The Effect of Biliary Decompression on Bacterial Translocation in Jaundiced Rats

Patients with obstructive jaundice are prone to septic complications after biliary tract operations. Restoring bile flow to the intestine may help to decrease the complication rate. The present study is aimed at evaluating the effect of biliary decompression on bacterial translocation in jaundiced rats. Sixty-six male Sprague-Dawley rats were randomly allocated to six groups subjected to common bile duct ligation (CBDL) and transection (groups 2–6) or sham operation (group 1). In groups and 2 the incidence of enteric bacterial translocation was determined 2 weeks after sham operation or CBDL. In groups 3–6, biliary decompression was achieved by performing a choledochoduodenostomy after 2 weeks of biliary decompression. Bacterial translocation was then studied 1,2,3 and 5 weeks following biliary decompression. The rate of bacterial translocation to mesenteric lymph nodes in obstructive jaundice was significantly higher as compared with controls, and decreased with time to nil three weeks following biliary decompression. The incidence of bacterial translocation was closely correlated (r = 0.844; p = 0.034) with serum alkaline phosphatase activity and seemed to fit with the morphological changes noted in the small intestine. The decrease in bacterial translocation, however, lags behind the recovery of liver function as measured by routine liver function tests and antipyrine clearance. Obstructive jaundice thus promotes bacterial translocation in the rat. Biliary decompression gradually decreases the rate of bacterial translocation.


INTRODUCTION
Obstructive jaundice is frequently associated with septic complications following biliary tract surgery1'2. The mechanisms for this clinical phenomenon are not entirely clear. Frequently, enteric, gram negative bacteria have been isolated from the infectious focus (mainly wound, bile or abscess)3'4. Recently, Deitch, et al. showed that bacterial translocation (i.e. enteric bacteria crossing the intestinal wall and invading extra-intestinal sites) occurred in obstructive jaundice in mice5. Bailey 6 and Koscr et al. have reported that jaundiced subjects are prone to endotoxaemia, due to the increased absorption and the decreased hpatic clearance capacity of the reticuloendothelial system (RES) in obstructive jaundice-. The recovery of intestinal bile flow has been considered to be crucial in the prevention of endotoxaemia and relief of the biliary obstruction has also been demonstrated to be important in the recovery of RES function11. Internal biliary drainage could both conduct bile flow back to the intestine and render relief of jaundice in biliary obstruction.
The purpose of the present study is to evaluate the effect of internal biliary drainage, by choledochoduodenostomy, on the incidence of bacterial translocation in rats with obstructive jaundice.

Animals and Experimental Design
Sixty-six male Sprague-Dawley (Mllegaard Ltd., Skensved, Denmark) rats, weighing 270-350g, were randomly allocated to six groups, in which groups 2-6 were subjected to obstructive jaundice by ligating the common bile duct, whereas group 1 underwent sham operation. The animals were kept in controlled temperature (22 C), humidity and 12-hour light/dark cycles and allowed rat pellets (ALTROMIN NR.1324, Altromin Spezialfutterwerke GHMB, Germany) and tap water ad libitum before and after the operations. The bacterial translocation study was performed 2 weeks following common bile duct ligation (CBDL, group 2) or sham operation (group 1) as well as 1, 2, 3 and 5 weeks after biliary decompression (BD) (groups 3-6).

