6th March 2009
A recent observational study has found that people with heart failure who took NSAIDs, particularly at higher doses, were at substantially greater risk of death or being admitted to hospital for treatment of heart failure or MI, than non-users. Diclofenac was associated with a particularly high risk.
Prescribers should consider the cardiovascular (CV) risk of NSAIDs and use them only when essential in people with heart failure. Patients with heart failure who are prescribed NSAIDs, particularly diclofenac or coxibs, should have their NSAID medication reviewed. If an NSAID is essential, the first choice drug is ibuprofen at 1200mg/day or less. Naproxen is probably next choice, although the risks with this drug are greater. Co-prescription with a PPI to manage gastrointestinal side effects and guard against ulceration may also be desirable, in line with NICE guidance on osteoarthritis.
We reviewed the CV safety of NSAIDs in November 2007, in MeReC Extra 30. Diclofenac (especially 150mg/day) is associated with a small increased thrombotic risk similar to that of etoricoxib and possibly other COX-2 selective inhibitors (coxibs). Many other traditional NSAIDs may also be associated with a small increased risk of thrombotic events. High-dose ibuprofen (e.g. 2400mg/day) appears to be associated with a small increased risk, whereas low-dose ibuprofen (<1200mg/day) and naproxen (1000mg/day) are not. Taking into account the high prescribing volume of diclofenac and the increased thrombotic risk, the prescribing of diclofenac could be associated with up to 2000 extra or premature CV events in England each year, compared with no treatment. The absolute risk of CV events for an individual person will depend on their baseline risk.
An earlier observational study had found that current users of NSAIDs had a 30% greater chance of being admitted to hospital for heart failure than non-users, and that the risk was substantially greater in people with a diagnosis of heart failure but who had not previously been admitted to hospital for heart failure.
What does this study claim?
This study looked at the risk of death, re-admission for heart failure or admission because of myocardial infarction (MI) associated with NSAID use among 107,092 Danish patients who survived their first admission for heart failure between 1995 and 2004. Diclofenac use was associated with a doubling in risk of death (hazard ratio [HR] 2.08, 95% confidence interval [CI] 1.95 to 2.21, P<0.001) and this risk was greatest at doses more than 100mg/day (HR 5.54, 95%CI 5.08 to 6.03, P<0.001). Overall, diclofenac was associated with an absolute risk increase (ARI) of 9% (95%CI 7.4% to 10.6%), with a number needed to harm (NNH) of 11 (95%CI 9 to 13) per year. In other words, for every 11 people discharged from hospital after their first admission for heart failure, who take diclofenac for a year, one extra person will die who would not have done otherwise. Diclofenac was also associated with increased risk of re-admission for heart failure (HR 1.35, 95%CI 1.24 to 1.48, P<0.001) and, at doses greater than 100mg/day, admission for MI (HR 2.43, 95%CI 1.74 to 3.40, P<0.001).
Celecoxib was associated with a 75% increased risk of death (HR 1.75, 95%CI 1.63 to 1.88, P<0.001). This was an ARI of 7% (95%CI 5.3% to 8.8%) with a 1 year NNH of 14 (95%CI 11 to 19). Its use increased the risk of re-admission for heart failure by 24% (HR 1.24, 95%CI 1.12 to 1.39, P<0.001) and admission for MI by 38% (HR 1.38, 95%CI 1.13 to 1.69, P=0.001). Again, risk was greater at higher doses.
At doses of 1200mg/day or less, ibuprofen did not statistically significantly increase the risk of death (P=0.65) but did at higher doses (HR 2.83, 95%CI 2.64 to 3.02, P<0.001). At all doses ibuprofen increased the risk of re-admission for heart failure (overall HR 1.16, 95%CI 1.10 to 1.23, P<0.001) and admission for MI (overall HR 1.33, 95%CI 1.19 to 1.50, P<0.001).
