30 and 50 mm Hg. Shoulder range of motion and the anterior surface of the fascia lata after being completely
laxity were examined under general anesthesia. Three por- detached from it (Figure 3F). All fatty tissue should be
tals were typically required for arthroscopic SCR. A poste- removed from the graft. Finally, the layers of fascia lata
rior portal was established for initial assessment of the were united very closely with nonabsorbable sutures to pre-
glenohumeral joint. An anterior portal through the rotator vent delamination after surgery (Figure 3G).
interval was established as the working portal for the Graft Attachment. All soft tissues were removed on the
treatment of intra-articular lesions (eg, labral tear and superior glenoid and greater tuberosity to expose cortical
biceps tear) or subluxation. The arthroscope was then bone. The graft was inserted into the subacromial space via
removed from the glenohumeral joint and redirected into the lateral portal. We usually use a 10-mL syringe as a can-
the subacromial space. A lateral portal was also estab- nula. The medial side of the fascia lata was then attached to
lished. Any pathological bursal tissue that impeded clear- the superior glenoid with 2 fully threaded titanium suture
ance of the space was removed. Arthroscopic subacromial anchors (diameter, 5 mm; Corkscrew II Suture Anchor,
decompression was performed to create a flat acromial Arthrex), each with 2 No. 2 FiberWire nonabsorbable sutures
undersurface. Bony spurs in the inferior part of the acro- (Arthrex), which were inserted into the superior glenoid at
mioclavicular joint and at the distal end of the clavicle the 10- to 11-o’clock and 11- to 12-o’clock positions on the gle-
were removed.29 The superior glenoid and rotator cuff foot- noid of the right shoulder (or the 1- to 2-o’clock and 12- to 1-
print of the greater tuberosity were debrided to expose cor- o’clock positions of the left shoulder). FiberWires from the
tical bone. If the subscapularis tendon tear was reparable, superior glenoid were placed through the fascia lata in a mat-
it was completely repaired with fully threaded titanium tress fashion outside the body before the graft was inserted.
suture anchors (diameter, 5 mm; Corkscrew II Suture The lateral side of the fascia lata was attached to the
Anchor, Arthrex). The size of the superior capsular defect rotator cuff footprint on the greater tuberosity with the
was evaluated with a measuring probe in the anteroposte- compression double-row technique, which is a combination
rior and mediolateral directions at 30° to 45° of shoulder of the conventional double-row technique and the suture
abduction. bridge,26,32 or the SpeedBridge technique with FiberTape
AJSM Vol. XX, No. X, XXXX ASCR Eliminates Pseudoparalysis for RCT 5

Figure 3. Harvesting the fascia lata and making the graft. (A, B) Location of fascia lata harvest in the proximal thigh (A, black
dotted line; B, white dotted line). (C) The fascia lata is flipped over after incision along 3 sides of the dotted rectangle (from
‘‘ant’’ to ‘‘prox’’ to ‘‘dist’’). The fascia lata includes an intermuscular septum (asterisk) that consists of the tissues of 2 tendons
and connects the fascia lata and the femur. To make the graft thicker, this intermuscular septum should be included. (D) Reverse
side of the harvested fascia lata. (E) The intermuscular septum is completely detached from the fascia lata. (F) The fascia lata
tends to be thinner anteriorly. Therefore, to make a flat graft of even thickness, the intermuscular septum is usually sutured to
the anterior surface of the fascia lata, and the fascia lata is then folded. (G) Finally, the layers of fascia lata are united with non-
absorbable sutures. The upper and lower pictures show the lateral and inferior views of the graft, respectively. ant, anterior; dist,
distal; post, posterior; prox, proximal.

(Arthrex), at 30° to 45° of shoulder abduction. Corkscrew II Postoperative Protocol

