A retrospective analysis of three trials studying beta-blockade to prevent cardiovascular events in at-risk patients undergoing non-cardiac surgery suggests that the regimen studied did not significantly increase stroke risk.

Peri-operative beta-blockade has been widely used in patients at risk of cardiovascular events who are undergoing non-cardiac surgery, as evidence suggests that it reduces the risk of serious cardiac events. A large controlled trial, however (the POISE trial; Lancet 2008; 371: 1839-47), found that the regimen used in that study, metoprolol m/r 200mg daily, was associated with an increased risk of stroke. This cast doubt on the safety of such treatment. An earlier, smaller study (DECREASE-I) had previously shown benefit from a lower-dose regimen of bisoprolol 10mg daily, started well before surgery and tailored according to heart rate, however DECREASE-I was significantly smaller than POISE (n=1,351 vs. n=8,331). As there have been three studies in the DECREASE series (I, II, and IV) that included beta-blocker treatment, the authors carried out a pooled analysis of the results.

The three studies included 3,884 participants and all used the same beta-blocker regimen; 2,336 (60%) were allocated to beta-blocker treatment. Baseline characteristics for the total pooled population were generally similar to those in the POISE trial, although POISE had rather more participants with diabetes and fewer with ischaemic heart disease. Most DECREASE patients (73%) were undergoing vascular surgery.

Overall incidence of stroke in the pooled population was 0.46% (18/3,884); 12 of these were in patients taking beta-blockers (0.5% of 2,336; odds ratio 1.16; 95% CI, 0.40 to 3.35). All were ischaemic. After adjustment for possible confounders, the only factor significantly related to increased risk of stroke was pre-existing cerebrovascular disease (OR 3.79, 95% CI 1.2 to 11.6). The authors calculated the proportion of the maximum recommended beta-blocker dose (according to US recommendations) for DECREASE and POISE: the dose of bisoprolol used in DECREASE was 10% to 20% of the maximum recommended, whereas POISE used 50% to 100% of the maximum recommended dose of metoprolol.

The authors conclude that in the pooled DECREASE studies, low-dose bisoprolol, started 30 days before non-cardiac surgery and titrated to effect was not associated with an increased risk of stroke in a high-risk population.  They discuss possible mechanisms for the differences between the studies and conclude that the benefits of the DECREASE regimen outweigh the risks.

[Editor's note: the incidence of stroke in DECREASE, overall and in the beta-blocker patients, was similar to the incidence in the placebo group of POISE (0.5%).]