Browse the corpus

78 SECTION 3: Resuscitation produces an 8-mmol increase in [H +], with little change in bicarbonate concentration (usually 1 mEq/L) or urinary acid excretion. If the [H+] is higher or lower than that suggested by the change in the P co 2, a mixed disorder is present. The adaptation to chronic respiratory acidosis is complex. Over time, chronic elevation of Pco 2 reduces carotid sinus sensitivity to hypercap nia, and ventilatory drive is then controlled by Pao2. The acidosis results in significant increases in renal HCO3 – generation and avid reclamation of filtered HCO3 –. It is frequently uncertain whether a patient has an acute respiratory acidosis, a chronic respiratory acidosis, or a mixed disorder. Evaluation of the acid-base status in such circumstances does not require “baseline” arterial blood gas values. Instead, the change in [H +] is compared with the change in Pco 2. If this ratio is 0.3, the patient has a chronic respiratory acidosis; if it is 0.8, the patient has an acute respiratory acidosis. Other ratios suggest a mixed acid-base disturbance, as shown in Table 15-4.  TREATMENT Treatment of respiratory acidosis is the improvement of alveolar ventilation. In general, if the minute ventilation is doubled, the P co 2 will be reduced by 50%. Treat chronic obstructive pulmonary disease, and provide noninvasive ventilatory support or tracheal intubation as needed. See Chapter 29B, “Mechanical Ventilation, ” for further discussion. In patients with chronic respiratory acidosis, the arterial Pco 2 should not be reduced by more than 5.0 mm Hg/h. Rapid correction of a chronic respiratory acidosis can cause sudden development of a severe metabolic alkalosis, with resulting dysrhythmias. A rapid rise in pH can cause an abrupt decrease in ionized calcium levels and hypokalemia. RESPIRATORY ALKALOSIS Respiratory alkalosis is alveolar hyperventilation and exists when P co 2 is less than expected. It is caused by conditions that stimulate respira tory centers, including CNS tumors or stroke, infections, pregnancy, hypoxia, and toxins (e.g., salicylates). Anxiety, pain, and iatrogenic overventilation of patients on mechanical ventilators also cause respiratory alkalosis. The clinical symptoms of acute respiratory alkalosis are predictable from its physiologic effects. Acute reduction in P co 2 produces a reduction in [H +], resulting in an increase in negative charge on anionic buffers. The now negatively charged proteins instead bind calcium, and if the effect is sufficiently large, the reduction in ionized calcium produces tetany (e.g., carpopedal spasm) and paresthesias. 26 Hypocapnia also produces substantial reductions in cerebral blood flow and results in reduced tissue oxygen delivery due to a leftward shift in the oxygen–hemoglobin dissociation curve (i.e., increased hemoglobinoxygen binding). The predicted relationship of [H +] and P co 2 is that a 1-mmol decrease in [H+] results from each 1-mm Hg reduction in P co 2. Chronic respiratory alkalosis is unique among the acid-base disorders in that its compensation may be complete. Compensatory events include bicarbonaturia and a reduction in acid excretion, requiring 6 to 72 hours to develop fully and at least 1 week to normalize pH.