suture anchors or SwiveLocks (Arthrex) were placed medi-
ally at the edge of the articular cartilage and laterally 5 to We recommended the use of an abduction sling (Block
10 mm inferior to the highest point of the greater tuberos- Shoulder Abduction Sling; Nagano Prosthetics & Orthotics
ity to minimize the possibility of the anchors pulling out. Co Ltd) for 4 weeks after reconstruction. After the immobi-
The sutures were placed through the fascia lata with either lization period, passive and active-assisted exercises were
a suture shuttle (Suture Lasso; Arthrex) or a suture-pass- initiated to promote scaption. Eight weeks after the sur-
ing device (Scorpion Suture Passer; Arthrex). Finally, side- gery, patients began to perform exercises to strengthen
to-side sutures with No. 2 FiberWire nonabsorbable the rotator cuff and the scapula stabilizers. Full activities
sutures (Arthrex) were added between the graft and the were allowed at 6 months if patients had sufficient range
infraspinatus tendon or teres minor tendon and between of motion and muscle strength. Physical therapists assis-
the graft and the residual anterior supraspinatus tendon ted all patients.
or subscapularis tendon to improve force coupling in the
shoulder joint. Careful attention should be paid to ensure
that the side-to-side suture at the anterior side was not Statistical Analysis
overtightened, to avoid shoulder contracture after surgery.
When the graft was attached securely at the medial, lat- A 1-way analysis of variance followed by a Tukey post hoc
eral, and posterior sides, an anterior suture might be test was performed to compare age, tear size, acromiohum-
unnecessary. eral distance, ASES score, JOA score, and active shoulder
6 Mihata et al The American Journal of Sports Medicine

