- Email: [email protected]
Are free radical scavengers beneficial in the treatment of compartment syndrome after acute arterial ischemia?
Are free radical scavengers beneficial in the treatment of compartment syndrome after acute arterial ischemia? Michael A. Ricci, M D , Alan M. G r a h a m , M D , Raphael Corbisiero, M D , R i c h a r d Baffour, P h D , Farida M o h a m e d , BSc, and James F. Symes, M D ,
Montreal, Quebec, Canada Because it is postulated that compartment syndrome developing secondary to an acute arterial occlusion may be due to reperfusion injury, oxygen-derived free radicals have been implicated in its genesis. To assess the possible beneficial effect of free radical scavengers in this setting, we used a previously established in vivo canine model of compartment syndrome to compare four groups: group I, no treatment; group II, prophylactic fascintomy; group III, intravenous albumin conjugated superoxide dismutase (SOD); group IV, intravenous mannitol (hydroxl radical scavenger). Both hind limbs were completely devascularized at the popliteal level except for an isolated pedicle to the anterior compartment. The right limb served as the nonischemic control, whereas the left underwent 8 hours of ischemia followed by reperfusion. Continuous monitoring of transfascial oxygen tension (tfPo2) demonstrated severe ischemia during occlusion (tfPo2 5.7 -+ 5.1 mm Hg) and restoration of blood flow with reperfusion (mean tfI'o2 50 to 60 mm Hg). Measurements of compartment pressure were significantly higher after reperfusion in groups I, HI, and IV when compared with those of group II (p < 0.001, groups I and II; p < 0.01, group IV). Extent of muscle necrosis assessed by technetium pyrophosphate scanning and expressed as a ratio of left to right legs was as follows: group I, 8.9 -+ 5.0; group II, 2.6 - 0.5; group III, 2.8 -+ 0.8; group IV, !.8 - 0.6. Muscle contraction studies 16 hours after reperfusion indicated abnormal findings in all but group II. In conclusion, administration of free radical scavengers did not preserve normal neuromusodar function despite a significant reduction in muscle damage. The effects of ischemiareperfusion and increased compartment pressure appear additive. Fasciotomy at the time of reperfusion after prolonged ischemia remains the treatment of choice to avoid functional impairment. (J VASc SURG 1989;9:244-50.)
Oxygen-derived free radicals have been implicated as a cause o f injury in various tissues after ischemia. When oxygen is supplied on reperfusion, these highly reactive compounds are generated and attack cell membranes, leading to increased vascular permeability, edema, and, ultimately, tissue necrosis. I3 Consequently, various free radical scavengers, which reduce free radicals to less toxic products, have been used to limit tissue injury. 2 Specifically, in skeletal
From the CardiovascularResearch Laboratory, The Department of Surgery, Royal Victoria Hospital--McGill University. Supported by the Medical Research Council of Canada and the National Institutes of Health (grant No. H107693). Presented at the Thirty-sixth Scientific Meeting of the North American Chapter, International Society for Cardiovascular Surgery, Chicago, Ill., June 14-15, 1988. Reprint requests: Alan M. Graham,MD, Dept. of Surgery, Royal VictoriaHospital, Suite $10.01,687 Pine Ave.West, Montreal, PQ H3A 1A1, Canada. 244
muscle, both superoxide dismutase (SOD), an oxygen radical ( 0 2 - ) scavenger, and mannitol, a hy droxyl radical (OH.) scavenger, have been shown ~o reduce muscle necrosis. 35 By replicating the clinical sequence o f limb ischemia followed by reperfusion and elevated compartment pressures, we have produced an in vivo canine model o f the compartment syndrome. In this exper- _ iment the effects o f these free radical scavengers (SOD and mannitol) are compared with those o f traditional treatment by fasciotomy. MATERIAL AND METHODS Mongrel dogs o f either sex weighing 9 to 11 kg were anesthetized with intravenous pentobarbital (25 mg/kg), intubated, and allowed to breathe room air. Intravenous normal saline solution was supplied at a rate o f 40 to 60 m l / h r over 9 hours. Anesthesia was supplemented as necessary with pentobarbital.
Volume 9 Number 2 February 1989
Free radical scave~zgersin compartment syndrome
'
245
TIBIALIS ANTERIOR
~
~
NERVE
- ~'~ ~
TO RECORDER
Fig. 1. Schematic illustration of muscle contraction apparatus. The care of the animals complied with the guidelines of the Canadian Council of Animal Care, "Principles of Laboratory Animal Care" and the "Guide for the Use of Laboratory Animals" (NIH Publication No. 80-23, revised 1985). The technical details of the compartment syndrome model have been described previously.6 In brief, all muscles and vessels of both hind limbs are divided at the popliteal level so that the distal limb and anterior compartment are isolated to a single pedicle of poplitcal artery and vein. The common peroneal nerve is carefully preserved with minimal mobilization. The right limb serves as a surgical control whereas the left has both vessels occluded for 8 hours. Ischemia and later reperfusion are confirmed t,y the measurement of venous blood gases and trans*ascial oxygen measurement over the anterior compartment (cutaneous PO2 monitor 820, Kontron Medical, Basel, Switzerland). Compartment pressure is continuously monitored by the slit catheter technique7 (Howmedica, Inc., Rutherford, N.J.). Four experimental groups were studied. Group I received no treatment (n = 13), group II had fasciotomy performed immediately before reperfusion (n = 10), group III received 6000 U/kg SOD intravenously 30 minutes before reperfusion (n = 9), and group IV received 25 gm of 25% mannitol (50 ml) intravenously 30 minutes before reperfusion (n = 8). Fasciotomy was not performed in group III or IV. The SOD used in these experiments was conjugated to albumin to prolong its half-life. 8,9 One hour after reperfusion, 35 ~Ci of technetium pyrophosphate (TcPYP) was injected intravenously.
In vitro studies with this radionucleotide have demonstrated uptake by calcium salts from damaged muscle up to 1 hour after injection; accumulation of TcPYP is a reliable indicator of muscle injury. 1° Sixteen hours later, after completion of the muscle function studies described later, the muscles of the anterior compartment were removed, and radioactivity was counted (per gram of wet muscle weight). Results are expressed as a ratio of the experimental left side to the control right leg (L/R ratio). Muscle function studies were carried out in the following fashion, adapted from the method described by Terzis et al. n Sixteen hours after reperfusion, the animal was again anesthetized with pentobarbital, immobilized on its back, and the anterior compartments of both limbs were exposed. The tibialis anterior muscle was isolated without disturbing its vascular supply; the proximal limb was suspended and fLxed to an overlying metal rod (Fig. 1). The distal tendon is then attached to a Grass FT-10 force transducer, which converts the mechanical force produced to an electrical signal that is recorded on a Grass model 79D polygraph (Grass Instrument Co., Quincy, Mass.). Resting length, at which maximal tension is produced, is maintained by a resting tension of 200 gm. The chart speed during trials was 10 mm/sec and the interval between trials was approximately 10 seconds. The record is analyzed by comparison to standard weight calibration done immediately before the experiment. The absolute value obtained (gram force) is standardized by dividing by the wet weight of the tibialis anterior muscle (gram force/gram of tissue).
