teaching
notes

10. RESPIRATORY FAILURE

Contents
10.1 Definition
10.2 Causes
10.3 Normal Values
10.4 Blood Gas Measurement
10.5 Indirect Measurement of Gas Tension
10.6 Pathophysiology of Respiratory Failure
10.7 Acid-Base Status
10.8 Chronic Respiratory Failure
10.9 Adult Respiratory Distress Syndrome

10.1 DEFINITION
Respiratory failure is present when the respiratory system no longer oxygenates the arterial blood and/or eliminates carbon dioxide adequately. A widely used working definition of respiratory failure is given as a PO2 of less than 60mmHg, with or without an elevated PCO2, when the subject is at rest breathing air at sea level. A PaO2 of less than 80 under these conditions is abnormal though not clinically serious unless under 60mmHg. Although such a definition does not tell us about the subject who only develops abnormal blood gases during exercise, or another who only develops abnormal blood gases during sleep (see sleep apnoea), it is a logical reference point.

10.2 CAUSES

• lung pathology (i.e. affecting transfer of gases and ventilation/perfusion relationships within the lungs)
  
• neuromuscular disease and skeletal deformities (i.e. affecting the respiratory pump)
  
• defects of respiratory control (affecting the regulation of gas exchange and therefore the respiratory pump)

The respiratory system is capable of a wide range of performances in response to a wide range of metabolic demands; in the basal state of sleep only 0.3 litres/minute of oxygen and carbon dioxide need to be transferred. The level of ventilation required to achieve this is only 7 litres/minute with a pulmonary blood flow of 5.0 litres/minute. In maximal exercise up to 3.0 litres/minute of oxygen are used and a similar volume of carbon dioxide produced. This requires ventilation in the order of 80 litres/minute and a pulmonary blood flow of 15 litres/minute. The upper performance range varies greatly according to fitness.

Because of its chemical properties, haemoglobin will take up and carry a maximum load of oxygen provided the dissolved partial pressure of oxygen is above 60mmHg. At 60mmHg, haemoglobin saturation begins to fall below 90%. Thus, for a wide physiologic range of alveolar and thus dissolved PO2 (i.e.. 60-100mmHg) haemoglobin carries its maximum or near maximum quantity of oxygen. In contrast, below a PO2 of 60mmHg, haemoglobin's ability to carry oxygen falls precipitously. For a normal subject, the arterial PO2 is typically 25-30mmHg above this level (i.e. 85-95mmHg) thus providing a large safety margin in the oxygen delivery chain.

Disease (due either to inefficiency of gas transfer within the lungs, or a disturbance of the control of neuromuscular systems) will initially affect the individual's maximum performance. Because performance ranges so widely, the definition of the level of arterial PO2 and PCO2, which indicate respiratory failure, is arbitrary.

10.4 BLOOD GAS MEASUREMENT
Respiratory failure as a concept in medicine is a relatively recent one, emerging in the 1950s with the development and clinical use of reliable methods of blood gas measurement. Only the extreme forms of respiratory failure can be recognised on physical examination. Respiratory failure and dyspnoea are not synonymous. Severe, life-threatening respiratory failure may be present without any symptoms (such as dyspnoea), and dyspnoea and other symptoms of respiratory difficulty may be present when there is no respiratory failure. Some signs can give clues to the presence of respiratory failure. Those signs are:

• breathing movements
  
• central cyanosis

Obviously an unconscious subject who is not making breathing movements is in 'respiratory failure'. Central cyanosis is a very useful physical finding, but its absence does not mean the absence of respiratory failure. Cyanosis is not reliably detectable until the saturation is about 80% (PaO2 of 45mmHg) and is dependent on Hb level. In gas poisoning (CO poisoning) the stable carboxyhaemoglobin compound is bright red even though the tissues are hypoxic. There are no reliable clinical signs of an elevated PCO2. Thus, respiratory failure should be suspected on clinical evidence (e.g. dyspnoea, airflow limitation, pneumonia, asthma etc.) but can be reliably diagnosed only by measuring arterial blood gases.

