Glucagon potentiates intestinal reperfusion injury Elizabeth T. Clark, M D , and Bruce L. Gewertz, M D ,
Chicago, Ill.
Vasoactive agents, including glucagon, have been used in treatment ofmesenteric ischemia. Such drugs change both intestinal blood flow and metabolism, Since reperfusion injury reflects the metabolic state o f an organ as well as the duration and severity of ischemia, we investigated the effect of glucagon in a standard model of intestinal ischemia. Data were generated from denervated isoperfused rat small intestinal preparations (n = 39). Arterial and venous pressures, intestinal blood flow, and oxygen consumption were monitored. Animals were subjected to 15, 30, or 45 minutes ofischemia followed by 1 hour reperfusion. Experiments were performed without drug infusion or during intravenous glucagon administration (0.1, 0.2, or 0.4 Ixg/kg/min). After the rats were killed, histologic sections of intestine were graded 1 through 5 in a blinded fashion with 1 = normal xSlli and 5 = severe injury. Results (mean ± SD) were analyzed by analysis of variance (~p < 0.05). Glucagon at all concentrations increased intestinal blood flow and oxygen consumption before isehemia. For example, with 0.2 b~g/kg/min glucagon, intestinal blood flow increased from 80.78 ± 13.5 to 114.79 ± 21.02 ml/min - 100 gm ~ and oxygen consumption increased from 3,65 ± 0.73 to 5.73 -+ 1.37 ml/min • 100 grn. ~ Mucosal injury after ischemia reflected duration of ischemia and glucagon infusion rate. At all ischemic intervals, increased glucagon concentrations were associated with greater mucosal injury. In fact the histologic injury with 15 minutes o f ischemia + 0.2 ~g/kg/min glucagon (3.04 ± 0.49) exceeded that o f 30 minutes of ischemia (2.87 -+ 0.06). Glucagon-mediated increases in intestinal metabolism before ischemia worsened eventual reperfusion injury. In the diverse clinical syndromes of mesenteric ischemia, often characterized by repeated episodes of flow interruption, treatment with glucagon or similar drugs may be detrimental. (J Vase SuR~ 1990;11:270-9.)
Despite improvements in diagnosis and therapy, acute embolic or thrombotic occlusions ofmesenteric vessels remain highly lethal complications o f advanced arterial occlusive disease. ~ Clinical efforts are currently limited to early restoration o f blood flow even though diverse experimental models have established that the eventual injury suffered by the gut reflects deleterious effects o f reperfusion as well as the hypoxia experienced during flow interruption. 2,3 It is generally accepted that hypoxic injuries are dependent on the duration and degree o f the flow disturbance. In contrast, reperfusion injury may occur
From the Department of Surgery, The Universityof Chicago. Presented at the Forty-third Annual Meeting of The Societyfor Vascular Surgery, New York, N.Y., June 20-21, 1989. Supported by American Heart Association grant fief. 880786). Dr. Clark supported by National Institutes of Health-Research Training grant (# H107665). Reprint requests: Bruce L. Gewertz, MD, Professorof Surgery, Universityof Chicago, 5841 S. MarylandAve., Box 129, Chicago, IL 60637. 24/6/16786 270
after even relatively brief periods o f flow interruption. The susceptibility o f tissue to reperfusion injury is quite variable and is thought to reflect, at least in part, the generation o f oxygen-derived free radicals. ~-7 These experiments specifically addressed the role o f the metabolic state o f the intestine at the time o f ischemic insult. We asked, would hypermetabolism (accentuating the imbalance between oxygen supply and demand during flow interruption) decrease the resistance o f the intestine to hypoxia and reperfusion? Such a relationship is likely since the intestinal metabolic state (e.g., fasted vs fed) has been shown to strongly influence other regulatory mechanisms such as pressure:flow autoregulation. 8-~-° Studies were performed in a rat small intestinal preparation in which the metabolic, hemodynamic, and histologic responses to ischemia have been well characterized. Glucagon was used to mediate the hypermetabolic state because the drug reliably increases both intestinal blood flow and metabolism, 11-13and has been used in the treatment ofmesenteric ischemia. 1 ~
4
++,= +
0
A
+
30 mNn
O - ,....
\+++
®
+..__
2,
•,
+++ I
I
I
2
3
4
~
!
I
5
6
HISTOLOGIC GRADE
Fig. 5. Oxygen extraction (402) and histologic grade as related to glucagon concentration and duration of ischemia.
pear to be a plateau both for length of ischemia and glucagon concentration. Concentrations of glucagon greater than 0.2 ~g/kg/min did not significantly alter histologic grade, and ischemic intervals greater than 30 minutes did not result in increased mucosal damage. The relationship between reperfusion VO2 and histologic inju.ry is depicted in Fig. 5. More profound depression of VO2 in the reperfusion period was correlated with a higher histologic grade. DISCUSSION
The outcome of an episode of intestinal ischemia doubtless reflects both the severity of the flow disturbance as well as its duration. As illustrated in Fig, 6, ischemic injury is likely if flow is completely interrupted even for 15 minutes, whereas less severe disturbances (e.g., IBF = 25 ml/min • 100 gm) can be tolerated for 30 minutes before injury. This concept of an "ischemic threshold," combining measures of blood flow and time can be extended to the grade of histologic injury. More severe mucosal in-
jury (grades 4 and 5) requires a longer period of flow interruption and, indeed, may not occur at all if perfusion is maintained greater than some arbitrary limit (in our example, IBF > 25 ml/min • 100 gm). We hypothesized that increasing the basal metabolic rate of the intestine would "shift" the ischemic threshold to the left. That is, when exposed to a comparable duration and severity of ischemia, intestine with increased metabolism would suffer a greater degree of histologic injury than that seen with a normat metabolic rate. This hypothesis is consistent with the most common clinical manifestation of intestinal ischemia--postprandial pain. This syndrome of"intestinal angina" results f?om the inability of stenosed or obstructed vessels to augment blood flow and oxygen delivery during the period of increased metabolic activity after a meal. Such mismatches in oxygen supply and demand may also influence the relative contributions ofhypoxia and reperfusion phenomena to ischemic injury. To test our hypothesis we used glucagon, which both vasodilates the mesenteric vascular bed and in-
~ournal of VASCULAR SURGERY
276 Clark and Gewertz
HISTOLOGIC
GRADE
o 50" o
E 25u,, In
2 15
3
30
4
5 60
90
TIME ( m i n )
Fig. 6. Concept of an "ischemic threshold" that relates mucosal injury to both blood flow and duration of flow disturbance. Increases in ischemic mucosal injury (e.g., grades 4 and 5) require longer periods of flow interruption and may not occur if perfusion is maintained above an arbitrary limit.
