What’s New In Pediatric ARDS Nancy G. Hoover, MD Medical Director, PICU Walter Reed AMC

New and Improved Acute Respiratory Distress Syndrome Ashbaugh, Lancet, 1967 Adult Respiratory Distress Syndrome To distinguish from neonatal HMD/RDS American-European Consensus conference, 1994 Ashbaugh first described the syndrome of severe respiratory failure similar to infant hyline membrane disease

ARDS: New Definition Criteria Acute onset Bilateral CXR infiltrates PA pressure < 18 mm Hg Classification Acute lung injury - PaO2 : F1O2 < 300 Acute respiratory distress syndrome - PaO2 : F1O2 < 200 PCWP higher than 18 mmHg are generally considered to be c/w left-heart failure and may be the cause of cargiogenic pulmonary edema. 1994 American-European Consensus Conference

Clinical Disorders Associated with ARDS Direct Injury Common Causes Pneumonia Gastric aspiration Less Common Causes Pulmonary contusion Fat emboli Near drowning Inhalational injury Indirect Injury Common Causes Sepsis Shock after severe trauma Less Common Causes Cardiopulm. bypass Drug overdose Acute pancreatitis Massive blood transfusions

The Problem: Lung Injury Davis et al., J Peds 1993;123:35 Noninfectious Pneumonia 14% Cardiac Arrest 12% Infectious Pneumonia 28% Hemorrhage 5% Trauma 5% Other 4% Septic Syndrome 32%

ARDS - Pathogenesis Instigation Endothelial injury: increased permeability of alveolar - capillary barrier Epithelial injury : alveolar flood, loss of surfactant, barrier vs. infection Proinflammatory mechanisms

ARDS Pathogenesis Resolution Equally important Alveolar edema - resolved by active sodium transport Alveolar type II cells - re-epithelialize Neutrophil clearance needed

ARDS - Pathophysiology Decreased compliance Alveolar edema Heterogenous “Baby Lungs” Gattinoni “Concept of Baby Lung”, Intensive Care Medicine, 2005: The "baby lung" concept originated as an offspring of computed tomography examinations which showed in most patients with acute lung injury/acute respiratory distress syndrome that the normally aerated tissue has the dimensions of the lung of a 5- to 6-year-old child (300-500 g aerated tissue). DISCUSSION: The respiratory system compliance is linearly related to the "baby lung" dimensions, suggesting that the acute respiratory distress syndrome lung is not "stiff" but instead small, with nearly normal intrinsic elasticity. Initially we taught that the "baby lung" is a distinct anatomical structure, in the nondependent lung regions. However, the density redistribution in prone position shows that the "baby lung" is a functional and not an anatomical concept. This provides a rational for "gentle lung treatment" and a background to explain concepts such as baro- and volutrauma. CONCLUSIONS: From a physiological perspective the "baby lung" helps to understand ventilator-induced lung injury. In this context, what appears dangerous is not the V(T)/kg ratio but instead the V(T)/"baby lung" ratio. The practical message is straightforward: the smaller the "baby lung," the greater is the potential for unsafe mechanical ventilation.

Axial CT of an experimental model of ARDS showing the heterogeneous distribution of lung disease. The gravitationally nondependent lung region (ventral) is relatively spared, whereas the dependent lung region (dorsal) exhibits greater involvement

Phases of ARDS Acute - exudative, inflammatory (0 - 3 days) Subacute - proliferative (4 - 10 days) Chronic - fibrosing alveolitis ( > 10 days)

Phases of ARDS

ARDS - Outcomes Most studies - mortality 40% to 60% Majority of deaths sepsis or MOD rather than primary respiratory Outcomes similar for adults and children Mortality may be decreasing 53/68 % 39/36 %

ARDS - Principles of Therapy Provide adequate gas exchange Avoid secondary injury

It would seem ironic that the very existence of humans is fully dependent on a gas that, in excess quantities, is toxic and lethal Lynn D. Martin

