Laurent Brochard

NIV in the ICU: Lessons learnt in the last

20 years

The Engström ventilator (1951)

Ventilatory problems in thoracic anesthesia; VO Björk, CG Engström et al J Thoracic Surg. 1956; 31:117

Sadoul, Bull.physio-pathol.Resp. 1965

Traitement par ventilation instrumentale de 100 cas d'insuffisance respiratoire aigue severe (pACO2> 70) chez des pulmonaires chroniques

Sadoul, Bull.physio-pathol.Resp. 1965

NIV and 2y mortality

•Mask ventilation difficult

•Need for respiratory stimulant

•Tracheal ventilation considered

much easier

1960. Pr A.B. Tonnel

Chest 1990

Chest 1994

Brochard L ARRD 1987

Brochard L et al. ARRD 1989; 139:513-521

NIV and exacerbations of NIV and exacerbations of COPDCOPDEffects on arterial blood Effects on arterial blood gasesgases

Brochard et al NEJM 1990, 323, 1523

Brochard et al. 1995Brochard et al. 1995

NIV and exacerbations

of COPD

Intubation Hosp Stay Mortality

NIV - PS 20 ( n=43 ) 26 % 23 ± 17 j 9 %

O2 ( n=42 ) 74 % 35 ± 33 j 29 %

NIV and exacerbations of NIV and exacerbations of COPDCOPD

Brochard et al NEJM 1995; 333, 817-22

Respiratory Rate Endotracheal Intubation Mortality Rate

Study name Odds ratio and 95% CI

Odds ratio p-Value

Mehta_1997 1,083 0,957

Park_2001 8,077 0,193

Cross_2003 4,250 0,206

Bellone_2004 0,524 0,608

Park_2004 1,000 1,000

Crane_2004 1,000 1,000

Bellone_2005 0,471 0,554

Ferrari_2006 0,135 0,189

Moritz_2007 0,414 0,477

Ferrari_2007 0,297 0,464

Rusterholtz_2008 0,640 0,565

Gray_2009 0,255 0,223

Ghannem_2009 1,282 0,725

Nouira_2010 0,639 0,499

0,776 0,371

0,01 0,1 1 10 100

CPAP Better BIPAP Better

CPAP vs BIPAP in CPE: Overall

No sedationNo sedationNo central lineNo central line

No urinary catheterNo urinary catheter

No endotracheal tubeNo endotracheal tube

NIV & Nosocomial infectionsNIV & Nosocomial infections

0

20

40

60

80

100

%

Inf Noso Pn Noso AB

VNIVM

p<0.0001p<0.0001

p=0.02p=0.02

p<0.001p<0.001

Girou et al. JAMA 2000Girou et al. JAMA 2000

Ferrer et al. AJRCCM 2003; 168: 1438

n = 51

n = 54

CCM 2007

NIV in hypoxemic or hypercapnic patients:

Is there a different impact?

Demoule A et al ICM 2006

Mortality

Reintubation 48%

Postextubation ARF

Reintubation 72%

Reintubation 8%

Reintubation 11%

Prevention of postextubation ARF

2004 Incidence study

353 ICU 26 countries Esteban et al, AJRCCM 2008

1998Incidence study

361 ICU 20 countries Esteban et al, JAMA 2002

2010 Incidence study

553 ICU 39 countries Esteban et al

Use of Mechanical ventilation

11stst, 2, 2dd, 3, 3rdrd International Studies of Mechanical Ventilation International Studies of Mechanical Ventilation

Countries 37ICUs 1,267Patients 18,321

19981998 20042004 20102010

n = 5183n = 5183 n = 4986n = 4986 n = 8313n = 8313

Age, mean (SD)Age, mean (SD) 59 (17)59 (17) 59 /1759 /17 61/1761/17

Gender, female (%)Gender, female (%) 39 %39 % 40 %40 % 38 %38 %

SAPS II, mean (SD)SAPS II, mean (SD) 44 (17)44 (17) 42 (18)42 (18) 45 (18)45 (18)

