BARCELONA -- Shifting blood pressure and oxygen targets failed to budge outcomes for comatose survivors of out-of-hospital cardiac arrest, the BOX trial showed.
A higher mean arterial blood-pressure target of 77 mm Hg -- hypothesized to potentially maintain better cerebral perfusion -- had no impact on the proportion of patients who died or left the hospital with a Cerebral Performance Category of 3 or 4, reflecting severe disability or coma, within 90 days compared with a target of 63 mm Hg (34% vs 32%, HR 1.08, 95% CI 0.84-1.37).
A restrictive oxygen target of a partial pressure of arterial oxygen (PaO2) of 68 to 75 mm Hg likewise didn't reduce the composite outcome compared with a liberal oxygen target of 98 to 105 mm Hg (32.0% vs 33.9%, HR 0.95, 95% CI 0.75-1.21).
All-cause mortality at 90 days was also similar between groups for both the higher versus lower blood pressure targets (31% vs 29%, HR 1.13, 95% CI 0.88-1.46) and liberal versus restrictive PaO2 targets (31.1% vs 28.7%).
Both parts of the two-by-two factorial trial were reported at the European Society of Cardiology (ESC) meeting here and simultaneously in separate papers in the New England Journal of Medicine.
"More is not better," concluded Jacob Eifer Møller, MD, of Copenhagen University Hospital Rigshospitalet, who presented the at an ESC press conference.
"This is reassuring that we don't need to give a lot of oxygen to these patients," he said. "If you target for a very high oxygen level, we may keep the patient in for longer time in the ventilator because you won't be able to wean it."
Median mechanical ventilation lasted 57 hours with the more restrictive target compared with 61 hours with the liberal target, although Møller cautioned against drawing solid conclusions about that secondary endpoint.
And that's a good thing for blood pressure too, said Jesper Kjaergaard, MD, also of Rigshospitalet, who presented those .
"Our take on the results is that excess doses of vasopressors that we know are harmful, because they are associated with side effects and potential complications, are not called for, and we plan to aim for the lower blood pressure target when treating future patients," he said.
While it didn't appear to matter which target was used overall, there could still be individual patients for whom it does matter, noted press conference moderator Susanna Price, MBBS, PhD, an intensivist at Royal Brompton Hospital in London.
"There are certain patient populations who are physiologically really, really different," she told Ƶ -- for example, patients with congenital heart disease who are usually hypoxic. "Looking at the exclusion criteria is really important for that study," she said.
The Blood Pressure and Oxygenation Targets in Post-Resuscitation Care (BOX) trial included 789 comatose adult patients who had been admitted to one of two large academic hospitals in Denmark after resuscitated cardiac arrest of presumed non-cardiac cause and who had a sustained return of spontaneous circulation.
Exclusion criteria included females for whom pregnancy couldn't be ruled out within the inclusion window, known bleeding diathesis, suspected or confirmed acute intracranial bleeding or acute stroke, unwitnessed asystole, known disease making 180 days survival unlikely, known poor cerebral performance before cardiac arrest, more than 4 hours from return of spontaneous circulation to screening, and systolic blood pressure under 80 mm Hg despite treatment, and temperature on admission under 30°C.
Participants were randomly assigned in a two-by-two factorial manner to either a PaO2 of 68 to 75 or 98 to 105 mm Hg (double blinded by offsetting the standard measurement modules clinicians saw) and to a blood-pressure target of 68 to 75 mm or 98 to 105 mm Hg maintained open label through use of vasopressors.
Across groups, patients received temperature control at 36°C with sedation and mechanical ventilation for at least 24 hours. The study interventions were initiated immediately after randomization and maintained until extubation or leaving the ICU.
In both portions of the trial, the intervention achieved a "clear separation" between groups, including more than 10 mm Hg in the blood pressure arm.
However, "the restrictive-target group still had a PaO2 at the upper limit of the oxygenation target," and some patients may have spontaneously gone above the target even without additional oxygen supplementation, the researchers pointed out. "Thus, we cannot rule out the possibility that the population was 'too healthy' for benefit. On arrival in the ICU, the PaO2 was higher than 15 kPa (113 mm Hg) in most patients, so the intervention was not evident until 2 hours after randomization."
However, both sets of targets reflect typical clinical practice across the field worldwide, the researchers said.
The trial was also limited by the number of patients who could be evaluated in person at 90 days, which was lower than expected, the team noted, adding that follow-up was challenged by pandemic restrictions that did not permit outpatient visits for extended periods in 2020 and 2021. Although this restriction had no effect on the primary outcome, it did affect the completeness of Montreal Cognitive Assessment scores at 90 days, the investigators said.
Also, for the oxygenation portion of the trial, the open-label design "may have biased choices regarding continued life-sustaining therapies, despite the use of a predefined algorithm for making such choices," the researchers noted.
It's been a long road filled with disappointing trial results in the journey toward improving outcomes in critical care, Price noted. One reason may be lumping patients together instead of more precise phenotyping, she suggested.
Møller was skeptical of finding any subgroups for benefit from oxygen or blood pressure targets, given the lack of any signal in the trial across patient characteristics.
But, he acknowledged that out-of-hospital cardiac arrest outcomes remain "dismal," with survival in less than 50% of patients.
What might be the next frontier to explore after BOX's neutral findings? "One thing that could be a target is inflammation," he told Ƶ. "A lot of the damage that we see is due to inflammation."
Hotline session study discussant Alexandre Mebazaa, MD, PhD, of Hospital Lariboisiere in Paris, argued that the trial was not neutral but positive, since future trials and industry can refocus on something with a better chance to work.
"In septic shock, a few years ago they had exactly the same path," he noted.
Disclosures
BOX was funded by the Novo Nordisk Foundation.
Møller disclosed relationships with Abbott Vascular, Abiomed, Boehringer Ingelheim, Novartis, Orion, and Roche.
Kjaergaard and Price disclosed no relevant relationships with industry.
Primary Source
New England Journal of Medicine
Kjaergaard J, et al "Blood-pressure targets in comatose survivors of cardiac arrest" N Engl J Med 2022; DOI: 10.1056/NEJMoa2208687.
Secondary Source
New England Journal of Medicine
Schmidt H, et al "Oxygen targets in comatose survivors of cardiac arrest" N Engl J Med 2022; DOI: 10.1056/NEJMoa2208686.