The authors reply

We read with great interest the letter to the editor written by Modesto i Alapont et al (1), suggesting to keep PaO2/FIO2 (PF) ratio instead of oxygenation index (OI) as proposed in the new pediatric acute respiratory distress syndrome (PARDS) definition (2). This change is indeed a major departure from previous and current definitions of acute respiratory distress syndrome (ARDS), and we encourage healthy debate from the global PARDS community. We appreciate the authors starting this much-needed debate. We fully agree with Modesto i Alapont et al (1) that hypoxemia or lung injury severity (however measured) is associated with mortality in ARDS and does not imply causation (3–6). This will be true regardless of the metric (PF ratio, OI, oxygen saturation index [OSI], SpO2/FIO2 [SF] ratio, and lung injury score). The ultimate causes of death in PARDS are multifactorial, and the hypoxemia severity is likely a surrogate for the severity of inflammation (not just in the lung) which when combined with preexisting comorbidities, the etiology of lung injury, and other organ dysfunction, among other variables, may explain PARDS mortality (3, 7). The authors primarily advocate maintaining PF ratio over OI to define PARDS because PF ratio could be a “physiopathologic measure of the intrapulmonary shunt, totally independent from the respiratory approach the patient (appropriately or not) is receiving.” There are several interesting issues and assumptions inherent to this statement relevant to the operational definition of PARDS. 1) PARDS severity is not simply about the degree of intrapulmonary shunt. Inherent to the pathophysiology of PARDS is shunt, V/Q mismatch, dead space, low end-expiratory lung volume, decreased compliance, etc. (8). The cited study (9) makes an argument that the PF ratio when logarithmically transformed on 100% oxygen reflects the degree of nonaerated lung tissue. In reality, this is not how PF ratio is used at the bedside, and to truly reflect intrapulmonary shunt, 100% oxygen is required, as it was in this study. 2) Such an approach (need for 100% oxygen) will also make it nearly impossible to use SpO2-based metrics to define PARDS (such as SF ratio or OSI) (10, 11) and will exclude many children who have PARDS but do not have an arterial blood gas (12). Although others have advocated an approach requiring standardization of ventilatory support (13, 14), we felt it was crucial to have a definition of PARDS that is simple to apply and does not require performance of an intervention to determine whether a given patient meets the criteria for PARDS. For the above two reasons, using PF ratio with 100% oxygen would further contribute to the under recognition of PARDS in global PICUs (12, 15). Further, Modesto i Alapont et al (1) argue that the interpretation of OI is confounded by whether or not an open lung approach (OLA) is used. This is also relevant for PF ratio. We believe that we are in agreement with the authors and the ARDS Task Force that developed the Berlin Definition of ARDS that ventilator management is relevant both to interpretation of metrics of oxygenation and applicability of these metrics to stratification of severity of illness in a manner that associates with a meaningful outcome such as mortality (16). This is well highlighted with the Acute Respiratory Distress System Network tidal volume trial where the 12 mL/kg cohort had higher PF ratios than the 6 mL/kg group. If the study had associated PF with mortality, the conclusion might have been that a higher PF is associated with higher mortality (17). The ARDS Task Force addressed the effect of mechanical ventilation strategy by adding the criterion of a minimum continuous positive airway pressure/positive end-expiratory pressure (PEEP) greater than or equal to 5 cm H2O into the definition. They considered higher PEEP (≥ 10 cm H2O) but found that it did not further discriminate mortality better than PF ratio for adult ARDS patients treated with PEEP greater than or equal to 5 cm H2O. This may be due to the fact that adult providers manage PEEP in a more protocolized fashion using PEEP (> 5 cm H2O)/FIO2 titration tables like those used in the tidal volume trial (18). The authors appear to argue that the association between PF ratio and mortality is not as influenced by ventilation strategy (i.e., OLA) as the association between OI and mortality. First, as mentioned previously and as the authors highlight, it is important to remember that the association between hypoxemia metrics and ARDS mortality is not implied to be causal. The OI is a combination metric that captures the relative cost (from the ventilator) to achieve a given PF ratio (indeed they are