Non-invasive ventilation –

setting up a service

Andrew Bentley
Critical Care & Chest Medicine
North Manchester General Hospital
Setting up an acute non-invasive
ventilation service

Why?

Who? Where?
NIV

How? When?
 Acute Non-invasive ventilation

Why? Mortality

Need Length
for NIV of
intubation stay

Cost
effectiveness

Cost effectiveness of ward based NIV for acute exacerbations

Of COPD: economic analysis of randomised controlled trial.
PK Plant, JL Owen, S Parrott, MW Elliott. BMJ 2003;326:956-959
 Respiratory failure in acute COPD
- role of non-invasive ventilation

Improvement in gas
exchange

Reduced work of
Lung hyperinflation
Increase in elastic breathing
-respiratory muscles
at maximum capacity
and resistive forces
Unloading of inspiratory
muscles

In-hospital mortality
is 20-40% despite
Tissue acidosis selective use of
worsens mechanical ventilation
respiratory
muscle function
 Bott et al, Lancet
1993;341:1555-1557
Entry criteria Exclusion Main findings Comments
criteria (standard vs
NIPPV)
PaO2<7.5kPa Severe non- Reduced No objective
respiratory mortality criteria for
PaCO2>6kPa disease (30%v4% intubation
p=0.014)
Age <80years Prior NIPPV Non-validated
Reduced dyspnoea
N=60 breathlessness questionnaires

3 centres
 Brochard et al, Lancet
1995;333:817-822
Inclusion criteria Exclusion criteria Main findings Comment

Increased dyspnoea RR <12/min Reduced mortality Carefully selected

for <2 weeks and: (29% v 9%, p=0.02) population

PaO2 <6kPa Need for immediate Reduced intubation Detailed criteria for
intubation (74% v 26%, intubation
p<0.001)
pH <7.35 Asthma, heart failure, Reduced hospital stay Sealed envelopes
pneumothorax (35 days v 23 days, used for treatment
p=0.02) allocation
RR >30/min Sepsis, post-operative Reduced life
threatening
complications (48% v
16%, p=0.001)
N=85
 Brochard et al, NEJM
1995;333:817-822
100 100
P<0.05 90
80
P<0.05 P<0.05
80
60 PaO2 (mmHg)
70
40 PaCO2 (mmHg)
60
P<0.05
20 50
n=32
0 40
Base line 1 hour 3 hour 12 hour
 Brochard et al, NEJM
1995;333:817-822

15
No. of patients

10
Standard treatment
5 (n=30)
Non-invasive
ventilation (n=39)
0
ys

ys
ys

ys

ys
da

da
da

da

da
14

8
7

>2
1-

-2

-2
8-

15

22

Hospital stay
 Brochard et al, NEJM
1995;333:817-822

20
No. of patients

15
Standard treatment
10

5 Non-invasive
ventilation
0
ys

ys

ys

>5 s
ys
hr
hr

hr

y
da

da

da

da

da
12
1

Endotracheal intubation
 Kramer et al, Am J Resp Crit
Care Med 1995;151:1799-1806
Entry criteria Exclusion criteria Main findings Comment

Moderate to severe Need for Reduced intubation Non-validated

dyspnoea and: immediate (73% v31%, dyspnoea
intubation p<0.05) questionnaires
PaCO2 >6kPa Inability to co- Reduced Lower mortality
operate or fit mask breathlessness

pH <7.35 Inability to clear Reduced RR rate at Intensive care

secretions 1 hour setting

RR >24/min Uncontrolled
arrhythmias,
systolic BP
<90mmHg
N=31
 Acute Non-invasive ventilation

Who?
Acute
hypercapnic
COPD

? Pneumonia

NIV

Heart
? failure
 Non-invasive ventilation – patient
groups

COPD
Acute Respiratory failure

Reduced mortality
Acute pneumonia & ARDS

Post surgery

Reduced morbidity related
Solid organ transplant
to endotracheal intubation
Immunosuppressed with