Arterial puncture is the only direct method of measurement giving all the required numbers:

10.5 INDIRECT MEASUREMENT OF GAS TENSIONS

Ear Oximetry
Modern instruments allow direct measurement of arterial oxygen saturation from arterialised blood in the ear lobe. The technique is based on the differential absorption of red and infrared light by the haemoglobin molecule. Measures only saturation, not partial pressures. This is used extensively in sleep tests and intensive care monitoring and will be in common use in operating theatres and emergency wards.

Transcutaneous Measurement of PCO2, PO2
Skin electrodes used extensively in the neonate but because of a slow equilibration time are not used in adults except in intensive care or sleep testing.

Rebreathing measure of mixed venous PCO2
A very simple measure gives an accurate value of the pulmonary mixed venous PCO2, and thus a guide to overall CO2 levels (normal mixed venous PCO2 is 45mmHg). Important method, as it can be used in hospitals etc. without blood gas machines.

10.6 PATHOPHYSIOLOGY OF RESPIRATORY FAILURE
Respiratory failure occurs in lung disease, chest wall disease, neuromuscular disease, and with upper airway dysfunction. These may occur in combinations. There are a number of pathophysiologic mechanisms involved, and these differ for each of the disease groups.

Mechanisms of low PaO2

• Hypoventilation - neuromuscular disease, disease of respiratory control centre, apnoea.
  
• Ventilation-perfusion inequality (abbreviated, V/Q inequality) - lung disease.
  
• Cardiovascular Shunt - lung disease.
  
• Diffusion impairment - lung disease.
  
• Special case - a low inspired PO2 (e.g. at high altitude, wrong anaesthetic gas!)

10.6.1 Hypoventilation
The cardinal evidence of hypoventilation is a raised arterial PCO2. Hypoventilation (alveolar hypoventilation) is commonly caused by non-pulmonary disease (i.e. respiratory muscle weakness, neurological disease, drug overdose, sleep apnoea) There is an inverse hyperbolic relationship between alveolar ventilation and arterial PCO2:

PCO2=	VCO2	x K
	VA	(where VA = alveolar ventilation)

 

Stated simply, the higher the alveolar ventilation, the lower the arterial PCO2, and vice versa.

Hypoxaemia induced by hypoventilation is readily abolished by increasing inspired PO2. Where hypoxaemia and hypercapnia co-exist, the question must be asked: 'how much of the hypoxaemia is due to hypoventilation?' This is readily answered by reference to a simplified alveolar gas equation:

PAO2 = PIO2 -

(Pa CO2) R

+ F

 
Where:
PaO2 = alveolar PO2
PIO2 = inspired PO2 (150mmHg for room air)
PaCO2 = arterial CO2
R = respiratory exchange quotient (i.e. the ratio of CO2 production to O2 consumption - usually 0.8).
F = correction factor (for clinical purposes can be deleted).

Thus, a simple clinical equation is: PaO2 = PIO2 - (PaCO2 x 1.25)

The use of the equation is as follows. Arterial blood gases are measured directly, so you will know the PaCO2 and PaO2 such that you can readily calculate the A - a (alveolar - arterial) gradient for oxygen (the AaDO2). Normal range is 5-15mmHg, and an increase is seen with age. For example, an arterial sample gives a PaCO2 of 60 and PaO2 of 50 (for a patient breathing room air). The A-a gradient is 25mmHg, therefore the hypoxaemia is not entirely the result of hypoventilation.

Note that the PIO2 can only be confidently assumed in patients breathing room air or a precisely controlled inspired oxygen concentration (e.g. on a ventilator). It cannot be confidently calculated in patients breathing supplemental oxygen via face masks or prongs.

