Keywords rotator cuff - arthroscopy - articular range of motion
Introduction
Rotator cuff tears affect 20% of the general population and up to 50% of patients
over 80 years of age.[1 ] Clinical improvement after surgery occurs in most patients,[2 ]
[3 ]
[4 ] but recurrence of tears takes place in 27% of the cases.[5 ]
The evaluation of predictive factors is important to define the patients at risk for
poor outcomes after rotator cuff repair. There are some studies that evaluate the
factors that increase the risk of healing failure[6 ]
[7 ]
[8 ]
[9 ]
[10 ]
[11 ]
[12 ] and unfavorable clinical outcomes.[13 ]
[14 ]
[15 ]
[16 ] The risk factors for worse clinical outcomes have been described as: older patients;[14 ]
[15 ] the female gender;[15 ] worse preoperative function; previous surgery and problems at work;[16 ] smoking;[17 ] degree of fatty degeneration; and dimension of the tear.[18 ]
Rotator cuff tears have distinct structural patterns, which are classically described
as crescent- (C), L- and U-shaped.[19 ] To date, few studies have evaluated the influence of the tear pattern on the postoperative
clinical outcomes[20 ]
[21 ] without evidence of difference between the groups. These studies included in their
series infraspinatus[20 ]
[21 ] and subscapularis tear,[20 ] which, although increase the external validity of the results, also increase the
confounding factors. The aim of the present study is to evaluate the influence of
the pattern of the supraspinatus rupture on the preoperative and postoperative functional
assessments.
Methods
Design
Retrospective cohort study comparing the preoperative and postoperative functional
assessments between two groups of patients according to supraspinatus rupture pattern:
C-shaped versus L- or U-shaped.
Location and Dates
We analyzed patients who underwent complete arthroscopic repair of the supraspinatus,
with procedures performed between November 2012 and November 2016, by one of the surgeons
of the Shoulder and Elbow Group of our institution.
Surgical procedure and rehabilitation
The surgeries were performed by arthroscopy, under general anesthesia and interscalene
block. The patients were positioned in the beach chair position or lateral decubitus,
according to the surgeon's preference. Bursectomy, acromioplasty and distal clavicle
resection were performed as needed. The rotator cuff was repaired after debridement
of the greater tubercle with a single-row technique using double-loaded anchors. The
long head tendon of the biceps was approached when it had subluxation or dislocation,
partial lesions greater than 50%, or in the presence of type 2, 3 and 4 slap lesions.
The procedure performed was tenotomy in patients aged 60 years or older, or tenodesis
in younger patients. Tenodesis, when indicated, was performed either with one of the
most anterior anchor, or with an anchor specifically for this purpose. Before the
rotator cuff repair was started, the lesion was measured with the aid of a millimeter
probe, and the pattern was evaluated according to its reducibility to the bone bed.
The number of anchors and the need for tendon-tendon stitches was decided during the
surgical procedure.
After the surgery, the patients remained immobilized for 6 weeks with a Velpeau sling.
Finger, wrist and elbow movements were encouraged from day one. Passive shoulder range
of motion was started at four weeks, and active movements, after sling removal. Strengthening
was performed after three months, and complete release for work and sports activities
at six months.
Magnetic resonance imaging
All patients underwent magnetic resonance imaging (MRI) prior to the surgical procedure
in a 1.5-T equipment (HDxt, GE Medical Systems, Milwaukee, WI, US) and shoulder coil
without intra-articular or intravenous contrast.
Participants (eligibility criteria)
We included patients who underwent arthroscopic surgery to treat isolated supraspinatus
tears, with complete repair. The patients also needed to have been submitted to a
preoperative MRI, a standardized collection of intraoperative findings, and have responded
the pre- and postoperative (6, 12 and 24 months) questionnaires. Patients with associated
or isolated rupture of the subscapular or infraspinatus tendons, those who underwent
open surgery, or those who had only partial repair achieved, were not included.
Groups
The patients were divided into two groups: C-shaped versus L- or U-shaped, according
to Burkhart e Lo.[19 ] The categorization was performed based on the arthroscopic inspection. Type-C tears
are those with medial to lateral mobility. L- or U-shaped tears show mobility primarily
in the anteroposterior direction, and may require tendon-tendon stitches ([Figure 1 ]).
Fig. 1 Rotator cuff tear patterns (A ) crescent-shaped lesion; (B ) L-shaped lesion; (C ) U-shaped lesion.
