The great tragedy of Science — the slaying of a beautiful hypothesis by an ugly fact.
–Thomas Henry Huxley
The use of lung protective ventilation has become ubiquitous in the culture of critical care medicine. And while its logistical implementation has not caught up with its theoretical acceptance, this is likely due more to a fault in our ability to recognize patients who are at risk for ARDS and ventilator induced lung injury (VILI), rather than a lack of acceptance of the premise of lung protective ventilatory strategies (1). Despite the universal acceptance of low-tidal volume for the management of ARDS, the evidence supporting lung protective ventilation is far more controversial than its fabled repute.
In fact, the first few trials examining the efficacy of low-tidal volume strategies failed to find a survival benefit when compared to more traditional ventilatory strategies (2,3,4). It was not until the ARMA trial, published in the NEJM in 2000 that low-tidal volume ventilation demonstrated a statistically significant improvement in mortality (5). This trial, performed by the ARDSNET research group, randomized patients admitted to 10 centers with ARDS (defined as P/F ratio of less than 300 and bilateral pulmonary infiltrates of non-cardiac origin) to two ventilator strategies. The first, which the called traditional, prescribed 12cc/kg of tidal volume, with a maximum plateau pressure of 50 mmHg. The low-tidal volume group prescribed 4-6 cc kg of ideal body weight with a maximum plateau pressure of 30 mmHg. The trial was stopped early after enrolling only 861 patients when an interim analysis demonstrated a significant improvement in mortality in the patients randomized to the low-tidal volume group.
The benefits demonstrated in the ARMA trial were impressive. The authors found an 8.8% increase in mortality at 180-days following intubation in patients randomized to the traditional ventilation group (31% vs 38.8%).
And yet despite these impressive results, it is important to realize what the ARMA trial truly demonstrated. The control arm, termed the traditional ventilation group, was exposed to extraordinarily high tidal volumes. In fact, the traditional ventilator strategy at the time ARMA was enrolling was to vary tidal volumes based on the individual lung compliance of the patient. The median prescribed tidal volume just prior to randomization was 10cc/kg, leading to an increase in tidal volumes in the majority of patients randomized to the traditional ventilation group. In the group of patients who met inclusion criteria but were not enrolled in the ARMA cohort because of technical reasons, the overall mortality was 31.7%, almost identical to the 31% reported in the low-tidal volume group (6). Even more interesting was a subgroup analysis of the ARMA data performed by Deans et al in Critical Care Medicine in 2005 (6). In this document, the authors examined the patients in the ARMA cohort based off the individual patient’s lung compliance at the time of randomization. In the patients with low-pulmonary compliance, patients randomized to the low-tidal volume arm demonstrated a significant improvement in mortality when compared to those in the traditional tidal volume group. Conversely, in the patients with high pre-randomization compliance, those randomized to the low tidal-volume group demonstrated an increased mortality.
Additionally, the three trials that failed to find a benefit of low-tidal volume ventilation employed a control group in which the mean tidal volume was approximately 10 cc/kg of ideal body weight, very similar to the pre-randomization tidal volumes in the ARMA trial. In the two trials that did identify a benefit of low-tidal volume ventilation, the ARMA cohort and the Amato et al cohort, both ventilated their control group at concerningly high volumes (12cc/kg) (7). Essentially, the ARMA trial demonstrated that an empiric low-tidal volume strategy is superior to an empiric high-tidal volume strategy. However, the ARMA cohort provides little guidance on whether such an empiric strategy is superior to a more individualized approach. It is unclear if this empirically low dose is appropriate for everyone. Anatomic studies have demonstrated that ideal body weight is a poor predictor of of lung volumes in patients with ARDS (8). Might these low-volumes lead to atelectrauma and worsening VILI in a certain subset of patients? Is there a way to quantify the degree of de-recruitment? To better estimate the size of the baby lung?