creases metabolism. Glucagon has recently been used clinically to treat mesenteric ischemia based on experimental evidence that glucagon administered after reperfusion improves survival in a rat model of intestinal ischemia. 19 In experimental preparations, glucagon has been a&ninistered both systemically (intravenously) and intraarterially (directly into the SMA). 2°22 Investigators have studied large pharmacologic doses (up to 100 ~g/kg/min) and more physiologic concentrations (100 to 500 ng/kg/min).23.24 Glucagon uniformly causes splanchnic hyperemia, as a result of changes in microvasculature dimensions as well as well-characterized central inotropic and chronotropic properties. 2s'26Work by Tiblin et al.27 first suggested that glucagon vasodilation is based on interference with alpha adrenergic receptors. Subsequently, glucagon has been shown to increase intracellular calcium and cyclic adenosine monophosphate levels by blockage of calcium extrusion from smooth muscle cells and/or increased calcium sequestration by a cyclic adenosine monophosphate-dependent mechanism? 8 The increased oxygen uptake associated with glucagon presumably reflects redistribution of blood flow to metabolically active tissues. Holliger et al.21 used in vivo videomicroscopy to demonstrate a greater sensitivity to glucagon in the villous arcade than in the larger vessels. At certain doses glucagon increased villous arcade flow by 50%, whereas SMA flow remained essentially unchanged. Studies by
Pawlik et al.29 also showed a redistribution of blood flow within the wall of the gut toward the mucosalsubmucosal compartment and an increase in the density of perfused capillaries. In our experiments infusion of 0.1, 0.2, and 0.4 btg/kg/min of glucagon consistently increased IBF and VO2. Maximal effects were noted even with the lowest doses. When glucagon infusions predated and were continued through the period of ischemia, we observed decremental worsening ofhistologic injury and more profound depression of reperfusion VO> If it is assumed that a histologic grade of 4 or 5 is indicative of severe ischemic injury, it seems dear that the "ischemic threshold" is reached sooner when higher amounts of glucagon are infused. For example, 15 minutes of ischemia with 0.4 p.g/kg/min glucagon resulted in a greater depression of reperfusion VO2 and a higher histologic grade than either 30 minutes of ischemia alone or 30 minutes of ischemia + 0.1 ~g/kg/min. Unfommately, with the current experimental protocol, we were unable to characterize the glucagon effect more specifically. That is, did the increase in metabolic rate augment hypoxic injury, reperfusion injury, or both? It could be argued that the adverse effects of glucagon in these experiments reflect more than just the increased metabolism associated with the drug. Indeed, glucagon-induced vasoditation could impair local blood flow regulation within the gut wail and in some way "shunt" blood from the most vulnerable mucosal regions, a°,al This protocol did not allow us
Volume 11 Number 2 February 1990
to determine whether hypermetabolism, vasodilation, or some other yet u n k n o w n mechanism was responsible for mucosal injury. H o w e v e r , since the small intestine was totally isolated and rendered uniformly and severely ischemic, "shunting" to adjacent well-perfused regions was n o t possible during the ischemic interval. Furthermore, previous studies with vasodilators alone have n o t d o c u m e n t e d adverse effects during reperfusion. 19,2°,a2 As previously noted, glucagon infusions b e g u n after reinstitution o f b l o o d flow improved survival in a different rat model o f acute intestinal ischemia.19 W h e n reconciled with o u r results, it appears that the timing o f glucagon infusion is crucial. I f infused before ischemia and revascularization, hypermetabolic effects o f glucagon m a y further embarrass an adverse oxygen s u p p l y : d e m a n d ratio and cxacerbate injury. I n contrast, limiting its administration to the reperfusion period m a y predispose to a favorable o u t c o m e because o f the well-documented vasodilatory properties o f the drug. It is always difficult to directly transpose experimental observations to the bedside. Nonetheless, in view o f the diverse clinical syndromes o f mesenteric ischemia, often characterized by repeated episodes o f flow interruption, it w o u l d seem prudent to avoid vasodilators such as glucagon that possess equally potent metabolic effects. This is especially true in patients suffering nonocclusive ischemia in which multiple episodes are the rule. The authors acknowledge the assistance of Michael Beach, PhD, in the statistical analysis of the data and the expert manuscript preparation of Mrs. Eileen Wayte. REFERENCES
1. Boley SJ, Gliedman ML. Circulatory responses to mesenteric ischemia. In: Boley SLed. Vascular disorders of the intestine. New York: Appleton-Ccnmry Crofts, 1971. 2. Parks DA, Granger DN. Contributions of ischemia and reperfusion to mucosal lesion formation. Am I Physiol 1986; 250:G749-53. 3. Haglund V, Bulldey GB, Granger DN. On the pathophysiology of intestinal ischemic injury. Acta Chir Scand 1987; 153:321-4. 4. Granger DN, Rutili G, McCord JM. 8uperoxide radicals in feline intestinal ischemia. Gastroenterology 1981;81: 22-9. 5. Granger DN, Hollwarth ME, Parks DA. Ischemiareperfusion injury: role of oxygen-derived free radicals. Acta Physiol Scan Suppl i986;548:47. 6. Parks DA, Granger DN. Ischemia reperfusion injury: a radical view. Hepatology 1988;8:680-2. 7. McCord JM. Oxygen derived radicals and reperfusion injury. Oxygen radicals and human disease. Ann Int Med i987; 107: 52&45.
