JOURNAL
OF SURGICAL
RESEARCH
51, 5-12 (1991)
Skeletal Muscle Function after Ischemia: “No Reflow” versus Reperfusion Injuty1p2 WILLIAM
J. QU&ONES-BALDRICH, MICHAEL
Department
of Surgery,
UCLA Medical
M.D., ARUN CHERVU, M.D., JUAN J. HERNANDEZ, COLBURN, M.D., AND WESLEY S. MOORE, M.D.
Center, Los Angeles, California; Submitted
for publication
and Sepulueda V.A. Medical
M.D.,
Center, Sepulueda, California
May 10, 1990
blood cells and platelets from the initial produced no better results than controlled alone. 0 1991 Academic Press, Inc.
The isometric contraction (supramaximal tetanic stimulation) of anterior tibialis muscle was studied in 32 New Zealand white rabbits after 5 hr of ischemia. Reperfusion was achieved after systemic heparinization (100 U/kg) by removal of vascular clamps (normal reperfusion, NR, iV = 10); isolated pump perfusion at 15 cc/min for 30 min followed by normal reperfusion (controlled reperfusion, CR, N = 8); CR with a Sepacell 500 filter in the circuit (leukopenic, thrombocytopenic, controlled reperfusion, L/TR, N = 9); or adding 25,000 U of urokinase to the initial reperfusate (UKR, N = 5). Experimental muscle is compared to control nonischemit contralateral muscle in each animal and expressed as percentage of control function. Specimens were studied by light microscopy. No significant difference in mean function at 2 hr was seen between the four groups, with NLR having 53% of control function, CR 55% of control function, L/TR 61% of control function, and UKR 48% of control function. “No reflow,” as defined by the absence of Doppler flow signals over the muscle pedicle with no recovery of function during reperfusion and continued incidence of persistent ischemia, was seen in NLR 4/10, CR 5/8, and L/TR 6/9 preparations with arteriolar, capillary, and venule thrombi documented by light microscopy. In contrast, “no reflow” was not seen in UKR (O/5, P < 0.05). Peak function at any interval (potential maximal recovery) in muscles that adequately reperfused was best in CR (73%) and L/TR (73%). No difference in the degree of injury in adequately reperfused muscles was seen between the four groups. These experiments suggest that “no reflow” (poor reperfusion) is an important component of reperfusion injury in skeletal muscle after 5 hr of ischemia. It suggests that this phenomenon is likely due to microvascular thrombus formed during ischemia and may respond to fibrinolysis during early reperfusion. In the absence of “no reflow,” removing white
reperfusate reperfusion
INTRODUCTION
Injury to cells following an ischemic interval has been shown to occur mostly during the early phases of reperfusion. Skeletal muscle is unique in its ability to tolerate periods of ischemia that are relatively long, compared to other tissues. The clinical manifestations following reperfusion of an acutely ischemic limb correlate best with the severity and length of the ischemic interval. Whereas a moderately ischemic limb may be revascularized without significant sequellae after long periods of ischemia, a severely ischemic limb will manifest significant changes after revascularization following a shorter ischemic interval. Research in the area of ischemia and reperfusion of skeletal muscle has concentrated on avoidance of the injury that occurs secondary to reestablishment of blood flow. In an effort to maximally recover potentially viable tissue, and based on current understanding of the consequences of reperfusion, oxygen and hydroxy radical scavengers, prostaglandin analogs, white cell inhibitors, enriched substrates, controlled rate of reperfusion, or combinations of these have been shown in one way or another to improve muscle viability after prolonged ischemia and variable periods of reperfusion [l-4]. Most of these investigators have used vital stain or other histologic criteria of viability to test their hypotheses. Skeletal muscle function as an indicator of viability has received sporadic interest mainly because of the difficulty in obtaining accurate measurements controlling important parameters such as temperature and resting tension [5]. These experiments were undertaken in an effort to evaluate the effects of reperfusion with blood (normal reperfusion, NR), controlled reperfusion at baseline flow (CR), leukopenic and thrombocytopenic controlled reperfusion (L/TR), or fibrinolytic reperfusion using uro-
’ This research is supported in part by a Merit Review Grant, Sepulveda V.A. Medical Center, Sepulveda, CA. ’ Presented at the Annual Meeting of the Association for Academic Surgery, Louisville, KY, November 15-18, 1989. 5
0022.4804/91
$1.50
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
6
JOURNAL
kinase (UKR) after prolonged cle using recovery of function viability [ 61. MATERIALS
AND
OF SURGICAL
RESEARCH:
ischemia of skeletal musas an indicator of tissue
METHODS
Thirty-two adult male New Zealand white rabbits weighing 3.5-4.0 kg each were anesthetized with 40-50 mg/kg of ketamine and lo-15 mg/kg of thorazine intramuscularly. The details of the preparation have been previously reported [6]. The abdomen is opened using a lower midline incision and the infrarenal aorta, bifurcation, and bilateral iliac arteries (internal and external) are dissected and all its branches identified. To eliminate collateral flow around the pelvis, the internal iliac and lateral circumflex arteries are ligated bilaterally. In addition, all lumbar branches originating from the infrarenal aorta are also ligated. The common femoral arteries are isolated through bilateral groin incisions. This dissection provides a nonbranching arterial conduit composed of the common and external iliac arteries from the infrarenal aorta to the femoral system in the thigh. Arterial flow is tested in each animal by Doppler flow signals using a Doppler probe over the origin of the anterior tibialis muscle pedicle. The venous system is left intact. In each animal, the right hind limb is made ischemic by applying microvascular atraumatic clamps on the right common iliac and common femoral arteries. Blood flow to the ischemic right hind limb is again assessed by Doppler probe and, in pilot studies, by arteriography. The left hind limb in each animal serves as an internal control and is similarly dissected without arterial clamping. The abdominal and groin incisions are closed and the anesthetized animal is placed under a warming lamp for the ischemic interval of 5 hr. During the ischemic interval, the animal’s vital signs are monitored. Blood pressure and heart rate are monitored and maintained with a carotid arterial line, and the core body temperature is measured with an intramuscular needle temperature probe connected to an analog recorder. Intravenous fluid boluses of normal saline are administered, as needed, to maintain normal systolic blood pressure of 60-70 mm Hg. Prior to reperfusion, the rabbits are placed supine on the operating board. Longitudinal incisions are made over both anterior tibialis muscles and fasciotomies performed. The anterior tibialis muscle tendons are detached distally and the muscles freed proximally to the level of their neurovascular pedicle. The vascular pedicles are carefully protected and preserved, and the broad muscular attachments to the adjacent tibia not dissected. Two orthopedic screws are used to immobilize each tibia to the operating board. The anterior tibialis tendons are attached to force transducers (Grass FTlO) using a steel wire and equal muscle length is maintained with a l-mm gradiation caliper. The muscles are continually moistened with warm saline gauze and their tem-
VOL.