Mechanical Ventilation Therapies for ARDS Mechanical Ventilation Innovations: iNO PLV Proning Surfactant Anti-Inflammatory Gentle ventilation: Permissive hypercapnia Low tidal volume Open-lung HFOV ARDS Extrapulmonary techniques ECMO IVOX IV gas exchange Total Implantable Artificial Lung Extrapulmonary Gas Exchange

The Dangers of Overdistention Repetitive shear stress inflammatory response air trapping Phasic volume swings: volutrauma Injury to normal alveoli

The Dangers of Atelectasis compliance intrapulmonary shunt FiO2 WOB inflammatory response

Lung Injury Zones Overdistention “Sweet Spot” Atelectasis

“Mechanical” Therapies in ARDS Lower tidal volumes but avoidance of atelectasis with higher PEEP Permissive hypercapnia HFOV Prone positioning

Lower Tidal Volumes for ARDS Multi-center trial, 861 adult ARDS Randomized: Tidal volume 12 cc/kg Plateau pressure < 50 cm H2O vs. Tidal volume 6 cc/kg Plateau pressure < 30 cm H2O ARDS Network, NEJM, 342: 2000

Lower Tidal Volumes for ARDS 22% decrease * * 40% vs. 31% mortality Vent-free days in the first 28 days was significantly higher in the low tidal volume group (12 +/- 11 vs. 10 +/-11;p=0.007 ARDS Network, NEJM, 342: 2000 * p < .001

Ventilator Goals Set the PEEP slightly higher than the lower inflection point Lower tidal volume (generally < 6 mL/kg) Static peak pressure <40 cm H20 Wean oxygen to <60% In Stewart’s study, If peak pressure is <30 and tidal volume is <8 mL/kg, then there was no difference in mortality.

Permissive Hypercapnia Defined: presence of hypercapnia in the setting of a mechanically ventilated patient receiving limited inspiratory pressures and reduced tidal volumes Hickling, Intensive Care Med 1990 Hickling, Int Care Med, 1990

Physiologic Effects of Hypercapnia RESP: Net effect is improvement in oxygenation by enhancing hypoxic pulmonary vasoconstriction and decreases intrapulmonary shunting Right-shift of oxygen-hemoglobin dissociation curve When hypercapnia is produced through the limitation of tidal volumes and inspiratory airway pressures without adequate PEEP, the Qs/Qt ratio increases secondary to progressive derecruitment of alveolar units. Under such conditions, oxygenation will be further impaired by the hypercapnia-induced increase in cardiac output. This increase in cardiac output induces a worsening Qs/Qt ratio as blood flow increases preferentially to the gravitationally dependent, poorly ventilated lung regions and results in additional intrapulmonary shunting The increase in Qs/Qt can frequently be counteracted through the optimization of lung volume by means of recruitment maneuvers and application of suitable levels of PEEP. Hypercapnic acidosis enhances hypoxic pulmonary vasoconstriction, thus improving ventilation-perfusion matching, decreasing Qs/Qt, and increasing the PaO2. Hypercapnic acidosis positively affects oxygen availability to tissues by promoting a right shift in the oxygen–hemoglobin dissociation curve. Net effect is improvement in oxygenation

Physiologic Effects of Hypercapnia CV: Net effect is often hemodynamic compromise Sympathetic stimulation with increased C.O. Increased HR and SV, decreased SVR Intracellular acidosis of cardiomyocyte is reversible when due to hypercarbia compared to metabolic acidosis When combined with high PEEP strategy, can lead to severely decreased preload and cardiovascular compromise Hypercapnic acidosis has been shown to cause marked sympathetic stimulation with predictable increases in cardiac output due to the augmentation of heart rate and stroke volume secondary to decreased systemic vascular resistance

Physiologic Effects of Hypercapnia RENAL: Compensatory bicarb reabsorption Acidosis leads to direct renal vasoconstriction Sympathetic-meditated release of norepinephrine (NE) Indirectly, hypercapnia causes a decrease in SVR that in turn releases NE, stimulates the renin-angiotensin-aldosterone system, leading to a further decrease in renal blood flow Within hours of the onset of sustained hypercapnic acidemia, the kidneys initiate compensatory net reabsorption of sodium bicarbonate, generally returning blood pH to physiologic levels within 2 days.

