|
Article 1
Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest. Dankiewicz J, Cronberg T, Lilja G, Jakobsen JC, et al. N Engl J Med. 2021 Jun 17;384(24):2283-2294.
Since the HACA and Bernard trials in the early 2000s there has been a suggestion that hypothermia may improve neurologic outcomes following a VF cardiac arrest with ROSC, but these studies were limited by sample size, blinding, and control patients who developed fever. The original Targeted Temperature Management (TTM) trial published in 2013 randomized 950 patients targeting 33°C vs 36°C with no significant difference in all-cause mortality. Other interim studies analyzed hypothermia after nonshockable cardiac arrest (HYPERION) which showed a modest improvement with hypothermia (6% vs 10%, p=0.047) and no change in mortality after TBI (POLAR, EUROTHERM). In spite of the results of TTM, there remains equipoise in the community about targeting true hypothermia versus just controlling fever after cardiac arrest.
Hence, the design of TTM2 as an international, multi-center, randomized and partially blinded superiority trial, which screened >3000 patients and randomized 1,900 adult patients from 2018 to early 2020 after cardiac arrest (both shockable and nonschockable rhythms). They compared 33°C to just apyrexia (febrile above 37.8°C). Note that patients were simultaneously enrolled in the Targeted Therapeutic Mild Hypercapnia after Resuscitated Cardiac Arrest (TAME) trial, but that did not affect the randomization. The outcomes measured were mortality (primary) and neurologic function (secondary) by Rankin scale at 30d, 6mos, and 24mos(ongoing). The intention to treat analysis was performed on 1,861 patients (only 39 withdrew consent or crossed over) with 930 and 931 in each arm.
The temperature curves mimic prior studies with effective separation of temperature, albeit slowly with hypothermia taking ~3h to achieve (no prehospital cooling), and 6% of patients were re-warmed early due to arrhthymias/hemodynamic instability. At 6mos, there was no significant difference in mortality (50% vs 48%; p=0.37; RR 1.04 95%CI 0.94-1.14;) or in good neurologic outcome (Rankin 4-6) (55% vs 55%; RR 1.00 95% CI 0.92-1.09). Importantly, the hypothermia arm had a higher rate of arrhythmias (24% vs 17%, p<0.001).
The strengths of the study are blinding of clinician who performed neuroprognostication at 96 hours, blinding of families, low rate of withdrawal of consent, and excellent follow up (only1% lost to follow up). A major weakness is co-enrollment in TAME trial (10% from each group) and it will not be known whether this was confounder until this study is published. Overall, the data reinforce the results previously seen with TTM, but we still need a study that examines whether managed normothermia (avoidance of fever) is better than doing nothing about the temperature at all.
Article 2
Early individualized positive end-expiratory pressure guided by electrical impedance tomography in acute respiratory distress syndrome: a randomized controlled clinical trial. He H, Yang Y, Yuan S, Long Y, et al. Crit Care 2021 Jun 30;25(1):230.
Today’s intensivist still struggles to convince the treatment team of the role of PEEP in ARDS, in spite of a decade of literature starting with the ARDSnet group demonstrating the importance of recruitment, the open lung concept and lung protective settings. Multiple techniques have been described to try to optimize PEEP, including esophageal measurements of tranpulmonary pressures and recently use of ultrasound to try to optimize recruitment. This study examines the individual patient optimization of PEEP using electrical impedance tomography (EIT) compared to use of the low PEEP/FiO2 table from ARDSnet. EIT uses current through electrodes in a belt surrounding the thoracic cavity to describe the conductivity allowing mapping of collapsed and overdistended lung; more electrodes means better resolution and mapping of the chest. Ideal recruitment occurs when the curve measuring collapse intercepts the curve measuring overdistension. Prior studies in severe ARDS suggested that EIT may be better than use of the pressure-volume curve (Hsu et al. Physiol Meas 2021. 22(7):707-10.
