my personal edition > rheumatoid arthritis > news


  E-Mail this DGDispatch to a colleague

DGDispatch

Proton-Pump Inhibitors Decrease Risk of NSAID-Induced Gastroduodenal Ulcers: Presented at EULAR

By Chris Berrie

AMSTERDAM, THE NETHERLANDS -- June 30, 2006 -- Elderly individuals with cardiovascular comorbidities are at high risk of serious gastroduodenal ulcers if they take nonsteroidal anti-inflammatory drugs (NSAID), according to results of a nested case-control study presented here at the Annual European Congress of Rheumatology (EULAR).

In addition, the study shows that use of cyclooxygenase-2 (COX-2) inhibitors does not reduce the risk of serious gastroduodenal ulcers, but that concomitant use of proton-pump inhibitors (PPIs) reduces this risk by 66%.

Harald Vonkeman, MD, rheumatologist, department of rheumatology and clinical immunology, Medisch Spectrum Twente Hospital and University of Twente, Twente, The Netherlands, presented the results on June 23^rd.

NSAID treatment is often complicated by gastrointestinal problems such as peptic ulcer bleeding, so a number of recommendations have been put forward for primary prevention of these adverse effects, including Helicobacter pylori eradication, use of concomitant prostaglandin E1 analogues, selective COX-2 inhibitors, concomitant double doses of H₂-receptor antagonists (H₂RAs), and concomitant use of PPIs.

Dr. Vonkeman and colleagues conducted a study to determine which patients are especially at risk for NSAID-attributable serious gastroduodenal ulcers, and to compare the effectiveness of different preventive strategies in these patients.

The researchers took advantage of the single large teaching hospital that serves a population of 153,000 in Enschede, The Netherlands. The study cohort was comprised of all NSAID users in this city, as revealed through the prescription databases of the local pharmacies. Cases were all consecutive patients hospitalized with serious gastroduodenal ulcers, and controls were randomly selected NSAID users from the remaining cohort that were matched to the cases for age, sex, and index date.

A 26-month observation period was specified (November 2001 to December 2003), during which time 104 cases (mean age, 70.4 years; female, 55.8%) of serious NSAID ulcer complications were recorded. This gave an annual incidence of 31.4 cases per 100,000 population, and of 168.1 cases per 100,000 NSAID users. Annual in-hospital mortality in these cases was 17.8 deaths per 100,000 NSAID users.

The clinical baseline characteristics of the cases for ulcer location were as follows: gastric, 51%; duodenal, 32.7%; both gastric and duodenal, 10.6%; unknown, 5.7%; perforation, 13.5%. For H. pylori, 20.2% were positive, 43.3% negative, and the remaining 36.5% were not tested.

A comparison of medical histories in cases versus matched controls (n = 284) showed that cases had significantly greater levels of heart failure (25.0% vs 1.3%; odds ratio [OR], 2.63; 95% confidence interval [CI], 1.48-4.67; P = .001), myocardial infarction (19.2% vs 11.3%; OR, 1.88; 95% CI, 1.02-3.45; P = .04) and stroke (17.3% vs 9.9%; OR, 1.91; 95% CI, 1.01-3.63; P = .04).

Comparison of nonselective NSAID use was similar for indomethacin (2.9% vs 1.4%), diclofenac (42.3% vs 38.0%), diclofenac-misoprostol (7.7% vs 6.7%), and other NSAIDs (2.9% vs 2.8%). However, the cases showed a trend towards greater use of naproxen (9.6% vs 4.9%; P = .09) and less use of ibuprofen (15.4% vs 24.3%; P = .06).

There were no significant differences in the use of the selective NSAIDs rofecoxib (15.4% vs 14.8%), celecoxib (1.0% vs 2.8%), and meloxicam (1.0% vs 4.2%), or in the gastroprotective H₂RAs (3.8% vs 3.2%) and misoprostol (7.7% vs 7.0%). However, there were significantly fewer cases than controls taking PPIs (13.5% vs 27.1%; P = .005).

There was no difference in use of corticosteroids (13.5% vs 11.3%), selective serotonin re-uptake inhibitors (5.8% vs 3.2%), clopidogrel/dipyridamole (4.8% vs 3.2%), and low-dose aspirin (30.8% vs 24.3%), while trends were seen for more cases than controls using high-dose NSAIDs (10.6% vs 6.0%; P = .12) and more controls than cases who were on more than 1 NSAID (11.5% vs 19.0%; P= .08). Significance was reached, however, for a greater use of coumarin by cases (13.5% vs 6.7%; P = .04).

When these factors were put through a multivariate analysis, the researchers found that significant predictors for gastroduodenal ulcers among cases were the following: use of low molecular weight heparin (adjusted OR [aOR], 17.33; 95% CI, 3.71-80.95; P < .001) and acetaminophen (aOR, 2.80; 95% CI, 1.64-4.79; P < .001), and heart failure (aOR, 2.44; 95% CI, 1.28-4.66; P = .007). While coumarin also showed a negative trend for gastroduodenal ulcers (aOR, 2.09; 95% CI, 0.93-4.70; P = .075), the only protective factor revealed through this analysis was the use of PPIs (aOR, 0.33; 95% CI, 0.17-0.67; P = .002).

The researchers initially concluded that the main at-risk NSAID-use population is elderly patients with cardiovascular comorbidities, where there is a high in-hospital mortality seen (10.6%). Conversely, there was a 66% risk reduction for serious gastroduodenal ulcers in cases using PPIs, while the use of selective COX-2 inhibitors provided no protection.


[Presentation title: Proton-Pump Inhibitors But Not Selective Cyclooxygenase-2 Inhibitors Decrease the Risk of Serious Gastroduodenal Ulcers Attributable to Non-Steroidal Anti-Inflammatory Drugs. Abstract OP0139]

E-Mail this DGDispatch to a colleague

To print, use this version