- Email: [email protected]
Fatty degeneration in the supraspinatus muscle after rotator cuff tear
Fatty degeneration in the supraspinatus muscle after rotator cuff tear Kimio Nakagaki, MD, Jiro Ozaki, MD, Yasuharu Tomita, MD, and Susumu Tamai, MD, Gojo, Nara, and Kashihara, Japan We studied the relationship between fatty degeneration in the supraspinatus muscle after rotator cuff tear and muscle atrophy and tendon retraction in 41 cadaveric shoulders. Eighteen shoulders had a supraspinatus tendon tear. Transverse sections of the supraspinatus muscle were made at the proximal end of the scapular notch. The tendon fiber areas were stained red and the muscle fiber areas were stained yellow by the van Gieson method. Areas of fatty degeneration were stained orange-red by the Sudan three method. These areas were measured with a computer color image analysis system. There were many areas of fatty degenerative change in the supraspinatus muscles associated with a cuff tear, mainly around the tendon fibers and the vessels. The fatty degeneration in the supraspinatus muscle after cuff tear was found to have a strong association with the degree of retraction of the tendon fibers rather than the reduction of muscle volume. (J SHOULDERELBOWSURG 1996;5:194-200.) Supraspinatus muscle atrophy is often associated with large rotator cuff tears. Knowledge of the residual function of an atrophic supraspinatus muscle associated with a torn cuff is useful in selecting the best operative method for repairing the cuff tear and evaluating the prognosis for recovery after surgery. The remaining function of the supraspinatus muscle depend on the degree of fatty degeneration.l' 6 We investigated the degree of fatty degeneration of the supraspinatus muscle belly and its association with reduction of the muscle volume and tendon retraction in cadaveric shoulders. MATERIAL A N D METHODS Forty-one supraspinatus muscles from 22 embalmed cadavers (13 men and 9 women) with a mean age at death of 74.1 years (range 44 to 92 years) were dissected. There was no evidence of neurologic deficit in the specimens. Eighteen shoulders had supraspinatus tendon tears. Of those, 13 shoulders had a full-thickness tear, 4 had a bursal
Fromthe departmentsof OrthopedicSurgery,Gojo PrefecturalHospital, Nara PrefecturalHospilal, and Nara MedicalUniversily. Reprint requests: Kimio Nakagaki, Departmenl of Orthopedic Surgery,Gojo PrefecturalHospital, Nohara, Gojo, 637Japan. Copyright 9 1996 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/96/$5.00 + 0 32/1/69053 194
Anterior
Sect~ :
Coracoid process ~JL Cross
~
~H~ead
d~
of Humerus
p,ne of Scapula Posterior
Figure 1 Axial section of supraspinatus muscle. Transverse section was made at proximal end of scapular notch.
surface tear, and 1 had an articular side tear. The supraspinatus muscles were dissected from the supraspinatus fossa. Transverse sections of the supraspinatus muscles at the proximal end of the scapular notch were made (Figure 1) and stained with the van Gieson method and the Sudan three method (Figure 2). The tendon fiber areas were stained red and the muscle fiber areas were stained yellow by the van Gieson method; these areas were measured by the "Speedy Particle Image and Color Computer Analysis" system (Japan Abionics, Tokyo, Japan). The ratios of the tendon fiber area and the muscle fiber area to the square of the distance from the greater tuberosity to the proximal end of
J. Shoulder Elbow Surg. Volume 5, Number 3
Nakagaki et al.
195
Figure 2 Transversesection of supraspinatusmuscle associatedwith a rotator cuff tear. A, Von Gieson stain. Tendonfiber is clearly shown. B, Sudan three stain. Curved arrow points to degenerated muscle fiber (original magnification x 100). the supraspinatus muscle were calculated and named the "adjusted tendon fiber area" and the "adjusted muscle fiber area," respectively. The area of fatty degeneration stained orange-red by the Sudan three method was measured in a similar fashion. The ratio of the area of fatty degeneration to the total muscle fiber area was calculated and named the "fatty degeneration ratio." In the normal cuff group, we analyzed the patient's age, the adjusted tendon fiber area, and the adjusted muscle fiber area, and these areas' association with the fatty degeneration ratio. In the group of shoulders with cuff tears, we measured the distance from the
greater tuberosity to the edge of the torn cuff. In the shoulder group with a bursal surface cuff tear or an articular side cuff tear, the distance from the greater tuberosity to the normal thickness cuff was measured. We did not measure the width of rotator cuff tear because the border between the supraspinatus tendon and the infraspinatus tendon was unclear. We then analyzed the size of cuff defect, the adjusted tendon fiber area, and the adjusted muscle fiber area, and their relationship to the fatty degeneration ratio. The data were analyzed by means of Pearson's correlation coefficient. A correlation was considered statistically significant when the abso-
196
Nakagaki et al.
