Patients receiving a NSAID (non-selective non-steroidal anti-inflammatory drug) as well as a PPI (proton pump inhibitor) – to treat inflammation and pain in arthritis – are more than four times more likely to suffer upper or lower gastrointestinal adverse clinical outcomes compared to those receiving a cyclo-oxygenase (COX)-2 selective NSAID.

The findings of the CONDOR study, reported in an Article Online First in the medical journal The Lancet, should encourage review of approaches to reduce risk of NSAID treatment. The study is written by Professor Francis K L Chan, Department of Medicine and Therapeutics, Prince of Wales Hospital, Hong Kong Special Administrative Region, China, and Professor Jay Goldstein, University of Illinois at Chicago, IL, USA, and colleagues. The study is being presented at the European League Against Rheumatism (EULAR) Annual Conference in Rome.

Previous studies have demonstrated that COX-2-selective NSAIDs and non-selective NSAIDs plus a proton-pump inhibitor (PPI) have comparable side effects in the upper gastrointestinal tract, but adverse effects in the entire gastrointestinal tract (including the lower part) might be less common with selective drugs than with non-selective drugs. This is because acid suppression does not avert damage to the lower gastrointestinal tract. In this study, the scientists aimed to compare risk of gastrointestinal events associated with celecoxib (COX-2 selective NSAID) versus diclofenac slow release (a non-selective NSAID) plus omeprazole (a PPI).

This randomised trial recruited patients with osteoarthritis or rheumatoid arthritis at increased gastrointestinal risk, that is, aged 60 years or over, or 18 years and older with previous gastrointestinal ulceration.

The study involved 196 centres in 32 nations or territories. Patients tested negative for Helicobacter pylori (a risk factor for peptic ulcer). Patients were assigned in a 1:1 ratio to receive celecoxib 200 mg twice a day or diclofenac slow release 75 mg twice a day plus omeprazole 20 mg once a day. Patients and researchers were masked to treatment allocation. The primary endpoint was a composite of clinically significant upper or lower gastrointestinal events adjudicated by an independent committee. Such events included several potential outcomes relevant to clinical practice, ranging from discontinuation of treatment due to presumed significant occult blood loss to admission to hospital for life-threatening complications.

4484 participants were randomly allocated to treatment (2238 celecoxib; 2246 diclofenac plus omeprazole) and were included in intention-to-treat analyses. 20 (0•9%) patients receiving celecoxib and 81 (3•8%) receiving diclofenac plus omeprazole met criteria for the primary endpoint, meaning those receiving diclofenac plus omeprazole were more than four times more likely to suffer upper or lower gastrointestinal adverse clinical outcomes than those in the celecoxib group. 114 (6%) patients taking celecoxib versus 167 (8%) taking diclofenac plus omeprazole withdrew early because of gastrointestinal adverse events.

The authors conclude:

Since guidelines recommend that selection of NSAID therapy be driven by consideration of both cardiovascular and gastrointestinal effects of treatment, CONDOR has provided new data relevant to patients requiring anti-inflammatory therapy who are at increased gastrointestinal but not increased cardiovascular risk. In this population, the gastrointestinal outcomes of a COX-2 selective NSAID were quite different to those of a non-selective NSAID plus a PPI. Further understanding of the cardiovascular outcomes of these two strategies requires the results of ongoing trials that have been designed directly to address that important clinical question. The findings of the CONDOR trial should encourage guideline committees to review their treatment recommendations for arthritis patients.

Dr Elham Rahme and Dr Sasha Bernatsky, McGill University, Montreal, QC, Canada, in an accompanying Comment wrote:

Chan and colleagues suggest that risk of lower gastrointestinal bleeding could be as important a consideration as is risk of upper gastrointestinal bleeding. However, decreased risk of gastrointestinal adverse events was driven mainly by haemoglobin decrease, not by documented lower gastrointestinal bleeds. Thus, although the researchers close by suggesting revision of existing guidelines (which currently recommend selection of anti-inflammatory therapy on the basis of the patient’s upper gastrointestinal and cardiovascular risks), this advice might be premature.

Celecoxib versus omeprazole and diclofenac in patients with osteoarthritis and rheumatoid arthritis (CONDOR): a randomised trial
Francis K L Chan, Angel Lanas, James Scheiman, Manuela F Berger, Ha Nguyen, Jay L Goldstein
The Lancet, Early Online Publication, 17 June 2010
doi:10.1016/S0140-6736(10)60673-3

Accompanying Comment
NSAIDs and risk of lower gastrointestinal bleeding
Elham Rahme, Sasha Bernatsky
The Lancet, Early Online Publication, 17 June 2010
doi:10.1016/S0140-6736(10)60839-2

Edited by: Christian Nordqvist