Essential respiratory medicine. Shanthi Paramothayan

Chapter 13. Respiratory failure

Learning objectives

 To understand the definition of respiratory failure

 To understand the physiology of respiratory failure

 To be able to interpret arterial blood gas results

 To be able to calculate the alveolar-arterial oxygen gradient

 To understand the difference between type 1 and type 2 respiratory failure

 To understand the causes, diagnosis, and management of acute type 1 respiratory failure

 To understand the causes, diagnosis, and management of acute type 2 respiratory failure

 To appreciate the management of chronic type 2 respiratory failure

 To understand how oxygen should be safely prescribed, delivered, and monitored

 To understand the indications and contraindications for non-invasive ventilation

 To appreciate the ethical dilemmas in the management of patients with type 2 respiratory failure


ABG arterial blood gas

APACHE acute physiology and chronic health evaluation

ARDS adult respiratory distress syndrome

BiPAP Bi-level positive airway pressure

BPM breaths per minute

BTS British Thoracic Society

CO2 carbon dioxide

COPD chronic obstructive pulmonary disease

CTPA computed tomography pulmonary angiogram

ECG electrocardiogram

EPAP expiratory positive airway pressure

FEV1 forced expiratory volume in one second

FiO2 fractional inspired oxygen concentration

GCS Glasgow Coma Scale

H+ hydrogen ions

H2O water

HDU high dependency unit

HRCT high-resolution computed tomography

ICU intensive care unit

IPAP inspiratory positive airway pressure

kPa kilo pascals

LTOT long term oxygen therapy

mm Hg millimetres of mercury

NICE National Institute for Health and Care Excellence

NIPPV non-invasive positive pressure ventilation

NIV non-invasive ventilation

NPSA National Patient Safety Agency

O2 oxygen

PCO2 partial pressure of carbon dioxide

PO2 partial pressure of oxygen

RCT randomised controlled trial

SaO2 oxygen saturation

UK United Kingdom

V-Q ventilation-perfusion

Respiratory failure

Respiratory failure occurs due to inadequate gas exchange resulting in an abnormally low oxygen (O2) level in blood. It is a potentially life-threatening condition that can lead to respiratory arrest and death if untreated. It can be categorised into type 1 respiratory failure or type 2 respiratory failure which can be acute or chronic at presentation. The diagnosis of type 1 and type 2 respiratory failure can be made by arterial blood gas (ABG) measurement. The mechanisms for developing these two types of respiratory failure are different (Box 13.1).

Box 13.1 Key definitions.

 Hypoxaemia is an arterial oxygen level that is below normal and which can result in hypoxia.

 Hypoxia is a reduction in the oxygen delivery to tissues despite adequate perfusion.

 Hypocapnoea is a reduced level of carbon dioxide (CO2) in blood.

 Hypercapnoea is an elevated level of COin blood.

 Respiratory failure is defined as hypoxaemia, with a partial pressure of oxygen (PaO2) of <60 mmHg or 8.0 kPa.

 Type 1 respiratory failure is hypoxaemia (PaO2 < 60 mmHg or 8.0 kPa) with a normal or reduced CO2 level. It arises from a disturbance of the ventilation and perfusion of the lungs and can be the result of any acute respiratory illness, such as pulmonary embolus, acute asthma, heart failure or pneumonia.

 Type 2 respiratory failure is defined as hypoxaemia (PaO2 < 60 mmHg or 8.0 kPa) with hypercapnoea, with a PaCO2 of >48 mmHg or 6.5 kPa. A patient who has type 2 respiratory failure is at risk of developing respiratory acidosis (pH < 735) which can be life-threatening. Type 2 respiratory failure can occur from any cause of alveolar hypoventilation as a result of ventilatory failure.


Table 13.1 depicts the results of a normal ABG measurement and ABG in patients with type 1 and type 2 respiratory failure.

The regulation of breathing

The physiology and control of breathing are discussed in Chapter 2. The respiratory centre consists of neurones in the medulla and in the floor of the fourth ventricle which initiate automated breathing activity under the regulation of chemical and physical reflexes. These can be overridden by voluntary cortical control, for example when sighing or breath-holding.

Table 13.1 Arterial blood gas measurements.



Type 1 respiratory failure

Acute type II respiratory failure

PaO2 mmHg (kPa)

80-100 (10.5-13.3)

<60 (8.0)

< 60 (8.0)

PaCO2 mmHg (kPa)


35-45 (4.5-6.0)

>48 (6.5)





HCO3- mmol/L




During inspiration, the inspiratory muscles contract, resulting in an increase in the volume of the thoracic cavity. This generates negative pressure in the alveoli and flow of air into the lungs. During expiration, the intra-alveolar pressure becomes slightly higher than atmospheric pressure and air flows out of the lungs.

The relationship between the partial pressure of oxygen in blood (PO2) and the oxygen saturation measured by pulse oximetry (SaO2) is described as the oxyhaemoglobin dissociation curve. Oxygen saturation is closely related to the partial pressure of oxygen in blood only over a short range of about 3—7 kPa. Above this level the dissociation curve begins to plateau and there is only a small increase in oxygen saturation as POrises. See Chapter 2 for details of the oxyhaemoglobin dissociation curve.

