Two Cases of Lactobacillus Bacteremia During Probiotic Treatment of Short Gut Syndrome

Lactobacillus, a gram-positive bacillus, is a constituent of the indigenous flora of the colon. Often used as a probiotic agent, it appears to help prevent both intestinal colonization by pathogenic organisms and bacterial overgrowth syndrome. Lactobacillus organisms in the intestine are thought to produce acetic acid, lactic acid, and hydrogen peroxide and to promote the secretion of antimicrobial substances. The production of short chain fatty acids and the resultant low luminal pH of the colon also appear to inhibit growth of bacterial pathogens (¹). Lactobacillus may also compete with pathogenic organisms for mucosal surface receptors and may decrease the incidence of antibiotic-associated diarrhea and Clostridium difficile colitis (²).

Although Lactobacillus generally is not considered a pathogen, it can cause disease in compromised hosts, including bacterial endocarditis, pleuropulmonary infections, gastrointestinal abscesses, urinary tract infection, conjunctivitis, dental caries, and endometritis (³). Lactobacillus bacteremia, presumably secondary to bacterial translocation from the gastrointestinal tract, has been reported in a patient with severe intestinal inflammation caused by ulcerative colitis (⁴). There are no published reports of bacteremia or sepsis secondary to Lactobacillus given as a probiotic agent. We report two cases of Lactobacillus GG sepsis occurring during the therapeutic use of this organism.

CASE REPORTS

Case 1

Patient 1 was a 36-week-gestation male infant with short gut syndrome secondary to congenital intestinal atresia and volvulus. The patient was dependent on total parenteral nutrition (TPN) because of his very short residual intestine. He had experienced cholestasis. A regimen of Lactobacillus GG (Culturelle, ConAgra, Omaha, Nebraska) supplementation was started on the 95th day of life, with the patient receiving the contents of 1 capsule daily via gastrostomy tube without apparent complication. On the 23rd day of supplementation, the patient experienced fever and diarrhea, with signs and symptoms consistent with sepsis or infection. Blood cultures were obtained, and antibiotic therapy with ceftriaxone and ampicillin was initiated. Within 48 hours, blood cultures grew a Lactobacillus species sensitive to ampicillin. Blood cultures taken after 10 days of treatment with ampicillin were negative. Although fingerprinting of the isolate was not performed, we suspected that the source of the lactobacillus sepsis might have been the exogenous probiotic. Lactobacillus GG supplementation was discontinued, and the patient had no additional Lactobacillus bacteremia. On endoscopy, the patient’s intestine was inflamed and friable. We suspect that the fragility of the intestine may have promoted transmigration of the supplemental Lactobacillus GG.

Case 2

Patient 2 was a 34-week-gestation male infant with gastroschisis. At delivery, most of the intestine was infarcted, leaving the infant with approximately 16 cm of bowel distal to the ligament of Treitz. He underwent a gastrostomy and jejunostomy shortly after birth, was dependent on TPN, and rapidly developed cholestatic liver disease. He began a regimen of Lactobacillus GG, 1 capsule by mouth daily, on day of life 17. It was hoped the regimen would prevent bacterial overgrowth of the small intestine. He tolerated the preparation well. On day of life 186, he experienced an elevated temperature, tachycardia, and increased apneic events. Cultures were obtained, and ceftriaxone and ampicillin were administered. Within 48 hours, the blood cultures grew Lactobacillus sensitive to both antibiotics. The Lactobacillus GG supplement was discontinued, and the patient completed a 10-day course of ceftriaxone and ampicillin without complications. His repeat blood cultures while receiving antibiotics were negative, and he had no additional episodes of Lactobacillus bacteremia. Upper endoscopy after the sepsis episode revealed mild inflammation of the intestine. As in case 1, we suspected that the patient’s intestinal inflammation led to translocation of Lactobacillus GG from the intestinal lumen. In this case, we confirmed through DNA fingerprinting that the strain of Lactobacillus in the probiotic supplement was identical to the strain causing sepsis. DNA fingerprinting was performed at the University of Washington Department of Clinical Microbiology using pulsed field gel electrophoresis of chromosomal DNA with SMA1, SFi1, and Not1 digests according to standard protocol. The supplemental strain was indistinguishable from the blood culture isolate (Fig. 1).

FIG. 1.: Pulsed-gel electrophoresis of Lactobacillus GG sample and Case 2 bloodsample using SFi1, Not1, and SMA1 digests. Lane #1 is the control for the gel. Lanes #2 and 4 for each digest are Case 2 blood sample. Lanes #3 and 5 for each digest are Lactobacillus GG sample.

DISCUSSION

Sepsis related to Lactobacillus GG supplementation in short bowel syndrome has not previously been reported. Both of our patients had the identical underlying medical conditions of ultra short gut, cholestasis, and chronic intestinal inflammation. Lactobacillus treatment was initiated in these patients to help prevent bacterial overgrowth in the setting of short bowel syndrome. Although the possibility of a primary intravenous line sepsis with Lactobacillus cannot be excluded, we speculate that the gastrointestinal tract was the most likely portal of entry because appropriate line care measures were being practiced and the Lactobacillus was administered a distance from the patients’ intravenous insertion sites. We suspect that the mechanism for bacteremia in our patients was transmigration of the probiotic across an inflamed intestinal mucosa. Previous studies have demonstrated that an intact intestinal mucosa is necessary for Lactobacillus GG adherence to enterocytes (¹) and their benefit in immune modulation of the intestinal mucosa. Friable mucosa could potentially decrease adherence of the Lactobacillus and increase intestinal permeability, thereby potentiating migration of the organism across the intestinal mucosal barrier.

Our cases suggest that patients treated with Lactobacillus GG supplementation for short gut syndrome, particularly when associated with intestinal friability, may be vulnerable to bacteremia or sepsis. These patients should be closely monitored for signs of sepsis, and empiric therapy for sepsis should include coverage of this organism pending specific culture results. Prevention of intestinal inflammation appears to be a potential means to prevent this complication of probiotic treatment. These two cases suggest that Lactobacillus GG may not be as benign a treatment as generally thought. Additional reports clarifying risk factors for the development of bacteremia with this probiotic would be useful. Surveillance endoscopy and the early use of anti-inflammatory therapy may aid in the prevention of sepsis when using probiotics in the treatment of patients with short bowel syndrome.