Pediatric Acute Respiratory Distress Syndrome: Consensus Recommendations From the Pediatric Acute Lung Injury Consensus Conference
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Pediatric Acute Respiratory Distress Syndrome: Consensus Recommendations From the Pediatric Acute Lung Injury Consensus Conference The Pediatric Acute Lung Injury Consensus Conference Group Objective: To describe the final recommendations of the Pediatric Interventions: None. Acute Lung Injury Consensus Conference. Methods: A panel of 27 experts met over the course of 2 years Design: Consensus conference of experts in pediatric acute lung to develop a taxonomy to define pediatric acute respiratory dis- injury. tress syndrome and to make recommendations regarding treat- Setting: Not applicable. ment and research priorities. When published, data were lacking Subjects: PICU patients with evidence of acute lung injury or a modified Delphi approach emphasizing strong professional agreement was used. acute respiratory distress syndrome. Measurements and Main Results: A panel of 27 experts met over Supported, in part, by the Department of Pediatrics, The Pennsylvania State the course of 2 years to develop a taxonomy to define pediatric University College of Medicine; Health Outcome Axis–Ste Justine Research acute respiratory distress syndrome and to make recommenda- Center, Montreal, Canada; Respiratory Research Network of Fonds de tions regarding treatment and research priorities. When published Recherche du Québec-Santé, QC, Canada; Mother and Children French- Speaking Network; French-Speaking Group in Pediatric Emergency and data were lacking a modified Delphi approach emphasizing strong Intensive Care, French-Speaking Intensive Care Society (SRLF); European professional agreement was used. The Pediatric Acute Lung Injury Society for Pediatric and Neonatal Intensive Care Society (travel support Consensus Conference experts developed and voted on a total for European experts); Australian and New Zealand Intensive Care Society (travel support for Australian expert); Children’s Hospital of Richmond of of 151 recommendations addressing the following topics related Virginia Commonwealth University; Division of Critical Care Medicine, CS to pediatric acute respiratory distress syndrome: 1) Definition, Mott Children’s Hospital at the University of Michigan; and Department of prevalence, and epidemiology; 2) Pathophysiology, comorbidities, Anesthesia and Critical Care, Children’s Hospital of Philadelphia. and severity; 3) Ventilatory support; 4) Pulmonary-specific ancillary Dr. Jouvet received grants from the respiratory research network of Fonds de Recherche du Québec-Santé, Réseau mère enfant de la francophonie, treatment; 5) Nonpulmonary treatment; 6) Monitoring; 7) Nonin- and Research Center of Ste-Justine Hospital related to the submitted work; vasive support and ventilation; 8) Extracorporeal support; and 9) and received equipment on loan from Philips and Maquet outside the submit- Morbidity and long-term outcomes. There were 132 recommenda- ted work. Dr. Thomas served on the Advisory Board for Discovery Labora- tories and Ikaria outside the submitted work; received a grant from United tions with strong agreement and 19 recommendations with weak States Food and Drug Administration Office of Orphan Product Development agreement. Once restated, the final iteration of the recommenda- outside the submitted work. Dr. Willson served on the Advisory Board for tions had none with equipoise or disagreement. Discovery Laboratories outside the submitted work. Drs. Khemani, Smith, Dahmer, and Watson received grants from the National Institutes of Health Conclusions: The Consensus Conference developed pediatric- (NIH) outside the submitted work. Dr. Zimmerman received research grants specific definitions for acute respiratory distress syndrome and from the NIH, Seattle Children's Research Institute, and ImmuneXpress out- side the submitted work. Drs. Flori and Sapru received grants from the NIH recommendations regarding treatment and future research priori- related to the submitted work. Dr. Cheifetz served as a consultant with Philips ties. These are intended to promote optimization and consistency and Hill-Rom outside the submitted work; and received grants from Philips, of care for children with pediatric acute respiratory distress syn- Care Fusion, Covidien, Teleflex, and Ikaria outside the submitted work. Drs. Rimensberger and Kneyber received travel support from the European Soci- drome and identify areas of uncertainty requiring further investiga- etiy of Pediatric and Neonatal Intensive Care (ESPNIC) related to this work. tion. (Pediatr Crit Care Med 2015; XX:00–00) Dr. Tamburro received a grant from United States Food and Drug Administra- Key Words: acute lung injury; acute respiratory distress syndrome; tion Office of Orphan Product Development outside the submitted work. Dr. Emeriaud received a grant from Respiratory Health Network of the Fonds de consensus development conference; guidelines; pediatrics la Recherche du Québec – Santé outside the submitted work. Dr. Newth served as a consultant for Philips Medical outside the submitted work. Drs. Erickson, Quasney, Curley, Nadkarni, Valentine, Carroll, Essouri, Dalton, Mac- rae, Lopez-Cruces, Santschi, and Bembea have disclosed that they do not S have any potential conflicts of interest. ince the first description of the “acute respiratory distress For information regarding this article, E-mail: nthomas@psu.edu syndrome” (ARDS) by Ashbaugh et al (1) in 1967, pediat- Copyright © 2015 by the Society of Critical Care Medicine and the World ric intensivists have recognized that ARDS in children is Federation of Pediatric Intensive and Critical Care Societies different from ARDS in adults. In the absence of identification DOI: 10.1097/PCC.0000000000000350 of these differences, however, children have been characterized Pediatric Critical Care Medicine www.pccmjournal.org 1 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Pediatric Acute Lung Injury Consensus Conference Group as having acute lung injury (ALI) and ARDS based on the adult METHODS definitions originating from the 1994 American-European Three members of the organizing committee met in March 2012 Consensus Conference (AECC) (2). Seventeen years later, a to define the methodology, to select the subtopics for study, and second consensus conference was convened with the intent of to identify the experts in the field. Experts were invited based improving the feasibility, reliability, and validity of the ALI/ on their record of publications in PARDS in the past 5 years and ARDS definitions. As with the previous AECC, however, this their participation in clinical research studies in pediatric criti- was conducted without specific consideration of children. The cal care. The final list of 27 experts, representing 21 academic new Berlin definitions (3) included several significant changes: institutions and eight countries, constituted the PALICC expert 1) the ALI