BMI. mean (SD)BMI. mean (SD) 27 (6)27 (6)

Baseline Demographics

NIV as first attempt

0

10

20

30

40

50

COPD CHF ARF

1998 2004 2010

Use of NIV 1998 - 2010

SRLF - 2012

Results of the 3rd NIV incidence study (France and Belgium)

A Demoule, S Jaber, A Kouatchet, J Lambert, F Meziani, S Perbet, L Camous, R Janssen-Langenstein, M Alves, B Zuber, F Collet, J Messika,

X Favre, O Guisset, B Misset, A Lafabrie, L Brochard, E Azoulay

For the oVNI group

Methods

• Prospective longitudinal study

• 54 ICUs

• 2 months (Nov 2010- April 2011)

• 2653 patients included

• 2445 invasive ventilatory assistance or NIV

Survey 2010 - 2011

0

20

40

60

80

100

All NIV NIV init NIV post

1997

2002

2011

Results 1 IncidenceNIV 23 vs.31% - pre 23% - post-extub 8%

**

0

20

40

60

80

100

exac CRF CPE de novo

1997

2002

2011

*

$

Results 2 IndicationsCPE: 53 vs.37% - « de novo » : 16 vs. 22%

*

*

0

20

40

60

80

100

1997

2002

2011

Results 3 successoverall success rate : 68 vs. 56%

$

Summary

• The global NIV rate is stable– Increase for post-extubation

• Variations across indications– Increase for CPE– Stable in exacerbations of CRF– Decrease in « de novo » ARF

• Success rate is increasing

Compared to 2002

Crit Care Med 2007;35:932–939

2653 patients required mechanical ventilation in 54 ICUs

1450 patients received mechanicalventilation for acute respiratory failure

780 patients received noninvasivemechanical ventilation (NIV)

72 (9.2%) patients received palliative NIV

(comfort care, SCCM class 3)

All died before day-90

708 patients received curative NIV

574 (81%) patients had no treatment 134 (19%) patients had alimitation decisions Do-Not-Intubate status

ICU survivors 529 (92.2%) 97 (72.4%)

Hospital survivors 505 (88%) 75 (56%)

D-90 survivors 474 (82.6%) 53 (39.6%)

Patients interviewed on D-90 120/237 (50.6%) 34/53 (64.2%)

Relatives interviewed on D-90 122/264 (42.5%) 53/134 (39.6%)

Azoulay E et al ICM 2012

P<0.0001Mortality (%)