And so a division was born. Those that believed in the ARMA data, and strictly followed a lung protective strategy, and those that would become known as the open-lung theorists, who believed the addition of higher positive end-expiratory pressure (PEEP) levels would decrease the alveolar collapse associated with low-tidal volume ventilator strategies optimizing the size of the baby lung. But when the ARDSNET trial group examined a high-PEEP protocol and compared it to their original lung protective strategy, they discovered it provided no additional benefit (9). In fact, there have been three large high quality RCTs examining various high-PEEP strategies, all of which failed to identify a benefit (10). Was this failure because the lung recruitment effects of PEEP are not as clinically important as physiologic reasoning would have us believe? Or did these PEEP trials fail due to the very same reasons the ARMA trial succeeded? The results of these trials merely state, high levels of empirically prescribed PEEP in an undifferentiated population of ARDS patients are no better than low levels of empirically prescribed PEEP. Maybe these trials failed because while the utilization of higher PEEP levels is beneficial in a portion of patients with ARDS, it is equally harmful in others (11). In fact, when these trials were analyzed in a patient level meta-analysis, there was a signal in benefit for a high-PEEP strategy in the patients with the potential for the most recruitable lung (P/F< 200) (12). The scientific literature consistently failed to support our physiologic love affair with the theoretical protective benefits of PEEP. Until 2015, when Amato et al published an analysis that offered us a chance for salvation(13). These authors proposed a simple and elegant explanation for the universally negative trials examining the benefits of PEEP.
Using individual patient data from 3,562 subjects from 9 different ARDSNET data sets, the authors examined the ability of the driving pressure (plateau pressure-PEEP), to predict mortality. What the authors found is that neither plateau pressure, nor PEEP level independently predicted mortality. Rather only when used in tandem to generate the driving pressure were their values prognostically useful. The authors noted that the damaging effects of an elevated plateau pressure were only observed in patients who had a correspondingly elevated driving pressure. Similarly, the benefits of high PEEP were only demonstrated in patients in whom a low driving pressure was observed. The protective effects of low-tidal volume ventilation were only perceived in patients where this decrease in tidal volume led to a concordant decrease in driving pressure. This simple and elegant formula not only proposes a solution to a generation of negative trials, but allows for a more personalized approach to the management of these complex patients without violating the tenets of lung-protective ventilation.
Of course none of this data is anything more than surrogates and associations. In order to truly learn the value of an individual titration strategy, it must be tested empirically in the clinical arena. The ART trial recently published in JAMA by Cavalcanti et al tested this very theory (14). The authors randomized 1013 patients with moderate to severe ARDS (P/F ratio < 200 for less than 72-hours). Patients were randomized to either a lung protective strategy similar to that in the ARMA trial or an open-lung strategy. Patients assigned to the open-lung strategy received a neuromuscular blocker and intravenous fluids and underwent a recruitment maneuver with incremental PEEP levels, followed by a decremental PEEP titration according to the best respiratory-system static compliance. This was followed by a second recruitment maneuver. After recruitment and PEEP titration, patients were ventilated under volume-assist control mode using the PEEP associated with highest respiratory-system compliance.
Patients randomized to the open-lung strategy had higher PEEP values throughout the first 7 days after enrollment. The authors also noted a decrease in driving pressure and an improvement in P/F ratio in patients in the open-lung strategy. Despite these improvements in respiratory mechanics, the authors reported 28-day mortality to be higher in the open-lung group. 28 day mortality was (55.3%) and (49.3%) in the experimental and control groups respectively, with a hazard ratio of 1.20 (95% CI, 1.01-1.42; P = .041). 6-month mortality was also higher in the open-lung strategy group (65.3% vs 59.9%;hazard ratio, 1.18; 95% CI, 1.01-1.38; P = .04). The authors also noted patients in the open-lung group had a higher 7-day mortality, fewer ventilator free days, and higher rates of barotrauma including pneumothoraces requiring drainage.