CO2 is carried in blood dissolved in plasma and is in equilibrium with the bicarbonate anion (HCO3-), which is an important buffer, ensuring acid-base homeostasis. The Henderson-Hasselbalch equation shows this:

In 1868, Pfluger was the first to show that the CO2 content of arterial blood directly affected breathing in animals. In 1905, Haldane and Priestley established that the respiratory centre was extremely sensitive to small changes in alveolar CO2 concentrations. CO2 readily crosses the blood-brain barrier, rapidly increasing the concentration of H+ in the cerebrospinal fluid. The chemoreceptors on the antero-lateral surfaces of the medulla are extremely sensitive to hydrogen ions (H+). A small rise in PCO2 therefore, results in an increase in the concentration of H+ in the cerebrospinal fluid and an increase in ventilation.

The peripheral chemoreceptors in the carotid and aortic bodies are also sensitive to CO2 and Hbut are activated only when there is a significant fall in the O2 level of arterial blood (< 60 mmHg or 8.0kPa). These chemoreceptors do not play an important role in the regulation of breathing under normal physiological conditions.

Patients with type 2 respiratory failure have chronically elevated CO2 levels in their blood and in the extracellular fluid surrounding the central chemoreceptors in the respiratory centre. These chemoreceptors, therefore, become relatively insensitive to the raised levels of CO2 Under these circumstances the response to hypoxia by the peripheral chemoreceptors becomes the key stimulant to breathing. Correcting the hypoxia by giving uncontrolled oxygen can prevent the response to hypoxia and worsen the hypoventilation, eventually resulting in respiratory arrest (see Chapter 2).

Mechanisms of respiratory failure

Respiratory failure can occur because of a disturbance of gas exchange at the alveolar level (type 1) or due to failure of the ventilatory muscle pump that enables air to enter and leave the lungs (type 2).

Type 1 respiratory failure

Type 1 respiratory failure occurs due to ventilationperfusion (V-Q) mismatch in the lungs, resulting in a reduction in gas exchange (Box 13.2). It is necessary to have an adequate surface area and sufficient blood flow through the pulmonary capillaries to maintain oxygenation. Any condition that results in a reduction in blood flow in the pulmonary arteries (such as pulmonary emboli) or a reduction in the surface area for gas exchange (such as emphysema) will result in type 1 respiratory failure.

Box 13.2 Causes of type 1 respiratory failure.

 Obstructive airways disease: severe asthma, COPD, bronchiectasis

 Parenchymal disease: pulmonary fibrosis of any cause, ARDS, pulmonary oedema, pneumonia

 Vascular disease: pulmonary hypertension, pulmonary emboli

 Other: pneumothorax, right-to-left shunt


Clinical presentation of type 1 respiratory failure

Patients with type 1 respiratory failure will usually present with breathlessness, increased respiratory rate and symptoms and signs of the underlying cause. This may include wheeze in asthma and crackles in patients with pulmonary oedema or pulmonary fibrosis. If the hypoxia is severe, the patient may appear cyanosed.

Investigations in type 1 respiratory failure

Initial investigations should include a full history, a thorough examination, pulse oximetry, ABG on air, chest X-ray, full blood count, urea and electrolytes, and blood glucose.

Further investigations can be carried out as indicated by the initial findings and may include a high-resolution CT scan (HRCT) if underlying fibrotic lung disease is suspected, ventilation-perfusion (VQ) scan or CTPA if a pulmonary embolus is suspected, or echocardiogram if left ventricular failure or pulmonary hypertension is suspected.

Alveolar-arterial oxygen gradient

The alveolar (A)-arterial (a) oxygen gradient is the difference between the alveolar concentration of oxygen (A) and the arterial concentration of oxygen (a). It is a more sensitive indicator of disturbance of gas exchange (VQ mismatch) than arterial blood gas measurement alone and can be particularly helpful in patients who appear breathless and are hypoxic without having respiratory failure (PaO2 > 8 kPa). Calculation of the A-a gradient in a patient with acute type 2 respiratory failure may also be helpful in determining the underlying cause of the lung disease.

Box 13.3 Calculating the Alveolar- arterial oxygen gradient.

FiO2 = fractional inspired oxygen concentration, which is 21% when breathing room air at atmospheric pressure

PAO2 = partial pressure of oxygen in the alveolus

PaO2 = partial pressure of arterial oxygen (obtained from ABG)

PaCO2 = partial pressure of arterial carbon dioxide (obtained from ABG)

R is the Respiratory Quotient and is 0.8 in an individual on a normal diet


In a young, healthy individual the A-a gradient will be around 2 or 3, increasing up to 4 with age. A high A-a gradient suggests a diffusion defect, a VQ mismatch or a right-to-left shunt. In routine clinical practice it is not unusual for the extent of the VQ mismatch to be underestimated because too much reliance is placed on simple oximetry and ABG measurement alone. A slightly reduced PaO2, for example between 10 and 12 kPa, is often ignored. Box 13.3 explains how to calculate the A-a gradient.

It is assumed that the PaCO2 is equal to the PACO2 because carbon dioxide crosses rapidly from the pulmonary vasculature to the alveoli. This equation can be simplified for ease of calculation as follows:

Appendix 13.A gives some examples of how to calculate the A-a gradient.

Management of type 1 respiratory failure

Type 1 respiratory failure can present as an emergency, requiring prompt assessment and treatment. Management of type 1 respiratory failure consists of immediately correcting the hypoxaemia and treating the underlying cause. The aim in type 1 respiratory failure is to maintain the oxygen saturation in the range of 94—98% using a suitable device. These patients are usually not at risk of developing hypercapnoea, so the concentration of oxygen that is required can be safely given, so long as the ABG is monitored regularly.

