AIM: To prospectively examine the association between presence of Streptococcus bovis (S. bovis) in colonic suction fluid and the endoscopic findings on colonoscopy.

METHODS: From May 2012 to March 2013, 203 consecutive patients who underwent colonoscopy for any reason were enrolled in the study. Exclusion criteria included: antibiotic use in the previous month, age younger than 18 years, and inadequate preparation for colonoscopy. The colonoscopy was performed for the total length of the colon or to the occluding tumor. The endoscopic findings were registered. Samples were obtained proximal to the colonoscopic part of the suction tube from each patient and sent to the clinical microbiology laboratory for isolation and identification of S. bovis. Samples were incubated in enrichment media with addition of antibiotic disks for inhibition of growth of Gram-negative rods. The samples were seeded on differential growth media; suspected positive colonies were isolated and identified with Gram staining, catalase, and pyrrolidonyl arylamidase tests, and further identified using a VITEK2 system. Statistical analyses were performed using the Student’s t and χ² tests.

RESULTS: Of the 203 patients recruited, 49 (24%) patients were found to be S. bovis carriers; of them, the endoscopic findings included: 17 (34.7%) cases with malignant tumors, 11 (22.4%) with large polyps, 5 (10.2%) with medium-sized polyps, 6 (12.2%) with small polyps, 4 (8.1%) with colitis, and 6 (12.2%) normal colonoscopies. Of 154 patients found negative for S. bovis, the endoscopic findings included: none with malignant tumors, 9 (5.8%) cases with large polyps, 11 (7.1%) with medium-sized polyps, 26 (16.9%) with small polyps, 7 (4.5%) with colitis, and 101 (65.6%) normal colonoscopies. S. bovis (Gram-positive coccus) is considered part of the normal intestinal flora. There is an association between S. bovis bacteremia and colonic neoplasia. It is not well understood whether the bacterium has a pathogenetic role in the development of neoplasia or constitutes an epiphenomenon of colorectal neoplasms. There was a clear relationship between positivity for S. bovis in colonic suction fluid and findings of malignant tumors and large polyps in the colon.

CONCLUSION: There is an association between S. bovis bacteremia and malignant colonic lesions; this should prompt for development of a reliable screening method for advanced colonic lesions.

Streptococcus bovis (S. bovis) is part of the normal flora of the human alimentary tract in 2.5%-15% of individuals[1]. The association between S. bovis bacteremia and colonic neoplasia is well described in the literature[1-4], and all patients with S. bovis bacteremia are advised to undergo investigation to rule out gastrointestinal tract neoplasia[5,6].

It has not yet been determined whether the relation between the bacterium and gastrointestinal neoplasia is etiologic or incidental. Data support the etiologic nature of this relation because of the proinflammatory potential and procarcinogenic properties of S. bovis, including the leukocyte recruitment driven by the bacterium, the tumor tissue-selective adhesion potential of S. bovis, selective colonization of S. bovis in tumor cells, tumor tissue microenvironment suitability for S. bovis proliferation, the local disruption of tumor tissues and capillaries that allow the entry of S. bovis into blood circulation, and the S. bovis-induced cytokines and transcriptional factors[6-9]. Some researchers have postulated that the increased load of S. bovis in the colon might be responsible for its association with colon cancer[1].

Studies that have examined the issue of S. bovis presence in stool and the risk of colorectal cancer have produced conflicting results. A correlation between S. bovis presence in stool and colorectal neoplasia was found in some of the studies[2,4,5], whereas that association was not found in others[6-13].

Our primary goal is a further examination of the nature of that association. We prospectively investigated the association of S. bovis presence in colonic content and the nature of endoscopic findings in colonoscopy of 203 patients.

Two hundred three consecutive patients with adequate preparation for the colonoscopy procedure were recruited and underwent whole colonoscopy procedures in the Gastroenterology Department (except in cases of occluding tumor). Forty-nine patients (24%) were found to be S. bovis carriers.

Among the study population, 17 malignant tumors were diagnosed. In all these cases S. bovis was identified in suction fluid. Among 49 patients who were found positive for S. bovis, the endoscopic findings included: 17 (34.7%) cases with malignant tumors, 11 (22.5%) with large polyps, 5 (10.2%) with medium-sized polyps, 6 (12.2%) with small polyps, 4 (8.2%) with colitis, and 6 (12.2%) with normal colonoscopies.

Among 154 patients who were found negative for S. bovis, the endoscopic findings included: none with malignant tumors, 8 (5.2%) cases with large polyps, 11 (7.1%) with medium-sized polyps, 26 (16.9%) with small polyps, 7 (4.6%) with colitis, and 102 (66.2%) normal colonoscopies (Table 1). In all cases, histopathologic examination was performed with confirmation of malignant or benign nature of every lesion.

A significant difference was observed between S. bovis carriers (patients positive for S. bovis) and patients without S. bovis (negative) among those who were diagnosed by colonoscopy with malignant tumors, large polyps, or normal-nature colonoscopy. There was no difference between S. bovis-positive and -negative patients among those who were diagnosed with small- and medium-sized polyps and with colitis/ileitis.

The mean age of S. bovis carriers was significantly higher (10 years older) than the non-carriers group (P < 0.05), which could thereby predict potential appearance of advanced lesions in later stages of life.

Colorectal cancer is one of the most common cancers (fourth among men and third among women)[14-17]. There is an association between S. bovis bacteremia and colonic neoplasia. It is not well understood whether the bacterium has a pathogenic role in the development of neoplasia or constitutes an epiphenomenon of colorectal neoplasms. Possible mechanisms that may explain this association include S. bovis overgrowth, breakdown of mucosal integrity, and subsequent bacterial translocation[18-21].

Performance of a routine colonoscopy examination in high-risk populations could result in detection, early treatment of premalignant lesions, and subsequent prevention of colorectal cancer. Early detection of a malignant lesion can be crucial in the success of treatment.

It is important to state that none of the available noninvasive tests are efficient enough tools for screening patients in high-risk populations, or to eventually predict the patients that must undergo colonoscopy examination. In this prospective study, we found a clear association between the presence of S. bovis in colonic suction fluid and findings of malignant tumor and large polyps in the colon. These findings confirm the previous data that correlates the presence of S. bovis with colorectal cancer[22-24]. Findings shown in these studies demonstrate an association of stool positivity with colorectal cancer, indicating that bowel suction fluid is representative of stool positivity for S. bovis. Stool samples in different forms can be used as screening material for detection of the population at risk for advanced colorectal lesion. This can become a preferred surveillance tool instead of repeated colonoscopy performance in patients who have undergone treatment of advanced colonic lesions by endoscopy or surgery[25,26].

Moreover, patients who had S. bovis-positive colonic content and whose colonoscopy examination showed a normal nature should be alerted and carefully observed in the future because of the possible pathogenic role of S. bovis in colorectal cancer. For this reason, culture of the suction fluid could shed light and predict the length of time required until the next colonoscopy. At present, there is no good noninvasive screening method for advanced colonic lesion detection.

In conclusion, we have found an association of S. bovis presence in colonic content and formation of malignant colonic tumors or large polyps, which could be detected in colonoscopy. We suggest considering our finding of this association in order to develop a reliable screening method for detection of advanced colonic lesions. In light of significant developments in molecular biology and greater sensitivity of these methods, it is a good idea to utilize this platform for identification of S. bovis[27-30].