This is the major mechanism by which lung disease causes hypoxaemia. The arterial CO2 may be normal or elevated. Stated simply, the ventilation and blood flow to various regions of the lung are mismatched such that transfer of all gaseous O2 and CO2 becomes inefficient. Some regions may receive too much ventilation and too little blood flow (high V/Q units), whereas others receive too little ventilation for too much blood flow (low V/Q units). Mismatch is the main cause of hypoxaemia in chronic obstructive lung disease, interstitial lung disease and pulmonary embolic disease. Normal lungs have some V/Q inequality because of inequality of ventilation and blood flow to the apical and basal regions of the lungs in the erect posture.

Measures of V/Q inequality

• A - a gradient for oxygen (maximum normal is 15mmHg), i.e. for a 60 year old man, an arterial PO2 of 85mmHg (with PCO2 of 40) represents an A - a gradient of 15mmHg (normal 'ideal' alveolar oxygen is 100mmHg). In a young adult an A - a gradient of 5mmHg is more typical
  
• Multiple inert gas elimination technique

Why can a low PaO2 co-exist with a Normal PaCO2?
The effects of V/Q mismatch in lung disease are not easily separated from the reflex ventilatory responses to abnormal blood gases (i.e. the ventilatory responses to hypercapnia mediated by central chemoreceptors, and to hypoxia mediated by peripheral chemoreceptors). V/Q mismatch always causes hypoxaemia (a low PaO2) and hypercapnia. However, provided the chemoreceptor responses are normal, these abnormal blood gases will in turn cause reflex adjustments to ventilation returning the CO2 towards normal. The AaDO2 and arterial PO2 is not returned to normal by these ventilatory adjustments. The sequence of events is outlined in Figure 10a.

Why does an increase in ventilation correct the hypercapnia of V/Q inequality but not the hypoxia?
This is a most important concept. The simple explanation is found in the way in which blood carries oxygen and CO2. An increase of ventilation of good lung units readily compensates for poor units where CO2 is concerned because the extraction of CO2 increases proportionally to increasing ventilation. In contrast, increasing ventilation to good units cannot increase the content of oxygen carried away by blood from those units because of the oxyhaemoglobin dissociation curve; above 70mmHg, blood will not take up any more oxygen - i.e. increasing the PaO2 to 110mmHg (as occurs in increased ventilation), does not significantly increase the content of oxygen in the blood leaving those units.

10.6.4 Diffusion Defects
Thickened alveolar walls do impede the diffusion of oxygen and CO2. However, even in interstitial lung disease, V/Q inequality is the main mechanism of hypoxia at rest. On exercise, diffusion becomes a major factor. At rest blood has 0.75 seconds exposure to oxygen in the alveoli - for normal alveolar walls, equilibration of blood is complete in 0.25 seconds. With a thickened alveolar membrane, greater time is required in the gas exchange region for diffusion equilibration to occur. Diffusion impairment will now cause hypoxaemia, initially on exercise (where blood flow increases), then later at rest.

To understand blood gases, you must also understand acid-base status.

Terms

• pH (normal: 7.40)
  
• HCO3-(normal: 27mEQ/L)
  
• Base excess (normal: +2 to -2)
  
• Acute respiratory acidosis
  
• Compensated respiratory acidosis (renal HCO3- retention)
  
• Acute respiratory alkalosis
  
• Metabolic acidosis (respiratory compensation)
  
• Metabolic alkalosis (respiratory compensation)

Clinical Syndromes of Respiratory Failure

• Acute (unstable or stable)
  
• Chronic
  
• Acute on chronic

None of these terms should be used in an absolute sense. Acute respiratory failure means that circumstances have recently arisen or occurred (minutes, hours). It is unstable if the blood gases are progressively worse over minutes, as in choking or drowning. Examples: pneumonia, acute severe asthma, drug overdose. Chronic respiratory failure occurs predominantly in patients with chronic lung diseases. Examples: COPD, restrictive lung disease (also sleep apnoea).