Oxygen therapy and monitoring in type 1 respiratory failure

Hypoxaemia (PaO2 < 60 mmHg or 8.0 kPa) must always be corrected to avoid the consequences, which includes respiratory arrest. Oxygen is a drug and must be prescribed, just like any other drug. Oxygen is indicated for all patients with hypoxaemia but is not a panacea for breathlessness in the absence of hypoxaemia except in the palliative care setting. The British Thoracic Society (BTS) Oxygen Guidelines, with agreement from 21 other societies, including the British Association for Emergency Medicine, the British Cardiovascular Society and the Royal College of Anaesthetists, has been disseminated nationally and these should be followed. The BTS oxygen audit is available at

It is the responsibility of the doctor to ensure that oxygen has been prescribed, specifying the appropriate target saturation range and the device through which oxygen is to be delivered.

Devices for giving oxygen

Patients who are hypoxic can be given oxygen through a variety of devices. Nasal cannulae are suitable for most patients with type 1 respiratory failure (Figure 13.1).

Figure 13.1 Patient using a nasal cannulae.

Figure 13.2 Patient using a simple face mask.

A flow rate of 2—6 L min-1 of oxygen can be delivered, giving a fractional inspired oxygen concentration (FiO2) of between 24 and 50%. The exact FiO2 will depend on the patient’s minute volume, inspiratory flow, and pattern of breathing. Nasal cannulae are cheap, comfortable to wear and well-tolerated. A simple face mask can also be used for patients with type 1 respiratory failure (Figure 13.2). The flow rate must be set at between 5 and 10 L min-1 to deliver an oxygen concentration of between 35 and 60%. A lower flow rate may result in CO2 build up. A reservoir or re-breathe mask (Figure 13.3) is used to deliver a higher concentration of oxygen of between 60% and 80% in patients with severe hypoxaemia who are not at risk of retaining CO2. It is often used in patients who are critically ill. Patients who are extremely hypoxic and require very high concentrations of inspired oxygen may benefit from continuous positive airways pressure (CPAP) device which can deliver over 90% of oxygen, often on the high dependency unit (HDU) (Figure 13.4). If oxygenation is inadequate through any of these devices, intubation and ventilation on the intensive care unit (ICU) should be considered. Patients who are at risk of type 2 respiratory failure should be given controlled oxygen via a venturi mask (see section on management of Type 2 respiratory failure).

Nursing staff are responsible for monitoring and documenting the oxygen saturation on the observation chart. In addition, the prescription chart should be signed at each drug round. This is to ensure that the oxygen prescription is being carefully followed and that the amount of oxygen given is adjusted according to the oxygen saturation measurement.

Figure 13.3 Patient using a reservoir (re-breathe) mask.

It must be remembered that high concentrations of oxygen given over a prolonged period may be harmful, resulting in coronary vasoconstriction, reduced cardiac index, and re-perfusion injury after a myocardial infarction. It may also have adverse effects in patients who have suffered a stroke.

Chronic type 1 respiratory failure

Patients with severe lung disease may be chronically hypoxic and are at risk of developing complications, such as cor pulmonale. They will require long term oxygen therapy (LTOT) at home. These patients should have a formal LTOT assessment, which is usually carried out by a respiratory nurse or physiotherapist. These patients are prescribed the amount of oxygen required (usually 1-4 L min-1) via a concentrator which is installed in their home to use for a specified number of hours. Chapter 3 discusses oxygen as a drug, how it is prescribed and delivered.

Type 2 respiratory failure

Type 2 respiratory failure occurs because of failure of ventilation resulting in alveolar hypoventilation. It can be acute or chronic, or present with an acute component overlying the chronic condition.

Figure 13.4 Patient being fitted with a CPAP device.

Patients who present with acute type 2 respiratory failure will be symptomatic and unwell.

Exacerbation of COPD is the commonest cause of acute type 2 respiratory failure in hospitals in the UK and is associated with significant morbidity and mortality. These patients may develop type 2 respiratory failure in transit to hospital or in the emergency department because they are given uncontrolled oxygen. The uncontrolled oxygen stops the hypoxic drive that the patient is reliant on, resulting in hypoventilation, hypercapnoea, and eventually respiratory arrest. It is therefore essential to be aware of the risk factors for developing type 2 respiratory failure. Box 13.4 lists the common causes of type 2 respiratory failure.

Acute type 2 respiratory failure can develop within minutes to hours. There is insufficient time for the renal buffering system to compensate, so the bicarbonate level remains in the normal range and the pH drops. Chronic type 2 respiratory failure develops over several days to weeks, during which period the kidneys excrete carbonic acid and reabsorb bicarbonate ions so that the pH is only slightly reduced and the bicarbonate level is elevated. Table 13.2 gives a quick and simple guide to working out ABG results in acute and chronic respiratory acidosis and in metabolic acidosis.

Box 13.4 Causes of type 2 respiratory failure.

 Chronic lung disease: COPD, severe chronic asthma, bronchiectasis, cystic fibrosis

 Chest wall deformity: kyphoscoliosis, thoracoplasty, extensive pleural calcification, chest wall trauma, obesity

 Neuromuscular and peripheral nerve disorders: myopathies, muscular dystrophy, motor neurone disease, spinal cord injury, poliomyelitis, Guillain-Barré syndrome, phrenic nerve injury, damage to diaphragm

 Disorders of the neuromuscular junction: myasthenia gravis, botulism

 Disorders of the respiratory centre: anaesthetics, respiratory depressants and sedatives, head injury, central sleep apnoea, cerebrovascular accident, multiple sclerosis


Table 13.2 Comparison of arterial blood gases in acute and compensated respiratory and metabolic acidosis.