51, NO. 1, JULY
1991
perature is recorded with a 22-gauge needle thermistor probe (YSI 500 series). A heat lamp is used to maintain ambient temperature and to minimize heat loss, maintaining equal muscle temperatures between 92 and 94°F. At the end of 5 hr of ischemia, reperfusion of the right hind limb is achieved by one of four separate methods. All animals, except the urokinase group, received 100 U/kg of intravenous heparin just prior to reperfusion to maintain clinical relevance. The urokinase group did not receive heparin to avoid excessive bleeding of dissected tissues. In 10 animals, reperfusion was obtained by removing the microvascular clamp and simply restoring normal hind limb reperfusion (NR, N = 10). In the second group of eight animals, the right lateral circumflex artery (distal to common iliac clamp but proximal to femoral clamp) is cannulated with a 22-gauge catheter and a controlled isolated pump perfusion through the circumflex artery into the femoral system (inflow to pump from left carotid cannula) started by removal of the femoral clamp and maintaining a rate of 15 cc/min for 30 min, followed by normal perfusion with removal of the common iliac clamp (CR, N = 8). Pilot experiments had shown normal blood flow in the iliac arteries of rabbits this size averaged between 14 and 18 cc/min. Therefore, our attempt was to maintain baseline flow and not allow the initial hyperemic response. The third group of nine animals underwent similar controlled reperfusion with the addition of a Sepacell500 (white blood cell and platelet) filter in the reperfusion circuit (between carotid cannula and pump). Systemic and reperfusate white blood cell count and platelet counts were obtained at the initiation and at the end of the controlled reperfusion interval (30 min). Thus, this group of animals underwent controlled leukopenic, thrombocytopenic reperfusion for the first 30 min, followed by normal perfusion with removal of the common iliac microvascular clamp (L/TR, N = 9). In the fourth group of five animals, reperfusion was achieved after the 5-hr ischemic interval by cannulation of the circumflex iliac artery and administration of 25,000 U of urokinase and removal of the microvascular clamps, thus allowing normal blood flow with the fibrinolytic agent (UKR, N = 5). In all animals, reperfusion was assessed by Doppler flow signals over the anterior tibialis vascular pedicle. The pattern of reperfusion was considered adequate if Doppler signals were heard over the muscle pedicle. “No reflow” was defined as absence of Doppler signals over the pedicle, with poor or no recovery of function, and persistent evidence of ischemia by gross inspection. Macroscopic clot was not present in the iliac and femoral vessels, with Doppler signals documenting flow in these arteries in all preparations. Reperfusion was continued for 2 hr in each animal. Immediately prior to and then at 30-min intervals during the reperfusion period, the isometric contractile response of both the experimental and control anterior tibialis muscle was tested. Supramaximal tetanic stimu-
QUINONES-BALDRICH
TABLE Number
ET AL.:
MUSCLE
1
of Limbs in Each Group to Pattern of Reperfusion
According
Reperfusion “No reflow” Normal reperfusion (N = 10) Controlled reperfusion \ (N = 8) Leukopenic/thrombocytopenic controlled reperfusion (N = 9) Fibrinolytic reperfusion
(N=5)
6
4
3
5
3
6
5*
0
*P i 0.05
lation of the muscle was achieved with a square wave pulse generator, generating 80 V with a pulse wave of 1 msec, a pulse interval of 10 msec, and a train duration of 250 msec. The stimulus, as generated by the stimulator (Grass Sll), was applied with platinum foils directly to the muscle. Muscle force was recorded on both a Grass 79-D polygraph and a Textronix 2230 digital storage oscilloscope. The maximum force (voltage) was recorded for each stimulation. Isometric contractile force obtained by direct muscle stimulation of the experimental ischemic reperfused limb was compared to that of the contralateral control limb, and the results are expressed as a ratio of control function. At the end of the 2-hr reperfusion period, the anterior tibialis muscles were harvested and the animal was sacrificed while still under anesthesia, using 0.5 ml/kg of T-61 Euthanasia solution. The muscles were fixed in a 10% buffered formalin solution and histologic slides were prepared using routine techniques for hematoxylin eosin staining. Specimens were studied by light microsCOPY*
Animal care complied with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals (NIH Publication 80-23, revised 1978). Statistical analysis was performed using Fisher’s exact (two-tail) test and the Wilcoxon rank sum. Results are expressed as means +. standard error of the mean (SEM).
FUNCTION
AFTER
out of 9 animals showed adequate reperfusion, with 6 showing no reflow. In contrast, fibrinolytic reperfusion was successful in reperfusing all five ischemic extremities. No reflow was not seen in the fibrinolytic reperfusion group. This difference is statistically significant, with a P value of t0.05. Poor reperfusion, or no reflow, invariably led to poor or no recovery of skeletalmuscle function. Muscles that did recover function are considered in the comparison of functional recovery after 5 hr of ischemia. All control muscles maintained their maximal contraction within 85% of the initial measurement at the beginning of reperfusion. Table 2 summarizes the ratio of the maximal tetanic contraction of the experimental, compared to the control, group after 5 hr of ischemia and 2 hr of reperfusion. No significant difference between the four groups can be seen. In fact, fibrinolytic reperfusion appears to have the lowest recovery at the end of 2 hr. This may be secondary to recovery of some function in muscles that would otherwise belong to the no reflow category. Leukopenic/thrombocytopenic controlled reperfusion, on the other hand, had better function at the end of 2 hr of reperfusion, although the difference was not statistically significant. Maximal recovery of function at any interval is a measure of potential recovery, and thus viable ischemic tissue. Table 2 summarizes the maximal recovery seen during the 2 hr of reperfusion for the four experimental groups. No statistically significant difference between the four groups is seen. Curiously, controlled reperfusion and leukopenic/thrombocytopenic controlled reperfusion had markedly similar maximal function. Figure 1 compares the four experimental groups during the 2 hr of reperfusion. It must be kept in mind that only those muscles that adequately reperfused are taken into consideration. This is done to separate the no reflow phenomenon from actual reperfusion injury. White blood cell count measurement, before and through the period of reperfusion, was not significantly TABLE Anterior
Tibialis Function Reperfusion (“No
2
after 5 hr of Ischemia Reflow” Excluded) Experimental/control After 2 hr reperfusion
RESULTS
Table 1 summarizes the patterns of reperfusion seen in the various groups. Of 10 animals that received normal reperfusion after the 5hr ischemic interval, six extremities adequately reperfused and four showed the no reflow phenomenon. In the controlled reperfusion group, only 3 animals adequately reperfused, whereas 5 evidenced no reflow. Similarly, in the leukopenicl thrombocytopenic controlled reperfusion group, only 3
7
ISCHEMIA
Normal
and
? SEM Maximal (any interval)
reperfusion
(N=6) Controlled
0.534 -t 0.069
0.574 k 0.073
0.552 +- 0.057
0.735 It 0.059
0.618 rt 0.151
0.733 + 0.102
0.482 f 0.066
0.656 + 0.111
reperfusion
(N=3) Leukopenic/thrombocytopenic Controlled reperfusion
(N=3) Fibrinolytic (N= 5)
reperfusion
a 1.0
JOURNAL
OF SURGICAL
RESEARCH:
EXPT/CONTROLFUNCTION
0.6 c 0.6 0.4 0.8 I 0
1 86
I 60
I so
I 180
TIME AFM3R FtEPERF’USION (MM) -
Nod
+
51, NO. 1, JULY
1991
show intact cellular andvascular structures (Fig. 3). Histologic sections of muscles in the fibrinolytic group showed minimal sludging, with mostly patent vascular systems. Figure 4 is typical of the group of muscles that showed adequate reperfusion after 5 hr of ischemia, with this particular section belonging to the fibrinolytic reperfusion group. Note that evidence of fiber injury is no different than that seen in Fig. 2, where no reflow occurred.
r
0.0 1
VOL.
controlled +
hkopenia
-a- FlbDinolgllc
FIG. 1. Experimental/control ratio of tetanic contraction to supramaximal stimulus during 2 hr of reperfusion for the four experimental groups.
