Come ventilare il paziente ipossiemico acuto Andrea Vianello Fisiopatologia e Terapia Intensiva Respiratoria Ospedale – Università di Padova

Non-Invasive Ventilation has demonstrated good efficacy in reducing risk of intubation and mortality; however, some potential etiologies of acute respiratory failure are associated with a high risk of NIV failure, which may lead to great hospital mortality. Walkey AJ. Annals ATS 2013; 10:10-7

Talking about NIV failure: 1.Epidemiology & Aetiology 2.Remedies to avoid failure 3.Transition to IMV: dancing in the dark

Rate of NIV failure is extremely different according to study design, severity of illness and level of monitoring

Overall NIV failure: 16.3% Sixty-two RCTs including a total of 5870 patients

Evaluation of all 449 patients receiving NPPV for a 1-yr period for acute or acute on chronic RF –CPE (n=97) –AECOPD (n=87) –non-COPD acute hypercapnic RF (n=35) –postextubation RF (n=95) –acute hypoxemic RF (n=144) Intubation rate was 18%, 24%, 38%, 40%, and 60%, respectively Hospital mortality for patients with acute hypoxemic RF who failed NPPV was 64% Schettino G. Crit Care Med 2008; 36:441-7 NIV – Real Life

Schettino G, Crit Care Med 2008; 36:441-7

Epidemiology Rationale: evidence supporting use of NIV varies widely for different causes of ARF. Population: 11,659,668 cases of ARF from the Nationwide Inpatient Sample during years 2000 to 2009; Objectives: To compare utilization trends and outcomes associated with NIV in patients with and without COPD.

Acute respiratory failure-associated diagnosis Changing etiologies of ARF among patients receiving NIV

20% of patients without a COPD diagnosis who received NIV IMV; 13.4% of patients with COPD who received NIV IMV; Patients experiencing NIV failure before transition to IMV had greater hospital mortality than those initially on IMV

Rationale: The patterns and outcomes of NIV use in patients hospitalized for AECOPD nationwide are unknown. Population: 7,511,267 admissions for acute AE occurred from 1998 to 2008; Objectives: To determine the prevalence and trends of NIV in AECOPD.

Use of NIPPV or IMV as first-line respiratory support in patients hospitalized with AECOPD

The percentage of patients transitioned from NIV to IMV ≈ 5% and did not increase from 1998 to 2008

Reasons for low rate of IMV use after NPPV, compared to clinical trial: End of life decision to not accept IMV Patients died before IMV could be started Good selection of appropriate patients

High mortality rate (≈30%) ;↑ over time OR for death:1.63, compared to those initially on IMV ↑hospital stay

Reasons for high mortality rate in patients transitioned to IMV Increased use of NIPPV in patients difficult to ventilate? Continuation of NIPPV despite a lack of early improvement?

Aetiology of NIV failure A.Failure to adequately ventilate/oxygenate A.Delayed NIV treatment B.Inappropriate ventilatory technique C.Patient’s clinical condition B. Dependence on non-invasive support Lack of improvement of acute illness C. Complications

NIV failure is predicted by: -Advanced age -High acuity illness on admission (i.e. SAPS-II >34) -Acute respiratory distress syndrome -Community-acquired pneumonia with or without sepsis -Multi-organ system failure NIV trial in hypoxemic RF is justified if patients are carefully selected according to available guidelines, known risk factors and predictors for NIV failure.

NIV failure is predicted by: -Advanced age -High acuity illness on admission (i.e. SAPS-II >34) -Acute respiratory distress syndrome -Community-acquired pneumonia with or without sepsis -Multi-organ system failure

 Retrospective analysis  59 episodes of ARF in 47 COPD patients NIV success: 46 NIV failure: 13  Predictors for NIV failure: Higher PaCO 2 at admission Worse functional condition Reduced treatment compliance Pneumonia NIV in acute COPD: correlates for success Ambrosino N, Thorax 1995;50:755-7

Summary of published studies for idiopathic pulmonary fibrosis patients in the ICU

 Retrospective analysis  60 episodes of ARF in ALS patients managed by IMV via ETI Initially on IMV: 31 Transitioned from NIV to IMV: 29 Dependence on NIV: 7  Predictors for survival: Age at admission Percent survival of patients after tracheostomy, stratified by age group ≥ 60 yrs < 60 yrs Patients with ALS

NIV complications ComplicationIncidence (%) Major Aspiration pneumonia<5 Haemodinamyc collapseInfrequent BarotraumaRare Minor Noise50-10 CO2 rebreathing Discomfort30-50 Claustrophobia5-20 Nasal skin lesions2-50

Pneumotorax associated with long-term non- invasive positive pressure ventilation in Duchenne muscular dystrophy Vianello A, Arcaro, G, Gallan F, Ori C, Bevilacqua M Neuromusc Dis 2004;14:353-55

NIV complications ComplicationIncidence (%) Major Aspiration pneumonia<5 Haemodynamic collapseInfrequent BarotraumaRare Minor Noise CO 2 rebreathing Leak/Discomfort30-50 Claustrophobia5-20 Nasal skin lesions2-50