Acute Respiratory Acidosis




Compensated (Chronic) Respiratory Acidosis

Normal but <740

Metabolic Acidosis



Compensated Metabolic Acidosis


Normal or↓

Normal but<740

Clinical presentation of acute type 2 respiratory failure

Patients presenting with type 2 respiratory failure are hypoventilating rather than hyperventilating, so may not appear dyspnoeic. Patients who initially present with type 1 respiratory failure and tachypnoea may become tired and thus begin to retain CO2, as occurs in life-threatening asthma discussed in Chapter 6. Patients may have symptoms and signs of the underlying cause of respiratory failure, for example, neuromuscular weakness, an abnormal chest wall or paradoxical abdominal movement suggesting diaphragmatic weakness. Patients with type 2 respiratory failure may display symptoms and signs of CO2 retention, which includes drowsiness, confusion, irritability, a CO2 retention flap, and a bounding pulse caused by vasodilatation. If untreated, the patient will become comatose when the PaCO2 rises above 10 kPa and will ultimately die.

Investigations for acute type 2 respiratory failure

Immediate investigations include a chest X-ray, measurement of oxygen saturation (pulse oximetry) and baseline ABG taken while breathing room air.

The initial ABG measurement will act as a guide to how much oxygen should be prescribed. Further, regular ABG measurements are compulsory once the patient has been commenced on oxygen. If the cause of the respiratory failure is not clear, then further investigations, such as a CT scan of the thorax or thoracic ultrasound may be indicated when the patient is stable.

Immediate management of acute type 2 respiratory failure

Once it has been established from an ABG measurement that the patient has acute type 2 respiratory failure, it is essential to treat it without delay. The key point in the management of type 2 respiratory failure is the use of controlled oxygen and treating the underlying cause, for example, with bronchodilators, antibiotics, corticosteroids, theophyllines, diuretics, and anticoagulants.

Controlled oxygen therapy and monitoring in type 2 respiratory failure

The challenge is to maintain the oxygen saturation between 88% and 92% without a significant increase in the level of CO2 and the development of respiratory acidosis. Controlled oxygen should be given by the use of a venturi (fixed performance) mask, which gives controlled inspired oxygen of 24%, 28%, 35% or 40%. These venturi masks are colour-coded to make it easier to identify which one to use (Figure 13.5, Figure 13.6).

If the patient is tachypnoeic with a respiratory rate of >30 breaths per minute, the oxygen supply should be increased by 50%. This does not increase the amount of inspired oxygen. If the patient is unable to tolerate a venturi mask, then a small concentration of inspired oxygen (0.5—1 L) can be given by nasal cannulae with careful monitoring. Once the patient has been commenced on oxygen therapy, the ABG should be measured every 20—30 minutes and the oxygen prescription adjusted until the patient is stable.

The risk of developing acute type 2 respiratory failure can be reduced by educating all healthcare professionals about those patients at risk, by always prescribing oxygen in the correct oxygen saturation range, and by carefully monitoring oxygen saturation to ensure that it is in the range prescribed. The British Thoracic Society (BTS) audit in 2008 highlighted concerns regarding the inappropriate prescription and monitoring of oxygen therapy in most UK hospitals. The BTS Guidelines (2008) have been disseminated to all Trusts with a recommendation to complete regular audits to ensure compliance. The National Patient Safety Agency (NPSA) has also emphasised the risk of oxygen therapy which can result in the development of type 2 respiratory failure.

Figure 13.5 A range of venturi valves. Source: ABC of COPD, 3rd Edition. Figure 11.4.

Figure 13.6 The venturi principle. Source: ABC of COPD, 3rd Edition, Figure 11.5.

Type 2 respiratory failure: to treat or not to treat?

As many patients who develop type 2 respiratory failure have severe chronic lung disease, some will deteriorate despite optimal management of the underlying condition and careful oxygen therapy. Their CO2 level will go up and they will develop respiratory acidosis. A decision has to be made by a senior doctor regarding the ceiling of treatment; whether to commence non-invasive ventilation (NIV), refer for intubation and ventilation, or whether palliative care is indicated. This decision should be based on the patient’s pre-morbid state (including their quality of life), the severity of their underlying disease, any co-morbid disease (particularly cardiac and neurological), any reversible component to their acute illness, any relative contra-indications to NIV, and their wishes.

Unless there are contraindications, non-invasive ventilation is the treatment of choice. Many patients with type 2 respiratory failure secondary to severe, chronic disease are not considered suitable for ventilation on the ICU as they can develop severe nosocomial infections and weaning them off the ventilator can be very difficult. Prior to the introduction of NIV, most of these patients would have died without any form of ventilation.

If it is clear that it is an ‘end-of-life’ situation, then palliative care should be the priority, with the aim of symptom control, especially relieving distressing breathlessness with medication such as opiates. Clear, careful, and sympathetic communication with the family and the patient is crucial. The palliative care team should be involved.

There should be clear documentation in the notes of the treatment decision and management plan. If NIV is to be initiated, there should be a clear plan regarding how long to continue with NIV, what to do if it fails and whether referral to the ICU for intubation and ventilation should be considered. A decision about the patient’s resuscitation status should be made and documented. Ideally, the decision regarding the ceiling of treatment in patients with end-stage lung disease should be made prior to an acute admission after a detailed discussion with the patient and their family, and should be clearly documented in the notes.

