Effect of Propranolol on the Factors Promoting Bacterial Translocation in Cirrhotic Rats With Ascites

Transcription

1 Effect of Propranolol on the Factors Promoting Bacterial Translocation in Cirrhotic Rats With Ascites MARÍ A PÉREZ-PARAMO, 1 JAVIER MUÑOZ, 2 AGUSTÍ N ALBILLOS, 5 ISABEL FREILE, 1 FRANCISCA PORTERO, 3 MARTÍ N SANTOS, 4 AND JOSÉ ORTIZ-BERROCAL 4 Abbreviations: SBP, spontaneous bacterial peritonitis; CCl 4, carbon tetrachloride; BT, bacterial translocation; MLN, mesenteric lymph nodes; IBO, intestinal bacterial overgrowth; 99m Tc-DTPA, 99m Tc-diethylenetriamine pentaacetic acid; CFU, colony forming units. From the Divisions of 1 Nuclear Medicine, 2 Gastroenterology, 3 Microbiology, and 4 Experimental Surgery, Clínica Puerta de Hierro, Madrid, Spain; and the 5 Division of Gastroenterology, Hospital Ramón y Cajal, Department of Medicine, University of Alcalá, Madrid, Spain. Received February 25, 1999; accepted October 8, Supported in part by grants from Fondo de Investigaciones Sanitarias (99/0356), Universidad de Alcalá (E001/98 y E002/99), and by the Programa Nacional de Salud from the Comisión Interministerial de Ciencia y Tecnología. Address reprint requests to: Agustín Albillos, M.D., Dept. de Medicina, Facultad de Medicina-Campus Universitario, Universidad de Alcalá, Ctra. Madrid-Barcelona, km Alcalá de Henares, Madrid, Spain. aalbillosm@meditex.es; fax: (34) Copyright 2000 by the American Association for the Study of Liver Diseases /00/ $3.00/0 Bacterial translocation appears to be an important mechanism in the pathogenesis of spontaneous infections in cirrhosis. Cirrhotic patients are commonly treated with -adrenoceptor blockers, but the impact of this treatment in the factors promoting bacterial translocation has not been investigated. This study was aimed at investigating in cirrhotic rats with ascites the effect of propranolol on intestinal bacterial load, transit, and permeability of the bowel and on the rate of bacterial translocation. Bacterial translocation to mesenteric lymph nodes and intestinal bacterial overgrowth, permeability (urinary excretion of 99m Tc-diethylenetriaminepentaacetic acid [ 99m Tc-DTPA]), and transit (geometric center ratio of 51 Cr) were assessed in 29 rats with carbon tetrachloride (CCl 4 ) cirrhosis and 20 controls. These variables were then measured in 12 placebo- and in 13 propranolol-treated ascitic cirrhotic rats. Bacterial translocation was present in 48% of the cirrhotic rats and in none of the controls. Cirrhotic rats with intestinal bacterial overgrowth had a significantly higher rate of translocation and slower intestinal transit than those without it. Among the 15 rats with overgrowth and a 99m Tc-DTPA excretion greater than 10%, 15 had translocation and 2 had bacterial peritonitis. Only 1 of the 14 rats with either intestinal overgrowth or a 99m Tc-DTPA excretion greater than 10% presented translocation. Compared with the placebo group, propranolol-treated animals had significantly lower portal pressure, faster intestinal transit, and lower rates of bacterial overgrowth and translocation. In ascitic cirrhotic rats, bacterial translocation results from intestinal overgrowth and severe damage to gut permeability. In this setting, intestinal overgrowth is associated with intestinal hypomotility. Propranolol accelerates the intestinal transit, decreasing the rates of bacterial overgrowth and translocation. (HEPATOLOGY 2000;31:43-48.) Cirrhotic patients are prone to bacterial infections, with spontaneous bacterial peritonitis (SBP) being the most characteristic. Clinical and experimental evidence indicates that translocation of bacteria from the intestinal lumen to the bloodstream is directly involved in the pathogenesis of these spontaneous infections. SBP is caused predominantly by enteric organisms, 1 and selective intestinal decontamination lowers the rate of first or recurrent SBP in cirrhotic patients. 2 Studies in carbon tetrachloride (CCl 4 )-induced cirrhotic rats have shown the frequent, simultaneous presence of bacterial translocation (BT) and SBP caused by enteric organisms, 3-5 a higher rate of BT in ascitic than in nonascitic animals, 6 frequent genotype identity between ileal flora and bacteria colonizing the mesenteric lymph nodes (MLN), 7 and a reduction, by selective intestinal decontamination, in the incidence of BT after hemorrhagic shock. 8 The factors that favor BT in cirrhosis are incompletely understood. Proposed mechanisms include disruption of the equilibrium of normal intestinal bacterial flora, increased permeability of the intestinal mucosal barrier, and deficiencies in host immune defenses. In fact, intestinal bacterial overgrowth (IBO) by gram-negative aerobes has been observed in patients with alcohol-induced cirrhosis and in ascitic cirrhotic rats with BT, 9,10 and its possible relationship with gut motor dysfunction has recently been suggested. 