S IRS, Perplexity is a consequence of cognitive distance! Calvet and colleagues suggest I am biased, but they may not be immune. They quote the only one of seven epidemiological studies to show a positive association between Helicobacter pylori and ulcer complications in NSAID users.1 Three of the other studies234 have shown no effect and three567 have claimed some evidence of protection by H. pylori. None of these studies is perfect, and a fair reading of the evidence is that there is no clear evidence for a strong positive or negative interaction between H. pylori and NSAID use in patients not taking acid suppression. As one whose first (and only) ballet report stated ‘has a nice smile’, intellectual pirouettes are the only ones I can make! Although studies specifically designed to investigate H. pylori and NSAID dyspepsia have shown that association, the argument that any resulting referral bias is irrelevant is probably a fair one. As regards therapy, the first Hong Kong study8 is extremely important in raising the possibility of genuine therapeutic benefit from H. pylori eradication prior to initiation of NSAID use. However, readers of AP & T will have to decide whether a recommendation to eradicate H. pylori in patients with ulcers, based upon one non-blinded short-term study that involved patients without ulcers and employed a potentially cytoprotective regimen is an evidence-based approach. They can make a similar judgement about eradication recommendations made before there was any evidence at all.9 A longer-term study is needed because of the potentially confounding effects of bismuth: I predict it will show no more than the 50% reduction that one would predict from the strongest epidemiological evidence in favour of a harmful effect of H. pylori. Calvet and colleagues are wrong to say that the second Hong Kong study concerned maintenance therapy after H. pylori eradication10—the study compared eradication therapy vs. maintenance therapy without eradication. It showed eradication therapy was around 10 times less effective than maintenance therapy with omeprazole in preventing recurrent gastric ulcer bleeding. As Calvet and colleagues acknowledge, antisecretory effectiveness diminishes markedly with H. pylori eradication. This resulted in delayed ulcer healing111213 in all published studies, including the one they quote as challenging this conclusion (although the reduction did not reach statistical significance, P=0.12).13 More importantly, as Singh and colleagues have pointed out, ineffective acid suppression does not prevent ulcer complications, which may even occur at an increased frequency, possibly because warning symptoms are masked.14 The assumption that even if H. pylori eradication reduces the effectiveness of acid suppression, the agents are still potent enough to take care of the problem, is based on hope rather than evidence. A ‘let’s keep it simple’ response to discordant information cannot advance thinking. New theories to explain apparent discrepancies can. As a rough (but not perfect) generalization, studies where H. pylori has been a relatively harmful or natural influence have tended to involve patients with no baseline ulcer8, 15, 16 or who had no baseline endoscopy,17 whilst those where it has been beneficial, at least in the presence of acid suppression, have concerned patients with an initial ulcer.10, 12, 181920 Were a management dichotomy to be proved for these two groups, it should not be difficult to convert this into simple prescribing messages. However, with the (appropriate) growing use of Cox-2 inhibitors in preference to non-selective NSAIDs, the relationship between H. pylori and NSAIDs is likely to become irrelevant. Current evidence suggests that the effect of H. pylori in patients using Cox-2 inhibitors is the same as in patients using placebo,21, 22 making it likely that management decisions about eradication can be harmonized. What to do in the growing number of people using low-dose aspirin is less clear.
Search from 250 M+ research papersSearch from 250 M+ research papers
Open Access
Jan 1, 2005
Nana Rokutanda + 7