Journal of VASCULAR SURGERY
246 Ricci et al.
Table I. Muscle contraction studies Normal 2T TFC MTFC 10T TFC MTFC
I
II
III
IV
A N O VA
105 + 65 207 + 22
21 + 14" 106 + 58*
66 + 2 8 t 227 + 6 0 t
16 + 4 ~: 121 + 605
13 +- 6":~ 80 + 2 8 " t
p = 0.004 p < 0.001
93 _+ 24 262 + 122
25 +_ 15 127 + 67*
75 + 2 9 t 254 - 7 9 t
28 + 165 156 + 56*
22 + 16"$ 104 + 43":~
p = 0.002 p = 0.01
Data are expressed as gram force divided by gram o f muscle and mean + one standard deviation. *Significant vs normal, p < 0.05. t Significant vs group I, p < 0.05. ¢Significant vs group II, p < 0.02.
The common peroneal nerve was isolated, carefully suspended away from adjacent tissue by a twopronged electrode, and stimuli were delivered by a Grass $48 stimulator. The lowest voltage capable of producing a measurable contraction was designated the threshold (T). Studies were then carried out at two times the threshold (2T) to activate all large myelinated fibers, and at 10 times threshold (10T), which activates both large and small fibers. Twitch force of contraction (TFC), from a single suprathreshold stimulus, was measured at 2T and 10T. Tetanic tension curves were generated at 2T and 10T by stimulation at the foUowing frequencies: 1, 3, 6, 9, 12, 15, 18, 21, 30, 40, 60, 80, and 100 Hz. The maximum developed tension was termed the "maximal tetanic force of contraction" (MTFC). Four dogs underwent neuromuscular testing without the experimental procedure and served as "normals." At the completion of the muscle function testing, the muscles were harvested for TcPYP scanning after which the dog was killed by an overdose of pentobarbital. All data are expressed as the mean _ one standard deviation. Statistical analysis (one-way analysis of variance [ANOVA] and unpaired t test) was performed on an IBM PC-AT computer with commercially available software.
RESULTS Severe ischemia after occlusion was confirmed in all experimental limbs by a decrease in transfascial oxygen (tfPo2) (89.6---7.0 to 5.7 + 5.1 mm Hg, p < 0.001) and venous pH (7.25 _+ 0.08 to 6.97 + 0.22, p < 0.001), as described previously. 6 After reperfusion tfPo2 levels increased to a mean level of 50 + 5.1 mm Hg except in group I (30 +_ 4,1 mm Hg). Oxygen saturation increased from 43.9% during the ischemic period to 80% to 85% in all groups. During the ischemic period, compartment pressures increased from a mean of
5 mm Hg to an average of 16 mm H g (range 11 to 18 mm Hg) in all groups. In the untreated group (I), compartment pressure increased on reperfusion to 106.2 + 30.8 mm H g (from baseline 6.8 + 5.3 mm Hg, p < 0.001) whereas prophylactic fasciotomy prevented this increase (group II, 15.9 ~-~ 13.3 mm Hg; vs group I,p < 0.001). In groups III and IV compartment pressure increased to 69.4 + 34.2 mm Hg and 73.6 _+ 56.8 mm Hg, respectively (ANOVA: p < 0.001; t test: group II vs III, p < 0.001; group II vs IV, p < 0.01). Technetium scanning demonstrated significant diminution of muscle necrosis in each treatment group. L / R ratios for each group are as follows: I, 8.9 _-+ 5.0; II, 2.6 -+ 0.8; III, 2.8 + 0.8; IV, 1.8 + 0.6 (ANOVA: p < 0.001; t test: vs group I, II, p = 0.002; III, p = 0.01; IV, p = 0.004). There was no significant difference between fasciotomy (group II) and SOD (group III) but maunitol (group IV) was significantly better than fasciotomy (II) (p = 0.05) and group III (p = 0.02). There was no significant difference among the muscle weights in any group. ,~ TFC and MTFC for 2T and 10T are listed in' Table I. Overall, prophylactic fasciotomy produced superior muscle function compared with no treatment, SOD, or mannitol. No difference was noted between the untreated group and the SOD or mannitol groups. Only with tetanic contraction at 10T did function after S O D equal that after prophylactic fasciotomy. Tetanic tension curves at 2T and 10T illustrate the significant effect of prophylactic fasciotomy (Fig. 2). No significant difference was found between control limbs in each group and normal limbs.
DISCUSSION An explosive increase in oxygen-derived free radical production occurs after reperfusion of ischemic tissues. A chain reaction is initiated when oxygen is
Volume 9 Number 2 February 1989
Free radical scavengers in compartment syndrome
247
MUSCLE CONTRACTION 300
E
280
o No Treatment (I)
260 240
o Prophylactic Fasciotomy (li)
220 E Z
c~ l-.-
2T
200
o SOD (111)
- -
~, Mannitol (IV)
/
16o-
~
o
~
o
o
/ /
120 1
O
/
/
180 -
140 i-Z
o
~ °
ooeo - ~o..O. °
/ °/°
^
~
_ ~ 0 ~ - - - - - ~
=n J
o
O ~ ¢~° " o ~_____------
~
o
~
O / ~ " - ° ~ / ~
,o4 2 0 "~=-~'~_ ~. ~,...~O |
I
I
,
,
,
I
,
t
I
,
lOT 300 280 260 I1
0/0/'----~--"
240
0~
0
/o/
220 E Dla
0~
200 180
Z
160 I.-,,el-Z i..IJ
--O-
010
140 120
o
.o~
/o-~o~
/
'
.O
°
100 8O
6O
o°/°]'~
z~/
40 20 0
i 10
i 20
f 30
I 40
i 50
i 60
I 70
i 80
i --1 90
100
FREQUENCY (Hz)
Fig. 2. Tetanic tension curves for each treatment group at (a) 2T and (b) 10T.
supplied to the two components that havc accumulated during the ischemic period: hypoxanthine, the product of adenosine triphosphate catabolism, and the enzyme xanthine oxidase. Superoxide and hydroxyl radicals produce massive tissue damage in various tissues by peroxidation of the lipid component of cell membranes and degradation of the hyaluronic acid of glycocalyx and basement membranes) "2'a2 Free radicals have been implicated in reperfusion injury in the heart, kidneys, intestine, and skeletal muscle).3,13 Several free radical scavengers have been used to
limit tissue injury after ischemia. Korthuis et al. s used allopurinol, catalase, SOD, and dirnethylsulfoxide to attenuate the free radical-induced increase in vascular permeability after 4 hours of ischemia in the canine gracilis muscle. Allopurinol, a xanthine oxidase inhibitor, prevents the formation of free radicals, whereas dimethylsulfoxide and catalase detoxify the OH. radical (a by-product of oxygen radical production), which may in fact be the primary damaging radical in skeletal muscle. 3,~,12 SOD, however, acts directly to convert the highly reactive O2- radical to hydrogen peroxide. The extremely short half-life (6
248
Ricci etal.