TABLE 2 46.6 6 9.3 points in the moderate pseudoparalysis group,

Concomitant Injuries and Surgery and 20.3 6 13.6 and 40.6 6 10.8 points in the severe pseudo-
paralysis group, respectively. The moderate and severe pseu-
Pseudoparalysis, n doparalysis groups had significantly lower preoperative
No Moderate Severe ASES and JOA scores than the no pseudoparalysis group
(n = 45) (n = 28) (n = 15) (P \ .001). Clinical outcome scores after arthroscopic SCR
were significantly improved at final follow-up (P \ .001)
Subscapularis among all 3 groups, and there were no significant differences
Intact or partial tear 34 24 9 in postoperative scores (Table 4): ASES (no pseudoparalysis,
(no treatment) 96.5 6 5.9 points; moderate, 92.2 6 13.9; severe, 91.8 6 11.8;
Repair for complete tear 11 4 6
P = .32-.99) and JOA (no pseudoparalysis, 95.2 6 5.5 points;
Biceps
moderate, 90.6 6 12.9; severe, 92.3 6 8.9; P = .19-.81).
Intact (no treatment) 12 10 5
Partial tear (no treatment) 18 9 2 Preoperative active shoulder elevation was 142.7° 6
Complete tear (no treatment) 11 6 7 23.4° in the no pseudoparalysis group, 54.3° 6 24.4° in
Tenodesis for dislocated biceps 4 3 0 moderate pseudoparalysis group, and 36.7° 6 19.1° in
Tenotomy for dislocated biceps 0 0 1 severe pseudoparalysis group. Active preoperative eleva-
Acromioplasty 45 28 15 tion was significantly less in the severe pseudoparalysis
group than in the no pseudoparalysis (P \ .001) and mod-
erate pseudoparalysis (P = .03) groups. Active preoperative
range of motion among the 3 groups (no, moderate, and elevation was significantly less in the moderate pseudopa-
severe pseudoparalysis). To calculate the mean MMT ralysis group than in the no pseudoparalysis group (P \
grade, we converted each grade to a scale of 0 to 10, where .001). Shoulder active elevation improved significantly
MMT 5 = 10, 5– = 9, 41 = 8, 4 = 7, 4– = 6, 31 = 5, 3 = 4, 3– = after arthroscopic SCR at final follow-up for all 3 groups
3, 2 = 2, 1 = 1, and 0 = 0. All data were compared with the t (no pseudoparalysis, 163.6° 6 15.4°; moderate, 146.8° 6
test (before surgery vs final follow-up). The number of torn 33.0°; severe, 150.0° 6 36.8°; P \ .001); there were no sig-
tendons, Hamada grade, Goutallier grade, the number of
nificant differences in this improvement among the 3
subscapularis repairs, biceps pathology/treatment, and groups (P = .12-.93) (Table 5). The severe pseudoparalysis
graft healing rates were compared among the 3 groups group had significantly less active shoulder external rota-
with a chi-square test. Data are shown as mean 6 SD. A
tion before surgery than the no pseudoparalysis group (P
significant difference was defined as P \ .05. = .02). Active shoulder external and internal rotation
improved significantly after arthroscopic SCR in all 3
groups (P \ .001), and there were no significant differences
RESULTS in the improvement among the 3 groups (P = .21-.99)
(Table 5). Shoulder muscle strength improved significantly
Tear size in the anteroposterior direction in the severe pseu- after arthroscopic SCR in all 3 groups (P \ .001) (Table 6).
doparalysis group (mean, 4.9 cm; range, 3-7 cm) was signifi- All patients in the moderate and severe pseudoparalysis
cantly greater than in the no pseudoparalysis group (mean, groups had a positive hornblower sign before surgery.
3.5 cm; range, 2-7 cm; P = .003) and moderate pseudoparaly- External rotation lag sign was positive for 3 of 45 patients
sis group (mean, 3.5 cm; range, 2-6 cm; P = .004) (Table 1). with no pseudoparalysis, 7 of 28 with moderate pseudopa-
There was no significant difference in age (P = .11), number ralysis, and 5 of 15 with severe pseudoparalysis before sur-
of torn tendons (P = .32), acromiohumeral distance (P = .74), gery. At the final follow-up after arthroscopic SCR, the
Hamada grade (P = .64), Goutallier grade of supraspinatus hornblower and external rotation lag signs became nega-
muscle (P = .82), number of subscapularis repairs (P = .17), tive for all patients if the graft was healed. Two patients
and biceps injury and treatment (P = .16) among the 3 groups with graft tear (1 with preoperative moderate pseudoparal-
(Tables 1 and 2). Acromioplasty was performed in all ysis and 1 with preoperative severe pseudoparalysis) had
patients. Acromiohumeral distance significantly increased positive external rotation lag and hornblower signs at the
after SCR in all group (P \ .001) (Table 1). final follow-up. For all patients with positive internal rota-
Pseudoparalysis was reversed in 96.4% of patients with tion lag sign (0 of 45 with no pseudoparalysis, 3 of 28 with
moderate preoperative pseudoparalysis (27 of 28; increased moderate pseudoparalysis, and 3 of 15 with severe pseudo-
active elevation, 60°-150°) and in 93.3% with severe preop- paralysis), the lag sign became negative at the final follow-
erative pseudoparalysis (14 of 15; increased active elevation, up after arthroscopic SCR.
70°-170°). Both patients with residual pseudoparalysis post- Some postoperative complications were experienced.
operatively had graft tears (Table 3). The graft healing rate Three patients had suture anchor pullout. Three patients
without retear was 98% (44 of 45) among patients with no had severe shoulder stiffness, which was treated with arthro-
preoperative pseudoparalysis, 96% (27 of 28) for those scopic release of rotator interval and inferior capsule. Two
with moderate pseudoparalysis, and 87% (13 of 15) in the patients had postoperative infection (Propionibacterium
severe pseudoparalysis group. Graft healing rates did not acnes). Both patients were treated by arthroscopic debride-
differ among the 3 groups (P = .73). ment and removal of some sutures and anchors, without
Preoperative scores were 43.6 6 17.2 (ASES) and 61.2 6 removal of the graft because we found that the graft had
8.7 (JOA) in the no pseudoparalysis group, 29.2 6 18.3 and already started to heal to the bone. One patient had
AJSM Vol. XX, No. X, XXXX ASCR Eliminates Pseudoparalysis for RCT 7

TABLE 3
Functional Outcomes of Patients With Postoperative Graft Teara

Pseudoparalysis Grade ASES Score JOA Score Active Elevation

Patient PRE POST PRE POST PRE POST PRE POST

1 No No 28.3 91.7 54.5 86 140 110

2 Moderate Severe 60 38.3 43 44 20 20
3 Severe Severe 20 65 54.5 72.5 40 50
4 Severe No 63.3 86.7 66.5 82 80 170

a
ASES, American Shoulder and Elbow Surgeons; JOA, Japanese Orthopaedic Association; POST, postoperative; PRE, preoperative.

