I
Inr J Radiorion Oncology Bml Phys Vol. 20. pp. 463-47 Printed I” the U.S.A. All nghts resend
0360-3016/91 $3.00 + .OO Copyright Q 1991 Pergamon Press plc
??Original Contribution
MUSCLE
INJURY FOLLOWING EXPERIMENTAL INTRAOPERATIVE IRRADIATION
B. E. POWERS, D.V.M., S. L. MCCHESNEY
GILLETTE,
PH.D.,* M.S.,
E. L. GILLETTE,
PH.D.,*
AND S. J. WITHROW,
D.V.M.,
R. A. LECOUTEUR,
PH.D.,* BVSc,
PH.D.+
D.V.M.+
Colorado State University, Fort Collins, CO 80523 The paraaortic region of beagle dogs was irradiated to 15 to 55 Gy intraoperative irradiation, 10 to 47.5 Gy intraoperative irradiation following 50 Gy external beam irradiation in 25 fractions, or 50 to 80 Gy external beam irradiation in 30 fractions. Six MeV electrons were used for intraoperative irradiation, and external beam irradiation was done using photons from a 6 MV linear accelerator. The psoas muscle in the irradiation field was examined histomorphometrically 2 or 5 years after irradiation. The percentage of muscle fibers and capillaries decreased, whereas the percentage of connective tissue increased with increased dose for both intraoperative irradiation only and intraoperative irradiation plus external beam irradiation. The dose causing a 50% decrease in the percentage of muscle fibers was 21.2 Gy and 33.8 Gy at 2 and 5 years, respectively, after intraoperative irradiation alone, and 22.9 Gy and 25.2 Gy at 2 and 5 years, respectively, after intraoperative irradiation combined with 50 Gy external beam irradiation. The EDa for severe vessel lesions was 19.2 Gy and 25.8 Gy at 2 and 5 years, respectively, after intraoperative irradiation alone and 16.0 Gy and 18.0 Gy at 2 and 5 years, respectively, after intraoperative irradiation combined with 50 Gy external beam irradiation. External beam irradiation alone caused a slight decrease in percentage of muscle fibers with increased dose, and vessel lesions were infrequent or mild. Radiation-induced muscle injury was characterized by loss of muscle fibers, decreased fiber size, severe vessel lesions, hemorrhage, inflammation, coagulation necrosis, and fibrosis. These histopathologic characteristics distinguish this muscle injury from that caused by neurogenic atrophy. These data indicate that radiation-induced muscle injury most likely was caused by injury of the supporting vasculature. The lesions produced were largely a function of the single intraoperative dose rather than the external beam fractionated doses. Furthermore, it appears that 20 to 25 Gy intraoperative irradiation combined with 50 Gy external beam irradiation may be near the maximum tolerated dose by sublumbar musculature and its supporting vasculature. lntraoperative
irradiation, Canine, Late effects, Muscle.
tumors (2, 7, 14). The advantage of IORT is that the ra-
INTRODUCTION
diation field can be directly visualized, and sensitive structures such as intestines and kidneys can be manually retracted from the field. Furthermore, the volume of tissue irradiated can be limited by the use of electrons which, depending on their energy, have limited penetration. Complications following IORT in humans are reported to be not significantly different from those following more conventional external beam irradiation (EBRT) (18).
Intraoperative irradiation (IORT) is being used with increasing frequency for the treatment of various malignancies in humans (14). Clinical results for the treatment of rectal cancer appear to be especially promising ( 14, 19). Other malignancies being treated by IORT include gastric, pancreatic, urinary bladder, biliary, retroperitoneal, prostatic, paraaortic nodal, and a variety of pediatric
Presented in part at the 3 1st Meeting of the American Society of Therapeutic Radiology and Oncology in San Francisco, Oc-
vided medical and surgical support. Special thanks to D. Madden for assistance at necropsy and the histopathology laboratory for histology support. B. A. Ensley, K. T. Crump, and M. M. Estabrook who provided technical assistance are acknowledged, as is Mrs. L. A. Ludwig for preparation of the manuscript. We also wish to thank Dr. T. B. Borak and R. J. Scott for radiation physics and engineering support. This investigation was supported by PHS grant number CA29 117 awarded by the National Cancer Institute, DHHS. Accepted for publication 12 September 1990.
tober 1989. * Department of Radiology and Radiation Biology. + Department of Clinical Sciences. Reprint requests to: Barbara E. Powers, D.V.M., Ph.D., Comparative Oncology Unit, Colorado State University, Fort Collins, CO 80523. Acknowledgements-The authors wish to thank Drs. Herman D. Suit and Joel Tepper for contributing to the design of the experiment and for continuing consultation. Dr. S. L. Allen pro-
463
464
I. J. Radiation
Oncology
0
Biology 0 Physics
Some complications reported include fibrosis (2, 16, 18). hemorrhage ( 18), peripheral neuropathies (2.9). bone necrosis (7), and pelvic pain (18). Most patients have been followed for a relatively short period of time, with few patients observed over 2 years after irradiation. Numerous studies in experimental dogs have been done to determine the effects of IORT on various normal tissues in the radiation field. Many of these studies had relatively short periods of observation or involved few animals at specific doses and time points. More recently, studies have been completed describing the effect of IORT or IORT combined with EBRT 2 and 5 years after irradiation of the paraaortic area in dogs. Lesions that were reported included aortic thromboses and aneurysms, branch artery injury (5, 6), peripheral neuropathies (IO), ureteral strictures ( 11) bone necrosis, and tumor induction ( 12). The purpose of this report is to describe injury to the sublumbar muscle that was included in the paraaortic radiation field of dogs 2 and 5 years after IORT and IORT combined with EBRT.
METHODS Experimenlal
AND
MATERIALS
March
I,
199
Volume
20. Number
Table 2. Experimental
Group IORT only 17.5 25.0 32.5 40.0 47.5 55.0 IORT plus IORT dose 10.0 17.5 25.0 32.5 40.0 47.5 EBRT only 60.0 70.0 80.0 Controls
3
design for 5-year studv
No. of dogs originally in group
No. of dogs alive at least 4 years after irradiation
(Gy) 5 5 6 6 2 50 Gy EBRT (Gy) 5
5 4 4 4
(Gy) 5 6
5 5 4
design
One hundred seventy-seven young adult beagle dogs of both sexes were randomized to receive various radiation doses of IORT, IORT combined with 50 Gy EBRT, or EBRT only to the paraaortic area (Tables 1 and 2). Ninety dogs that were to be observed for 2 years after irradiation received 15 to 50 Gy IORT only; 10 to 42.5 Gy IORT combined with 50 Gy EBRT in 2 Gy fractions; or 50, 60,
Table I. Experimental
Group
design for 2-year study No. of dogs originally in group
No. of dogs alive at 2 years after irradiation
5 5 5 5 5 5 5
5 5 4 4 2 3 2
5 5 5 5 5 5
5 5 5 5 5 5
4 6 5 5 5
4 6 5 4 5
IORT only (Gy)
15.0 20.0 25.0 30.0 35.0 42.5 50.0 IORT plus 50 Gy EBRT IORT dose (Gy) 10.0 15.0 20.0 21.5 35.0 42.5 EBRT only (Gy) 50.0 60.0 70.0 80.0 Controls
70, or 80 Gy EBRT only in 30 fractions. Four to six dogs were in each dose group, and five sham irradiated dogs served as controls (Table 1). Eighty-seven dogs that were to be observed for 5 years after irradiation received 17.5 to 55 Gy IORT only: 10 to 47.5 Gy combined with 50 Gy EBRT in 2 Gy fractions: or 60. 70, or 80 Gy EBRT only in 30 fractions. Five or six dogs were in each dose group, and seven age-matched non-irradiated dogs served as controls (Table 2). Irradiations
For IORT, dogs were anesthetized, placed in dorsal recumbency, and had a celiotomy performed under sterile conditions as previously described (5. 6, IO, 1 1, 12). A 5 X 8 cm plexiglass cone was placed over the paraaortic region. and the intestines, kidney, and urinary bladder were retracted from the field. In the 2-year study, the left ureter was also retracted, whereas it was included in ‘the field for the j-year study. Other tissues included in the held were the aorta, vena cava, lumbar nerves, ventral cortex of lumbar vertebrae 3-4 through 7, and the left psoas muscle. Radiation was delivered with 6 MeV electrons at a rate of 6.6 Gy/min and an SSD of 125 cm. The D mal was 15 mm. Doses were calculated such that irradiated tissues of interest were within 90% of the prescribed dose. For dogs receiving EBRT in addition to IORT, the 50 Gy was delivered in 2 Gy fractions over a j-week period immediately prior to IORT. The EBRT field was 5 X 10 cm and included the aorta, vena cava, lumbar nerves, both ureters, a portion of the left kidney, ventral cortex of lumbar vertebrae 2 through 7, and the psoas muscle.
