Article Text


  1. G Aithal,
  2. R Charnley,
  3. R F A Logan,
  4. P Moayyedi,
  5. I D Penman

Statistics from

Screening or surveillance for Barrett’s?

Deciding how to manage our Barrett’s patients is a headache for many gastroenterologists. Barrett’s oesophagus is common, cancer risk is reportedly high and surveillance for dysplasia seems no less feasible than it is in ulcerative colitis. Nevertheless the difficulties are all too obvious.

In the absence of randomised trials decision analyses are perhaps our best guide. Several have tackled Barrett’s screening and surveillance and this one by Inadomi et al is the latest and arguably the most robust. They modelled the cost effectiveness of screening men aged 50 with reflux symptoms. Screening followed by surveillance in patients with Barrett’s and dysplasia would cost $10 400 per quality adjusted life year (QALY) while surveillance for those with Barrett’s but no dysplasia would cost an additional $596 000 per QALY. In addition performing oesophagectomy only when cancer was present cost less and yielded more QALYs than oesophagectomy for high grade dysplasia as well as cancer. Their conclusion is that initial screening to detect Barrett’s and dysplasia is probably cost effective but subsequent surveillance in the absence of dysplasia, even 5 yearly, is not.

A dip into Ascitic fluid!

Spontaneous bacterial peritonitis (SBP) in patients with cirrhosis has a mortality of over 20%. The diagnosis of SBP is based on ascitic fluid neutrophil count of ⩾250/ mm3, but the neutrophil count is not often available in an emergency basis especially “out of hours”. Castellote and colleagues evaluated 228 paracentesis performed in 128 cirrhotic patients to identify 52 episodes of SBP and 5 cases of secondary peritonitis. Ascitic fluid was tested using a reagent strip for leukocyte esterase designed for testing urine. The strip was read at 90 seconds using the colorimetric 5 grade (0 to 4) scale. A scale of 3 or 4 was highly specific (99%) with a positive predictive value of 98% for the diagnosis of bacterial peritonitis. Conversely, a scale of 0 or 1 excluded the diagnosis because of a negative predictive value of 99%. Using inexpensive urine “dipsticks” to detect neutrophils in ascetic fluid, the time for the diagnosis of SBP could be reduced down to as little as 90 seconds!

Normal EUS avoids ERCP

EUS and MRCP have emerged as less invasive alternatives to ERCP for biliary imaging in patients at low or intermediate risk of harbouring CBD stones. While studies have demonstrated the accuracy of EUS, little data exists about the long term outcome of patients who have a normal biliary EUS and thereby avoid ERCP.

Napoleon et al prospectively followed a cohort of 238 such patients for one year to determine the frequency with which this group eventually required ERCP and the results of this. At one year 30 patients (13%) had undergone ERCP because of lingering suspicion of stones. Only 11 of these were found to have stones, 10 within one week of the EUS. Overall, only 4.5% of patients had confirmed ductal calculi and the negative predictive value of EUS in this group was 95.4%. A negative EUS, at least in this centre of excellence, is highly accurate and allows the vast majority of such patients to avoid ERCP (87% in this study). While there seems little to choose between EUS and MRCP, the sensitivity of the former for very small stones is probably greater and, at least in theory, EUS offers the potential for endoscopic stone removal during the same session. The choice of imaging used will, however, likely depend on local expertise and availability, which is little comfort to those practising in many parts of the world where neither is readily available.

No bar to surgery?

Liver resection is a well established treatment for localised colorectal liver metastases. The presence of disease outside the liver has always been considered a contra-indication to liver resection. This paper shows that patients with extra-hepatic recurrence such as lung metastases and local tumour recurrence at the site of the primary operation, if carefully selected for radical surgery can expect a 20% chance of long term survival. The mortality rate from surgery was 4%. This study evaluated patients treated since 1987 and the outlook is likely to be even better now with modern chemotherapy agents.

Dyspepsia management—moving away from endoscopy

Several randomised controlled trials have suggested H pylori test and treat reduces dyspepsia as effectively as early endoscopy at a lower overall cost. There is a paucity of evidence as to whether H pylori test and treat is more cost effective than empirical acid suppression. Allison et al have evaluated these strategies in patients labelled as having peptic ulcer disease (PUD) on long term acid suppression. 650 patients were randomised to H pylori test and treat or continued acid suppression. The prevalence of H pylori was only 38% presumably reflecting the inaccuracy of the peptic ulcer disease label (only 17% had PUD confirmed by endoscopy or radiography). At 12 months the H pylori test and treat group had significantly less dyspepsia in the previous week than the control group (50% v 64%, p=0.01) and fewer were taking antisecretory therapy (58% of the H pylori test and treat compared with 70% of the control group, p=0.001). The mean 12 month health service dyspepsia costs, however, were higher in the H pylori test and treat (mean = $496/person) compared to the control group (mean = $380/person) because of the extra 13C-urea breath tests and eradication therapy prescriptions in the intervention arm. The extra 14% benefit in curing dyspepsia was therefore offset by an extra $116 cost. A larger study is needed in an unselected dyspepsia patient group to give a more precise estimate of costs and benefits to establish the most cost- effective strategy. In the mean time, both antisecretory therapy and H pylori test and treat can be considered appropriate management strategies for dyspepsia patients without alarm symptoms.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.