TABLE 4
Pre- and Postoperative Shoulder Scoresa

ASES Score JOA Score

Pseudoparalysis Preoperative Postoperative Preoperative Postoperative

No
Total 43.6 (18.3-78.3) 96.5 (70-100) 61.2 (33-76) 95.2 (76.5-100)
Healed graft 44.0 (18.3-78.3) 96.6 (70-100) 61.4 (33-76) 95.4 (76.5-100)
Moderate
Total 29.2b (3.3-68.3) 92.2 (38.3-100) 46.6b (30-74) 90.6 (44-100)
Healed graft 28.0b (3.3-68.3) 94.2 (61.7-100) 46.8b (30-74) 92.3 (59.5-100)
Severe
Total 20.3b (6.7-63.3) 91.8 (65-100) 40.6b (26.5-66.5) 92.3 (72.5-100)
Healed graft 17.1b (6.7-28.3) 94.2 (70-100) 37.5b (26.5-51.5) 94.6 (79.5-100)

a
Data are expressed as mean (range). Postoperative data were recorded at the final follow-up. All pre- vs postoperative values per score
were significant at P \ .001.
b
Score was significantly lower than that in the no pseudoparalysis group (P \ .05).

discomfort in the gluteal muscle, although she could play developed and reported to restore shoulder function for
badminton at the competitive level without any pain. No patients with irreparable rotator cuff tear.2,19,27-31 We found
patients had pain with the harvest site at the final follow-up. here that 96.4% (27 of 28) of patients with moderate preoper-
ative pseudoparalysis and 93.3% (14 of 15) with severe preop-
erative pseudoparalysis had no pseudoparalysis (.90° of
active elevation) at final follow-up after SCR. Also, both
DISCUSSION patients with residual pseudoparalysis had graft tears post-
operatively. Furthermore, complication after SCR (anchor
Whereas rotator cuff repair reverses preoperative pseudopar- pullout, infection, and shoulder stiffness) was not so severe,
alysis in most patients with reparable rotator cuff tears,9,34 and the rate of complication was relatively low. Thus, SCR
pseudoparalysis attributed to irreparable rotator cuff tear is completely eliminated pseudoparalysis in patients with irrep-
considered a difficult situation in which to restore shoulder arable rotator cuff tears, with a low rate of complications if
function, even surgically. Reverse total shoulder arthroplasty the graft healed very well.
(TSA) is an alternative treatment for irreparable rotator cuff Our previous clinical study showed that postoperative
tears and was reported to improve shoulder function, even clinical outcome scores were significantly better among
though it is not completely restorative6,40,41; however, rela- healed patients (ASES, 96.0; JOA, 94.9) than unhealed
tively high rates of complications5,35,37 and severe limitation patients who had graft tears or retears of the repaired rota-
of shoulder external or internal rotation, or both,6 were tor cuff tendon (ASES, 77.1 [P \ .0001]; JOA, 81.1 [P \
reported, even after reverse TSA. Otto et al35 documented .001]).27 Similarly, our results showed that preoperative
a total complication rate of 22.4% after reverse shoulder pseudoparalysis was not eliminated after SCR in 2 of 4
arthroplasty among patients aged \55 years. Wall et al39 cases with graft tears (Table 3). These results suggest
indicated that the risk of complication associated with revi- that graft healing is the key to improved shoulder function
sion surgery (36.7%, 18 of 49) was significantly higher than after SCR, especially for patients with pseudoparalysis.
that with primary surgery (13.3%, 20 of 150). When patients The common symptoms and signs of irreparable rotator
have severe glenohumeral osteoarthritis, reverse TSA is cuff tear—including pain from subacromial impingement,
thought to be a good surgical option.24,33 SCR was recently muscle weakness, and limitation of active shoulder range
8 Mihata et al The American Journal of Sports Medicine

TABLE 5
Pre- and Postoperative Active Shoulder Range of Motiona

Active Elevation Active External Rotation Active Internal Rotation

Pseudoparalysis Preoperative Postoperative Preoperative Postoperative Preoperative Postoperative

None
Total 142.7 (90 to 170) 163.6 (110 to 180) 33.4 (–60 to 80) 45.4 (0 to 70) L4 (S to T12) L1 (S to T7)
Healed graft 142.8 (90 to 170) 164.8 (120 to 180) 33.3 (–60 to 80) 46.0 (0 to 70) L4 (S to T12) L1 (S to T7)
Moderate
Total 54.3b (20 to 80) 146.8 (20 to 180) 22.9 (0 to 50) 37.9 (10 to 60) L5 (S to T7) L2 (S to T7)
Healed graft 55.6b (20 to 80) 151.5 (90 to 180) 23.3 (0 to 50) 38.9 (10 to 60) L5 (S to T7) L2 (S to T7)
Severe
Total 36.7b,c (10 to 80) 150.0 (50 to 180) 16.7b (–20 to 60) 44.0 (20 to 90) L5 (S to T12) L1 (S to T10)
Healed graft 33.1b,c (10 to 60) 156.2 (100 to 180) 14.6b (–20 to 60) 46.2 (20 to 90) L5 (S to T12) L1 (L3 to T12)