Muscle injury following experimental IORT 0 B. E. POWERSet al.
For EBRT, dogs were tranquilized and restrained in a plexiglass box in a standing position. Irradiation was with photons from a 6 MV linear accelerator. The dose rate was 2.5 Gy/min and the SSD was 100 cm. Doses were calculated to the dorsal ventral midline, and the psoas muscle was within the 90% isodose level. Dogs receiving EBRT only received 50 to 80 Gy in 30 equal fractions over a 6-week period. Radiation was delivered as described above.
Dogs were observed daily for any adverse signs. Some dogs were euthanatized or died prior to scheduled necropsy date (Tables 1 and 2). At 2 or 5 years after irradiation, dogs were anesthetized, heparinized, and given an overdose of barbiturates. Immediately dogs were perfused through the left ventricle of the heart with the aid of a perfusion pump set at physiologic pressure. The perfusion solution consisted of 4 liters of 1.0% paraformaldehyde and 1.25% glutaraldehyde followed by 4 liters of 4.0% paraformaldehyde and 5.0% glutaraldehyde. Tissues, including the psoas muscles, were dissected from the dog, labeled, and placed in 10% neutral buffered formalin. Transverse sections of the left psoas muscle were embedded in paraffin, sectioned at 5 pm, and stained with hematoxylin and eosin and Masson’s trichrome. Three sections, spaced 50 pm apart, were made of each muscle. Histomorphometry was done using a 36 point ocular grid in a microscope (3). A total of 500 points were counted for each of the three sections and recorded as to whether the grid line intersection covered muscle, connective tissue, capillaries, larger vessels, fat (Fig. 1A) or hemorrhage and fibrin. Lesions in arteries and arterioles were scored on a scale of 0 to 3 where grade 0 was no lesions, grade 1 was mild fibrosis and/or mild intimal proliferation, grade 2 was marked fibrosis and/or intimal proliferation, and grade 3 was thrombosis, disruption, or medial necrosis. Inflammation was noted as being present or not. All slides were examined without prior knowledge of treatment received. Statistical analysis
Only dogs that survived at least 20 months in the 2year study or at least 4 years in the 5-year study were included for statistical analysis (Tables 1 and 2). From histomorphometry, the percentage of each tissue component was calculated by dividing the number of “hits” on a tissue of interest over the total number of points counted. The mean and standard error of the mean (SEM) were calculated for each dose group and plotted against dose. Linear regression analysis was used to determine the dose response and r values and confidence intervals were calculated. For graded vessel lesions, a quanta1 response was determined and grades of 2 or 3 were counted as positive. The quanta1 data were plotted against dose, and using probit analysis, the dose to cause the lesion in
465
50% of the dogs, EDso, and confidence intervals were calculated.
RESULTS Two years after irradiation
In the study designed for a 2-year observation period, 11 dogs died or were euthanatized prior to 2 years, primarily because of peripheral neuropathies or spinal cord damage (Table 1) ( 10). These dogs were not included in analysis of the psoas muscle. At the higher IORT doses 2 years after irradiation, the psoas muscle was grossly smaller than normal, firm, pale, and adherent to surrounding tissues and bone. Histologically, there were focal areas of increased connective tissue associated with the loss or shrinkage of muscle fibers. At higher radiation doses, large areas of loss of muscle were present, and scattered muscle fibers were either swollen and hypereosinophilic or smaller than normal. Fibrous connective tissue or masses of hemorrhage and fibrin replaced areas of muscle fiber loss (Fig. 1B and 2). Histomorphometry revealed that with increasing dose there was a decrease in the percentage of muscle fibers (? = 0.70), an increase in the percentage of connective tissue (? = 0.67) a decrease in the percentage of capillaries (3 = 0.64), and an increase in the percentage of hemorrhage and fibrin (? = 0.75) after IORT only (Fig. 3). Similarly, after IORT combined with 50 Gy EBRT with increasing IORT dose, there was a decrease in the percentage of muscle fibers (? = 0.88), an increase in the percentage of connective tissue (? = 0.79), a decrease in the percentage of capillaries (? = 0.87), and an increase in the percentage of hemorrhage and fibrin (? = 0.87) (Fig. 3). The dose causing a 50% decrease in the percentage of muscle fibers from a normal of 84.0% was 2 1.2 Gy (confidence intervals could not be calculated) IORT only and 22.9 Gy (I 5.228.6,95% C.I.) IORT when combined with 50 Gy EBRT (Table 3). The dose which caused a 20% increase in the percentage of hemorrhage and fibrin was 38.8 Gy (30.559.0, 95% C.I.) IORT only and 30.8 Gy (24.5-42.5, 95% C.I.) IORT when combined with 50 Gy EBRT (Table 3). After EBRT only there was a minimal increase in percentage of connective tissue with increasing dose (9 = 0.77) but other parameters did not change significantly. Vessel lesions were present 2 years after irradiation and consisted of mild perivascular fibrosis and mild intimal proliferation (grade 1 lesion) at lower doses. At higher doses, severe perivascular fibrosis and severe intimal proliferation (grade 2 lesion) or severe hyalinization, medial necrosis, and disruption of the vessel walls with thrombosis (grade 3 lesion) was present in some arteries and arterioles (Fig. 2). The EDso for grade 2 or 3 vessel lesions was 19.2 Gy (17.5-2 1.1 95% CI) IORT only and 16.0 Gy (12.620.3 95% CI) IORT when combined with 50 Gy EBRT (Table 3). Grade 2 or 3 vessel lesions were not seen in any dogs that received EBRT only.
466
1. J. Radiation
Oncology
0 Biology
0 Physics
March
1991, Volume
20, Number
3
Fig. l.(A) Normal muscle with large fibers (M), capillaries (arrows), and fat (F). Massons X200. (B) Muscle from a dog 5 years after 25 Gy IORT plus 50 Gy EBRT. Muscle fibers (M) with markedly increased fibrous connective tissue (F). Massons X200. (Figures shown 75% of original size.)
Inflammation, characterized by infiltrates of lymphocytes, plasma cells, and a few macrophages was present and predominately perivascular in location. Inflammation was present in 79% of the dogs receiving IORT only, mostly at doses of 20 Gy and above and in 73% of the dogs receiving IORT with 50 Gy EBRT, mostly at doses of 15 Gy IORT and above. Inflammation was not present in dogs receiving EBRT only.
Five years ajer irradiation In the study designed to evaluate dogs for 5 years, 13 dogs died or were euthanatized prior to 4 years, primarily because of peripheral nerve or ureteral ( 11) damage (Table
2). These dogs were not included in analysis of the psoas muscle. At 5 years after irradiation at the higher doses, the psoas muscle was grossly small, firm, pale, and adherent to surrounding tissues and bone. In two dogs that received 47.5 Gy IORT with 50 Gy EBRT, the muscle was abscessed and consisted of a firm connective tissue capsule filled with thick, foul-smelling, tan-red exudate. In three dogs that received 50 Gy EBRT and 25, 40, and 47.5 Gy IORT, the muscle was replaced by a firm gritty tumor ( 12). Histologically, lesions were similar to those observed at 2 years after irradiation, but swollen eosinophilic fibers were less frequent. Connective tissue appeared more dense (Fig. 1B) with occasional mineral formation,
Muscle injury following experimental IORT 0 B. E.
POWERS
et al.
Fig. 2. Muscle from a dog 2 years after 27.5 Gy IORT plus 50 Gy EBRT. Muscle replaced by fibrous connective tissue (F), edema (E), and hemorrhage (arrows). A thrombosed artery (A) is also present. Massons X 100. (Figure shown 80% of original size.)
remaining muscle fibers appeared larger, and inflammation, hemorrhage, and fibrin was less frequent than at 2 years after irradiation. The two dogs with abscess formation had pockets of neutrophils, masses of bacteria, total loss of muscle fibers, and fibrous tissue replacement. The three dogs with tumor had a complete replacement of the muscle by the tumor that were osteosarcomas (12); these three dogs were not included in histomorphometric data. Histomorphometrically, with increasing dose there was a decrease in the percentage of muscle (? = 0.93), an increase in the percentage of connective tissue (9 = 0.79)
90
r
0 15
a decrease in the percentage of capillaries (? = 0.94) and an increase in the percentage of hemorrhage and fibrin (r’ = 0.86) after IORT only (Fig. 4). Similarly, after IORT combined with 50 Gy EBRT, with increasing dose there was a decrease in the percentage of muscle (? = 0.95) an increase in the percentage ofconnective tissue (? = 0.78), a decrease in the percentage of capillaries (f = 0.92), and an increase in the percentage of hemorrhage and fibrin (i = 0.67) (Fig. 4). The dose which caused a 50% decrease in the percentage of muscle from a normal of 80.4% was 33.8 Gy (27.838.6, 95% C.I.) IORT only and 25.2 Gy (20.4-29.2, 95%
EBRT
IORT
20
25
30
35
40
45 DOSE
Fig. 3. Graph of percentage of tissue components Control values are open symbols.
