番茄社区

The common practice of using proton-pump inhibitors (PPIs) as prophylaxis against stress ulcers during invasive ventilation significantly reduced the risk of problematic gastrointestinal (GI) bleeds, without affecting mortality, a large randomized trial indicated.

Among nearly 5,000 critically ill patients in the intensive care unit (ICU), clinically important upper GI bleeding turned up in 1% of those assigned to intravenous pantoprazole versus 3.5% of placebo recipients (HR 0.30, 95% CI 0.19-0.47, P<0.001), reported Deborah Cook, MD, of St. Joseph's Healthcare Hamilton in Ontario, and colleagues.

Importantly, there was no difference in deaths at 90 days between the two groups (29.1% vs 30.9%, respectively; HR 0.94, 95% CI 0.85-1.04, P=0.25), according to findings of the REVISE trial published in the and presented simultaneously at the meeting in Belfast, Ireland.

Investigators also observed less patient-important bleeding with pantoprazole (1.5% vs 4.2% with placebo, P<0.001), while all other key secondary outcomes -- including rates of ventilator-associated pneumonia and Clostridioides difficile infection -- were similar between groups.

"We did not observe an increased risk of death among the most severely ill patients receiving pantoprazole," noted Cook and co-authors, addressing concerns raised by the and PEPTIC trials.

"The aphorism 'an ounce of prevention is worth a pound of cure' persists because it feels true. So, when trial data suggest that an ounce of prevention may actually kill you, it cuts hard against the grain," said Samuel Brown, MD, of Intermountain Medical Center in Murray, Utah, writing in an .

That potential signal for increased mortality combined with observational data suggesting a possible link between PPI use and more C. difficile infections and ventilator-associated pneumonia has "continued to vex the critical care community regarding the safety of proton-pump inhibitors in the ICU," he noted.

But the new findings, along with a simultaneously reported meta-analysis in that combines data from REVISE and SUP-ICU (which also tested pantoprazole vs placebo), do not support an association between pantoprazole and the risk for either pneumonia or C. difficile infection, said Brown.

On the mortality front, he continued, the meta-analysis suggested a decreased risk at 90 days among the less severely ill subgroup in the two trials, but confidence intervals for the more severely ill subset "included unity for higher mortality":

• Less severely ill: relative risk (RR) 0.89 (95% CI 0.80-0.98)
• More severely ill: RR 1.08 (95% CI 0.96-1.20)

Subgroup analysis of mortality in REVISE alone showed an HR of 0.85 (95% CI 0.74-0.98) for those with Acute Physiology and Chronic Health Evaluation (APACHE) II scores below 25 and an HR of 1.04 (95% CI 0.89-1.20) for those with APACHE II scores of 25 or above.

"Personally, with no partisan fervor or false certainty, I plan to prescribe prophylactic proton-pump inhibitors to patients during mechanical ventilation if they have an APACHE II score of less than 25," or a reasonable equivalent, Brown wrote. "For sicker patients, I would probably reserve the use of proton-pump inhibitors for those who are being treated with antiplatelet agents, especially in the presence of therapeutic anticoagulants."

The international from Cook and colleagues randomized 4,821 critically ill adults undergoing invasive ventilation at 68 ICUs to intravenous pantoprazole (40 mg daily) or placebo for 90 days or until discontinuation of invasive ventilation.

Participants in the study had a mean age of 58 years, 64% were men, and 10% had an active COVID infection. Respiratory illness was the most common admitting diagnostic category (31-32%), followed by neurologic (22-23%), trauma (13-15%), cardiovascular (10-11%), and other reasons. Nearly two-thirds had APACHE II scores below 25.

About one-third had received glucocorticoids at least a week before their hospitalization, more than three-fourths had no prior acid suppression, and about one-fifth were previously on a PPI. "We did not observe more gastrointestinal bleeding in the subgroup of patients in the placebo group who received prehospital acid suppression, despite concern about rebound gastric acid hypersecretion," wrote Cook and colleagues.

Clinically important upper GI bleeds was the primary efficacy endpoint of REVISE, while mortality at 90 days was the primary safety endpoint. Subgroup analyses showed consistent benefit with pantoprazole on the GI bleeding endpoint, including in those with an active COVID infection and by type of admitting diagnostic category.

As noted, most secondary outcomes were not significantly different between the pantoprazole and placebo groups, including ventilator-associated pneumonia in the ICU (23.2% vs 23.8%, respectively), C. difficile infection during hospitalization (1.2% vs 0.7%), new renal-replacement therapy in the ICU (6.1% vs 6%), and death either in the ICU (20.3% vs 21.5%) or hospital (26.3% vs 28.4%).

Cook's team also reported no important differences in tertiary outcomes, which included the number of red-cell units transfused, peak serum creatinine level in the ICU, time on mechanical ventilation (median 6 days for both groups), and time spent either in the ICU (10 days for both) or hospital (20 days in the PPI group and 21 days in the placebo group).

Excluding events related to the trial outcomes, only one suspected serious adverse drug reaction was reported, but in the placebo group.

Comment