minutes) of intravenous SOD may limit its effectiveness because considerably more time may be required before effective tissue concentrations are reached? Conjugation with albumin significantly increases the half-life to 6 hours while up to 90% of the native enzymatic activity is maintained, s,9 Another hydroxyl radical scavenger readily available is mannitol. Although its abifity to increase glomerular filtration, 14 reduce cellular swelling, is and possibly reduce compartment pressure ~6 has been documented, its hydroxyl scavenging activity is less well understood. It has been suggested that the aldehyde moiety of mannitol reacts with OH. to produce a less reactive "mannitol radical." To clarify the dual activities of mannitol, Magovern et al?7 compared it with hyperosmolar glucose (at the same osmolarity as mannitol) and standard crystalloid solution in a rabbit heart preparation. Significantly improved function occurred after reperfusion with mannitol compared with crystalloid controls or hyperosmolar glucose, suggesting that hydroxyl radical scavenging, rather than hyperosmolarity alone, is the primary benefit of mannitol. With an isolated gracilis muscle model, Walker et al.4 demonstrated a significant reduction in muscle damage after administration of free radical scavengers. However, they found it necessary to control the oxygen content of reperfused blood and to use a combination of mannitol, SOD, and catalase before significant improvement was obtained. Similarly, with the same model, Blebea et al. 5 compared intraarterial mannitol and SOD. With technetium scanning used to assess muscle injury, mannitol produced a 15% decrease in muscle necrosis but SOD was no better than controls. This is in agreement with our finding that mannitol significantly attenuated muscle damage in experimental limbs compared with that in the untreated group. Indeed , mannitol produced a significantly lower L / R ratio compared with that produced by SOD (group III) and with fasciotomy (group II). However, in contrast to the gracilis model, SOD and fasciotomy preserved muscle compared with no treatment; this may be due to differences in the experimental model because our preparation isolates the entire anterior compartment in vivo rather than a single muscle. Despite a significant decrease in muscle necrosis in all three treatment groups in this study, statistically significant preservation of muscle function occurred only when fasciotomy was performed before reperfusion (Table I). With functional studies similar to this experiment, Patterson et al?s observed up to an 80% reduction in gastrocnemius function 6 days after 5 hours of tourniquet ischemia in mon-
Journal of VASCULAR SURGERY
keys. Six days after a 3-hour ischemic period, the muscle varied from normal to only a 36% reduction. This is consistent with ultrastructure studies, which demonstrate sublethal injury after 3 hours ofischemia but irreversible capillary and myofibrillar changes at 4 hours? 9 Our model produces significant ischemia over 8 hours resulting in severe muscle damage, as evidenced by the high L / R ratio of group I. Despite the extent of devascularization in our model, some perfusion of the muscle occurred, necessitating longer periods of ischemia than in a truly isolated preparation. Although muscle necrosis is significantly reduced in the treatment groups, an additional component capable of causing neuromuscular injury (i.e., elevated compartment pressure) remains. Pressure alone, induced by infusing autogenous plasma into the anterior compartment of the dog, produces histologic and TcPYP evidence of muscle injury after pressure of 30 mm H g for 8 hours. 2° When muscl',* function was studied in this model, 21 after 8 hours of pressure at 40 mm Hg, twitch force of contraction and maximal tetanic contraction were significantly decreased from baseline when measured 2 days later. By 4 weeks, no muscle weakness was noted. The nerve itself, particularly the neuromuscular junction, 22 is also sensitive to the effects of ischcmia or pressure. Nerve function rapidly deteriorates with the onset of ischemia but after up to 6 hours of ischemia rapidly recovers with restoration of blood flow. 2a After 8 hours, however, the endoneural vessels are irreversibly damaged and no recovery of function occurs. ='2a In addition, after pressure elevation in the anterior compartment from 30 to 40 mm Hg, nerve conduction is slowed, but small, slow fibers continue to function for 14 hours. = Complete block occurs only after compartment pressure is greate~ than 50 mm Hg, although recovery occurs when the pressure is reduced. = A threshold pressure level that produces irreversible nerve damage was suggested by Rorabeck and Clarke, 23 who found that 40 mm H g for 12 hours prevented nerve recovery, even if fasciotomy was performed. It appears that ultrastructural evidence of muscle damage occurs after 4 hours of ischemia and histologically at 12 hours of pressure elevation to 30 mm Hg.~S'19Muscle function recovers, at least in part, after either insult. However, nerve function is exquisitely sensitive to ischemia but recovers if perfusion is restored before 8 hours. The nerve can endure a longer period of elevated compartment pressure, up to 12 hours, before irreversible damage occurs.='2a However, it may be that the combined effects of ischemia and elevated pressure are additive in dam-
Volume 9 Number 2 February 1989
a g i n g the n e u r o m u s c u l a r unit. 22 F r o m the results o f this study, it w o u l d a p p e a r t h a t muscle necrosis after an 8 - h o u r ischemic p e r i o d f o l l o w e d b y reperfusion a n d the d e v e l o p m e n t o f a c o m p a r t m e n t s y n d r o m e , can be l i m i t e d b y f a s c i o t o m y o r free radical scavengers. F u n c t i o n o f t h e n e u r o m u s c u l a r u n i t as a whole, h o w e v e r , was n o t i m p r o v e d b y S O D o r m a n n i t o l . F a s c i o t o m y was the o n l y t r e a t m e n t that reliably prev e n t e d a pressure increase after p r o l o n g e d ischemia and, at least in the s h o r t term, preserved muscle function. Thus, a l t h o u g h free radical scavengers m a y p r o v e useful adjunctive t h e r a p y , early f a s c i o t o m y rem a i n s the m o s t effective t r e a t m e n t for t h e ischemiareperfusion compartment syndrome. We thank Dr. M. J. Poznansky o f the University of Alberta for preparing the SOD and Nicholas Giannou, Lise Proulx, and Giselle Pouliot for their technical assistance. , _EFERENCES
1. McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med I985;312:159-63. 2. Bulkley GB. The role of oxygen free radicals in human disease processes. Surgery 1983;94:407-11. 3. Korthuis RJ, Granger DN, Townsley MI, Taylor AE. The role of oxygen-derived free radicals in ischemia-induced increases in canine skeletal musde vascular permeability. Circ Res 1985;57:599-609. 4. Walker PM, Lindsay TF, Labbe 1% et al. Salvage of skeletal muscle with free radical scavengers. I VASCSUR6 1987;5:6875. 5. Blebea J, Kerr JC, Hobson RW II, Padberg FT Jr. The effects of oxygen free radical scavengers on skeletal musde ischerma and reperfusion injury. Curr Surg 1987;44:396-8. 6. Corbisiero RM, Graham AM, Riccl MA, et al. A canine model of the ischemia-reperfusion compartment syndrome: evaluation of delayed vs prophylactic fasciotomy. Presented at the Annual Meeting of the Committee on Trauma of the American College of Surgeons, Nashville, Tenn., Feb. 25, 1988 7. Rorabeck CH, Castle GSP, Hardie 1% Logan J. Compartmental pressure measurements: an experimental investigation using the slit catheter. J Trauma 1981;21:446-9. 8. Wong K, Cleland LG, Poznansky MJ. Enhanced anaLinflammatory effect and reduced immunogenicity of bovine fiver superoxide dismutase by conjugation with homologous albumin. Agents Actions 1980;10:231-9.