11 Portal hypertension in itself also causes structural changes in the bowel wall that may disrupt the gut barrier and favor BT. 6,8,12 Strategies to reduce the enteric bacterial load of the bowel and the rate of BT in patients and in experimental models of cirrhosis have mainly involved selective intestinal decontamination. 2,8 On the other hand, many cirrhotic patients are currently treated with nonselective -adrenergic blockers to prevent variceal bleeding. The effects of -blockers on the transit and the bacterial load of the bowel in cirrhosis have not been studied to date. Sympathetic overactivity is a characteristic feature of cirrhosis with ascites, and sympathetic stimulation delays intestinal transit, an effect that can 43 be reversed by -adrenoceptor blockade. 13,14 In addition, -blockers by lowering portal pressure may ameliorate the disruption of the gut barrier in cirrhosis, reducing the rate of BT. This study was aimed at investigating the effect of longterm therapy with propranolol on the intestinal transit and the permeability of the gut barrier in cirrhotic rats with ascites and whether these changes influence BT. To accomplish this aim we previously determined the relative impor-

2 44 PÉREZ-PARAMO ET AL. HEPATOLOGY January 2000 tance of changes in intestinal transit, enteric bacterial load, intestinal permeability, and portal pressure in promoting BT in this animal model. MATERIALS AND METHODS One hundred forty-eight male Sprague-Dawley rats with an initial weight of 120 to 150 g were included in this study. Animals were housed in a controlled environment with a 12-hour light/dark cycle and were fed standard rat diet with water ad libitum. Studies were performed in accordance with the National Institutes of Health guidelines. Animal Model Cirrhosis was induced by intragastric administration of CCl 4 (Carbon Tetrachloride Carlo Erba; Farmitalia, Milan, Italy) according to previously described methods. 3,15 To increase the yield of cirrhosis, rats received phenobarbital (Sigma Chemical Co., St. Louis, MO) in the drinking water (35 mg/dl). Animals were gavaged weekly under light isoflurane anesthesia (Forane; Abbott Laboratories, Madrid, Spain) and weighed twice a week to assess each animal s response to CCl 4. An initial dose of 20 µl of CCl 4 was given 2 weeks after starting phenobarbital. The weight loss 48 hours after each dose of CCl 4 was used to estimate the subsequent doses. Animals were studied only if they developed ascites. The onset of ascites was heralded by a rapid weight gain associated with bulging flanks and occurred, on average, 10 weeks (range, 6-19 weeks) after the initial dose of CCl 4. Once ascites was present, CCl 4 was continued for 2 weeks at a fixed dose of 40 µl. The studies were performed 7 days after the last dose of CCl 4. On average, the rate of mortality in this series was 57%. Experimental Protocols The present study comprises 2 protocols. The goal of protocol 1 was to determine the relationship among intestinal transit, IBO, intestinal permeability, portal pressure, and BT. Twenty-nine cirrhotic rats with ascites and 20 healthy age-matched rats receiving phenobarbital were included in this protocol. The goal of protocol 2 was to investigate the effect of propranolol on the previous variables, specifically on BT. Twenty-five cirrhotic rats with ascites were divided into 2 groups to receive propranolol (n 13) or placebo (n 12). DL-propranolol (Sigma) was dissolved in the drinking water and administered at an approximate dose of 10 mg/kg/d for 2 weeks starting the second week after ascites development. Study Design Before use, animals were fasted for 8 hours, but water was allowed ad libitum. All experiments were performed under strict sterile conditions. Anesthesia was induced with isoflurane. The first day of the experiment, the rats were given 1 ml of water containing approximately 5 µci of 99m Tc-diethylenetriamine pentaacetic acid ( 99m Tc-DTPA) by orogastric gavage and were then placed into individual metabolic cages to collect urine for the following 24-hour period. The second day of the experiment, after another period of 8-hour fasting, animals received 1 ml of water containing 2 µci of 51 Cr through a gastric tube and, thereafter, they were allowed to recover and move freely. Thirty minutes later, animals were anesthetized with isoflurane, and thereafter a small