We are faced with a beautiful physiological hypothesis that when tested in an empiric fashion leads to an increase in mortality. Was it the open-lung strategy that led to these deleterious outcomes, or the impressively aggressive recruitment maneuver these patients received prior to PEEP titration. In 15.6% of patients, the maneuver had to be interrupted, most often due to hypotension or a decrease in oxygen saturation. Within 1 hour, recruitment patients randomized to the open-lung strategy more frequently experienced hypotension or the need for increased doses of vasopressors. After approximately 50% of the patients were enrolled, the authors modified their recruitment maneuver following 3 patients experiencing peri-maneuver arrest. But the therapeutic burden cannot be shouldered by the recruitment maneuver alone. When analyzed, it did not appear the authors’ attenuation of their recruitment maneuver had any influence on 28-day mortality, shifting at least some of the blame to the PEEP titration strategy. We now have four large high quality RCTs examining the efficacy of high PEEP, all with slightly different titration strategies, but all failing to demonstrate benefit of an open-lung strategy. Is it because we have yet to test the optimal protocol? Maybe, but the likelihood of this reality is becoming diminishingly small with each negative trial. What can be said is the optimization of surrogate endpoints to guide ventilator strategies, though physiologically pleasing, does not seem to directly translate into improvements in patient important outcomes, and at times can lead to harm. An ugly truth indeed.
Sources Cited:
- Bellani G, Laffey JG, Pham T, et al. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 2016;315(8):788-800.
- Stewart TE, Meade MO, Cook DJ, Granton JT, Hodder RV, Lapinsky SE, Mazer CD, McLean RF, Rogovein TS, Schouten BD, et al. Evaluation of a ventilation strategy to prevent barotrauma in patients at high risk for acute respiratory distress syndrome. N Engl J Med 1998;338: 355–361.
- Brochard L, Roudot-Thoraval F, Roupie E, Delclaux C, Chastre J, Fernandez-Mondejar E, Clementi E, Mancebo J, Factor P, Matamis D, et al. Tidal volume reduction for prevention of ventilator-induced lung injury in acute respiratory distress syndrome. Multicenter Trial Group on Tidal Volume Reduction in ARDS. Am J Respir Crit Care Med 1998;158:1831–1838.
- Brower RG, Shanholtz CB, Fessler HE, Shade DM, White P Jr, Wiener CM, Teeter JG, Doddo JM, Almog Y, Piantadosi S. Prospective, randomized, controlled clinical trial comparing traditional versus re- duced tidal volume ventilation in acute respiratory distress syndrome patients. Crit Care Med 1999;27:1492–1498.
- Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308.
- Deans KJ, Minneci PC, Cui X, Banks SM, Natanson C, Eichacker PQ. Mechanical ventilation in ARDS: One size does not fit all. Crit Care Med. 2005;33(5):1141-3.
- Eichacker PQ, Gerstenberger EP, Banks SM, Cui X, Natanson C. Meta-analysis of acute lung injury and acute respiratory distress syndrome trials testing low tidal volumes. Am J Respir Crit Care Med. 2002;166(11):1510-4.
- Chiumello D, Carlesso E, Cadringher P, Caironi P, Valenza F, Polli F, et al. Lung stress and strain during mechanical ventilation for acute respiratory distress syndrome. Am J Respir Crit Care Med. 2008;178:346–55.
- Brower RG, Lanken PN, MacIntyre N, et al; National Heart, Lung, and Blood Institute ARDS Clinical Trials Network. Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med. 2004;351(4):327-336
- Briel M, Meade M, Mercat A, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and meta-analysis. JAMA. 2010;303(9):865-73.
- Grasso S, Fanelli V, Cafarelli A, et al. Effects of high versus low positive end-expiratory pressures in acute respiratory distress syndrome. Am J Respir Crit Care Med. 2005;171(9):1002-8.
- Briel M, Meade M, Mercat A, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and meta-analysis. JAMA. 2010;303(9):865-73.
- Amato MB, Meade MO, Slutsky AS, et al. Driving pressure and survival in the acute respiratory distress syndrome. N Engl J Med. 2015;372(8):747-55.