[H+] results from each 1-mm Hg reduction in P co 2. Chronic respiratory alkalosis is unique among the acid-base disorders in that its compensation may be complete. Compensatory events include bicarbonaturia and a reduction in acid excretion, requiring 6 to 72 hours to develop fully and at least 1 week to normalize pH. The steady-state TABLE 15-4 Evaluation of Acid-Base Status in Respiratory Acidosis *Ratio = Δ[H +]/ΔP co 2 Ratio < 0.3 Ratio = 0.3 0.3 < Ratio < 0.8 Ratio = 0.8 Ratio > 0.8 Change in hydrogen ion concentration is less than accounted for by chronic change in Pco 2. Metabolic alkalosis is also present. Change in hydrogen ion concentration matches chronic change in Pco 2. Chronic respiratory acidosis is present. Change in hydrogen ion concentration is larger than accounted for by chronic change in Pco 2. Chronic respiratory acidosis plus either acute respiratory acidosis or metabolic acidosis is present; examine pH. Change in hydrogen ion concentration matches acute change in Pco 2. Acute respiratory acidosis is present. Change in hydrogen ion concentration is larger than accounted for by acute or chronic change in Pco 2. Metabolic acidosis is also present. Abbreviation: Pco 2 = partial pressure of carbon dioxide. *See method of calculation in the Box, “Clinical Approach to Acid-Base Disorders,” in this chapter. relationship between [H +] and P co 2 in chronic respiratory alkalosis observed in normal human subjects at high altitude is roughly half that expected at sea level. Each 1-mm Hg reduction in P co 2 results in a 0.4-mmol reduction in [H+].  TREATMENT Treat the underlying cause. For patients with anxiety and hyperventila tion, do not use “paper-bag” rebreathing because it may lead to hypoxia. Chronic respiratory alkalosis is seen at high altitudes, in particular among mountaineers climbing over 3700 m (12,000 ft), where the partial pressure of oxygen is significantly diminished. Acetazolamide is frequently prescribed to counter the physiologic respiratory effects of such ascents. REFERENCES The complete reference list is available online at www.TintinalliEM.com. Blood Gases, Pulse Oximetry, and Capnography Casey M. Glass INTRODUCTION Blood gases provide important clinical information for patients with respiratory disorders, compromised circulation, or abnormal metabo lism. This chapter briefly reviews respiratory physiology to aid selection of assessment options and discusses appropriate use of arterial and venous blood gases and the advantages and limitations of noninvasive monitoring methods. The material focuses on the evaluation of oxygen and carbon dioxide levels; for information on carbon monoxide, please refer to Chapter 222, “Carbon Monoxide. ” RESPIRATORY PHYSIOLOGY Several factors contribute to overall gas exchange in the lungs. Each breath (tidal volume ) moves air in and out of the alveolus for gas exchange but also moves air through large airways and nonper fused areas of lung, the physiologic dead space . The physiologic dead space is approximately 30% of the tidal volume. The air remaining in the chest at the end of exhalation is the functional residual capacity . Dead space and the functional residual capacity do not contribute to gas exchange. Minute ventilation is a product of the respiratory rate and tidal volume. Relatively small changes in the usable alveolar space require CHAPTER Tintinalli_Sec03_p0053-0142.indd 78 8/2/19 2:57 PM

tion is the functional residual capacity . Dead space and the functional residual capacity do not contribute to gas exchange. Minute ventilation is a product of the respiratory rate and tidal volume. Relatively small changes in the usable alveolar space require CHAPTER Tintinalli_Sec03_p0053-0142.indd 78 8/2/19 2:57 PM CHAPTER 16:  Blood Gases, Pulse Oximetr y, and Capnogr aphy        79 Additionally, the alveolar space must be perfused by the pulmonary circulation. When portions of lung are perfused but not ventilated (as in pneumonia) or ventilated but not perfused (as in pulmonary embo lism), there is a ventilation-perfusion mismatch. Either scenario may lead to hypoxemia, the first as deoxygenated blood passes through the nonfunctional lung and mixes with oxygenated blood in the left atrium, and the second when too little perfused lung is available for adequate oxygen loading. The alveolar-arterial gradient or the P ao 2/Fio2 ratio can estimate the effectiveness of alveolar oxygenation. The alveolar-arterial gradient is the difference between the partial pressure of oxygen in the alveolar space (estimated from F io 2 at atmospheric pressure) and the measured partial pressure of the gas in an arterial blood gas sample. The following formula estimates the alveolar-arterial gradient [the P (A-a)O2]: P(A-a)O2 = [Fio2(Patm – PH2O) – (PaCO2 × 1.25)] – Pao2 Fio2 is the fraction of oxygen in the inspired air (21% if room air), P atm is the atmospheric pressure (760 mm Hg at sea level), P H2O is the vapor pressure at body temperature (47 mm Hg), and Pa CO2 and Pao 2 are the partial pressures of carbon dioxide and oxygen (respectively) from the patient’s blood gas. A normal gradient for young adults is <15 mm Hg. The gradient increases with age and is estimated with the following formula: age/4 + 4. The Pao 2/Fio2 ratio correlates with the relative venous shunt across the pulmonary circulation and is calculated as the measured Pao 2 divided by the F io2 in decimal form. A healthy person on 40% oxygen would be expected to have a ratio of approximately 600, representing a normal physiologic shunt of approximately 5%. As the shunt increases, the ratio decreases. Table 16-2 illustrates the decrease in the Pao 2/Fio2 ratio with increasing physiologic shunt in hypothetical patients all receiving oxygen with an Fio2 of 40%. The Pao2/Fio2 ratio is the most frequently used parameter for evaluating the severity of lung failure and is included in the current definition for acute lung injury/acute respiratory distress syndrome. 