DISCUSSION
Dr. Dhiraj M. Shah (Albany, N.Y.). The authors have shown that after 8 hours of ischcmia and 16 hours of reperfusion in a canine lower limb model, there is a demonstrable reperfusion syndrome manifested by increased compartmental pressure, muscle necrosis, and neuromus-
Free radical scavengers in compartment syndrome 249
9. CIeland LG, Betts WH. Conjugated superoxide dismutase complexes. In: Greenwald RA, ed. CRC handbook of methods for oxygen radical research. Boca Raton: CRC Press, 1985:31-2. 10. Buja LM, Tofe AJ, Kulkami PV, et al. Sites and mechanisms of localization of technetium-99m phosphorous radiopharmaceuticals in acute myocardial infarcts and other tissues. J Clin Invest 1977;60:724-70. 11. Terzis IK, Sweet RC, Dykes RW, Williams HB. Recover), of function in free muscle transplants using microneurovascular anastomoses. J Hand Surg 1978;3:37-59. 12. DelMaestro RF, Thaw HH, Bjork J, et al. Free radicals as mediators of tissue injurY". Acta Physiol Scand 1980;492 (Suppl).:43-57. 13. Harris K, Walker PM, Miclde DAG, et al. Metabolic response of skeletal muscle to ischemia. Am J Physiol 1985;19:H213H220. 14. Valdes ME, Landau SE, Shah DM, et al. Increased glomemlar filtration rate following mannitol administration in man. J Surg Res 1979;26:473-7. 15. Buchbinder D, Karmody AM, Leather RP, Shah DM. Hypertonic mannitol; its use in the prevention of revascularization syndrome after acute arterial ischemia. Arch Surg 1981; 116:414-21. 16. Hutton M, Rhodes RS, Chapman G. The lowering ofpostischemic compartment pressures with mannitol. J Surg Res 1982;32:239-42. 17. Magovern GJ Jr, Boiling SF, Casale AS, et al. The mechanism of mannitol in reducing ischemic injury: hypersmolarity or hydroxyl scavenger. Circulation 1984;70(Suppl I):I-91-I-95. 18. Patterson S, Klenerman L, Kiswas M, Rhodes A. The effect of pneumatic tourniquets on skeletal muscle physiology. Acta Orv,hop Scand 1981;52:171-5. 19. Santavirta S, Luoma A, Arstila AU. Ultrastrucmral changes in striated muscle after experimental tourniquet ischernia and short reflow. Eur Surg Res 1978;10:415-24. 20. Hargeus AR, Schmidt DA, Evans KL, et al. Quantitation of skeletal-muscle necrosis in a model compartment syndrome. J Bone Joint Surg [Am] 1981;63:63L6. 21. Mortenson WW, Hargens A1% Gershuni DH, et al. Longterm myoneural function after an induced compamnent syndrome in the canine hindlimb. Clin Orthop 1985;195:289i 93. 22. Lundborg G. Structure and function of the intraneural microvessels as related to trauma, edema formation, and nerve function. J Bone Joint Surg [Am] 1975;57:938-48. 23. Rorabeck CH, Clarke KM. The pathophysiology of the anterior tibial compartment syndrome: an experimental investigation. J Trauma 1978;18:299-304.
cular dysfianction. They hypothesized that this reperfusion syndrome is due to oxygen-derived free radical injury. Fasciotomy, which releases compartmental pressure, offered complete recovery from this insult. However, when free radical scavengers such as superoxide dismutase (SOD) or mannitol, is used, partial relief is achieved.
250
Ricci et al.
Our experience in the treatment of this syndrome is primarily based on the use o f hypertonic mannitol and fasciotomy. In an experimental design similar to this but using the entire lower limb, we have shown that after 90 minutes o f hypoperfusion followed by 2 hours of reperfusion, reperfusion syndrome occurs, manifested by decreased blood flow, increased resistance, decreased oxygen uptake, and muscle necrosis. On the basis of our experimental findings we have treated patients who have acute arterial ischemia with hypertonic mannitol before revascularization. The use o f hypertonic mannitol in conjunction with revascularization after acute arterial occlusion caused by thromboembolism decreased the necessity for fasciotomy. Only 3 of 24 patients required fasciotomy compared with the group of untreated patients, of which 13 of 17 patients required fasciotomy. Furthermore, the neuromuscular dysfunction was minimal and similar in both groups. After acute arterial occlusion from trauma to the popliteal artery, we treated 20 patients with mannitol; 15 patients were not treated with mannitol. Two o f the 20 patients receiving mannitol required fasciotomy and 12 of the 15 untreated patients required fasciotomy. Again, neuromuscular dysfunction was minimal in both groups and were similar. It is possible to compare this trauma group with this experimental model because of similar popliteal ischemia. There may be certain reasons for this difference in the effect of mannitol that the authors have seen and we have observed in terms of neuromuscular dysfunction. Although the authors have corrected for short half-life of SOD, they gave only a single treatment of mannitol. In the doses that they have used, if dogs had normal glomerular filtration rate, its half-life should be an hour or less. Therefore mannitol should have been repeated or continuously given to achieve maximal effect. The second critique is that the authors hypothesized that the injury is due to superoxide radical. If that is so, then why would only compartment decompression improve the situation? Do you have general hemodynamic data on these dogs relating to cardiac depression, hypotension, and thromboxane release? Finally, I suggest that although fasciotomy offered complete relief, if the authors had used marmitol in appropriate doses, it also would have achieved similar neuromuscular fimctional recovery in these dogs. Dr. William J. Quifiones-Baldrich (Los Angeles, Calif.). In our laboratory we have been interested in studying ischemia and reperfusion in a rabbit model of hind limb ischemia. One of the difficulties is measurement of skeletal muscle function in contrast to that of cardiac muscle, where function is rather easy to measure. We have noticed a significant difference in the recovery ofnervc and muscle after ischemia and reperfusion. In our model, peripheral nerve has poor recovery after 3 hours of ischemia in contrast to skeletal muscle, which has about 60% of control function after 3 hours of ischemia and 2 hours of reperfusion. When skeletal muscle fimction is studied, it is critical
Journal of VASCULAR SURGERY
to control for temperature. Temperature is a variable that will significantly alter function o f any muscle. In addition, it is important to determine the optimal resting tension of the muscle. The authors used 200 gm; I would like to know how that was determined. In two of the experimental groups the authors noted that there was significant edema. No edema was noted in the group with fasdotomy. This is interesting, considering that one o f the mechanisms by which fasciotomy works is by allowing the muscle to develop edema and thus not compromise reperfusion. Regardless, expressing muscle function as gram force per gram of tissue can be misleading because edema is present in some specimens and not in others. Dr. Ricci, did you control for temperature? Did you perform direct muscle stimulation and compare it with nerve stimulation? H o w was the optimal tension at rest determined? H o w do you adjust for the differences when expressing your results as gram force per gram tissue when you have variable edema in some of the specimens? Dr. Rieei (closing). Dr. Shah, of course we are we,. aware of your work and the work at Albany with mannitol; in fact that was, at least in part, a stimulus for our experiments. We did not repeat the dose, as you pointed out. We used a fairly large dose to start. We believed that a single dose would be more comparable to the single dose of SOD. Indeed if we were using these compounds as free radical scavengers, the damage done by the free radicals is immediate and that's when we needed to have the effects of SOD or marmitol. It may very well be that a drip might produce different results. We do not have hemodynamic data, as you suggested, with regards to fasciotomy. Fasciotomy certainly allows better perfusion of the muscle; I am not sure we can explain why it does not produce a reperfusion syndrome. Dr. Quifiones-Baldrich, we certainly agree it is difficult to study muscle function. Our next set of experiments is determined to study the nerve as well. As of now we have not done direct muscle stimulation, which may tell us more about which component of the neuromuscular problem most prominent. Temperature was controlled. Initially all the flaps were closed and the animal was kept warm. There were no special efforts to keep the muscle at the same temperature or warmed during the ischemic period or overnight. Again, the next day when muscle function testing was done; it was primarily the animal that was warmed. Two hundred grams were determined by some previous work that was done by our plastic surgery department with the same technique, which we adopted. In fact, I did not comment about edema. The fasciotomy produced less muscle necrosis and a lower L / R ratio. Edema in each of the legs judged by muscle weight was not significantly different. This is somewhat different than others have reported, but in our animals there is no significant difference in muscle weights; therefore we felt justified in expressing our results in terms of the muscle weight. In addition, the left leg is always compared with the right leg so each animal in a sense serves as his own control.