Glucagon for treatment of int,estinal reperfusion injury 277
8. Shepherd AP, Granger HI. Autoregulatory escape in the gut: a systems analysis. Gastroenterology 1973;65:77-91. 9. Shepherd AP. Intestinal capillary blood flow during metabolic hyperemia. Am J Physiol 1979;237:E548-54. I0. Shepherd AP. Metabolic control ofintestinal oxygenation and blood flow. Federation Proc 1982;41:2084-9. i 1. Bowen JC, Pawlik W, Fang W, Jacobson ED. Pharmacologic effects of gastrointestinal hormones on intestinal oxygen consumption and blood flow. Surgery 1975;78:515-9. 12. Kock NG, Tibblin S, Schenk WG. Mesenteric blood flow response to glucagon administration. Arch Surg 1970;100: 280-3. 13. Kvietys PR, Granger DN. Vasoactive agents and splanchnic oxygen uptake. Am J Physiol 1982;243:61-9. 14. Athanasoutis CA, Wittenberg J, Bernstein R, Williams LF. Vasodilatory drugs in the management ofnonocclusive bowel ischemia. Gastroenterology 1975;68:146-50. 15. Roark KM, Suzuki NT. Glucagon use in mesenteric ischemia. Drug Intell Clin Pharm 1987;21:660-1. 16. Anzueto L, Benoit JN, Granger DN. Arat model for studying the intestinal circulation. Am J Physiol 1984;246:656-61. 17. Kim EH, Gewertz BL. Chronic digitalis administration alters mesenteric vascular reactivity. J VAsc SURG 1987;5:382-9. 18. Chin Chu-Jeng, McArdle AH, Brown R, et al. Intestinal mucosal lesion in low-flow states. Arch Surg 1970;101:478-83. 19. Cronenwett JL, Ayad M, Kazmers A. Effect of intravenous glucagon on the survival of rats after acute occlusive ruesenteric ischemia. J Surg likes 1985;38:446-52. 20. Boorstein JM, Dacey LJ, Cronenwett JL. Pharmacologic treatment of occlusive mesenteric ischemia in rats. J Surg Res 1988;44:555-60. 21. Holliger CH, Radzyner M, Knoblauch M. Effects of glucagon, vasoactive intestina{ peptide and vasopressin on villous rnicrocirulation and superior mesenteric artery blood flow of the rat. Gastroenterology 1983;85:1036-43. 22. Tibblin S, Kock NG, Schenk WG. Response of mesenteric blood flow to glucagon. Arch Surg 1971;102:65-70. 23. Fasth S, Hulton L. The effect of glucagon on intestinal motility and blood flow. Acta Physiol Scand 1971;83:169-73. 24. Krarup N, Larson JA. The effect of glucagon on hepatosplanchnic hemodynamics, functional capacity and metabolism of the liver in cats. Acta Physiol Scand 1974;91:42-52. 25. Finke V, Seifert J. Differential effects of gastrointestinal hormones on the blood flow of the alimentary tract of the dog. Res Exp Med 1986;186:151-65. 26. Kvietys PR, Granger DN. Relation between intestinal blood flow and oxygen uptake. Am J Physiol I982;243:G202-8. 27. Tibblin S, Kock NG, Schenk WG. Splanchnic hemodynamic responses to glucagon. Arch Surg 1970;100:84-9. 28. Farah AE. Glucagon and the circulation. Pharmacol Rev 1983; 35:181-217. 29. Pawlik, Wiselaw W, Fondcaro JD, Jacobson ED. Metabolic hyperemia in canine gut. Am J Physiol 1980;239:612-7. 30. BulldeyGB, Womack WA, Downey JM, KvietysPR, Granger DN. Collateral blood flow in segmental intestinal ischemia: effects of vasoactive agents. Surgery 1986;i00:157-66. 31. Granger DN, KvietysPR, PerryMA. Role ofexchangevessels in the regulation of intestinal oxygenation. Am I Physiol 1982;242:G570-4. 32, Boley SJ, Sprayregan S, Sigalman SS, Veith FJ. Initial results from an aggressive roentgenological and surgical approach to acute mesenteric ischemia. Surgery 1977;82:848-55.
