Higher BP, Oxygen Targets No Help in Comatose OHCA Patients: BOX

For patients who remain comatose after having an out-of-hospital cardiac arrest (OHCA), neither more-liberal use of oxygenation nor a higher blood-pressure target appear to affect outcomes, researchers reported today at the European Society of Cardiology Congress 2022.

The dual findings from the BOX trial, which had a 2x2 factorial design, are the latest to show discouraging results for strategies aimed at reducing mortality in this challenging condition. Just last year, the long-awaited TTM2 trial upended decades of practice by showing no benefit to therapeutic hypothermia.

Jesper Kjaergaard, MD, DMSc (Copenhagen University Hospital Rigshospitalet, Denmark), lead author on a New England Journal of Medicine paper summarizing the BOX blood-pressure analysis, stressed the need to continue searching for ways to improve outcomes in OHCA. “In Europe alone, approximately 100,000 patients are resuscitated from out-of-hospital cardiac arrest each year. And when they’re admitted to hospital in a comatose state, less than 50% of them will survive. So it’s a dangerous disease with dismal outcomes,” he said in a press briefing.

Jacob Eifer Møller, MD, DMSc (Copenhagen University Hospital Rigshospitalet), sharing the BOX trial’s oxygen analysis, also published in NEJM, told journalists that there’s a shared message for both analyses: “More is not better.”

Møller said that they had wanted to see what would happen during the vulnerable period when OHCA patients are intubated and ventilated. “There are translational data and some observational data that suggest that a lot of oxygen may harm the brain,” while other studies have shown “of course that if you give too little oxygen that can be harmful as well,” he explained.

As it turns out, BOX was “very conclusive” on this question, said Møller. “There was absolutely no signal of difference between these groups, and it was very stable in all subgroup analyses and also for the secondary endpoints. So even though it’s neutral, it’s a very consistent signal in this study.”

Regarding the blood-pressure analysis, Kjaergaard said the take-home message for practice is twofold. Beyond the lack of improvement with higher BP, it’s clear that “excess doses of vasopressors, [which] we know may be harmful because they’re associated with side effects and potential complications, are not called for,” he noted.

However, neither investigator ruled out the possibility that some individual patients might still see an advantage with more oxygen or higher blood pressure.

BOX Times Two

BOX enrolled 789 comatose patients at two tertiary cardiac arrest centers in Denmark. All had been admitted to the hospital after resuscitated cardiac arrest and had a sustained return to spontaneous circulation. Additionally, all received temperature control at 36°C with mechanical ventilation while under sedation for at least 24 hours, after which they were returned to normothermia and weaned off sedation.

For the oxygen comparison, patients were randomized to either a restrictive target (partial pressure of arterial oxygen [PaO2] 9-10 kPa, or 68-75 mm Hg) or a liberal target (PaO2 13-14 kPa, or 98-105 mm Hg). The blood-pressure comparison was double blinded thanks to adjustments made to the internal calibration of BP-monitoring units, which displayed values 10% above or 10% below the true measurements. Thus, while the study’s stated goal for mean target blood pressure was 70 mm Hg, in reality patients were randomized to targets of 63 or 77 mm Hg. The higher pressure goal was obtained by using high doses of vasopressors, Møller said.

The primary outcome in both arms was a composite of all-cause death or hospital discharge with severe disability or coma (Cerebral Performance Category [CPC] of 3 or 4), whichever came first within 90 days of randomization.

Rates were similar between the restrictive and liberal oxygen targets (32.0% vs 33.9%), and between the higher and lower BP targets (34% vs 32%). Within 48 hours, the median neuron-specific enolase level did not differ by strategy. By 90 days, around 30% of patients had died, while various measures of disability (CPC, Montreal Cognitive Assessment, and modified Rankin Scale score) were similar irrespective of oxygen or BP targets. Nor were there differences in adverse events, including bleeding, infection, arrhythmia, electrolyte or metabolic derangement, acute kidney damage with renal replacement therapy, and seizures.

Results followed similar patterns across various subgroups, with no interaction between the oxygen and BP targets.

Evolving Strategies

To TCTMD, Kjaergaard said the two BP goals in their study intentionally reflect what clinicians already are doing across Europe. “A lot of us were aiming for 65 [mm Hg], and a lot of other centers were aiming higher,” particularly in neurointensive care, he reported. “There’s definitely a discussion going on out there [about] which is more beneficial for the brain.” Based on their findings, “we will be staying on the lower target for future patients,” Kjaergaard added.

From his perspective, the oxygen results are “reassuring,” Møller said. “It’s not necessary to give a lot of oxygen to these patients.” Both here and with blood pressure, a simplified strategy could be helpful, as more-intensive targets translate into patients being kept on a ventilator for longer periods, he suggested.

Susanna Price, MBBS, PhD (Royal Brompton Hospital, London, England), who moderated the press conference, pointed out that there’s a challenge when studying such critically ill patients. “What you’ve shown us is potentially that overall it doesn’t matter if you have a higher or a lower pressure, but that doesn’t mean for an individual patient in front of me that they may not benefit from a higher or lower pressure,” she noted, asking if there’s a way to tease out who could still benefit from one target or the other, who might sustain harm, and who might see no effect.

Subgroup analyses or even future RCTs might help identify the best candidates for particular BP targets, Kjaergaard agreed. “But we’re not there yet.” For the researchers, Møller said that a next step will be to look closer for “hypothesis-generating” findings related to the various subgroups, though he urged caution given that the overall data are “so neutral.”

In a field that’s seen so many dead ends as of late, he suggested that one avenue for future research may relate to addressing inflammation. Not only is it common in OHCA patients, but also “a lot of the damage we see is due to inflammation,” Møller observed.