category was eliminated and replaced with a grada- group (Appendix 1). Of note, only one expert declined to par- tion of ARDS severity (mild, moderate, and severe) based on ticipate due to personal reasons; two experts initially agreed to the degree of oxygenation disturbance; 2) a minimum of 5 cm participate but were subsequently unable for personal reasons. H2O of positive end-expiratory pressure (PEEP) was required; The first PALICC meeting took place in Chicago, IL, on and 3) the determination of cardiac failure was rendered more October 2, 2012, in conjunction with the fall meeting of the subjective in view of the decreased utilization of pulmonary PALISI Network. At this meeting, we discussed and agreed artery catheters. upon conference subtopics, the project timeline, and the con- Both the AECC and Berlin ARDS definitions were focused sensus methodology (Fig. 1). Experts were also assigned to on adult lung injury and have limitations when applied to each of the nine subtopics. The modified Delphi approach pre- children. For example, a major shortcoming is the necessity viously employed by the French Society of Pediatric Intensive of invasive measurement of arterial oxygen. Pulse oximetry Care (13) was chosen as the methodology to achieve consen- is increasingly obviating the use of arterial blood gas mea- sus. This approach was necessary because of the limited data surement in pediatrics, and consequently, definitions requir- and low level of available evidence, as well as the high vari- ing direct measurement of PaO2 may underestimate ARDS ability in clinical practice in PARDS. A detailed description of prevalence in children. This may result in the selection of this methodology is available in the supplement published in children with more severe hypoxemia and/or cardiovascular Pediatric Critical Care Medicine (14). compromise. A second limitation is the use of the PaO2/FIO2 Between the first and second meeting, each group of experts (P/F) ratio. In addition to requiring measurement of PaO2, undertook a comprehensive, standardized literature review. this ratio is greatly influenced by ventilator pressures (4–7). Upon completion, each group drafted their recommendations Although the Berlin definition requires a minimum PEEP of along with detailed arguments to support them. The second 5 cm H2O, other ventilator manipulations and the practice meeting occurred in Montreal, QC, Canada, on April 18–19, patterns around PEEP management can also alter this ratio. 2013. At this 2-day meeting, the recommendations were dis- Consequently, differences in clinical practice may influence the cussed and the wording of each agreed upon by the majority diagnosis, particularly in the PICU where there is greater vari- of experts. Possible omissions for any of the nine topics were ability in ventilator management relative to adult ICUs (8, 9). also discussed. After the second meeting, recommendations This has led some pediatric practitioners to adopt the oxygen- with their respective arguments (long texts) were distributed ation index (OI) ([FIO2 × mean airway pressure (Paw) × 100]/ to each expert for electronic scoring by the Research ANd PaO2) and oxygen saturation index (OSI) ([FIO2 × Paw × 100]/ Development/University of California Los Angeles (RAND/ SpO2) to assess hypoxemia in children (10, 11). Finally, differ- UCLA) appropriateness method (15). Experts with a disclosed ences in risk factors, etiologies, pathophysiology, and outcomes conflict of interest were excluded from voting on areas where between adults and children were not considered in either the any real or perceived conflict was identified. After the initial AECC or Berlin definitions. scoring, all recommendations were consolidated by the orga- These concerns prompted the organization of the Pediatric nizing committee. Acute Lung Injury Consensus Conference (PALICC) (12). The Agreement was determined by voting using the RAND/ concept originated with the Pediatric Acute Lung Injury and Sepsis UCLA scale (scores range from 1 to 9), with each expert having Investigators (PALISI) Network but was subsequently supported by an equal vote but with the highest and lowest scores discarded the Australian and New Zealand Intensive Care Society, Canadian after each vote. “Strong” agreement required that all experts rank Critical Care Trials Group, World Federation of Pediatric Intensive the recommendation 7 or higher. “Weak” agreement meant that and Critical Care Societies, European Society for Pediatric and at least one more expert ranked the recommendation below 7, Neonatal Intensive Care, and French Group for Pediatric Intensive but the median vote was at least 7. Those with strong agreement Care and Emergency Medicine. The goals of the conference were were considered complete, and those with weak agreement were 1) to develop a taxonomy to define pediatric ARDS (PARDS), revised based on comments by the experts. These revised rec- specifically predisposing factors, etiology, and pathophysiology; ommendations were then distributed for a second round of 2) to offer recommendations regarding therapeutic support of electronic voting. After this voting, some reworded recommen- the patient with PARDS; and 3) to identify priorities for future dations obtained a strong agreement. For the remaining recom- research in PARDS, including defining short- and long-term out- mendations with a weak agreement after the second round, the comes of interest. We also hoped to foster collaborative relation- percentage of experts who rated 7 or above was calculated and ships for future international research in PARDS. is reported after each weak recommendation. With this method 2 www.pccmjournal.org 888 s 6OLUME s .UMBER 888 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Feature Article Figure 1. Plan for the three meetings of the Pediatric Acute Lung Injury Consensus Conference (PALICC). The timeline, including the tasks, that has been completed by the PALICC experts. PALISI = Pediatric Acute Lung Injury and Sepsis Investigators. of calculation, a strong agreement corresponded to a percent- Aspiration Syndrome, and pneumonia and sepsis acquired age of agreement more than 95% (no more than one expert during delivery), or other congenital abnormalities (e.g., con- rated below 7 on the RAND/UCLA scale). genital diaphragmatic hernia or alveolar capillary dysplasia). The third and final meeting took place on October 9, 2013, Strong agreement in Paris, France. Each group presented their final recommen- 1.1.2 We recommend that in the absence of a compel- dations, and a third round of voting was conducted for several ling rationale related to physiology or feasibility, studies specific but unresolved recommendations related to the defi- of PARDS should not include age limits. In order to better nitions. The organizers believed it was vital to achieve strong understand the pathobiology of PARDS across the spectrum agreement regarding definitions, and this was accomplished of age, and in the absence of a clear break point in the epide- after much dialog and debate. Additionally, each group of miology of PARDS, adult and pediatric investigators should experts presented their consensus regarding key areas of con- engage in collaborative studies targeting adolescents and troversy and future research. young adults. Future studies are needed to evaluate potential age-dependent differences in the pathophysiology of PARDS RESULTS across the entire pediatric age spectrum. Strong agreement The nine topics studied by PALICC resulted in 151 total rec- 1.2 Timing and Triggers for PARDS. 