+ IORT
(Gy)
versus dose 2 years after irradiation.
Mean and SEM are shown.
468
I.
J.
Radiation
Oncology 0 Biology 0 Physics
March 1991. Volume 20. Number 3
Table 3. Doses causing muscle and vessel lesions Time (years)
Lesion 50% decrease
IORT (Gy)
IORT plus 50 Gy EBRT (Gy)
Difference (GY)
in muscle
2 5
2 I .2 (-) 33.8 (27.8-38.6)*
22.9 (15.2-28.6) 25.2 (20.4-29.2)
8.6
20% increase in fibrin and hemorrhage
2 5
38.8 (30.5-59.0) 45.5 (38.5-59.8)
30.8 (24.5-42.5) 44.5 (-)
8.0
2 5
19.2 (17.5-21.1) 25.8 (23.8-28.1)
16.0 (12.6-20.3) 18.0 (16.2-20.0)
3.2 7.8
EDso for grade 2 & 3 vessel lesions
* 95% confidence
I.0
interval.
Neurogenic muscle atrophy
C.I.) IORT when combined with 50 Gy EBRT (Table 3). The dose which caused a 20% increase in the percentage of hemorrhage and fibrin was 45.5 Gy (38.5-59.8, 95% C.I.) IORT only and 44.5 Gy (confidence intervals could not be calculated) IORT when combined with 50 Gy EBRT (Table 3). After EBRT only there was a mild decrease in percentage of muscle fibers with increasing dose (8 = 0.92), but the decrease was only to 76.5% muscle. There was no significant change in the other parameters. Vessel lesions were present 5 years after irradiation and were similar in type to those observed 2 years after irradiation. The EDso for grade 2 or 3 vessel lesions was 25.8 Gy (23.8-28.1 95% CI) IORT only and 18.0 Gy ( 16.220.0 95% CI) IORT when combined with 50 Gy EBRT (Table 3). One dog receiving 80 Gy EBRT had a grade 2 vessel lesion. Chronic inflammation was similar to that observed 2 years after irradiation but was less frequent. Inflammation was present in 39% of the dogs receiving IORT only, mostly at doses of 40 Gy IORT and above, and in 54% of the dogs receiving IORT with 50 Gy EBRT. mostly at doses of 32.5 Gy IORT and above. Inflammation was not present in dogs receiving EBRT only.
90
I
0
15
At both 2 and 5 years after irradiation, many dogs had clinical peripheral neuropathies caused by radiation damage of the lumbar nerves (10). The quadriceps muscles, which were outside of the radiation field but supplied by those nerves, were small, firm, and pale. Histologically, the muscle fibers were small and angular, but there was no significant increase in connective tissue, no vessel lesions nor hemorrhage or fibrin, and no inflammation (Fig. 5). DI!XXJSSION
This study was designed to determine the late effects of irradiation of normal tissues in the paraaortic region of dogs after IORT, IORT combined with EBRT. and EBRT only. Some of the late effects that have occurred in this study include aortic thrombosis and aneurysms (5, 6). peripheral neuropathies (lo), ureteral strictures ( 1 I), bone necrosis, and tumor induction (12). The left psoas muscle, which was also included in the radiation field, showed injury characterized by loss of muscle fibers, atrophy. and fibrosis. The doses to cause a 50% decrease in
EBRT+
IORT
20
25
30
35
40
Fig. 4. Graph of percentage of tissue components Control values are open symbols.
45
50
55
60
DOSE
(Gy)
65
70
75
IORT
80
85
versus dose 5 years after irradiation.
90
95
100
Mean and SEM are
Muscle injury following experimental IORT 0 B. E. POWERSef al.
469
Fig. 5. Quadriceps muscle out of the field of irradiation showing neurogenic atrophy. Muscle fibers are small and angular with no fibrosis. Capillaries (arrows) and arteries (A) are normal. Adjacent nerve (N) has decreased nerve fibers. Massons X200. (Figure shown 80% of original size.)
muscle fibers 2 years after irradiation were 2 I .2 Gy IORT only and 22.9 Gy IORT when combined with 50 Gy EBRT. The doses required to cause a 50% decrease in
muscle fibers 5 years after irradiation were 33.8 Gy IORT only and 25.2 Gy IORT when combined with 50 Gy EBRT (Table 3). The higher doses needed to cause an equivalent effect at 5 years versus 2 years after irradiation may imply that some regeneration of muscle occurred. This process was not observed histologically, but could have occurred between 2 and 5 years. The satellite cells or myoblasts have some limited potential for regeneration (4). A more likely explanation for the increase in percentage of muscle at 5 years compared to 2 years is a condensation of the hemorrhage, fibrin, and immature loose connective tissue into more mature, dense connective tissue. This is evident as the doses to cause a 20% increase in fibrin and hemorrhage were 7 to 14 Gy higher 5 years after irradiation compared to 2 years after irradiation (Table 3). Since the fibrin and hemorrhage were less abundant and connective tissue more dense at a given dose 5 years after irradiation, because of the point count methods used, an apparent corresponding increase in muscle would occur. Some of the remaining muscle fibers may also have undergone hypertrophy, rather than regeneration, as some fibers were subjectively larger at 5 years after irradiation, compared to 2 years after irradiation. Hypertrophy of muscle fibers would also cause a relative increase in the percentage of muscle counted. None of the dogs in this study showed any clinical manifestation of damage to the irradiated muscle. The psoas muscle’s action is to rotate the thigh medially and
to flex the lumbar vertebrae. As these are subtle move-
ments in dogs, their lack of function would likely not be detected. In humans, one of the most preponderant complications following abdominal IORT is that of soft tissue injury. Most soft tissue complications are related to abscesses, hemorrhages, fibrosis, and pelvic pain (1, 2, 17). Two dogs in this study had abscess formation, but these occurred after 5 years after 47.5 Gy plus 50 Gy EBRT. The occasional occurrence of abscesses in humans at the lower doses of 10 to 20 Gy is likely because of the additional trauma caused by the tumor itself or by the surgical procedures performed. Hemorrhage into the muscle did occur in the dogs of this study, and a small amount was present at doses as low as 20 Gy IORT only or 15 Gy IORT when combined with 50 Gy EBRT. A 20% increase in hemorrhage and fibrin occurred at IORT doses of 30 Gy and above. Clinical signs related to this hemorrhage were not apparent. Pelvic pain in humans may be caused by damage to peripheral nerves or due to fibrosis and adhesions that could be caused by the tumor, surgery, or irradiation. Pelvic pain was not noted in the dogs of this study; however, this would be difficult to detect in dogs. In humans examined 1 to 18 months after IORT, mild fibrotic changes were observed in the retroperitoneal soft tissue, and these increased in severity with increasing time after irradiation ( 16). Soft tissue necrosis and fibrosis occurred in young patients treated with IORT for a variety
of pediatric tumors (2). Fibrotic changes also routinely occurred in the retroperitoneal tissues of dogs given IORT at doses of 30 Gy or more in another study (17). The present study confirms these results as the muscle atrophy
470
I. J.
Radiation
Oncology
0 Biology
0 Physics
and fibrosis seen at doses below 20 Gy, as would be used in humans. were mild, and were more consistent and severe at doses of 30 Gy and above. The pathogenesis of radiation-induced muscle injury has been studied to a limited extent. In rat thighs, 4 to 6 hr after a single dose of 15 Gy there was an increase in release of amino acids, suggesting increased protein breakdown caused by metabolic disturbance (15). In rabbits, 24 hr after a single dose of 1 1 to 13 Gy interfibrillar edema, myofilament disruption and endothelial swelling were present. At I week and 1 month there were areas of myofiber necrosis, focal atrophy, mesenchymal cell proliferation, and fibrinoid necrosis of vessels (8). At 3 to 10 months after a single dose of 30 to 55 Gy to rats, variable necrosis of myocytes, fiber atrophy. fiber hypertrophy areas of regeneration, endomysial and perimysial fibrosis, mineralization, and arteriole hyalinization with obliterative intimal proliferation were observed (2 1). Similar lesions of muscle fibrosis, vessel lesions, and inflammation were observed in the muscle of pigs after 40 to 80 Gy given in a single dose (13, 20). These studies suggest that there is an early direct radiation effect on the myocyte causing metabolic abnormalities which could result in cell death (8. 15, 2 1). The later effect. which results in further muscle injury, is more likely caused by vascular lesions resulting in ischemia (8, 13, 20, 2 I ). The fibrosis which occurs may also be related to vessel damage and ischemia. which may be further mediated by inflammation that accompanies the vessel damage (20). The results of the present study support a role for vascular damage in causing late muscle atrophy and fibrosis. as a consistent decrease in the percentage of capillaries was noted with increasing IORT dose at both 2 and 5 years after irradiation. Furthermore, consistent severe vascular injury was seen with an EDs0 of 16.0 to 25.8 Gy IORT only or when combined with 50 Gy EBRT at 2 and 5 years after irradiation. The presence of fibrin and hemorrhage, which was present within the muscle, is also an indication of severe vessel injury that resulted in vessel rupture. Finally, the presence of inflammation that was present in most dogs after IORT supports the hypothesis of mediators of inflammation contributing to late fibrosis as monocytes and macrophages may produce mediators that induce fibroblast recruitment and proliferation (20). The somewhat lower doses needed to cause the given effect of fibrin. hemorrhage, and vessel lesions and more frequent inflammation at 2 years compared to 5 years after irradiation suggest that a portion of the vascular injury and inflammation had occurred by 2 years. However, in the abdominal aorta and large branch arteries, there was a progression of vessel injury between 2 and 5 years (5. 6). This likely reflected a difference in vessel size with injury to the smaller arteries.
