[rs] Pneumonia With Severe Hypoxia

THE RESPIRATORY SYSTEM

Pneumonia with Severe Hypoxia

Mr. D., a previously healthy 35-year-old man, is seen in the emergency department for cough and shortness of breath; the evaluation takes place at sea level. Chest x-ray examination shows a large infiltrate, consistent with the clinical diagnosis of pneumonia, in his left lung. When the patient is breathing ambient air (FIO2 is 0.21), his arterial PO2 (PaO2) is 52 mm Hg, PaCO2 is 39 mm Hg, and pH is 7.42. Minute ventilation is 12 L/min.

1. Listed below are the major physiologic mechanisms that can diminish PaO2. For each mechanism, explain how it may or may not play a role in causing this patient's low PaO2. Ventilation/perfusion imbalance Right to left intrapulmonary shunt Diffusion barrier to gas transfer Hypoventilation

Ventilation/perfusion imbalance. V/Q imbalance is found in virtually all forms of airway and parenchymal lung disease (asthma, pneumonia, emphysema, bronchitis, etc.). Areas of low V/Q ratios (one part of the spectrum of V/Q imbalance) cause a low PO 2 that cannot be compensated for by other areas of high V/Q ratios. In other words, low and high V/Q areas do not average out the pulmonary capillary PO2 values, so that the final result is a lower than normal PaO2 and a widened alveolararterial PO2 difference.

Using the alveolar gas equation presented in the questions, we can calculate the patient's PAO2:

PAO2 = 0.21* (760 - 47) - 40 * (0.21 + (1 -0.21)/0.8) = 102 mm Hg

With normal blood gases, PaO2 would be about 90 mm Hg, and P(A-a)O2 about 12 mmHg. His P(A-a)O2 is increased to 102 - 52 or 50 mm Hg, caused by pneumonia and the resulting V/Q imbalance. The V/Q imbalance in this case might also encompass some actual right to left shunting, i.e., areas of perfusion with no ventilation [V/Q = 0]).

Right to left intrapulmonary shunt. A right to left shunt can be anatomic or physiologic. An anatomic right to left shunt (e.g., a connection between a pulmonary artery and pulmonary vein that bypasses capillaries) is "fixed." Based on the clinical information (previously healthy, now acutely ill), his low PaO2 is probably not from an anatomic shunt (which is rare in any case).

1

A physiologic right to left shunt arises from parenchymal lung disease, for example, pneumonia or pulmonary edema. Ventilation to affected areas of the lung is effectively blocked. Some perfusion continues to that area, but it does not take part in gas exchange (i.e., the blood is shunted past the unventilated alveoli). This shunted blood (with venous PO2 and PCO2 values) returns to the left side of the heart. As previously pointed out, a right to left intrapulmonary shunt is one extreme of V/Q imbalance (V/Q = 0).

Diffusion barrier. Diffusion block or barrier across the alveolar capillary membranes does not cause much hypoxemia in resting patients. The reason is that the reserve for oxygen transfer is quite large; normally, hemoglobin is fully loaded with oxygen after the blood has travelled about 1/3 of the way through the pulmonary capillaries. Even with a diffusion barrier, such as from fluid in the lung interstitium, oxygen usually has time to cross the alveolar-capillary membrane and fully saturate hemoglobin. The exception is if blood flow is markedly accelerated through the capillaries, as happens with strenuous exercise; then a diffusion barrier may lead to oxygen desaturation and hypoxemia.

Diseases that cause diffusion block also lead to V/Q imbalance (e.g., pulmonary edema, pneumonia), so that both physiologic abnormalities are commonly present. In such cases diffusion block is less important then V/Q imbalance in causing hypoxemia.

Hypoventilation. Hypoventilation is defined by a high PaCO2, which this patient does not have; his PaCO2 is low, so hypoventilation cannot be causing his hypoxemia.

In summary, the cause of his hypoxemia is V/Q imbalance, which might encompass some right to left intrapulmonary ("physiologic") shunting.

2. Does this minute ventilation, together with the PaCO2, indicate an increase in wasted ventilation? How is an increase in wasted ventilation explained by ventilation/perfusion imbalance?

His increased minute ventilation of 12 L/min (normally about 6 L/min at rest) with a normal PaCO 2 indicates an increase in wasted ventilation. Physiologically, there are two possible causes: panting and ventilation/perfusion imbalance. Because his respiratory rate is not given, it is possible that he was breathing, for example, 50 times a minute at 200 ml/breath. Because normal anatomic dead space is about 150 ml, a tidal breath of 200 ml would give him an alveolar volume of only 50 ml/breath or an alveolar ventilation of 2.5 L (normally about 4 L/min at rest). Although hypoventilation from panting can theoretically occur without any parenchymal lung disease, it is extremely unlikely. A much more likely explanation is that V/Q imbalance has created dead space in previously normal lung tissue, by reducing perfusion more than ventilation. This is very common, particularly in patients with chronic lung diseases, such as emphysema. Because inhaled air will go to alveolar spaces whether or not they are perfused, much of each tidal volume is wasted (i.e., it goes to poorly ventilated or unventilated alveoli). Thus, V/Q imbalance causes not only hypoxemia through creation of low V/Q areas, but also wasted ventilation through creation of high V/Q areas. The latter is a principal explanation for the increased work of breathing in many patients with V/Q imbalance. They have to breathe more to bring in enough air for adequate gas exchange.