1.2.1 We recommend ommendations, including 132 recommendations with strong that symptoms of hypoxemia and radiographic changes must agreement and 19 with weak agreement. Once restated, the occur within 7 days of a known clinical insult to qualify for final iteration of the recommendations had none with equi- PARDS. Strong agreement poise or disagreement, according to the predefined definitions 1.3 Defining PARDS in Children With Left Ventricular by the RAND/UCLA appropriateness methodology. The rec- Dysfunction. 1.3.1 We recommend that children with left ven- ommendations for each topic are listed below, with the justifi- tricular heart dysfunction that fulfill all other PARDS criteria cation for these recommendations detailed in the supplement have PARDS if the acute hypoxemia and new chest imaging in this issue of Pediatric Critical Care Medicine. changes cannot be explained by acute left ventricular heart fail- ure or fluid overload. Strong agreement Section 1: Definition, Prevalence, and Epidemiology 1.4 Radiographic Findings. 1.4.1 We recommend that chest 1.1 Age. 1.1.1 We recommend that there should not be age imaging findings of new infiltrate(s) consistent with acute criteria for the definition of PARDS. However, exclusion cri- pulmonary parenchymal disease are necessary to diagnose teria for PARDS should include causes of acute hypoxemia PARDS. Strong agreement that are unique to the perinatal period, such as prematurity- 1.4.2 We recommend that future clinical trials for PARDS related lung disease, perinatal lung injury (e.g., Meconium should stratify patients by the presence or absence of bilateral Pediatric Critical Care Medicine www.pccmjournal.org 3 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Pediatric Acute Lung Injury Consensus Conference Group infiltrates on chest imaging. In order to minimize variability in PARDS criteria (acute onset, a known clinical insult, and these studies, investigators should standardize interpretation chest imaging supporting new onset pulmonary parenchy- of all chest imaging. Strong agreement mal disease) and have an acute deterioration in oxygenation 1.4.3 We recommend that future studies are needed to from baseline which meets oxygenation criteria for PARDS. determine the optimal common training or effect of auto- Strong agreement mated methodologies to reduce interobserver variability 1.9.2 We recommend that patients with cyanotic con- in the interpretation of chest imaging for PARDS. Strong genital heart disease are considered to have PARDS if they agreement fulfill standard criteria (acute onset, a known clinical insult, 1.5 Measures of Oxygenation in the Definition. 1.5.1 We and chest imaging supporting new onset pulmonary paren- recommend that OI, in preference to P/F ratio, should be the chymal disease) and have an acute deterioration in oxygen- primary metric of lung disease severity to define PARDS for ation not explained by the underlying cardiac disease. Strong all patients treated with invasive mechanical ventilation. Strong agreement agreement 1.9.3 We recommend that children with chronic lung 1.5.2 We recommend that P/F ratio should be used to diag- disease who are not on mechanical ventilation at baseline nose PARDS for patients receiving noninvasive, full-face mask or cyanotic congenital heart disease with acute onset of ill- ventilation (continuous positive airway pressure [CPAP] or bi- ness that satisfy PARDS criteria should not be stratified by level positive airway pressure [BiPAP]) with a minimum CPAP OI or OSI risk categories. Future studies are necessary to of 5 cm H2O. Strong agreement determine PARDS risk stratification of patients with acute- 1.6 Pulse Oximetry Versus Pao2. 1.6.1 We recommend that on-chronic hypoxemic respiratory failure. Strong agreement OSI should be used when an OI is not available for stratifica- 1.9.4 We recommend that future studies of PARDS should tion of risk for patients receiving invasive mechanical ventila- endeavor to include children with preexisting pulmonary and tion. Strong agreement cardiac disease. Strong agreement 1.6.2 We recommend that oxygen saturation/FIO2 ratio can Based on the recommendations above, Figure 2 details the be used when P/F ratio is not available to diagnose PARDS proposed definitions of PARDS, and Figure 3 details the pro- in patients receiving noninvasive full-face mask ventilation posed definitions for those children at risk for PARDS. (CPAP or BiPAP) with a minimum CPAP of 5 cm H2O. Strong agreement Section 2: Pathophysiology, Comorbidities, and 1.7 Other Markers of Lung Disease Severity. 1.7.1 We rec- Severity ommend that given the limited published data on dead space 2.1 Pathophysiology. 2.1.1 There may be a difference in the in PARDS, there is insufficient evidence to recommend a mea- progression and outcome from ARDS in children as compared sure of dead space as part of the diagnostic criteria for PARDS. with adults. We recommend that future studies be designed to Strong agreement examine whether there are differences in the progression and/ 1.7.2 We recommend that future study is needed to deter- or outcome of ARDS between adults and children or between mine the potential relevance of elevated dead space for the children of different ages. Strong agreement definition of PARDS. Strong agreement 2.1.2 There is a paucity of studies related to the patho- 1.7.3 We recommend that measures of respiratory system physiology of PARDS. The impact of postnatal maturational compliance should not be used for the definition of PARDS. development on the pathophysiology of PARDS is unknown. Future studies of respiratory system compliance with reliable We recommend that biomarker and genetic studies that may and standardized methods for measurement are warranted provide insight into the pathophysiology of PARDS in chil- to determine the relevance of respiratory system compli- dren, and study of pathophysiology in animals of different ance to the diagnosis and risk stratification of PARDS. Strong ages with age cutoffs informed by chronology of postnatal lung agreement and immune system development, should be a focus of future 1.8 Characterizing Oxygen Delivery for Noninvasive Ven- research protocols. Strong agreement tilation. 