March
1991. Volume
20. Number
3
arterioles, and capillaries within the psoas muscle occurring earlier than injury to the larger arteries and abdominal aorta. The severe injury that occurred to the aorta and larger branch arteries at 5 years after irradiation could be responsible for the ongoing vessel lesions, hemorrhage, and fibrin that were seen in the psoas muscle 5 years after irradiation. Neurologic atrophy has been proposed to be a cause of muscle injury after irradiation and may have contributed to the muscle atrophy seen in the dogs in this study. However, the doses required to produce electrophysiologic abnormalities and severe histologic lesions of nerve fiber loss and fibrosis were lower than those required to cause severe muscle injury (10). The EDso for severe vessel lesions in and around nerves (10) were very similar to the EDso for vessel lesions in the psoas muscle. This is to be expected, as these were similarly sized vessels and are likely different sections of the same vessel. Furthermore, neurogenic atrophy of muscle is characterized by small, angulated fibers with relatively little fibrosis (4) and was seen in muscles outside of the radiation field (Fig. 5). In contrast, the irradiated muscle had, in addition to fiber atrophy, abundant fibrosis. vascular lesions, inflammation. hemorrhage, and fibrin formation. These latter lesions would not be found if neurogenic atrophy were the sole cause for the muscle injury seen in this study. In comparing the difference between IORT and IORT combined with 50 Gy EBRT in causing a given level of effect. the 50 Gy EBRT in 2 Gy fractions resulted in changes in the isoeffect values ranging from approximately 0 to 10 Gy (Table 3). The lack of an effect of 50 Gy EBRT in causing some lesions may be because the large single IORT dose obscured the relatively smaller contribution of the 50 Gy EBRT given in 2 Gy fractions. In either case, the IORT dose was responsible for producing the majority of the lesions seen in the psoas muscle. At both 2 and 5 years after irradiation, dogs receiving 50 to 80 Gy EBRT only in 2 to 2.67 Gy fractions had either no or only minimal lesions. In summary, this study demonstrated a dose response for muscle atrophy and fibrosis at 2 and 5 years after large single doses. Muscle was resistant to fractionated irradiation up to 80 Gy. Radiation-induced muscle atrophy was associated with significant fibrosis and vascular lesions, which distinguishes this muscle injury from that caused solely by neurogenic atrophy. The lesions produced were largely a function of the single IORT dose rather than EBRT given in fractions. In general, 20 to 25 Gy in combination with 50 Gy EBRT was reasonably well-tolerated by the sublumbar musculature. Doses above 25 Gy IORT could lead to significant muscle fibrosis and complications associated with soft tissue injury.
REFERENCES 1. Calve, F. A.; Algarra. S. M.; Azinovic, 1.: Santos, M.; Escude. L.; Hemidez,
J. L.; Zornoza,
G.: Voltas. J. Intraoperative
radiotherapy for recurrent and/or residual colorectal cancer. Radiother. Oncol. 1.5:133-140: 1989.
Muscle injury following experimental IORT 0 B. E. POWERSet al. 2. Calvo. F. A.; Sierrasesumaga, L.; Martin, I.; Santos, M.: Voltas, J.; Barian, J. M.; Canadell, J. Intraoperative radiotherapy in the multidisciplinary treatment of pediatric tumors. Acta. Oncol. 28:257-260; 1989. 3. Chalkley. H. W. Methods for quantitative morphologic analysis of tissue. JNCI 4:47-53; 1943. 4. Cotran, R. S.; Kumar, V.; Robbins, S. L. Diseases of muscle. In: Cotran, R. S., Kumar, V., Robbins, S. L., eds. Robbins pathologic basis of disease, 4th edition. Philadelphia: W. B. Saunders, Co.; 1989:1363-1371. 5. Gillette, E. L.; Powers, B. E.; McChesney, S. L.; Park, R. D.; Withrow. S. J. Response of aorta and branch arteries to experimental intraoperative irradiation. Int. J. Radiat. Oncol. Biol. Phys. 17: 1247- 1255; 1989. 6. Gillette, E. L.; Powers, B. E.; McChesney, S. L.: Withrow, S. J. Aortic wall injury following intraoperative irradiation. Int. J. Radiat. Oncol. Biol. Phys. 15:1401-1406; 1988. 7. Hoekstra. H. J.; Sindelar, W. F.; Kinsella, T. J. Surgery with intraoperative radiotherapy for sarcomas of the pelvic girdle: a pilot experience. Int. J. Radiat. Oncol. Biol. Phys. 15: 1013-1016: 1988. 8. Khan, M. Y. Radiation-induced changes in skeletal muscle: An electron microscopic study. J. Neuropathol. Exp. Neurol. 33:42-57: 1974. 9. Kinsella, T. J.: Sindelar. W. F.; DeLuca, A. M.; Pezeshkpour, G.; Smith, R.; Maher. M.; Terre]], R.; Miller, R.; Mixon, A.; Harwell. J. F.; Rosenberg, S. A.; Glatstein, E. Tolerance of peripheral nerve to intraoperative radiotherapy (IORT): clinical and experimental studies. Int. J. Radiat. Oncol. Biol. Phys. 11:1579-1585; 1985. 10. LeCouteur. R. A.: Gillette, E. L.; Powers, B. E.; Child, G.; McChesney. S. L.; Ingram, J. T. Peripheral neuropathies following experimental intraoperative radiation therapy (IORT). Int. J. Radiat. Oncol. Biol. Phys. 17:583-590; 1989. I 1. McChesney Gillette, S. L.; Gillette, E. L.: Powers, B. E.: Park, R. D.; Withrow. S. J. Ureteral injury following ex-
12.
13.
14. 15.
16.
17.
18.
19.
20.
2 1.
471
perimental intraoperative radiation. Int. J. Radiat. Oncol. Biol. Phys. 17:791-798; 1989. Powers, B. E.; Gillette, E. L.; McChesney, S. L.; LeCouteur, R. A.; Withrow, S. J. Bone necrosis and tumor induction following experimental intraoperative irradiation. Int. J. Radiat. Oncol. Biol. Phys. 17:559-567; 1989. Remy, J.; Martin, M.; Lefaix, J. L.; Daburon, F. Radiation induced fibrosis in pig muscle: pathological and cellular observations. Brit. J. Cancer 53(Suppl. VII):232-233; 1986. Rich, T. A. Intraoperative radiotherapy. Radiother. Oncol. 6:207-22 I ; 1986. Schwenen, M.; Altman, K. 1.; Schroder, W. Radiation-induced increase in the release of amino acids by isolated, perfused skeletal muscle. Int. J. Radiat. Biol. 55:257-269; 1989. Sindelar. W. F.; Hoekstra, H.; Restrepo, C.; Kinsella, T. J. Pathological tissue changes following intraoperative radiotherapy. Am. J. Clin. Oncol. 9:504-509; 1986. Sindelar, W. F.; Tepper, J.; Travis, E. L.; Terrill, R. Tolerance of retroperitoneal structures to intraoperative radiation. Ann. Surg. 196:601-608; 1982. Tepper. J. E.: Gunderson. L. L.; Orlow, E.: Cohen, A. M.; Hedberg. S. E.; Shipley, W. U.; Blitzer, P. H.: Rich, T. Complications of intraoperative radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 10: 183 l-l 839; 1984. Tepper, J. E.; Wood, W. C.; Cohen, A. M. Treatment of locally advanced rectal cancer with external beam radiation, surgical resection, and intraoperative radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 16:1437-1444; 1989. Wegrowski. J.; Lafuma, C.; Lefaix, J. L.; Daburon, F.; Robert, L. Modification of collagen and noncollagenous proteins in radiation-induced muscular fibrosis. Exper. Molecular Pathol. 48:273-285; 1988. Zeman. W.; Solomon, M. Effects of radiation on striated muscle. In: Berdjis, C. C., ed. Pathology of irradiation. Baltimore; Williams and Wilkins, Co.: I7 l-l 85; 197 I.