Reduced ICU admissions pulmonary infiltrates

Reduced hospital length of
Haematological
stay malignancy
Antonelli et al. NEJM 1998;339:429-
1998;339:429-
435
Antonelli et al. JAMA 2000;283:235-
2000;283:235-
241
Hilbert et al. NEJM 2001;344:481-
2001;344:481-487
Confalonieri et al.
AmJRespCritCareMed1999;160:1585-
AmJRespCritCareMed1999;160:1585-
1591
 Acute non-invasive Ventilation in
COPD -
predictors of poor outcome

Low pH

Pneumonia (consolidation) on CXR

Low body weight

Bronchiectasis (excessive secretions)

Poor neurological status
Ambrosino et al, Thorax 1995;50:755-757
Simonds et al, Thorax 1995;50:595-596
 Acute Non-invasive ventilation

How?
Equipment

Training NIV Protocols

Monitoring
 Non-invasive ventilation
at NMGH

1996 – Medical HDU

Sullivan ST VPAPs

Non-invasive monitoring

Entrained supplemental oxygen via mask

Respiratory physio led service

1999 – 12 bedded medical & surgical HDU

Vision BiPAPs

Invasive monitoring

Nurse led service

Protocol driven (for acute hypercapnic COPD)

Automatic referral to chest consultant
 Non-invasive modalities of positive
pressure ventilation in acute
exacerbations of COPD

Non-invasive pressure
IPAP v IPAP +EPAP v CPAP
support ventilation (NPSV) v volume cycled NIPPV
vs NIPPV (assist-control)
No difference between

Success rate (NPSV 87.5%; Pressure support, CPAP &
NIPPV 77%) volume cycled NIPPV

Compliance score (NPSV 4 vs
No advantage conferred by
NIPPV 3, p<0.02)
EPAP
Vitacca et al, Int Care Med 1993;19:450-455 Meecham-Jones et al , Thorax 1994;49:1222-1224

Reduced work of breathing

assist control>NPSV

Patient comfort NPSV>assist
control
Girault et al, Chest 1997;111:1639-1648
 Non-invasive ventilation at NMGH

Documented resuscitation and ICU
Exclusion criteria

admission status.
Hypotension

Medical treatment:
Primary metabolic acidosis
Controlled oxygen therapy
Untreated pneumothorax
Nebulised bronchodilators
Compromised airway
Antibiotics
IV aminophylline
Systemic corticosteroids
NIPPV to be considered if:

No improvement in oxygenation

Inclusion criteria and the same or deteriorating pH

pH <7.36 after 2 hours of medical therapy.

pCO2 > 45 mmHg
Improvement in oxygenation but

pO2 < 60 mmHg same or worsening pH after 2
hours of medical therapy.

Obvious clinical deterioration.
 Acute Non-invasive ventilation

Standard medical
NIV
treatment
BiPAP through face mask

Controlled oxygen (SaO2 or nasal mask
85-90%)
IPAP 10cm H2O, increased

Nebulised salbutamol 5mg to 20 cm H2O
every 4-6 hours
EPAP 5 cm H2O

Nebulised ipratroprium
Target duration first day 24
500µg 6 hourly hours, second day 16

Prednisolone 30mg daily hours, third day 8 hours,
for minimum of 5 days fourth day discontinued

Antibiotic agent
Oxygen in circuit to
maintain SaO2 85-90%
Plant PK, Owen JL , Elliott MW. Early use of NIV for acute exacerbations
exacerbations
Of COPD on general respiratory wards: a multicentre randomised controlled
trial. Lancet 2000;355:1931-
2000;355:1931-1935
Acute Non-invasive ventilation

Training

On the job, self
directed, protocol
driven

Locally organised
sessions & study days

Company organised

Courses – national &
international
Eg. ERS School courses 2004: NIPPV June 10-12th
Pisa, Italy
 Acute Non-invasive ventilation