J. Shoulder Elbow Surg. May/June 1996
Figure 3 Sectionof supraspinatusmuscle belly with normal rotator cuff. A, Van Gieson stain. Littletendon fiber is seen. B, Sudan three stain (original magnification x100). Littlefatty degeneration is in muscle. lute value of the correlation coefficient was >0.5 and the p value was <0.05.
RESULTS In the supraspinatus muscle with a normal rotator cuff, tendon fibers and fatty degenerative changes were seldom present (Figure 3). The adjusted tendon fiber area averaged 0.000359 (range 0 to 0.00169). The fatty degeneration ratio averaged 0.00583 (range 0 to 0.0316) and was not associated with age (correlation coefficient = 0.269, p = 0.214), the adjusted tendon fiber area
(correlation coefficient = 0.388, p = 0.0673), or the adjusted muscle fiber area (correlation coefficient = 0.404, p = 0.056) (Figures 4, 5, and 6). Many fatty degenerative changes were present around the tendon fibers and the vessels in the supraspinatus muscle associated with cuff tearing (Figure 7). The adjusted tendon fiber area in the cuff tear group averaged 0.00147 (range 0.00005 to 0.00379). The fatty degeneration ratio averaged 0.0260 (range 0.0001 to 0.126). The fatty degeneration ratio correlated weft with the size of the cuff defect (correlation coefficient = 0.758, p =
J. Shoulder Elbow Surg. Volume 5, Number 3
Nakagaki et al.
Fatty Degeneration Ratio
R-0269
(X0"~i)
p=O1214
197
9
ee
2] 11
98
w g
1
2
3
4
5
9149 ~
6
7
Age
8 9 xl0years
Figure 4 Relationship between age and fatty degeneration ratio in shoulders with normaJ rotator cuffs.
Fatty Degeneration Ratio (xo.1) (9 torncuff group R=0.673(p=0.0022) 2-
,:o:)
normalcuff group R=0.388(p=0.0673)
_
~ .o~176176176176176176176176176176176 ~176176176176176
9 __Oo
o
.~.qt......... 9
......
o
1
9 ,
,
,
2
3
4
Adjusted Tendon Fiber Are(~~176176 Figure 5 Relationship between adjusted tendon fiber area and fatty degeneration ratio. Dotted line indicates regression )ine in torn cuff group.
198
J. Shoulder Elbow Surg. May/June 1996
Nakagaki et al.
Fatty Degeneration Ratio (xo.1) 2 t o r n c u f f g r o u p
R=-0.295 (p=0.235)
~0~ normal cuff group R=0.404 (p:0.056)
_
9 o 9
1
o 0 I
I
2
3
9
9 9
9
_~
o
9
O0 I
4
5
6
7
q
8
Adjusted Muscle Fiber Area(x~176 Figure 6 Relationshipbetween adjusted musclefiber area and fatty degeneration ratio.
0.0003) (Figure 8). The correlation coefficient between the adjusted tendon fiber area and the fatty degeneration ratio was 0.673 (p = 0.0022) (Figure 5). Thus the fatty degeneration ratio was associated with both the size of cuff defect and the adjusted tendon fiber area. The correlation coefficient between the adjusted muscle fiber area and the fatty degeneration ratio was -0.295 (p = 0.235) (Figure 6). The fatty degeneration ratio was therefore not associated with reduction of the supraspinatus muscle volume. DISCUSSION It has been reported that fatty degenerative changes are seen in the muscles with a torn rotator cuff tendon. ~ Seeger et al. 5 reported that magnetic resonance imaging can depict the extent of muscle retraction in patients with complete rupture of the supraspinatus tendon. The supraspinatus muscle atrophy associated with a cuff tear is characterized by bands of bright signal within the muscle belly, indicative of fatty replacement. Goutallier et al. 3 reported that muscular fatty degeneration of the rotator cuff seen on computed tomography scans implicates a tear of the corresponding tendon. Bj6rkenheim 1 reported that in rabbits with resected supraspinatus tendons, the supraspinatus
muscle has decreased contractile activity and increased development of fatty degeneration. Jozsa et al. 4 measured by automatic image analyzer the connective tissue to muscle ratio from histologic sections of the muscle from an immobilized rat limb. However, the quantitative analysis of fatty degeneration in a muscle associated with a rotator cuff tear has never been reported. We quantitatively compared the degree of the fatty degeneration in the supraspinatus muscle belly with the size of cuff defect, the degree of reduction of the muscle volume, and the degree of retraction of the tendon fiber. The fatty degeneration correlated well with the size of cuff defect and the tendon fiber area but not with the muscle fiber area in shoulders with cuff tears. This finding implies that the fatty degeneration in the supraspinatus muscle after a cuff tear is associated with retraction of the tendon fiber rather than with reduction of the muscle size. The symptomatic period could not be related to the atrophic changes of the supraspinatus muscle because it could not be assessed for the cadaveric specimens. Dastur et al. 2 reported that accumulated lipofuscin was often found in muscle fibers in an immobilized limb and occasionally in the endothelial cells of intramuscular blood vessels. We
J. Shoulder Elbow Surg. Volume 5, Number 3
Nakagaki et al.
