Oral Antibiotics Attenuate Bowel Segment Reversal–Induced Systemic Inflammatory Response and Body Weight Loss in Massively Bowel-Resected Rats  ,
From the * Division of Pediatric Surgery,
Department of Surgery, Changhua Christian Hospital, Changhua, Taiwan; Correspondence: Hui-Chen Lo, PhD, Department of Medical Education and Research, Changhua Christian Hospital, 135 Nanhsia Street, Changhua 500, Taiwan. Electronic mail may be sent to 66161{at}cch.org.tw.
Background: Using a massively bowel-resected rat model, our
previous study demonstrated that small bowel segment reversal stimulates
jejunal hyperplasia but may also increase the possibility of bacterial
translocation and the elevation of circulating white blood cells and serum
interleukin-6 that may reduce the whole-body anabolism. The aim of this study
is to investigate whether oral antibiotics might attenuate the inflammatory
responses and might therefore facilitate the beneficial effects of bowel
segment reversal. Methods: Male Wistar rats ( Diarrhea, weight loss, and malnutrition are the major features of short bowel syndrome brought about by the removal of a large portion of the small intestine.1–3 Even though massive bowel resection is a common and lifesaving operation for patients with extensive bowel abnormalities, thrombosis, or injuries, this operation still causes some adverse effects, especially in nutrition status. Various surgical procedures have been used in an attempt to prolong the transit time, to expand absorptive surface area of the small intestine, and to further improve the nutrition status,4–9 and several nutrition approaches, such as parenteral nutrition (PN),10 early enteral feeding,11–14 and nutrient supplementation,15,16 have been applied to prevent the occurrence of malnutrition in patients with massive bowel resection. However, the clinical experiences of these surgical procedures and nutrition care are still limited and controversial. Our previous study demonstrated that the surgical reversal of a small bowel segment stimulates jejunal hyperplasia and increases intestinal transit time in rats with massive bowel resection and early enteral feeding.17 However, bowel segment reversal–induced bacterial overgrowth in the intestine might provoke extended acute-phase responses, such as elevation in the numbers of circulating white blood cells and increase of serum interleukin-6 concentrations that may reduce the whole-body anabolism. These bowel segment reversal–induced adverse responses might be related to the disturbed gut-associated defense barrier, because the broken intestinal mucosa and muscularis may allow the translocation of intestinal pathogens to the internal environment.6,18 It is worth noting that patients with reversed bowel segments and continuous enteral alimentation complain about abdominal pain. Pigot and his coworkers6 attributed these adverse symptoms to the bacterial overgrowth in the reversed jejunum, which can be attenuated with oral antibiotic therapy. Because our previous study observed that bowel segment reversal significantly increased fecal and bacterial load in the intestine, we believe that the bowel segment reversal– induced adverse response may be attenuated by antibiotic therapy. Therefore, in our study, we aimed to investigate the beneficial effects of oral antibiotics on whole-body growth, inflammatory responses, and jejunal histology in massively bowel-resected rats with small bowel segment reversal.