Montreal, Quebec, Canada Because it is postulated that compartment syndrome developing secondary to an acute arterial occlusion may be due to reperfusion injury, oxygen-derived free radicals have been implicated in its genesis. To assess the possible beneficial effect of free radical scavengers in this setting, we used a previously established in vivo canine model of compartment syndrome to compare four groups: group I, no treatment; group II, prophylactic fascintomy; group III, intravenous albumin conjugated superoxide dismutase (SOD); group IV, intravenous mannitol (hydroxl radical scavenger). Both hind limbs were completely devascularized at the popliteal level except for an isolated pedicle to the anterior compartment. The right limb served as the nonischemic control, whereas the left underwent 8 hours of ischemia followed by reperfusion. Continuous monitoring of transfascial oxygen tension (tfPo2) demonstrated severe ischemia during occlusion (tfPo2 5.7 -+ 5.1 mm Hg) and restoration of blood flow with reperfusion (mean tfI'o2 50 to 60 mm Hg). Measurements of compartment pressure were significantly higher after reperfusion in groups I, HI, and IV when compared with those of group II (p < 0.001, groups I and II; p < 0.01, group IV). Extent of muscle necrosis assessed by technetium pyrophosphate scanning and expressed as a ratio of left to right legs was as follows: group I, 8.9 -+ 5.0; group II, 2.6 - 0.5; group III, 2.8 -+ 0.8; group IV, !.8 - 0.6. Muscle contraction studies 16 hours after reperfusion indicated abnormal findings in all but group II. In conclusion, administration of free radical scavengers did not preserve normal neuromusodar function despite a significant reduction in muscle damage. The effects of ischemiareperfusion and increased compartment pressure appear additive. Fasciotomy at the time of reperfusion after prolonged ischemia remains the treatment of choice to avoid functional impairment. (J VASc SURG 1989;9:244-50.)
Oxygen-derived free radicals have been implicated as a cause o f injury in various tissues after ischemia. When oxygen is supplied on reperfusion, these highly reactive compounds are generated and attack cell membranes, leading to increased vascular permeability, edema, and, ultimately, tissue necrosis. I3 Consequently, various free radical scavengers, which reduce free radicals to less toxic products, have been used to limit tissue injury. 2 Specifically, in skeletal
From the CardiovascularResearch Laboratory, The Department of Surgery, Royal Victoria Hospital--McGill University. Supported by the Medical Research Council of Canada and the National Institutes of Health (grant No. H107693). Presented at the Thirty-sixth Scientific Meeting of the North American Chapter, International Society for Cardiovascular Surgery, Chicago, Ill., June 14-15, 1988. Reprint requests: Alan M. Graham,MD, Dept. of Surgery, Royal VictoriaHospital, Suite $10.01,687 Pine Ave.West, Montreal, PQ H3A 1A1, Canada. 244
muscle, both superoxide dismutase (SOD), an oxygen radical ( 0 2 - ) scavenger, and mannitol, a hy droxyl radical (OH.) scavenger, have been shown ~o reduce muscle necrosis. 35 By replicating the clinical sequence o f limb ischemia followed by reperfusion and elevated compartment pressures, we have produced an in vivo canine model o f the compartment syndrome. In this exper- _ iment the effects o f these free radical scavengers (SOD and mannitol) are compared with those o f traditional treatment by fasciotomy. MATERIAL AND METHODS Mongrel dogs o f either sex weighing 9 to 11 kg were anesthetized with intravenous pentobarbital (25 mg/kg), intubated, and allowed to breathe room air. Intravenous normal saline solution was supplied at a rate o f 40 to 60 m l / h r over 9 hours. Anesthesia was supplemented as necessary with pentobarbital.
Volume 9 Number 2 February 1989
Free radical scave~zgersin compartment syndrome
'
245
TIBIALIS ANTERIOR
~
~
NERVE
- ~'~ ~
TO RECORDER
Fig. 1. Schematic illustration of muscle contraction apparatus. The care of the animals complied with the guidelines of the Canadian Council of Animal Care, "Principles of Laboratory Animal Care" and the "Guide for the Use of Laboratory Animals" (NIH Publication No. 80-23, revised 1985). The technical details of the compartment syndrome model have been described previously.6 In brief, all muscles and vessels of both hind limbs are divided at the popliteal level so that the distal limb and anterior compartment are isolated to a single pedicle of poplitcal artery and vein. The common peroneal nerve is carefully preserved with minimal mobilization. The right limb serves as a surgical control whereas the left has both vessels occluded for 8 hours. Ischemia and later reperfusion are confirmed t,y the measurement of venous blood gases and trans*ascial oxygen measurement over the anterior compartment (cutaneous PO2 monitor 820, Kontron Medical, Basel, Switzerland). Compartment pressure is continuously monitored by the slit catheter technique7 (Howmedica, Inc., Rutherford, N.J.). Four experimental groups were studied. Group I received no treatment (n = 13), group II had fasciotomy performed immediately before reperfusion (n = 10), group III received 6000 U/kg SOD intravenously 30 minutes before reperfusion (n = 9), and group IV received 25 gm of 25% mannitol (50 ml) intravenously 30 minutes before reperfusion (n = 8). Fasciotomy was not performed in group III or IV. The SOD used in these experiments was conjugated to albumin to prolong its half-life. 8,9 One hour after reperfusion, 35 ~Ci of technetium pyrophosphate (TcPYP) was injected intravenously.
In vitro studies with this radionucleotide have demonstrated uptake by calcium salts from damaged muscle up to 1 hour after injection; accumulation of TcPYP is a reliable indicator of muscle injury. 1° Sixteen hours later, after completion of the muscle function studies described later, the muscles of the anterior compartment were removed, and radioactivity was counted (per gram of wet muscle weight). Results are expressed as a ratio of the experimental left side to the control right leg (L/R ratio). Muscle function studies were carried out in the following fashion, adapted from the method described by Terzis et al. n Sixteen hours after reperfusion, the animal was again anesthetized with pentobarbital, immobilized on its back, and the anterior compartments of both limbs were exposed. The tibialis anterior muscle was isolated without disturbing its vascular supply; the proximal limb was suspended and fLxed to an overlying metal rod (Fig. 1). The distal tendon is then attached to a Grass FT-10 force transducer, which converts the mechanical force produced to an electrical signal that is recorded on a Grass model 79D polygraph (Grass Instrument Co., Quincy, Mass.). Resting length, at which maximal tension is produced, is maintained by a resting tension of 200 gm. The chart speed during trials was 10 mm/sec and the interval between trials was approximately 10 seconds. The record is analyzed by comparison to standard weight calibration done immediately before the experiment. The absolute value obtained (gram force) is standardized by dividing by the wet weight of the tibialis anterior muscle (gram force/gram of tissue).
Journal of VASCULAR SURGERY
246 Ricci et al.
Table I. Muscle contraction studies Normal 2T TFC MTFC 10T TFC MTFC
I
II
III
IV
A N O VA
105 + 65 207 + 22
21 + 14" 106 + 58*
66 + 2 8 t 227 + 6 0 t
16 + 4 ~: 121 + 605
13 +- 6":~ 80 + 2 8 " t
p = 0.004 p < 0.001
93 _+ 24 262 + 122
25 +_ 15 127 + 67*
75 + 2 9 t 254 - 7 9 t
28 + 165 156 + 56*
22 + 16"$ 104 + 43":~
p = 0.002 p = 0.01
Data are expressed as gram force divided by gram o f muscle and mean + one standard deviation. *Significant vs normal, p < 0.05. t Significant vs group I, p < 0.05. ¢Significant vs group II, p < 0.02.