278
Clark and Gewertz
DISCUSSION
Dr. lames C. Stanley (Ann Arbor, Mich.). Mechanisms of intestinal ischemic injury are complex and include cellular hypoxia, oxygen-free radical generation, effects of digestive enzymes on gut tissues, and bacterial invasion of intestinal tissues. The authors' study is one of severe ischemia in which reperfusion appears important. However, mucosal damage does not necessarily require reperfiasion. In fact, the entire cellular surface of an ischemic villus may be destroyed before reperfusion. In isolated canine small bowel preparations, oxygen uptake becomes flow dependent at rates below 30 ml/min/100 gm of tissue. Although oxygen extraction at the onset o f flow dependentT is not worse in the gut compared to the whole body, the critical point in intestinal perfusion is reached at a time when nonintestinal oxTgen uptake is still independent of supply. Blood flows in the range of the critical point may be manipulated pharmacologically to lessen tissue injury, but it is unlikely that prolonged interruption of flow with reperfusion will be affected by such interventions. The authors' conclusion that glucagon worses reperfusion injury in their experimental model is clearly supported by the histologic data and should be accepted. However, extrapolation of this conclusion to flow and oxygen consumption data should be tempered, in that both are expressed to weights of intestinal tissue measured after reperfusion. At such times tissue edema alone might give the impression that flow or oxygen consumption was altered, when such might simply reflect only an increase in tissue weight. Your data might be somewhat soft in this regard and your comments on weight changes in this study would be appreciated. Second, in the absence of pressure or cardiac output data, the question arises as to just what was responsible for flow reductions. Was it a decrease in cardiac output or was it an increase in tissue resistance? I would favor, and am sure the authors would concur, that tissue resistance was the issue, but do hard data exist regarding this issue in your study? Last, the infusion of glucagon beginning 20 minutes before SMA occlusion makes the data in this experiment nonrelevant to many clinical settings. The authors' intent was not to do this--they were interested in creating a hypermetabolic gut that then could be studied regarding the effects of ischemia--but one must reemphasize this point to those of you in the audience who might assume that glucagon has no potential efficacy in the clinical management of intestinal ischemia. In this regard, a number of studies not cited by the authors do exist, with intact animals of varying species, that reveal important differences in the net effect of glucagon a&ninistration on gut ischemia. In one, a canine study by Shapiro and Cronenwett and colleagues (J Surg Res 1984;36:535-46), glucagon administered after a 75% reduction in S/vIA flow, not total occlusion, caused a fur-
Journal of VASCULAR SURGERY
ther decrease in ileal blood flow as determined by microsphere injection. Such data were consistent with the authors' study. However, in this same experiment when glucagon was given 30 minutes after reperfusion, the SMA flow nearly doubled, and oxygen consumption returned to its normal baseline. In a second study, Kazmers (J VASe SURG 1984;1:472-81) administered gtucagon 15 minutes after the initiation of an 85-minute period of SMA total occlusion before blood flow was restored to the intestine in Wistar rats. He observed a 48-hour survival of 85% in these subjects, which was markedly greater than the survival of 54% after saline administration alone or the 38% survival in untreated controls. Thus glucagon in the intact subject may have a different effect on intestinal ischemia/reperfusion injury. In such a setting its action may be both central with inotropic and chronotrophic cardiac effects as well as peripheral in the gut, causing direct vascular smooth muscle relaxation as well as by eliminating gut motility--something that very few other agents d o - - w i t h cessation of the muscularis mucosa tonicity. The latter muscle is often chronically contracted after severe ischemic insults and under such circumstances it may act as a precapillary sphincter itself, preventing blood flow to the villose tip where ischemic injury is most evident. Dr. Elizabeth Clark. Again, I stress that the goal of our study was to define the role of the metabolic rate and the basic mechanisms ofischemia/reperfusion injury. Thus the timing of glucagon infusion we used was to alter metabolism and not to be so much as a treatment ofmesenteric ischemia. In terms of the quantification of oxygen extraction data, given the weight of the intestine and the possibility of alterations between control and ischemic intestine, we found that the intestine weight varied between 7 and 10 gin, with a variance of about 30%. There was no trend toward increasing weight in the ischemic rats as compared to the control rats. Although we considered edema as a possible problem in the interpretation of these data, we did not find that there was prominent evidence of this in our histologic studies. Second, Dr. Stanley's points about the measurement of cardiac output are well taken, and we did not measure cardiac output in these particular studies. However, as I mentioned, we did carefully monitor mean arterial pressure, and it was regulated to between 65 and 75 mm Hg. Therefore, by definition, any changes in IBF reflect a change in resistance as the primary mechanism. Dr. lack L. Cronenwett (Hanover, N.H.). I believe that the results of this study may help explain our previous observation that glucagon improved survival when given after, but not during, ischemia in an intact rat model. Although our recent, as yet unpublished, data would suggest that this beneficial effect of glucagon is related to its inotropic activity, the authors have nicely shown that the met-
Volume 11 Number 2 February 1990
Glucagonfor treatment of intestinal reperfusion injury 279
abolic effects of glucagon are ones that we cannot ignore when considering its potential mechanism. I have three questions: (1) Have you had an opportunity to study glucagon in different temporal relationships to ischemia, particularly when given after ischemia during reperfusion? (2) Have you looked at mucosal injury after ischemia alone, that is, before reperfusion, since your data .would suggest that the metabolic effects of this injury occur very early during the ischemic period? (3) Finally, do you know if glucagon has comparable metabolic versus vasodilating effects in humans.~ When given to patients with chronic mesenteric ischemia during duplex studies of the SMA, glucagon increases blood flow, but I am not aware that it causes abdominal pain, which might be predicted if significant metabolic changes occurred. Dr. Elizabeth Clark. In answer to your first question, we have not examined different temporal relationships and that is an excellent suggestion.
In terms of looking at the mucosal injury after ischemia alone, the model that we used lends itself" well to do that as biopsies can be taken at various time intervals. Certainly that is something we will look into. In answer to your third question, I agree that it is difficult at times to translate from the laboratory to the clinical arena. The effects of glucagon in man may in fact be different from those we see in the rat intestine preparation. In man, neurologic input and the presence of recruitable collaterals may alter the effect of glucagon. In conclusion, we do not recommend any abrupt changes in the current therapy of mesenteric ischemia. Rather, glueagon's mediation of the hypermetabolic state was used to focus on basic mechanisms of severe ischemia/reperfilsion injury. The presence of preexisting stenoses and recruitable cot.laterals, as seen in our patients, will certainly affect clinical treatment.
B O U N D VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the Jot:KNAL oF VASCULAR Sue, fiERy for 1990 are available to subscribers only. They may be purchased from the publisher at a cost of $54.00 (S68.00 international) for Vol. 11 (January to June) and Vol, 12 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Circulation Fulfillment, The C.V. Mosby Company, 11830 West_line Industrial Drive, St. Louis, MO 6314:6-3318, USA. In the United States call toll free: (800)325-4177, ext. 351. In Missouri call collect: (314)872-8370, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.