different between the four groups just prior to reperfusion. The group with controlled reperfusion had a somewhat lower white blood cell count, probably secondary to trapping of cells in the circuit. The leukopenic/thrombocytopenic reperfusion group had undetectable white blood cells in the reperfusate for the first 30 min, with platelet count below 10,000. The white blood cell count normalized in the controlled reperfusion group by 60 min, with the leukopenic/thrombocytopenic group persistently showing a lower white blood cell count during the remainder of the reperfusion. At the end of 2 hr of reperfusion, the controlled reperfusion and leukopenic/ thrombocytopenic reperfusion groups had similar white blood cell counts, averaging 1300, with the normal reperfusion and the fibrinolytic groups showing similar white blood cell counts of around 3300. The only group that showed any significant decrease in the platelet count was the leukopenic/thrombocytopenic reperfusion group. In the initial 30 min of reperfusion in this group, the platelet count was less than 10,000. The systemic platelet count decreased to 20,000 in this group and increased to an average of 50,000 after filtration was discontinued. The white blood cell count findings are summarized in Table 3. Histologic examination of the control and experimental specimens showed marked difference between no reflow and adequately reperfused muscles. Regardless of the experimental group, muscles that suffered no reflow invariably showed extensive arteriolar, capillary, and venular occlusions. Histologic evidence of muscle fiber injury was seen in all four groups, without any significant difference noted on gross histologic examination. Figure 2 shows a typical section of a muscle with no reflow. This particular section belongs to the group where normal reperfusion was carried out. This appearance is markedly different than control muscles which
DISCUSSION
Several conclusions may be drawn from these experiments. Based on our previous experience, the no reflow phenomenon is usually not seen when the ischemic interval has been limited to less than 4 hr. When graded periods of ischemia of 1, 2, and 3 hr were studied in our laboratory, all muscles adequately reperfused, as evidenced by signals by Doppler flow detector over the pedicle and recovery of function. Some reduction in the recovery of function was seen after 3 hr of ischemia and 2 hr of reperfusion [7]. In the experiments herein described, the ischemic interval has been increased to 5 hr and in up to 40% of these preparations, no recovery of function is seen, with absent Doppler signals over the pedicle. The percentage of preparations that do reperfuse appear to recover somewhere between 40 and 50% of control function. The exact reason as to why this phenomenon is not more often reported may be related to the use of heparin prior to ischemia in some experimental preparations. In addition, the investigator may consider the experiment as having a technical problem, and thus is not included in the report. In analyzing those muscles that did reperfuse, it appears that potential for recovery (peak function at any interval) was similar in muscles where reperfusion was carried out in a controlled fashion with no additional benefit seen during removal of leukocytes and partial removal of platelets, suggesting a limited role for these components beyond that achieved by controlled reperfusion. In fact, removal of white blood cells did not prevent
TABLE Reperfusate
3 WBC Count
0
30 min
60 min
120 min
5.5 + 1.1
3.3 + 0.8
4.2 + 1.0
3.8 k 0.9
3.2 + 1.5
1.1 +
3.9
1.3 f 0.4
3.1 -c 1.1
0
4.2 r 0.3
3.6 + 0.4
Heparin
(N=6) Heparin pump (N= 3) Filter
(N=3)
.9
1.6 +
.l
1.4 + 0.3
Urokinase
(N=5)
3.8 + 0.6
3.3 + 0.5
QUINONES-BALDRICH
ET AL.:
MUSCLE
FUNCTION
AFTER
ISCHEMIA
FIG. 2. Histologic section of anterior tibialis muscle after 5 hr of &hernia and 2 hr of reperfusion (NR) showing extensive arteriolar, capillary, and venular occlusions typically seen in all muscles where “no reflow” occurred (H&E, 40X). Note histologic evidence of fiber injury.
the no reflow phenomenon, suggesting that white blood cells, in and of themselves, are not required in the establishment of no reflow. Thus, control of the initial hyperemit response in those tissues that are capable of reperfusing appears an important method of reducing reperfusion injury. This has been suggested by others using different preparations [8]. Continued ischemia after reestablishment of axial blood flow in the clinical arena has been reported since the early attempts at embolectomy [9]. Experimentally, Dunnant and Edwards documented this process using mongrel dogs. With the most severe ischemia, there was histologic evidence of thrombus formation in small muscular arteries. This was not observed when collateral blood flow was maintained during axial flow interruption. Radiologic visualization of these small arteries, however, was impaired even after shorter ischemic intervals, with resistance to flow progressively increasing after 6 hr of ischemia. The latter was ascribed to arteriolar thrombus which was documented histologically [lo]. The severity and incidence of the no reflow phenomenon appeared to be related to the length of the ischemic interval and the severity of the ischemia. Thus, in skeletal muscle where long periods of ischemia are required be-
fore significant injury is produced by reperfusion, the no reflow phenomenon may be seen with increasing frequency. When the rate and distribution of blood flow in muscle after prolonged ischemia are examined, reduced blood flow is noted after 5 hr of ischemia when compared to 4 hr of ischemia, suggesting that prolongation of the ischemic interval leads to increased resistance in the microcirculation with reduction in the rate of reflow. No improvement was seen with vasodilators, suggesting that local vasoconstriction is not a primary mechanism [ll]. Other factors may play a significant role in preventing adequate reperfusion of skeletal muscle after prolonged periods of ischemia. Cellular swelling occurs during ischemia and may impede flow during reperfusion [12-141. Perfusion with hyperosmolar agents, such as Mannitol, has been shown to improve reperfusion in skin flaps and other organs. Deformability of white blood cells has been suggested and observed to cause plugging in skeletal muscles of cats during shock [15]. Increasing vascular resistance is seen with an increasing number of leukocytes, specifically at low flows [16]. Our experiments, however, would suggest a limited role for white blood cells as contributors to the no reflow phenomenon, as
JOURNAL
FIG. 3.
OF SURGICAL
RESEARCH:
VOL. 51, NO. 1, JULY 1991
Histologic section of anterior tibialis control muscle (H&E, 40x). Note intact cellular and vascular structures.
leukopenic reperfusion did not prevent no reflow. Other contributors to this phenomenon may include capillary collapse with inability to reach an opening pressure because of the factors mentioned above, endothelial cell damage and dysfunction, and other as yet unrecognized changes occurring during both ischemia and reperfusion. Since the no reflow phenomenon has been observed in other organs such as kidneys, skin flaps, and brain [12,16-B], it appears to be an important component impeding successful retrieval of ischemic tissue. When histologic specimens are analyzed in control and experimental muscles, a remarkable difference is seen. Experimental muscles showing no reflow demonstrate marked red cell sludging and thrombi in arterioles, venules, and capillaries (Fig. 2). No such changes are seen in the control specimens (Fig. 3). Muscles treated with urokinase, on the other hand, resembled muscles from the control group, with the sporadic appearance of arteriolar sludging and thrombi (Fig. 4). This leads us to postulate that temporarily increasing fibrinolytic activity during the initial phase of reperfusion may revert changes, contributing to the no reflow phenomenon. Our experiments suggest that one of the principal causes of no reflow is small vessel occlusion
due to fibrin deposition, which in turn traps blood elements and is seen as sludging under the microscope. When the process is advanced, frank thrombus is seen. Platelets and white cells may be trapped in this process releasing oxygen radicals and, thus, contribute to reperfusion injury [3]. Endothelial cell damage, with opening of cellular gaps and exposure of the subendothelial elements secondary to processes during ischemia, would lead to fibrin deposition and potentially play a major role in triggering the no reflow phenomenon. Some of these changes have been shown to occur experimentally following ischemia and reperfusion [ 191. The fact that recovery of function in the urokinase-treated animals was no better and perhaps somewhat worse than those animals that did reperfuse with controlled reperfusion suggests a limited role for urokinase in preventing reperfusion injury. Nevertheless, the assurance of adequate reflow would open the possibilities for other manipulations to be effective in preventing reperfusion injury in the overall process. Thus, two major components of ischemia and reperfusion are suggested. On the one hand, the no reflow phenomenon, which prolongs the ischemia and prevents adequate restoration of flow, and second, cellular injury resulting from processes occurring be-
QUINONES-BALDRICH
ET AL.:
MUSCLE
FUNCTION
AFTER
ISCHEMIA
F‘IG. 4. Histologic section of anterior tibialis muscle after 5 hr of ischemia and fibrinolytic reperfusion. Note minimal patf :nt vascular system. There is evidence of fiber injury, no different than that in other experimental groups.