- Writing Group for the Alveolar Recruitment for Acute Respiratory Distress Syndrome Trial (ART) Investigators. Effect of lung recruitment and titrated positive end-expiratory pressure (PEEP) vs low PEEP on mortality in patients with acute respiratory distress syndrome: a randomized clinical trial. JAMA. doi:10.1001/jama.2017.14171
Some additional thoughts from Emily Damuth (@EDamuth), EM Critical Care Physician at Cooper, regarding the recruitment maneuver used in the ART Trial:
The bundled protocol of two aggressive alveolar recruitment maneuvers performed before and after prolonged decremental PEEP titration makes it difficult to interpret the variable(s) responsible for the increased 28-day mortality in the ART trial experimental group. Setting PEEP is a careful balance between achieving alveolar recruitment and avoiding alveolar overdistention in an effort to minimize the dead space fraction. For years, the most widely described recruitment maneuver (RM) was a sustained hyperinflation (CPAP) of 35-40 cm H20 for 30-40 seconds (Brower NEJM 2004, Meade JAMA 2008, ARDSNet Crit Care Med 2003, Stewart AJRCCM 2007). A meta-analysis by Fan et al. showed an association between RMs delivered as sustained hyperinflation and improved oxygenation in patients with acute lung injury (AJRCCM 2008). The stepwise RM was first described in 2011 (Hodgson, Crit Care) and subsequently by Kacmarek (Crit Care Med 2016). These two trials utilizing stepwise RMs also showed improved oxygenation in patients with ARDS. Suzumura performed a meta-analysis of 10 RCTs examining alveolar recruitment maneuvers in patients with moderate-severe ARDS (majority CPAP, 2 stepwise RMs) and demonstrated an association with reduced mortality, although heterogeneity limited the strength of the evidence (Int Care Med 2014). In addition, there was no increased risk of barotrauma, further supporting the safety of RMs, as described in the ARDS literature through 2014.
In contrast, the stepwise RMs and decremental PEEP titration utilized in the ART trial were far more aggressive than any protocol previously described with respect to both inflation pressure and duration (4-min RM + 22-min decremental PEEP titration + 2-min RM totaling 28 minutes of elevated airway pressures). The safety of this methodology has not been established, and must be questioned, considering the worse outcomes in the experimental group, as well as the 3 instances of resuscitated cardiac arrest provoked by the maneuver. In contrast to the prolonged RMs of the ART trial, Arnal et al. (Int Care Med 2011) demonstrated that 98% of alveolar volume is recruited in the first 10 seconds of a RM. The majority of hemodynamic perturbation is observed beyond 10 seconds, while only accounting for <2% of the total alveolar volume recruited during the maneuver. We learned from Gatinoni’s work (NEJM 2006) that patients with greater recruitability are those with greater lung water, lower P/F and worse compliance (i.e. severe ARDS population). Therefore, similar to the prone positioning literature, have we studied the wrong population (e.g. severe ARDS) and dose to produce a mortality benefit? Just like any other medical therapy, we need to determine the appropriate population (severe ARDS vs. moderate-severe), route of administration (CPAP vs. PCV with high PEEP and PIP), dose (duration and inflation pressure), frequency (once, twice, more?) and timing (early exudative phase vs. fibroproliferative phase).
Is the ART trial data really the “ugly truth” about the open lung approach or just an overly aggressive protocol for alveolar recruitment and PEEP titration?
- EM Nerd-The Case of the Partial Cohort - May 24, 2020
- EM Nerd: The Case of the Sour Remedy Continues - January 20, 2020
- EM Nerd-The Case of the Adjacent Contradictions - December 23, 2019
I find it difficult to interpret the results of the ART study. The very aggressive recruitment maneuvers that were watered down mid-study as well as the incidence of double triggering asynchrony both may have meaningfully contributed to the harm seen in the results. As I attend the Critical Care Canada Forum this week, the thought leaders in mechanical ventilation do not seem to be shying away from open lung as a concept, but more focused on improved physiological targets, largely derived from esophageal manometry. The best way of setting PEEP has been elusive in RCTs. Certainly this trial was important,… Read more »
This trial has a brutal recruitment strategy and was bound to fail. In fact we need to reduce the TV significantly and a PEEP of more than 20 are detrimental. If you see recruitment when the chest is open, it is easily seen that a milder more hemofynamically stable method of lowering tidal volumes to 3ml/Kg and 20 PEEP for 45 seconds and then lowering PEEP and titrating it from 10 mmHg and 6-7 ml/Kg works better. On thing is if you are using higher PEEP we need to also use lower TV to prevent hemodynamic instability. Also earlier initiation… Read more »
Fantastic post, Rory. I’ll be honest – I like this article because it supports my pre-existing biases. In a busy unit, I don’t generally have time to do really fancy PEEP titrations on every patient. This paper indicates that a simple approach to PEEP is as good (or probably better) than going nuts with inflection points, compliance, etc. Physiologically ugly, perhaps, but clinically beautiful. This reminds me a bit of the volume-responsiveness debate. If the patient is volume-responsive, that doesn’t mean that they will necessarily benefit from volume (often they won’t). Likewise, if a patient has PEEP-recruitable lung, that doesn’t… Read more »
Josh. While I have never been a fan of recruitment maneuvers, I do sometimes titrate my PEEP to minimize the driving pressure. Because of the aggressive recruitment maneuvers in the treatment group, I don’t think that this study debunks the use of driving pressure to find optimal PEEP. That being said, I was wondering if you ever use driving pressure to titrate PEEP and, if so, how long do you wait (minutes or hours?) after changes in PEEP to re-measure driving pressure and assess changes in compliance?