parasagittal incision ( cm) was made in the skin over the left upper abdominal quadrant to expose the spleen and measure the intrasplenic pressure. The abdomen was opened through a midline incision and a sample of ascitic fluid taken from the peritoneal cavity. Then, the small intestine was ligated at both ends, the cranial and caudal MLN removed, and samples of blood withdrawn from the inferior vena cava and from a mesenteric vein. Finally, the intestine and portions of the spleen and of the liver were removed, and samples of the ileum content were taken. Methods Intestinal Permeability. Epithelial permeability was assessed by measuring the 24-hour urinary excretion of 99m Tc-DTPA. The level of radioactivity in the urine samples was determined on a gamma counter. Data are expressed as fractional excretion of the radioactive marker because the total administered amount of the probe was unknown. Increased permeability to macromolecules, such as 99m Tc- DTPA, is generally assumed to represent a breakdown of the mucosal physical barrier, a critical factor preventing BT. 16 Portal Pressure. Intrasplenic pressure was measured by inserting into the splenic pulp a polyethylene tubing connected to a quartz transducer, and this was registered using a multichannel recorder (Lectromed Holding Ltd., St. Peters, Jersey, UK). Intestinal Transit. Intestinal transit was assessed by measuring the distribution of 51 Cr throughout the intestine, as previously described. 17,18 Once both ends of the small intestine were ligated, special care was taken during intestine handling to avoid movement of intestinal contents. The small intestine was mobilized from the ileum to the duodenum, freed from the mesentery, removed from the abdomen, placed longitudinally on a moist gauze, and divided into 5-cm segments. Individual segments were numbered consecutively from orad to aborad and their ends ligated. The stomach and the colon were also removed and their radioactivity measured in a gamma-scintillation counter along that of the individual segments of small intestine. Intestinal transit was expressed as the geometric center of the distribution of the 51 Cr within the intestine and was calculated as the sum of the products of the fraction of the total administered dose of radioactivity per segment and the segment number. To correct for differences in the length of the small intestine among rats, results of the geometric center were divided by the total number of intestinal segments in each rat and expressed as the geometric center ratio. The geometric center is actually the center of gravity for the distribution of the radionuclide and represents the most accurate measure for quantifying intestinal transit. 17 Bacteriologic Study. Samples of ascitic fluid and of blood were immediately inoculated into tubes containing brain-heart broth. The MLN, portions of the spleen and of the liver, and ileal content were weighed, placed in measured amounts of sterile saline, and homogenized, and aliquots were plated on blood agar. After incubation for 48 hours at 37 C, the number of viable colonies on each plate was counted and specific microorganisms identified by means of a manual biochemical test or an automated system (Microscan; Baxter, Irvine, CA) when necessary. Blood culture bottles were incubated at 37 C for 7 days, and the presence of bacteria was monitored by subculturing on blood agar plates when turbidity was evident. BT from the intestinal lumen and SBP were defined on the basis of a positive MLN and ascitic fluid culture, respectively. Positivity of blood or solid organ cultures was considered indicative of passage of bacteria into the portal or systemic circulation. Total intestinal aerobic count was defined as the sum of all the aerobic bacteria present in the stool, and expressed as log 10 colony-forming units (CFU) per gram of stool. IBO of a specific organism was defined as a stool bacterial count higher than the mean plus 2 SDs of the count of the same organism in the stool of normal rats. Statistics. Results are expressed as mean SD. Statistical analysis was performed using the unpaired Student s t test or the Mann- Whitney test. Comparison of proportions between groups was done with Fisher s Exact test. Linear regression analysis was performed for selected variables. Statistical significance was set at a P.05. RESULTS Protocol 1: Role of Intestinal Transit, Bacterial Overgrowth, and Intestinal Permeability in Bacterial Translocation Twenty-nine cirrhotic rats with ascites and 20 healthy controls were studied. Forty-two rats died during the induction of cirrhosis or were found dead after developing ascites.