1.8.1 We recommend that to apply SpO2 criteria to 2.2 Severity of Illness. Disease severity measures can be diagnose PARDS, oxygen therapy should be titrated to achieve subdivided into measures that can be made at the bedside, the SpO2 between 88% and 97%. Strong agreement measures requiring more in-depth calculation, biochemical 1.8.2 We recommend that defining a group of patients at measurements, and early responsiveness to therapy. risk for PARDS is necessary to determine the epidemiology of 2.2.1 Of the measures available at the bedside, both oxygen- disease progression and potential avenues for disease preven- ation defect and ventilation defect have generally been found tion. Strong agreement to be associated with outcome. There is great inconsistency in 1.9 Defining PARDS in Children With Chronic Car- the literature, however, concerning the optimal timing of these diorespiratory Disease. 1.9.1 We recommend that patients measurements. We recommend evaluating respiratory indi- with preexisting chronic lung disease who are treated with ces and biomarkers at the onset of PARDS, within the first 24 supplemental oxygen, noninvasive ventilation, or inva- hours of onset, as well as serial measures beyond that is indi- sive ventilation via tracheostomy should be considered to cated according to treatment and/or clinical studies. Strong have PARDS if they have acute changes that meet standard agreement 4 www.pccmjournal.org 888 s 6OLUME s .UMBER 888 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Feature Article mortality or length of mechanical ventilation have resulted in con- flicting results; some studies exhibit associations with outcomes while others do not. We recommend that future studies incorporat- ing variables such as tidal volume, peak and plateau airway pressures, PEEP, or Paw use explicit proto- cols and definitions such that these measures can be more robustly evaluated. Strong agreement 2.2.5 Among measures requir- ing more in-depth calculation, we recommend that the use of an estimate of multiple organ system failure should be included in any studies of clinical risk factors asso- Figure 2. Pediatric acute respiratory distress syndrome definition. OI = oxygenation index, OSI = oxygen ciated with outcome in patients saturation index. aUse PaO2-based metric when available. If PaO2 not available, wean FIO2 to maintain with PARDS. Strong agreement SpO2 ≤ 97% to calculate OSI or oxygen saturation/FIO2 ratio. bFor nonintubated patients treated with supplemental oxygen or nasal modes of noninvasive ventilation, see Figure 3 for at-risk criteria. cAcute 2.2.6 With respect to evaluating respiratory distress syndrome severity groups stratified by OI or OSI should not be applied to children risk factors related to organ failure with chronic lung disease who normally receive invasive mechanical ventilation or children with cyanotic in a research related to PARDS, congenital heart disease. OI = (FIO2 × mean airway pressure × 100)/PaO2. OSI = (FIO2 × mean airway pressure × 100)/SpO2. caution should be exercised in the use of organ failure scoring sys- 2.2.2 For disease severity measures that can be made at tems that include indices of respiratory failure. We recommend the bedside, we recommend that future research studies the development of a validated, nonpulmonary organ failure evaluating both trajectory of illness and recovery should use definition for use in PARDS research. Strong agreement standardized, minimal datasets with adequately explicit defi- 2.2.7 We recommend further research into the potential use nitions. Strong agreement of combinations of biomarker levels in providing a stronger 2.2.3 Recent adult studies evaluating the effect of dead- prediction of outcome. Strong agreement space ventilation, thereby reflecting lung perfusion, have been 2.2.8 We recommend that early response to therapy should highly predictive of outcome. We recommend that future mul- not be used as a primary outcome measure in phase III clinical ticenter studies should examine the association of dead space research trials. Future research should explore the relationship and outcome of PARDS. Strong agreement of early response to therapy as an intermediate process vari- 2.2.4 Studies examining the relationship between tidal vol- able linked to more clinically relevant, long-term outcomes ume, peak airway pressures, PEEP, or mean airway pressure with (e.g., ventilator-free days and mortality). Strong agreement Section 3: Ventilatory Support 3.1 Modes of Conventional Ven- tilation. 3.1.1 There are no out- come data on the influence of mode (control or assisted) during conventional mechanical ventila- tion. Therefore, no recommenda- tion can be made on the ventilator mode to be used in patients with PARDS. Future clinical studies should be designed to assess con- trol and assisted modes of ventila- tion on outcome. Strong agreement 3.2 Tidal Volume/Plateau Pressure Limitations. 3.2.1 In Figure 3. At risk of pediatric acute respiratory distress syndrome definition. aGiven lack of available data, for patients on an oxygen blender, flow for at-risk calculation = FiO2 × flow rate (L/min) (e.g., 6 L/min flow at 0.35 any mechanically ventilated pedi- FiO2 = 2.1 L/min). bIf PaO2 not available, wean FiO2 to maintain SpO2 ≤ 97% to calculate oxygen saturation index. atric patient, we recommend in Pediatric Critical Care Medicine www.pccmjournal.org 5 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Pediatric Acute Lung Injury Consensus Conference Group controlled ventilation to use tidal volumes in or below the range 3.4.4 We recommend that, in addition to the use of HFOV, of physiologic tidal volumes for age/body weight (i.e., 5–8 mL/kg HFJV might be considered in patients with severe air leak syn- predicted body weight) according to lung pathology and respi- drome. Weak agreement (64% agreement) ratory system compliance. Weak agreement (88% agreement) 3.4.5 High-frequency percussive ventilation (HFPV) is not 3.2.2 We recommend using patient-specific tidal vol- recommended for routine ventilatory management of PARDS. umes according to disease severity. Tidal volumes should Strong agreement be 3–6 mL/kg predicted body weight for patients with poor 3.4.6 We recommend that HFPV can be considered in respiratory system compliance and closer to the physiologic patients with PARDS and secretion-induced lung collapse, range (5–8 mL/kg ideal body weight) for patients with bet- which cannot be resolved with routine clinical care (e.g., inha- ter preserved respiratory system compliance. Weak agree- lational injuries). Weak agreement (72% agreement) ment (84% agreement) 3.5 Liquid Ventilation. 3.5.1 The clinical use of liquid ven- 3.2.3 In the absence of transpulmonary pressure measure- tilation cannot be recommended. Strong agreement ments, we recommend an inspiratory plateau pressure limit 3.6 Endotracheal Tubes. 3.6.1 Cuffed endotracheal tubes of 28 cm H2O, allowing for slightly higher plateau pressures (ETTs) are recommended when conventionally ventilating a (29–32 cm H2O) for patients with increased chest wall elas- patient with PARDS. Strong agreement tance (i.e., reduced chest wall compliance). Weak agreement 3.6.2 We recommend allowing for an ETT air leak during (72% agreement) HFOV to augment ventilation, if needed, assuming Paw can be 3.3 PEEP/Lung Recruitment. 