cause of restoration of flow. The latter is probably secondary to a multitude of factors, some of which will need to be addressed for successful retrieval of the ischemic cell. Our goal is to develop a clinically applicable method of reperfusion in severely ischemic limbs. In this regard, the intraoperative administration of fibrinolytic agent has been reported. In our own experience, both clinical and experimental data would suggest that they can be used with safety and efficacy in these acutely ischemic limbs [20-221. More recently, we have tried intraoperative and continuous postoperative fibrinolytic therapy in patients with severely ischemic extremeties. In these patients, we have performed operative thrombectomy and placement of an arterial catheter, starting a urokinase infusion intraoperatively and continuing into the postoperative period. In six patients treated, the results have been encouraging [23]. Although the problem of reperfusion injury in ischemit skeletal muscle has yet to be resolved, accumulating evidence suggests that controlled reperfusion may be a clinically feasible and effective alternative. These experiments and those of others suggest an additive role for fibrinolytic agents during the initial phases of reper-
sludging with mo
fusion [24]. This may improve our ability to reestablish flow in areas where fibrin deposition has occurred, and thus limit the ischemic insult. Additional experiments are required to determine other manipulations during the early phases of reperfusion which may have additional beneficial effects in restoring viability to skeletal muscle after prolonged periods of ischemia. REFERENCES 1.
2.
3.
Lee, K. R., Cronenwett, J. L., Shlafer, M., Corpron, C., and Zelenock, G. B. Effect of superoxide dismutase plus catalase on calcium transport in ischemic and reperfused skeletal muscle. J. Surg. Res. 42: 24-32, 1987. Perry, M. O., and Fantini, G. Ischemia: Profile of an enemy. Reperfusion injury of skeletal muscle. J. Vast. Surg. 6: 231-234, 1987. Beyersdorf, F., Matheis, G., Kruger, S., Hanselmann, A., Freisleben, H. G., Zimmer, G., and Satter, P. Avoiding reperfusion injury after limb revascularization: Experimental observations and recommendations for clinical application. J. Vusc. Surg. 9: 757-
766,1989. 4.
5.
Belkin, M., LaMorte, W. L., Wright, J. G., and Hobson, R. W. The role of leukocytes in the pathophysiology of skeletal muscle ischemic injury. J. Vmc. Surg. 10: 14-19, 1989. Fish, J. S., McKee, N. H., Pynn, B. R., Kuzon, W. M., and Plyley,
12
6.
7.
8.
9.
10.
11.
12.
13. 14.
JOURNAL
OF SURGICAL
RESEARCH:
M. J. Isometric contractile function recovery following tourniquet ischemia. J. Surg. Res. 47: 365-370, 1989. Chervu, A., Moore, W. S., Homsher, E., and Quifiones-Baldrich, W. J. Differential recovery of skeletal muscle and peripheral nerve function after ischemia and reperfusion. J. Surg. Res. 47: 12-19,1989. Chervu, A., Moore, W. S., and Quifiones-Baldrich, W. J. Skeletal muscle function after ischemia and reperfusion. Presented at the Society of University Surgeons 49th Annual Meeting, San Antonio, Texas. February 12, 1988. Wright, J. G., Fox, D., Kerr, J. C., Valeri, C. R., and Hobson, R. W. Rate of reperfusion blood flow modulates reperfusion injury in skeletal muscle. J. Surg. Res. 44: 754-764, 1988. Green, R. M., DeWeese, J. A., and Rob, C. G. Arterial embolectomy before and after the Fogerty catheter. Surgery 77: 24-33, 1975. Dunnant, J. H., and Edwards, W. S. Small vessel occlusion in the extremity after various periods of arterial obstruction: An experimental study. Surgery 73: 240-2451973. Forrest, I., Lindsay, T., Romaschin, A., and Walker, P. The rate and distribution of muscle blood flow after prolonged ischemia. J. Vast. Surg. 10: 83-88,1989. Chait, L. A., May, J. W., O’Brien, B. M., and Hurley, J. V. The effects of the perfusion of various solutions on the no reflow phenomenon in experimental free flaps. Pk. Recon. Surg. 61: 421430, 1978. Leaf, A. Cell swelling: A factor in ischemic tissue injury. Circulution 8: 455, 1963. Strock, P. E., and Majno, G. Microvascular changes in experimental tourniquet ischemia. Surg. Gynecol. O&et. 129: 12131223,1969.
15.
16.
17. 18. 19.
20.
21.
22.
23.
24.
VOL.
51, NO. 1, JULY
1991
Bagge, U., Maundson, B., and Lauritzen, C. White blood cell deformability and plugging of skeletal muscle capillaries in hemorrhagic shock. Actu Physiol. Stand. 180: 159-163,198O. Braide, M., Amundson, B., Chien, S., and Bagge, U. Quantitative studies on the influence of leukocytes on the vascular resistance in a skeletal muscle preparation. Microvasc. Res. 27: 331-352, 1984. Ames, A., Wright, R. L., Kowada, M., et al. Cerebral ischemiaThe no reflow phenomenon. Am. J. Pathol. 52: 437-453,1968. Summers, W. K., and Jamison, R. L. The no reflow phenomenon in renal ischemia. Lab. Invest. 25: 635, 1971. Fonkalsrud, E. W., Sanchez, M., Zerubavel, R., Lassaletta, L., Smeesters, C., and Mahoney, A. Arterial endothelial changes after ischemia and perfusion. Surg. Gynecol. Obstet. 142: 715721,1976. Quiiiones-Baldrich, W. J., Zierler, E., and Hiatt, J. R. Intraoperative fibrinolytic therapy: An adjunct to catheter thromboembolectomy. J. Vast. Surg. 2: 319-326, 1985. Quiiiones-Baldrich, W. J., Baker, J. D., Busuttil, R. W., Machleder, H. I., and Moore, W. S. Intraoperative infusion of lytic drugs for thrombotic complications of revaacularization. J. Vast. Surg. 10: 408, 1989. Quiiiones-Baldrich, W. J., Ziomek, S., Henderson, T., and Moore, W. S. Intraoperative fibrinolytic therapy: Experimental evaluation. J. Vast. Surg. 4: 229-236, 1986. Quifiones-Baldrich, W. J. The role of fibrinolysis during reperfusion of ischemic skeletal muscle. Microcirculation, Endothelium, and Lymphutics 5: 299-314,1989. Belkin, M., Valeri, C. R., and Hobson, R. W. Intraarterial urokinase increases skeletal muscle viability after acute ischemia. J. Vax. Surg. 9: 161-168, 1989.
