This Week in Critical Care — May 28, 2026

Welcome to This Week in Critical Care. This week we're covering 9 notable papers spanning cardiovascular dynamics in the ICU, the long-term systemic effects of critical illness, and practical questions around ICU procedures and interventions. Let's dive in.

We start this week with a focus on cardiovascular hemodynamics, with three papers examining how we assess and phenotype shock and arrhythmias. The central paper comes from *Critical Care Medicine*, which attempts to bring precision medicine to sepsis-associated circulatory failure by identifying distinct cardiovascular subphenotypes [9].

The Study

Investigators at a single center in the United Kingdom retrospectively analyzed hemodynamic and transthoracic echocardiography data from ICU patients with sepsis. Using unsupervised clustering methods on a derivation cohort of nearly one thousand patients, they sought to identify distinct cardiovascular profiles, which they then validated in a separate cohort of over 800 patients.

Results

The analysis consistently identified a four-class model. Class 1, the largest group, comprised about half the patients and had mostly normal left and right ventricular function. Class 2, about a quarter of patients, was a hyperdynamic group with a high cardiac index. Class 3, about 10 percent of patients, was characterized by a dilated right ventricle with impaired systolic function. And Class 4, also about 10 percent, represented a low cardiac output state with depressed left ventricular ejection fraction. These phenotypes were not just academic; they carried starkly different prognoses. While the 'normal' Class 1 had a 90-day mortality around 20 percent, mortality for the other three classes was more than double, ranging from 41 to 58 percent. Importantly, the association between these classes and mortality was influenced by the specific vasopressor and fluid strategies used, suggesting these phenotypes represent potentially treatable traits.

One of the high-mortality phenotypes identified in that sepsis study was characterized by right ventricular dysfunction. But how reliably can we actually assess the RV at the bedside with a quick visual 'eyeball' assessment? A related international study, also in *Critical Care Medicine*, provides some crucial context [8]. Investigators had 73 raters with varying levels of echocardiography expertise—from novice to expert—visually assess RV function, size, and septal motion on 115 recordings from critically ill patients. They then compared these visual assessments to guideline-based quantitative parameters. The results showed that experience matters significantly. For experts, the agreement between visual grading of RV systolic function and the quantitative fractional area of change, or FAC, was strong. For novices, it was only moderate. Agreement with other measures like TAPSE, S' wave, and RV free wall strain was only moderate to low, regardless of experience level. The clinical implication here is that while a quick visual assessment of the RV by an experienced user can be a reliable part of a multiparametric evaluation, it is not a standalone tool, and its reliability is highly dependent on the operator's expertise. This underscores the need for structured training in point-of-care ultrasound.

Finally, in our cardiovascular theme, we turn from diagnosis to prevention. Another study in *Critical Care Medicine* revisited the common practice of giving intravenous magnesium to prevent postoperative atrial fibrillation in patients undergoing cardiac surgery [1]. This was a double-blind, randomized, placebo-controlled trial aiming to maintain a serum magnesium concentration between 1.5 and 2.0 mmol/L. The trial, however, was stopped early at an interim analysis for futility. The data showed no benefit. In fact, the incidence of postoperative atrial fibrillation was numerically higher in the magnesium group, occurring in about 38 percent of patients, compared to about 29 percent in the placebo group, although this difference was not statistically significant. The findings do not support the routine prophylactic use of magnesium to prevent post-op atrial fibrillation in this setting.

Next, we turn to the prolonged phase of critical illness, exploring both the metabolic and cellular consequences of inflammatory stress. For patients with a protracted ICU stay, nutritional support is a constant challenge. A prospective multicenter study in *Critical Care* tracked the metabolic journey of these patients to better understand their needs [4].

The Study

Conducted across seven European and Australian ICUs, this observational study included over 400 patients who had an ICU stay of at least 10 days. The investigators performed serial indirect calorimetry measurements to track energy expenditure and substrate utilization over time.

Results

They discovered a distinct biphasic pattern. Energy expenditure initially increased, peaking around day 10, and then subsequently declined. This pattern remained significant even after adjusting for multiple confounding factors like age, sex, inflammation, and fever. This inflection point, around day 10, coincides with the empirical onset of what we call persistent critical illness. Furthermore, a latent class analysis identified three distinct metabolic trajectories within this population: groups of patients who were persistently hypometabolic, normometabolic, or hypermetabolic. The clinical takeaway is that a patient's metabolic rate is not static during a long ICU stay. Our feeding strategies may need to adapt over time, and a one-size-fits-all approach to caloric targets is likely suboptimal.