Type A:
Maintain relatively good blood gases and, in particular, keep arterial CO2 at a normal level despite severe airflow limitation. These have been called 'pink puffers'. They tend to be thin and present with dyspnoea as their major symptom.

Type B:
Have poor blood gases (in particular, elevated PaCO2) at an earlier stage of their disease and are more likely to develop right heart failure. These patients have been called 'blue bloaters' and are said to have a dominant history of cough and sputum, their underlying pathology being airways disease. These two patterns overlap considerably. The underlying difference is in the breathing control systems. In particular, the reflex hypoxic responsiveness (which is mediated by the carotid bodies), is depressed or absent in patients who develop CO2 retention. There is evidence that this reflex responsiveness increases in the patient who maintains normal CO2 levels.

To understand how these patterns of respiratory failure develop, the operation of the breathing pump must be reviewed. Muscle work must be done to achieve tidal volume and minute ventilation. This work is necessary to operate the pump to expand the lungs elastic work, and to force air through the conducting tubes resistive work. The control system, which determines how much effort is needed on each breath, is dominated by the chemoreflexes. Lung disease increases the load. For example, with obstructed airways a great driving pressure is needed simply to produce any given tidal volume; more work must be done by the respiratory muscles and more 'neural drive' is needed to sustain the effort. Thus, as lung disease progresses and the load to breathing increases, greater effort must be used simply to maintain tidal volume and minute ventilation. At some point in the progression of the lung disease, the load may become so great or the destruction of lung tissue so extensive, that it is simply impossible to maintain sufficient alveolar ventilation; and arterial CO2 must then rise.

Although there are many complex interactions between the type and extent of V/Q abnormality, the level of ventilation and the resultant blood gases, there are some rough guidelines which are of use clinically. For example, if the FEV1 is in the range of 0.5 litres (or less) then it is likely that severe respiratory failure with hypercapnia will develop, regardless of the control system - the lung damage is simply too severe.

However, if the FEV1 is around the 1.0 litre mark (or better), then patients should be able to maintain a normal arterial CO2. If the CO2 is elevated, the patient will usually have depressed chemoreceptor drive (as well as the lung disease). It is this group who fall into the clinical category of the 'blue bloated' respiratory failure. They will have a lower PaO2 (because of the added hypoventilation) and will die sooner than their counterparts of cor-pulmonale. Of the male patients in this group approximately 50% will have, in addition to their lung disease, upper airway obstructive sleep apnoea, and it is this additional factor (coupled with depressed chemoreceptor reflex drive) which causes the awake (as well as asleep) alveolar hypoventilation. These individuals often have marked falls in their oxygen saturation in sleep. The majority of these patients will be obese and will have a long history of snoring. As well as their extensive smoking history, they will often have a long history of very heavy alcohol consumption (usually well in excess of 50gm alcohol/day).

The other 50% of this group, although not displaying sleep apnoea currently, will often give a very similar history of past heavy snoring and heavy alcohol intake.

While alcohol and heavy snoring appear to be important factors in producing this type of respiratory failure, a significant proportion of such patients will be non-drinkers.

10.9 ADULT RESPIRATORY DISTRESS SYNDROME (ARDS)

a. Aetiology

• Sepsis (including pneumonia)
  
• Extensive burns
  
• Multiple fractures
  
• Disseminated intravascular coagulation
  
• Severe chest trauma
  
• Multiple transfusions
  
• Aspirations

b. Pathogenesis - production or activation of:

• Tumour necrosis factor
  
• C5a - activation of complement
  
• Arachidonic acid metabolites
  
• Free oxygen radicals

c. Overall function abnormalities

• Permeability pulmonary oedema
  
• PaO2 < 50 mmHg despite FiO2 > 60 %
  
• Respiratory compliance < 0.05 l/cm H20
  

Figure 10b