OF SURGICAL
RESEARCH
51, 5-12 (1991)
Skeletal Muscle Function after Ischemia: “No Reflow” versus Reperfusion Injuty1p2 WILLIAM
J. QU&ONES-BALDRICH, MICHAEL
Department
of Surgery,
UCLA Medical
M.D., ARUN CHERVU, M.D., JUAN J. HERNANDEZ, COLBURN, M.D., AND WESLEY S. MOORE, M.D.
Center, Los Angeles, California; Submitted
for publication
and Sepulueda V.A. Medical
M.D.,
Center, Sepulueda, California
May 10, 1990
blood cells and platelets from the initial produced no better results than controlled alone. 0 1991 Academic Press, Inc.
The isometric contraction (supramaximal tetanic stimulation) of anterior tibialis muscle was studied in 32 New Zealand white rabbits after 5 hr of ischemia. Reperfusion was achieved after systemic heparinization (100 U/kg) by removal of vascular clamps (normal reperfusion, NR, iV = 10); isolated pump perfusion at 15 cc/min for 30 min followed by normal reperfusion (controlled reperfusion, CR, N = 8); CR with a Sepacell 500 filter in the circuit (leukopenic, thrombocytopenic, controlled reperfusion, L/TR, N = 9); or adding 25,000 U of urokinase to the initial reperfusate (UKR, N = 5). Experimental muscle is compared to control nonischemit contralateral muscle in each animal and expressed as percentage of control function. Specimens were studied by light microscopy. No significant difference in mean function at 2 hr was seen between the four groups, with NLR having 53% of control function, CR 55% of control function, L/TR 61% of control function, and UKR 48% of control function. “No reflow,” as defined by the absence of Doppler flow signals over the muscle pedicle with no recovery of function during reperfusion and continued incidence of persistent ischemia, was seen in NLR 4/10, CR 5/8, and L/TR 6/9 preparations with arteriolar, capillary, and venule thrombi documented by light microscopy. In contrast, “no reflow” was not seen in UKR (O/5, P < 0.05). Peak function at any interval (potential maximal recovery) in muscles that adequately reperfused was best in CR (73%) and L/TR (73%). No difference in the degree of injury in adequately reperfused muscles was seen between the four groups. These experiments suggest that “no reflow” (poor reperfusion) is an important component of reperfusion injury in skeletal muscle after 5 hr of ischemia. It suggests that this phenomenon is likely due to microvascular thrombus formed during ischemia and may respond to fibrinolysis during early reperfusion. In the absence of “no reflow,” removing white
reperfusate reperfusion
INTRODUCTION
Injury to cells following an ischemic interval has been shown to occur mostly during the early phases of reperfusion. Skeletal muscle is unique in its ability to tolerate periods of ischemia that are relatively long, compared to other tissues. The clinical manifestations following reperfusion of an acutely ischemic limb correlate best with the severity and length of the ischemic interval. Whereas a moderately ischemic limb may be revascularized without significant sequellae after long periods of ischemia, a severely ischemic limb will manifest significant changes after revascularization following a shorter ischemic interval. Research in the area of ischemia and reperfusion of skeletal muscle has concentrated on avoidance of the injury that occurs secondary to reestablishment of blood flow. In an effort to maximally recover potentially viable tissue, and based on current understanding of the consequences of reperfusion, oxygen and hydroxy radical scavengers, prostaglandin analogs, white cell inhibitors, enriched substrates, controlled rate of reperfusion, or combinations of these have been shown in one way or another to improve muscle viability after prolonged ischemia and variable periods of reperfusion [l-4]. Most of these investigators have used vital stain or other histologic criteria of viability to test their hypotheses. Skeletal muscle function as an indicator of viability has received sporadic interest mainly because of the difficulty in obtaining accurate measurements controlling important parameters such as temperature and resting tension [5]. These experiments were undertaken in an effort to evaluate the effects of reperfusion with blood (normal reperfusion, NR), controlled reperfusion at baseline flow (CR), leukopenic and thrombocytopenic controlled reperfusion (L/TR), or fibrinolytic reperfusion using uro-
’ This research is supported in part by a Merit Review Grant, Sepulveda V.A. Medical Center, Sepulveda, CA. ’ Presented at the Annual Meeting of the Association for Academic Surgery, Louisville, KY, November 15-18, 1989. 5
0022.4804/91
$1.50
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
6
JOURNAL
kinase (UKR) after prolonged cle using recovery of function viability [ 61. MATERIALS
AND
OF SURGICAL
RESEARCH:
ischemia of skeletal musas an indicator of tissue
METHODS
Thirty-two adult male New Zealand white rabbits weighing 3.5-4.0 kg each were anesthetized with 40-50 mg/kg of ketamine and lo-15 mg/kg of thorazine intramuscularly. The details of the preparation have been previously reported [6]. The abdomen is opened using a lower midline incision and the infrarenal aorta, bifurcation, and bilateral iliac arteries (internal and external) are dissected and all its branches identified. To eliminate collateral flow around the pelvis, the internal iliac and lateral circumflex arteries are ligated bilaterally. In addition, all lumbar branches originating from the infrarenal aorta are also ligated. The common femoral arteries are isolated through bilateral groin incisions. This dissection provides a nonbranching arterial conduit composed of the common and external iliac arteries from the infrarenal aorta to the femoral system in the thigh. Arterial flow is tested in each animal by Doppler flow signals using a Doppler probe over the origin of the anterior tibialis muscle pedicle. The venous system is left intact. In each animal, the right hind limb is made ischemic by applying microvascular atraumatic clamps on the right common iliac and common femoral arteries. Blood flow to the ischemic right hind limb is again assessed by Doppler probe and, in pilot studies, by arteriography. The left hind limb in each animal serves as an internal control and is similarly dissected without arterial clamping. The abdominal and groin incisions are closed and the anesthetized animal is placed under a warming lamp for the ischemic interval of 5 hr. During the ischemic interval, the animal’s vital signs are monitored. Blood pressure and heart rate are monitored and maintained with a carotid arterial line, and the core body temperature is measured with an intramuscular needle temperature probe connected to an analog recorder. Intravenous fluid boluses of normal saline are administered, as needed, to maintain normal systolic blood pressure of 60-70 mm Hg. Prior to reperfusion, the rabbits are placed supine on the operating board. Longitudinal incisions are made over both anterior tibialis muscles and fasciotomies performed. The anterior tibialis muscle tendons are detached distally and the muscles freed proximally to the level of their neurovascular pedicle. The vascular pedicles are carefully protected and preserved, and the broad muscular attachments to the adjacent tibia not dissected. Two orthopedic screws are used to immobilize each tibia to the operating board. The anterior tibialis tendons are attached to force transducers (Grass FTlO) using a steel wire and equal muscle length is maintained with a l-mm gradiation caliper. The muscles are continually moistened with warm saline gauze and their tem-
VOL.