Providing a fascinating molecular underpinning to the concept of prolonged illness, a paper in *Nature* reveals that our own hematopoietic stem cells can 'remember' past inflammatory stress [5]. Researchers identified a specific subset of human hematopoietic stem cells, which they termed 'HSC-inflammatory memory' or HSC-iM. After an inflammatory insult, these cells become quiescent and exhibit restrained blood cell production. This molecular memory program was found in stem cells from patients recovering from COVID-19, as well as in conditions like sickle cell disease and aging. This program could be passed down to differentiated immune cells, and most strikingly, its enrichment in circulating blood cells was associated with a heightened risk for all-cause mortality in large population cohorts. This provides a powerful cellular mechanism to help explain why a single severe inflammatory event, like sepsis, can have such long-lasting negative health consequences.

Rounding out this theme on systemic illness, a research statement from the American Thoracic Society, published in the *American Journal of Respiratory and Critical Care Medicine*, highlights major gaps in our understanding of a key public health tool: the United States Air Quality Index, or AQI [3]. The committee found very little research evaluating whether the AQI actually leads people to change their behavior, whether using it improves health outcomes, or even how patients with respiratory disease interpret and act on its messages. The report calls for a new research agenda to strengthen the scientific foundation of the AQI to better protect patients.

Our final section covers three papers that address common questions in resuscitation, patient management, and the care of complex populations. First, in cardiac arrest, does using a supraglottic airway, or SGA, improve outcomes when placed by Basic Life Support providers compared to standard bag-valve-mask ventilation? A new systematic review conducted for the International Liaison Committee on Resuscitation, published in *Resuscitation*, provides an answer [6].

The Evidence

The review synthesized data from 3 randomized trials and 13 observational studies. The randomized trials, which provide the highest quality evidence, showed with low certainty no significant difference in survival to hospital discharge between SGA and bag-valve-mask use. The much larger body of observational studies provided very low-certainty evidence with inconsistent findings and no clear signal of benefit for SGAs on outcomes like survival or return of spontaneous circulation. The review also noted a complete lack of data for first rescuers like lifeguards or first responders like police and firefighters.

Conclusions

Based on the current evidence, there is no demonstrated benefit of SGAs compared to bag-valve-mask ventilation when used by BLS-trained providers during cardiac arrest. This supports current guidelines and suggests that a focus on high-quality chest compressions and effective bag-valve-mask technique remains paramount.

Next, a look at the challenging issue of physical restraints in the ICU. A multicenter crossover trial in *Critical Care Medicine* tested a novel device designed as an alternative to traditional wrist restraints [2]. Fifty-four ICU patients alternated between the novel device and standard restraints in 4-hour blocks. The primary outcome was upper extremity activity, measured by wrist actigraphy. The study found no difference in activity between the two devices. There were also no significant differences in sedation levels or delirium. While satisfaction surveys showed only moderate satisfaction from patients, families, and clinicians, the qualitative feedback provided important insights into what users are looking for in a restraint alternative, highlighting themes that could guide the development of better future devices.

Finally, for our colleagues managing critically ill oncology patients, a small phase II trial in *PLoS Medicine* explored a de-escalation strategy for older or frail patients with stage III non-small-cell lung cancer who were ineligible for standard concurrent chemoradiotherapy [7]. Patients received sequential chemo-immunotherapy followed by either standard-dose or reduced-dose thoracic radiotherapy. The primary endpoint was the 1-year progression-free survival rate. In this non-comparative trial, the rate was 84 percent in the standard radiotherapy group versus 71 percent in the reduced radiotherapy group. As expected, severe adverse events were more common in the standard-dose group. While the study is small and was not powered for a formal comparison, the authors suggest that a reduced-dose radiation strategy might be a feasible approach for this vulnerable population, warranting confirmation in larger, adequately powered randomized trials.

If you only have time for one paper this week, make it the study on cardiovascular subphenotypes in sepsis by Chotalia and colleagues in *Critical Care Medicine* [9]. This work moves us closer to personalized hemodynamic management by identifying four distinct, reproducible phenotypes with vastly different mortality risks, potentially paving the way for targeted therapies beyond our current one-size-fits-all approach to sepsis.

Here are the key takeaways from this week in Critical Care.

First: For sepsis, consider that patients may fall into one of four cardiovascular subphenotypes—normal, hyperdynamic, RV failure, or LV failure—which carry different prognoses and may respond differently to fluids and vasopressors.

Second: Prophylactic magnesium infusion to a target of 1.5 to 2.0 mmol/L does not appear to reduce post-operative atrial fibrillation in cardiac surgery patients and should not be used routinely for this indication.

Third: In BLS-managed cardiac arrest, current evidence does not show a survival benefit for supraglottic airways over bag-valve-mask ventilation.

Fourth: Energy expenditure in patients with persistent critical illness is dynamic, typically peaking around day 10 before declining. This suggests our nutritional targets may need to be adjusted over the course of a long ICU stay.

Fifth: Visual assessment of right ventricular function by experienced echocardiographers shows good agreement with quantitative measures like FAC, supporting its use as part of a rapid, multiparametric bedside evaluation.

That's your roundup for This Week in Critical Care. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.