a
Data are expressed in degrees, mean (range). Postoperative data were recorded at the final follow-up. All pre- vs postoperative values per
range of motion category were significant at P \ .001.
b
Range of motion was significantly less than that in the no pseudoparalysis group (P \ .05).
c
Range of motion was significantly less than that in the moderate pseudoparalysis group (P \ .05).

TABLE 6
Pre- and Postoperative Muscle Strength Gradesa
Abduction External Rotation Internal Rotation

Preoperative Postoperative Preoperative Postoperative Preoperative Postoperative

Pseudoparalysis MMT 10 Scale MMT 10 Scale MMT 10 Scale MMT 10 Scale MMT 10 Scale MMT 10 Scale

None
Total 4 (3 to 5–) 7.0 (4 to 9) 5– (4– to 5) 9.3 (6 to 10) 4 (1 to 5) 7.0 (1 to 10) 5– (3 to 5) 9.0 (4 to 10) 41 (3– to 5) 8.4 (3 to 10) 5 (4 to 5) 9.6 (7 to 10)
Healed graft 4 (3 to 5–) 7.0 (4 to 9) 5– (4 to 5) 9.3 (7 to 10) 4 (1 to 5) 7.0 (1 to 10) 5– (3 to 5) 9.1 (4 to 10) 41 (3– to 5) 8.4 (3 to 10) 5 (4 to 5) 9.6 (7 to 10)
Moderate
Total 3– (2 to 3–) 2.8 (2 to 3) 41 (1 to 5) 8.4 (1 to 10) 31 (1 to 4) 4.5 (1 to 7) 5– (3 to 5) 8.6 (4 to 10) 4 (1 to 5) 7.2 (1 to 10) 5– (3 to 5) 9.2 (4 to 10)
Healed graft 3– (2 to 3–) 2.8 (2 to 3) 5– (3 to 5) 8.7 (4 to 10) 31 (1 to 4) 4.6 (1 to 7) 5– (3 to 5) 8.8 (4 to 10) 4 (1 to 5) 7.2 (1 to 10) 5– (3 to 5) 9.3 (4 to 10)
Severe
Total 3–b,c (1 to 3–) 2.5 (1 to 3) 41 (2 to 5) 7.9 (2 to 10) 3–b,c (1 to 4) 2.9 (1 to 7) 4b (2 to 5) 7.4 (2 to 10) 4b (1 to 5) 6.7 (1 to 10) 5– (4 to 5) 9.3 (7 to 10)
Healed graft 3–b,c (1 to 3–) 2.6 (1 to 3) 5– (3 to 5) 8.5 (3 to 10) 3–b,c (1 to 4) 3.2 (1 to 7) 41 (3– to 5) 8.2 (3 to 10) 41 (1 to 5) 7.5 (1 to 10) 5 (4 to 5) 9.5 (7 to 10)

a
Data are expressed in degrees, mean (range). Postoperative data were recorded at the final follow-up. All strength grades were significantly different
between preoperative and postoperative (P \ .05). MMT, manual muscle testing.
b
Muscle strength was significantly less than that in the no pseudoparalysis group (P \ .05).
c
Muscle strength was significantly less than that in the moderate pseudoparalysis group (P \ .05).