51, NO. 1, JULY
1991
perature is recorded with a 22-gauge needle thermistor probe (YSI 500 series). A heat lamp is used to maintain ambient temperature and to minimize heat loss, maintaining equal muscle temperatures between 92 and 94°F. At the end of 5 hr of ischemia, reperfusion of the right hind limb is achieved by one of four separate methods. All animals, except the urokinase group, received 100 U/kg of intravenous heparin just prior to reperfusion to maintain clinical relevance. The urokinase group did not receive heparin to avoid excessive bleeding of dissected tissues. In 10 animals, reperfusion was obtained by removing the microvascular clamp and simply restoring normal hind limb reperfusion (NR, N = 10). In the second group of eight animals, the right lateral circumflex artery (distal to common iliac clamp but proximal to femoral clamp) is cannulated with a 22-gauge catheter and a controlled isolated pump perfusion through the circumflex artery into the femoral system (inflow to pump from left carotid cannula) started by removal of the femoral clamp and maintaining a rate of 15 cc/min for 30 min, followed by normal perfusion with removal of the common iliac clamp (CR, N = 8). Pilot experiments had shown normal blood flow in the iliac arteries of rabbits this size averaged between 14 and 18 cc/min. Therefore, our attempt was to maintain baseline flow and not allow the initial hyperemic response. The third group of nine animals underwent similar controlled reperfusion with the addition of a Sepacell500 (white blood cell and platelet) filter in the reperfusion circuit (between carotid cannula and pump). Systemic and reperfusate white blood cell count and platelet counts were obtained at the initiation and at the end of the controlled reperfusion interval (30 min). Thus, this group of animals underwent controlled leukopenic, thrombocytopenic reperfusion for the first 30 min, followed by normal perfusion with removal of the common iliac microvascular clamp (L/TR, N = 9). In the fourth group of five animals, reperfusion was achieved after the 5-hr ischemic interval by cannulation of the circumflex iliac artery and administration of 25,000 U of urokinase and removal of the microvascular clamps, thus allowing normal blood flow with the fibrinolytic agent (UKR, N = 5). In all animals, reperfusion was assessed by Doppler flow signals over the anterior tibialis vascular pedicle. The pattern of reperfusion was considered adequate if Doppler signals were heard over the muscle pedicle. “No reflow” was defined as absence of Doppler signals over the pedicle, with poor or no recovery of function, and persistent evidence of ischemia by gross inspection. Macroscopic clot was not present in the iliac and femoral vessels, with Doppler signals documenting flow in these arteries in all preparations. Reperfusion was continued for 2 hr in each animal. Immediately prior to and then at 30-min intervals during the reperfusion period, the isometric contractile response of both the experimental and control anterior tibialis muscle was tested. Supramaximal tetanic stimu-
QUINONES-BALDRICH
TABLE Number
ET AL.:
MUSCLE
1
of Limbs in Each Group to Pattern of Reperfusion
According
Reperfusion “No reflow” Normal reperfusion (N = 10) Controlled reperfusion \ (N = 8) Leukopenic/thrombocytopenic controlled reperfusion (N = 9) Fibrinolytic reperfusion
(N=5)
6
4
3
5
3
6
5*
0
*P i 0.05
lation of the muscle was achieved with a square wave pulse generator, generating 80 V with a pulse wave of 1 msec, a pulse interval of 10 msec, and a train duration of 250 msec. The stimulus, as generated by the stimulator (Grass Sll), was applied with platinum foils directly to the muscle. Muscle force was recorded on both a Grass 79-D polygraph and a Textronix 2230 digital storage oscilloscope. The maximum force (voltage) was recorded for each stimulation. Isometric contractile force obtained by direct muscle stimulation of the experimental ischemic reperfused limb was compared to that of the contralateral control limb, and the results are expressed as a ratio of control function. At the end of the 2-hr reperfusion period, the anterior tibialis muscles were harvested and the animal was sacrificed while still under anesthesia, using 0.5 ml/kg of T-61 Euthanasia solution. The muscles were fixed in a 10% buffered formalin solution and histologic slides were prepared using routine techniques for hematoxylin eosin staining. Specimens were studied by light microsCOPY*
Animal care complied with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals (NIH Publication 80-23, revised 1978). Statistical analysis was performed using Fisher’s exact (two-tail) test and the Wilcoxon rank sum. Results are expressed as means +. standard error of the mean (SEM).
FUNCTION
AFTER
out of 9 animals showed adequate reperfusion, with 6 showing no reflow. In contrast, fibrinolytic reperfusion was successful in reperfusing all five ischemic extremities. No reflow was not seen in the fibrinolytic reperfusion group. This difference is statistically significant, with a P value of t0.05. Poor reperfusion, or no reflow, invariably led to poor or no recovery of skeletalmuscle function. Muscles that did recover function are considered in the comparison of functional recovery after 5 hr of ischemia. All control muscles maintained their maximal contraction within 85% of the initial measurement at the beginning of reperfusion. Table 2 summarizes the ratio of the maximal tetanic contraction of the experimental, compared to the control, group after 5 hr of ischemia and 2 hr of reperfusion. No significant difference between the four groups can be seen. In fact, fibrinolytic reperfusion appears to have the lowest recovery at the end of 2 hr. This may be secondary to recovery of some function in muscles that would otherwise belong to the no reflow category. Leukopenic/thrombocytopenic controlled reperfusion, on the other hand, had better function at the end of 2 hr of reperfusion, although the difference was not statistically significant. Maximal recovery of function at any interval is a measure of potential recovery, and thus viable ischemic tissue. Table 2 summarizes the maximal recovery seen during the 2 hr of reperfusion for the four experimental groups. No statistically significant difference between the four groups is seen. Curiously, controlled reperfusion and leukopenic/thrombocytopenic controlled reperfusion had markedly similar maximal function. Figure 1 compares the four experimental groups during the 2 hr of reperfusion. It must be kept in mind that only those muscles that adequately reperfused are taken into consideration. This is done to separate the no reflow phenomenon from actual reperfusion injury. White blood cell count measurement, before and through the period of reperfusion, was not significantly TABLE Anterior
Tibialis Function Reperfusion (“No
2
after 5 hr of Ischemia Reflow” Excluded) Experimental/control After 2 hr reperfusion
RESULTS
Table 1 summarizes the patterns of reperfusion seen in the various groups. Of 10 animals that received normal reperfusion after the 5hr ischemic interval, six extremities adequately reperfused and four showed the no reflow phenomenon. In the controlled reperfusion group, only 3 animals adequately reperfused, whereas 5 evidenced no reflow. Similarly, in the leukopenicl thrombocytopenic controlled reperfusion group, only 3
7
ISCHEMIA
Normal
and
? SEM Maximal (any interval)
reperfusion
(N=6) Controlled
0.534 -t 0.069
0.574 k 0.073
0.552 +- 0.057
0.735 It 0.059
0.618 rt 0.151
0.733 + 0.102
0.482 f 0.066
0.656 + 0.111
reperfusion
(N=3) Leukopenic/thrombocytopenic Controlled reperfusion
(N=3) Fibrinolytic (N= 5)
reperfusion
a 1.0
JOURNAL
OF SURGICAL
RESEARCH:
EXPT/CONTROLFUNCTION
0.6 c 0.6 0.4 0.8 I 0
1 86
I 60
I so
I 180
TIME AFM3R FtEPERF’USION (MM) -
Nod
+
51, NO. 1, JULY
1991
show intact cellular andvascular structures (Fig. 3). Histologic sections of muscles in the fibrinolytic group showed minimal sludging, with mostly patent vascular systems. Figure 4 is typical of the group of muscles that showed adequate reperfusion after 5 hr of ischemia, with this particular section belonging to the fibrinolytic reperfusion group. Note that evidence of fiber injury is no different than that seen in Fig. 2, where no reflow occurred.
r
0.0 1
VOL.
controlled +
hkopenia
-a- FlbDinolgllc
FIG. 1. Experimental/control ratio of tetanic contraction to supramaximal stimulus during 2 hr of reperfusion for the four experimental groups.