199
Figure 7 Section of supraspinatus muscle associated with cuff tear. Sudan three stain; F,, fatV degeneration tissue; T, tendon fiber; V,, vessel; M, muscle fiber. A, Fatty degeneration of muscle near tendon fiber (original magnification x20). B, Fatl?' degeneration of muscle near vessels (original magnification x20).
believe that the fatty degeneration is increased in the supraspinatus muscle because the retracted tendon narrows the length of muscle contraction and lipofuscin is increased. Such fatty degeneration is thought to impair the remaining function of the supraspinatus muscle with a cuff tear and lead to a suboptimal result after surgical rotator cuff repair.
REFERENCES
1. Bj6rkenheim J-M. Structure and function of the rabbit's supraspinatus muscle after resection of its tendon. Acta Orthop Scand 1989;60:461-3. 2. DasturDK, Gagrat BM, Manghani DK. Fine structureoFmuscle in human disuse atrophy: significance of proximal muscle involvement in muscle disorders. Neuropathd Appl Neurobiol 1979;5:85-101. 3. Goutallier D, BernageauJ, Patte D. Assessmentof the trophicily
200
Nakagaki et al.
J. Shoulder Elbow Surg. May~June 1996
Fatty Degeneration Ratio (x0.1)2
R=0.758 p=0.0003
m
9
.o,.o.o,~176176176176176176176 "~17617617617617617617617617617617617
9
9 9 9 ..oe.- 9 e...o~....-~"
_
|
l
l
I
I
I
i
I
1
2
3
4
5
6
7
8
Size of Cuff Defect
(X10mm)
F i g u r e 8 Relationship between size of cuff defect and fatly degeneration ratio. Dotted line indicates regression line of the muscles of the ruptured rotator cuff by CT scan In: Post M, Morrey BF, Hawkins Rj, eds. Surgery of the shoulder. St Louis: Mosby-Year Book, 1990:11-3. 4. jozsa L, Thoring j, Jarvinen M, Kannus P, Lehto M, Kvist M. Quantitative alterations in inlramuscular connective tissue following immobilization: an experimental study in lhe rat calf muscles. Exp Mol Pathd i9B8;49:267-78.
5. Seeger LL, Gold RH, Basset LW, EIJman H Shoulder impingement syndrome: MR findings in 53 shoulders. AJR Am J Roentgenol 1988;150:343-7. 6. Zlatkin MB, Dalinka MK, Kressd HY Magnetic resonance imaging of the shoulder Magn Reson Q 1989;5:3-22
Fromthe departmentsof OrthopedicSurgery,Gojo PrefecturalHospital, Nara PrefecturalHospilal, and Nara MedicalUniversily. Reprint requests: Kimio Nakagaki, Departmenl of Orthopedic Surgery,Gojo PrefecturalHospital, Nohara, Gojo, 637Japan. Copyright 9 1996 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/96/$5.00 + 0 32/1/69053 194
Anterior
Sect~ :
Coracoid process ~JL Cross
~
~H~ead
d~
of Humerus
p,ne of Scapula Posterior
Figure 1 Axial section of supraspinatus muscle. Transverse section was made at proximal end of scapular notch.
surface tear, and 1 had an articular side tear. The supraspinatus muscles were dissected from the supraspinatus fossa. Transverse sections of the supraspinatus muscles at the proximal end of the scapular notch were made (Figure 1) and stained with the van Gieson method and the Sudan three method (Figure 2). The tendon fiber areas were stained red and the muscle fiber areas were stained yellow by the van Gieson method; these areas were measured by the "Speedy Particle Image and Color Computer Analysis" system (Japan Abionics, Tokyo, Japan). The ratios of the tendon fiber area and the muscle fiber area to the square of the distance from the greater tuberosity to the proximal end of
J. Shoulder Elbow Surg. Volume 5, Number 3
Nakagaki et al.