Outcomes
The clinical evaluation was made using The American Shoulder and Elbow Surgeons Standardized
Shoulder Assessment (ASES)[22 ]
[23 ] and the Modified-University of California at Los Angeles Shoulder Rating Scale (UCLA).[24 ]
[25 ]
Other variables analyzed
All variables related to the lesion, except for fatty degeneration, were analyzed
during the arthroscopy. Fatty degeneration was measured in the oblique sagittal section
T1 of the 1.5-T MRI.
Statistical analysis
We submitted the continuous variables to the evaluation of normality through the Kolmogorov-Smirnov
test, and homogeneity through the test of Levene. We presented the continuous variables
in means and standard deviations, and the categorical variables, in absolute and percentage
values.
The comparison between the supraspinatus tear pattern (C-shaped versus L- or U-shaped)
and the functional results, according to the ASES and UCLA scales, was performed by
the test of Mann-Whitney. For the other variables, we used the Mann-Whitney test for
continuous variables, and the Chi-squared test for categorical variables.
The Statistical Package for the Social Sciences (SPSS, IBM Corp., Armonk, NY, US)
software, version 21.0, was used for the data analysis, with a significance level
of 5%.
Results
During the study period, we performed 341 arthroscopic rotator cuff repairs. A total
of 174 cases were not included because they had undergone subscapular and/or infraspinatus
repair, or because only partial repair was possible. Thus, we analyzed a sample of
167 shoulders (from 163 patients).
The variables of the patients showed that the C-shaped pattern has a lower proportion
of female patients and a lower frequency of diabetic patients ([Table 1 ]).
Table 1
Supraspinatus tear
Crescent-shaped (n = 104)
L- or U-shaped (n = 63)
p -value
n (%)
n (%)
Gender*
Male
44 (42)
16 (25)
0.027*
Female
60 (58)
47 (75)
Dominant side
Yes
71 (68)
48 (76)
0.273
No
33 (32)
15 (24)
Diabetes*
Yes
9 (9)
14 (22)
0.014*
No
95 (91)
49 (78)
Smoking
No
73 (70)
42 (67)
0.175
Former smoker
21 (20)
9 (14)
Smoker
10 (10)
12 (19)
Age, years (mean ± standard deviation)
53.9 ± 7.9
54.7 ± 7.9
0.822
∗
p < 0.05.
The variables of the surgery showed that the C-shaped pattern has lower retraction,
lower fatty degeneration of the supraspinatus, and less need for procedures in the
tendon of the long head of the biceps ([Table 2 ]).
Table 2
Supraspinatus tear
Crescent-shaped (n = 104)
L- or U-shaped (n = 63)
p -value
n (%)
n (%)
Retraction*
Small
64 (62)
17 (27)
< 0.001*
Medium
37 (36)
31 (49)
Large
2 (2)
11 (17)
Massive
1 (1)
4 (6)
Extension*
Anterior supraspinatus region
57 (55)
41 (65)
0.95
Posterior supraspinatus region
28 (27)
8 (13)
Full extension
19 (18)
14 (22)
Supraspinatus fatty degeneration*
0
54 (52)
18 (29)
0.004*
1
44 (42)
32 (51)
2
6 (6)
11 (17)
3
0 (0)
2 (3)
Subscapularis tear
No
66 (63)
38 (60)
0.685
Partial
38 (37)
25 (40)
Number of anchors
1
33 (32)
29 (46)
0.179
2
67 (64)
32 (51)
3
4 (4)
2 (3)
Acromioplasty
Yes
94 (90)
56 (89)
0.757
No
10 (10)
7 (11)
Mumford procedure
Yes
6 (6)
1 (2)
0.191
No
98 (94)
62 (98)
Biceps procedure*
None
79 (76)
33 (52)
0.007*
Tenotomy
11 (11)
12 (19)
Tenodesis
14 (13)
18 (29)
Tense suture*
Yes
5 (5)
2 (3)
0.61
No
99 (95)
61 (97)
∗
p < 0.05.
Preoperatively, the ASES scale was significantly higher in the C-shaped pattern (43.5 ± 17.6
versus 37. ± 13.8; p = 0.034). The UCLA scale had the same behavior (15.2 ± 4.6 versus 13.5 ± 3.6; p = 0.028). Postoperatively, however, there was no significant difference. According
to the ASES scale, the C-shaped pattern scored 83.7 ± 18.7 points, and the L-or U-shaped
patterns, 82.9 ± 20.1 points (p = 0.887). In the UCLA scale, the values were of 30.9 ± 4.9 and 30.5 ± 5.6 (p = 0.773) respectively ([Table 3 ]).