2

3. While he is breathing 50% oxygen via a face mask, the following blood gas results are obtained: PaO 2 is 65 mm Hg, PaCO2 is 35 mm Hg, and pH is 7.46. Do these results change your assessment of the physiologic cause of his hypoxemia?

Mr. D. is admitted to intensive care and continues to manifest severe hypoxemia. Overnight the pneumonia spreads to involve his entire left lung, but it spares the right lung. He is now receiving 100% inspired oxygen via face mask, plus broad spectrum antibiotics. He remains awake and alert, and breathes at a rate of 28 times per minute. Arterial blood gases show the following: PaO 2 is 60 mm Hg, and PaCO2 is 30 mm Hg.

When the patient breathes 50% oxygen, his PaO2 is only marginally better. This information does not indicate any different physiologic causes, but does implicate right to left shunting more directly as a significant feature. Without any V/Q imbalance, his PaO2 would increase by an amount predicted by the increase in FIO2. Again by the alveolar gas equation:

PAO2 = 0.5 (760 - 47) - 40 (0.5 + (1 - 0.5)/0.8) = 317 mm Hg

The variables in the equation show that PAO 2 is not affected by V/Q imbalance (except to the extent V/Q imbalance might affect the PaCO2). His calculated P(A-a)O2 is 317 - 65 = 252 mm Hg. With normal lungs his PaO2 would be much closer to the PAO2; when he breathes 50% oxygen, PAO 2 should normally rise to at least 250 mm Hg. The fact that his PaO2 improved only slightly points to some degree of right to left shunting; the extra inhaled oxygen is simply not getting into his blood. We cannot reliably estimate the degree of shunt unless the subject is breathing 100% oxygen; on less than 100% oxygen, the poorly ventilated units cannot be distinguished from those not ventilated at all.

4. Based on the above information, what is the approximate percentage of the right to left shunt?

When a patient breathes 100% oxygen, we can estimate the percentage of right to left shunt because all units of V/Q >0 should contain pure oxygen and should saturate the blood that perfuses them. A PaO2 lower than predicted when the patient breathes 100% oxygen reflects a right to left shunt. The percentage of shunt can be estimated as 1% of cardiac output for each 20 mm Hg P(A-a)O2. First, calculate PAO2.

PAO2 = 1

*

(760 - 47) - 30 (1 + (1 - 1)/0.08) = 673 mm Hg

Next, calculate P(A-a)O2; in this case it is 673 - 60 = 613 mm Hg. His estimated shunt is therefore 613/20 = 31 %. About 31 % of his cardiac output is blood that flows through his lungs but does not become oxygenated. One point to ponder: if he has such a large right to left physiologic shunt, why is his PaCO2 not elevated (i.e., how is he able to exchange CO2 so well but not O2)?

5. These blood gases are obtained while he is lying on his left side. His physician, concerned about the low PO2, is considering tracheal intubation and mechanical ventilation. However, the physician first decides

3

to change the patient's body position to improve PaO2. What is the change asked for, and why? Mr. D. appears in the emergency department a year later. He took an overdose of sleeping pills and is comatose. His chest x-ray examination is clear and shows no residual of the pneumonia. Vital signs reveal a normal blood pressure, pulse is 100 beats/min and regular, and respirations are 6 per minute and shallow. A friend states that the patient has been despondent over domestic problems, and he has no prior history of drug overdose. Arterial blood gases show that pH is 7.34, PaO 2 is 55 mm Hg, and PaCO2 is 70 mm Hg.

The physician asks the patient to lie with his right side down. The aim is to improve perfusion to the good lung. Because his left lung is involved with pneumonia, most of his ventilation is to the right lung. Unlike ventilation, perfusion is gravity dependent. Increasing perfusion to the right lung should improve V/Q and therefore PaO2. In fact, this change does help, and his PaO 2 increases to 77 mm Hg; intubation may therefore be avoided.

6. With reference to the same four mechanisms of hypoxemia listed in question 1, explain the patient's low PaO2.

He has a high PaCO2 and therefore is hypoventilating. First use the alveolar gas equation to calculate his PAO2.

PAO2= 0.21 * (760 - 47) - 70 * (0.21 + (1- 0.21)/0.8) = 65 mm Hg

Next, subtract his measured PaO2 from the calculated PAO2; 65 - 55 = 10 mm Hg. Because (PAO2 PaO2) is normal, the cause of his low PaO2 is simply hypoventilation (high PaCO2) from the drug overdose, not a V/Q imbalance. 7. What is the principal physiologic process that needs correction in this patient: ventilation or oxygenation? What would be appropriate treatment? The principal physiologic problem is inadequate ventilation, which is solely responsible for the hypoxemia. A comatose patient with inadequate ventilation usually requires intubation and mechanical ventilation. Giving supplemental oxygen alone, without augmenting ventilation, would transiently improve PaO2but not PaCO2. His PaCO2could actually increase further while he receives oxygen, and lead to a fatal acidosis.

4