Bacteremia and Bacterial Translocation in Horses with Enterocolitis
ACVIM 2008
Imogen Johns, BVSc, DACVIM, MRCVS
North Mymms, UK

Introduction

Bacteremia is defined as the presence of viable bacteria within the bloodstream. The diagnosis of bacteremia relies on isolation of organisms from blood cultures. Bacteremia can result in initiation of the systemic inflammatory response syndrome (SIRS), as well as septic foci development throughout the body. The presence of bacteremia is associated with increased morbidity and mortality in human patients1, critically ill dogs and cats2 and cows with acute coliform mastitis3. In horses, bacteremia is considered a rare event4, and is most frequently diagnosed in neonatal foals, with the gastro-intestinal tract (GIT) presumed to be the source of the bacteria.5-7 Several case reports of bacteremia in adult horses with GIT disease have also implicated the gut as the source of the bacteria.8-10 The idea of the gut as the source bacteria in septic human patients with no obvious infectious focus has gained increasing acceptance over the last several decades.11-13 Bacterial translocation (BT), defined as the passage of viable enteric bacteria into the mesenteric lymph nodes (MLN) or other extra-intestinal sites, can result in localized and systemic inflammatory responses, as well as bacteremia.14-16 The 'gut-origin of sepsis' is now widely accepted as an explanation for the development of SIRS and sepsis in experimental models and human patients. The route by which bacteria and inflammatory mediators access the systemic circulation has been the topic of intense research. Initially, it was thought that bacteria that crossed the gut mucosal barrier entered the portal circulation, and from there disseminated to other organs after overwhelming the phagocytic abilities of the liver. Evidence for this route was provided by experimental studies in rats which identified portal vein bacteremia after intestinal-reperfusion injury.12 However, failure to consistently identify either intact bacteria or endotoxin within the portal vein of patients considered at high risk of BT led to the search for an alternate route.17 It is now thought that the predominant route by which enteric bacteria and associated inflammatory mediators access the circulation is via the lymphatics.18 Furthermore, it has been shown that although bacteria may access the systemic circulation in these models, their presence at sites beyond the MLN is not required to initiate SIRS after BT. Inflammatory mediators, such as endotoxin, and cytokines induced after immune system destruction of bacteria within the MLN, are sufficient to initiate this response. In a murine trauma-hemorrhagic shock model, mesenteric lymph from shock-affected animals induced neutrophil activation, endothelial cell activation, and increased epithelial cell monolayer permeability, whereas lymph from control mice did not. As the lymph was sterile and contained minimal amounts of endotoxin, other inflammatory mediators were implicated. 19

Defense Mechanisms of the GIT

Although BT is believed to be a normal phenomenon, in health it occurs at a rate which does not allow establishment of infection within the MLN.11 Clinically significant BT, either by indigenous normal flora or pathogenic bacteria, occurs when there is a breakdown in the gut barrier.15,16 The defenses which exist to prevent BT include the normal flora and peristaltic activity of the gut, a mucus layer, tight junctions between enterocytes, the mucosa itself, as well as lymphoid tissue associated with the gut.13 Within the lumen of the gut, bacteria that are essential to the normal digestive and absorptive processes co-habit with organisms that may act as potential pathogens. The normal flora predominates, and prevents overgrowth of these potential pathogens by occupying epithelial cell attachment sites, by secreting antimicrobial substances and by competing for nutrients.16,20,21 Disturbance of this fine balance, as may occur following administration of antimicrobials, can result in death of the protective normal flora, overgrowth of the potential pathogens, and possibly increase the chance that these pathogens will access extra-luminal sites. The normal peristaltic activity of the gut can also aid in prevention of bacterial epithelial cell adherence.13,22 The physical barrier between the bacteria-laden lumen of the gut, and the normally sterile circulation consists of a layer of mucus, the epithelial cells themselves, and the tight junctions between those cells. The mucus layer helps prevent bacteria from gaining access to sites of adherence on epithelial cells.23 Inflammation can cause a decrease in this mucus layer, as can therapy with drugs such as non-steroidal anti-inflammatory drugs (NSAIDs).24 The epithelial cells of the gut are connected via tight junctions, which function to prevent passage of large macromolecules and potential pathogens, whilst allowing for absorption of smaller molecules via the paracellular route. A break in the integrity of these tight junctions can allow previously excluded substances, such as bacteria, access across the mucosal barrier.25 The mucosal cells themselves provide an effective barrier to bacterial translocation. Damage to the mucosa, as may occur secondary to infection with Salmonella spp, can increase mucosal permeability and may allow for increased BT.26 The mucosa can be directly damaged during such infections, or may be injured secondary to conditions which can decrease splanchnic perfusion, such as hypovolemic or endotoxic shock.16 The gut-associated lymphoid tissue (GALT) provides a secondary line of defense should bacteria invade across the mucosal barrier. The GALT is considered the largest immunological organ of the body, and consists of Peyer's patches, lymphoid follicles, lamina propria lymphocytes, intraepithelial lymphocytes and mesenteric lymph nodes.20,21 Should bacteria breach the mucosa, phagocytes and lymphocytes within this system generally effectively contain their progress. Bacteria and endotoxin that gain access to the portal circulation are typically effectively scavenged in the liver by the Kupffer cells.20 A breach in any of these either functional or physical barriers may result in BT.

