Critical Care: A Special Issue of the Blue Journal

Welcome to the March 1, 2024, issue of the Journal, a special issue dedicated to critical care. Readers will find a broad array of topics, research, and debate in the following pages, from translational to clinical to emerging perspectives on the design and interpretation of studies in our field.

Clinical Research

Accordingly, bringing all these topics together, the special edition includes a special perspective on the international landscape of adaptive trials in critical care (pp. 491–496; 1). With authors from five major adaptive trial consortiums that are, or will soon be, under way (i.e., the PRACTICAL, PANTHER, TRAITS, INCEPT, and REMAP-CAP investigators), they provide a vision (that is increasingly becoming a reality) for experimental research in critical care. In doing so, they walk readers through the innovative design aspects and scope of each adaptive trial and take a broader look at what is needed to fully realize the potential benefits of adaptive trial designs in critical care.

In a pair of Viewpoints, de Grooth and Cremer (pp. 483–484) and Goligher and Harhay (pp. 485–487) scrutinize the growing number of Bayesian (re)analyses of critical care trials (2, 3). Together, the authors offer new insights and perspectives on how Bayesian approaches should (and should not) be used as the field increasingly moves toward Bayesian trial interpretations after decades of disappointing clinical trial results with the more familiar P value–based framework.

Sepsis

Mouse Models

The value of mouse models of sepsis has been debated for decades, with particularly robust discussion since a high-profile paper published more than a decade ago raised concerns that biological responses in humans were not well replicated in mouse models. In a Viewpoint, Matthay and colleagues (pp. 488–490) provide a strong argument that mouse models do have major value for advancing our understanding of biology and potential therapeutics in sepsis, highlighting their role in the development of therapies ranging from corticosteroids for community-acquired pneumonia to lung protective ventilation (4). Although limitations of these models are acknowledged, the Viewpoint emphasizes the utility of small animal models for identifying specific biological pathways relevant to human disease and for drug development, being mindful of animal welfare and the potential benefits of new therapeutic approaches for veterinary medicine.

Overall Prognosis

Prescott and colleagues (pp. 507–516) leveraged a unique dataset made up of two overlapping, broadly representative ICU cohorts containing more than 400,000 patients admitted over a period of two decades in the United Kingdom to assess the temporal changes in sepsis-related outcomes (5). Using patient-level risk adjustment with the Acute Physiology and Chronic Health Evaluation, site of infection, and organ dysfunction–related variables, the group found a consistent decrease in the sepsis-related in-hospital mortality rate from 54.6% to 32.4% over the study period. Moreover, the authors found that 40% of the total decrease was not explained by differences in case mix and may be attributed to improvements in care. Increased awareness of sepsis, standardized resuscitation, and the broad application of lung protective ventilation, among others, are potential causes of this improvement. Although the role of ICU admission practices is not completely accounted for, this powerful analysis highlights the patient-level direct benefits of critical care research.

Geographic Differences

Volume expansion with isotonic crystalloids is a cornerstone of resuscitation for the treatment of sepsis-related circulatory failure. Clinical trials comparing higher versus lower resuscitative volumes have had discordant results. Gendreau and colleagues (pp. 517–528) performed a systematic review and meta-analysis of available trials that examined higher versus lower volumes of resuscitation and used the income status of the trial country to explore the heterogeneity of treatment across countries (6). The investigators found that location was instrumental. Higher fluid volumes were harmful in countries with low and middle incomes compared with those with high incomes. This finding was robust to risk adjustment. A proportion of treatment effect was explained by the availability of mechanical ventilation. The results point to the urgent need to contextualize evidence accounting for variable resources and to expand the reach of clinical trials across the globe.

Mechanical Ventilation

CO₂ Removal

Duggal and colleagues (pp. 529–542) report the results of the first modern randomized clinical trial of extracorporeal carbon dioxide removal (ECCO₂R) in patients with acute exacerbations of chronic obstructive pulmonary disease (7). They randomized 113 patients from 41 hospitals in the United States over a 4-year period in two strata: 1) those deemed to be at high risk of failed noninvasive ventilation (NIV) and 2) those already receiving NIV who were not ready for ventilator liberation to usual care or ECCO₂R using a dedicated device (A-Lung). The primary outcome was NIV-free days to Day 5. The study was stopped early for low enrollment. There was no statistically significant difference in the primary outcome in either stratum, although, in the invasive ventilation stratum, there was a numeric difference of 1.75 days. Notably, in the NIV group, the risk of failed NIV and requirement for intubation was much lower than anticipated, with only 1 of 22 patients in the control group requiring intubation; both groups had a median of 5 NIV-free days to Day 5. This trial highlights the importance of understanding the current event rate in the control group: in this example, the trial was destined to find no improvement in the NIV stratum because there was almost no intubation needed in the control group. They also demonstrate the challenge of recruiting patients to such trials, with an average of fewer than 1 patient per center per year included. Overall, the role of ECCO₂R remains to be defined in these and other potential indications.