3 HEPATOLOGY Vol. 31, No. 1, 2000 PÉREZ-PARAMO ET AL. 45 Although the amount of ascitic fluid at laparotomy varied among animals, the volume was sufficient to allow its easy withdrawal and culture in all of them. Portal pressure ranged between 11 and 24.4 mm Hg in cirrhotic rats, whereas it was lower than 10 mm Hg in all normal rats (Table 1). The mean weight of the spleen was g in cirrhotic rats and g in control animals (P.01). BT was present in 14 cirrhotic rats (48%) and in none of the controls (P.01) (Table 1). The organisms isolated were all aerobes of enteric origin, with Escherichia coli being the most common (Table 2). SBP was present in 10 rats (34%), 8 with BT and 2 without it (57% vs. 13%, P.05) (Fig. 1). In those cases in which SBP coincided with BT, the same bacterial species was isolated in the MLN and in ascitic fluid. In the 2 cases of SBP without BT, the same bacteria grew in the portal blood and/or in the liver sample. Portal pressure was similar in cirrhotic rats with and without BT. Among the 29 cirrhotic rats, IBO was present in 13 animals with BT and in 7 without it (93% vs. 47%, P.05) (Tables 1 and 2). BT was observed in 13 of the 20 rats with IBO, and in only 1 of the 9 without it (65% vs. 11%, P.01) (Fig. 1). Although more than one microorganism could overgrow in the same animal, the species isolated in the MLN of the 13 rats with BT and IBO were always one of the overgrowing bacteria (Table 2). The total aerobic bacterial count in stool was greater in cirrhotic rats than in controls (P.01), mainly because of an increased intestinal count of E. coli in the former ( vs logcfu/g, P.05). The total aerobic bacterial count in stool was also greater (P.05) in rats with BT than in those without it. Portal pressure was similar in cirrhotic rats with and without BT. Intestinal transit was delayed in cirrhotic rats compared with controls (Table 1). This was the result of a lower geometric center ratio in those animals with IBO than in those without it ( vs , P.01) (Fig. 2), because the bowel transit time in the latter was similar to that TABLE 1. Characteristics of Normal Rats and Ascitic Cirrhotic Rats With and Without Bacterial Translocation Control Rats Cirrhotic Rats With Bacterial Translocation Cirrhotic Rats Without Bacterial Translocation Number of animals Portal pressure (mm Hg) * Bacterial translocation (%) 0 48* Spontaneous bacterial peritonitis (%) Intestinal bacterial overgrowth (%) 0 69* Aerobic bacterial stool count (logcfu/g) * Intestinal transit (geometric center ratio) Intestinal permeability (% urinary excretion of 99m Tc-DTPA) * *P.01 vs. control rats. P.05 vs. cirrhotic rats with BT. P.05 vs. control rats. P.01 vs. cirrhotic rats with BT. Rat TABLE 2. Bacteria Species Isolated From Mesenteric Lymph Nodes, Ascitic Fluid, and Other Specimens, and Overgrowing in the Bowel Mesenteric Lymph Node Ascitic Fluid Portal Blood Peripheral Blood Liver Spleen Intestinal Bacterial Overgrowth* 1 E. coli E. coli E. coli E. coli E. coli E. faecalis 2 E. coli E. coli E. coli E. coli E. coli E. coli 3 E. coli E. coli E. coli E. coli E. faecalis 4 E. coli E. coli E. coli E. coli 5 E. coli E. coli E. coli E. coli E. coli 6 E. coli E. coli E. coli E. faecalis 7 8 E. coli E. coli E. coli 9 E. coli E. coli 10 E. coli E. coli 11 E. coli E. coli 12 E. coli E. coli 13 E. coli E. coli 14 E. coli E. coli 15 E. coli E. coli E. coli E. coli 16 E. coli E. coli E. coli E. coli Abbreviations: E. coli, Escherichia coli;, Proteus mirabilis; Ps. aeruginosa, Pseudomonas aeruginosa. *Intestinal overgrowth of E. coli without BT or SBP was present in 5 other cirrhotic rats. It was associated with overgrowth in one of these animals and with E. faecalis overgrowth in another. of the controls ( ). The geometric center was