Diagnosis of Bacterial Translocation

The most reliable method of diagnosing BT is considered to be culture of the mesenteric lymph nodes.15,16 This method however limits subjects to those undergoing laparotomy. Surrogate tests of BT, such as permeability tests, are unreliable, as increased permeability does not necessarily equate to increased BT. Blood cultures are also used to diagnose BT, although the sensitivity is low, and the gut origin of the organism cannot be determined. The use of PCR performed on blood to diagnose BT may be more sensitive than conventional blood culture.27

Significance and Prevention of Bacterial Translocation

Bacterial translocation has been diagnosed in human patients with many disease states, including hemorrhagic shock, cirrhosis, abdominal surgery, neoplasia and intestinal obstruction.16 In a study of 927 human patients undergoing laparotomy, BT (positive MLN cultures) was identified in 130 (14%).28 Post-operative septic complications were significantly more likely to occur in those patients with BT. Additionally, the organisms responsible for the clinical infections were similar to those isolated in the MLN. The potentially devastating sequelae to BT have led to investigation of preventative strategies aimed at decreasing its occurrence in at-risk patients. Glutamine supplementation, targeted nutritional intervention, maintaining splanchnic flow, the judicious use of antimicrobials and directed selective gut decontamination may limit BT, although currently only limited data support these interventions.15

Bacterial Translocation in Veterinary Species

Methods to diagnose BT in veterinary patients have included blood cultures, culture of mesenteric lymph nodes in surgical patients, and post-mortem studies.2,29-32 The incidence of bacteremia and bacterial translocation was investigated in dogs with naturally occurring gastric dilatation-volvulus. Blood cultures, as well as cultures from the liver, mesenteric lymph nodes and stomach were collected during surgery in 21 dogs with GDV, and in 5 healthy control dogs.30 Although positive blood cultures were identified in 9/21 dogs with GDV, only three had positive mesenteric lymph node cultures, and the same bacterial species was not isolated from both sites. Despite the apparently high percentage of bacteremic dogs, this was not significantly different to control dogs, where 2/5 (40%) had positive blood cultures. In a study of 50 healthy dogs undergoing elective ovariohysterectomy, positive lymph node cultures were identified in 26 (52%) dogs.31 One dog was also bacteremic. A retrospective post-mortem study of 137 dogs with parvoviral enteritis revealed that 88/98 cases that had liver or lung cultures performed at necropsy were positive for E coli. Whilst not direct evidence of BT, it was assumed that the gut was the origin of the E coli.29 Bacteremia of presumed gut origin has been reported in up to 30% of calves with diarrhea.32 In adult cows, the well recognized phenomenon of liver abscesses secondary to ruminal acidosis occurs secondary to bacterial translocation across the damaged rumen wall.33

Evidence for Bacterial Translocation in Horses

Bacteremia is considered a common occurrence in foals with diarrhea.5,6,7 In contrast, evidence for its occurrence is limited in adult horses. Bacteremia has been reported in two adult horses with Salmonellosis, one of which had a positive blood culture for Salmonella,8 the other Pseudomonas aeruginosa.9 A horse diagnosed with granulomatous enteritis had a positive blood culture for Campylobacter fetus subsp fetus.10 Embolic fungal pneumonia, presumably secondary to fungemia, has been reported in critically ill adult horses.34 Twenty five of 29 horses in one report had a primary or secondary disease compatible with loss of integrity of the GI tract. The authors believed that the damaged GI mucosa allowed gut derived fungi to access the systemic circulation resulting in embolic mycotic disease. In addition, the profound neutropenia that typically accompanies severe colitis was thought to be a contributory factor, perhaps indicative of more generalized immunosuppression.

Based on an assessment of the risk factors for the development of BT, horses affected by inflammatory diseases of the intestine appear to be at great risk of developing BT and bacteremia. Antimicrobial administration is a common predisposing factor for development of diarrhea in horses, presumably secondary to loss of normal bacterial flora and overgrowth of pathogens such as Clostridia spp. Horses with colitis are typically treated with NSAIDs, which interfere with normal mucus production via inhibition of prostanoids. This inhibition can induce or potentiate inflammation in the gastrointestinal tract by injuring the mucosa (increased mucosal permeability and ulceration) or by delaying mucosal healing.35,36 Infection with Salmonella spp and Clostridium spp are well recognized causes of colitis in horses. Experimental infection with Salmonella spp has been shown to result in in vitro alteration of tight junctions, resulting in increased bacterial translocation of both pathogenic and non-pathogenic bacteria across cell monolayers.25 Similarly, Clostridium difficile toxins A and B have also been shown to increase mucosal permeability and bacterial penetration across mucosal cells in vitro.37,38 Acute enterocolitis results in loss of large volumes of fluid through the GIT, and absorption of endotoxin into the systemic circulation. The resultant hypovolemia and systemic inflammatory response may predispose to bacterial translocation via hypoxic injury to the epithelial cells, and subsequent ischemia-reperfusion injury. Despite these potential risk factors, the actual incidence of bacterial translocation, as determined by detection of bacteremia, is considered to be low in adult horses with inflammatory diseases of the GIT.4,39 However, the true incidence is unknown, as evidence is limited to case reports/series and anecdotal reports.