Surgical Procedures
An upper-midline abdominal incision was made through which the common bile duct was identified, double ligated with 5-0 silk (ETHICON, GmbH, Germany) and transected between the ligatureslz. Sham operation was performed by mobilizing the common bile duct from the surrounding tissues without ligation and division. Internal biliary drainage was then achieved by performing a choledochoduodenostomy. A longitudinal incision (0.8 cm) was made in the duodenum and the dilated common bile duct remnant, respectively. The side to side anastomosis was established by a continuous single suture layer with 6-0 silk (ETHICON, GmbH, Germany). All operations were carried out under sterile conditions using light ether anaesthesia.
Liver Function Tests Blood samples for the determination of serum concentrations of bilirubin, alanine aminotransferase (ALAT), aspartate aminotransferase (ASAT) and alkaline phosphatase (ALP), were taken at the time of bacterial translocation studies. Bilirubin was determined in accordance with the method of Kuffer et al. 13 and the other enzymes were determined using the methods of the Committee on Enzymes of the 14 15 Scandinavian Society of Clinical Chemistry and Clinical Physiology Plasma Antipyrine Clearance Plasma antipyrine clearance, a determinant of microsomal function of hepatocytes, was analysed by a single sample method assay16. Four mg of antipyrine (Sigma Chemical Co., St Louis, Mo., USA) were dissolved in sterile saline and injected intravenously (penile vein) 5 hours before performing common bile duct ligation, biliary decompression or bacterial translocation studies 2, 3 and 5 weeks after biliary decompression. Five hours after the injection, blood samples for determining antipyrine clearance were taken into heparinized tubes and centrifuged at 2000 rpm for 15 min, after which plasma was removed and frozen at -20 C. Antipyrine was analysed by high performance liquid chromatography (HPLC) as described previously16. The antipyrine plasma clearance was calculated by using the onesample method as Bacteriological Study A modification of the methodology described by Deitch et al. and Jones II et al. 17 was used in the present study. The abdomen was shaved, twice soaked with chlorhexidine in spirit (KabiVitrum AB, Stockholm, Sweden) and aseptically opened with sterile instruments. Aliquots of 0.5 ml blood from the portal vein and inferior vena cava were transferred into tubes containing 2 ml of Brain Heart Infusion (BHI) broth (Difco Laboratories, Detroit, Michigan, USA). The liver, spleen and mesenteric lymph nodes were removed and placed in preweighed tubes containing Tryptic Soy Broth (TSB) (Oxoid, Basingstoke, Hampshire, UK). The peritoneal cavity and the peritoneal viscera were swabbed with sterile cotton tipped applicator sticks and cultured in BHI broth in order to demonstrate accidental bacterial contamination. Finally, the caecum was opened and 0.5 ml caecal content was inoculated into TSB.
Aliquots of 0.2 ml of homogenized organ samples and the supernatant of the caecal content after settlement of vortexing were inoculated into 3 ml BHI broth, and plated on blood agar and McConkey agar plates (Oxoid, Basingstoke) for the culture of aerobic and microaerophilic bacteria. Bacto-Rogosa SL broth (Difco Laboratories) was used for culture of Lactobacilli and anaerobic bacteria. Aerobic and anaerobic cultures of blood were incubated at 37 C for 7 days and read daily. If any macroscopic bacterial growth was noted, subcultures on blood agar were performed. All organ samples were incubated aerobically at 37 C overnight, as well as anaerobically at 37 C for 7 days before being discarded. Enteric Gram-negative bacteria were identified by using the API 20 system (BioMeriux SA, Marcy 1' Etoile, France), while all other aerobic, microaerophilic and anaerobic bacteria were identified with standard proceduresTM. For identification of Lactobacillus acidophilus, API 50CH (Analytab Products Inc., Plainview, New York, USA) was used.

Morphological Study
The terminal ileum was removed after the animals were killed. The specimens were fixed in 10% formalin in 0.15 M phosphate buffer, pH7.2 (effective osmolar pressure 300 m OSM), embedded in paraffin, stained with haematoxylin and eosin and examined under the light microscope.

Scanning Electron Microscopy (SEM)
The terminal ileum was removed from both controls and the jaundiced animals immediately after the animals were killed. The small intestine was cannulated and carefully rinsed with lactated Ringer's solution. The specimens were then drip-fixed in 2.5% Glutaraldehyde in phosphate buffer. Further fixation was achieved by immersion in 2.5% Glutaraldehyde for 24 hours and subsequently the specimens were extensively rinsed in Millonig's phosphate solution. The specimens were prepared for scanning electron microscopy (SEM) by dehydration in ethanol and transferred to the critical point drier. The mounted specimens were sputtered in a polaron 5400 sputtercoater unit with gold-palladium and studied in a JEOL T 330 SEM (Tokyo, Japan).

