Octreotide does not prevent bacterial translocation in an infant piglet model of intestinal ischemia-reperfusion

Octreotide does not prevent bacterial translocation in an infant piglet model of intestinal ischemia-reperfusion

Octreotide Does Not Prevent Bacterial Translocation Piglet Model of Intestinal Ischemia-Reperfusion By Grant E. Taylor, Andre Hebra, Karin L. McGow...

457KB Sizes 0 Downloads 39 Views

Octreotide

Does Not Prevent Bacterial Translocation Piglet Model of Intestinal Ischemia-Reperfusion

By Grant E. Taylor,

Andre Hebra, Karin L. McGowan, Ming Yu, Stephen Myers, Melissa Chris Boardman, Arthur J. Ross III, and James A. O’Neill, Jr Philadelphia. Denns ylvania

0 The process of bacterial translocation (BT) after ischemia/ reperfusion (I/R) injury is reported to be mediated by local mucosal factors, the effects of pancreatic enzymes, epithelial disruption, and by dysfunctional intestinal motility. Octreotide (OCT), a somatostatin analog, has been postulated to protect against BT by influencing one or more of these factors. Twenty-two formula-fed piglets (weight, 3.5 -L 0.5 kg; age, 20 & 5 days) were divided into four groups: control (no drug given; no I/R; n = 61, I/R (no drug given; n = 5). I/R plus low-dose OCT (LD OCT, 0.08 pg/ kg; n = 6). and I/R plus high-dose OCT (HD OCT. 8 pg/kg; n = 5). All experimental subjects had nonocclusive mesenteric ischemia induced by reversible pericardial tamponade with mesenteric flow decreased to 25 f 5% of baseline for 5 hours followed by 15 f 5 hours of reperfusion. Mesenteric lymph nodes (MLN), liver, spleen, blood, and peritoneum were harvested for blind microbial analysis. None of the animals in the control group experienced translocation to the tissues tested. All of the animals in the I/R group experienced BT to the MLN. The subjects in the LD OCT and HD OCT groups experienced BT to the MLN 66% and 80% of the time, respectively. Despite the reported clinical evidence that OCT can protect the intestinal mucosa from injury and increase the clearance of bacteria from the gastrointestinal tract, in this study in which variables other than I/R known to promote bacterial translocation were eliminated, OCT failed to modify or prevent the occurrence of translocation to the MLN after I/R injury. Copyright o 1995 by W.B. Saunders Company INDEX WORDS: Bacterial translocation, sion, intestinal motility, octreotide.

ischemia-reperfu-

E PREVIOUSLY demonstrated that the migrating motor complex (MMC) becomes dysfunctional during periods of intestinal ischemia and reperfusion, and that the occurrence of ischemia followed by reperfusion (I/R) is associated with bacterial translocation (BT) to mesenteric lymph nodes.le2 These results taken together suggest that the dysfunctional motility permits bacterial overgrowth and subsequent translocation. If this is indeed the case, then pharmacological maintenance of the MMC during ischemia and reperfusion should provide protection against bacterial translocation. Octreotide (OCT), a synthetic analog of somatostatin, has been shown to increase gut motility, particularly phase 3 (MMC) activity, in severa models.3-6 We hypothesized that administering octreotide during ischemia might modify or prevent the process of BT from the gastrointestinal (GI) tract in our infant piglet model.

W

JournalofPediatric

Surgery,

Vol30,

in an Infant

No 7 (July),

1995:

pp 967-970

MATERIALS

L. McKernan,

AND METHODS

Twenty-two weaned Yorkshire piglets (age, 20 f 5 days; weight, 3.5 f 0.5 kg) were carefully selected to ensure adequate intestinal function. Under genera1 anesthesia, a Left external jugular central line was placed and tunneled subcutaneously to a position behind the ear. A pericardial catheter was placed via a subxiphoid approach and tunneled subcutaneously to the animal’s back. Using a retroperitoneal approach, a perivascular ultrasonic flow probe (Transonic Systems, Ithaca, NY; No. 3R) was placed around the superior mesenteric artery (SMA) and tunneled subcutaneously to the back. Subjects, after recovery, were assigned to four different groups. Control animals (n = 6) underwent a sham operation with no I/R and no OCT. The second group of animals (I/R, n = 5) were subjected to 5 hours of ischemia followed by 15 f 5 hours of reperfusion. Ischemia was induced by injecting dextran 40 into the pericardial sac to establish pericardial tamponade. SMA blood flow was measured by a 3R perivascular flow probe (Transonic Systems). SMA flow was monitored continuously and strictly kept within the experimental parameters. SMA flow was maintaitled at 25 2 5% of baseline during the entire tamponade period by injecting or withdrawing dextran 40 as needed. Reperfusion was accomplished by removal of the dextran 40 from the pericardial sac. The technique for induction, maintenance, and reversal of pericardial tamponade has been described previously.’ Two additional groups of animals underwent the same I/R procedure and, m addition, received low-dose (LD) or high-dose (HD) octreotide. LD OCT animals (n = 6) received LD OCT (0.08 kg/kg of OCT per dose), whereas HD OCT animals (n = 5) received HD OCT (8 kg/kg per dose). Doses of OCT were given at the start of the tamponade and every hour thereafter until the tamponade was complete; the animals were then given one final dose. Control animals were killed by barbiturate overdose between postoperative days 5 and IO; the other animal groups were killed immediately after the reperfusion phase. Under sterile conditions, all animals underwent necropsy: mesenteric lymph nodes (MLN), liver, spleen, terminal ileum, a peritoneal swab, and blood were harvested for quantitative microbial analysis as previously described? Cultures were positive if growth of enteric organisms was detected in cultures. Additionally, samples of ileum and jejunum were obtained for light microscopy with standard hematoxylin and eosin preparation. The culture results of each of the tissues studied from the four groups were compared by the extended Fisher’s exact test From the Division of General Surgery The Children> Hospital of Philadelphia, Philadelphia, PA. Presented at the 1994 Annual Meeting of the Section on Surgery of the American Academy of Pediatrics, Dallas, Texas, October 21-23, 1994. Address reprint requests to Graizt E. Taylor, MD, Department of General Surgery, Hahnemann University Hospital, Broad and Vine Sts, Philadelphia, PA 19102. Copyright o 1995 by W.B. Saunders Company 0022-3468195/3007-0012$03.00/O 967

966

TAYLOR

to compare the proportions of positive cultures. The groups were all compared with control and with each other. Pvalues of less than .05 were considered significant. The experimental protocol was approved by the Institutional Animal Care and Use Committee of the Children’s Hospital of Philadelphia.

RESULTS

Each subject was carefully examined before and after instrumentation to ensure the absence of clinically apparent infection, and this determination was confirmed at the time of necropsy. Intestinal cultures from all animals showed normal flora, consisting typically of Klebsiella pneumoniae, Escherichia coli, Proteus spp, Enterococcus spp, Staphylococcus aureus, and Bacteroides. In all but two of the animals, the quantitative stool cultures had more than lo8 colonyforming units per millijiter of homogenate, which is considered within normal limits. The animals in the control group did not experience BT to any of the tissues tested. All of the animals in the I/R group experienced BT to the MLN (P < .05). The animals receiving LD OCT showed BT to the MLN in four of six trials, significantly more than controls (P < .OS). The HD OCT group had BT to the MLN in four of five trials (P < .05 versus controls). Bacteremia did not occur at the time of blood sampling in either OCT-treated group. BT in the I/R group compared with the OCT groups was not statistically different, nor was the rate of BT different in comparing the LD OCT and HD OCT groups. The results of blinded, quantitative microbial analysis of the samples from the MLN, liver, spleen, blood, and peritoneum from the four groups are presented in Fig 1. No histological damage to the mucosa was noted on light microscopy of the intestinal sections. There was mild edema of some of the reperfusion sections; these n

Q .z .a:: 100 03 p! 80 2 a

is 8 d 5 $2 a”

MLN

60 40 20

0 CONTROL

I/R

LD OCT

HD OCT

Experimental Groups Fig 1. the MLN, experimental

Culture results. Bar graph comparing liver, spleen, blood, and peritoneum group.

