Infection following lower gastrointestinal endoscopy: silent risk or non-event?

Colonoscopy and flexible sigmoidoscopy are widely performed with diagnostic and therapeutic intent in a broad range of gastrointestinal (GI) diseases. The implementation and dissemination of national population-based colorectal cancer screening programs have led to a rapid and exponential increase in the number of colonoscopies and sigmoidoscopies performed, and the volume of procedures carried out annually is expected to rise further.

“Infections are undoubtedly under-reported and important potential complications of lower GI endoscopy.”

Serious pre-, intra-, and post-procedural complications are uncommon in colonoscopy, being reported in 2.8 per 1000 procedures [1]; however, numerous minor, transient symptoms are reported in up to 33 % of patients [2]. Over 85 % of serious colonoscopy complications follow polypectomy, which carries a 7-fold increased risk of bleeding and perforation [3]. In a UK study, Rutter et al. found an overall risk of post-polypectomy bleeding of 0.65 % and a perforation rate of 0.06 % [4]. There is, however, a paucity of reliable comprehensive prospective data reporting the detection, monitoring, and management of all adverse events, other than these very serious complications.

The development of noninvasive diagnostic tests, such as computed tomography virtual colonoscopy and video capsule endoscopy, along with the establishment of national cancer screening programs are leading to an increasing number of straight-to-therapy colonoscopies. More studies focusing on endoscopic complications may be required as a wider range of individuals may be more freely exposed to these risks. Appropriate risk stratification and consent is a fundamental necessity and is especially important where there is minimal clinician contact before the procedure. Two major studies performed in screening populations [4] [5] have not reported any significant adverse events during or after colonoscopy, confirming that lower GI endoscopy is a safe procedure, particularly if performed by skilled endoscopists.

In this issue of Endoscopy, Lin et al. present results of a large nationwide retrospective study investigating the incidence of infections after diagnostic and therapeutic colonoscopy and flexible sigmoidoscopy, a relatively little-studied topic that has been approached mainly in small population studies. These are undoubtedly under-reported and important potential complications of lower GI endoscopy.

Documented cases of infectious complications after GI endoscopy are rare events, with an estimated incidence of 1 in 1.8 million procedures [6]. However, the reported incidence of bacteremia from biopsies and polypectomy varies, from 0 % to 25 % for colonoscopy, and from 0 % to 1 % for flexible sigmoidoscopy [7] [8] [9], and infections are usually transient and asymptomatic in immunocompetent patients. Among the other infectious complications after colonoscopy and sigmoidoscopy are appendicitis, with a reported incidence of 0.04 % [10], peritonitis, and endocarditis [11].

The mechanisms underlying endoscopy-related infections can be divided into exogenous and endogenous (mainly through a damaged intestinal membrane). Exogenous transmission is exceedingly rare, and reported cases are undoubtedly attributable to failures in endoscope reprocessing protocols or to defective equipment [6].

In the retrospective, nationwide case–control study from Taiwan, Lin et al. investigated the incidence of infections 30 days after the procedure in patients undergoing colonoscopy or flexible sigmoidoscopy for GI symptoms, compared with a control group of healthy individuals. A multivariate analysis including statistically selected comorbidities was performed and an infection risk stratification tool was devised.

The data showed that the overall incidence of infections in patients who underwent either a diagnostic or therapeutic procedure was 0.37 %, which was significantly higher than that of the control group (P < 0.001), with a 9.38-fold increased risk. Interestingly, mucosal biopsies and polypectomy were not found to be a significant risk factor for infections; full colonoscopy was the only significant discriminating factor. The incidence rate analysis showed that the most common infections were diverticulitis (37.96 %; P < 0.001), appendicitis (13.38 %; P < 0.001), and peritonitis (25.55 %; P < 0.001). Peritonitis presented in 105/411 patients among whom 10/105 were diagnosed with concomitant perforation or bleeding. The data suggest that only one case of peritonitis was as a result of a perforation but no details of this case were given. However, if a new diagnosis of peritonitis follows a colonoscopy, the usual assumption would be that a clinical or subclinical perforation had occurred during the procedure. The multivariate analysis resulted in colonoscopy and flexible sigmoidoscopy being independent factors for infectious complications, regardless of the presence of cardiovascular, metabolic, liver and pancreaticobiliary disease, renal failure, or autoimmune disease, which is perhaps somewhat surprising.

A few questions are raised by this paper. The only three complications found to be statistically significant (appendicitis, diverticulitis, and peritonitis) could be considered as inflammatory rather than infectious conditions; possible underlying pathogenic mechanisms are either barotrauma from over-insufflation in patients with suboptimal bowel preparation or direct trauma from mucosal damage during instrumentation. However, post-colonoscopy appendicitis has been reported and endoscopic appearances of early appendicitis have also been described [12], suggesting that appendicitis may be an incidental finding rather than a consequence of endoscopy. There are no studies investigating the role of colonoscopy as a triggering factor for the development of diverticulitis, despite diverticulosis being a risk factor for perforation [13].

Infectious conditions potentially related to sedation, such as pneumonia, have not been considered in this study. Post-colonoscopy urinary tract infections, previously attributed to Salmonella spp. and Pseudomonas spp., have also not been considered [6].

With the advent of widespread use of biological agents and immunomodulators in patients with inflammatory bowel disease, it is somewhat surprising that infection is not more common following endoscopic procedures in this group, but the literature remains controversial in this regard. Antibiotic prophylaxis is recommended for patients with severe neutropenia and/or if profoundly immunocompromised [14]. A systematic review published in 2015 suggested that antibiotics should be given in patients with a neutrophil count of less than 500/mm³, regardless of the presence of fever, even for low-risk endoscopic procedures [15].

Colonoscopy remains an invasive procedure with potential risks and complications. The implementation of robust mechanisms to follow all patients and record complication, readmission, morbidity, and mortality data is an imperative. Alongside this, the development of a validated risk stratification tool is to be welcomed. The nomogram proposed in the Lin et al. study presents a model for such a tool, which may also be applicable in other contexts.

• References

• 1 Whitlock EP, Lin JS, Liles E. et al. Screening for colorectal cancer: a targeted, updated systematic review for the U.S. Preventive Services Task Force. Ann Intern Med 2008; 149: 638-658
• 2 Ko CW, Dominitz JA. Complications of colonoscopy: magnitude and management. Gastrointest Endosc Clin N Am 2010; 20: 659-671
• 3 Rabeneck L, Paszat LF, Hilsden RJ. et al. Bleeding and perforation after outpatient colonoscopy and their risk factors in usual clinical practice. Gastroenterology 2008; 135: 1899-1906
• 4 Rutter MD, Nickerson C, Rees CJ. et al. Risk factors for adverse events related to polypectomy in the English Bowel Cancer Screening Programme. Endoscopy 2014; 46: 90-97
• 5 Nelson DB, McQuaid KR, Bond JH. et al. Procedural success and complications of large-scale screening colonoscopy. Gastrointest Endosc 2002; 55: 307-314
• 6 Kimmey MB, Burnett DA, Carr-Locke DL. et al. Transmission of infection by gastrointestinal endoscopy. Gastrointest Endosc 1993; 36: 885-888
• 7 Coughlin GP, Bulter RN, Alp MH. et al. Colonoscopy and bacteraemia. Gut 1977; 18: 678-679
• 8 Kiss A, Ferenci P, Granniger W. et al. Endotoxemia following colonoscopy. Endoscopy 1983; 15: 24-26
• 9 Llach J, Elizalde JI, Bordas JM. et al. Prospective assessment of the risk of bacteremia in cirrhotic patients undergoing lower intestinal endoscopy. Gastrointest Endosc 1999; 49: 214-217
• 10 Chae HS, Jeon SY, Nam WS. et al. Acute appendicitis caused by colonoscopy. Korean J Intern Med 2007; 22: 308-311
• 11 ASGE Standards of Practice Committee. Complications of colonoscopy. Gastrointest Endosc 2011; 74: 745-752
• 12 Petro M, Minocha A. Asymptomatic early acute appendicitis initiated and diagnosed during colonoscopy: a case report. World J Gastroenterol 2005; 11: 5398-5400
• 13 Gatto NM, Frucht H, Sundararajan V. et al. Risk of perforation after colonoscopy and sigmoidoscopy: a population-based study. J Natl Cancer Inst 2003; 95: 230-236
• 14 Allison MC, Sandoe JAT, Tighe R. et al. Antibiotic prophylaxis in gastrointestinal endoscopy. Gut 2009; 58: 869-880
• 15 Tong MC, Tadros M, Vaziri H. Endoscopy in neutropenic and/or thrombocytopenic patients. World J Gastroenterol 2015; 21: 13166-13176