Statistical Methods
One way analysis of variance (ANOVA) was used for parametric data, while the comparison of bacterial translocation rates between the groups was calculated by Chi-square with Yates' correction. All values were expressed as mean + SEM. A probability less than 5% was considered statistically significant.

RESULTS
All rats with common bile duct ligation were jaundiced and bile was present in the urine. No bilio-intestinal patency could be found in jaundiced rats when performing choledochoduodenstomy or following the bacterial translocation study (sacrifice).

Liver Function
Serum levels of bilirubin, alanine aminotransferase, aspartate aminotransferase and alkaline phosphatase were significantly elevated (p < 0.001) in animals with 2 weeks of biliary obstruction as compared with controls and biliary decompressed animals, and decreased to normal levels one week following biliary decompression, without significant difference (p > 0.05) among the decompressed animals ( Table  1). The antipyrine clearance rate decreased 20% (p < 0.01) after 2 weeks of biliary obstruction and normalized (p < 0.01) within 2 weeks following biliary decompression, after which the clearance rate did not further increase ( Table 2).

Bacterial Translocation
Mesenteric lymph nodes (MLNs) were the sole source of bacterial growth. The incidence of positive bacterial cultures in MLNs in animals with 2 weeks of :I: p<0.001 as compared with controls and all biliary decompressed groups.
1W BD: one week following biliary decompression. 2W BD" Two weeks following biliary decompression. 3W BD: Three weeks following biliary decompression. 5W BD: Five weeks following biliary decompression. "I" p<0.01 as compared with controls and biliary decompressed groups.
2W BD: Two weeks following biliary decompression. 3W BD: Three weeks following biliary decompression. 5W BD" Five weeks following biliary decompression. obstructive jaundice (47%) was significantly higher (p < 0.05) than in controls (10%). The rate of positive cultures in MLNs consistently decreased from (29%) 1 week after biliary decompression to nil both 3 and 5 weeks following the relief of obstructive jaundice ( Table 3). The bacteria most frequently identified from the MLNs included Escherichia coli, Enterobacter sp., Pseudomonas sp., Proteus sp. and in some instances Staphylococcus sp. Thus, enteric bacteria dominated culture findings in MLNs. One positive culture from the liver was noted in all groups except for group 5 (3W BD). Two positive bacterial cultures from the spleen and one positive blood culture were found in group 3 (lW BD).
Caecal levels of E. coli did not significantly differ among the various groups (p > 0.05), though the level in animals with 2 weeks of biliary obstruction tended to be highest. Viable counts of aerobes and microaerobes in rats 1 week after biliary decompression were significantly higher (p < 0.01) as compared with both controls, jaundiced and all other biliary decompressed animals. The counts of Lactobacilli in jaundiced and decompressed animals were lower than in controls, though without statistical significance (p > 0.05). A tendency of recovery of Lactobacilli levels after biliary decompression was noted ( Table 4).
bacterial translocation rate significantly correlated (R 0.844; p 0.034) with alkaline phosphatase activity and that it tended to correlate, though not significantly, with bilirubin, alanine aminotransferase and aspartate aminotransferase (Table 5).

Intestinal Morphology
The intestinal specimens from animals with 1, 3 and 5 weeks of biliary decompression were examined with light microscopy. Dilated lymph vessels and oedematous subepithelial regions with infiltration of inflammatory cells in the villi were demonstrated in jaundiced animals, findings that were reduced with time after biliary decompression (Figure 1).

Scanning Electron Microscopy
Numerous Lactobacilli adherent to the surfaces of the intestinal epithelial cells could be demonstrated using SEM in control animals, while only sparse Lactobacilli was noted on the surfaces of the epithelial cells in jaundiced rats ( Figure 2).