The common peroneal nerve was isolated, carefully suspended away from adjacent tissue by a twopronged electrode, and stimuli were delivered by a Grass $48 stimulator. The lowest voltage capable of producing a measurable contraction was designated the threshold (T). Studies were then carried out at two times the threshold (2T) to activate all large myelinated fibers, and at 10 times threshold (10T), which activates both large and small fibers. Twitch force of contraction (TFC), from a single suprathreshold stimulus, was measured at 2T and 10T. Tetanic tension curves were generated at 2T and 10T by stimulation at the foUowing frequencies: 1, 3, 6, 9, 12, 15, 18, 21, 30, 40, 60, 80, and 100 Hz. The maximum developed tension was termed the "maximal tetanic force of contraction" (MTFC). Four dogs underwent neuromuscular testing without the experimental procedure and served as "normals." At the completion of the muscle function testing, the muscles were harvested for TcPYP scanning after which the dog was killed by an overdose of pentobarbital. All data are expressed as the mean _ one standard deviation. Statistical analysis (one-way analysis of variance [ANOVA] and unpaired t test) was performed on an IBM PC-AT computer with commercially available software.
RESULTS Severe ischemia after occlusion was confirmed in all experimental limbs by a decrease in transfascial oxygen (tfPo2) (89.6---7.0 to 5.7 + 5.1 mm Hg, p < 0.001) and venous pH (7.25 _+ 0.08 to 6.97 + 0.22, p < 0.001), as described previously. 6 After reperfusion tfPo2 levels increased to a mean level of 50 + 5.1 mm Hg except in group I (30 +_ 4,1 mm Hg). Oxygen saturation increased from 43.9% during the ischemic period to 80% to 85% in all groups. During the ischemic period, compartment pressures increased from a mean of
5 mm Hg to an average of 16 mm H g (range 11 to 18 mm Hg) in all groups. In the untreated group (I), compartment pressure increased on reperfusion to 106.2 + 30.8 mm H g (from baseline 6.8 + 5.3 mm Hg, p < 0.001) whereas prophylactic fasciotomy prevented this increase (group II, 15.9 ~-~ 13.3 mm Hg; vs group I,p < 0.001). In groups III and IV compartment pressure increased to 69.4 + 34.2 mm Hg and 73.6 _+ 56.8 mm Hg, respectively (ANOVA: p < 0.001; t test: group II vs III, p < 0.001; group II vs IV, p < 0.01). Technetium scanning demonstrated significant diminution of muscle necrosis in each treatment group. L / R ratios for each group are as follows: I, 8.9 _-+ 5.0; II, 2.6 -+ 0.8; III, 2.8 + 0.8; IV, 1.8 + 0.6 (ANOVA: p < 0.001; t test: vs group I, II, p = 0.002; III, p = 0.01; IV, p = 0.004). There was no significant difference between fasciotomy (group II) and SOD (group III) but maunitol (group IV) was significantly better than fasciotomy (II) (p = 0.05) and group III (p = 0.02). There was no significant difference among the muscle weights in any group. ,~ TFC and MTFC for 2T and 10T are listed in' Table I. Overall, prophylactic fasciotomy produced superior muscle function compared with no treatment, SOD, or mannitol. No difference was noted between the untreated group and the SOD or mannitol groups. Only with tetanic contraction at 10T did function after S O D equal that after prophylactic fasciotomy. Tetanic tension curves at 2T and 10T illustrate the significant effect of prophylactic fasciotomy (Fig. 2). No significant difference was found between control limbs in each group and normal limbs.
DISCUSSION An explosive increase in oxygen-derived free radical production occurs after reperfusion of ischemic tissues. A chain reaction is initiated when oxygen is
Volume 9 Number 2 February 1989
Free radical scavengers in compartment syndrome
247
MUSCLE CONTRACTION 300
E
280
o No Treatment (I)
260 240
o Prophylactic Fasciotomy (li)
220 E Z
c~ l-.-
2T
200
o SOD (111)
- -
~, Mannitol (IV)
/
16o-
~
o
~
o
o
/ /
120 1
O
/
/
180 -
140 i-Z
o
~ °
ooeo - ~o..O. °
/ °/°
^
~
_ ~ 0 ~ - - - - - ~
=n J
o
O ~ ¢~° " o ~_____------
~
o
~
O / ~ " - ° ~ / ~
,o4 2 0 "~=-~'~_ ~. ~,...~O |
I
I
,
,
,
I
,
t
I
,
lOT 300 280 260 I1
0/0/'----~--"
240
0~
0
/o/
220 E Dla
0~
200 180
Z
160 I.-,,el-Z i..IJ
--O-
010
140 120
o
.o~
/o-~o~
/
'
.O
°
100 8O
6O
o°/°]'~
z~/
40 20 0
i 10
i 20
f 30
I 40
i 50
i 60
I 70
i 80
i --1 90
100
FREQUENCY (Hz)
Fig. 2. Tetanic tension curves for each treatment group at (a) 2T and (b) 10T.
supplied to the two components that havc accumulated during the ischemic period: hypoxanthine, the product of adenosine triphosphate catabolism, and the enzyme xanthine oxidase. Superoxide and hydroxyl radicals produce massive tissue damage in various tissues by peroxidation of the lipid component of cell membranes and degradation of the hyaluronic acid of glycocalyx and basement membranes) "2'a2 Free radicals have been implicated in reperfusion injury in the heart, kidneys, intestine, and skeletal muscle).3,13 Several free radical scavengers have been used to
limit tissue injury after ischemia. Korthuis et al. s used allopurinol, catalase, SOD, and dirnethylsulfoxide to attenuate the free radical-induced increase in vascular permeability after 4 hours of ischemia in the canine gracilis muscle. Allopurinol, a xanthine oxidase inhibitor, prevents the formation of free radicals, whereas dimethylsulfoxide and catalase detoxify the OH. radical (a by-product of oxygen radical production), which may in fact be the primary damaging radical in skeletal muscle. 3,~,12 SOD, however, acts directly to convert the highly reactive O2- radical to hydrogen peroxide. The extremely short half-life (6
248
Ricci etal.