Thanks for you input
Cyrus
This is going to sound dumb, but I generally go by the standard ARDSnet PEEP table. Sometimes I will titrate up/down a bit based on oxygen saturation and hemodynamics (esp. in morbid obesity where the ARDSnet table can under-PEEP some patients). I think this is a reasonable strategy for now, until we have more prospective evidence on various alternative strategies. There are certainly many other ways to to this, and insufficient evidence to know which way is the best.
One of the factors often overlooked in all the open lung strategies, is the influence of intra-cardiac shunts opening up during Recruitment manoeuvres. Given the fact that up to 30% of the population has a Persistent Foramen Ovale (PFO), it stands to reason that this subset of patients will develop a right-to-left intra-cardiac shunt, when their right sided pressures go up, as a result of sustained lung inflation. Obviously, this results in worsening hypoxemia, in response to PEEP. I have a feeling that this aspect of overlapping cardiovascular and respiratory physiology is often not stressed upon, in our training. The… Read more »
Agree with most of this but two things worth pointing out 1.) Brower et al, after Eichacker and Natanson’s disruptive article, have taken great pains to show that 12 cc/kg PIBW was indeed the standard. Part of the problem is people were using 10 cc/kg ACTUAL body weight so 12 seems higher than what they thought they were doing. I don’t have references in hand but I was working with Roy at the time and this highly contentious matter created quite the ruckus and Roy had multiple hour long presentations on the matter and indeed we invited Natanson to give… Read more »
Great points Scott thanks for the feedback. I see your point regarding the similarities to the proning literature though I would argue that the hypothesis examined by the PROSEVA authors was derived from the Sud et al meta-analysis demonstrating that the patients who potentially benefit from prone positioning where those with severe ARDS. PROSEVA confirmed these findings. In contrast the Briel et al meta-analysis published in JAMA demonstrated a statistically significant benefit for high PEEP in patients with P/F < 200, the very group tested in the ART trial. This time unfortunately the results are different. Even if a trial… Read more »
thank you, Rory.
informative and cool analysis.
tom
The bundled protocol of two aggressive alveolar recruitment maneuvers performed before and after prolonged decremental PEEP titration makes it difficult to interpret the variable(s) responsible for the increased 28-day mortality in the ART trial experimental group. Setting PEEP is a careful balance between achieving alveolar recruitment and avoiding alveolar overdistention in an effort to minimize the dead space fraction. For years, the most widely described recruitment maneuver (RM) was a sustained hyperinflation (CPAP) of 35-40 cm H20 for 30-40 seconds (Brower NEJM 2004, Meade JAMA 2008, ARDSNet Crit Care Med 2003, Stewart AJRCCM 2007). A meta-analysis by Fan et al.… Read more »
Actually it’s even more complicated than you think. Gattinoni’s group showed quite elegantly that there is no significant connection between amount of recruitability and best titrated peep (using several different methods including esophageal pressure). I think he himself found this puzzling and I haven’t found a good explanation from him except to say this has something to do with the nature of one’s ARDS (ie different subphenotypes which have been described).
https://www.ncbi.nlm.nih.gov/m/pubmed/24196193/