Hi everyone,
I’ve generally been an early adopter of new ideas. Of course new isn’t always better than old, and, like many fields, critical care has had its share of busts and dead ends. For example, in the early 2000s, we embraced tight glucose control and a recombinant form of activated Protein C, neither of which stood the test of time. I even extolled those therapies in an article that hasn’t aged well.
But we do need to challenge paradigms, including the widely held and seemingly self-evident. This week, we’ll highlight five paradigm-changing studies in critical care:
- Mechanical Controlled Hypoventilation in Status Asthmaticus by Darioli and Perret (1984): During critical care’s infancy, mortality was extremely high among ventilated asthmatics, above 40% in one study. Aggressive efforts to correct hypercapnia caused tension pneumothoraces and shock. Survival soared after Darioli and Perret’s study, which showed that most patients tolerated hypercapnia well, thus ushering in the era of permissive hypercapnia. It is now rare for asthmatics to die on the ventilator.
- The Effectiveness of Right Heart Catheterization in the Initial Care of Critically Ill Patients by Connors et al.(1996): When I was a trainee, right heart catheters (aka Swan Ganz or pulmonary artery catheters) were used routinely in patients with shock and ARDS to measure filling pressures and cardiac output in order to craft treatments. Occasional skeptics raised concern about the safety and effectiveness of these catheters, but their voices were marginalized in a field where mastering right heart catheterization was core to an intensivist’s identity. Critical care was transformed by Dr. Connors’ landmark study, which suggested no benefit, and potential harm, when right heart catheters were used. Follow up studies reinforced his observation, and right heart catheters are now rarely seen in the MICU.
- A National Survey of End-of-life Care for Critically Ill Patients by Prendergast et al (1998): It may seem astonishing now, but in the early years of critical care, most ICU deaths were preceded by CPR. But in the 1990s, practice shifted, as Prendergast showed in a study demonstrating that more than half of patients dying in American ICUs passed after decisions to limit life sustaining treatment. This study was my first foray into a multicenter project: for half a year I catalogued every death in the YNHH MICU, labeling each one as “full code,” “DNR,” “withhold,” “withdrawal,” and “brain death.” Since then, addressing goals of care has become central to our practice, and routine ACLS has been replaced in many cases by a shift to CMO.
- One-Year Outcomes in Survivors of the Acute Respiratory Distress Syndrome by Herridge et al (2003): As a young intensivist, I accepted the widely held notion that surviving critical illness was the key metric of success. Studies at the time focused primarily on short term outcomes, like ICU or 30-day survival. In a groundbreaking study, Dr. Margaret Herridge and colleagues showed that many ARDS survivors were left with functional limitations, particularly muscle weakness, in the year after their illness. In a follow up study, she showed that up to five years out, even after lung function returned to normal, many patients struggled with physical, neuropsychological, and social deficits. Herridge’s work helped introduce a new focus on functional outcomes in ICU survivors, which is why we now provide rehabilitation in the MICU, for example in the STEPS-ICU program led by Dr. Lauren Ferrante.
- Intensive versus Conventional Glucose Control in Critically Ill Patients by The NICE-SUGAR Investigators (2009): As an early attending, I would encourage residents to start addressing blood sugars when they rose to the 200s to 300s, implying that controlling blood glucose was an afterthought once we addressed hypoxemia and hypotension. This approach changed in 2001 with a study by Van den Berghe et al., which suggested that normalizing blood sugars would decrease ICU mortality. For years afterwards, tight glucose control was standard, but in 2009, a large multicenter trial (NICE-SUGAR) showed that a more liberal target was associated with fewer deaths, in part because patients had less hypoglycemia. In an accompanying editorial, Dr. Silvio Inzucchi and I supported a higher glucose target, 140-180, which remains the goal today. An updated version of the Yale insulin protocol, originally developed by Drs. Inzucchi, Philip Goldberg, and others, remains the standard used in ICUs at Yale and other medical centers.
If you stick around long enough, many of your most deeply held assumptions will be pushed aside. I’m less of an early adopter than I used to be, instead holding out for scientifically sound, replicable studies to guide my practice. I wonder how many of today’s practices will be overturned tomorrow. This is why it’s so important to ask questions, support science, and challenge paradigms.
Next week, we’ll return with five more studies, and by then we’ll be able to share our Match list!
Enjoy your Sunday, everyone. I’m off to New York for a big day in the city.
Mark
P.S. For anyone who missed it: congratulations to the 2026-27 Chief Residents!
P.P.S. What I’m reading:
- Musk Said No One Has Died Since Aid Was Cut. That Isn’t True. By Nicholas Kristof
- We Need a Reminder of What the Pre-Vaccine Era Was Like By Elena Conis
- 15 Lessons Scientists Learned About Us When the World Stood Still By Claire Cain Miller and Irineo Cabreros