Non-invasive ventilation (NIV)

The term non-invasive positive pressure ventilation (NIPPV) is synonymous with non-invasive ventilation (NIV). A variety of ventilator units are available (Figure 13.7), but the commonest in UK hospitals is the bi-level positive airways pressure (BiPAP) unit. Negative-pressure tank-type ventilators were first developed by Dalziel in 1832 and Drinker-Shaw used the iron lung for patients with respiratory failure secondary to poliomyelitis in 1928. The tank ventilator was further developed by Emerson in 1931.

Figure 13.7 A non-invasive ventilator. Source: ABC of COPD, 3rd Edition, Figure 13.1.

NIV has revolutionised the management of type 2 respiratory failure and is the treatment of choice for patients with acute type 2 respiratory failure and decompensated respiratory acidosis, with a pH < 7.35 and a PaCO2 of > 6.5 kPa, who have not responded to optimal medical treatment and careful oxygen therapy. NIV improves clinical parameters within a few hours of being commenced, with a reduction in respiratory rate, reduction in the work of breathing, an increase in tidal volume, improvement in oxygenation, a reduction in the CO2 level and improvement in acidosis. NIV has been shown in randomised controlled trials (RCTs) to reduce the need for intubation, reduce the length of stay in hospital, and reduce mortality in patients with type 2 respiratory failure secondary to a variety of causes, but particularly COPD. NICE recommends that NIV is available in all hospitals treating patients with acute type 2 respiratory failure. Box 13.5 lists the inclusion and exclusion criteria for NIV.

Box 13.5 Inclusion and exclusion criteria for NIV.

Inclusion criteria

Exclusion criteria

pH < 735 and PaCO2 > 6 kPa

Metabolic acidosis



Not requiring immediate intubation and ventilation

Requires immediate intubation and ventilation


Not co-operative

Relatively calm

Severe agitation

Not cognitively impaired

Severe cognitive impairment

Little or no confusion

Severe confusion

Patient and/or their relatives wish to have NIV

Patient and/or their relatives do not wish to have NIV

Able to protect airways

Unable to protect airways

Reasonable quality of life

Poor quality of life

Facial surgery



Haemodynamically unstable

Some of these are relative contraindications and the benefit versus the risks of NIV should be considered carefully by a senior doctor after full discussion with the patient and/or their relatives. NIV is not indicated for patients with metabolic acidosis.

Table 13.2 describes the differences in ABG measurements between respiratory and metabolic acidosis. It is beyond the scope of this book to discuss the causes and management of metabolic acidosis.

It is important to be realistic about the use of BiPAP in patients with severe type 2 respiratory failure and patient selection is essential. Factors predicting a successful outcome include a co-operative patient with normal neurological function, a moderately high APACHE II score (acute physiology and chronic health evaluation) and a pH>7.10. If the patient deteriorates despite NIV, there should be a clear decision regarding the ceiling of treatment made by a senior physician after discussion with the patient and their family. If it is felt that the patient is a candidate for invasive ventilation, then referral to ICU should be made without delay.

There is evidence that the prognosis is better if the patient is commenced on NIV within 60 minutes of presentation. There is also evidence that patients who are managed on a high dependency unit (HDU) by experienced doctors and nurses have a better outcome. NIV may also be indicated for patients who are slow to wean from ventilation as it reduces the total length of time on a ventilator and reduces mortality.


The usual device used is a bi-level positive airways pressure device (BiPAP) which can be set to deliver different pressures during inspiration and expiration. BiPAP should be readily available in the emergency department, acute medical unit, and respiratory ward in all hospitals and should be initiated by an experienced doctor or other healthcare professional. The patient should be in a sitting or semi-recumbent position and should have a full face mask fitted in the first instance which can be changed to a nasal mask after 24 hours if the patient prefers it.

It is recommended that the inspiratory positive airways pressure (IPAP), which blows off the CO2, is commenced at 10 cm water (H2O), and then increased incrementally, by 2 cm, up to a maximum of 24 cm H2O if there is persistent hypercapnia. The expiratory positive airways pressure (EPAP) should be set at 4 cm H2O to begin with and then increased up to 10 cm H2O (Figure 13.8). The number of breaths per minute (BPM) is set between 12 and 18 in the patient flow-triggered/time-triggered (S/T) mode and the synchrony of ventilation should be checked. The limiting factor is often patient tolerance. As patient comfort will improve compliance and the success of this treatment, it is important to begin with the low settings to allow the patient time to get used to the feeling of the mask and the ventilator as this can be quite frightening and uncomfortable. The exact settings of the BiPAP will depend on the individual patient and includes factors such as the size of the patient and the severity of bullous disease.

Figure 13.8 NIV pressures.

A variety of full face and nasal masks in small, medium and large sizes are available. Measurement of the facial dimension should be taken to ensure that the mask fits properly. A mask that fits firmly but comfortably will improve compliance and prevent the leakage of air which will compromise the close circuit functioning of the system. A well-fitting mask will reduce pressure sores and skin lacerations from developing in areas of close fitting, such as the bridge of the nose. Other types of patient-ventilator interfaces, such as mouthpieces, nasal pillows, total face masks, and helmet devices are also available.

Patients who feel claustrophobic should be given reassurance, encouragement, and frequent (but short) breaks to allow them to get used to the mask and the device. This can be time-consuming for the doctor or nurse who initiates the BiPAP. Patients on BiPAP can become dry, so will need intravenous fluids, humidification, and nasal saline. Gastric insufflation, aspiration, and pneumothorax are rare complications of NIV.