3.3.1 We recommend mod- maintained. Strong agreement erately elevated levels of PEEP (10–15 cm H2O) titrated to the 3.7 Gas Exchange. 3.7.1 We recommend that oxygenation observed oxygenation and hemodynamic response in patients and ventilation goals are titrated based on the “perceived” with severe PARDS. Weak agreement (88% agreement) risks of the toxicity of the ventilatory support required. Strong 3.3.2 We recommend that PEEP levels greater than 15 cm agreement H2O may be needed for severe PARDS, although attention 3.7.2 We recommend that for mild PARDS with PEEP less should be paid to limiting the plateau pressure as previously than 10 cm H2O, SpO2 should generally be maintained at 92– described. Strong agreement 97%. Weak agreement (92% agreement) 3.3.3 We recommend that markers of oxygen delivery, 3.7.3 We recommend that after optimizing PEEP, lower SpO2 respiratory system compliance, and hemodynamics should be levels (in the range of 88–92%) should be considered for those closely monitored as PEEP is increased. Strong agreement with PARDS with PEEP at least 10 cm H2O. Strong agreement 3.3.4 We recommend that clinical trials should be designed 3.7.4 Insufficient data exist to recommend a lower SpO2 to assess the effects of elevated PEEP on outcome in the pedi- limit. Strong agreement atric population. Strong agreement 3.7.5 When SpO2 is less than 92%, monitoring of central 3.3.5 We recommend careful recruitment maneuvers in venous saturation and markers of oxygen delivery is recom- the attempt to improve severe oxygenation failure by slow mended. Strong agreement incremental and decremental PEEP steps. Sustained inflation 3.7.6 We recommend that permissive hypercapnia should maneuvers cannot be recommended due to lack of available be considered for moderate-to-severe PARDS to minimize data. Weak agreement (88% agreement) ventilator-induced lung injury. Strong agreement 3.3.6 We recommend that clinical trials should be designed 3.7.7 We recommend maintaining pH 7.15–7.30 within to assess optimal recruitment strategies in infants and children lung protective strategy guidelines as previously described. with PARDS. Strong agreement There are insufficient data to recommend a lower limit for pH. 3.4 High-Frequency Ventilation. 3.4.1 We recommend Exceptions to permissive hypercapnia should include intra- that high-frequency oscillatory ventilation (HFOV) should cranial hypertension, severe pulmonary hypertension, select be considered as an alternative ventilatory mode in hypoxic congenital heart disease lesions, hemodynamic instability, respiratory failure in patients in whom plateau airway pres- and significant ventricular dysfunction. Weak agreement (92% sures exceed 28 cm H2O in the absence of clinical evidence of agreement) reduced chest wall compliance. Such an approach should be 3.7.8 Bicarbonate supplementation is not routinely recom- considered for those patients with moderate-to-severe PARDS. mended. Strong agreement Weak agreement (92% agreement) 3.4.2 In HFOV, we recommend that the optimal lung vol- Section 4: Pulmonary-Specific Ancillary Treatment ume be achieved by exploration of the potential for lung 4.1 Inhaled Nitric Oxide. 4.1.1 Inhaled nitric oxide is not rec- recruitment by a stepwise increase and decrease of the Paw ommended for routine use in PARDS. However, its use may be (continuous distending pressure) under continuous monitor- considered in patients with documented pulmonary hyperten- ing of the oxygenation and CO2 response as well as hemody- sion or severe right ventricular dysfunction. In addition, it may namic variables. Strong agreement be considered in severe cases of PARDS as a rescue from or 3.4.3 We cannot recommend the routine use of high-fre- bridge to extracorporeal life support. When used, assessment of quency jet ventilation (HFJV) in children with PARDS. Strong benefit must be undertaken promptly and serially to minimize agreement toxicity and to eliminate continued use without established 6 www.pccmjournal.org 888 s 6OLUME s .UMBER 888 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Feature Article effect. Finally, future study is needed to better define its role, if sedation and to facilitate interprofessional communication. any, in the treatment of PARDS. Strong agreement Strong agreement 4.2 Exogenous Surfactant. 4.2.1 At this time, surfactant 5.1.3 We recommend that sedation monitoring, titration, therapy cannot be recommended as routine therapy in PARDS. and weaning should be managed by a goal-directed protocol Further study should focus on specific patient populations that with daily sedation goals collaboratively established by the may be likely to benefit and specific dosing and delivery regi- interprofessional team. Strong agreement mens. Strong agreement 5.1.4 We recommend that clinical trials in PARDS should 4.3 Prone Positioning. 4.3.1 Prone positioning cannot report their sedation goal, strategy, and exposures. Strong be recommended as routine therapy in PARDS. However, it agreement should be considered an option in cases of severe PARDS. Fur- 5.1.5 We recommend that the reporting of sedation strategy ther pediatric study is warranted, particular study stratifying and monitoring in clinical trials should be adequately explicit on the basis of severity of lung injury. Weak agreement (92% to allow comparison across studies. Strong agreement agreement) 5.1.6 We recommend that when physiologically stable, pedi- 4.4 Suctioning. 4.4.1 We recommend that maintaining a atric patients with PARDS should receive a periodic assessment clear airway is essential to the patient with PARDS. However, of their capacity to resume unassisted breathing (e.g., extuba- endotracheal suctioning must be performed with caution to tion) that is synchronized with sedative titration to an aroused minimize the risk of derecruitment. Strong agreement state. Strong agreement 4.4.2 There are insufficient data to support a recommenda- 5.1.7 We recommend an individualized sedation weaning tion on the use of either an open or closed suctioning system. plan, guided by objective withdrawal scoring and assessment However, in severe PARDS, consideration should be given to of patient tolerance that is developed by the clinical team and the technique of suctioning with careful attention to minimize managed by the bedside nurse. Strong agreement the potential for derecruitment. Strong agreement 5.2 Neuromuscular Blockade. 5.2.1 We recommend that 4.4.3 The routine instillation of isotonic saline prior to if sedation alone is inadequate to achieve effective mechani- endotracheal suctioning is not recommended. However, the cal ventilation, neuromuscular blockade (NMB) should be instillation of isotonic saline prior to endotracheal suctioning considered. When used, pediatric patients with PARDS should may be indicated at times for lavage to remove thick tenacious receive minimal yet effective NMB with sedation to facilitate secretions. Strong agreement their tolerance to mechanical ventilation and to optimize oxy- 4.5 Chest Physiotherapy. 