Animals and Study Design The animal facilities and protocols used in this study were approved by the Laboratory Animal Care and Use Committee in Changhua Christian Hospital, Changhua, Taiwan. Male Wistar rats (National Taiwan University, Taipei, Taiwan) initially weighing 180–250 g, were acclimatized to the animal facility, with free access to water and chow diet in a room maintained at 22°C on a 12-hour light-dark cycle for 2 weeks before surgery. These animals were randomly assigned to 4 groups (ie, sham-resected [small bowel transection and reanastomosis; SHAM], control-resected [70% small bowel resection; CON], and resected plus reversal [70% small bowel resection and 3-cm reversed segment] with intragastric administration of clindamycin and amoxicilline [50 and 50 mg/kg/d; ANT], or distilled water [REV]). After being fasted overnight, animals were anesthetized by intramuscular injection with 80 mg of ketamine and 8 mg of xylazine per kg of body weight and then underwent massive bowel resection with or without bowel segment reversal (day 0). To perform the 70% intestinal resection, the small bowel was cut 3 cm distal to the ligament of Treitz and 6 cm proximal to the cecum after the mesenteric vessels were ligated with 5-0 silk suture. This part of the intestine, approximately 70% of total length of the small intestine, was removed. To perform the reversal, a 3-cm small bowel segment, which was 3 cm distal to the resected segment, was reversed, and then the remaining jejunum and ileum were constructed by 2 end-to-end anastomotic sutures (Figure 1), as described in the previous study.17,19 The length and position of the small bowel segment to be reversed did not cause intestinal blockade, as our experience has shown from preliminary experiments and the observation of fecal excretion. One day before the surgery, animals in the ANT group were gavaged with clindamycin and amoxicillin (50 mg/kg and 50 mg/kg dissolved in 0.2 mL of distilled water). The reasons for choosing these 2 antibiotics are that amoxicillin (Synpac-Kingdom Pharmaceutical Co., Ltd., Taipei, Taiwan), an analog of ampicillin, with a broad spectrum of bactericidal activity against many Gram-positive and Gram-negative microorganisms, is used as the drug of first choice in many medical situations,20 and clindamycin (Nang Kuang Pharmaceutical Co., Ltd., Tainan, Taiwan) is prescribed for patients with intra-abdominal infections caused by susceptible anaerobic bacteria.21 This antibiotic treatment was administered daily during the experiment. Animals in the CON and SHAM groups, on the other hand, were gavaged with distilled water instead. The final sample size was 8–9 rats per group, with an animal loss of 25%, mostly from day 0 to day 3, because of internal abdominal bleeding confirmed by abdominal anatomy. Animals were allowed free access to 5% glucose water for prevention of hypoglycemia during the first 2 postoperative days (day 0 and day 1). From day 2 to day 21, animals were fed with a semipurified powder diet (AIN-76; ICN Biomedicals Inc., Costa Mesa, CA). On day 14, rats were gavaged with 1 g of carbon powder in 4 mL of distilled water to assess the intestinal transit time, which seems to be a less quantifiable but easy, inexpensive, and clinically applicable approach. Thereafter, feces were observed daily to check for the appearance of carbon powder. Food intake and body weights were recorded daily. Feeding efficiency (g body weight gain per 100 kcal) was also calculated. Twenty-one days after surgery, rats were killed under anesthesia, with intramuscular injections of 100 mg ketamine and 10 mg xylazine per kg of body weight. The order of killing was randomized among the groups. Blood was collected by cardiac puncture, after which the serum and whole blood were isolated for further assays. The heart, lung, liver, thymus, spleen, kidney, cecum, and gastrocnemius muscles were dissected and weighed, and the data were recorded. The entire gastrointestinal tract was also removed. The total weight and length of small intestine were determined after removal of the gut content using a cold saline rinse. After removal of the above mentioned organs and tissues, carcass weights were recorded and carcasses were stored at –20°C for carcass composition analysis (ie, water, fat, and protein).