and (P.01) in cirrhotic animals with and without IBO, respectively. The fractional urinary excretion of 99m Tc-DTPA was greater in cirrhotic animals than in controls (Table 1) and similar in cirrhotic animals with and without IBO ( vs %, not significant) (Fig. 2). Among cirrhotic rats with IBO, the derangement in intestinal permeability was of greater magnitude in those with BT than in those without it ( vs %, P.01). BT to MLN required the simultaneous presence of IBO and severe disturbance in the intestinal barrier (Fig. 1). Among the 15 cirrhotic rats with IBO and a urinary excretion of the radiolabel greater than 10%, 13 developed BT and 2 developed SBP without BT. BT was present in only 1 of the 14 cirrhotic rats in which only 1 of these 2 factors was present or both were absent. There was no correlation between the portal pressure and the urinary excretion of the radiolabel. Protocol 2: Effect of Propranolol on Bacterial Translocation, Intestinal Transit, Bacterial Overgrowth, and Intestinal Permeability Thirty-two rats died during the induction of cirrhosis, the total survival being similar in propranolol- and placebotreated animals. Portal pressure was lower (P.05) in propranolol- than in placebo-treated cirrhotic rats (Table 3).

4 46 PÉREZ-PARAMO ET AL. HEPATOLOGY January 2000 TABLE 3. Effect of Propranolol on Bacterial Translocation, Intestinal Bacterial Overgrowth, Intestinal Transit, and Intestinal Permeability of Cirrhotic Rats With Ascites Placebo Propranolol Number of animals Portal pressure (mm Hg) * Bacterial translocation (%) 58 15* Spontaneous bacterial peritonitis (%) 33 8 Intestinal bacterial overgrowth (%) 67 15* Aerobic bacterial stool count (logcfu/g) Intestinal transit (geometric center ratio) Intestinal permeability (% urinary excretion of 99m Tc-DTPA) *P.05 vs. placebo. P.01 vs. placebo. FIG. 1. Intestinal permeability of cirrhotic rats with ascites as measured by the fractional urinary excretion of an oral dose of 99m Tc-DTPA. Filled circles and triangles represent animals with intestinal bacterial overgrowth (IBO ), and empty circles represent animals without intestinal bacterial overgrowth (IBO ). The triangles represent animals with spontaneous bacterial peritonitis. Bacterial translocation was present in most cases of spontaneous bacterial peritonitis. Bacterial translocation to mesenteric lymph nodes required the simultaneous presence of intestinal bacterial overgrowth and severe damage to intestinal permeability, as expressed by a urinary excretion of 99m Tc-DTPA 10%. Thirteen of the 15 animals with intestinal bacterial overgrowth and a urinary excretion of 99m Tc-DTPA 10% developed bacterial translocation, and 2 developed spontaneous bacterial peritonitis without bacterial translocation. Bacterial translocation was present in only 1 of those animals with intestinal bacterial overgrowth, severe damage to intestinal permeability, or none of them. The weight of the spleen was also lower in animals receiving propranolol ( vs g, P.05). BT was present in 7 of the 12 cirrhotic rats receiving placebo and in 2 of the 13 rats treated with propranolol (58% vs. 15%, P.05). The translocated bacteria were E. coli (3 cases in the placebo group, 2 cases in the propranolol group), E. coli and Proteus (3 cases in the placebo group), and E. faecalis (1 case in the placebo group). SBP developed in 4 of the 7 rats with BT in the placebo group and in 1 of the 2 in the propranolol group. On the whole, the rate of SBP was lower in propranolol- than in placebo-treated rats, although the difference did not achieve statistical significance (33% vs. 8%, P.16). IBO was present in 8 of the 12 animals receiving placebo and in 2 of the 13 receiving propranolol (67% vs. 15%, P.05); 9 of these 10 animals also showed