minutes) of intravenous SOD may limit its effectiveness because considerably more time may be required before effective tissue concentrations are reached? Conjugation with albumin significantly increases the half-life to 6 hours while up to 90% of the native enzymatic activity is maintained, s,9 Another hydroxyl radical scavenger readily available is mannitol. Although its abifity to increase glomerular filtration, 14 reduce cellular swelling, is and possibly reduce compartment pressure ~6 has been documented, its hydroxyl scavenging activity is less well understood. It has been suggested that the aldehyde moiety of mannitol reacts with OH. to produce a less reactive "mannitol radical." To clarify the dual activities of mannitol, Magovern et al?7 compared it with hyperosmolar glucose (at the same osmolarity as mannitol) and standard crystalloid solution in a rabbit heart preparation. Significantly improved function occurred after reperfusion with mannitol compared with crystalloid controls or hyperosmolar glucose, suggesting that hydroxyl radical scavenging, rather than hyperosmolarity alone, is the primary benefit of mannitol. With an isolated gracilis muscle model, Walker et al.4 demonstrated a significant reduction in muscle damage after administration of free radical scavengers. However, they found it necessary to control the oxygen content of reperfused blood and to use a combination of mannitol, SOD, and catalase before significant improvement was obtained. Similarly, with the same model, Blebea et al. 5 compared intraarterial mannitol and SOD. With technetium scanning used to assess muscle injury, mannitol produced a 15% decrease in muscle necrosis but SOD was no better than controls. This is in agreement with our finding that mannitol significantly attenuated muscle damage in experimental limbs compared with that in the untreated group. Indeed , mannitol produced a significantly lower L / R ratio compared with that produced by SOD (group III) and with fasciotomy (group II). However, in contrast to the gracilis model, SOD and fasciotomy preserved muscle compared with no treatment; this may be due to differences in the experimental model because our preparation isolates the entire anterior compartment in vivo rather than a single muscle. Despite a significant decrease in muscle necrosis in all three treatment groups in this study, statistically significant preservation of muscle function occurred only when fasciotomy was performed before reperfusion (Table I). With functional studies similar to this experiment, Patterson et al?s observed up to an 80% reduction in gastrocnemius function 6 days after 5 hours of tourniquet ischemia in mon-
Journal of VASCULAR SURGERY
keys. Six days after a 3-hour ischemic period, the muscle varied from normal to only a 36% reduction. This is consistent with ultrastructure studies, which demonstrate sublethal injury after 3 hours ofischemia but irreversible capillary and myofibrillar changes at 4 hours? 9 Our model produces significant ischemia over 8 hours resulting in severe muscle damage, as evidenced by the high L / R ratio of group I. Despite the extent of devascularization in our model, some perfusion of the muscle occurred, necessitating longer periods of ischemia than in a truly isolated preparation. Although muscle necrosis is significantly reduced in the treatment groups, an additional component capable of causing neuromuscular injury (i.e., elevated compartment pressure) remains. Pressure alone, induced by infusing autogenous plasma into the anterior compartment of the dog, produces histologic and TcPYP evidence of muscle injury after pressure of 30 mm H g for 8 hours. 2° When muscl',* function was studied in this model, 21 after 8 hours of pressure at 40 mm Hg, twitch force of contraction and maximal tetanic contraction were significantly decreased from baseline when measured 2 days later. By 4 weeks, no muscle weakness was noted. The nerve itself, particularly the neuromuscular junction, 22 is also sensitive to the effects of ischcmia or pressure. Nerve function rapidly deteriorates with the onset of ischemia but after up to 6 hours of ischemia rapidly recovers with restoration of blood flow. 2a After 8 hours, however, the endoneural vessels are irreversibly damaged and no recovery of function occurs. ='2a In addition, after pressure elevation in the anterior compartment from 30 to 40 mm Hg, nerve conduction is slowed, but small, slow fibers continue to function for 14 hours. = Complete block occurs only after compartment pressure is greate~ than 50 mm Hg, although recovery occurs when the pressure is reduced. = A threshold pressure level that produces irreversible nerve damage was suggested by Rorabeck and Clarke, 23 who found that 40 mm H g for 12 hours prevented nerve recovery, even if fasciotomy was performed. It appears that ultrastructural evidence of muscle damage occurs after 4 hours of ischemia and histologically at 12 hours of pressure elevation to 30 mm Hg.~S'19Muscle function recovers, at least in part, after either insult. However, nerve function is exquisitely sensitive to ischemia but recovers if perfusion is restored before 8 hours. The nerve can endure a longer period of elevated compartment pressure, up to 12 hours, before irreversible damage occurs.='2a However, it may be that the combined effects of ischemia and elevated pressure are additive in dam-
Volume 9 Number 2 February 1989
a g i n g the n e u r o m u s c u l a r unit. 22 F r o m the results o f this study, it w o u l d a p p e a r t h a t muscle necrosis after an 8 - h o u r ischemic p e r i o d f o l l o w e d b y reperfusion a n d the d e v e l o p m e n t o f a c o m p a r t m e n t s y n d r o m e , can be l i m i t e d b y f a s c i o t o m y o r free radical scavengers. F u n c t i o n o f t h e n e u r o m u s c u l a r u n i t as a whole, h o w e v e r , was n o t i m p r o v e d b y S O D o r m a n n i t o l . F a s c i o t o m y was the o n l y t r e a t m e n t that reliably prev e n t e d a pressure increase after p r o l o n g e d ischemia and, at least in the s h o r t term, preserved muscle function. Thus, a l t h o u g h free radical scavengers m a y p r o v e useful adjunctive t h e r a p y , early f a s c i o t o m y rem a i n s the m o s t effective t r e a t m e n t for t h e ischemiareperfusion compartment syndrome. We thank Dr. M. J. Poznansky o f the University of Alberta for preparing the SOD and Nicholas Giannou, Lise Proulx, and Giselle Pouliot for their technical assistance. , _EFERENCES
1. McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med I985;312:159-63. 2. Bulkley GB. The role of oxygen free radicals in human disease processes. Surgery 1983;94:407-11. 3. Korthuis RJ, Granger DN, Townsley MI, Taylor AE. The role of oxygen-derived free radicals in ischemia-induced increases in canine skeletal musde vascular permeability. Circ Res 1985;57:599-609. 4. Walker PM, Lindsay TF, Labbe 1% et al. Salvage of skeletal muscle with free radical scavengers. I VASCSUR6 1987;5:6875. 5. Blebea J, Kerr JC, Hobson RW II, Padberg FT Jr. The effects of oxygen free radical scavengers on skeletal musde ischerma and reperfusion injury. Curr Surg 1987;44:396-8. 6. Corbisiero RM, Graham AM, Riccl MA, et al. A canine model of the ischemia-reperfusion compartment syndrome: evaluation of delayed vs prophylactic fasciotomy. Presented at the Annual Meeting of the Committee on Trauma of the American College of Surgeons, Nashville, Tenn., Feb. 25, 1988 7. Rorabeck CH, Castle GSP, Hardie 1% Logan J. Compartmental pressure measurements: an experimental investigation using the slit catheter. J Trauma 1981;21:446-9. 8. Wong K, Cleland LG, Poznansky MJ. Enhanced anaLinflammatory effect and reduced immunogenicity of bovine fiver superoxide dismutase by conjugation with homologous albumin. Agents Actions 1980;10:231-9.