During 2005, 32 adult horses with acute diarrhea were treated at the George D Widener Hospital for Large Animals, New Bolton Center. Blood cultures, taken at the discretion of the supervising clinician in animals suspected of being bacteremic, were positive in two cases. In addition, septic foci outside the GI tract were identified in five horses at necropsy (one of which had a positive blood culture). Although suggestive that bacteremia may be more frequent in adult horses with diarrhea than previously thought, the results were likely biased as blood cultures were not routinely taken on all patients. As early and appropriate antimicrobial treatment is vital in preventing morbidity and mortality due to bacteremia, information regarding the frequency with which bacteremia occurs in this population is necessary. The purpose of the study was thus to prospectively determine the incidence of bacteremia in adult horses with AEC. Blood cultures were aseptically obtained from horses aged over one year suffering from acute enterocolitis. Cultures were obtained at admission and 24 hours after this. Horses were eligible for the study if they were admitted to the isolation facility with a history of diarrhea of less than three days duration. In addition, horses which developed diarrhea during hospitalization and which were transferred to the isolation facility according to hospital protocol, were included. Blood cultures were obtained from these animals at admission to the isolation facility and at 24 hours. Clinico-pathologic data was collected at admission and at 24 hours. Outcome was recorded as survival/non-survival, and complications during hospitalization were recorded. Administration of antimicrobials was determined prior to admission, as well as during hospitalization.

The results of this study will be presented at the meeting.

References

1.  Bearmann GML, et al. Archives of Medical Research 2005;36: 646

2.  Dow ST, et al. J Am Vet Med Assoc 1989;195:113.

3.  Wenz JR, et al. J Am Vet Med Assoc 2001;219:976.

4.  Roy MF. Vet Clin Equine 2004;20:41.

5.  Marsh PS, et al. J Am Vet Med Assoc 2001;218:1608.

6.  Stewart AJ, et al. J Vet Intern Med 2002;16:464.

7.  Hollis AR, et al. Research abstract 2008 ACVIM Forum.

8.  Hondalus MK. J Am Vet Med Assoc 1988; 192: 527.

9.  Johns IC, et al. J Vet Emerg Crit Care 2006;16:219.

10. Johnson PJ, et al. J Am Vet Med Assoc 1993;203:1039.

11. Berg RD. Trends in Microbiology 1995;3:149.

12. Luo CC. Biol neonate 2004;85:151.

13. Swank GM, et al. World J Surg 1996;20:411.

14. Berg, et al. Infect Immun 1979;23:403.

15. Gatt M, et al. Aliment Pharmacol Ther 2007;25:741.

16. Balzan S, et al. J Gastroenterol Hepatol 2007;22:464.

17. Moore FA, et al. J Trauma 1991;31:629.

18. Deitch EA. Surgery 2002;131:241.

19. Magnotti LJ, et al. Annals Surg 1998;228:518.

20. Macintire DK, et al. Vet Clin Small Anim 2002;32:1165.

21. Acheson DW, et al. Best Pract Res Clin Gastroenterol 2004;18:387.

22. Erbil Y, et al. Acta Chir Belg. 1998;98:245.

23. Swidsinski A, et al. Gut 2007;56:343.

24. Tomlinson J, et al. J Am Vet Med Assoc 2003;222:946.

25. Kohler H, et al. Am J Physiol gastrointest physiol. 2007;293:G178.

26. Jones SL. In: Equine Internal Medicine 2004. 884.

27. Ono S, et al. World J Surg. 2005;29:535.

28. MacFie J, et al. Br J Surg. 2006;93:87.

29. Turk J, et al. J Am Vet Med Assoc. 1990;196:771.

30. Winkler KP, et al. J Am Anim Hosp Assoc 2003;39:351.

31. Dahlinger J, et al. J Vet Intern Med 1997;11:319.

32. Constable PD. J Vet Intern Med 2004;18:8.

33. Garry FB. Indigestion in ruminants. In: Smith BP (ed). Large Animal Internal Medicine. 2002; 731.Mosby.

34. Sweeney CR, et al. J Am Vet Med Assoc 1999;214:808.

35. Bilkslager AT, et al. J Am Vet Med Assoc 1998;211:1437.

36. Tomlinson J, et al. J Am Vet Med Assoc 2003;222:946.

37. Hecht G, et al. Gastroenterology. 1992;102:416.

38. Feltis BA, et al. Shock. 2000;14:629.

39. Papich MG. Vet Clin Equine.2003;19:645

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Imogen Johns, BVSc, DACVIM, MRCVS
The Royal Veterinary College
Hatfield, Hertfordshire, United Kingdom


MAIN : ACVIM Equine Specialist : Horses with Enterocolitis
Powered By VIN
SAID=27