Inr J Radiorion Oncology Bml Phys Vol. 20. pp. 463-47 Printed I” the U.S.A. All nghts resend
0360-3016/91 $3.00 + .OO Copyright Q 1991 Pergamon Press plc
??Original Contribution
MUSCLE
INJURY FOLLOWING EXPERIMENTAL INTRAOPERATIVE IRRADIATION
B. E. POWERS, D.V.M., S. L. MCCHESNEY
GILLETTE,
PH.D.,* M.S.,
E. L. GILLETTE,
PH.D.,*
AND S. J. WITHROW,
D.V.M.,
R. A. LECOUTEUR,
PH.D.,* BVSc,
PH.D.+
D.V.M.+
Colorado State University, Fort Collins, CO 80523 The paraaortic region of beagle dogs was irradiated to 15 to 55 Gy intraoperative irradiation, 10 to 47.5 Gy intraoperative irradiation following 50 Gy external beam irradiation in 25 fractions, or 50 to 80 Gy external beam irradiation in 30 fractions. Six MeV electrons were used for intraoperative irradiation, and external beam irradiation was done using photons from a 6 MV linear accelerator. The psoas muscle in the irradiation field was examined histomorphometrically 2 or 5 years after irradiation. The percentage of muscle fibers and capillaries decreased, whereas the percentage of connective tissue increased with increased dose for both intraoperative irradiation only and intraoperative irradiation plus external beam irradiation. The dose causing a 50% decrease in the percentage of muscle fibers was 21.2 Gy and 33.8 Gy at 2 and 5 years, respectively, after intraoperative irradiation alone, and 22.9 Gy and 25.2 Gy at 2 and 5 years, respectively, after intraoperative irradiation combined with 50 Gy external beam irradiation. The EDa for severe vessel lesions was 19.2 Gy and 25.8 Gy at 2 and 5 years, respectively, after intraoperative irradiation alone and 16.0 Gy and 18.0 Gy at 2 and 5 years, respectively, after intraoperative irradiation combined with 50 Gy external beam irradiation. External beam irradiation alone caused a slight decrease in percentage of muscle fibers with increased dose, and vessel lesions were infrequent or mild. Radiation-induced muscle injury was characterized by loss of muscle fibers, decreased fiber size, severe vessel lesions, hemorrhage, inflammation, coagulation necrosis, and fibrosis. These histopathologic characteristics distinguish this muscle injury from that caused by neurogenic atrophy. These data indicate that radiation-induced muscle injury most likely was caused by injury of the supporting vasculature. The lesions produced were largely a function of the single intraoperative dose rather than the external beam fractionated doses. Furthermore, it appears that 20 to 25 Gy intraoperative irradiation combined with 50 Gy external beam irradiation may be near the maximum tolerated dose by sublumbar musculature and its supporting vasculature. lntraoperative
irradiation, Canine, Late effects, Muscle.
tumors (2, 7, 14). The advantage of IORT is that the ra-
INTRODUCTION
diation field can be directly visualized, and sensitive structures such as intestines and kidneys can be manually retracted from the field. Furthermore, the volume of tissue irradiated can be limited by the use of electrons which, depending on their energy, have limited penetration. Complications following IORT in humans are reported to be not significantly different from those following more conventional external beam irradiation (EBRT) (18).
Intraoperative irradiation (IORT) is being used with increasing frequency for the treatment of various malignancies in humans (14). Clinical results for the treatment of rectal cancer appear to be especially promising ( 14, 19). Other malignancies being treated by IORT include gastric, pancreatic, urinary bladder, biliary, retroperitoneal, prostatic, paraaortic nodal, and a variety of pediatric
Presented in part at the 3 1st Meeting of the American Society of Therapeutic Radiology and Oncology in San Francisco, Oc-
vided medical and surgical support. Special thanks to D. Madden for assistance at necropsy and the histopathology laboratory for histology support. B. A. Ensley, K. T. Crump, and M. M. Estabrook who provided technical assistance are acknowledged, as is Mrs. L. A. Ludwig for preparation of the manuscript. We also wish to thank Dr. T. B. Borak and R. J. Scott for radiation physics and engineering support. This investigation was supported by PHS grant number CA29 117 awarded by the National Cancer Institute, DHHS. Accepted for publication 12 September 1990.
tober 1989. * Department of Radiology and Radiation Biology. + Department of Clinical Sciences. Reprint requests to: Barbara E. Powers, D.V.M., Ph.D., Comparative Oncology Unit, Colorado State University, Fort Collins, CO 80523. Acknowledgements-The authors wish to thank Drs. Herman D. Suit and Joel Tepper for contributing to the design of the experiment and for continuing consultation. Dr. S. L. Allen pro-
463
464
I. J. Radiation
Oncology
0
Biology 0 Physics
Some complications reported include fibrosis (2, 16, 18). hemorrhage ( 18), peripheral neuropathies (2.9). bone necrosis (7), and pelvic pain (18). Most patients have been followed for a relatively short period of time, with few patients observed over 2 years after irradiation. Numerous studies in experimental dogs have been done to determine the effects of IORT on various normal tissues in the radiation field. Many of these studies had relatively short periods of observation or involved few animals at specific doses and time points. More recently, studies have been completed describing the effect of IORT or IORT combined with EBRT 2 and 5 years after irradiation of the paraaortic area in dogs. Lesions that were reported included aortic thromboses and aneurysms, branch artery injury (5, 6), peripheral neuropathies (IO), ureteral strictures ( 11) bone necrosis, and tumor induction ( 12). The purpose of this report is to describe injury to the sublumbar muscle that was included in the paraaortic radiation field of dogs 2 and 5 years after IORT and IORT combined with EBRT.
METHODS Experimenlal
AND
MATERIALS
March
I,
199
Volume
20. Number
Table 2. Experimental
Group IORT only 17.5 25.0 32.5 40.0 47.5 55.0 IORT plus IORT dose 10.0 17.5 25.0 32.5 40.0 47.5 EBRT only 60.0 70.0 80.0 Controls
3
design for 5-year studv
No. of dogs originally in group
No. of dogs alive at least 4 years after irradiation
(Gy) 5 5 6 6 2 50 Gy EBRT (Gy) 5
5 4 4 4
(Gy) 5 6
5 5 4
design
One hundred seventy-seven young adult beagle dogs of both sexes were randomized to receive various radiation doses of IORT, IORT combined with 50 Gy EBRT, or EBRT only to the paraaortic area (Tables 1 and 2). Ninety dogs that were to be observed for 2 years after irradiation received 15 to 50 Gy IORT only; 10 to 42.5 Gy IORT combined with 50 Gy EBRT in 2 Gy fractions; or 50, 60,
Table I. Experimental
Group
design for 2-year study No. of dogs originally in group
No. of dogs alive at 2 years after irradiation
5 5 5 5 5 5 5
5 5 4 4 2 3 2
5 5 5 5 5 5
5 5 5 5 5 5
4 6 5 5 5
4 6 5 4 5
IORT only (Gy)
15.0 20.0 25.0 30.0 35.0 42.5 50.0 IORT plus 50 Gy EBRT IORT dose (Gy) 10.0 15.0 20.0 21.5 35.0 42.5 EBRT only (Gy) 50.0 60.0 70.0 80.0 Controls
70, or 80 Gy EBRT only in 30 fractions. Four to six dogs were in each dose group, and five sham irradiated dogs served as controls (Table 1). Eighty-seven dogs that were to be observed for 5 years after irradiation received 17.5 to 55 Gy IORT only: 10 to 47.5 Gy combined with 50 Gy EBRT in 2 Gy fractions: or 60. 70, or 80 Gy EBRT only in 30 fractions. Five or six dogs were in each dose group, and seven age-matched non-irradiated dogs served as controls (Table 2). Irradiations
For IORT, dogs were anesthetized, placed in dorsal recumbency, and had a celiotomy performed under sterile conditions as previously described (5. 6, IO, 1 1, 12). A 5 X 8 cm plexiglass cone was placed over the paraaortic region. and the intestines, kidney, and urinary bladder were retracted from the field. In the 2-year study, the left ureter was also retracted, whereas it was included in ‘the field for the j-year study. Other tissues included in the held were the aorta, vena cava, lumbar nerves, ventral cortex of lumbar vertebrae 3-4 through 7, and the left psoas muscle. Radiation was delivered with 6 MeV electrons at a rate of 6.6 Gy/min and an SSD of 125 cm. The D mal was 15 mm. Doses were calculated such that irradiated tissues of interest were within 90% of the prescribed dose. For dogs receiving EBRT in addition to IORT, the 50 Gy was delivered in 2 Gy fractions over a j-week period immediately prior to IORT. The EBRT field was 5 X 10 cm and included the aorta, vena cava, lumbar nerves, both ureters, a portion of the left kidney, ventral cortex of lumbar vertebrae 2 through 7, and the psoas muscle.