of motion—result mainly from loss of superior stability of necessary to improve external and internal rotation lag
the glenohumeral joint.11,15 This loss of superior stability signs. When the graft is healed, the capsular continuity
is due to defects in the superior capsule and the supraspina- in the anterior-posterior direction is recovered, and the
tus and infraspinatus tendons.1,20 Reduced glenohumeral residual external rotators (infraspinatus and teres minor)
stability may cause more severe symptoms,7 such as pseu- and subscapularis work well because (1) the side-to-side
doparalysis. Therefore, a stiffer graft is needed to improve suture is utilized between the graft and the residual rota-
glenohumeral stability among patients with pseudoparaly- tor cuff tendons and (2) subscapularis tendon repair is per-
sis.28 Here, we found relatively high rates of graft healing formed. These biomechanical improvements make the
(95%, 84 of 88 patients) with MRI after arthroscopic SCR external and internal rotation lag signs recover. Even
using fascia lata autografts. Furthermore, graft healing though all patients with graft healing can hold an exter-
rates after arthroscopic SCR did not differ between patients nally or internally rotated position after SCR, the muscle
with and without pseudoparalysis. These results suggest strength of external and internal rotation was still weaker
that a fascia lata graft 6 to 8 mm thick in our current tech- than the other side in some cases.
nique has the appropriate stiffness for SCR, even for According to the current study, arthroscopic SCR is rec-
patients with pseudoparalysis. ommended for irreparable medium, large, and massive
In this study, all external and internal rotation lag signs tears of posterior-superior rotator cuff tendons with Ham-
became negative when the graft was healed. However, for ada grades 1 to 3 and Goutallier grades 3 and 4. Age is
2 patients with postoperative graft tear, external rotation not a limiting factor for arthroscopic SCR. A contraindica-
lag signs did not resolve. Therefore, graft healing is tion is deltoid weakness attributed to cervical or axillary
AJSM Vol. XX, No. X, XXXX ASCR Eliminates Pseudoparalysis for RCT 9