4.5.1 There are insufficient data gen delivery, oxygen consumption, and work of breathing. to recommend chest physiotherapy as a standard of care in the Strong agreement patient with PARDS. Strong agreement 5.2.2 We recommend that when used, NMB should be 4.6 Corticosteroids. 4.6.1 At this time, corticosteroids can- monitored and titrated to the goal depth established by the not be recommended as routine therapy in PARDS. Further interprofessional team. Monitoring may include effective ven- study should focus on specific patient populations that are tilation, clinical movement, and train-of-four response. Strong likely to benefit from corticosteroid therapy and specific dos- agreement ing and delivery regimens. Strong agreement 5.2.3 We recommend that if full chemical paralysis is used, 4.7 Other Ancillary Therapies. 4.7.1 No recommendation the team should consider a daily NMB holiday to allow peri- for the use of the following ancillary treatment is supported: odic assessment of the patient’s level of NMB and sedation. helium-oxygen mixture, inhaled or IV prostaglandins therapy, Strong agreement plasminogen activators, fibrinolytics, or other anticoagulants, 5.2.4 We recommend that clinical trials in PARDS should inhaled β-adrenergic receptor agonists or ipratropium, IV report their NMB goal, strategy, and exposure. Strong agreement N-acetylcysteine for antioxidant effects or intratracheal N-ace- 5.2.5 We recommend that the reporting of NMB strategy tylcysteine for mobilizing secretions, dornase alpha outside of and monitoring in clinical trials should be adequately explicit the cystic fibrosis population, and a cough-assist device. Strong to allow comparison across studies (e.g., type of NMB agent agreement and use of steroids). Strong agreement 4.7.2 No recommendation for the use of stem cell therapy 5.2.6 We recommend that further studies are needed to bet- can be supported. It must be considered experimental therapy ter understand the short- and long-term outcomes of NMB at this point. Strong agreement use. Strong agreement 5.3 Nutrition. 5.3.1 We recommend that pediatric patients Section 5: Nonpulmonary Treatment with PARDS should receive a nutrition plan to facilitate their 5.1 Sedation. 5.1.1 We recommend that pediatric patients with recovery, maintain their growth, and meet their metabolic PARDS should receive minimal yet effective targeted sedation needs. Strong agreement to facilitate their tolerance to mechanical ventilation and to 5.3.2 We recommend that enteral nutrition, when tolerated, optimize oxygen delivery, oxygen consumption, and work of should be used in preference to parenteral nutrition. Strong breathing. Strong agreement agreement 5.1.2 We recommend that valid and reliable pain and 5.3.3 We recommend that enteral nutrition monitoring, sedation scales should be used to monitor, target, and titrate advancement, and maintenance should be managed by a Pediatric Critical Care Medicine www.pccmjournal.org 7 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Pediatric Acute Lung Injury Consensus Conference Group goal-directed protocol that is collaboratively established by the Section 6: Monitoring interprofessional team. Strong agreement 6.1 General Monitoring. 6.1.1 We recommend that all children 5.3.4 We recommend that clinical trials in PARDS should with or at risk of PARDS should receive the minimum clini- report their nutritional/feeding goals, strategy, and exposure. cal monitoring of respiratory frequency, heart rate, continu- Strong agreement ous pulse oximetry, and noninvasive blood pressure. Strong 5.3.5 We recommend that the reporting of the nutrition agreement strategy, exposure, and monitoring in clinical trials should be 6.1.2 We recommend that specific alarms should be avail- adequately explicit to allow comparison across studies (e.g., able when the monitored variables are outside predefined route, composition, calories delivered, use of additives, and ranges. Strong agreement time to reach nutrition goal). Strong agreement 6.1.3 We recommend that some monitored values (e.g., tidal 5.4 Fluid Management. 5.4.1 We recommend that pediatric volume and compliance of the respiratory system) should be patients with PARDS should receive total fluids to maintain interpreted after standardization to body weight. Hence, accu- adequate intravascular volume, end-organ perfusion, and opti- rate weight is critical. Predicted body weight should be used, mal delivery of oxygen. Strong agreement based on calculation from gender and from height or length or 5.4.2 After initial fluid resuscitation and stabilization, we from ulna length. Strong agreement recommend goal-directed fluid management. Fluid balance 6.2 Respiratory System Mechanics. 6.2.1 We recommend should be monitored and titrated to maintain adequate intra- that during invasive ventilation in children with PARDS, the vascular volume while aiming to prevent positive fluid balance. exhaled tidal volume should be continuously monitored to Strong agreement prevent injurious ventilation. Strong agreement 5.4.3 We recommend that fluid titration be managed by a 6.2.2 We recommend that monitoring of ventilatory inspi- goal-directed protocol that includes total fluid intake, output, ratory pressure is important to prevent ventilator-induced and net balance. Strong agreement lung injury. It should be based on peak pressure in pressure- 5.4.4 We recommend that clinical trials in PARDS should regulated modes and plateau pressure during ventilation in report their fluid management goals, strategy, and exposure. volume-control modes. It should be interpreted with caution Strong agreement in patients with suspected abnormal chest wall compliance or 5.4.5 We recommend that the reporting of fluid strategy with spontaneous breathing. Strong agreement and monitoring in clinical trials should be adequately explicit 6.2.3 We recommend the monitoring of flow-time and pres- to allow comparison across studies (e.g., fluid bolus trigger, sure-time curves to assess the accuracy of respiratory timings type of fluid, central venous pressure goal, use of ultrasound, and to detect expiratory flow limitation or patient-ventilator or impedance monitoring). Strong agreement asynchrony. Strong agreement 5.4.6 We recommend that clinical trials in PARDS should 6.2.4 We recommend that in infants and smaller children, use a clinical protocol to guide fluid management. Strong the exhaled tidal volumes should be monitored at the end of agreement the endotracheal tube and/or with appropriate compensation 5.4.7 We recommend that further studies are needed to for circuit compliance. Strong agreement definitively determine the optimal fluid management strategy 6.2.5 There is insufficient evidence to recommend the sys- in pediatric patients with PARDS. Strong agreement tematic monitoring of the following variables of respiratory 5.5 Transfusion. 