the incidence in subjects

of BT to from each

would be grade 0 in the histological previously reported by our group.7

ET AL

grading scale

DISCUSSION

The GI tract not only serves the role of nutrient absorption, but also provides endocrine, immunological, metabolic, and barrier functions. Research implicates the gut as the source of the bacteria or bacterial products that lead to sepsis and eventual multiple organ failure.8-14 Shock is known to be associated with three major factors, each thought to contribute to BT: alterations in the composition of intestinal flora, impairment of host immune responses, and disruption of normal intestinal mucosal barrier function.l* Our model of cardiogenic shock uses pericardial tamponade to create low systemic flow with resultant intestinal ischemia; gut reperfusion occurs when the tamponade is reversed. Our prior work using this tamponade model has shown marked alterations of intestinal moti1ity.l Using the same model without manipulation of the bowel, we also showed that bowel ischemia and reperfusion leads to increased BT to the MLN. Here we also found that tamponade was accompanied by BT to MLN in 75% of subjects, whereas the addition of reperfusion to the protocol led to 100% occurrence of BT to the MLN. These findings suggested that a decrease in intestinal motility, perhaps combined with other factors, could lead to bacterial overgrowth and translocation. This led us to postulate that the administration of a drug that has protective mucosal effects as well as prokinetic activity in the GI tract should be able to minimize BT in I/R injury. Somatostatin is a tetradecapeptide that was first isolated from bovine hypothalamus in 1976.15 Somatostatin has limited clinical use because of its very short half-life. Octreotide (Sandostatin; Sandoz, Inc, Basel, Switzerland), a synthetic cyclic octapeptide analog of somatostatin with a longer half-life shares most biological activities of somatostatin, but has significant differences that are of importance to this study.16 Cullen et al5 have suggested that this compound might inhibit secretion of the substance that inhibits phase 3 electrical activity of GI motility. Thus, OCT could serve to protect from BT in I/R injury by preserving the occurrence of normal MMC patterns in the GI tract of infant piglets during periods of low flow or ischemia. In the experimental model reported, we did not attempt to record MMCs during ischemia or reperfusion in order to avoid entering the peritoneal cavity and manipulating the intestine, either of which could lead to BT. In -addition, other factors that might contribute to BT were avoided: The abdominal cavity and its contents were free of any inflammatory

OCTREOTIDE

AND

BACTERIAL

TRANSLOCATION

969

response, subjects were studied without the influence of any drugs or anesthesia, hypovolemia was avoided, nutrition was maintained, the indigenous intestinal flora was not altered, and mesenteric ischemia was accomplished without manipulation of the SMA. Our experimental study shows that the addition of OCT to our protocol did not alter the presence of bacterial translocation to the MLN, liver, and spleen either qualitatively or quantitatively. These observations do not support the role of OCT as a protective agent against BT. Disruption of the intestinal mucosal barrier is a factor in BT and may be induced by the generalized ischemia and by pancreatic proteases in this experimental modeli In a model of severe ischemic injury, Morris et alis showed that OCT was able to attenuate some of the mucosal effects of ischemic intestinal injury, which they speculated was caused by reduction in pancreatic protease output by OCT.l* Our model, however, used a less severe reduction in flow, and none of the animals in our study showed histological damage to the mucosa with or without OCT treatment. Despite the finding of normal histology in our

model, bacteria must be crossing the barrier because BT occurred. This finding of BT with normal histology is consistent with other investigators having shown BT through intact mucosa.19Jo It is possible that the alterations in the mucosal barrier function might not be caused by pancreatic proteases at this level of ischemia because OCT had no effect in our model. In conclusion, because the results of the OCT groups are not statistically different from those of the I/R group, we conclude that this drug alone cannot prevent the occurrence of BT in the infant piglet subjected to mesenteric I/R. We believe that BT is the result of multiple pathophysiologic phenomena and further studies are still needed to determine the effects of OCT during low-flow states on motility and pancreatic proteases with respect to the pathogenesis of BT. ACKNOWLEDGMENT The authors are grateful to Eduardo Ruchelli. MD, from the Department of Pathology at the Children’s Hospital of Philadelphia for his assistance with the interpretation of histology.

REFERENCES 1. Hebra A, Brown MF, McGeehin K, et al: The effects of ischemia and reperfusion on intestinal motility. J Pediatr Surg 28:362-366,1993 2. Hebra A, Hong J, McGowan KL, et al: Bacterial translocation in mesenteric &hernia-reperfusion injury: Is dysfunctional motility the link? J Pediatr Surg 29:280-287, 1994 3. Peeters TL, Romanski KW, Janssens J, et al: Effect of the long-acting somatostatin analogue SMS 201-995 on small intestinal interdigestive motility in the dog. Stand J Gastroenterol 23:769774,198s 4. Richards WO, Greer R, ODorisio TM, et al: Octreotide acetate induces fasting small bowel motility in patients with dumping syndrome. J Surg Res 49:483-487,199O 5. Cullen JJ, Eagon JC, Dozois EJ, et al: Treatment of acute postoperative ileus with octreotide. Am J Surg 165:113-120.1993 6. Soudah HC, Hasler WL, Owyang C: Effect of octreotide on intestinal motility and bacterial overgrowth in scleroderma. N Engl J Med 325:1461-1467,199l 7. Brown MF, Ross AJ, Dasher J, et al: The role of leukocytes in mediating mucosal injury of intestinal ischemia/reperfusion. J Pediatr Surg 25:214-217,199O 8. Zhi-yong S, Yuan-lin D, Xiao-hong W: Bacterial translocation and multiple system organ failure in bowel ischemia and reperfusion. J Trauma 32:148-153, 1992 9. Redan JA, Rush BF, McCullough JN, et al: Organ distribution of radiolabeled enteric escherichia coli during and after hemorrhagic shock. Ann Surg 211:663-668,199O

