Abstract
Introduction: Among vulnerable individuals, cyclo-oxygenase-2
(COX-2) inhibition increases the risk of cardiovascular thrombotic
events. Promotional safety messages led to ‘channeling’ of selective
COX-2 inhibitors to individuals with gastrointestinal risks.
Aims: We investigated whether this was associated with inappropriate
channeling of COX-2 inhibitors to patients at risk of cardiovascular events.
Methods: Case–control study, August 2003–October 2006. Cases:
patients admitted to hospital with acute coronary syndrome (ACS, myocardial infarction/unstable angina) (n = 814). Controls:
Patients admitted for reasons other than acute vascular ischaemia, heart
failure, renal failure, or upper gastro-intestinal ulceration/bleeding
(n = 1500). Structured interviews gathered information on cardiovascular
events, risk factors, and ingested drugs, including selective COX-2 inhibitors and non-selective non-steroidal anti-inflammatory drugs (NSAIDs).
Controls were frequency matched to cases by age and sex. Relative risks
were estimated from exposure odds ratios (ORs), adjusted for multiple
potential confounders.
Results: Among cases, 386 (47.6%) had experienced prior vascular
events; 425 reported no vascular history. In multivariate modeling, the
strongest association with ACS was a vascular history (adjusted OR 4.47
(95% Confidence Interval 3.33, 6.00). However, cases with a vascular
history were significantly less likely than those without to have used
anti-inflammatory drugs in the preceding week: OR 0.65 (0.47, 0.89).
This effect was seen with non-selective NSAIDs, 0.51 (0.33, 0.79), but
not with selective COX-2 inhibitors (rofecoxib/celecoxib/diclofenac/meloxicam), 0.90 (0.60, 1.35).
Conclusions: Prescribers avoided non-selective NSAIDs but not selective COX-2 inhibitors in some patients. This sidestepped gastrointestinal
risk but caused inappropriate channeling of selective COX-2 inhibitors to
patients with vascular risks.
(COX-2) inhibition increases the risk of cardiovascular thrombotic
events. Promotional safety messages led to ‘channeling’ of selective
COX-2 inhibitors to individuals with gastrointestinal risks.
Aims: We investigated whether this was associated with inappropriate
channeling of COX-2 inhibitors to patients at risk of cardiovascular events.
Methods: Case–control study, August 2003–October 2006. Cases:
patients admitted to hospital with acute coronary syndrome (ACS, myocardial infarction/unstable angina) (n = 814). Controls:
Patients admitted for reasons other than acute vascular ischaemia, heart
failure, renal failure, or upper gastro-intestinal ulceration/bleeding
(n = 1500). Structured interviews gathered information on cardiovascular
events, risk factors, and ingested drugs, including selective COX-2 inhibitors and non-selective non-steroidal anti-inflammatory drugs (NSAIDs).
Controls were frequency matched to cases by age and sex. Relative risks
were estimated from exposure odds ratios (ORs), adjusted for multiple
potential confounders.
Results: Among cases, 386 (47.6%) had experienced prior vascular
events; 425 reported no vascular history. In multivariate modeling, the
strongest association with ACS was a vascular history (adjusted OR 4.47
(95% Confidence Interval 3.33, 6.00). However, cases with a vascular
history were significantly less likely than those without to have used
anti-inflammatory drugs in the preceding week: OR 0.65 (0.47, 0.89).
This effect was seen with non-selective NSAIDs, 0.51 (0.33, 0.79), but
not with selective COX-2 inhibitors (rofecoxib/celecoxib/diclofenac/meloxicam), 0.90 (0.60, 1.35).
Conclusions: Prescribers avoided non-selective NSAIDs but not selective COX-2 inhibitors in some patients. This sidestepped gastrointestinal
risk but caused inappropriate channeling of selective COX-2 inhibitors to
patients with vascular risks.
Original language | English |
---|---|
Article number | TP23 |
Pages (from-to) | 90 |
Journal | Basic and Clinical Pharmacology and Toxicology |
Volume | 105 |
Issue number | s1 |
DOIs | |
Publication status | Published - 2009 |
Externally published | Yes |