Chicago, Ill.
Vasoactive agents, including glucagon, have been used in treatment ofmesenteric ischemia. Such drugs change both intestinal blood flow and metabolism, Since reperfusion injury reflects the metabolic state o f an organ as well as the duration and severity of ischemia, we investigated the effect of glucagon in a standard model of intestinal ischemia. Data were generated from denervated isoperfused rat small intestinal preparations (n = 39). Arterial and venous pressures, intestinal blood flow, and oxygen consumption were monitored. Animals were subjected to 15, 30, or 45 minutes ofischemia followed by 1 hour reperfusion. Experiments were performed without drug infusion or during intravenous glucagon administration (0.1, 0.2, or 0.4 Ixg/kg/min). After the rats were killed, histologic sections of intestine were graded 1 through 5 in a blinded fashion with 1 = normal xSlli and 5 = severe injury. Results (mean ± SD) were analyzed by analysis of variance (~p < 0.05). Glucagon at all concentrations increased intestinal blood flow and oxygen consumption before isehemia. For example, with 0.2 b~g/kg/min glucagon, intestinal blood flow increased from 80.78 ± 13.5 to 114.79 ± 21.02 ml/min - 100 gm ~ and oxygen consumption increased from 3,65 ± 0.73 to 5.73 -+ 1.37 ml/min • 100 grn. ~ Mucosal injury after ischemia reflected duration of ischemia and glucagon infusion rate. At all ischemic intervals, increased glucagon concentrations were associated with greater mucosal injury. In fact the histologic injury with 15 minutes o f ischemia + 0.2 ~g/kg/min glucagon (3.04 ± 0.49) exceeded that o f 30 minutes of ischemia (2.87 -+ 0.06). Glucagon-mediated increases in intestinal metabolism before ischemia worsened eventual reperfusion injury. In the diverse clinical syndromes of mesenteric ischemia, often characterized by repeated episodes of flow interruption, treatment with glucagon or similar drugs may be detrimental. (J Vase SuR~ 1990;11:270-9.)
Despite improvements in diagnosis and therapy, acute embolic or thrombotic occlusions ofmesenteric vessels remain highly lethal complications o f advanced arterial occlusive disease. ~ Clinical efforts are currently limited to early restoration o f blood flow even though diverse experimental models have established that the eventual injury suffered by the gut reflects deleterious effects o f reperfusion as well as the hypoxia experienced during flow interruption. 2,3 It is generally accepted that hypoxic injuries are dependent on the duration and degree o f the flow disturbance. In contrast, reperfusion injury may occur
From the Department of Surgery, The Universityof Chicago. Presented at the Forty-third Annual Meeting of The Societyfor Vascular Surgery, New York, N.Y., June 20-21, 1989. Supported by American Heart Association grant fief. 880786). Dr. Clark supported by National Institutes of Health-Research Training grant (# H107665). Reprint requests: Bruce L. Gewertz, MD, Professorof Surgery, Universityof Chicago, 5841 S. MarylandAve., Box 129, Chicago, IL 60637. 24/6/16786 270
after even relatively brief periods o f flow interruption. The susceptibility o f tissue to reperfusion injury is quite variable and is thought to reflect, at least in part, the generation o f oxygen-derived free radicals. ~-7 These experiments specifically addressed the role o f the metabolic state o f the intestine at the time o f ischemic insult. We asked, would hypermetabolism (accentuating the imbalance between oxygen supply and demand during flow interruption) decrease the resistance o f the intestine to hypoxia and reperfusion? Such a relationship is likely since the intestinal metabolic state (e.g., fasted vs fed) has been shown to strongly influence other regulatory mechanisms such as pressure:flow autoregulation. 8-~-° Studies were performed in a rat small intestinal preparation in which the metabolic, hemodynamic, and histologic responses to ischemia have been well characterized. Glucagon was used to mediate the hypermetabolic state because the drug reliably increases both intestinal blood flow and metabolism, 11-13and has been used in the treatment ofmesenteric ischemia. 1 ~
4
++,= +
0
A
+
30 mNn
O - ,....
\+++
®
+..__
2,
•,
+++ I
I
I
2
3
4
~
!
I
5
6
HISTOLOGIC GRADE
Fig. 5. Oxygen extraction (402) and histologic grade as related to glucagon concentration and duration of ischemia.
pear to be a plateau both for length of ischemia and glucagon concentration. Concentrations of glucagon greater than 0.2 ~g/kg/min did not significantly alter histologic grade, and ischemic intervals greater than 30 minutes did not result in increased mucosal damage. The relationship between reperfusion VO2 and histologic inju.ry is depicted in Fig. 5. More profound depression of VO2 in the reperfusion period was correlated with a higher histologic grade. DISCUSSION
The outcome of an episode of intestinal ischemia doubtless reflects both the severity of the flow disturbance as well as its duration. As illustrated in Fig, 6, ischemic injury is likely if flow is completely interrupted even for 15 minutes, whereas less severe disturbances (e.g., IBF = 25 ml/min • 100 gm) can be tolerated for 30 minutes before injury. This concept of an "ischemic threshold," combining measures of blood flow and time can be extended to the grade of histologic injury. More severe mucosal in-
jury (grades 4 and 5) requires a longer period of flow interruption and, indeed, may not occur at all if perfusion is maintained greater than some arbitrary limit (in our example, IBF > 25 ml/min • 100 gm). We hypothesized that increasing the basal metabolic rate of the intestine would "shift" the ischemic threshold to the left. That is, when exposed to a comparable duration and severity of ischemia, intestine with increased metabolism would suffer a greater degree of histologic injury than that seen with a normat metabolic rate. This hypothesis is consistent with the most common clinical manifestation of intestinal ischemia--postprandial pain. This syndrome of"intestinal angina" results f?om the inability of stenosed or obstructed vessels to augment blood flow and oxygen delivery during the period of increased metabolic activity after a meal. Such mismatches in oxygen supply and demand may also influence the relative contributions ofhypoxia and reperfusion phenomena to ischemic injury. To test our hypothesis we used glucagon, which both vasodilates the mesenteric vascular bed and in-
~ournal of VASCULAR SURGERY
276 Clark and Gewertz
HISTOLOGIC
GRADE
o 50" o
E 25u,, In
2 15
3
30
4
5 60
90
TIME ( m i n )
Fig. 6. Concept of an "ischemic threshold" that relates mucosal injury to both blood flow and duration of flow disturbance. Increases in ischemic mucosal injury (e.g., grades 4 and 5) require longer periods of flow interruption and may not occur if perfusion is maintained above an arbitrary limit.
