Given that the new batch of house officers have just started, it seemed like a good time to update this blog. Today’s post is the first in a two part series about the ABCs. When called to the bedside of an unwell patient, the advice given is to “attend to the ABCs” first. Without further explanation of what the ABCs actually entail, this is only minimally helpful. Today we shall focus on the first two letters.
The best way to start is to make explicit that the lung really only has two functions. The first is to maintain an appropriate blood level of oxygen. Appropriate can be considered to be an oxygen saturation somewhere in the 90s. Now, an oxygen saturation of 92% may not be normal, but it is generally sufficient for the purposes of oxygen delivery to organs. The second is to get rid of CO2. More specifically, to maintain a CO2 of 40mmHg (around 5kpa).
These two processes proceed very differently. To get oxygen into blood one needs an open alveolus and a capillary that supplies it. The blood flow through that capillary must be appropriately matched to the amount of oxygen in the alveolus- if the blood flow is too much for the amount of oxygen in the alveolus, the blood in that capillary will be desaturated. This explains why you can have patients who are much more hypoxic than you might expect from looking at their chest xray. Collapse of alveoli occurs on a microscopic level and is not necessarily visible on chest Xray. Many patients are on vasodilatory medications (antianginal or antihypertensives) that impair their ability to match capillary blood flow with the amount of oxygen in the alveolus. It can also be easy to miss collapse of an entire lobe on xray unless you know what you are looking for. The relevance of this is that it is easy to go on a wild goose chase of other causes of hypoxia if you believe that the Xray must correlate with the patient’s clinical condition. Consider the following case.
This (not real) elderly post operative day 1 patient has saturations of 92% on 10L of oxygen. He has clear bibasal atelectasis on chest Xray. Since it was believed that the chest Xray could not possibly explain the degree of hypoxia, he proceeded for an unnecessary CTPA with its radiation exposure, and risk of contrast reaction.
This entire left lower lobe is collapsed … easy to miss.
There are only two ways to correct hypoxia on the ward. One is to increase the amount of oxygen the patient is getting. That is why it is crucial to know what “dose” of oxygen the patient is getting. Patients approaching 50% inspired oxygen are sick and may not respond to further increases in oxygen, and may need to be transferred to the ICU. As a rule of thumb, nasal prongs cannot provide more than 30% oxygen at the very most. It is difficult to estimate what fraction of inspired oxygen one gets from a facemask at various flow rates other than that it varies greatly. This is where putting the patient on high flow nasal prongs comes in handy, because they display the fraction of inspired oxygen. Some caution is required with this, as this assumes that the patient is getting all their inspired air flow from the high flow device. Patients however may suck in room air around the nasal prongs if they do not fit snugly into their nose. Hence patients with big noses may be getting less inspired oxygen than the high flow is displaying. The same happens with mouth breathers. To see the FiO2 simply press the l> button on the box. A sure fire way to provide 100% oxygen is to let the patient breathe spontaneously through a bag valve mask with the oxygen flow cranked up to max.
The other way is to expand more alveoli. Most patients you see will be slumped over in bed. Simply sitting them up at 90 degrees (or better, in a chair) may improve oxygenation dramatically because the abdominal contents are not pushed up against the lungs. Getting pain under control to the point that the patient is able to cough up secretions and take deep breaths comfortably is crucial, hence there should be a low threshold to ask for help from the pain team. Draining excessive stomach contents with a nasogastric tube if appropriate can also produce huge results in oxygenation.
The other thing one can do to remedy poor oxygenation is to avoid vasodilatory drugs. Salbutamol is a potent vasodilator, so don’t give it unless you actually suspect bronchospasm. Driving people tachycardic with overzealous salbutamol also increases their oxygen demands/consumption, which may affect the oxygen saturations.