5.5.1 In clinically stable children with evi- system mechanics: flow-volume loop, static pressure-volume dence of adequate oxygen delivery (excluding cyanotic heart loop, dynamic pressure-volume loop, dynamic compliance and disease, bleeding, and severe hypoxemia), we recommend that resistance, stress index, intrinsic PEEP, esophageal manometry a hemoglobin concentration up to 7.0 g/dL be considered a and transpulmonary pressure, work of breathing, corrected trigger for RBC transfusion in children with PARDS. Strong minute ventilation, functional residual capacity, dead space/ agreement tidal volume ratio, assessment of respiratory muscle activity 5.5.2 We recommend that clinical trials in PARDS should using airway occlusion pressure (P0.1), esophageal pressure rate report their blood product transfusion triggers, strategies, and product, electrical activity of diaphragm, ultrasonography of the exposures. Strong agreement diaphragm, or thoracoabdominal asynchrony quantification by 5.5.3 We recommend that the reporting of transfusion trig- respiratory inductance plethysmography. Weak agreement (92% ger, strategy, and monitoring in clinical trials should be ade- agreement) quately explicit to allow comparison across studies (e.g., whole 6.3 Oxygenation Variables, Severity Scoring, and Co2 vs packed RBCs, age of blood, use of leukoreduction, fresh- Monitoring. 6.3.1 Monitoring of FIO2, SpO2 and/or PaO2, Paw, frozen plasma, and platelets). Strong agreement and PEEP is recommended to detect PARDS, to assess PARDS 5.5.4 We recommend that clinical trials in PARDS should severity, and to guide the management of oxygenation failure. use a clinical protocol to guide blood product transfusion. Strong agreement Strong agreement 6.3.2 We recommend that blood pH and PaCO2 measure- 5.5.5 We recommend that further studies are needed to ment frequency should be adjusted according to PARDS sever- definitely determine the risks and benefits of transfusion in ity, noninvasive monitoring data, and stage of the disease. pediatric patients with PARDS. Strong agreement Strong agreement 8 www.pccmjournal.org 888 s 6OLUME s .UMBER 888 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Feature Article 6.3.3 Peripheral venous blood gas sampling is not recom- 7.1.2 We recommend that selected populations of children, mended. Weak agreement (83% agreement) such as children with immunodeficiency who are at greater risk 6.3.4 Continuous monitoring of CO2 is recommended in of complications from invasive mechanical ventilation, may children with invasive mechanical ventilation, using end-tidal benefit more from earlier NPPV to avoid invasive mechanical CO2/time curves, volumetric capnography, and/or transcutane- ventilation. Weak agreement (80% agreement) ous CO2 measurements. Strong agreement 7.2 Team Training. 7.2.1 We recommend that although 6.4 Specific Weaning Considerations. 6.4.1 We recommend noninvasive, NPPV should be delivered in a setting with at least daily assessment of predefined clinical and physiologic trained experienced staff and where close monitoring is criteria of extubation readiness in order to avoid unnecessary available to rapidly identify and treat deterioration. Strong prolonged ventilation. Strong agreement agreement 6.4.2 We recommend that Spontaneous Breathing Trials 7.3 Noninvasive Support Ventilation Management. 7.3.1 and/or Extubation Readiness Tests should be performed. We recommend that intubation should be considered in Strong agreement patients receiving NPPV who do not show clinical improve- 6.4.3 We recommend that for research studies, Spontaneous ment or have signs and symptoms of worsening disease, Breathing Trials and Extubation Readiness Tests should be including increased respiratory rate, increased work of breath- standardized. Strong agreement ing, worsening gas exchange, or an altered level of conscious- 6.5 Imaging. 6.5.1 We recommend that chest imaging is ness. Strong agreement necessary for the diagnosis of PARDS and to detect complica- 7.3.2 We recommend the use of an oronasal or full facial tions such as air leak or equipment displacement. Frequency of mask to provide the most efficient patient-ventilator syn- chest imaging should be predicated on patient clinical condi- chronization for children with PARDS. Weak agreement (84% tion. Strong agreement agreement) 6.5.2 There is insufficient evidence to recommend the sys- 7.3.3 We recommend that children using NPPV should be tematic use of chest CT scan, lung ultrasonography, and elec- closely monitored for potential problems, such as skin break- trical impedance tomography. Strong agreement down, gastric distention, barotrauma, and conjunctivitis. 6.6 Hemodynamic Monitoring. 6.6.1 Hemodynamic mon- Strong agreement itoring is recommended during PARDS, in particular, to guide 7.3.4 Heated humidification is strongly recommended for volume expansion in the context of fluid restrictive strategy, NPPV in children. Strong agreement to evaluate the impact of ventilation and disease on right and 7.3.5 We recommend that to allow the most efficient left cardiac function, and to assess oxygen delivery. Strong patient-ventilator synchronization and tolerance, sedation agreement should be used only with caution in children receiving NPPV 6.6.2 In patients with suspected cardiac dysfunction, echo- for PARDS. Weak agreement (88% agreement) cardiography is recommended for noninvasive evaluation of 7.3.6 To reduce inspiratory muscle effort and improve both left and right ventricular function, the preload status, and oxygenation, we recommend noninvasive pressure support pulmonary arterial pressures. Strong agreement ventilation combined with PEEP in patients with PARDS. 6.6.3 We recommend that a peripheral arterial catheter Continuous positive airway pressure alone may be suitable for should be considered in patients with severe PARDS for con- those children who are unable to attain patient ventilatory syn- tinuous monitoring of arterial blood pressure and arterial chrony or when using nasal interface. Weak agreement (92% blood gas analysis. Strong agreement agreement) 6.6.4 There is insufficient evidence to recommend the 7.4 Other Modes of Noninvasive Support Ventilation. systematic use of the following hemodynamic monitoring 7.4.1 We recommend that further studies are needed to identify devices: pulse contour with transpulmonary dilution technol- clinical indications for high-flow nasal cannula in patients at risk ogy, pulmonary artery catheters, alternative devices to monitor of PARDS. High-flow nasal cannula has not been demonstrated cardiac output (ultrasonic cardiac output monitoring, trans- to be equivalent to NPPV. Strong agreement esophageal aortic Doppler, and noninvasive monitoring of car- 7.4.2 NPPV is not recommended for children with severe diac output based on changes in respiratory CO2 concentration disease. Strong agreement caused by a brief period of rebreathing), central venous oxy- genation monitoring, and B-type natriuretic peptide measure- Section 8: Extracorporeal Support ments. Strong agreement 8.1 Indications for Extracorporeal Membrane Oxygenation in Children With PARDS. 8.1.1 We recommend that extracorpo- Section 7: Noninvasive Support and Ventilation real membrane oxygenation (ECMO) should be considered to 7.1 Indications for Noninvasive Support Ventilation. 7.1.1 support children with severe PARDS where the cause of the We recommend that noninvasive positive pressure ventilation respiratory failure is believed to be reversible or the child is (NPPV) is considered early in disease in children at risk for likely to be suitable for consideration for lung transplantation. PARDS to improve gas exchange, decrease work of breathing, Strong agreement and potentially avoid complications of invasive ventilation. 8.1.2 It is not possible to apply strict criteria for the selec- Weak agreement (88% agreement) tion of children who will benefit from ECMO in PARDS. We Pediatric Critical Care Medicine www.pccmjournal.org 9 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