Naproxen was not associated with a statistically significant increased risk of death when used at doses of 500mg/day or less (P=0.15), but did increase the risk at higher doses (HR 1.97, 95%CI 1.64 to 2.36, P<0.001). Its use at 500mg/day or less, or at higher doses, was not associated with a statistically significant increased risk of re-admission for heart failure (P=0.10 and 0.27 respectively). Perhaps surprisingly, daily doses of 500mg or less were associated with an increased risk of admission for MI (HR 1.47, 95%CI 1.02 to 2.10, P=0.04) but at higher doses the observed increased risk was not statistically significant (HR 1.62, 95%CI 0.97 to 2.72, P=0.07).
These findings are in keeping with what we already know. Coxibs (e.g. celecoxib) and diclofenac increase the risk of CV events. In most patients, heart failure arises from ischaemic damage to their myocardia, and they are likely to be at greater baseline risk of thrombotic events. This higher baseline risk means that the absolute risk of thrombotic events is greater than in other people.
In addition, NSAIDs have been shown to increase the risk of worsening heart failure, probably owing to their effects on fluid balance. In this group of people with a history of heart failure sufficiently severe to require admission, most NSAIDs increased the risk of the heart failure worsening such that patients needed re-admission.
Use of any drug involves a trade-off between risks and benefits and this is especially true of NSAID use in people with heart failure. If an NSAID is essential, the first choice drug is probably ibuprofen at 1200mg/day or less. If this is not sufficiently effective, naproxen is probably next choice, although the risks with this drug are greater. Co-prescription with a PPI to manage gastrointestinal side effects and guard against ulceration may also be desirable, in line with NICE guidance on osteoarthritis.
Patients: 107,092 patients who survived their first hospitalisation because of heart failure between January 1,1995, and December 31, 2004.
Intervention and comparison: Subsequent use of NSAIDs from individual-level linkage of nationwide registries of hospitalisation and drug dispensing by pharmacies in Denmark. A total of 36,354 patients (33.9%) claimed at least one prescription of an NSAID after discharge.
Outcome: (more results are included in the full paper)
- Risk of death: Diclofenac HR 2.08, 95%CI 1.95 to 2.21, P<0.001). HR at 100mg/day or less 1.31, 95%CI 1.20 to 1.42, P<0.001. HR at more than 100mg/day 5.54, 95%CI 5.08 to 6.03, P<0.001). Celecoxib HR 1.75, 95%CI 1.63 to 1.88, P<0.001. Ibuprofen 1200mg/day or less HR 0.99, 95%CI 0.94 to 1.04, P=0.65. HR at higher doses 2.83, 95%CI 2.64 to 3.02, P<0.001. Naproxen at 500mg/day or less HR 0.88, 95%CI 0.73 to 1.05, P=0.15. HR at higher doses 1.97, 95%CI 1.64 to 2.36, P<0.001.
- Re-admission for heart failure: Diclofenac HR 1.35, 95%CI 1.24 to 1.48, P<0.001; Celecoxib HR 1.24, 95%CI 1.12 to 1.39, P<0.001). Ibuprofen HR 1.16, 95%CI 1.10 to 1.23, P<0.001, Naproxen 500mg/day or less HR 1.18, 95%CI 0.97 to 1.44, P=0.10, higher doses HR 1.18, 95%CI 0.88 to 1.57, P=0.27.
- Admission for MI: Diclofenac at 100mg/day or less, 1.14, 95%CI 0.91 to 1.43, P=0.26. HR at more than100mg/day 2.43, 95%CI 1.74 to 3.40, P<0.001. Celecoxib HR 1.38, 95%CI 1.13 to 1.69, P=0.001). Ibuprofen HR 1.33, 95%CI 1.19 to 1.50, P<0.001). Naproxen at 500mg or less HR 1.47, 95%CI 1.02 to 2.10, P=0.04, HR at higher doses 1.62, 95%CI 0.97 to 2.72, P=0.07.
Sponsorship: No financial disclosure reported.