Adenoma Detection Rate: Life or Death Quality Measure

Many physicians bristle at numerous quality measures due to concerns that they may have little relevance to clinical practice or that they are not representative since they measure only a tiny fraction of their work.  For adult gastroenterologists, adenoma detection rate may become a quality measure that will be more difficult to dismiss (NEJM 2014; 370: 1298-306).

Background: In this study, the authors evaluated 314,872 colonoscopies by 136 gastroenterologists by using data from an integrated health care delivery organization.  They determined associations between adenoma detection rate and the risks of colorectal cancer/cancer-related deaths diagnosed 6 months to 10 years after colonoscopy.  Estimates of attributable risk were adjusted for the demographics of the patients, indications for colonoscopy, and coexisting conditions.

Key Results:

  • The adenoma detection rate varied considerably, from 7.4% to 52.5%.
  • 712 colorectal adenocarcinomas were identified during followup, including 255 advanced-stage cancers.
  • 147 deaths from interval colorectal cancer occurred.
  • Among patients of physicians with adenoma detection rates in the highest quintile, compared with patients of physicians in the lowest quintile, the adjusted hazard ratio for any interval cancer was 0.52; it was 0.43 for advanced-stage interval cancer and 0.38 for fatal interval cancer.
  • For each 1% increase in the adenoma detection rate, there was an associated decrease in the risk of cancer by 3%.

Bottomline: Previous studies have shown that spending longer than 6 minutes on a screening colonoscopy increases detection rates.  This study shows that doing a high quality colonoscopy really does alter the outcome.

Related blog posts:


Polyps: Clinical Decision Tool

The best approach to polyps from the U.S. Multi-Society Task Force: Gastroenterology 2014; 146: 305-306.  This paper’s simple chart on page 306 could help reduce many follow-up colonoscopies that are performed at shorter than recommended intervals.

Polyp Guideline

Related blog post:  Consensus guidelines after polypectomy | gutsandgrowth)

More bad news for smokers

Add two more cancer risks for tobacco smoke (Gastroenterology 2012: 142: 233-40, 242-47).  There is now evidence linking tobacco smoke to 18 different cancers and tobacco smoke is probably the most preventable cause of death in the world.

In the first study, the investigators examined 3167 patients with Barrett’s esophagus.  This retrospective study followed patients for 7.5 years.  Patients who were current smokers (any form of tobacco) had double the risk of developing high-grade dysplasia or cancer compared to those who had never smoked.  Former cigarette smokers had a hazard ratio of 1.53.

In the second study, 386 patients with Lynch syndrome were analyzed during a 10 month period.  The hazard ratio for developing colorectal adenomas was 6.13 for current smokers and 3.03 for former smokers compared with patients who never smoked.  In addition, the authors identified a trend for developing adenomas based on pack-years.

Two more reasons to quit smoking.  On a side note, my grandmother said quitting smoking was the easiest thing that she ever did.  So easy, she did it a thousand times.

Additional references:

  • -Gastroenterolgy 2005; 129: 1825-31.  1.6% incidence of BE in adult Swedish population. Alcohol & smoking increase risk.
  • -NEJM 2011; 365: 1222. Treating smokers -useful review.
  • -NEJM 2011; 365: 1193. Cytisine -inexpensive- helps with smoking cessation (8.4% success vs 2.4%in placebo)
  • -NEJM 2008 358; 2249. Smoking and role of social networks.
  • -Gastroenterology 2011; 141: 2000. Lower risk of Barrett’s in pts taking NSAIDs & statins. n=570.
  • -Gastroenterology 2011; 141: 1179. Lower risk of Barrett’s in pts with low-grade dysplasia than previously noted -similar to non-dysplastic Barrett’s.
  • -NEJM 2011; 365: 1375. Large Danish study, n=11028. Lower incidence of Barrett’s than previous estimates. Relative risk of 11.3 compared to general population for adenoca of Esophagus with absolute annual risk of 0.12%. Barrett’s patients have the same life expectancy as general population (ed. pg 1437). Detecting cancer only ~1 in 1460 scopes with screening whereas Barrett’s detected in 10% of pts.
  • -Gastroenterology 2011; 140: 1084. AGA statement on Barrett’s . Recs screening only in those with multiple risk factors (age 50, male, chronic GERD, white, incr BMI)
  • -NEJM 2005; 352: 1851. Cases of Lynch can be missed when following screening guidelines.
  • -Gastroenterology 2010; 138: 207-2177 (entire issue) Colon cancer, Lynch syndrome
  • -Gastroenterology 2008; 135: 380.  Review of colon cancer screening and prevention -2008 up-to-date- literature review
  • -Gastroenterology 1967; 53: 517-27.  Seminal article.  Lynch HT showed gene-related cancer in family cancer syndrome -different than polyposis syndromes.