2,3 Carbon dioxide is transported by the red blood cells bound to hemoglobin and other proteins and dissolved in the plasma. Most carbon dioxide (60% to 70%) combines with plasma water to form carbonic acid, which dissociates into bicarbonate and hydrogen ions. Carbonic anhydrase in the erythrocyte catalyzes this near-instantaneous reaction. The deoxygenated hemoglobin is a willing receptor for released hydrogen ions. As hemoglobin is oxygenated in the lung, hydrogen ions are returned to the plasma, driving the reaction back toward the production of carbon dioxide. ARTERIAL BLOOD GAS ANALYSIS The amount of oxygen and carbon dioxide in the blood can be sampled and reported as the partial pressure of the gas. Blood gas analysis also typically includes a direct measurement of the serum pH and estimates of serum bicarbonate derived from the measured partial pressure of carbon dioxide (Pco 2) and pH (see Chapter 15, “ Acid-Base Disorders”). Current tests often include other useful information such as direct measurement of lactic acid as lactate, total hemoglobin, and serum electrolytes.

nt of the serum pH and estimates of serum bicarbonate derived from the measured partial pressure of carbon dioxide (Pco 2) and pH (see Chapter 15, “ Acid-Base Disorders”). Current tests often include other useful information such as direct measurement of lactic acid as lactate, total hemoglobin, and serum electrolytes. TABLE 16-1 Expected Pao2 in Patients Inhaling Various Concentrations of Oxygen (mm Hg) Fraction of inspired oxygen 0.21 (room air) 0.4 0.6 0.8 1.0 Expected Pao2 * (approximate) 105† 200 300 400 500 Abbreviation: Pao2 = partial pressure of arterial oxygen. *Assuming a patient with normal lung function at sea level and a partial pressure of carbon dioxide (Pco2) of 40 mm Hg. †Calculation demonstrates small overestimation using this technique. large increases in the minute ventilation to maintain the same rate of gas exchange. Raising the fraction of inspired oxygen (F io 2) or increasing the surface area or functional residual capacity of the lung can increase total alveolar oxygen content. Positive-pressure ventilation increases the functional residual capacity through recruitment of collapsed nonventilated alveolar space.  ESTIMATING OXYGEN DELIVERY The Fio2 is the percentage of oxygen in each breath. At sea level, room air is 21% oxygen. As F io2 increases, so does the alveolar concentra tion of oxygen (P ao 2). Each liter per minute of oxygen flow delivered via nasal cannula increases the F io2 by about 4%. Flow rates >4 L/min through a nasal cannula are poorly tolerated because of upper airway irritation, although some noninvasive devices can supply a higher F io via nasal cannula. A simple mask provides an Fio2 of 35% to 60% at flows of 10 to 15 L/min. A nonrebreather mask with a reservoir can deliver 95% Fio 2 with a supply flow rate of 10 to 12 L/min. The nature and severity of respiratory disorders can be estimated by comparing the measured arterial concentration of oxygen to the expected concentration of oxygen. The approximate arterial oxygen concentration (Pao 2) values that are expected in normal persons who are inhaling various concentrations of oxygen are listed in Table 16-1. The expected Pao 2 from supplemental oxygen can be roughly estimated by multiplying the actual F io2 by 5, the Five Times Rule. Thus, a patient getting 60% O 2 would be expected to have a Pao 2 of about 60 × 5, or 300 mm Hg. For every 1000-ft (305-m) rise in altitude, the atmospheric partial pressure of oxygen (Po 2) drops about 5 mm Hg, and the arterial oxygen estimate would therefore be reduced by that value. The total alveolar oxygen pressure cannot be greater than atmospheric pressure. See Chapter 214, “Diving Disorders, ” and Chapter 216, “High-Altitude Disorders, ” for further discussion of oxygenation and ventilation at altitude and depth. Cardiac output and the patient’s hemoglobin level are the biggest determinants of oxygen delivery to the tissues. Relatively little oxygen is dissolved in the plasma itself. Both hypoxia and anemia stimulate intrinsic mechanisms to increase cardiac output. If this is insufficient, increasing the hemoglobin level or oxygen saturation will improve the systemic delivery of oxygen. Based on survival data and the known complications of transfusions, the recommended threshold for transfu sion in the absence of acute bleeding is 7.0 grams/dL (70 grams/L). See Chapter 13, “ Approach to Traumatic Shock, ” and Chapter 238, “Transfusion Therapy, ” for further discussion of transfusion indications. Therefore, in most clinical situations, increasing the oxygen saturation of existing hemoglobin is the fastest method to increase oxygen delivery.  ALVEOLAR GAS EXCHANGE Once oxygen and carbon dioxide reach the lung, the gases diffuse across the interstitial space to either the red blood cell or alveolar space, respectively.