Muscle injury following experimental IORT 0 B. E. POWERSet al.
For EBRT, dogs were tranquilized and restrained in a plexiglass box in a standing position. Irradiation was with photons from a 6 MV linear accelerator. The dose rate was 2.5 Gy/min and the SSD was 100 cm. Doses were calculated to the dorsal ventral midline, and the psoas muscle was within the 90% isodose level. Dogs receiving EBRT only received 50 to 80 Gy in 30 equal fractions over a 6-week period. Radiation was delivered as described above.
Dogs were observed daily for any adverse signs. Some dogs were euthanatized or died prior to scheduled necropsy date (Tables 1 and 2). At 2 or 5 years after irradiation, dogs were anesthetized, heparinized, and given an overdose of barbiturates. Immediately dogs were perfused through the left ventricle of the heart with the aid of a perfusion pump set at physiologic pressure. The perfusion solution consisted of 4 liters of 1.0% paraformaldehyde and 1.25% glutaraldehyde followed by 4 liters of 4.0% paraformaldehyde and 5.0% glutaraldehyde. Tissues, including the psoas muscles, were dissected from the dog, labeled, and placed in 10% neutral buffered formalin. Transverse sections of the left psoas muscle were embedded in paraffin, sectioned at 5 pm, and stained with hematoxylin and eosin and Masson’s trichrome. Three sections, spaced 50 pm apart, were made of each muscle. Histomorphometry was done using a 36 point ocular grid in a microscope (3). A total of 500 points were counted for each of the three sections and recorded as to whether the grid line intersection covered muscle, connective tissue, capillaries, larger vessels, fat (Fig. 1A) or hemorrhage and fibrin. Lesions in arteries and arterioles were scored on a scale of 0 to 3 where grade 0 was no lesions, grade 1 was mild fibrosis and/or mild intimal proliferation, grade 2 was marked fibrosis and/or intimal proliferation, and grade 3 was thrombosis, disruption, or medial necrosis. Inflammation was noted as being present or not. All slides were examined without prior knowledge of treatment received. Statistical analysis
Only dogs that survived at least 20 months in the 2year study or at least 4 years in the 5-year study were included for statistical analysis (Tables 1 and 2). From histomorphometry, the percentage of each tissue component was calculated by dividing the number of “hits” on a tissue of interest over the total number of points counted. The mean and standard error of the mean (SEM) were calculated for each dose group and plotted against dose. Linear regression analysis was used to determine the dose response and r values and confidence intervals were calculated. For graded vessel lesions, a quanta1 response was determined and grades of 2 or 3 were counted as positive. The quanta1 data were plotted against dose, and using probit analysis, the dose to cause the lesion in
465
50% of the dogs, EDso, and confidence intervals were calculated.
RESULTS Two years after irradiation
In the study designed for a 2-year observation period, 11 dogs died or were euthanatized prior to 2 years, primarily because of peripheral neuropathies or spinal cord damage (Table 1) ( 10). These dogs were not included in analysis of the psoas muscle. At the higher IORT doses 2 years after irradiation, the psoas muscle was grossly smaller than normal, firm, pale, and adherent to surrounding tissues and bone. Histologically, there were focal areas of increased connective tissue associated with the loss or shrinkage of muscle fibers. At higher radiation doses, large areas of loss of muscle were present, and scattered muscle fibers were either swollen and hypereosinophilic or smaller than normal. Fibrous connective tissue or masses of hemorrhage and fibrin replaced areas of muscle fiber loss (Fig. 1B and 2). Histomorphometry revealed that with increasing dose there was a decrease in the percentage of muscle fibers (? = 0.70), an increase in the percentage of connective tissue (? = 0.67) a decrease in the percentage of capillaries (3 = 0.64), and an increase in the percentage of hemorrhage and fibrin (? = 0.75) after IORT only (Fig. 3). Similarly, after IORT combined with 50 Gy EBRT with increasing IORT dose, there was a decrease in the percentage of muscle fibers (? = 0.88), an increase in the percentage of connective tissue (? = 0.79), a decrease in the percentage of capillaries (? = 0.87), and an increase in the percentage of hemorrhage and fibrin (? = 0.87) (Fig. 3). The dose causing a 50% decrease in the percentage of muscle fibers from a normal of 84.0% was 2 1.2 Gy (confidence intervals could not be calculated) IORT only and 22.9 Gy (I 5.228.6,95% C.I.) IORT when combined with 50 Gy EBRT (Table 3). The dose which caused a 20% increase in the percentage of hemorrhage and fibrin was 38.8 Gy (30.559.0, 95% C.I.) IORT only and 30.8 Gy (24.5-42.5, 95% C.I.) IORT when combined with 50 Gy EBRT (Table 3). After EBRT only there was a minimal increase in percentage of connective tissue with increasing dose (9 = 0.77) but other parameters did not change significantly. Vessel lesions were present 2 years after irradiation and consisted of mild perivascular fibrosis and mild intimal proliferation (grade 1 lesion) at lower doses. At higher doses, severe perivascular fibrosis and severe intimal proliferation (grade 2 lesion) or severe hyalinization, medial necrosis, and disruption of the vessel walls with thrombosis (grade 3 lesion) was present in some arteries and arterioles (Fig. 2). The EDso for grade 2 or 3 vessel lesions was 19.2 Gy (17.5-2 1.1 95% CI) IORT only and 16.0 Gy (12.620.3 95% CI) IORT when combined with 50 Gy EBRT (Table 3). Grade 2 or 3 vessel lesions were not seen in any dogs that received EBRT only.
466
1. J. Radiation
Oncology
0 Biology
0 Physics
March
1991, Volume
20, Number
3
Fig. l.(A) Normal muscle with large fibers (M), capillaries (arrows), and fat (F). Massons X200. (B) Muscle from a dog 5 years after 25 Gy IORT plus 50 Gy EBRT. Muscle fibers (M) with markedly increased fibrous connective tissue (F). Massons X200. (Figures shown 75% of original size.)
Inflammation, characterized by infiltrates of lymphocytes, plasma cells, and a few macrophages was present and predominately perivascular in location. Inflammation was present in 79% of the dogs receiving IORT only, mostly at doses of 20 Gy and above and in 73% of the dogs receiving IORT with 50 Gy EBRT, mostly at doses of 15 Gy IORT and above. Inflammation was not present in dogs receiving EBRT only.
Five years ajer irradiation In the study designed to evaluate dogs for 5 years, 13 dogs died or were euthanatized prior to 4 years, primarily because of peripheral nerve or ureteral ( 11) damage (Table
2). These dogs were not included in analysis of the psoas muscle. At 5 years after irradiation at the higher doses, the psoas muscle was grossly small, firm, pale, and adherent to surrounding tissues and bone. In two dogs that received 47.5 Gy IORT with 50 Gy EBRT, the muscle was abscessed and consisted of a firm connective tissue capsule filled with thick, foul-smelling, tan-red exudate. In three dogs that received 50 Gy EBRT and 25, 40, and 47.5 Gy IORT, the muscle was replaced by a firm gritty tumor ( 12). Histologically, lesions were similar to those observed at 2 years after irradiation, but swollen eosinophilic fibers were less frequent. Connective tissue appeared more dense (Fig. 1B) with occasional mineral formation,
Muscle injury following experimental IORT 0 B. E.
POWERS
et al.
Fig. 2. Muscle from a dog 2 years after 27.5 Gy IORT plus 50 Gy EBRT. Muscle replaced by fibrous connective tissue (F), edema (E), and hemorrhage (arrows). A thrombosed artery (A) is also present. Massons X 100. (Figure shown 80% of original size.)
remaining muscle fibers appeared larger, and inflammation, hemorrhage, and fibrin was less frequent than at 2 years after irradiation. The two dogs with abscess formation had pockets of neutrophils, masses of bacteria, total loss of muscle fibers, and fibrous tissue replacement. The three dogs with tumor had a complete replacement of the muscle by the tumor that were osteosarcomas (12); these three dogs were not included in histomorphometric data. Histomorphometrically, with increasing dose there was a decrease in the percentage of muscle (? = 0.93), an increase in the percentage of connective tissue (9 = 0.79)
90
r
0 15
a decrease in the percentage of capillaries (? = 0.94) and an increase in the percentage of hemorrhage and fibrin (r’ = 0.86) after IORT only (Fig. 4). Similarly, after IORT combined with 50 Gy EBRT, with increasing dose there was a decrease in the percentage of muscle (? = 0.95) an increase in the percentage ofconnective tissue (? = 0.78), a decrease in the percentage of capillaries (f = 0.92), and an increase in the percentage of hemorrhage and fibrin (i = 0.67) (Fig. 4). The dose which caused a 50% decrease in the percentage of muscle from a normal of 80.4% was 33.8 Gy (27.838.6, 95% C.I.) IORT only and 25.2 Gy (20.4-29.2, 95%
EBRT
IORT
20
25
30
35
40
45 DOSE
Fig. 3. Graph of percentage of tissue components Control values are open symbols.