0

10

20

30

40

50

60

ChronicRespiratoryInsufficiency

Heart FailurePneumonia

Cancer

Azoulay E et al ICM 2012

0

10

20

30

40

50

60

No Treatment-limitationDecisions

Do-Not-IntubateDecisions

SF-36 Physical Pre-ICU Pre-ICUPost-ICU Post-ICU

0

10

20

30

40

50

60

No Treatment-limitationDecisions

Do-Not-IntubateDecisions

SF-36 Mental

Pre-ICU Pre-ICUPost-ICU Post-ICU

Azoulay E et al ICM 2012

0

10

20

30

40

50

60

No Treatment-limitationDecisions

Do-Not-IntubateDecisions

SF-36 Physical Pre-ICU Pre-ICUPost-ICU Post-ICU

0

10

20

30

40

50

60

No Treatment-limitationDecisions

Do-Not-IntubateDecisions

SF-36 Mental

Pre-ICU Pre-ICUPost-ICU Post-ICU

Azoulay E et al ICM 2012

0

5

10

15

20

25

30

35

40

45

D-90 IES Scale

D-90 HADS Anxiety Subscale

D-90 HADS Depression

Subscale

0

5

10

15

20

25

30

35

40

45

D-90 IES Scale

D-90 HADS Anxiety Subscale

D-90 HADS Depression

Subscale

P=0.33 P=0.49 P=0.06

P=0.13 P=0.16 P=0.81

Patients

Relatives

Patients

Azoulay E et al ICM 2012

0

5

10

15

20

25

30

35

40

45

D-90 IES Scale

D-90 HADS Anxiety Subscale

D-90 HADS Depression

Subscale

0

5

10

15

20

25

30

35

40

45

D-90 IES Scale

D-90 HADS Anxiety Subscale

D-90 HADS Depression

Subscale

P=0.33 P=0.49 P=0.06

P=0.13 P=0.16 P=0.81

Patients

RelativesRelatives

Azoulay E et al ICM 2012

Non invasive ventilation for acute respiratory failure

• Reduces the complications associated with mechanical ventilation

• Improves the outcome of COPD exacerbations and cardiogenic pulmonary edema

• A careful use in hypoxemic respiratory failure may be interesting

• NIV represents an interesting therapy when intensity of therapy is limited

Merci! Thank you!

Blog

• http://mechanicalventilation.wordpress.com

19981998 20042004 20102010

NIV rate (%)NIV rate (%) 4.4 4.4 (0.001)(0.001) 8.78.7 14.6%14.6%

SAPS IISAPS II 39.5 (16)39.5 (16) 39.0 (15)39.0 (15) 38.5 (15)38.5 (15)

Failure (intubation) (%)Failure (intubation) (%) 3131 (0.39)(0.39) 37 %37 % 30.5%30.5%

Mortality in NIV failureMortality in NIV failure 47%47% (0.94)(0.94) 48%48% 49%49%

Mortality in NIV successMortality in NIV success 21%21% (0.36)(0.36) 15%15% 9 %9 %

Non-Invasive Ventilation

NIVNIV POSTEXTUBATION POSTEXTUBATION 11.1 %11.1 %

FOR PREVENTIONFOR PREVENTION 5.9 %5.9 %

FOR TREATMENTFOR TREATMENT 5.2 %5.2 %

19981998 20042004 20102010

n = 5183n = 5183 n = 4986n = 4986 n = 8576n = 8576

Main Reason for M.V:Main Reason for M.V:

COPDCOPD 10 %10 % 5 %5 % 6 %6 %

AsthmaAsthma 1 %1 % 1 %1 % 1 %1 %

ComaComa 17 %17 % 19 %19 % 19 %19 %

Neuromuscular diseaseNeuromuscular disease 2 %2 % 1 %1 % 1 %1 %

ARFARF 68.5 %68.5 %

ARDSARDS 4.5 %4.5 % 5 %5 % 3 %3 %

PostoperativePostoperative 21 %21 % 21 %21 % 21 %21 %

CHFCHF 10 %10 % 6 %6 % 8 %8 %

PneumoniaPneumonia 14 %14 % 11 %11 % 10 %10 %

SepsisSepsis 9 %9 % 9 %9 % 9 %9 %

TraumaTrauma 8 %8 % 6%6% 4.5 %4.5 %

COPD 10 % 5 % 6 %

ARDS 4.5 % 5 % 3 %

Oxygen (15) CPAP (15)

P/F 169 (71-240) 167 (76-270)

RR 30 (18-44) 2 (20-60)

BAL(ml) 66 15 43 19 p<0.03

ETI 4 0 p<0.03

NIV 1 0

0

100

200

300

400

500

Baseline Bacou Koo Oracle Airvie

PT

Pes

(cm

H 2

O*s

/min

)

CCM 2009

Rouzé A et al Mémoire M2 Univ Paris Est

Secular Trends in Mortality and Use of Secular Trends in Mortality and Use of NIVNIV

in COPD and Severe CPEin COPD and Severe CPE

0

10

20

30

40

50

60

70

80

90

100

1994(n=40)

1995(n=53)

1996(n=66)

1997(n=64)

1998(n=70)

1999(n=63)

2000(n=58)

2001(n=72)

Year

Per

cent

age

of p

atie

nts

DC

NIV

P<0.0001

P=0.012

Girou et al. JAMA 2003