BT. The total aerobic bacterial count in stool was lower in the propranolol group (P.01). Intestinal transit was faster in the propranolol group than in the placebo group, as shown by a greater geometric center ratio in the former (Table 3). The geometric center was in the propranolol group and in the placebo group (P.01). The urinary excretion of 99m Tc- DTPA was similar in both groups. DISCUSSION This study investigates the effect of propranolol therapy on the transit and permeability of the intestine of cirrhotic rats with ascites and the influence of these changes on the rate of BT. It also provides data regarding the influence of IBO, intestinal transit, and intestinal permeability on the pathogenesis of BT to MLN in this experimental model. We have shown that in cirrhotic rats with ascites IBO results from intestinal hypomotility and that BT occurs only in the presence of IBO and of a severe disruption of the gut barrier. -adrenoceptor blockade with propranolol accelerates the intestinal transit, decreasing the rate of IBO and of BT. In this animal model of cirrhosis, passage of bacteria from the intestinal lumen to MLN requires the concurrent presence of overgrowth of the microorganism in the small bowel and severe damage to the mucosal barrier. IBO is a first, necessary step for translocation, because it was present in all but 1 of the 14 rats with BT. However, the fact that bacteria did not translocate in all animals with IBO indicates that the process of translocation requires the presence of factors other than FIG. 2. Bacterial translocation, intestinal transit and intestinal permeability in ascitic cirrhotic rats with and without intestinal bacterial overgrowth (IBO, IBO, respectively). Cirrhotic rats with intestinal bacterial overgrowth showed a higher rate of bacterial translocation and slower intestinal transit than cirrhotic rats without intestinal bacterial overgrowth, whereas the degree of damage to intestinal permeability was similar in both groups.

5 HEPATOLOGY Vol. 31, No. 1, 2000 PÉREZ-PARAMO ET AL. 47 IBO. The urinary excretion of 99m Tc-DTPA was greater than 10% in all but 1 of the rats with BT and in 2 animals with SBP and no BT. Therefore, in spite of IBO, BT did not develop in cirrhotic rats unless a severe impairment in intestinal permeability was present. Studies of intestinal grafting in rats have also shown a relationship between BT to MLN and the degree of damage to intestinal permeability caused by rejection, with the rate of BT strikingly elevated in those animals with a urinary excretion of 51 Cr-ethylenediaminetetraacetic acid greater than 10%. 19 The relationship between BT and intestinal permeability is further supported by studies in burn patients, in whom the impairment in permeability during the postburn period was associated with infection by enteric organisms. 20 Several mechanisms may be involved in promoting the disruption of the gut barrier leading to BT in this animal model. Significant structural changes have been described in the ileal and cecal mucosa of CCl 4 -induced cirrhotic rats. 6,8,12 Portal hypertension may play a relevant role in the genesis of these abnormalities. This contention is supported by the lower rate of BT in cirrhotic rats without ascites and in rats with chronic prehepatic portal hypertension, which show lower portal pressure and milder histological changes in the bowel than cirrhotic rats with ascites and rats with acute prehepatic portal hypertension, respectively. 