195
Figure 2 Transversesection of supraspinatusmuscle associatedwith a rotator cuff tear. A, Von Gieson stain. Tendonfiber is clearly shown. B, Sudan three stain. Curved arrow points to degenerated muscle fiber (original magnification x 100). the supraspinatus muscle were calculated and named the "adjusted tendon fiber area" and the "adjusted muscle fiber area," respectively. The area of fatty degeneration stained orange-red by the Sudan three method was measured in a similar fashion. The ratio of the area of fatty degeneration to the total muscle fiber area was calculated and named the "fatty degeneration ratio." In the normal cuff group, we analyzed the patient's age, the adjusted tendon fiber area, and the adjusted muscle fiber area, and these areas' association with the fatty degeneration ratio. In the group of shoulders with cuff tears, we measured the distance from the
greater tuberosity to the edge of the torn cuff. In the shoulder group with a bursal surface cuff tear or an articular side cuff tear, the distance from the greater tuberosity to the normal thickness cuff was measured. We did not measure the width of rotator cuff tear because the border between the supraspinatus tendon and the infraspinatus tendon was unclear. We then analyzed the size of cuff defect, the adjusted tendon fiber area, and the adjusted muscle fiber area, and their relationship to the fatty degeneration ratio. The data were analyzed by means of Pearson's correlation coefficient. A correlation was considered statistically significant when the abso-
196
Nakagaki et al.
J. Shoulder Elbow Surg. May/June 1996
Figure 3 Sectionof supraspinatusmuscle belly with normal rotator cuff. A, Van Gieson stain. Littletendon fiber is seen. B, Sudan three stain (original magnification x100). Littlefatty degeneration is in muscle. lute value of the correlation coefficient was >0.5 and the p value was <0.05.
RESULTS In the supraspinatus muscle with a normal rotator cuff, tendon fibers and fatty degenerative changes were seldom present (Figure 3). The adjusted tendon fiber area averaged 0.000359 (range 0 to 0.00169). The fatty degeneration ratio averaged 0.00583 (range 0 to 0.0316) and was not associated with age (correlation coefficient = 0.269, p = 0.214), the adjusted tendon fiber area
(correlation coefficient = 0.388, p = 0.0673), or the adjusted muscle fiber area (correlation coefficient = 0.404, p = 0.056) (Figures 4, 5, and 6). Many fatty degenerative changes were present around the tendon fibers and the vessels in the supraspinatus muscle associated with cuff tearing (Figure 7). The adjusted tendon fiber area in the cuff tear group averaged 0.00147 (range 0.00005 to 0.00379). The fatty degeneration ratio averaged 0.0260 (range 0.0001 to 0.126). The fatty degeneration ratio correlated weft with the size of the cuff defect (correlation coefficient = 0.758, p =
J. Shoulder Elbow Surg. Volume 5, Number 3
Nakagaki et al.
Fatty Degeneration Ratio
R-0269
(X0"~i)
p=O1214
197
9
ee
2] 11
98
w g
1
2
3
4
5
9149 ~
6
7
Age
8 9 xl0years
Figure 4 Relationship between age and fatty degeneration ratio in shoulders with normaJ rotator cuffs.
Fatty Degeneration Ratio (xo.1) (9 torncuff group R=0.673(p=0.0022) 2-
,:o:)
normalcuff group R=0.388(p=0.0673)
_
~ .o~176176176176176176176176176176176 ~176176176176176
9 __Oo
o
.~.qt......... 9
......
o
1
9 ,
,
,
2
3
4
Adjusted Tendon Fiber Are(~~176176 Figure 5 Relationship between adjusted tendon fiber area and fatty degeneration ratio. Dotted line indicates regression )ine in torn cuff group.
198
J. Shoulder Elbow Surg. May/June 1996
Nakagaki et al.
Fatty Degeneration Ratio (xo.1) 2
R=-0.295 (p=0.235)
~0~ normal cuff group R=0.404 (p:0.056)
_
9 o 9
1
o 0 I
I
2
3
9
9 9
9
_~
o
9
O0 I
4
5
6
7
q
8
Adjusted Muscle Fiber Area(x~176 Figure 6 Relationshipbetween adjusted musclefiber area and fatty degeneration ratio.