Permissive Hypercapnia Is it worth it? Early adult ARDS trial showed a reduction in expected mortality of 56% to an actual mortality of 26% Included in adult trauma patients protocol for mechanical ventilation Several pediatric studies showing benefit when used in conjunction with low TV and high PEEP Caution in patients with elevated ICP Hickling, CCM, 1994 Nathens, J Trauma, 2005 Sheridan, J Trauma, 1995 Paulson, J Pediatr, 1996

High Frequency Oscillation: A Whole Lotta Shakin’ Goin’ On

It’s not absolute pressure, but volume or pressure swings that promote lung injury or atelectasis. Reese Clark

High Frequency Ventilation Rapid rate Low tidal volume Maintain open lung Minimal volume swings

Differences Between CMV and HFOV

HFOV vs. CMV in Pediatric Respiratory Failure: Results Greater survival without severe lung disease Greater crossover to HFOV and improvement Failure to respond to HFOV strong predictor of death Arnold et al, CCM, 1994

HFOV vs. CMV in Pediatric Respiratory Failure * -Arnold et al, CCM, 1994

HFOV: Outcomes of Randomized Controlled Trials Reduces cost, severity of chronic lung disease and decreases airleak in neonatal RDS Decreases need for ECMO in eligible neonates Improves survival without CLD in pediatric ARDS

Indications for HFOV Severe persistent airleak Neonatal: HMD (*) Pneumonia Meconium aspiration Lung hypoplasia Acute respiratory distress syndrome

Is turning the ARDS patient “prone” helpful?

Prone Positioning in ARDS Theory: let gravity improve matching perfusion to well-ventilated lung Improvement is immediate Decreased shunt: improved PaO2 but variable (75%) Uncertain effect on outcome

Effect of prone position on ventilation distribution Effect of prone position on ventilation distribution. In the supine position, the distribution of ventilation is preferentially distributed to the ventral regions. When the patient is prone, the stiffness of the dorsal chest wall favors the distribution of ventilation to the dorsal regions, facilitating reinflation in this area.

Prone Positioning in Adult ARDS Randomized trial Standard therapy vs. standard + prone positioning Improved oxygenation No difference in mortality, time on ventilator No difference in complications Post-hoc analysis showed an improvement in 10-day mortality in the prone group but overall ventilator-free days, ICU discharge and length of hospitalization was unchanged Gattinoni et al., NEJM, 2001

Conflicting Evidence for Proning? Mancebo, Am J of Resp & CCM, 2006 136 adults, randomized to 20 h/day proning within 48h of intubation for severe ARDS Same ventilator treatment protocols in both groups 25 % relative reduction in ICU mortality Curley, JAMA, 2005 Shorter proning times and multiple protocols for vent mgt with lung-protective stragegy and weaning, sedation, nutrition, etc Only 8% mortality and no benefit from prone positioning In the study by Curely et al, you probably won’t see a benefit because with all of the other therapies, the mortality was so low, it would take a huge number of patients to show a mortality difference.