Hematological Malignancies

In decades past, it was a common assumption that patients with hematological malignancies and respiratory failure admitted to the ICU were destined for certain death. Significant progress has been made in the oncological and critical treatment of these patients, and now many have positive outcomes. Herasevich and colleagues (pp. 543–552) expand on this topic, developing and validating a lung injury prediction score for patients undergoing hematopoietic stem cell transplantation who had a subsequent hospitalization for ⩾24 hours in the subsequent year (8). A 22-point score was developed with patients in one treatment center and then validated in patients from two other centers. The model performed well in the validation cohort, with an areas under the receiver operating characteristic curve of 0.87 for predicting the need for assisted ventilation and 0.84 for the development of acute respiratory distress syndrome. This study highlights the spectrum of positive and negative outcomes for patients undergoing stem cell transplantation. The ability to predict respiratory failure could be used for prognostic enrichment in future trials to prevent respiratory progression in these patients.

Transition from the Operating Room

To limit the impact of positive ventilation on the lungs, lung protective ventilation is recommended in patients with acute respiratory distress syndrome. However, in patients with relatively normal respiratory compliance, the impact of low tidal volume ventilation—the cornerstone of lung protection—on clinical outcomes is limited. Therefore, the recommendations for lung protective ventilation during and after elective surgery are unclear. In a large prospective observational study, von Wedel and colleagues (pp. 553–562) evaluated the impact of changes in ventilator settings with the transition from the operating room to the ICU (9). After the transition, tidal volume slightly decreased, but respiratory rate increased, resulting in an increase in mechanical power. Increasing mechanical power during the transition phase was associated with higher mortality rates after adjusting for confounding variables. Although these findings should be confirmed in prospective clinical trials, they are of clinical importance. It is advisable to monitor mechanical power during the transition phase and think twice before increasing the respiratory rate.

Expiratory Muscles and Opioids

The expiratory muscles are an important component of the respiratory pump. The expiratory muscles are inactive during tidal breathing in healthy subjects, but recruitment may occur with high respiratory demands (e.g., exercise), diaphragm weakness, or pulmonary hyperinflation. In an elegant physiological study published in this issue of the Journal, Plens and colleagues (pp. 563–572) demonstrate that, in 50% of patients undergoing invasive mechanical ventilation, administration of neuromuscular blockers increases end-expiratory lung impedance (a surrogate for end-expiratory lung volume) by >10% (responders) (10). The authors convincingly demonstrate that this increase in end-expiratory impedance results from relaxation of the expiratory muscles. Clinical characteristics were not different between responders and nonresponders, although changes in end-expiratory lung impedance moderately correlated with fentanyl dose. These findings are of scientific importance because they shed new light on the role of the expiratory muscles in lung-protective ventilation. The clinical implications remain to be established.

Resilience

Resilience, the process of adapting well in the face of adversity or significant stressors, is a critical psychological attribute of ICU clinicians. Azoulay and coworkers (pp. 573–583) conducted a cross-sectional survey during the coronavirus disease (COVID-19) surge in late 2021 among more than 900 clinicians in 21 French ICUs to examine predictors of resilience (11). They found that having managed more than 10 patients who died of COVID-19, having felt frightened or isolated, and reporting greater distress from the COVID-19 “infodemic” was associated with lower resilience. They also found several factors that were associated with higher resilience, including male sex, having managed more than 50 patients with COVID-19, and more often involving families in end-of-life decisions. Notably, they also found that significant symptoms of anxiety, depression, and posttraumatic stress disorder were present in 61%, 39%, and 36% of clinicians, respectively, signaling ongoing threats to ICU clinician wellness that demand urgent attention. The study raises important questions about what organizational strategies might best promote resilience among critical care clinicians.

Tight Glucose Control

The issue of tight glucose control has been one of the most complex problems in the ICU and has had a huge impact on the delivery of care during the past 20 years. Van den Berghe and colleagues (pp. 497–506) have provided a fascinating description of their scientific journey through this topic from bench to bedside (12).

We hope our readers will have as much pleasure in discovering this edition as we had in preparing it, highlighting the breadth and depth of critical care research.