Most patients in type 2 respiratory failure will require supplemental oxygen while on BiPAP in order to correct their hypoxaemia without worsening the hypercapnoea and acidosis. This is given through a port on the BiPAP mask until oxygen saturation in the range of 88—92% is achieved.

Monitoring on NIV

NIV should be prescribed on a chart with documentation of the initial IPAP and EPAP settings, the flow rate of supplemental oxygen given, the baseline ABG measurement, respiratory rate, heart rate and GCS. It must be emphasised that a patient commenced on BiPAP must have ABG measurement done 30—60 minutes later, and after every change in setting. Acutely unwell patients may require more frequent ABG measurements and may benefit from an arterial line. It is recommended that these patients are monitored on the HDU by experienced nurses. There should be documentation of any change in the settings and a record of the hours of use and time taken for breaks. Clinical progress can be gauged by observing the patient’s respiratory rate, heart rate and use of accessory muscles. Patients on BiPAP should have continuous oxygen pulse oximetry and cardiac monitoring.

Figure 13.9 Oxygen Alert Card. Source: ABC of COPD, 3rd Edition, Figure 11.3.

Weaning off NIV

Patients should be continued on BiPAP until there is clinical improvement and their acidosis resolves. On average, this takes 48—72 hours. Patients are usually weaned off BiPAP gradually, with extended periods off BiPAP, initially during the day and then at night, until they no longer require it. Some patients improve very quickly and may only require NIV for a few hours. Some patients with severe lung disease may remain in type 2 respiratory failure and may require domiciliary ventilatory support.

Respiratory stimulants

Doxapram stimulates the respiratory centre in the medulla to increase the tidal volume and respiratory rate. Although not used as first line treatment, doxapram can be used in patients with type 2 respiratory failure who are unable to tolerate BiPAP or when there is a contra-indication to BiPAP. Doxapram should be initiated after consultation with a senior doctor who has experience in using the drug. Doxapram is given intravenously at 1—3 mg min-1 and usually for a short period of time. Careful monitoring of the patient will be required.

Prognosis and outcome

In many cases, patients with acute type 2 respiratory failure will improve over a few days if their respiration is supported over the critical period. However, their prognosis depends on their underlying disease and they may be at risk of developing respiratory failure again. They should be made aware of this, given written information about their condition and the dangers of uncontrolled oxygen therapy. Before discharge from hospital, they should be given an oxygen alert card (Figure 13.9) specifying how much oxygen they can have and a venturi mask to use at home if they are on long term oxygen or if they become acutely unwell and require oxygen, for example, while being transferred to hospital via an ambulance.

Management of chronic type 2 respiratory failure

Patients with musculoskeletal abnormalities, neuromuscular problems and obesity can develop type 2 respiratory failure gradually over a period of time. These patients hypoventilate at night when they are supine. Patients with chronic type 2 respiratory failure may not appear particularly breathless or very unwell despite high CO2 levels because of the compensatory mechanisms which correct the acidosis (Table 13.2 describes the ABG in a patient with a compensated respiratory acidosis). These patients often report symptoms of tiredness, lethargy, and morning headaches from the high CO2 levels overnight. An overnight full sleep study, which includes pulse oximetry and nocturnal CO2 monitoring, will be required to make the diagnosis.

If they become unwell, for example, with an infection, they can present with an acute respiratory failure and acidosis on top of their chronic respiratory failure. Once the acute element has been treated with antibiotics, diuretics, and NIV, the patient will return to their baseline but will continue to have elevated CO2 levels. Patients with chronic type 2 respiratory failure are managed with domiciliary NIV with regular follow up in a dedicated centre.

 Respiratory failure can be defined as a PaO2 of less than 8.0 kPA.

 Type 1 respiratory failure is hypoxia with normal CO2 levels and is caused by a diffusion defect, VQ mismatch or a right-to- left shunt.

 Type 2 respiratory failure is hypoxia with hypercapnoea (PaCO2 level above 6.5 kPA) and is due to failure of ventilation.

 Type 2 respiratory failure can occur if a patient is given uncontrolled oxygen which can stop their hypoxic drive.

 Interpretation of the ABG measurement is critical in deciding what the diagnosis is and in prescribing the correct concentration of oxygen.

 The widened alveolar-arterial (A-a) oxygen gradient is more sensitive than ABG measurement at detecting a VQ mismatch, a diffusion defect or a right-to-left shunt in a breathless patient who is hypoxic.

 Patients with type 1 respiratory failure should be prescribed oxygen to maintain their oxygen saturation in the range of 94-98%.

 Oxygen is not indicated for breathlessness alone except in the palliative context.

 When prescribing oxygen, it is important to use the correct device.

 Patients at risk of developing type 2 respiratory failure should be identified and prescribed oxygen to maintain their oxygen saturation in the range of 88-92%.

 It is possible to work out from the ABG measurement whether the patient has an acute or chronic type 2 respiratory failure.

 Patients with type 2 respiratory failure have a high morbidity and mortality and should be managed by specialists on the HDU.

 Controlled oxygen should be prescribed for type 2 respiratory failure using a venturi mask. Serial ABG measurements should be made and oxygen concentration titrated accordingly.

 Patients with type 2 respiratory failure should be commenced on NIV if pH < 7.35, so long as there are no absolute contraindications.

 Supplemental oxygen can be given via nasal cannulae in a patient on BiPAP

 It is essential to choose the right mask to ensure a firm and comfortable fit and to get the settings right in order to ensure compliance. This can take time.