The authors screened 191 adult patients from a single-center from Nov 2018-Sept 2020 and included nonobese patients with ARDS (PaO2/FiO2 <300mmHg) who could tolerate electrical impedance monitoring (lacking implanted cardiac devices). Of these patients 126 were randomized to EIT vs control (low PEEP table) within 24h with blinding to nurses and research personnel, but 9 patients were erroneously randomized leading to n=117. The primary outcome was 28d mortality and secondary outcomes included ventilator-free days, ICU length of stay, barotrauma (pneumothorax/mediastinum/peritoneum or subcutaneous emphysema).
Although there was a trend toward decreased mortality in the EIT group compared to control (21% vs 27%, p=0.634) , there was no statistically significant difference in mortality, ventilator-free days, ICU LOS, and no incidence of barotrauma. Of note, a higher proportion of patients in the EIT group had severe ARDS.
The major limitations of this study were the small sample size and the heterogeneous population (mild-severe ARDS) compared to prior studies focusing on moderate-severe ARDS. The study was significantly underpowered as the expected mortality rate was significantly lower than previously described (which should have been expected given inclusion of mild ARDS). However, this study and the prior one mentioned are compelling examples of personalized critical care medicine. Future studies should focus on severe ARDS and include EIT-based PEEP optimization throughout mechanical ventilation instead of just at initiation.
Article 3
The Effect of Reconnection to Mechanical Ventilation for 1 Hour After Spontaneous Breathing Trial on Reintubation Among Patients Ventilated for More Than 12 Hours: A Randomized Clinical Trial. Dadam MM, Goncalves AR, Mortari GL, Klamt AP, Hippler A, Lago JU, Ponikieski C, Catelano BA, Delvan D, Westphal GA. Chest. 2021 Jul;160(1):148-156.
Spontaneous breathing trials (SBT) are often used in mechanically-ventilated patients as part of an assessment of readiness for ventilator liberation. As SBTs may be associated with an increased work of breathing, some clinicians return the patient to mechanical support to rest accessory muscles of respiration, putatively to mitigate the risk of subsequent post-extubation fatigue and re-intubation.
Dadam and colleagues, in this month’s Chest, investigated the need to rest mechanically-ventilated patients after SBT, hypothesizing that returning ventilator support after a successful SBT would decrease the likelihood of subsequent reintubation.
Three-hundred and thirty-six eligible patients were randomized to receive 1 hour of reattachment to mechanical ventilation or immediate extubation. The study group excluded patients who were unable to obey commands, were predetermined to require NIV after extubation (eg, hypercapnia during SBT), had an unplanned extubation, neuromuscular disease and spinal cord injury, or tracheostomy. The authors powered the trial to identify a hypothesized 50% decrease in reintubation at 24 hours.
While the majority of enrolled patients in the study’s mixed ICUs were intubated for medical morbidities, and 22.4-28% were surgical patients with post-operative respiratory failure, and a further 15-18% were admitted to the ICU following trauma; thus there is, with caveats, some generalizability to the surgical critical care patient cohort.
Prolonged intubation (>72 hours) and clinical signs suggestive of equivocal SBT success (such as increased respiratory effort) appeared to predict early failure of extubation on univariate analysis. However, on direct comparison, ‘resting’ patients following SBT failed to statistically-significantly decrease the requirement to re-intubate within 24 hours (12.9% in the rest group vs 18.2% in the direct extubation group, P = .18). The authors thus concluded that rest for 1 hour after a successful SBT does not improve sustained ventilator liberation outcomes.
One potential explanation of the results (such as that in the editorial accompanying Dadam et al.’s manuscript) is that this well-planned study was overly ambitious in its hypothesized therapeutic effect of rest, thus underpowering the study to detect smaller incremental improvements. Conversely, however, it may also be reasonable to consider that the small difference in reintubation rate between groups may not be attributable to a therapeutic effect of rest, but instead simply unmasked that patient cohort not quite ready for ventilator liberation. These marginal patients exist on the cusp of the SBTs discriminatory ability, and may have failed with or without rest.
|
