Dexamethasone for Cardiac Surgery II trial

Open-heart surgery with cardiopulmonary bypass (CPB) is a unique model of
systemic inflammation in humans, because treatment can begin before the
inflammatory stimulus (surgery, CPB). The inflammatory response associated with
cardiac surgery and CPB likely plays an important role in the development of a
significant number of adverse outcomes. Deregulated inflammation is associated
with haemodynamic instability, respiratory compromise and organ dysfunction
leading to a severe systemic inflammatory response syndrome (SIRS). Patient
recovery is delayed and the risk of postoperative complications and prolonged
hospital stay are greatly increased.

High-dose corticosteroids attenuate the inflammatory response to surgery with
CPB and are commonly used in some countries including The Netherlands, but less
common in others, such as Australia, the US and Canada. Steroids can reliably
attenuate activation of the complement pathways associated with cardiac
surgery, but clinical trials measuring clinically relevant outcomes have had
mixed results.5 The current evidence is dominated by the results of two large
randomised trials: DECS (n=4,494)6 and SIRS (n=7,507). Both DECS and SIRS
assigned patients undergoing cardiac surgery to receive either a
highintraoperative dose of steroids (dexamethasone 1 mg/kg, or
methylprednisolone 500 mg, respectively) or placebo. The point estimates of
both trials suggested a possible reduction in serious complications and
mortality. Planned subgroup analyses in the DECS trial found steroids reduced
the incidence of respiratory failure (3.0 % vs. 4.3%, P=0.02), infection (9.5%
vs. 14.8%, P=0.009), and shortened hospital stay (median 8 [7-13] vs. 9 [7-13]
days, P=0.009). Severe renal failure (need for RRT) was reduced, 0.4% vs. 1.0%,
P=0.04. But SIRS found methylprednisolone was associated with a higher
incidence of myocardial injury (as measured by elevation of CK-MB enzyme).
Neither trial identified a higher risk of myocardial infarction (MI). The
methylprednisolone-induced elevation of CK-MB may therefore not be a class
effect. Another compelling finding in pre-planned subgroup analysis of patient
age groups is that when limiting analysis to those aged less than 75 years in
the DECS trial, dexamethasone reduced the risk of the primary composite
endpoint, RR 0.74 (95% CI: 0.58-0.95), P=0.017; as well as respiratory failure
RR 0.62 (95% CI: 0.42-0.91), P=0.014; and possibly mortality RR 0.53 (95% CI:
0.26-1.10), P=0.08.6 This age-interaction effect is supported by the
demonstration of increased C-reactive protein concentrations in younger
patients enrolled in the DECS trial. Therefore, it is highly plausible that
prophylactic steroids can suppress deregulated inflammation and improve
outcomes in cardiac surgery, but only when used in a less elderly (i.e. <75
years) patient population.

In the proposed study, a novel trial design will be used in order to improve
the efficiency of clinical trials in routine care settings. Such trials must be
robust (internal validity) and pragmatic (external validity) in their design if
they are to have maximal impact. They typically require enrolment of many
thousands of patients, and so are costly and take years to complete, and longer
still for their results to be implemented into routine clinical practice. In
this context investigator-initiated, public good, clinical trials are under
threat. There are growing calls for improved efficiencies in medical research,
and this has led to interest in novel trial designs. There is also renewed
interest in low-cost, large, simple randomized trials, including
registry-nested trials.

We therefore propose to use the *standard practice-preference randomised
consent* (SPP-RC) design. In this design, a modified consent procedure will be
used in participating centres in The Netherlands. For this, patients will first
be asked to participate in the DECS-II cohort, and also to be randomised
between use of standard care (in this case, the intraoperative use of
dexamethasone as part of the cardiac anaesthesia protocol for cardiac surgery)
of use of care that is different from the standard (in this case, not using
dexamethasone intraoperatively as part of the cardiac anaesthesia protocol for
cardiac surgery). Randomisation will be in a 2:1 ratio favouring standard care.
Patients who are randomised tot non-standard care will be approached for
additional consent for this alternative treatment.

The primary objective of this study is to study the effects of intraoperative
dexamethasone administration on postoperative recovery following cardiac
surgery.

The DECS-II study is a pragmatic, assessor-blinded randomised clinical trial of
administering a single dose of dexamethasone,1 mg/kg (standard care in the UMC
Utrecht), versus no dexamethasone (non-standard care in the UMC Utrecht) as
part of anaesthesia care for cardiac surgery.

Not giving dexamethasone as part of routine cardiac anaesthesia care. This
intervention will be compared to standard care in the UMC Utrecht, which is of
administering a single dose of dexamethasone,1 mg/kg.

Dexamethasone has been used during many decades for a large variety of
indications. In the setting of cardiac surgery, it*s routine use to reliably
suppress the perioperative inflammatory response has been common practice
already for decades in many of the Dutch cardiac surgical centres, while
practice is much more variable across countries around the world. However, both
practices (of giving and not giving steroids) are considered safe, with
comparable postoperative outcomes. Since both practices are common and
considerably safe, the risks for patients to participate in this study are
anticipated to be low, and as such acceptable for subjects participating in
this study.