Training requirements

Understanding rationale for assisted ventilation

Mask & headgear assembly

Ventilator circuit assembly

Theory of operation & adjusting ventilation to desired
outcome

Cleaning & general maintenance

Problem solving, recognise serious situations and act
accordingly

General overall acceptance that technique works
Acute non-invasive ventilation

Monitoring

Pulse oximetry

NIBP

Peripheral venous
access

Arterial blood gas
sampling

ECG

Capillary gases

Arterial lines
 Acute Non-invasive ventilation

When?
Early

Demand Protocols

NIV

Later Appropriateness
 Acute Non-invasive ventilation

Where?
A&E
/MEU

HDU NIV ICU

Ward
 Acute hypercapnic exacerbations of
COPD in A&E

Little advantage of NIV over conventional therapy
Barbe et al. EurRespJ 1996;9:1240-1245
Wood et al. Chest 1998;113:1339-1346

1 year prevalence study of acute COPD

exacerbations in Leeds A&E departments (n=954)

25% acidotic on arrival to A&E and 25% of these

had corrected pH on arrival to ward

Relationship between PaO2 on arrival and

presence of respiratory acidosis
Plant et al. Thorax 2000;55:550-554
 Non-invasive ventilation -
Location of provision of service

YONIV study (Plant et al, Lancet 2000;355:1931-
1935)

NIV can be applied successfully outside of ICU/HDU setting

Outcome not as good as in HDU setting if pH<7.30

Outside of ICU cost efficacy related to prevention of ICU
admission

Training, patient throughput, skill retention –
single location (Doherty et al, Thorax
1998;53:863-866)

1998 – acute NIV service (48% hospitals)

Ward (40%), HDU (12%), ICU (13%)

Acute Respiratory Care Units

NHS Modernisation Agency (Critical Care Programme) weaning &
long term ventilation (April 2002)
 Acute non-invasive ventilation

Where - factors to consider

Location of staff with training & expertise

Adequate staff available over 24 hour period

Rapid access to endotracheal intubation and

invasive mechanical ventilation

Severity of respiratory failure and liklihood of
success

Facilities for monitoring
Non-invasive ventilation on HDU at
NMGH

Audit of practice 1999

“Uncontrolled “ oxygen therapy prior to arrival in
A+E.

Poor documentation

High mortality despite treatment (45%)

Low pH on admission (mean pH <7.20)

Multiple comorbid factors as predictors of poor

outcome
Non-invasive ventilation on HDU at
NMGH

Audit March 2000 –March 2001
• Appropriate for NIPPV n = 69

• NIPPV instituted n = 43 (62%)

• NIPPV not instituted n = 26 (38%)

Recovered with medical therapy n = 14 (20%)
Admitted to ICU n = 1 (1.5%)
Contraindication to NIPPV n = 2 (3%)
No documentation / unclear n = 9 (13%)
Non-invasive ventilation on HDU at
Findings:
NMGH
NIPPV instituted 43
Resuscitation state documented 15 (35%)
Maximal medical treatment 26 (60%)
2nd blood gas not documented 11 (25%)

Documentation of termination of NIPPV:

weaned 1
not tolerated 9
hypotension 2
ICU 1
Unclear 30
Outcome of NIPPV:
- Survived with no re-admission to date: 14 (33%)
- Re-admission within study time period: 9 (21%)
- Death same admission: 20 (46%)
- HDU / Ward 19
- ICU 1
Interrogating KSM, James Wheeler,March 5th 2003
“The Washington Times” recently published a
method for the efficient interrogation of Al
Quaeda suspect Khalid Shaikh Mohammed,
suggested by the president of the Freedom
Research Foundation. This involved ventilation
by nasal mask of a paralysed subject, with the
ventilator turned off to provide transient
suffocation whenever the interrogator was
dissatisfied.”
Summerfield D. BMJ 2003;326:773-
2003;326:773-774