10. Mainous MR, Tso P, Berg RD, et al: Studies of the route, magnitude, and time course of bacterial translocation in a model of systemic inflammation. Arch Surg 126:33-37,199l 11. Redan JA, Rush BF, Lysz TW, et al: Organ distribution of gut-derived bacteria caused by bowel manipulation or ischemia. Am J Surg 159:85-90,199O 12. Arden WA, Yacko MA, Jay M, et al: Scintigraphic evaluation of bacterial translocation during hemorrhagic shock. J Surg Res 54:102-106,1993 13. Sori AJ, Rush BF, Lysz TW, et al: The gut as a source of sepsis after hemorrhagic shock. Am J Surg 155:188-192, 1988 14. Fushikima R, Gianotti L, Alexander JW: The primary site of bacterial translocation. Arch Surg 12953-58, 1994 15. Brazeau P, Vale WL, Bargus R, et al: Hypothalamic polypeptide that inhibits the secretion of immunoreactive pituitary growth hormone. Science 179:77-79,1973 16. Shulkes A: Somatostatin: Physiology and clinical appiications. Baillieres Clin Endocrinol Metab 8:215-236,1994 17. Montgomery A, Borgstrom A, Haglund U: Pancreatic proteases and intestinal mucosal injury after ischemia and reperfusion in the pig. Gastroenterology 102:216-222, 1992 18. Morris JB, Guerrero NH, Furth EE, et al: Somatostatin attenuates ischemic intestinal injury. Am J Surg 165:676-680, 1993 19. Smith SD, Cardona MA, Wishnev SA, et al: Unique characteristics of the neonatal intestinal mucosal barrier. J Pediatr Surg 27:333-338,1992 20. Alexander WJ, Boyce ST, Babcock GF, et al: The process of microbial translocation. Ann Surg 212:496-512,199O

Discussion IQ’. Garcia (Cincinnati, OH): In earlier work you have shown that your model results in an intestinal ischemia-reperfusion that alters intestinal motility

and prolongs transit time. In this study you postulate that by using a prokinetic agent such as octreotide, you would inhibit bacterial translocation.

970

I think it is important to note some of the pharmacokinetic and pharmacodynamic effects of octreotide. Depending on your route of administration the plasma concentration will differ. Secondly, the half-life is on the order of about 1% hours. Both these pharmacokinetic and pharmacodynamic effects I think raise the following questions for you. Would you explain to us how you gave octreotide? And do you think that it would have made a difference if you administered the agent throughout the entire reperfusion period? In the manuscript you indicated that you administered it for about 1 hour in the reperfusion period. Bacterial translocation is probably species-dependent and bacterial dose-dependent. You performed some intraluminal cultures. Did you do so at the end of your experiment? And a related question is, when you looked at the bacteria in the lymph nodes, was this per-organ weight? And then a final question, it is well appreciated that malnutrition and ischemia-reperfusion will affect translocation. Were your animals pair-fed?

TAYLOR

ET AL

G.E. Taylor (response): Thank you, Dr Garcia, for your comments. Octreotide was given intermittently via central venous line at l-hour intervals during the ischemic phase only. We did not attempt to give the drug during the reperfusion phase, and we do not know if that would have changed the rate of translocation. However, our previous experience has demonstrated that significant translocation occurs during the ischemic phase, which was the focus of this particular study. All of the gastrointestinal (intraluminal) cultures were obtained at the end of the experiment, and results were expressed in colony forming units of bacteria per fixed amount of tissue (standard volumes were obtained in all animals). The nutritional parameters of all animals were maintained before the experiment, based on the feeding and nutritional guidelines of the animal care facility. The animals were fasted, but given free access to water the night before the experiment. Animals with poor nutrition or any other problems were excluded.