+ IORT
(Gy)
versus dose 2 years after irradiation.
Mean and SEM are shown.
468
I.
J.
Radiation
Oncology 0 Biology 0 Physics
March 1991. Volume 20. Number 3
Table 3. Doses causing muscle and vessel lesions Time (years)
Lesion 50% decrease
IORT (Gy)
IORT plus 50 Gy EBRT (Gy)
Difference (GY)
in muscle
2 5
2 I .2 (-) 33.8 (27.8-38.6)*
22.9 (15.2-28.6) 25.2 (20.4-29.2)
8.6
20% increase in fibrin and hemorrhage
2 5
38.8 (30.5-59.0) 45.5 (38.5-59.8)
30.8 (24.5-42.5) 44.5 (-)
8.0
2 5
19.2 (17.5-21.1) 25.8 (23.8-28.1)
16.0 (12.6-20.3) 18.0 (16.2-20.0)
3.2 7.8
EDso for grade 2 & 3 vessel lesions
* 95% confidence
I.0
interval.
Neurogenic muscle atrophy
C.I.) IORT when combined with 50 Gy EBRT (Table 3). The dose which caused a 20% increase in the percentage of hemorrhage and fibrin was 45.5 Gy (38.5-59.8, 95% C.I.) IORT only and 44.5 Gy (confidence intervals could not be calculated) IORT when combined with 50 Gy EBRT (Table 3). After EBRT only there was a mild decrease in percentage of muscle fibers with increasing dose (8 = 0.92), but the decrease was only to 76.5% muscle. There was no significant change in the other parameters. Vessel lesions were present 5 years after irradiation and were similar in type to those observed 2 years after irradiation. The EDso for grade 2 or 3 vessel lesions was 25.8 Gy (23.8-28.1 95% CI) IORT only and 18.0 Gy ( 16.220.0 95% CI) IORT when combined with 50 Gy EBRT (Table 3). One dog receiving 80 Gy EBRT had a grade 2 vessel lesion. Chronic inflammation was similar to that observed 2 years after irradiation but was less frequent. Inflammation was present in 39% of the dogs receiving IORT only, mostly at doses of 40 Gy IORT and above, and in 54% of the dogs receiving IORT with 50 Gy EBRT. mostly at doses of 32.5 Gy IORT and above. Inflammation was not present in dogs receiving EBRT only.
90
I
0
15
At both 2 and 5 years after irradiation, many dogs had clinical peripheral neuropathies caused by radiation damage of the lumbar nerves (10). The quadriceps muscles, which were outside of the radiation field but supplied by those nerves, were small, firm, and pale. Histologically, the muscle fibers were small and angular, but there was no significant increase in connective tissue, no vessel lesions nor hemorrhage or fibrin, and no inflammation (Fig. 5). DI!XXJSSION
This study was designed to determine the late effects of irradiation of normal tissues in the paraaortic region of dogs after IORT, IORT combined with EBRT. and EBRT only. Some of the late effects that have occurred in this study include aortic thrombosis and aneurysms (5, 6). peripheral neuropathies (lo), ureteral strictures ( 1 I), bone necrosis, and tumor induction (12). The left psoas muscle, which was also included in the radiation field, showed injury characterized by loss of muscle fibers, atrophy. and fibrosis. The doses to cause a 50% decrease in
EBRT+
IORT
20
25
30
35
40
Fig. 4. Graph of percentage of tissue components Control values are open symbols.
45
50
55
60
DOSE
(Gy)
65
70
75
IORT
80
85
versus dose 5 years after irradiation.
90
95
100
Mean and SEM are
Muscle injury following experimental IORT 0 B. E. POWERSef al.
469
Fig. 5. Quadriceps muscle out of the field of irradiation showing neurogenic atrophy. Muscle fibers are small and angular with no fibrosis. Capillaries (arrows) and arteries (A) are normal. Adjacent nerve (N) has decreased nerve fibers. Massons X200. (Figure shown 80% of original size.)
muscle fibers 2 years after irradiation were 2 I .2 Gy IORT only and 22.9 Gy IORT when combined with 50 Gy EBRT. The doses required to cause a 50% decrease in
muscle fibers 5 years after irradiation were 33.8 Gy IORT only and 25.2 Gy IORT when combined with 50 Gy EBRT (Table 3). The higher doses needed to cause an equivalent effect at 5 years versus 2 years after irradiation may imply that some regeneration of muscle occurred. This process was not observed histologically, but could have occurred between 2 and 5 years. The satellite cells or myoblasts have some limited potential for regeneration (4). A more likely explanation for the increase in percentage of muscle at 5 years compared to 2 years is a condensation of the hemorrhage, fibrin, and immature loose connective tissue into more mature, dense connective tissue. This is evident as the doses to cause a 20% increase in fibrin and hemorrhage were 7 to 14 Gy higher 5 years after irradiation compared to 2 years after irradiation (Table 3). Since the fibrin and hemorrhage were less abundant and connective tissue more dense at a given dose 5 years after irradiation, because of the point count methods used, an apparent corresponding increase in muscle would occur. Some of the remaining muscle fibers may also have undergone hypertrophy, rather than regeneration, as some fibers were subjectively larger at 5 years after irradiation, compared to 2 years after irradiation. Hypertrophy of muscle fibers would also cause a relative increase in the percentage of muscle counted. None of the dogs in this study showed any clinical manifestation of damage to the irradiated muscle. The psoas muscle’s action is to rotate the thigh medially and
to flex the lumbar vertebrae. As these are subtle move-
ments in dogs, their lack of function would likely not be detected. In humans, one of the most preponderant complications following abdominal IORT is that of soft tissue injury. Most soft tissue complications are related to abscesses, hemorrhages, fibrosis, and pelvic pain (1, 2, 17). Two dogs in this study had abscess formation, but these occurred after 5 years after 47.5 Gy plus 50 Gy EBRT. The occasional occurrence of abscesses in humans at the lower doses of 10 to 20 Gy is likely because of the additional trauma caused by the tumor itself or by the surgical procedures performed. Hemorrhage into the muscle did occur in the dogs of this study, and a small amount was present at doses as low as 20 Gy IORT only or 15 Gy IORT when combined with 50 Gy EBRT. A 20% increase in hemorrhage and fibrin occurred at IORT doses of 30 Gy and above. Clinical signs related to this hemorrhage were not apparent. Pelvic pain in humans may be caused by damage to peripheral nerves or due to fibrosis and adhesions that could be caused by the tumor, surgery, or irradiation. Pelvic pain was not noted in the dogs of this study; however, this would be difficult to detect in dogs. In humans examined 1 to 18 months after IORT, mild fibrotic changes were observed in the retroperitoneal soft tissue, and these increased in severity with increasing time after irradiation ( 16). Soft tissue necrosis and fibrosis occurred in young patients treated with IORT for a variety
of pediatric tumors (2). Fibrotic changes also routinely occurred in the retroperitoneal tissues of dogs given IORT at doses of 30 Gy or more in another study (17). The present study confirms these results as the muscle atrophy
470
I. J.
Radiation
Oncology
0 Biology
0 Physics
and fibrosis seen at doses below 20 Gy, as would be used in humans. were mild, and were more consistent and severe at doses of 30 Gy and above. The pathogenesis of radiation-induced muscle injury has been studied to a limited extent. In rat thighs, 4 to 6 hr after a single dose of 15 Gy there was an increase in release of amino acids, suggesting increased protein breakdown caused by metabolic disturbance (15). In rabbits, 24 hr after a single dose of 1 1 to 13 Gy interfibrillar edema, myofilament disruption and endothelial swelling were present. At I week and 1 month there were areas of myofiber necrosis, focal atrophy, mesenchymal cell proliferation, and fibrinoid necrosis of vessels (8). At 3 to 10 months after a single dose of 30 to 55 Gy to rats, variable necrosis of myocytes, fiber atrophy. fiber hypertrophy areas of regeneration, endomysial and perimysial fibrosis, mineralization, and arteriole hyalinization with obliterative intimal proliferation were observed (2 1). Similar lesions of muscle fibrosis, vessel lesions, and inflammation were observed in the muscle of pigs after 40 to 80 Gy given in a single dose (13, 20). These studies suggest that there is an early direct radiation effect on the myocyte causing metabolic abnormalities which could result in cell death (8. 15, 2 1). The later effect. which results in further muscle injury, is more likely caused by vascular lesions resulting in ischemia (8, 13, 20, 2 I ). The fibrosis which occurs may also be related to vessel damage and ischemia. which may be further mediated by inflammation that accompanies the vessel damage (20). The results of the present study support a role for vascular damage in causing late muscle atrophy and fibrosis. as a consistent decrease in the percentage of capillaries was noted with increasing IORT dose at both 2 and 5 years after irradiation. Furthermore, consistent severe vascular injury was seen with an EDs0 of 16.0 to 25.8 Gy IORT only or when combined with 50 Gy EBRT at 2 and 5 years after irradiation. The presence of fibrin and hemorrhage, which was present within the muscle, is also an indication of severe vessel injury that resulted in vessel rupture. Finally, the presence of inflammation that was present in most dogs after IORT supports the hypothesis of mediators of inflammation contributing to late fibrosis as monocytes and macrophages may produce mediators that induce fibroblast recruitment and proliferation (20). The somewhat lower doses needed to cause the given effect of fibrin. hemorrhage, and vessel lesions and more frequent inflammation at 2 years compared to 5 years after irradiation suggest that a portion of the vascular injury and inflammation had occurred by 2 years. However, in the abdominal aorta and large branch arteries, there was a progression of vessel injury between 2 and 5 years (5. 6). This likely reflected a difference in vessel size with injury to the smaller arteries.