nerve palsy because postoperative functional improvement epidemiology and characteristics. Orthop Traumatol Surg Res.
mainly results from deltoid muscle strength with stabilized 2009;95:S13-S18.
5. Boileau P. Complications and revision of reverse total shoulder
glenohumeral joint by SCR. Also severe glenohumeral osteo-
arthroplasty. Orthop Traumatol Surg Res. 2016;102:S33-S43.
arthritis is not a good indication for arthroscopic SCR. 6. Boileau P, Chuinard C, Roussanne Y, Bicknell RT, Rochet N, Trojani
In this study, we could repair the subscapularis tendon C. Reverse shoulder arthroplasty combined with a modified latissi-
tear in all cases. Therefore, subscapularis tendon tear can- mus dorsi and teres major tendon transfer for shoulder pseudoparal-
not be a contraindication for SCR, and we can expect good ysis associated with dropping arm. Clin Orthop Relat Res. 2008;
clinical outcomes after SCR when the torn subscapularis 466:584-593.
tendon is repaired. Even in 4 tendon tears (1 patient in 7. Chung SW, Kim JY, Kim MH, Kim SH, Oh JH. Arthroscopic repair of
massive rotator cuff tears: outcome and analysis of factors associ-
the moderate pseudoparalysis group), we could obtain ated with healing failure or poor postoperative function. Am J Sports
good clinical outcome after SCR (active elevation and Med. 2013;41:1674-1683.
ASES score: 50° and 38.3 preoperatively, 140° and 88.3 8. Daniels L, Worthingham C. Muscle Testing. Philadelphia, PA: WB
points postoperatively). Regarding irreparable subscapula- Saunders; 1980.
ris tendon tear, we will conduct a study to examine if it can 9. Denard PJ, Koo SS, Murena L, Burkhart SS. Pseudoparalysis: the impor-
be an indication of SCR. tance of rotator cable integrity. Orthopedics. 2012;35:e1353-e1357.
10. Denard PJ, Ladermann A, Brady PC, et al. Pseudoparalysis from
Our study had some limitations. First, we did not com-
a massive rotator cuff tear is reliably reversed with an arthroscopic
pare the clinical outcome of SCR with any other treatment, rotator cuff repair in patients without preoperative glenohumeral
such as debridement alone, partial cuff repair, repair with arthritis. Am J Sports Med. 2015;43:2373-2378.
a patch, or reverse TSA. Second, we investigated clinical 11. Duralde XA, Bair B. Massive rotator cuff tears: the result of partial
outcomes only after fascia lata SCR. Different grafts, rotator cuff repair. J Shoulder Elbow Surg. 2005;14:121-127.
such as allografts or synthetic grafts, may not have given 12. Farshad M, Gerber C. Reverse total shoulder arthroplasty—from the
most to the least common complication. Int Orthop. 2010;34:1075-1082.
similar clinical results after SCR. Third, we presented
13. Fink Barnes LA, Grantham WJ, Meadows MC, Bigliani LU, Levine
only retrospective results after arthroscopic SCR. Fourth, WN, Ahmad CS. Sports activity after reverse total shoulder arthro-
all the procedures in this study were performed by a single plasty with minimum 2-year follow-up. Am J Orthop (Belle Mead
experienced shoulder surgeon, which could have biased the NJ). 2015;44:68-72.
results, as it may not be easy for less-experienced surgeons 14. Gerber C, Krushell RJ. Isolated rupture of the tendon of the subscap-
to obtain the same outcomes. Fifth, we did not include ularis muscle: clinical features in 16 cases. J Bone Joint Surg Br.
patients with deltoid weakness attributed to cervical or 1991;73:389-394.
15. Gerber C, Maquieira G, Espinosa N. Latissimus dorsi transfer for the
axillary nerve palsy or with severe shoulder stiffness (pas- treatment of irreparable rotator cuff tears. J Bone Joint Surg Am.
sive shoulder elevation \90°) to investigate the effect of 2006;88:113-120.
pseudoparalysis on postoperative outcomes. If we investi- 16. Goutallier D, Postel JM, Bernageau J, Lavau L, Voisin MC. Fatty mus-
gate the postoperative outcomes of all patients, including cle degeneration in cuff ruptures: pre- and postoperative evaluation
those with deltoid weakness or severe shoulder stiffness, by CT scan. Clin Orthop Relat Res. 1994;304:78-83.
postoperative ASES and JOA scores should be worse 17. Hamada K, Fukuda H, Mikasa M, Kobayashi Y. Roentgenographic
findings in massive rotator cuff tears: a long-term observation. Clin
than those of the current study.
Orthop Relat Res. 1990;254:92-96.
18. Hertel R, Ballmer FT, Lombert SM, Gerber C. Lag signs in the diag-
nosis of rotator cuff rupture. J Shoulder Elbow Surg. 1996;5:307-313.
CONCLUSION 19. Hirahara AM, Adams CR. Arthroscopic superior capsular reconstruc-
tion for treatment of massive irreparable rotator cuff tears. Arthrosc
Tech. 2015;4:e637-e641.
Arthroscopic SCR restored superior glenohumeral stability
20. Ishihara Y, Mihata T, Tamboli M, et al. Role of the superior shoulder
and improved shoulder function in patients with or with- capsule in passive stability of the glenohumeral joint. J Shoulder
out pseudoparalysis who had previously irreparable rota- Elbow Surg. 2014;23:642-648.
tor cuff tears. If the graft did not tear postoperatively, 21. Itoi E, Minagawa H, Yamamoto N, Seki N, Abe H. Are pain location
arthroscopic SCR reversed preoperative pseudoparalysis. and physical examinations useful in locating a tear site of the rotator
Graft healing rates after arthroscopic SCR did not differ cuff? Am J Sports Med. 2006;34:256-264.
between patients with and without pseudoparalysis. 22. Kelly BT, Kadrmas WR, Speer KP. The manual muscle examination
for rotator cuff strength: an electromyographic investigation. Am J
Sports Med. 1996;24:581-588.
23. McFarland EG. Strength testing. In: McFarland EG, ed. Examination
REFERENCES of the Shoulder. New York, NY: Thieme Medical Publishers Inc;
2005:88-125.
1. Adams CR, DeMartino AM, Rego G, Denard PJ, Burkhart SS. The 24. McFarland EG, Huri G, Hyun YS, Petersen SA, Srikumaran U.
rotator cuff and the superior capsule: why we need both. Arthros- Reverse total shoulder arthroplasty without bone-grafting for severe
copy. 2016;32:2628-2637. glenoid bone loss in patients with osteoarthritis and intact rotator
2. Adams CR, Denard PJ, Brady PC, Hartzler RU, Burkhart SS. The cuff. J Bone Joint Surg Am. 2016;98:1801-1807.
arthroscopic superior capsular reconstruction. Am J Orthop (Belle 25. Mihata T. Superior capsule reconstruction for irreparable rotator cuff
Mead NJ). 2016;45:320-324. tears. In: Tamai K, Itoi E, Takagishi K, eds. Advances in Shoulder Sur-
3. Bacle G, Nove-Josserand L, Garaud P, Walch G. Long-term out- gery. Tokyo, Japan: Springer; 2015:173-183.
comes of reverse total shoulder arthroplasty: a follow-up of a previous 26. Mihata T, Fukuhara T, Jun BJ, Watanabe C, Kinoshita M. Effect of
study. J Bone Joint Surg Am. 2017;99:454-461. shoulder abduction angle on biomechanical properties of the
4. Berhouet J, Collin P, Benkalfate T, et al; Société d’Orthopédie de repaired rotator cuff tendons with 3 types of double-row technique.
l’Ouest. Massive rotator cuff tears in patients younger than 65 years: Am J Sports Med. 2011;39:551-556.
10 Mihata et al The American Journal of Sports Medicine