different between the four groups just prior to reperfusion. The group with controlled reperfusion had a somewhat lower white blood cell count, probably secondary to trapping of cells in the circuit. The leukopenic/thrombocytopenic reperfusion group had undetectable white blood cells in the reperfusate for the first 30 min, with platelet count below 10,000. The white blood cell count normalized in the controlled reperfusion group by 60 min, with the leukopenic/thrombocytopenic group persistently showing a lower white blood cell count during the remainder of the reperfusion. At the end of 2 hr of reperfusion, the controlled reperfusion and leukopenic/ thrombocytopenic reperfusion groups had similar white blood cell counts, averaging 1300, with the normal reperfusion and the fibrinolytic groups showing similar white blood cell counts of around 3300. The only group that showed any significant decrease in the platelet count was the leukopenic/thrombocytopenic reperfusion group. In the initial 30 min of reperfusion in this group, the platelet count was less than 10,000. The systemic platelet count decreased to 20,000 in this group and increased to an average of 50,000 after filtration was discontinued. The white blood cell count findings are summarized in Table 3. Histologic examination of the control and experimental specimens showed marked difference between no reflow and adequately reperfused muscles. Regardless of the experimental group, muscles that suffered no reflow invariably showed extensive arteriolar, capillary, and venular occlusions. Histologic evidence of muscle fiber injury was seen in all four groups, without any significant difference noted on gross histologic examination. Figure 2 shows a typical section of a muscle with no reflow. This particular section belongs to the group where normal reperfusion was carried out. This appearance is markedly different than control muscles which
DISCUSSION
Several conclusions may be drawn from these experiments. Based on our previous experience, the no reflow phenomenon is usually not seen when the ischemic interval has been limited to less than 4 hr. When graded periods of ischemia of 1, 2, and 3 hr were studied in our laboratory, all muscles adequately reperfused, as evidenced by signals by Doppler flow detector over the pedicle and recovery of function. Some reduction in the recovery of function was seen after 3 hr of ischemia and 2 hr of reperfusion [7]. In the experiments herein described, the ischemic interval has been increased to 5 hr and in up to 40% of these preparations, no recovery of function is seen, with absent Doppler signals over the pedicle. The percentage of preparations that do reperfuse appear to recover somewhere between 40 and 50% of control function. The exact reason as to why this phenomenon is not more often reported may be related to the use of heparin prior to ischemia in some experimental preparations. In addition, the investigator may consider the experiment as having a technical problem, and thus is not included in the report. In analyzing those muscles that did reperfuse, it appears that potential for recovery (peak function at any interval) was similar in muscles where reperfusion was carried out in a controlled fashion with no additional benefit seen during removal of leukocytes and partial removal of platelets, suggesting a limited role for these components beyond that achieved by controlled reperfusion. In fact, removal of white blood cells did not prevent
TABLE Reperfusate
3 WBC Count
0
30 min
60 min
120 min
5.5 + 1.1
3.3 + 0.8
4.2 + 1.0
3.8 k 0.9
3.2 + 1.5
1.1 +
3.9
1.3 f 0.4
3.1 -c 1.1
0
4.2 r 0.3
3.6 + 0.4
Heparin
(N=6) Heparin pump (N= 3) Filter
(N=3)
.9
1.6 +
.l
1.4 + 0.3
Urokinase
(N=5)
3.8 + 0.6
3.3 + 0.5
QUINONES-BALDRICH
ET AL.:
MUSCLE
FUNCTION
AFTER
ISCHEMIA
FIG. 2. Histologic section of anterior tibialis muscle after 5 hr of &hernia and 2 hr of reperfusion (NR) showing extensive arteriolar, capillary, and venular occlusions typically seen in all muscles where “no reflow” occurred (H&E, 40X). Note histologic evidence of fiber injury.
the no reflow phenomenon, suggesting that white blood cells, in and of themselves, are not required in the establishment of no reflow. Thus, control of the initial hyperemit response in those tissues that are capable of reperfusing appears an important method of reducing reperfusion injury. This has been suggested by others using different preparations [8]. Continued ischemia after reestablishment of axial blood flow in the clinical arena has been reported since the early attempts at embolectomy [9]. Experimentally, Dunnant and Edwards documented this process using mongrel dogs. With the most severe ischemia, there was histologic evidence of thrombus formation in small muscular arteries. This was not observed when collateral blood flow was maintained during axial flow interruption. Radiologic visualization of these small arteries, however, was impaired even after shorter ischemic intervals, with resistance to flow progressively increasing after 6 hr of ischemia. The latter was ascribed to arteriolar thrombus which was documented histologically [lo]. The severity and incidence of the no reflow phenomenon appeared to be related to the length of the ischemic interval and the severity of the ischemia. Thus, in skeletal muscle where long periods of ischemia are required be-
fore significant injury is produced by reperfusion, the no reflow phenomenon may be seen with increasing frequency. When the rate and distribution of blood flow in muscle after prolonged ischemia are examined, reduced blood flow is noted after 5 hr of ischemia when compared to 4 hr of ischemia, suggesting that prolongation of the ischemic interval leads to increased resistance in the microcirculation with reduction in the rate of reflow. No improvement was seen with vasodilators, suggesting that local vasoconstriction is not a primary mechanism [ll]. Other factors may play a significant role in preventing adequate reperfusion of skeletal muscle after prolonged periods of ischemia. Cellular swelling occurs during ischemia and may impede flow during reperfusion [12-141. Perfusion with hyperosmolar agents, such as Mannitol, has been shown to improve reperfusion in skin flaps and other organs. Deformability of white blood cells has been suggested and observed to cause plugging in skeletal muscles of cats during shock [15]. Increasing vascular resistance is seen with an increasing number of leukocytes, specifically at low flows [16]. Our experiments, however, would suggest a limited role for white blood cells as contributors to the no reflow phenomenon, as
JOURNAL
FIG. 3.
OF SURGICAL
RESEARCH:
VOL. 51, NO. 1, JULY 1991
Histologic section of anterior tibialis control muscle (H&E, 40x). Note intact cellular and vascular structures.
leukopenic reperfusion did not prevent no reflow. Other contributors to this phenomenon may include capillary collapse with inability to reach an opening pressure because of the factors mentioned above, endothelial cell damage and dysfunction, and other as yet unrecognized changes occurring during both ischemia and reperfusion. Since the no reflow phenomenon has been observed in other organs such as kidneys, skin flaps, and brain [12,16-B], it appears to be an important component impeding successful retrieval of ischemic tissue. When histologic specimens are analyzed in control and experimental muscles, a remarkable difference is seen. Experimental muscles showing no reflow demonstrate marked red cell sludging and thrombi in arterioles, venules, and capillaries (Fig. 2). No such changes are seen in the control specimens (Fig. 3). Muscles treated with urokinase, on the other hand, resembled muscles from the control group, with the sporadic appearance of arteriolar sludging and thrombi (Fig. 4). This leads us to postulate that temporarily increasing fibrinolytic activity during the initial phase of reperfusion may revert changes, contributing to the no reflow phenomenon. Our experiments suggest that one of the principal causes of no reflow is small vessel occlusion
due to fibrin deposition, which in turn traps blood elements and is seen as sludging under the microscope. When the process is advanced, frank thrombus is seen. Platelets and white cells may be trapped in this process releasing oxygen radicals and, thus, contribute to reperfusion injury [3]. Endothelial cell damage, with opening of cellular gaps and exposure of the subendothelial elements secondary to processes during ischemia, would lead to fibrin deposition and potentially play a major role in triggering the no reflow phenomenon. Some of these changes have been shown to occur experimentally following ischemia and reperfusion [ 191. The fact that recovery of function in the urokinase-treated animals was no better and perhaps somewhat worse than those animals that did reperfuse with controlled reperfusion suggests a limited role for urokinase in preventing reperfusion injury. Nevertheless, the assurance of adequate reflow would open the possibilities for other manipulations to be effective in preventing reperfusion injury in the overall process. Thus, two major components of ischemia and reperfusion are suggested. On the one hand, the no reflow phenomenon, which prolongs the ischemia and prevents adequate restoration of flow, and second, cellular injury resulting from processes occurring be-
QUINONES-BALDRICH
ET AL.:
MUSCLE
FUNCTION
AFTER
ISCHEMIA
F‘IG. 4. Histologic section of anterior tibialis muscle after 5 hr of ischemia and fibrinolytic reperfusion. Note minimal patf :nt vascular system. There is evidence of fiber injury, no different than that in other experimental groups.