March
1991. Volume
20. Number
3
arterioles, and capillaries within the psoas muscle occurring earlier than injury to the larger arteries and abdominal aorta. The severe injury that occurred to the aorta and larger branch arteries at 5 years after irradiation could be responsible for the ongoing vessel lesions, hemorrhage, and fibrin that were seen in the psoas muscle 5 years after irradiation. Neurologic atrophy has been proposed to be a cause of muscle injury after irradiation and may have contributed to the muscle atrophy seen in the dogs in this study. However, the doses required to produce electrophysiologic abnormalities and severe histologic lesions of nerve fiber loss and fibrosis were lower than those required to cause severe muscle injury (10). The EDso for severe vessel lesions in and around nerves (10) were very similar to the EDso for vessel lesions in the psoas muscle. This is to be expected, as these were similarly sized vessels and are likely different sections of the same vessel. Furthermore, neurogenic atrophy of muscle is characterized by small, angulated fibers with relatively little fibrosis (4) and was seen in muscles outside of the radiation field (Fig. 5). In contrast, the irradiated muscle had, in addition to fiber atrophy, abundant fibrosis. vascular lesions, inflammation. hemorrhage, and fibrin formation. These latter lesions would not be found if neurogenic atrophy were the sole cause for the muscle injury seen in this study. In comparing the difference between IORT and IORT combined with 50 Gy EBRT in causing a given level of effect. the 50 Gy EBRT in 2 Gy fractions resulted in changes in the isoeffect values ranging from approximately 0 to 10 Gy (Table 3). The lack of an effect of 50 Gy EBRT in causing some lesions may be because the large single IORT dose obscured the relatively smaller contribution of the 50 Gy EBRT given in 2 Gy fractions. In either case, the IORT dose was responsible for producing the majority of the lesions seen in the psoas muscle. At both 2 and 5 years after irradiation, dogs receiving 50 to 80 Gy EBRT only in 2 to 2.67 Gy fractions had either no or only minimal lesions. In summary, this study demonstrated a dose response for muscle atrophy and fibrosis at 2 and 5 years after large single doses. Muscle was resistant to fractionated irradiation up to 80 Gy. Radiation-induced muscle atrophy was associated with significant fibrosis and vascular lesions, which distinguishes this muscle injury from that caused solely by neurogenic atrophy. The lesions produced were largely a function of the single IORT dose rather than EBRT given in fractions. In general, 20 to 25 Gy in combination with 50 Gy EBRT was reasonably well-tolerated by the sublumbar musculature. Doses above 25 Gy IORT could lead to significant muscle fibrosis and complications associated with soft tissue injury.
REFERENCES 1. Calve, F. A.; Algarra. S. M.; Azinovic, 1.: Santos, M.; Escude. L.; Hemidez,
J. L.; Zornoza,
G.: Voltas. J. Intraoperative
radiotherapy for recurrent and/or residual colorectal cancer. Radiother. Oncol. 1.5:133-140: 1989.
Muscle injury following experimental IORT 0 B. E. POWERSet al. 2. Calvo. F. A.; Sierrasesumaga, L.; Martin, I.; Santos, M.: Voltas, J.; Barian, J. M.; Canadell, J. Intraoperative radiotherapy in the multidisciplinary treatment of pediatric tumors. Acta. Oncol. 28:257-260; 1989. 3. Chalkley. H. W. Methods for quantitative morphologic analysis of tissue. JNCI 4:47-53; 1943. 4. Cotran, R. S.; Kumar, V.; Robbins, S. L. Diseases of muscle. In: Cotran, R. S., Kumar, V., Robbins, S. L., eds. Robbins pathologic basis of disease, 4th edition. Philadelphia: W. B. Saunders, Co.; 1989:1363-1371. 5. Gillette, E. L.; Powers, B. E.; McChesney, S. L.; Park, R. D.; Withrow. S. J. Response of aorta and branch arteries to experimental intraoperative irradiation. Int. J. Radiat. Oncol. Biol. Phys. 17: 1247- 1255; 1989. 6. Gillette, E. L.; Powers, B. E.; McChesney, S. L.: Withrow, S. J. Aortic wall injury following intraoperative irradiation. Int. J. Radiat. Oncol. Biol. Phys. 15:1401-1406; 1988. 7. Hoekstra. H. J.; Sindelar, W. F.; Kinsella, T. J. Surgery with intraoperative radiotherapy for sarcomas of the pelvic girdle: a pilot experience. Int. J. Radiat. Oncol. Biol. Phys. 15: 1013-1016: 1988. 8. Khan, M. Y. Radiation-induced changes in skeletal muscle: An electron microscopic study. J. Neuropathol. Exp. Neurol. 33:42-57: 1974. 9. Kinsella, T. J.: Sindelar. W. F.; DeLuca, A. M.; Pezeshkpour, G.; Smith, R.; Maher. M.; Terre]], R.; Miller, R.; Mixon, A.; Harwell. J. F.; Rosenberg, S. A.; Glatstein, E. Tolerance of peripheral nerve to intraoperative radiotherapy (IORT): clinical and experimental studies. Int. J. Radiat. Oncol. Biol. Phys. 11:1579-1585; 1985. 10. LeCouteur. R. A.: Gillette, E. L.; Powers, B. E.; Child, G.; McChesney. S. L.; Ingram, J. T. Peripheral neuropathies following experimental intraoperative radiation therapy (IORT). Int. J. Radiat. Oncol. Biol. Phys. 17:583-590; 1989. I 1. McChesney Gillette, S. L.; Gillette, E. L.: Powers, B. E.: Park, R. D.; Withrow. S. J. Ureteral injury following ex-
12.
13.
14. 15.
16.
17.
18.
19.
20.
2 1.
471
perimental intraoperative radiation. Int. J. Radiat. Oncol. Biol. Phys. 17:791-798; 1989. Powers, B. E.; Gillette, E. L.; McChesney, S. L.; LeCouteur, R. A.; Withrow, S. J. Bone necrosis and tumor induction following experimental intraoperative irradiation. Int. J. Radiat. Oncol. Biol. Phys. 17:559-567; 1989. Remy, J.; Martin, M.; Lefaix, J. L.; Daburon, F. Radiation induced fibrosis in pig muscle: pathological and cellular observations. Brit. J. Cancer 53(Suppl. VII):232-233; 1986. Rich, T. A. Intraoperative radiotherapy. Radiother. Oncol. 6:207-22 I ; 1986. Schwenen, M.; Altman, K. 1.; Schroder, W. Radiation-induced increase in the release of amino acids by isolated, perfused skeletal muscle. Int. J. Radiat. Biol. 55:257-269; 1989. Sindelar. W. F.; Hoekstra, H.; Restrepo, C.; Kinsella, T. J. Pathological tissue changes following intraoperative radiotherapy. Am. J. Clin. Oncol. 9:504-509; 1986. Sindelar, W. F.; Tepper, J.; Travis, E. L.; Terrill, R. Tolerance of retroperitoneal structures to intraoperative radiation. Ann. Surg. 196:601-608; 1982. Tepper. J. E.: Gunderson. L. L.; Orlow, E.: Cohen, A. M.; Hedberg. S. E.; Shipley, W. U.; Blitzer, P. H.: Rich, T. Complications of intraoperative radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 10: 183 l-l 839; 1984. Tepper, J. E.; Wood, W. C.; Cohen, A. M. Treatment of locally advanced rectal cancer with external beam radiation, surgical resection, and intraoperative radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 16:1437-1444; 1989. Wegrowski. J.; Lafuma, C.; Lefaix, J. L.; Daburon, F.; Robert, L. Modification of collagen and noncollagenous proteins in radiation-induced muscular fibrosis. Exper. Molecular Pathol. 48:273-285; 1988. Zeman. W.; Solomon, M. Effects of radiation on striated muscle. In: Berdjis, C. C., ed. Pathology of irradiation. Baltimore; Williams and Wilkins, Co.: I7 l-l 85; 197 I.