The removal of CO2 proceeds by much different mechanisms. CO2 that can’t escape through a collapsed alveolus will just escape through the ones that aren’t collapsed, as long as the bulk flow of air in and out of the lungs is adequate. Hence, CO2 removal is dependent on the bulk flow of air- the minute ventilation. If this falls to below normal, the CO2 will start to rise. As the CO2 concentration rises in the alveolus it will decrease the oxygen concentration, leading to hypoxia. This “competitive” effect is easily reversed by administering supplemental oxygen. Hence, hypercarbic patients will become hypoxic if they are breathing room air. However, if they are breathing extra oxygen, even as little as 2L of nasal prongs, you may not be alerted to rising CO2 levels through dropping oxygen levels. In fact, patients can become significantly hypercarbic and require only minimal supplemental oxygen. Consider the following case.
A (hypothetical) 32 year old has had an ankle fracture fixed. He has been given 30mg of IV morphine post-operatively. He is drowsy. You are reassured that he does not have respiratory depression by the fact his oxygen saturations are 96% on 2L nasal prongs. An hour later he has a respiratory arrest.
The mistake in this case was to assume that there is a correlation between hypoxia and the degree of hypercarbia. An arterial blood gas taken at the time of review would have shown a significantly raised CO2.
There are two corollaries to this. Firstly, any condition which acts to depress ventilation e.g. opiate narcosis, must cause hypercarbia. A hypoxic patient with a low or normal CO2 on blood gas does not have opiate narcosis as the cause of hypoxia.
The second is that any patient with an upper airway problem not involving the gas exchange surface of the lungs (a great example in children would be croup) is very very sick by the time they start to require any amount of supplemental oxygen. The only way for them to become hypoxic is by becoming hypercarbic, which implies the condition has progressed in severity to the point that they are no longer able to ventilate properly.
By all the same logic as above, patients may have partial airway obstruction without significant hypoxia. You will usually be alerted to a partially obstructed airway by noisy breathing. However as the degree of obstruction gets worse the noise will get quieter. Another giveaway will be “see-saw” breathing- as the abdomen moves out, the chest gets sucked in rather than moving out. However it can be difficult to tell sometimes when you arrive to an unconscious patient to tell if the airway is obstructed or not. The best thing to do is initiate simple airway manoeuvres (head tilt, chin lift, jaw thrust), and observe for improved chest movement and awakeness. The patient may push your hand away which is a good sign they do not have an obstructed airway. Other signs of a protected airway include being cursed at or spitting out airway adjuncts. A completely obstructed airway however will produce severe hypoxia and not make any noise (because there is no air at all getting to the lungs).
A patient unable to ventilate is in trouble. CO2 is the main driver of the sensation of shortness of breath and how fast and hard one breathes. Oxygen levels, in contrast, exert very little control over this. Hence the respiratory rate should be considered as the effort the patient needs to make to maintain their CO2 level. If one is breathing fast and the CO2 is low then there is an extra driver of the respiratory rate (usually the sensation of shortness of breath, or a metabolic acidosis). The patient’s lungs are performing their ventilatory functions just fine and they will be able to sustain this for a long time. Patients breathing fast with a high CO2 are in trouble. Their lungs are not ventilating adequately and they are highly likely to need some kind of assisted ventilation. Patients with CO2 in the normal range are a judgemental call. One thing to consider is that the lungs try to keep the CO2 at 5kPA (40mmHg) when the pH is normal. If the patient is acidotic, this CO2 target gradually drops in an effort to compensate the acidosis. A patient with a metabolic acidosis and a pH of 7.25 should achieve a CO2 of about 25mmHg (3.5kpa). Hence a patient in this situation with a normal CO2 and tachypnea is hypoventilating and quite likely to progress to respiratory arrest.
This may be a bit difficult to digest. To summarise, patients try their hardest to maintain a normal CO2, and a progressively lower CO2 as they get more acidotic. This is to avoid worsening acidosis and a death spiral. Patients will increase their depth and rate of breathing to achieve this. Those who can’t achieve this despite increased respiratory effort have ventilatory muscle insufficiency and are likely to quickly succumb to progressive respiratory fatigue, acidosis and death. Those who are overachieving this are likely to be able to sustain their respiratory efforts indefinitely. Tachypneic and drowsy patients should be assumed to have high CO2 until proven otherwise.