cause of restoration of flow. The latter is probably secondary to a multitude of factors, some of which will need to be addressed for successful retrieval of the ischemic cell. Our goal is to develop a clinically applicable method of reperfusion in severely ischemic limbs. In this regard, the intraoperative administration of fibrinolytic agent has been reported. In our own experience, both clinical and experimental data would suggest that they can be used with safety and efficacy in these acutely ischemic limbs [20-221. More recently, we have tried intraoperative and continuous postoperative fibrinolytic therapy in patients with severely ischemic extremeties. In these patients, we have performed operative thrombectomy and placement of an arterial catheter, starting a urokinase infusion intraoperatively and continuing into the postoperative period. In six patients treated, the results have been encouraging [23]. Although the problem of reperfusion injury in ischemit skeletal muscle has yet to be resolved, accumulating evidence suggests that controlled reperfusion may be a clinically feasible and effective alternative. These experiments and those of others suggest an additive role for fibrinolytic agents during the initial phases of reper-
sludging with mo
fusion [24]. This may improve our ability to reestablish flow in areas where fibrin deposition has occurred, and thus limit the ischemic insult. Additional experiments are required to determine other manipulations during the early phases of reperfusion which may have additional beneficial effects in restoring viability to skeletal muscle after prolonged periods of ischemia. REFERENCES 1.
2.
3.
Lee, K. R., Cronenwett, J. L., Shlafer, M., Corpron, C., and Zelenock, G. B. Effect of superoxide dismutase plus catalase on calcium transport in ischemic and reperfused skeletal muscle. J. Surg. Res. 42: 24-32, 1987. Perry, M. O., and Fantini, G. Ischemia: Profile of an enemy. Reperfusion injury of skeletal muscle. J. Vast. Surg. 6: 231-234, 1987. Beyersdorf, F., Matheis, G., Kruger, S., Hanselmann, A., Freisleben, H. G., Zimmer, G., and Satter, P. Avoiding reperfusion injury after limb revascularization: Experimental observations and recommendations for clinical application. J. Vusc. Surg. 9: 757-
766,1989. 4.
5.
Belkin, M., LaMorte, W. L., Wright, J. G., and Hobson, R. W. The role of leukocytes in the pathophysiology of skeletal muscle ischemic injury. J. Vmc. Surg. 10: 14-19, 1989. Fish, J. S., McKee, N. H., Pynn, B. R., Kuzon, W. M., and Plyley,
12
6.
7.
8.
9.
10.
11.
12.
13. 14.
JOURNAL
OF SURGICAL
RESEARCH:
M. J. Isometric contractile function recovery following tourniquet ischemia. J. Surg. Res. 47: 365-370, 1989. Chervu, A., Moore, W. S., Homsher, E., and Quifiones-Baldrich, W. J. Differential recovery of skeletal muscle and peripheral nerve function after ischemia and reperfusion. J. Surg. Res. 47: 12-19,1989. Chervu, A., Moore, W. S., and Quifiones-Baldrich, W. J. Skeletal muscle function after ischemia and reperfusion. Presented at the Society of University Surgeons 49th Annual Meeting, San Antonio, Texas. February 12, 1988. Wright, J. G., Fox, D., Kerr, J. C., Valeri, C. R., and Hobson, R. W. Rate of reperfusion blood flow modulates reperfusion injury in skeletal muscle. J. Surg. Res. 44: 754-764, 1988. Green, R. M., DeWeese, J. A., and Rob, C. G. Arterial embolectomy before and after the Fogerty catheter. Surgery 77: 24-33, 1975. Dunnant, J. H., and Edwards, W. S. Small vessel occlusion in the extremity after various periods of arterial obstruction: An experimental study. Surgery 73: 240-2451973. Forrest, I., Lindsay, T., Romaschin, A., and Walker, P. The rate and distribution of muscle blood flow after prolonged ischemia. J. Vast. Surg. 10: 83-88,1989. Chait, L. A., May, J. W., O’Brien, B. M., and Hurley, J. V. The effects of the perfusion of various solutions on the no reflow phenomenon in experimental free flaps. Pk. Recon. Surg. 61: 421430, 1978. Leaf, A. Cell swelling: A factor in ischemic tissue injury. Circulution 8: 455, 1963. Strock, P. E., and Majno, G. Microvascular changes in experimental tourniquet ischemia. Surg. Gynecol. O&et. 129: 12131223,1969.
15.
16.
17. 18. 19.
20.
21.
22.
23.
24.
VOL.
51, NO. 1, JULY
1991
Bagge, U., Maundson, B., and Lauritzen, C. White blood cell deformability and plugging of skeletal muscle capillaries in hemorrhagic shock. Actu Physiol. Stand. 180: 159-163,198O. Braide, M., Amundson, B., Chien, S., and Bagge, U. Quantitative studies on the influence of leukocytes on the vascular resistance in a skeletal muscle preparation. Microvasc. Res. 27: 331-352, 1984. Ames, A., Wright, R. L., Kowada, M., et al. Cerebral ischemiaThe no reflow phenomenon. Am. J. Pathol. 52: 437-453,1968. Summers, W. K., and Jamison, R. L. The no reflow phenomenon in renal ischemia. Lab. Invest. 25: 635, 1971. Fonkalsrud, E. W., Sanchez, M., Zerubavel, R., Lassaletta, L., Smeesters, C., and Mahoney, A. Arterial endothelial changes after ischemia and perfusion. Surg. Gynecol. Obstet. 142: 715721,1976. Quiiiones-Baldrich, W. J., Zierler, E., and Hiatt, J. R. Intraoperative fibrinolytic therapy: An adjunct to catheter thromboembolectomy. J. Vast. Surg. 2: 319-326, 1985. Quiiiones-Baldrich, W. J., Baker, J. D., Busuttil, R. W., Machleder, H. I., and Moore, W. S. Intraoperative infusion of lytic drugs for thrombotic complications of revaacularization. J. Vast. Surg. 10: 408, 1989. Quiiiones-Baldrich, W. J., Ziomek, S., Henderson, T., and Moore, W. S. Intraoperative fibrinolytic therapy: Experimental evaluation. J. Vast. Surg. 4: 229-236, 1986. Quifiones-Baldrich, W. J. The role of fibrinolysis during reperfusion of ischemic skeletal muscle. Microcirculation, Endothelium, and Lymphutics 5: 299-314,1989. Belkin, M., Valeri, C. R., and Hobson, R. W. Intraarterial urokinase increases skeletal muscle viability after acute ischemia. J. Vax. Surg. 9: 161-168, 1989.