mathematically tied) and theoretically ties together other elements of the pathophysiology of PARDS in addition to shunt (end-expiratory lung volume, compliance, etc.). Pediatric data suggest that there are few, if any, agreed-upon methods for PEEP management in children (19, 20). Furthermore, there are insufficient data to support that the generally lower PEEP/higher FIO2 strategy used by PICU versus adult ICU practitioners leads to worse outcome. Hence, how do we truly define the OLA in children? Do we ventilate above the lower inflection point in the inflation limb? Do we ventilate above the closing capacity of the expiration limb? For some patients with severe PARDS, OLA cannot be achieved until total lung capacity is reached (21). Modesto i Alapont et al (1) suggest that physicians use PF ratios to guide PEEP management to determine whether the lung is open and that this PF ratio when the lung is open should be used for risk stratification (perhaps without reference to PEEP). This mandates more uniform ventilator practice for the PF ratios to be meaningful. Data demonstrate that there is large variability in ventilator practice in pediatrics (19, 20). Furthermore, using the Modesto i Alapont et al (1) scenarios, is it truly correct to interpret that a patient with high mean airway pressure and low FIO2 (patient 4) had less severe lung injury than the patient with lower mean airway pressure and higher FIO2 but similar OI (patient 3)? In current practice, there is no marker of PARDS severity that is completely independent of ventilator management. In contrast to PF ratio, the OI factors the important variable of ventilator management into the severity assessment, and will attempt to capture current practice, without mandating universal agreement on ventilator strategies with limited pediatric evidence. This is particularly important given that OI will also capture the relative cost (i.e., ventilatory support) with high-frequency oscillatory ventilation, which is frequently used in PARDS management. Existing data and published reports demonstrate similar associations between PF ratio and OI with mortality in PARDS, and the secondary analysis presented in the Pediatric Acute Lung Injury Consensus Conference (PALICC) article demonstrates that OI may be slightly superior to PF ratio in its ability to discriminate mortality (22). As Modesto i Alapont et al (1) pointed out, PF ratio and OI are colinear. Hence, most authors have correctly created comparative analyses with different regression models, ultimately comparing the fit of these models to each other to evaluate discrimination ability. This approach, or the multiplicative interaction approach suggested by the authors, would enable direct comparison of these two metrics. Although the authors stress the importance of the OLA in the interpretation of OI, studies have demonstrated the association between OI and mortality for over 20 years, both prior to and after the implementation (to various degrees) of lung-protective ventilation (3, 5, 6, 12, 23, 24). As such, all the PALICC investigators strongly agreed to recommend OI over PF ratio (2). Lastly, we want to emphasize that although this definition was tested and validated with existing PARDS datasets, we view this definition as merely a first step. These definitions require prospective study for validation. We agree with the authors that prioritizing simplicity is important, and no matter how sensitive and specific the definition is, if it is cumbersome to use, it has low value to clinicians and researchers. We encourage international collaboration to test this definition and invite the participation of the authors or anyone in the global PARDS community to share in this endeavor. Lincoln S. Smith, MD , Department of Pediatrics, Seattle Children's Hospital, University of Washington School of Medicine, Seattle, WA; Robinder G. Khemani, MD, MsCI , Department of Pediatrics, Children’s Hospital Los Angeles, University of Southern California Keck School of Medicine, Los Angeles, CA; Simon Erickson, MBBS, FRACP, FCICM , Department of Pediatrics, Princess Margaret Hospital for Children, University of Western Australia, Subiaco, Western Australia, Australia; Douglas F. Willson, MD , Department of Pediatrics, Children's Hospital of Richmond at Virginia Commonwealth University, Richmond, VA; Philippe Jouvet, MD, PhD , Department of Pediatrics, Sainte-Justine Hospital, Montreal, QB, Canada; Neal J. Thomas, MD, MSc , Department of Pediatrics, Penn State Hershey Children's Hospital, Pennsylvania State University College of Medicine, Hershey, PA; on behalf of the Pediatric Acute Lung Injury Consensus Conference

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.