NIV reasons for failure ACPECOPDNon COPD ALI/ARDS Hypoxemia+++++ Hypercapnia++++ Leak/Mask intol++++ Secretion+++ Mentation++++ Agitation+++ Progression++++ Failure Rate18%24%37%60% Schettino G, Crit Care Med 2008; 36:441-7

Remedies to avoid NIV failure 1.Select patients carefully 2.Assess risk for failure; diagnosis, etc 3.Select comfortable mask 4.Optimize vent settings 5.Facilitate secretion removal 6.Treat agitation 7.Monitor closely in proper location 8.Assess response after 1 to 2 hrs

Remedies to avoid NIV failure 1.Select patients carefully 2.Assess risk for failure; diagnosis, etc 3.Select comfortable mask 4.Optimize vent settings 5.Facilitate secretion removal

Respiratory arrest Inability to tolerate the device, because of claustrophobia, agitation or uncooperativeness Inability to protect the airway, due to swallowing impairment Excessive secretions not sufficiently managed by clearance techniques Recent upper airway surgery NIV should not be used in:

COPD CHF/CPE PNA Asthma OHS NMD UAO post-op post-extub trauma ARDS MOF IPF Tight UAO

NIV fails more frequently for de novo ARF than for acute-on-chronic RF De Moule, Intensive Care Med 2006; 32:

Remedies to avoid NIV failure 1.Select patients carefully 2.Assess risk for failure; diagnosis, etc 3.Select comfortable mask 4.Optimize vent settings 5.Facilitate secretion removal

Mask selection - a crucial issue! Noise (50-100%) CO 2 rebreathing (50-100%) Leak/Discomfort (30-50%) Claustrophobia (5-20%) Nasal skin lesions (2-50%)

Choose correct interface and size! Although there is no difference in ETI or mortality, RCT have reported that nasal mask is less tolerated than oronasal mask Girault, Crit Care Med 2009;37: Cuvelier, Intensive Care Med 2009;35: The use of an oronasal mask is suggested rather than a nasal mask in patients who have ARF. No recommendation about the use of an oronasal mask versus full face mask.

Remedies to avoid NIV failure 1.Select patients carefully 2.Assess risk for failure; diagnosis, etc 3.Select comfortable mask 4.Optimize vent settings 5.Facilitate secretion removal

ICU Ventilator Ventilators Factors influencing preferred ventilator: Personal experience Location Available monitoring Leak compensation Trigger sensitivity Handling Flexibility Alarms Portable Ventilator

The golden rule does not apply to ventilator setting, however: Pressure preset modes reduce the risk of failure and are recommended for COPD decompensated patients. French Guidelines for NIV treatment, 2008 The “lung-protective” strategy may reduce the risk of barotrauma in patients with exacerbated ILD. Fernandez-Perez ER, Chest 2008; 133:1113-9

Remedies to avoid NIV failure 1.Select patients carefully 2.Assess risk for failure; diagnosis, etc 3.Select comfortable mask 4.Optimize vent settings 5.Facilitate secretion removal

Enhancement of secretion clearance Adequate hydration/humidification Manually assisted cough Cough assist Antibiotics Expectorants, mucolytics not of known value

Am J Phys Med Rehabil 2005;84:83-8

Transition to IMV: when is in the interest of a patient? Hospital mortality: 64% (Schettino, 2008) Mortality rate: 30%; prolonged hospitalization (Chandra, 2011) Great hospital mortality (Walkey, 2013)

Transition to IMV (personal experience, ) Number of subjects62 Age (mean ± SD), yrs65.4±19.3 Gender (males, females)26, 36 Ineffective NIV, n (%) Severe hypercapnia Severe hypoxemia 52 (83.8) 25 (42.4) 21 (35.6) Dependence on NIV, n (%)8 (13.3) NIV complication, n (%)2 (3.4) Tracheotomy, n (%)16 (28.8) Outcome, n (%) Died during hosp Discharged from hosp 41 (66.1) 21 (33.9)

Kaplan-Meier function of overall survival Median survival: 46 days (95% CI, 43 to 162)

Kaplan-Meier function of survival according to baseline condition Mean survival: NM/CW = ±36.9 COPD = 53.90±7.3 ILD = 31.13±7.8 ] p= ] p<0.0001

Kaplan-Meier function of survival for dichotomus age (  50 and >50) Median survival:  50 = d (95%CI, 15.0 to n.c.) >50 = 45.0 d (95%CI,24.0 to 54.0) ] p=0.0071

Remarks Mortality rate among patients transitioned to IMV is very high; The outcome of patients with ILD is extremely poor. Should IPF/COPD patients be excluded from IMV after failing a NIV trial?

Conclusions The boundaries for the use of NIV continue to expand, however: Transitioning from NIV to IMV may not be in the interest of some categories of patients. The routine use of NIV in all patients with severe ARF is not yet supported Caution should be used with NIV among patients at high risk of failure