creases metabolism. Glucagon has recently been used clinically to treat mesenteric ischemia based on experimental evidence that glucagon administered after reperfusion improves survival in a rat model of intestinal ischemia. 19 In experimental preparations, glucagon has been a&ninistered both systemically (intravenously) and intraarterially (directly into the SMA). 2°22 Investigators have studied large pharmacologic doses (up to 100 ~g/kg/min) and more physiologic concentrations (100 to 500 ng/kg/min).23.24 Glucagon uniformly causes splanchnic hyperemia, as a result of changes in microvasculature dimensions as well as well-characterized central inotropic and chronotropic properties. 2s'26Work by Tiblin et al.27 first suggested that glucagon vasodilation is based on interference with alpha adrenergic receptors. Subsequently, glucagon has been shown to increase intracellular calcium and cyclic adenosine monophosphate levels by blockage of calcium extrusion from smooth muscle cells and/or increased calcium sequestration by a cyclic adenosine monophosphate-dependent mechanism? 8 The increased oxygen uptake associated with glucagon presumably reflects redistribution of blood flow to metabolically active tissues. Holliger et al.21 used in vivo videomicroscopy to demonstrate a greater sensitivity to glucagon in the villous arcade than in the larger vessels. At certain doses glucagon increased villous arcade flow by 50%, whereas SMA flow remained essentially unchanged. Studies by
Pawlik et al.29 also showed a redistribution of blood flow within the wall of the gut toward the mucosalsubmucosal compartment and an increase in the density of perfused capillaries. In our experiments infusion of 0.1, 0.2, and 0.4 btg/kg/min of glucagon consistently increased IBF and VO2. Maximal effects were noted even with the lowest doses. When glucagon infusions predated and were continued through the period of ischemia, we observed decremental worsening ofhistologic injury and more profound depression of reperfusion VO> If it is assumed that a histologic grade of 4 or 5 is indicative of severe ischemic injury, it seems dear that the "ischemic threshold" is reached sooner when higher amounts of glucagon are infused. For example, 15 minutes of ischemia with 0.4 p.g/kg/min glucagon resulted in a greater depression of reperfusion VO2 and a higher histologic grade than either 30 minutes of ischemia alone or 30 minutes of ischemia + 0.1 ~g/kg/min. Unfommately, with the current experimental protocol, we were unable to characterize the glucagon effect more specifically. That is, did the increase in metabolic rate augment hypoxic injury, reperfusion injury, or both? It could be argued that the adverse effects of glucagon in these experiments reflect more than just the increased metabolism associated with the drug. Indeed, glucagon-induced vasoditation could impair local blood flow regulation within the gut wail and in some way "shunt" blood from the most vulnerable mucosal regions, a°,al This protocol did not allow us
Volume 11 Number 2 February 1990
to determine whether hypermetabolism, vasodilation, or some other yet u n k n o w n mechanism was responsible for mucosal injury. H o w e v e r , since the small intestine was totally isolated and rendered uniformly and severely ischemic, "shunting" to adjacent well-perfused regions was n o t possible during the ischemic interval. Furthermore, previous studies with vasodilators alone have n o t d o c u m e n t e d adverse effects during reperfusion. 19,2°,a2 As previously noted, glucagon infusions b e g u n after reinstitution o f b l o o d flow improved survival in a different rat model o f acute intestinal ischemia.19 W h e n reconciled with o u r results, it appears that the timing o f glucagon infusion is crucial. I f infused before ischemia and revascularization, hypermetabolic effects o f glucagon m a y further embarrass an adverse oxygen s u p p l y : d e m a n d ratio and cxacerbate injury. I n contrast, limiting its administration to the reperfusion period m a y predispose to a favorable o u t c o m e because o f the well-documented vasodilatory properties o f the drug. It is always difficult to directly transpose experimental observations to the bedside. Nonetheless, in view o f the diverse clinical syndromes o f mesenteric ischemia, often characterized by repeated episodes o f flow interruption, it w o u l d seem prudent to avoid vasodilators such as glucagon that possess equally potent metabolic effects. This is especially true in patients suffering nonocclusive ischemia in which multiple episodes are the rule. The authors acknowledge the assistance of Michael Beach, PhD, in the statistical analysis of the data and the expert manuscript preparation of Mrs. Eileen Wayte. REFERENCES
1. Boley SJ, Gliedman ML. Circulatory responses to mesenteric ischemia. In: Boley SLed. Vascular disorders of the intestine. New York: Appleton-Ccnmry Crofts, 1971. 2. Parks DA, Granger DN. Contributions of ischemia and reperfusion to mucosal lesion formation. Am I Physiol 1986; 250:G749-53. 3. Haglund V, Bulldey GB, Granger DN. On the pathophysiology of intestinal ischemic injury. Acta Chir Scand 1987; 153:321-4. 4. Granger DN, Rutili G, McCord JM. 8uperoxide radicals in feline intestinal ischemia. Gastroenterology 1981;81: 22-9. 5. Granger DN, Hollwarth ME, Parks DA. Ischemiareperfusion injury: role of oxygen-derived free radicals. Acta Physiol Scan Suppl i986;548:47. 6. Parks DA, Granger DN. Ischemia reperfusion injury: a radical view. Hepatology 1988;8:680-2. 7. McCord JM. Oxygen derived radicals and reperfusion injury. Oxygen radicals and human disease. Ann Int Med i987; 107: 52&45.