Table 3
Supraspinatus tear
Crescent-shaped (n = 104)
L- or U-shaped (n = 63)
p -value
American Shoulder and Elbow Surgeons Standardized Shoulder Assessment
Preoperatively (mean ± standard deviation)
43.5 ± 17.6
37.7 ± 13.8
0.034
24 months postoperatively (mean ± standard deviation)
83.7 ± 18.7
82.9 ± 20.1
0.887
Modified-University of California at Los Angeles Shoulder Rating Scale
Preoperatively (mean ± standard deviation)
15.2 ± 4.6
13.5 ± 3.6
0.028
24 months postoperatively (mean ± standard deviation)
30.9 ± 4.9
30.5 ± 5.6
0.773
Discussion
The present study showed that the pattern of the supraspinatus rupture did not affect
the postoperative functional scales. The C-shaped pattern scored 83.7 ± 18.7 points
in the ASES scale, and 30.9 ± 4.9 points in the UCLA scale, while the L- or U-shaped
pattern scored 82.9 ± 20.1 and 30.5 ± 5.6 points respectively. This result is consistent
with that of other articles.[20 ]
[21 ] Park et al.,[20 ] studying large tear, compared the moving patterns (C- and L-shaped) with the U-shaped
pattern, and did not observe significant differences between the groups either. Watson
et al.,[21 ] evaluating posterosuperior tears, did not observe any differences between the groups
as well. However, these authors observed that the improvement obtained compared to
the preoperative period was greater in the C-shaped group, although not significantly
(34.7 points versus 29.5 in the L-shaped group).
We observed that the C-shaped pattern presented statistically higher values preoperatively
according to the ASES (43.5 ± 17.6 versus 37.7 ± 13.8; p = 0.034) and UCLA (15.2 ± 4.6 versus 13.5 ± 3.6; p = 0.028) scales. Although the clinically significant minimum difference was not reached,[27 ] this finding differs from that of other studies.[20 ]
[21 ] Similarly to Watson et al.,[21 ] C-shaped lesions presented a smaller size, but unlike these authors, our sample
showed a significantly lower number of women and diabetics with C-shaped tears. In
addition, we observed greater fatty degeneration and greater need for the biceps procedure
in the L- or U-shaped patterns, a set of data not analyzed by these authors.
The functional improvement with the procedure was greater in L- or U-shaped lesions,
starting from a worse functional state and reaching the same level as patients with
C-shaped lesions. This occurred despite the fact that L- or U-shaped lesions had greater
retraction and greater fatty degeneration. A possible explanation for this is the
fact that we evaluated a predominant sample of patients with degeneration classified
up to grade 2 according to Goutallier in both groups. Only 3% of L- or U-shaped group
were classified as grade 3, and there were no patients classified as grade 4 in either
group. In addition, the tears were restricted to the supraspinatus. Fatty degeneration,
especially in the infraspinatus, is known to generate worse structural results,[9 ] although the effect on the clinical outcome is not statistically significant.[14 ]
[15 ] The size of the tear, in turn, is a risk factor for worse clinical outcomes.[14 ] Our data demonstrate that the tear pattern influenced the degree of fatty degeneration,
but not the postoperative functional outcome.
The rotator cable is important to transmit force from the supraspinatus to the humerus,
even in the presence of a tear.[19 ] This structure is usually preserved in C-shaped tears, which may explain the worse
preoperative function in L- or U-shaped lesions, and the greater functional gain after
its anatomical restoration. Similarly, we consider that this may be the reason for
the higher degree of preoperative fatty degeneration in L- or U-shaped tears.
The present study has some limitations. First, we analyzed only the supraspinatus,
excluding repairs involving the subscapularis and/or infraspinatus. Although this
option decreases the external validity, it was chosen as a means of increasing the
internal validity and reducing the confounding factors. The retrospective cohort design,
although similar to that of previous studies,[20 ]
[21 ] is also a possible source of bias. The intraoperative analysis by only one surgeon
adds subjectivity to the classification. Finally, we did not perform a structural
analysis of the repair, unlike Park et al.[20 ] However, it is known that the structural integrity does not correlate with clinically
significant functional outcomes after repair of the rotator cuff,[28 ] and clinical analysis alone has been already performed by other authors.[21 ]
As favorable points, we highlight the standardized analysis of supraspinatus tears
in a large sample, which was superior to that of previous studies,[20 ]
[21 ] and the demonstration that, although it does not influence the postoperative results,
the pattern of the rupture may influence the preoperative evaluation.
Conclusions
Crescent- and L- or U-shaped tear of the supraspinatus have similar postoperative
functional results. Preoperatively, C-shaped tears have a statistically superior function.