27. Mihata T, Lee TQ, Watanabe C, et al. Clinical results of arthroscopic 34. Oh JH, Kim SH, Shin SH, et al. Outcome of rotator cuff repair in large-
superior capsule reconstruction for irreparable rotator cuff tears. to-massive tear with pseudoparalysis: a comparative study with pro-
Arthroscopy. 2013;29:459-470. pensity score matching. Am J Sports Med. 2011;39:1413-1420.
28. Mihata T, McGarry MH, Kahn T, Goldberg I, Neo M, Lee TQ. Biome- 35. Otto RJ, Clark RE, Frankle MA. Reverse shoulder arthroplasty in
chanical effect of thickness and tension of fascia lata graft on gleno- patients younger than 55 years: 2- to 12-year follow-up. J Shoulder
humeral stability for superior capsule reconstruction in irreparable Elbow Surg. 2017;26(5):792-797.
supraspinatus tears. Arthroscopy. 2016;32:418-426. 36. Rhee YG, Cho NS, Moon SC. Effects of humeral component retrover-
29. Mihata T, McGarry MH, Kahn T, Goldberg I, Neo M, Lee TQ. Biome- sion on functional outcomes in reverse total shoulder arthroplasty for
chanical effects of acromioplasty on superior capsule reconstruction cuff tear arthropathy. J Shoulder Elbow Surg. 2015;24:1574-1581.
for irreparable supraspinatus tendon tears. Am J Sports Med. 37. Riley C, Idoine J, Shishani Y, Gobezie R, Edwards B. Early outcomes
2016;44:191-197. following metal-on-metal reverse total shoulder arthroplasty in
30. Mihata T, McGarry MH, Kahn T, Goldberg I, Neo M, Lee TQ. Biome- patients younger than 50 years. Orthopedics. 2016;39:e957-e961.
chanical role of capsular continuity in superior capsule reconstruction 38. Walch G, Boulahia A, Calderone S, Robinson AH. The ‘‘dropping’’
for irreparable tears of the supraspinatus tendon. Am J Sports Med. and ‘‘hornblower’s’’ signs in evaluation of rotator-cuff tears. J Bone
2016;44:1423-1430. Joint Surg Br. 1998;80:624-628.
31. Mihata T, McGarry MH, Pirolo JM, Kinoshita M, Lee TQ. Superior 39. Wall B, Nove-Josserand L, O’Connor DP, Edwards TB, Walch G.
capsule reconstruction to restore superior stability in irreparable rota- Reverse total shoulder arthroplasty: a review of results according
tor cuff tears: a biomechanical cadaveric study. Am J Sports Med. to etiology. J Bone Joint Surg Am. 2007;89:1476-1485.
2012;40:2248-2255. 40. Werner BS, Boehm D, Gohlke F. Revision to reverse shoulder arthro-
32. Mihata T, Watanabe C, Fukinishi K, et al. Functional and structural plasty with retention of the humeral component. Acta Orthop.
outcomes of single-row vs double-row vs combined double-row 2013;84:473-478.
and suture-bridge repair for rotator cuff tears. Am J Sports Med. 41. Werner CM, Steinmann PA, Gilbart M, Gerber C. Treatment of painful
2011;39(10):2091-2098. pseudoparesis due to irreparable rotator cuff dysfunction with the
33. Mizuno N, Denard PJ, Raiss P, Walch G. Reverse total shoulder Delta III reverse-ball-and-socket total shoulder prosthesis. J Bone
arthroplasty for primary glenohumeral osteoarthritis in patients Joint Surg Am. 2005;87:1476-1486.
with a biconcave glenoid. J Bone Joint Surg Am. 2013;95:1297- 42. Wiater JM, Fabing MH. Shoulder arthroplasty: prosthetic options and
1304. indications. J Am Acad Orthop Surg. 2009;17:415-425.

For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav.