Skip navigation

Harm from endoscopy or colonoscopy


Clinical bottom line

Endoscopy is associated with morbidity and mortality, but is a crucial method in the diagnosis of gastrointestinal disease. Upper gastrointestinal mortality occurs in between 1 in 2000 and 1 in 12,000 patients, and some morbidity in 1 in 200.

Adequate training of endoscopists is essential in maintaining and improving safety standards.

Frequent and severe symptoms of reflux over a long period are associated with very much higher risks of gastric ulceration and oesophageal cancer cancer. Endoscopy may be conducted for diagnostic purposes to detect the aetiology of a condition or for the purposes of treatment. It is a major diagnostic tool in the detection of gastric cancers.



Froehlich F et al. Appropriateness of gastrointestinal endoscopy: Risk of complications. Endoscopy 1999 31: 684-686.


Quine MA et al. Prospective audit of upper gastrointestinal endoscopy in two regions of England: safety, staffing, and sedation methods. Gut 1999; 36: 462-467.

US surgical training programme

Schauer PR et al. Complications of surgical endoscopy. A decade of experience from a surgical residency training program. Surgical Endoscsopy 1997; 11:8-11.


Reading these reports together raises some interesting points.

The review claimed to have a systematic search, but gave no inclusion and exclusion criteria. It did not include the UK survey data (Quine et al, 1999). It provided information mainly available from retrospective case reports and reviews. Endoscopists in the studies were described as being atypical of those in practice because of their extensive expertise. Morbidity included perforation, bleeding, cardiorespiratory events mainly associated with sedation/premedication, need for surgery, and other complications. Summarised results are described below, with similar complication rates for diagnostic upper gastrointestinal endoscopy and for diagnostic colonoscopy. Complications occurring up to 30 days following the procedure were recorded. The overall risks, given as percentages and odds, are in Table 1.

Upper gastro-intestinal diagnostic endoscopy:

Information was available on nearly 577,000 patients. Between 0.1-0.2% of complications considered directly related to the procedure resulted in morbidity and 0-0.7% were fatal.

Diagnostic colonoscopy

In about 100,000 patients mortality was about 0.02% and morbidity was 0.25%.

Therapeutic colonoscopy:

This resulted in complications in 1.0-7.2% of 32,000 patients and about 0.04% of complications were fatal.

Table 1: Risk of harm from endoscopy and colonoscopy


Number of patients surveyed

Mortality (%)

Mortality odds

Morbidity (%)

Morbidity odds

Diagnostic upper gastrointestinal endoscopy



1 in 12,000


1 in 230

Diagnostic colonoscopy



1 in 5000


1 in 440

Therapeutic colonoscopy



1 in 3,500


1 in 50

The audit assessed services and complications arising from upper gastrointestinal endoscopy in 36 hospitals, with 383 doctors, including roughly 150 consultants, in the UK. Information from over 14,000 procedures was provided, of which 92% were diagnostic and 8% therapeutic. Most (85%) patients received intravenous medazolam or diazepam for sedation. Problems identified included endoscopies being conducted by inexperienced personnel with insufficient supervision, a lack of basic facilities, and poor recovery areas. The most common complications were cardiorespiratory in nature.

Seven patients (0.05%) died as a direct result of a diagnostic gastroscopy, which the authors rated as a risk of 1 death per 2,000 procedures. A further 36 complications were considered related to the performance of the procedure. The risk of morbidity was reported as 1 in 200. These figures are roughly in line with those from the review, though death is more frequent.

The US surgical training programme provided information from 9,200 upper and lower gastrointestinal endoscopies. The complication rate was 1.4% for upper and 0.42% for lower GI procedures, with overall mortality rates of 0.76% and 0.6% respectively. No fatalities were regarded as a direct consequence of a procedure related complication. The conclusions drawn were that adequate training and supervision indicates endoscopy as a safe procedure.


There is a risk of morbidity and mortality associated with endoscopy. Results were similar in the studies. Dyspeptic symptoms are common [1] with the long-term consequence of oesophageal or gastric cancer in some patients [2]. It cannot be disputed that endoscopy is needed for the detection and treatment of gastro-intestinal disease, though adequate training and supervision of endoscopists is crucial to improve its safety.


  1. RC Heading. Prevalence of upper gastrointestinal symptoms in the general population: a systematic review. Scandinavian Journal of Gastroenterology 1999 34 Suppl 231: 3-8.
  2. J Lagergren et al. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. New England Journal of Medicine 1999 340: 825-31.