DISCUSSION
Dr. Dhiraj M. Shah (Albany, N.Y.). The authors have shown that after 8 hours of ischcmia and 16 hours of reperfusion in a canine lower limb model, there is a demonstrable reperfusion syndrome manifested by increased compartmental pressure, muscle necrosis, and neuromus-
Free radical scavengers in compartment syndrome 249
9. CIeland LG, Betts WH. Conjugated superoxide dismutase complexes. In: Greenwald RA, ed. CRC handbook of methods for oxygen radical research. Boca Raton: CRC Press, 1985:31-2. 10. Buja LM, Tofe AJ, Kulkami PV, et al. Sites and mechanisms of localization of technetium-99m phosphorous radiopharmaceuticals in acute myocardial infarcts and other tissues. J Clin Invest 1977;60:724-70. 11. Terzis IK, Sweet RC, Dykes RW, Williams HB. Recover), of function in free muscle transplants using microneurovascular anastomoses. J Hand Surg 1978;3:37-59. 12. DelMaestro RF, Thaw HH, Bjork J, et al. Free radicals as mediators of tissue injurY". Acta Physiol Scand 1980;492 (Suppl).:43-57. 13. Harris K, Walker PM, Miclde DAG, et al. Metabolic response of skeletal muscle to ischemia. Am J Physiol 1985;19:H213H220. 14. Valdes ME, Landau SE, Shah DM, et al. Increased glomemlar filtration rate following mannitol administration in man. J Surg Res 1979;26:473-7. 15. Buchbinder D, Karmody AM, Leather RP, Shah DM. Hypertonic mannitol; its use in the prevention of revascularization syndrome after acute arterial ischemia. Arch Surg 1981; 116:414-21. 16. Hutton M, Rhodes RS, Chapman G. The lowering ofpostischemic compartment pressures with mannitol. J Surg Res 1982;32:239-42. 17. Magovern GJ Jr, Boiling SF, Casale AS, et al. The mechanism of mannitol in reducing ischemic injury: hypersmolarity or hydroxyl scavenger. Circulation 1984;70(Suppl I):I-91-I-95. 18. Patterson S, Klenerman L, Kiswas M, Rhodes A. The effect of pneumatic tourniquets on skeletal muscle physiology. Acta Orv,hop Scand 1981;52:171-5. 19. Santavirta S, Luoma A, Arstila AU. Ultrastrucmral changes in striated muscle after experimental tourniquet ischernia and short reflow. Eur Surg Res 1978;10:415-24. 20. Hargeus AR, Schmidt DA, Evans KL, et al. Quantitation of skeletal-muscle necrosis in a model compartment syndrome. J Bone Joint Surg [Am] 1981;63:63L6. 21. Mortenson WW, Hargens A1% Gershuni DH, et al. Longterm myoneural function after an induced compamnent syndrome in the canine hindlimb. Clin Orthop 1985;195:289i 93. 22. Lundborg G. Structure and function of the intraneural microvessels as related to trauma, edema formation, and nerve function. J Bone Joint Surg [Am] 1975;57:938-48. 23. Rorabeck CH, Clarke KM. The pathophysiology of the anterior tibial compartment syndrome: an experimental investigation. J Trauma 1978;18:299-304.
cular dysfianction. They hypothesized that this reperfusion syndrome is due to oxygen-derived free radical injury. Fasciotomy, which releases compartmental pressure, offered complete recovery from this insult. However, when free radical scavengers such as superoxide dismutase (SOD) or mannitol, is used, partial relief is achieved.
250
Ricci et al.
Our experience in the treatment of this syndrome is primarily based on the use o f hypertonic mannitol and fasciotomy. In an experimental design similar to this but using the entire lower limb, we have shown that after 90 minutes o f hypoperfusion followed by 2 hours of reperfusion, reperfusion syndrome occurs, manifested by decreased blood flow, increased resistance, decreased oxygen uptake, and muscle necrosis. On the basis of our experimental findings we have treated patients who have acute arterial ischemia with hypertonic mannitol before revascularization. The use o f hypertonic mannitol in conjunction with revascularization after acute arterial occlusion caused by thromboembolism decreased the necessity for fasciotomy. Only 3 of 24 patients required fasciotomy compared with the group of untreated patients, of which 13 of 17 patients required fasciotomy. Furthermore, the neuromuscular dysfunction was minimal and similar in both groups. After acute arterial occlusion from trauma to the popliteal artery, we treated 20 patients with mannitol; 15 patients were not treated with mannitol. Two o f the 20 patients receiving mannitol required fasciotomy and 12 of the 15 untreated patients required fasciotomy. Again, neuromuscular dysfunction was minimal in both groups and were similar. It is possible to compare this trauma group with this experimental model because of similar popliteal ischemia. There may be certain reasons for this difference in the effect of mannitol that the authors have seen and we have observed in terms of neuromuscular dysfunction. Although the authors have corrected for short half-life of SOD, they gave only a single treatment of mannitol. In the doses that they have used, if dogs had normal glomerular filtration rate, its half-life should be an hour or less. Therefore mannitol should have been repeated or continuously given to achieve maximal effect. The second critique is that the authors hypothesized that the injury is due to superoxide radical. If that is so, then why would only compartment decompression improve the situation? Do you have general hemodynamic data on these dogs relating to cardiac depression, hypotension, and thromboxane release? Finally, I suggest that although fasciotomy offered complete relief, if the authors had used marmitol in appropriate doses, it also would have achieved similar neuromuscular fimctional recovery in these dogs. Dr. William J. Quifiones-Baldrich (Los Angeles, Calif.). In our laboratory we have been interested in studying ischemia and reperfusion in a rabbit model of hind limb ischemia. One of the difficulties is measurement of skeletal muscle function in contrast to that of cardiac muscle, where function is rather easy to measure. We have noticed a significant difference in the recovery ofnervc and muscle after ischemia and reperfusion. In our model, peripheral nerve has poor recovery after 3 hours of ischemia in contrast to skeletal muscle, which has about 60% of control function after 3 hours of ischemia and 2 hours of reperfusion. When skeletal muscle fimction is studied, it is critical
Journal of VASCULAR SURGERY
to control for temperature. Temperature is a variable that will significantly alter function o f any muscle. In addition, it is important to determine the optimal resting tension of the muscle. The authors used 200 gm; I would like to know how that was determined. In two of the experimental groups the authors noted that there was significant edema. No edema was noted in the group with fasdotomy. This is interesting, considering that one o f the mechanisms by which fasciotomy works is by allowing the muscle to develop edema and thus not compromise reperfusion. Regardless, expressing muscle function as gram force per gram of tissue can be misleading because edema is present in some specimens and not in others. Dr. Ricci, did you control for temperature? Did you perform direct muscle stimulation and compare it with nerve stimulation? H o w was the optimal tension at rest determined? H o w do you adjust for the differences when expressing your results as gram force per gram tissue when you have variable edema in some of the specimens? Dr. Rieei (closing). Dr. Shah, of course we are we,. aware of your work and the work at Albany with mannitol; in fact that was, at least in part, a stimulus for our experiments. We did not repeat the dose, as you pointed out. We used a fairly large dose to start. We believed that a single dose would be more comparable to the single dose of SOD. Indeed if we were using these compounds as free radical scavengers, the damage done by the free radicals is immediate and that's when we needed to have the effects of SOD or marmitol. It may very well be that a drip might produce different results. We do not have hemodynamic data, as you suggested, with regards to fasciotomy. Fasciotomy certainly allows better perfusion of the muscle; I am not sure we can explain why it does not produce a reperfusion syndrome. Dr. Quifiones-Baldrich, we certainly agree it is difficult to study muscle function. Our next set of experiments is determined to study the nerve as well. As of now we have not done direct muscle stimulation, which may tell us more about which component of the neuromuscular problem most prominent. Temperature was controlled. Initially all the flaps were closed and the animal was kept warm. There were no special efforts to keep the muscle at the same temperature or warmed during the ischemic period or overnight. Again, the next day when muscle function testing was done; it was primarily the animal that was warmed. Two hundred grams were determined by some previous work that was done by our plastic surgery department with the same technique, which we adopted. In fact, I did not comment about edema. The fasciotomy produced less muscle necrosis and a lower L / R ratio. Edema in each of the legs judged by muscle weight was not significantly different. This is somewhat different than others have reported, but in our animals there is no significant difference in muscle weights; therefore we felt justified in expressing our results in terms of the muscle weight. In addition, the left leg is always compared with the right leg so each animal in a sense serves as his own control.