Analytic Measurements
The harvested small intestine without gut contents was divided into 3 parts: proximal, reversed, and distal segments. About 0.5 cm of small intestine in each segment was fixed in 10% buffered formalin for routine paraffin embedding, hematoxylin and eosin (H & E) staining, and morphometric measurement under a light microscope. The villus height, crypt depth, and muscularis thickness were then determined. At least 10 villus-crypt axes were measured per animal. One investigator was assigned to perform all measurements to prevent interobserver differences in the measuring technique. The gut content from the small intestine was collected separately and cultured on the tryptic soy agar for examining bacterial growth (DIFCO Laboratories, Detroit, MI). To determine the mucosal wet and dry weights, half of the remaining intestine was also collected and the mucosa obtained by scraping the small intestine with a glass slide were performed by the same individual to reduce variation. The rest of the small intestine from each segment was used for the analysis of protein (bioinchoninic acid protein assay; Pierce Chemical Co, Rockford, IL) and DNA contents.9
Statistical Analysis
Whole-Body Growth and Energy Intake Body weight gain and the amount of food intake (day 3– day 21) are shown in Figure 2. There were no significant differences in body weight among groups before surgery, on the day of surgery (day 0), and after the early postoperative days (day 1– day 3). Animals in the SHAM group started to have significantly greater body weight on day 4 than those in the other groups. Animals with segment reversal (REV group) weighed significantly less on days 20 and 21 than animals with sham operation (SHAM group) or massive bowel resection (CON group, data not shown). Body weight gain was shown to be approximately 25% less in the CON group, 50% less in the REV group, and 33% less in the ANT group compared with that in the SHAM group (Figure 2A). Animals in the CON, REV, and ANT groups had a significantly lower amount of food intake than those in the SHAM group (data not shown), with no significant difference among the 3 groups. When calculating feeding efficiency from body weight gain divided by caloric intake, we found that animals with resection plus segment reversal had significantly decreased feeding efficiency compared with those with sham and control resection, whereas antibiotic administration significantly attenuated this decrease (Figure 2B).
Relative Weights of Organs and Tissues and Carcass Composition
The relative weights of organs and tissues are shown in
Table I. The relative weights
of liver and heart were significantly greater in the REV group than in the CON
group, but that of the heart was significantly lower in the ANT group than in
the REV group. Likewise, the relative weights of kidneys were significantly
lower and that of the cecum was significantly greater in the ANT group than in
the CON and REV groups. The length of small intestine was also significantly
greater in the SHAM group (
Blood and Serum Substrate Concentrations
Serum concentrations of albumin, glucose, and insulin-like growth factor-I
were significantly decreased, whereas those of nitric oxide and interleukin-6
were significantly increased in the REV group compared with the SHAM and CON
groups (Table III). Antibiotic
administration significantly attenuated the bowel segment reversal-induced
decrease in serum insulin-like growth factor-I and the increases in the serum
nitric oxide and interleukin-6. In addition, antibiotic administration
significantly decreased serum concentration of glutamic oxaloacetic
transaminase and increased serum concentration of creatinine in massively
bowel-resected rats with bowel segment reversal. There were no significant
differences in serum concentrations of cholesterol, blood urea nitrogen,
glutamic pyruvic transaminase, insulin, and tumor necrosis factor-
The results of Pearson's method indicated that feeding efficiency had
significantly positive correlations with serum concentrations of albumin
(r = 0.338; p = .033) and insulin-like growth factor-I
(r = 0.614; p < .001), but had significantly negative
correlations with glutamic pyruvate transaminase (GPT; r =
–0.362; p = .035) and interleukin-6 (r =–0.329;
p = .046). There were no significant correlations between feeding
efficiency and serum concentrations of glucose, cholesterol, triglyceride,
glutamic oxaloacetic transaminase, blood urea nitrogen, creatinine, insulin,
nitric oxide, and tumor necrosis factor-
Composition and Architecture of the Small Intestine
The results of mucosal composition are shown in Figures 6 and 7, respectively. In the mucosa, massive bowel resection significantly increased the dry weight and protein content of the proximal and reversed segments (Figures 6A and 6B), whereas antibiotic administration significantly decreased the dry weight of the distal segment (Figure 6C). The DNA content of the mucosa was significantly increased in animals with massive bowel resection in the proximal but not the distal segment (Figure 6A and 6C) and was significantly increased in animals with massive bowel resection plus segment reversal in the reversed segment (Figure 6B). In the muscularis, massive bowel resection significantly increased the dry weight and DNA content, and bowel segment reversal significantly increased the protein content in the proximal segment (Figure 7A). There were no significant differences in the muscularis dry weight, protein, and DNA contents in the reversed and distal segments (Figure 7B and 7C).