Glucagon for treatment of int,estinal reperfusion injury 277
8. Shepherd AP, Granger HI. Autoregulatory escape in the gut: a systems analysis. Gastroenterology 1973;65:77-91. 9. Shepherd AP. Intestinal capillary blood flow during metabolic hyperemia. Am J Physiol 1979;237:E548-54. I0. Shepherd AP. Metabolic control ofintestinal oxygenation and blood flow. Federation Proc 1982;41:2084-9. i 1. Bowen JC, Pawlik W, Fang W, Jacobson ED. Pharmacologic effects of gastrointestinal hormones on intestinal oxygen consumption and blood flow. Surgery 1975;78:515-9. 12. Kock NG, Tibblin S, Schenk WG. Mesenteric blood flow response to glucagon administration. Arch Surg 1970;100: 280-3. 13. Kvietys PR, Granger DN. Vasoactive agents and splanchnic oxygen uptake. Am J Physiol 1982;243:61-9. 14. Athanasoutis CA, Wittenberg J, Bernstein R, Williams LF. Vasodilatory drugs in the management ofnonocclusive bowel ischemia. Gastroenterology 1975;68:146-50. 15. Roark KM, Suzuki NT. Glucagon use in mesenteric ischemia. Drug Intell Clin Pharm 1987;21:660-1. 16. Anzueto L, Benoit JN, Granger DN. Arat model for studying the intestinal circulation. Am J Physiol 1984;246:656-61. 17. Kim EH, Gewertz BL. Chronic digitalis administration alters mesenteric vascular reactivity. J VAsc SURG 1987;5:382-9. 18. Chin Chu-Jeng, McArdle AH, Brown R, et al. Intestinal mucosal lesion in low-flow states. Arch Surg 1970;101:478-83. 19. Cronenwett JL, Ayad M, Kazmers A. Effect of intravenous glucagon on the survival of rats after acute occlusive ruesenteric ischemia. J Surg likes 1985;38:446-52. 20. Boorstein JM, Dacey LJ, Cronenwett JL. Pharmacologic treatment of occlusive mesenteric ischemia in rats. J Surg Res 1988;44:555-60. 21. Holliger CH, Radzyner M, Knoblauch M. Effects of glucagon, vasoactive intestina{ peptide and vasopressin on villous rnicrocirulation and superior mesenteric artery blood flow of the rat. Gastroenterology 1983;85:1036-43. 22. Tibblin S, Kock NG, Schenk WG. Response of mesenteric blood flow to glucagon. Arch Surg 1971;102:65-70. 23. Fasth S, Hulton L. The effect of glucagon on intestinal motility and blood flow. Acta Physiol Scand 1971;83:169-73. 24. Krarup N, Larson JA. The effect of glucagon on hepatosplanchnic hemodynamics, functional capacity and metabolism of the liver in cats. Acta Physiol Scand 1974;91:42-52. 25. Finke V, Seifert J. Differential effects of gastrointestinal hormones on the blood flow of the alimentary tract of the dog. Res Exp Med 1986;186:151-65. 26. Kvietys PR, Granger DN. Relation between intestinal blood flow and oxygen uptake. Am J Physiol I982;243:G202-8. 27. Tibblin S, Kock NG, Schenk WG. Splanchnic hemodynamic responses to glucagon. Arch Surg 1970;100:84-9. 28. Farah AE. Glucagon and the circulation. Pharmacol Rev 1983; 35:181-217. 29. Pawlik, Wiselaw W, Fondcaro JD, Jacobson ED. Metabolic hyperemia in canine gut. Am J Physiol 1980;239:612-7. 30. BulldeyGB, Womack WA, Downey JM, KvietysPR, Granger DN. Collateral blood flow in segmental intestinal ischemia: effects of vasoactive agents. Surgery 1986;i00:157-66. 31. Granger DN, KvietysPR, PerryMA. Role ofexchangevessels in the regulation of intestinal oxygenation. Am I Physiol 1982;242:G570-4. 32, Boley SJ, Sprayregan S, Sigalman SS, Veith FJ. Initial results from an aggressive roentgenological and surgical approach to acute mesenteric ischemia. Surgery 1977;82:848-55.
