This study was a multicentre, double-blind, randomised, placebo-controlled parallel group study consisting of 4 visits over a period of 6 months. The primary variable was to assess theefficacy of esomeprazole 40 mg orally qd (E40) or esomeprazole 20 mg orally qd (E20) versus placebo orally qd after 6 months of treatment for the prevention of relapse of upper GI symptoms associated with NSAID use, including COX-2 selective NSAIDs, in patients receiving daily NSAID therapy.

Inclusion Criteria:

• Signed informed consent.
• Completed the SH-NEN-0003 study with relief of upper GI symptom. Relief is defined as "last 7 days with a diary assessment of None or Minimal, but allowing up to 2 days to be rated as Mild during the 7-day period".
• A clinical diagnosis of a chronic condition (e.g., osteoarthritis or rheumatoid arthritis) that requires continuous daily NSAID treatment for at least 6 months. Daily NSAID treatment is defined as taking the prescribed dose for at least 5 of 7 days in any given week and can include COX-2 selective NSAIDs, multiple NSAIDs, and high-dose NSAIDs.
• Daily NSAID treatment dose and type:
• Must have been stable for at least 9 weeks prior to inclusion
• Are expected to remain stable for the duration of the study.
• Must be administered orally. If more than one type of NSAID treatment is used, at least one type must be administered orally.
• Hp negative on Histology performed at baseline endoscopy in the study SHNEN-0003.

Exclusion Criteria:

• Discontinuation from study SH-NEN-0003
• Pain, discomfort or burning in the upper abdomen precipitated by exercise or relieved by defecation.
• Pain, discomfort or burning in the upper abdomen not associated with the use of NSAIDs, including COX-2 selective NSAIDs.
• Pregnancy or lactation. Women of childbearing potential must maintain effective contraception during the study period as judged by the investigator.