Gastrointestinal Transit Time and Bacterial Growth All animals had normal fecal excretion from day 3 to day 21. Using carbon powder as the indicator of intestinal transit time, we found that the carbon powder was excreted into the feces on the day (day 14) it was fed to the animals with (the CON, REV, and ANT groups) or without (the SHAM group) massive bowel resection. In addition, on the day that the animals were killed (day 21), a heavy amount of carbon powder was found in the REV group, an intermediate amount was found in the ANT group, and a trace amount was found in the SHAM and CON groups. The REV and ANT groups had about 10- and 2-fold amounts of gut content than the CON and SHAM groups, respectively. In addition, the SHAM, CON, and ANT groups had about 15-fold numbers of bacteria per g of gut content (2.12 x 106, 1.80 x 106, and 1.65 x 106 colony forming units [CFU]/g, respectively) than the R group (4.50 x 103 CFU/g). The REV group had about 21-fold numbers of bacteria per g of gut content (4.54 x 107 CFU/g) than the SHAM, CON, and ANT groups.
It has been demonstrated that the reversed jejunal segment acts as an antiperistaltic segment to retrograde peristalsis, and disrupts the motility of the proximal intestine by reducing the frequency of the basic electrical rhythm.9 Clinical evidence indicates that the length and location of the reversed small bowel segment may result in varying outcomes.9 In our previous study, animals underwent a 5-cm reversed segment, approximately 13% of the remaining small intestine after resection, which is, for rats, a relatively long reversed segment. In the present study, we reversed 3 cm instead and used the combination of amoxicillin and clindamycin to control for the bacterial overgrowth in the small intestine. This combination treatment not only can eliminate the growth of aerobic and anaerobic organisms, but also can avoid the possibility of bacterial resistance and the side effects resulting from a high dose of a single antibiotic drug. Utilizing carbon powder as a simple indicator of intestinal transit time, animals with 3 cm of reversed jejunal segment and oral antibiotics had a greater amount of carbon powder than those with sham operation or massive bowel resection alone. In addition, the numbers of bacteria per gram of feces were significantly decreased when oral antibiotics were administered. These results suggest that a 3-cm jejunal segment reversal with oral antibiotic administration is efficient in increasing gastrointestinal transit time and in preventing bacterial overgrowth in massive bowel resection in rats. An added beneficial effect of oral antibiotics is the improvement of the whole-body growth in massively bowel-resected rats with bowel segment reversal. For instance, bowel segment reversal significantly decreased body weight gain and carcass weight, and oral antibiotics tended to attenuate this adverse response (Figure 2A). Interestingly, the results of Pearson's method clearly showed that serum albumin and insulin-like growth factor-I, an anabolic hormone that is regulated by nutrition and modulates somatic and tissue growth,24 positively correlated with feeding efficiency. The increased feeding efficiency (Figure 2B) and the elevated serum concentrations of albumin and insulin-like growth factor-I (Table III) confirmed the advantageous effects of oral antibiotics in massively bowel-resected rats with bowel segment reversal.