278
Clark and Gewertz
DISCUSSION
Dr. lames C. Stanley (Ann Arbor, Mich.). Mechanisms of intestinal ischemic injury are complex and include cellular hypoxia, oxygen-free radical generation, effects of digestive enzymes on gut tissues, and bacterial invasion of intestinal tissues. The authors' study is one of severe ischemia in which reperfusion appears important. However, mucosal damage does not necessarily require reperfiasion. In fact, the entire cellular surface of an ischemic villus may be destroyed before reperfusion. In isolated canine small bowel preparations, oxygen uptake becomes flow dependent at rates below 30 ml/min/100 gm of tissue. Although oxygen extraction at the onset o f flow dependentT is not worse in the gut compared to the whole body, the critical point in intestinal perfusion is reached at a time when nonintestinal oxTgen uptake is still independent of supply. Blood flows in the range of the critical point may be manipulated pharmacologically to lessen tissue injury, but it is unlikely that prolonged interruption of flow with reperfusion will be affected by such interventions. The authors' conclusion that glucagon worses reperfusion injury in their experimental model is clearly supported by the histologic data and should be accepted. However, extrapolation of this conclusion to flow and oxygen consumption data should be tempered, in that both are expressed to weights of intestinal tissue measured after reperfusion. At such times tissue edema alone might give the impression that flow or oxygen consumption was altered, when such might simply reflect only an increase in tissue weight. Your data might be somewhat soft in this regard and your comments on weight changes in this study would be appreciated. Second, in the absence of pressure or cardiac output data, the question arises as to just what was responsible for flow reductions. Was it a decrease in cardiac output or was it an increase in tissue resistance? I would favor, and am sure the authors would concur, that tissue resistance was the issue, but do hard data exist regarding this issue in your study? Last, the infusion of glucagon beginning 20 minutes before SMA occlusion makes the data in this experiment nonrelevant to many clinical settings. The authors' intent was not to do this--they were interested in creating a hypermetabolic gut that then could be studied regarding the effects of ischemia--but one must reemphasize this point to those of you in the audience who might assume that glucagon has no potential efficacy in the clinical management of intestinal ischemia. In this regard, a number of studies not cited by the authors do exist, with intact animals of varying species, that reveal important differences in the net effect of glucagon a&ninistration on gut ischemia. In one, a canine study by Shapiro and Cronenwett and colleagues (J Surg Res 1984;36:535-46), glucagon administered after a 75% reduction in S/vIA flow, not total occlusion, caused a fur-
Journal of VASCULAR SURGERY
ther decrease in ileal blood flow as determined by microsphere injection. Such data were consistent with the authors' study. However, in this same experiment when glucagon was given 30 minutes after reperfusion, the SMA flow nearly doubled, and oxygen consumption returned to its normal baseline. In a second study, Kazmers (J VASe SURG 1984;1:472-81) administered gtucagon 15 minutes after the initiation of an 85-minute period of SMA total occlusion before blood flow was restored to the intestine in Wistar rats. He observed a 48-hour survival of 85% in these subjects, which was markedly greater than the survival of 54% after saline administration alone or the 38% survival in untreated controls. Thus glucagon in the intact subject may have a different effect on intestinal ischemia/reperfusion injury. In such a setting its action may be both central with inotropic and chronotrophic cardiac effects as well as peripheral in the gut, causing direct vascular smooth muscle relaxation as well as by eliminating gut motility--something that very few other agents d o - - w i t h cessation of the muscularis mucosa tonicity. The latter muscle is often chronically contracted after severe ischemic insults and under such circumstances it may act as a precapillary sphincter itself, preventing blood flow to the villose tip where ischemic injury is most evident. Dr. Elizabeth Clark. Again, I stress that the goal of our study was to define the role of the metabolic rate and the basic mechanisms ofischemia/reperfusion injury. Thus the timing of glucagon infusion we used was to alter metabolism and not to be so much as a treatment ofmesenteric ischemia. In terms of the quantification of oxygen extraction data, given the weight of the intestine and the possibility of alterations between control and ischemic intestine, we found that the intestine weight varied between 7 and 10 gin, with a variance of about 30%. There was no trend toward increasing weight in the ischemic rats as compared to the control rats. Although we considered edema as a possible problem in the interpretation of these data, we did not find that there was prominent evidence of this in our histologic studies. Second, Dr. Stanley's points about the measurement of cardiac output are well taken, and we did not measure cardiac output in these particular studies. However, as I mentioned, we did carefully monitor mean arterial pressure, and it was regulated to between 65 and 75 mm Hg. Therefore, by definition, any changes in IBF reflect a change in resistance as the primary mechanism. Dr. lack L. Cronenwett (Hanover, N.H.). I believe that the results of this study may help explain our previous observation that glucagon improved survival when given after, but not during, ischemia in an intact rat model. Although our recent, as yet unpublished, data would suggest that this beneficial effect of glucagon is related to its inotropic activity, the authors have nicely shown that the met-
Volume 11 Number 2 February 1990
Glucagonfor treatment of intestinal reperfusion injury 279
abolic effects of glucagon are ones that we cannot ignore when considering its potential mechanism. I have three questions: (1) Have you had an opportunity to study glucagon in different temporal relationships to ischemia, particularly when given after ischemia during reperfusion? (2) Have you looked at mucosal injury after ischemia alone, that is, before reperfusion, since your data .would suggest that the metabolic effects of this injury occur very early during the ischemic period? (3) Finally, do you know if glucagon has comparable metabolic versus vasodilating effects in humans.~ When given to patients with chronic mesenteric ischemia during duplex studies of the SMA, glucagon increases blood flow, but I am not aware that it causes abdominal pain, which might be predicted if significant metabolic changes occurred. Dr. Elizabeth Clark. In answer to your first question, we have not examined different temporal relationships and that is an excellent suggestion.
In terms of looking at the mucosal injury after ischemia alone, the model that we used lends itself" well to do that as biopsies can be taken at various time intervals. Certainly that is something we will look into. In answer to your third question, I agree that it is difficult at times to translate from the laboratory to the clinical arena. The effects of glucagon in man may in fact be different from those we see in the rat intestine preparation. In man, neurologic input and the presence of recruitable collaterals may alter the effect of glucagon. In conclusion, we do not recommend any abrupt changes in the current therapy of mesenteric ischemia. Rather, glueagon's mediation of the hypermetabolic state was used to focus on basic mechanisms of severe ischemia/reperfilsion injury. The presence of preexisting stenoses and recruitable cot.laterals, as seen in our patients, will certainly affect clinical treatment.
B O U N D VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the Jot:KNAL oF VASCULAR Sue, fiERy for 1990 are available to subscribers only. They may be purchased from the publisher at a cost of $54.00 (S68.00 international) for Vol. 11 (January to June) and Vol, 12 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Circulation Fulfillment, The C.V. Mosby Company, 11830 West_line Industrial Drive, St. Louis, MO 6314:6-3318, USA. In the United States call toll free: (800)325-4177, ext. 351. In Missouri call collect: (314)872-8370, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.