6,12,21 However, the relationship between the degree of portal pressure and the impairment in intestinal permeability does not seem to be linear. This was shown in our study by the absence of correlation among portal pressure, the urinary excretion of the probe and the rate of BT, as well as by the lack of improvement in permeability after portal pressure reduction by propranolol. Alternatively, we cannot exclude a contributory role of the overgrowing microorganisms in the damage to the intestinal barrier, although the similar values of 99m Tc- DTPA urinary excretion in rats with and without IBO makes this possibility unlikely. Coordinated motor function of the bowel is the most important mechanism in the prevention of IBO. Our results suggest a possible association between intestinal hypomotility and IBO in this animal model. This can be supported by the delayed intestinal transit in cirrhotic rats with IBO and by the lower incidence of IBO in cirrhotic animals treated with propranolol, a drug that accelerated the intestinal transit. Intestinal hypomotility had not been previously reported in rats with cirrhosis, but our finding agrees with the observation of prolonged small intestinal transit in rats with portal vein stenosis and in patients with cirrhosis. 11,18,22 Moreover, cirrhotic patients with a history of SBP show higher incidence of IBO and more severe small intestine dismotility than cirrhotic patients without a history of SBP. 11 The pathogenesis of these abnormalities in small bowel motility remains speculative and is probably of multifactorial origin. Sympathetic stimulation, as present in cirrhosis with ascites, delays intestinal transit, an effect that seems to be modulated by a -adrenoceptor mediated pathway. 13,14 Nitric oxide, a vasodilator whose production is increased in cirrhosis, relaxes the circular muscle of human and canine small bowel and can contribute to the intestinal dismotility. 23 Besides these factors, the structural damage to the intestinal wall described in cirrhosis may also be relevant. Taking into account the possible association between intestinal hypomotility and BT, modulation of intestinal transit offers a new approach for the prophylaxis of gutderived bacterial infections in cirrhosis. The effect of nonselective -adrenoceptor blockade with propranolol on the enteric bacterial load and the rate of BT in ascitic cirrhotic rats constitutes an important finding of this study. Besides lowering portal pressure, propranolol hastened the intestinal transit, decreasing the enteric bacterial load, the rate of IBO, and in consequence, the rate of BT as well. It has been shown that under conditions of hyperactivity of the sympathetic nervous system -adrenoceptor blockade accelerates the intestinal transit by increasing the propulsive force generated by intestinal contractions. 13,14 These findings further support the relationship among small bowel motility, IBO, and BT in cirrhosis. In keeping with them, Bartolí etal. 24 have recently observed that the administration of a prokinetic agent, such as cisapride, to ascitic cirrhotic rats also reduces the bacterial content of the small bowel and the incidence of BT. In conclusion, severe damage to the gut mucosal barrier and IBO determine BT in cirrhotic rats with ascites. In this setting, IBO is associated with intestinal hypomotility. -adrenergic blockade with propranolol accelerates the intestinal transit and, in consequence, reduces the bacterial load of the bowel and the rate of BT. These findings, if extensible to humans, widen the beneficial effects of propranolol to the prophylaxis of spontaneous bacterial infections in cirrhosis. Acknowledgment: The authors are indebted to Martha Messman for her editorial assistance. REFERENCES 1. Albillos A, Cuervas-Mons V, Millán I, Cantón T, Montes J, Barrios C, Garrido A, et al. Ascitic fluid polymorphonuclear cell count and serum to ascites albumin gradient in the diagnosis of bacterial peritonitis. Gastroenterology 1990;98: Ginés P, Rimola A, Planas R, Vargas V, Marco F, Almeda A, Forne M, et al. Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis: results of a double-blind, placebo controlled trial. HEPATOLOGY 1990;12: Runyon B, Sugano S, Kanel G, Mellencamp M. A rodent model of cirrhosis, ascites, and bacterial peritonitis. Gastroenterology 1991;100: Runyon BA, Squier S, Borzio M. Translocation of gut bacteria in rats with cirrhosis to mesenteric lymph nodes partially explains the pathogenesis of spontaneous bacterial peritonitis. J Hepatol 1994;21: Llovet JM, Bartoli R, Planas R, Cabré E, Jiménez M, Urban A, Ojanguren I, et al. Bacterial translocation in cirrhotic rats. Its role in the development of spontaneous bacterial peritonitis. Gut 1994;35: Garcia-Tsao G, Lee FY, Barden GE, Cartun R, West B. Bacterial translocation to mesenteric lymph nodes is increased in cirrhotic rats with ascites. Gastroenterology 1995;108: Llovet JM, Bartolí R, March F, Planas R, Vinado B, Cabré E, Arnal J, et al. Translocated intestinal bacteria cause spontaneous bacterial peritonitis in cirrhotic rats: molecular epidemiologic evidence. J Hepatol 1998;28: Llovet JM, Bartoli R, Planas R, Vinado B, Pérez J, Cabré E, Arnal J, et al. Selective intestinal decontamination with norfloxacin reduces bacterial translocation in ascitic cirrhotic rats exposed to hemorrhagic shock. HEPATOLOGY 1996;23: Morencos FC, De las Heras G, Martín-Ramos L, López-Arias MJ, Ledesma F, Pons-Romero F. Small bowel bacterial overgrowth in patients with alcoholic cirrhosis. Dig Dis Sci 1996;41: Guarner C, Runyon B, Young S, Heck M, Sheikh MY. Intestinal bacterial overgrowth and bacterial translocation in cirrhotic rats with ascites. J Hepatol 1997;26: Chan CS, Chen GH, Lien HC, Yeh HZ. Small intestine dismotility and bacterial overgrowth in cirrhotic patients with spontaneous bacterial peritonitis. HEPATOLOGY 1998;28: Garcia-Tsao G, Albillos A, Barden GE, West AB. Bacterial translocation in acute and chronic portal hypertension. HEPATOLOGY 1993;17: McIntyre AS, Thompson DG, Burnham WR, Walker ER. Modulation of

6 48 PÉREZ-PARAMO ET AL. HEPATOLOGY January 2000 human upper intestinal nutrient transit by a beta-adrenoceptor mediated pathway. Gut 1992;33: Ahluwalia NK, Thompson DG, Barlow J, Heggie L. -adrenergic modulation of human upper intestinal propulsive forces. Gut 1994;35: Proctor E, Chatamra K. High yield micronodular cirrhosis in the rat. Gastroenterology 1982;83: Bjarnason I, Macpherson A, Hollander D. Intestinal permeability: an overview. Gastroenterology 1995;108: Miller MS, Galligan JJ, Burks TF. Accurate measurement of intestinal transit in the rat. J Pharmacol Methods 1981;6: Reilly JA, Quigley EMM, Forst CF, Rikkers LF. Small intestinal transit in the portal hypertensive rat. Gastroenterology 1991;100: Grant D, Hurlbut D, Zhong R, Wang P, Chen H, Garcia B, Behme R, et al. Intestinal permeability and bacterial translocation following small bowel transplantation in the rat. Transplantation 1991;52: Ziegler TR, Smith RJ, O Dwyer ST, Demling RH, Wilmore DW. Increased intestinal permeability associated with infection in burn patients. Arch Surg 1988;123: Geraghty JG, Angerson WJ, Carter DC. Portal venous pressure and portasystemic shunting in experimental portal hypertension. Am J Physiol 1989;257:G52-G Galati JS, Holdeman KP, Bottjen PL, Quigley EM. Gastric emptying and orocecal transit in portal hypertension and end-stage chronic liver disease. Liver Transplant Surg 1997;3: Stark ME, Bauer AJ, Sarr MG, Szurszewski JH. Nitric oxide mediates inhibitory nerve input in human and canine jejunum. Gastroenterology 1993;104: Bartolí R, Pardo A, Planas R, Rodríguez MP, Ausina V, Gassull MA. Effect of cisapride upon jejunal bacterial flora and bacterial translocation in ascitic cirrhotic rats. Gastroenterology 1998;114:A1209.