It has been reported that circulating interleukin-6 is a better biomarker
of postoperative trauma, inflammation, and sepsis than circulating tumor
necrosis factor-
The bowel segment reversal–induced hepatic dysfunction, as evidenced
by the elevated serum levels of glutamic oxaloacetic transaminase, has been
confirmed in the present study. The results of our previous study suggested
that massive bowel resection might cause decreases in hepatocytic numbers and
function, as evidenced by the decreased protein content and the decreased
levels of tumor necrosis factor- Intestinal adaptation after massive bowel resection is characterized by increases in mucosal mass, villus height, and crypt depth, thickness of bowel wall, and production of brush border enzymes and occurs within a short period.1,32 It has been demonstrated that the enteral intake of food is important for intestinal compensatory hyperplasia after resection.33 In our study, we observed that animals with massive bowel resection and early enteral feeding have significant intestinal adaptation as shown in the increases in mucosal dry weight, protein, and DNA contents in the proximal and reversed segments (Figure 6A and B) and in the relative weights of cecum (Table I). Animals with surgical reversal of a small bowel segment show further intestinal adaptation in the reversed segment, such as increases in crypt depth and villus height in the proximal segment and in mucosal DNA content in the reversed segment. However, disadvantages of oral antibiotics on intestinal histology have been shown in the present study. For example, the intestinal adaptation in crypt depth and villus height in the proximal segment and the DNA content in the reversed segment were partially withdrawn in animals given oral antibiotics. Therefore, oral antibiotics should be used under careful monitoring in patients with intestinal surgery, especially in their side effects on intestinal physiology, as reported in animal studies.31 The result of this study noticeably raises an idea that is worth further study, that is, the attenuation of intestinal adaptation by antibiotics might have important implications in the long-term effect of reversed segment compared with the short-term benefit of promoting anabolism.
In summary, oral antibiotic administration attenuates the segment reversal–induced body weight loss and systemic inflammatory responses in rats with massive bowel resection and early enteral feeding. Massively bowel-resected rats with bowel segment reversal and oral antibiotics had significantly increased feeding efficiency and serum insulin-like growth factor-I, as well as having significantly decreased numbers of intestinal bacteria, serum nitric oxide, and serum interleukin-6 compared with those rats with bowel segment reversal alone. In patients with massive bowel resection, the surgical procedure of bowel segment reversal has been used to increase the intestinal transit time; however, the copious fecal and bacterial load might increase the risk of bacterial translocation and subsequent acute-phase responses. The results of the present study revealed that oral antibiotic administration might be useful in controlling the bacterial overgrowth and inflammatory responses. It is worth noting that oral antibiotic administration would partially decrease the effects of bowel segment reversal in jejunal hyperplasia. Therefore, attenuating bacterial overgrowth in the intestine by oral antibiotic administration, monitoring the systemic inflammatory response by circulating cytokine measurement, and increasing nutrient intake by early enteral feeding may be potential strategies to enhance the beneficial effects of surgical reversal in massively bowel-resected patients.
The work was supported by the National Science Council of the Republic of China under grant number NSC 92-2320-B-371-001. Ming-Chih Chou and Chien-Hsing Lee contributed equally. We thank Su-Chen Lin, Fu-Ann Tsai, and Ya-Chi Lai for their technical support. Received for publication November 6, 2006. Accepted for publication March 30, 2007.
Journal of Parenteral and Enteral Nutrition, Vol. 31, No. 5,
397-405 (2007) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
270 g) underwent a
70% small bowel resection with (REV group) or without (CON group) a 3-cm small
bowel segment reversal, or underwent a sham operation (SHAM group). After
surgeries, half of the animals in the REV group were given oral clindamycin
plus amoxicillin (50 plus 50 mg/kg/d, ANT group) for 3 weeks.
Results: Oral antibiotics administration significantly attenuated the
decreases in feeding efficiency (g of body weight/100 kcal diet) and increases
in the circulation of white blood cells, serum nitric oxide, and interleukin-6
(1-way ANOVA, p < .05), which are associated with bowel segment
reversal. In addition, antibiotics significantly increased serum
concentrations of insulin-like growth factor-I, significantly decreased the
total numbers of bacteria in the intestine, and tended to reduce the extent of
jejunal hyperplasia in rats with bowel segment reversal. Conclusions:
Our results suggest that oral antibiotics may be used as an adjuvant to
attenuate the inflammatory responses and to enhance the anabolic responses in
massively bowel-resected patients with bowel segment reversal. 
Top
MATERIALS AND METHODS
RESULTS
, and interleukin-6 (DuoSet;
R&D System, Minneapolis, MN; and Pharmingen Inc, San Diego, CA). All of
the samples were then analyzed in 1 assay, with a duplicate. The interassay
coefficients of variance are 