 Type 2 respiratory failure can be prevented by educating doctors, nurses, and patients about the risks, issuing alert cards and by giving the patient a venturi mask to take home.

 Patients with type 1 or type 2 respiratory failure who do not respond to any treatment should be discussed with the intensivists for consideration of intubation and ventilation.

 In patients with type 2 respiratory failure and severe lung disease, a decision will have to be made by a senior doctor regarding the ceiling of treatment.


13.1 In a healthy individual the chemoreceptors in the medulla oblongata are most sensitive to changes in the concentration of what in the blood?

A Bicarbonate

B Carbon dioxide

C Carbonic acid

D Hydrogen ion

E Oxygen

Answer: B

The chemoreceptors in the medulla are sensitive to hydrogen ions, CO2 and O2. However, CO2 crosses the blood-brain barrier more quickly than hydrogen ions and therefore changes in CO2 level in the blood result in the most rapid change in ventilation.

13.2 Which of the following statements about the alveolar-arterial gradient is true?

A The A-a gradient decreases with age

B The A-a gradient increases in Type 2 respiratory failure

C The A-a gradient decreases in Type 1 respiratory failure

D The A-a gradient can be calculated using the Henderson-Hasselbalch equation

E The A-a gradient is a more sensitive measure of VQ mismatch than arterial blood gas measurement

Answer: E

The A-a gradient increases with age, from two up to four in those with normal lungs. It increases in any condition that causes a diffusion defect, a VQ mismatch or a right- to-left shunt, all of which present with type 1 respiratory failure. It can be calculated using the PaO2, PaCO2, and the FiO2. It is a more sensitive measure of VQ mismatch than ABG measurement.

13.3 A 65-year-old woman with kyphoscoliosis is admitted to hospital and has ABG measurement while breathing 2 l oxygen via nasal cannulae. The results show a pH of 7.38 kPa, PaO2 of 8.6 kPa, PaCO2 of 10.0 kPa and a HCO3- (bicarbonate) of 41.2 mmol/L. What does this indicate?

A Respiratory acidosis

B Compensated respiratory acidosis

C Respiratory alkalosis

D Compensated respiratory alkalosis

E Metabolic acidosis

Answer: B

The ABG indicates that although the COis high, the pH is within normal limits and the bicarbonate is high. This is consistent with a compensated respiratory acidosis.

13.4 A 50-year-old man with severe COPD exacerbation is rushed into the emergency department and is given 10 l oxygen via a re-breathe mask. His ABG is as follows: pH 7.25, PaO2 of 10.82 kPa, PaCO2 of 8.50 kPa and HCO3- of 31.2 mmol/L. How would you manage this patient?

A Commence BiPAP

B Commence CPAP

C Give oxygen via a venturi mask and redo ABG

D Give intravenous bicarbonate

E Call ICU

Answer: C

This patient with COPD has developed type 2 respiratory failure because he has been given uncontrolled oxygen. He is over-oxygenated as his PaO2 is well above 8 kPa. Although he is acidotic, the oxygen should be reduced in the first instance using a venturi device and the ABG should be rechecked. Only if the patient and the ABGs do not improve should BiPAP be commenced. If that does not work, then intubation and ventilation should be considered.

13.5 A 68-year-old man with moderately severe COPD is admitted with an infective exacerbation. His respiratory rate is 22 and his GCS is 15. His ABG on air is as follows: pH 7.38, PaCO2 6.92kPa, PaO2 6.50 kPa, HCO3—24.2 mmol/L. What would you do? A Commence BiPAP alone

B Commence BiPAP and 2 l of oxygen via nasal cannulae

C Give controlled oxygen via nasal cannulae

D Give controlled oxygen via face mask

E Give controlled oxygen via a venturi device

Answer: E

This man is in type 2 respiratory failure and therefore needs controlled oxygen. This can only be delivered safely through a venturi device. It is not possible to know the exact concentration of oxygen delivered through the face mask or nasal cannulae. He is not acidotic enough to need BiPAP

13.6 Which of the following statements about type 1 respiratory failure is true?

A It can occur in patients with COPD

B One should aim for a target oxygen saturation range of 88—92%

C Any patient with a PaO2 of <10 kPa should receive oxygen

D It can be alleviated by BiPAP and oxygen

E It should always be managed by giving high concentration of oxygen through a rebreathe mask

Answer: A

Type 1 respiratory failure can occur in any respiratory disease, including COPD, although these patients are at risk of developing type 2 respiratory failure so should be

monitored closely with regular ABG measurements. The target oxygen saturation to aim for is 94—98% in those who do not retain Co2. The definition of respiratory failure is a PaO2 < 8.0 kPA so a patient with a PaO2 of 10 kPa does not require oxygen. BiPAP is not indicated in type 1 respiratory failure.

13.7 Which of the following statements about type 2 respiratory failure is true?

A It is possible to predict who will develop type 2 respiratory failure from the initial ABG measurement

B It can be managed with CPAP and controlled oxygen on the HDU

C A PaCO2 of >6.5 kPA is an indication for immediate BiPAP

D A bicarbonate level of >35 in the ABG suggests a chronic type 2 respiratory failure

E Chronic type 2 respiratory failure responds well to respiratory stimulants, such as doxapram

Answer: D

It is not possible to predict whether someone will retain CO2 from the first ABG measurement which is why serial readings are required. CPAP is not a treatment for type 2 respiratory failure. Slight hypercap- noea in the absence of acidosis is not an indication for BiPAP. A high bicarbonate level is indicative of a chronic process as there has been time for the kidneys to retain bicarbonate. Respiratory stimulants will not have a significant effect on chronic type 2 respiratory failure.