Pharmacological Therapies in ARDS Surfactant iNO Steroids Partial Liquid Ventilation

Surfactant in ARDS ARDS: surfactant deficiency surfactant present is dysfunctional Surfactant replacement improves physiologic function

Calf’s Lung Surfactant Extract in Acute Pediatric Respiratory Failure Multicenter trial-uncontrolled, observational Calf lung surfactant (Infasurf) - intratracheal Immediate improvement and weaning in 24/29 children with ARDS and 14% mortality In several other studies, there is no evidence for sustained benefit from Surfactant administration Wilson et al, CCM, 24:1996 Wilson et al, JAMA, 2005

Steroids in ARDS Theoretical anti-inflammatory, anti-fibrotic benefit Previous randomized studies Acute use (1st 5 days) No benefit Increased 2 infection

Effects of Prolonged Steroids in Unresolving ARDS Randomized, double-blind, placebo-controlled trial Adult ARDS ventilated for > 7 days without improvement Randomized: Placebo Methylprednisolone 2 mg/kg/day x 4 days, tapered over 1 month Meduri et al, JAMA, 1998

Steroids in Unresolving ARDS By day 10, steroids improved: PaO2/FiO2 ratios Lung injury/MOD scores Static lung compliance Steroids decreased procollagen metabolites 24 patients enrolled; study stopped due to survival difference Meduri et al, JAMA, 1998

Steroids in Unresolving ARDS * * * p<.01

What about after first 28 days? NHLBI ARDS Clinical Trials Network, NEJM, 2006 180 adult patients with ARDS >7 days No difference in mortality with steroids EXCEPT, if the patient was entered into the study after 14 days of ARDS THEN, there was an increase in 60 and 180 day mortality

Inhaled Nitric Oxide in Respiratory Failure Neonates Beneficial in term neonates with PPHN Decreased need for ECMO Adults/Pediatrics Benefits - lowers PA pressures, improves gas exchange Randomized trials: No difference in mortality or days of ventilation

ECMO and NO in Neonates ECMO improves survival in neonates with PPHN (UK study) iNO decreases need for ECMO in neonates with PPHN: 64% vs 38% Clark et al, NEJM, 2000

Effects of Inhaled Nitric Oxide In Children with AHRF Randomized, controlled, blinded multi-center trial 108 children, median age 2.5 years Entry: OI > 15 x 2 Randomized: Inhaled NO 10 ppm vs. mechanical ventilation alone Dobyns, et al., J. Peds, 1999

Inhaled NO and HFOV In Pediatric ARDS Dobyns et al., J Peds, 2000

Partial Liquid Ventilation Mechanisms of action oxygen reservoir recruitment of lung volume alveolar lavage redistribution of blood flow anti-inflammatory

Liquid Ventilation Pediatric trials started in 1996 Adult study 2001 Partial: FRC (15 - 20 cc/kg) Study halted 1999 due to lack of benefit Adult study 2001 no effect on outcome

ARDS- “Mechanical” Therapies Low tidal volumes Outcome benefit in large study Prone positioning Unproven outcome benefit Open-lung strategy Outcome benefit in small study HFOV Outcome benefit in small study ECMO Proven in neonates unproven in children

Pharmacologic Approaches to ARDS: Randomized Trials Steroids - acute no benefit - fibrosing alveolitis lowered mortality, small study Surfactant possible benefit in children Inhaled NO no benefit PLV no benefit

“…We must discard the old approach and continue to search for ways to improve mechanical ventilation. In the meantime, there is no substitute for the clinician standing by the ventilator…” Martin J. Tobin, MD

If you think about ECMO, it is worth a call to consider ECMO

Pediatric ECMO Potential candidates Neonate - 18 years Reversible disease process Severe respiratory/cardiac failure < 10 days mechanical ventilation Acute, life-threatening deterioration

Impact of ECMO on Survival in Pediatric Respiratory Failure Retrospective, multicenter cohort analysis 331 patients, 32 hospitals Use of ECMO associated with survival (p < .001) 53 diagnosis and risk-matched pairs: ECMO decreased mortality (26% vs 47%, p < .01) -Green et al, CCM, 24:1996

Impact of ECMO on Survival in Pediatric Respiratory Failure % Mortality mortality risk quartile Green et al, CCM, 1996 p<0.05