0.0003) (Figure 8). The correlation coefficient between the adjusted tendon fiber area and the fatty degeneration ratio was 0.673 (p = 0.0022) (Figure 5). Thus the fatty degeneration ratio was associated with both the size of cuff defect and the adjusted tendon fiber area. The correlation coefficient between the adjusted muscle fiber area and the fatty degeneration ratio was -0.295 (p = 0.235) (Figure 6). The fatty degeneration ratio was therefore not associated with reduction of the supraspinatus muscle volume. DISCUSSION It has been reported that fatty degenerative changes are seen in the muscles with a torn rotator cuff tendon. ~ Seeger et al. 5 reported that magnetic resonance imaging can depict the extent of muscle retraction in patients with complete rupture of the supraspinatus tendon. The supraspinatus muscle atrophy associated with a cuff tear is characterized by bands of bright signal within the muscle belly, indicative of fatty replacement. Goutallier et al. 3 reported that muscular fatty degeneration of the rotator cuff seen on computed tomography scans implicates a tear of the corresponding tendon. Bj6rkenheim 1 reported that in rabbits with resected supraspinatus tendons, the supraspinatus
muscle has decreased contractile activity and increased development of fatty degeneration. Jozsa et al. 4 measured by automatic image analyzer the connective tissue to muscle ratio from histologic sections of the muscle from an immobilized rat limb. However, the quantitative analysis of fatty degeneration in a muscle associated with a rotator cuff tear has never been reported. We quantitatively compared the degree of the fatty degeneration in the supraspinatus muscle belly with the size of cuff defect, the degree of reduction of the muscle volume, and the degree of retraction of the tendon fiber. The fatty degeneration correlated well with the size of cuff defect and the tendon fiber area but not with the muscle fiber area in shoulders with cuff tears. This finding implies that the fatty degeneration in the supraspinatus muscle after a cuff tear is associated with retraction of the tendon fiber rather than with reduction of the muscle size. The symptomatic period could not be related to the atrophic changes of the supraspinatus muscle because it could not be assessed for the cadaveric specimens. Dastur et al. 2 reported that accumulated lipofuscin was often found in muscle fibers in an immobilized limb and occasionally in the endothelial cells of intramuscular blood vessels. We
J. Shoulder Elbow Surg. Volume 5, Number 3
Nakagaki et al.
199
Figure 7 Section of supraspinatus muscle associated with cuff tear. Sudan three stain; F,, fatV degeneration tissue; T, tendon fiber; V,, vessel; M, muscle fiber. A, Fatty degeneration of muscle near tendon fiber (original magnification x20). B, Fatl?' degeneration of muscle near vessels (original magnification x20).
believe that the fatty degeneration is increased in the supraspinatus muscle because the retracted tendon narrows the length of muscle contraction and lipofuscin is increased. Such fatty degeneration is thought to impair the remaining function of the supraspinatus muscle with a cuff tear and lead to a suboptimal result after surgical rotator cuff repair.
REFERENCES
1. Bj6rkenheim J-M. Structure and function of the rabbit's supraspinatus muscle after resection of its tendon. Acta Orthop Scand 1989;60:461-3. 2. DasturDK, Gagrat BM, Manghani DK. Fine structureoFmuscle in human disuse atrophy: significance of proximal muscle involvement in muscle disorders. Neuropathd Appl Neurobiol 1979;5:85-101. 3. Goutallier D, BernageauJ, Patte D. Assessmentof the trophicily
200
Nakagaki et al.
J. Shoulder Elbow Surg. May~June 1996
Fatty Degeneration Ratio (x0.1)2
R=0.758 p=0.0003
m
9
.o,.o.o,~176176176176176176176 "~17617617617617617617617617617617617
9
9 9 9 ..oe.- 9 e...o~....-~"
_
|
l
l
I
I
I
i
I
1
2
3
4
5
6
7
8
Size of Cuff Defect
(X10mm)
F i g u r e 8 Relationship between size of cuff defect and fatly degeneration ratio. Dotted line indicates regression line of the muscles of the ruptured rotator cuff by CT scan In: Post M, Morrey BF, Hawkins Rj, eds. Surgery of the shoulder. St Louis: Mosby-Year Book, 1990:11-3. 4. jozsa L, Thoring j, Jarvinen M, Kannus P, Lehto M, Kvist M. Quantitative alterations in inlramuscular connective tissue following immobilization: an experimental study in lhe rat calf muscles. Exp Mol Pathd i9B8;49:267-78.
5. Seeger LL, Gold RH, Basset LW, EIJman H Shoulder impingement syndrome: MR findings in 53 shoulders. AJR Am J Roentgenol 1988;150:343-7. 6. Zlatkin MB, Dalinka MK, Kressd HY Magnetic resonance imaging of the shoulder Magn Reson Q 1989;5:3-22