13.8 Which of the following statements about NIV is true?

A It is recommended for all patients with type 2 respiratory failure

B It should be used with sedation in an agitated and non-compliant patient

C It has significantly improved the mortality of patients with type 2 respiratory failure

D It is contra-indicated if the patient is very hypoxic with a PaO2 < 5 kPa

E It is commonly complicated by a pneumothorax

Answer: C

NIV has improved the management of type 2 respiratory failure and reduced the need for intubation. It is only recommended for patients who are acidotic with a pH < 7.35 and who can tolerate it. Sedation is contra-indicated in any patient with respiratory failure and no patient should be forced to have NIV. Patients who are very hypoxic and in type 2 respiratory failure can have NIV and supplemental oxygen as required. Pneumothorax can occur, especially in patients with bullous disease, but is not a common complication.

13.9 Which of the following factors best predicts a successful outcome with NIV for a patient presenting with type 2 respiratory failure?

A The APACHE 11 score

B The initial PaO2 on air

C The age of the patient

D The length of time they have had their lung disease

E Their baseline FEV1

Answer: A

A moderately high APACHE 11 score best predicts a successful outcome. The other factors may also indicate severity of disease and influence prognosis.

13.10 Which of the following statements about metabolic acidosis is true?

A It can never occur together with respiratory acidosis

B The bicarbonate in the ABG is usually low

C The PaCO2 in ABG is usually high though the compensatory mechanism

D It can be successfully managed with NIV.

E It can improve slightly with respiratory stimulants

Answer: B

Metabolic acidosis can occur together with respiratory acidosis and results in a low bicarbonate. The PaCO2 is usually normal or slightly low from the increased respiratory rate. Neither NIV nor respiratory stimulants are indicated for metabolic acidosis.

Appendix 13.A Calculation of the alveolar-arterial oxygen gradient

The equation for calculating the A-a gradient Example 13.1:

Normal individual breathing room air.

FiO2 = 21 PaCO2 = 4 kPa

PaO2 = 13 kPa

A - a gradient = 21 - (5 x 1.2) - 13 = 2 Example 13.2:

Young woman with a small pulmonary embolus making her breathless, breathing room air. Her

PaO2 appears to be almost normal.

FiO2 = 21 PaCO2 = 3 kPa

PaO2 = 12 kPa

A - a gradient = 21 — (3 x 1.2) — 12 = 5.4 Example 13.3:

A young man with a right-to-left shunt breathing room air.

FiO2 = 21 PaCO2 = 4 kPa

PaO2 = 9 kPa

A - a gradient = 21 — (4 x 1.2) — 9 = 7.2


Albert, R., Spiro, S., and Jett, J. (2001). Comprehensive Respiratory Medicine. St. Louis, MO: Mosby Chapter 12.

Bott, J., Carroll, M.P, Conway, J.H. et al. (1993). Randomised controlled trial of nasal ventilation in acute ventilator failure due to chronic obstructive airways disease. Lancet 341: 1555—1557.

Brochard, L., Mancebo, J., Wysocki, M. et al. (1995). Noninvasive ventilation for acute exacerbations of chronic obstructive pulmonary disease. New England Journal of Medicine 333: 817—822.

Flenley, D.C. (1978). Interpretation of blood-gas and acid-base data. British Journal of Hospital Medicine 20: 384-394.

Kramer, N., Meyer, TJ., Meharg, J. et al. (1995). Randomised, prospective trial of noninvasive positive pressure ventilation in acute respiratory failure. American Journal of Respiratory and Critical Care Medicine 151 (6): 1799-1806.

Lumb, A.B. (2000). Nunn's Applied Respiratory Physiology, 5the. Oxford: Butterworth-Heinemann Chapter 5.

Martin, T.J., Hovis, J.D., Constantino, J.P et al. (2000). A randomised, prospective evaluation of noninvasive ventilation for acute respiratory failure. American Journal of Respiratory and Critical Care Medicine 161 (3): 807-813.

National Institute for Health and Care Excellence (2010). Guideline on Chronic Obstructive Pulmonary Disease, June 2010, clinical guideline 12.

Nava, S., Bruschi, C., Orlando, A. et al. (1998). Noninvasive mechanical ventilation (NINMV) facilitates the weaning of patients with respiratory failure due to chronic obstructive pulmonary disease. Annals of Internal Medicine 128:721-728.

O’Driscoll, B.R., Howard, L.S., and Davison, A.G. (2008). BTS guidelines for emergency oxygen use in adult patients. Thorax 68: vi1-vi 68.

Pingleton, S.K. (1988). Complications of acute respiratory failure. American Review of Respiratory Diseases 137: 1463-1493.

Royal College of Physicians (2008). Non-invasive ventilation in chronic obstructive pulmonary disease: management of acute type 2 respiratory failure, National Guidelines, Concise Guidance to Good Practice, vol. 11. London: Royal College of Physicians.

Soo Hoo, G.W., Santiago, S., and Williams, J.(1994). Nasal mechanical ventilation for hypercapnic respiratory failure in chronic obstructive pulmonary disease: determinants of success and failure. Critical Care Medicine 27: 417-434.

Woollam, C.H.M. (1976). The development of apparatus for intermittent negative pressure respiration. Anaesthesia 3: 666-668.

If you find an error or have any questions, please email us at Thank you!