Pediatric Acute Lung Injury Consensus Conference Group recommend that children with severe PARDS should be con- pulmonologist for further assessment, treatment, and long- sidered for ECMO when lung protective strategies result in term pulmonary follow-up. Strong agreement inadequate gas exchange. Strong agreement 9.2 Neurocognitive Development. 9.2.1 We recommend 8.1.3 We recommend that decisions to institute ECMO that physical, neurocognitive, emotional, family, and social should be based on a structured evaluation of case history and function be evaluated within 3 months of hospital discharge clinical status. Strong agreement for children who survive moderate-to-severe PARDS. Strong 8.1.4 We recommend that serial evaluation of ECMO eligibil- agreement ity is more useful than single-point assessment. Strong agreement 9.2.2 We recommend that for younger patients (infants and 8.1.5 We recommend that careful consideration of qual- toddlers), additional evaluation of physical, neurocognitive, ity of life and likelihood of benefit should be assessed. Strong emotional, family, and social function should be performed agreement prior to entering school. Strong agreement 8.2 Contraindications to ECMO in Children With Severe 9.2.3 We recommend that when abnormalities are identi- PARDS. 8.2.1 We recommend that ECMO should not be fied, children should be treated or referred for more in-depth deployed in patients in whom life-sustaining measures are assessment and treatment by appropriate subspecialists and likely to be limited. Strong agreement educators (e.g., when learning deficits are identified). Strong 8.3 Team Training and Organization. 8.3.1 We recom- agreement mend that ECMO programs should have clearly defined 9.3 Outcome Measures. 9.3.1 Given decreasing mortal- leadership structure, including administrative support. ity among children with PARDS, we recommend research Strong agreement into the following potential alternative endpoints for clini- 8.3.2 We recommend that all personnel directly caring cal trials: longer term mortality (e.g., 90 d), rates of new or for the patient should have an understanding of the ECMO progressive organ dysfunction, organ failure- or treatment- circuit and the physiologic interactions between it and the free days, ventilator-free days (with and without noninvasive patient. Competencies for physicians with primary patient ventilation), duration of oxygen therapy (or a higher con- care duties and ECMO specialists should be required. Strong centration of oxygen for subjects on chronic supplemental agreement oxygen), risk-adjusted hospital and PICU lengths of stay, 8.3.3 We recommend that all centers providing ECMO hospital and PICU readmissions (e.g., within 30 d of dis- support should belong to The Extracorporeal Life Support charge), quality of life, neurocognitive function, and emo- Organization (ELSO) and report all patient activity to ELSO tional health. Strong agreement or similar organization. Strong agreement 8.3.4 We recommend that ECMO programs should bench- DISCUSSION mark themselves against other programs via the ELSO registry The PALICC was convened to identify and articulate differ- or similar. Strong agreement ences between adult and pediatric ARDS. The conference 8.4 Other Modes of Extracorporeal Lung Support. 8.4.1 We made important first steps in this process. We recognize recommend that patients suffering from extreme hypercarbia that further work is required to build on these initial efforts, and mild-to-moderate hypoxia may benefit from new extracor- and we hope these recommendations provide a roadmap to poreal devices which provide partial respiratory support. Such future areas of investigation. The details of each section along devices may be effective in removing all carbon dioxide and may with the extensive literature researched are presented in the not require a pump to provide blood flow but may instead use supplement to this issue of Pediatric Critical Care Medicine the patient’s own generated systemic blood pressure to drive published with this article. The Conference identified many blood through a low-resistance oxygenator. Weak agreement areas of agreement, but its primary benefit may well be in (63% agreement) illustrating how little is known about this relatively common condition in children. Section 9: Morbidity and Long-Term Outcomes The process by which the recommendations were devel- 9.1 Pulmonary Function. 9.1.1 We recommend screening oped was based on previously published methods (13) and for pulmonary function abnormalities within the first year was chosen due to the relative paucity of data in PARDS. The after discharge, including a minimum of respiratory symp- experts in each group were tasked with synthesizing the data tom questionnaires and pulse oximetry for all children with on their specific topic and developing recommendations PARDS who undergo invasive mechanical ventilation. Strong based on peer-reviewed, pediatric-specific data. If no pediat- agreement ric data were available, experts were directed to use data gen- 9.1.2 We recommend that for all children with PARDS who erated from either adults with ARDS or neonates with lung undergo invasive mechanical ventilation and are of sufficient injury, to solidify their recommendations. Finally, expert developmental age and capabilities, spirometry should also be opinion was used when no data were available. Once the ini- performed for the screening for pulmonary function abnor- tial recommendations were presented, each of the PALICC malities within the first year after discharge. Strong agreement members had equal input on each recommendation. One 9.1.3 We recommend that when deficits in pulmonary func- advantage of the RAND/UCLA appropriateness method tion are identified, patients should be referred to a pediatric is that it diminishes the “leader effect” and provides every 10 www.pccmjournal.org 888 s 6OLUME s .UMBER 888 Copyright © 2015 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. Unauthorized reproduction of this article is prohibited
You can also read