On the other hand there will be patients with high CO2 who are not breathing faster and harder to eliminate this. These patients have respiratory depression- either caused by medications or acute pathology such as an encephalopathy or by diseases that give them a chronically elevated CO2 and cause them to readjust their CO2 “set point”. Patients with acute respiratory depression need urgent attention. Patients with chronic lung diseases who are at their baseline CO2 are identified by a pH that is generally > 7.35 and a high bicarbonate. Patients with pH <7.3 generally have some acute element of respiratory failure. Those residing between 7.3 and 7.35 can be difficult to tell and require some degree of clinical judgement.
The other thing to take into account in certain situations is the degree of CO2 production. You will see patients not infrequently who are in the midst of a temperature spike who become very tachypneic. They are generally very febrile, up to 39 or 40 degrees Celsius and with rigors. CO2 production is greatly increased when you are that hot and shaking, and the normal response to this CO2 load is to breathe it off. Hence it is quite normal to get quite tachypneic in this situation in order to maintain a normal CO2, as long as the patient is alert and does not have worsening hypoxia. Once the temperature has reached that high, it does not take long for it to break. Usually one returns in an hour to find the temperature has broken and the patient looks just fine. The clue here is that generally the patient is breathing quite fast, but the degree of respiratory muscle activation is not high.
Respiratory muscle activation just refers to how hard the muscles of breathing need to contract to achieve adequate ventilation. Other people call it the work of breathing, but technically this terminology includes the respiratory rate as well. It is incorrect to say someone has no work of breathing. This means they are dead. By placing a finger over the sternocleidomastoid belly and then over the trachea, as well as observing the degree of indrawing, one can judge the degree of muscle activation. People using a lot of their accessory muscle will generally demonstrate paradoxical breathing i.e. the abdomen moves in as the chest moves out. A patient with lots of respiratory muscle activation has worse lung disease than one with minimal and is more in trouble.
It can be difficult to distinguish between atelectasis and a pneumonia. One clue on Xray is that the lung volume shrinks when it is atelectasis, and stays normal or is expanded when it is a pneumonia. See the Xray above for an example of reduced lung volumes on the left side. You will encounter this dilemma in post operative patients. Generally speaking, post operative day 1 patients who were not hospitalised prior to their surgery do not develop pneumonias so quickly. The biggest clue is the rate of resolution. An opacity that has resolved the next day on repeat Xray is not a pneumonic consolidation.
Pulmonary oedema can be subtle on Xray. The only sign may be upper lobe diversion. The Xray below is of a hypothetical patient who was misdiagnosed with bilateral pneumonia, showing the features of pulmonary oedema.
Bilateral pleural effusion should generally be assumed to be on the basis of cardiac failure. It is important to remember that volume overload is a nebulous term. Many patients in pulmonary oedema are not volume overloaded, but rather the fluid is backed up in the wrong places. The feature that makes pulmonary oedema most likely is a documented history of poor LV function. Patients come in two flavours- the generally oedematous patient who has been getting fluid overloaded over a number of days and week and is generally oedematous, who benefits most from frusemide. The other is probably more common- the patient with relatively acute pulmonary oedema that develops as a result of sympathetic stimulation causing increased afterload on the left ventricle. They are generally hypertensive and NOT oedematous, and benefit most from some sublingual GTN first to vasodilate them (assuming they are not hypotensive of course). The pulmonary oedema often clears without diuretics. Extremely rapid onset of respiratory distress should make one strongly suspect pulmonary oedema and rapid improvement with vasodilators is more diagnostic of pulmonary oedema than any BNP or chest Xray. A patient does not need to have a reduced ejection fraction to develop pulmonary oedema. Patients with diastolic heart failure are just as likely to develop it. When scouring an echo report one has to pay attention to specific comments about diastology. A dilated left atrium on a previous echo report means that the patient is predisposed to developing pulmonary oedema- they have some kind of pathology that is stretching the left atrium.
That’s all for now, next time we will talk about the letter C…