DISCUSSION
Biliary tract surgery in jaundiced patients is frequently associated with septic complications1-3, such as biliary sepsis, wound infection, intraabdominal abscess formation, and renal insufficiency19'2. Gunn reported that wound infection, subphrenic abscess and septicemia are 4 times more common in patients with infected bile than in patients in whom the bile is sterile. Bacteriaemia and septicemia in obstructive jaundice were usually demonstrated to be secondary to both the occurrence of infected bile and increased biliary pressure21. The precise aetiology of infectious complications in jaundiced patients without biliary infection is, however, still uncertain. Furthermore, bile acids have been demonstrated to inhibit bacterial growth both in vivo and in oitro28'29, in the current study the incidence of bacterial translocation gradually, but slowly decreased following biliary drainage by performing a choledochoduodenostomy. The incidence of bacterial translocation after relieving obstructive jaundice for one week was still high (29%). Long term obstructive jaundice and subsequent operative trauma might act as a major insult leading to suppression of host immunity and resultant bacterial overgrowth. No bacterial translocation was noted three weeks after relief of biliary obstruction. The tendency towards recovery of intestinal Lactobacilli counts and diminished rate of bacterial translocation after biliary decompression implies the importance of biliary decompression in order to reduce the susceptibility to septic complications. Compromised host immunity promotes bacterial translocation in obstructive jaundice. Nonspecific 3 and specific cellular immunity31, as well as reticuloendothelial dysfunction 8-1 have all been reported in experimental biliary obstruction. The clinical significance of bacterial translocation and transient bacteriaemia, however, depends upon several factors like the integrity of the host defense and invasiveness of the microorganisms. Bacterial translocation from the gut may thus be of no clinical significance if host defenses are adequate to control and ultimately eradicate the translocating bacteria. However, if the host defenses are impaired enough, then bacterial translocation might lead to a systemic infection31. Biliary decompression has been shown to reverse the impaired immunity in biliary obstruction31'33. The present results on bacterial translocation supports previous findings by this group1, Where RES function was demonstrated to be in a depressed state still one and two weeks after biliary decompression, recovering in the third week after relief of obstructive jaundice.
The integrity of the intestinal mucosal barrier function is critical for preventing microorganisms from invading. Subepithelial oedema, dilated lymph vessels and inflammatory cell infiltration were noted in obstructive jaundice in the present study. It appears that these pathological changes might diminish the efficiency of the intestinal barrier function, though intestinal integrity otherwise is maintained. The histological changes found in the intestine following biliary obstruction and initially following biliary decompression, might thus contribute to an impaired local host defence and increased permeability. These pathological alterations reduced with time after biliary decompression. Further studies are warranted to investigate the intestinal morphological changes and their implications in obstructive jaundice. Liver function, especially the activity of alkaline phosphatase, correlated with the incidence of bacterial translocation. Since the liver is involved in a variety of metabolic processes, either liver disease 4' or obstructive jaundice36, could lter various metabolic mechanisms in the body, which might then affect intestinal function and lead to bacterial translocation. Antipyrine clearance capacity is a measurement of liver microsomal function and correlates with other "vital hepatic functions" of non-microsomal nature such as galactose elimination capacity and the capacity for urea-N-synthesis37. Auranen et al. 38 reported a significant decrease in glucose-6-phosphatase activity, in aminopyrine demethylation, hexobarbital metabolism and cytochrome P-450 content, indicating impairment of liver microsomal function in obstructive jaundice. Antipyrine clearance capacity returned to normal levels two weeks after biliary decompression in our study. It seems that the recovery of liver function after biliary decompression preceded the reduction of the rate of bacterial translocation.
In conclusion, bacterial translocation occurs in obstructive jaundice in the rat and decreases after biliary decompression.

Acknowledgement
This study has been supported by grants from the Medical Faculty, Lund University and the Lars Hierta Memorial Foundation.