re, in most clinical situations, increasing the oxygen saturation of existing hemoglobin is the fastest method to increase oxygen delivery.  ALVEOLAR GAS EXCHANGE Once oxygen and carbon dioxide reach the lung, the gases diffuse across the interstitial space to either the red blood cell or alveolar space, respectively. The efficiency of diffusion depends on the distance across the alveolar-capillary membrane (interstitial space), the partial pres sure of the gas in the alveolar space, and the solubility of the gas. Both oxygen and carbon dioxide are highly soluble, and carbon dioxide is 20 times more soluble than oxygen. As a result, increases in the distance across the interstitial space (as in pulmonary edema, for example) have a greater effect on oxygen diffusion than carbon dioxide diffusion. Gas diffusion requires functioning alveolar space, and conditions that damage the alveoli prevent effective oxygen and carbon dioxide transport. TABLE 16-2 Interpretation of Pao2/Fio2 Pao2 (mm Hg) Fio2 (mm Hg) Ratio QS/QT (%) Impairment of Oxygenation 240 0.4 600 5 None 120 0.4 300 10 Minimal 100 0.4 250 15 Mild 80 0.4 200 20 Moderate 60 0.4 150 30 Severe * 40 0.4 100 40 Very severe* Abbreviations: Fio2 = fraction of inspired oxygen; Pao2 = partial pressure of arterial oxygen; QS/QT = venous-arterial admixture (shunt). *Ventilatory support and positive end-expiratory pressure to increase functional residual capacity and reduce the QS/QT to 15% should be considered. Tintinalli_Sec03_p0053-0142.indd 79 8/2/19 2:57 PM

o2 = fraction of inspired oxygen; Pao2 = partial pressure of arterial oxygen; QS/QT = venous-arterial admixture (shunt). *Ventilatory support and positive end-expiratory pressure to increase functional residual capacity and reduce the QS/QT to 15% should be considered. Tintinalli_Sec03_p0053-0142.indd 79 8/2/19 2:57 PM 80 SECTION 3: Resuscitation PvO2 20 40 60 80 100 Partial pressure O2 P50 PaO2 % Saturation O2 FIGURE 16-1. The oxyhemoglobin dissociation curve. This curve demonstrates the relationship of the partial pressure of oxygen (Po2) in the plasma to the saturation of hemoglobin molecules  with  oxygen  (O2). The  P50  is  the  Po2  at  which  hemoglobin  is  50%  saturated  and correlates with Po2 of 27 mm Hg normally. Normal mixed venous blood has an oxygen partial pressure (Pmvo2) of 40 mm Hg and an oxyhemoglobin saturation of 75%. A partial pressure of arterial oxygen (Pao2) of 60 mm Hg normally results in approximately 90% saturation of hemoglobin. not necessarily reflect normal arterial saturations and may overestimate arterial oxygen saturation.23 The pulse oximeter becomes more inaccu rate when arterial saturations range between 80% and 90%. 23-25 Carbon monoxide also causes hemoglobin to conform to the saturated state and will cause a false elevation in measured saturation (see Chapter 222). Pulse oximetry does not allow for determination of the partial pres sure of oxygen in the arterial blood, which, in combination with total hemoglobin content, is the primary determinant of peripheral oxygen delivery. A basic understanding of the oxygen saturation/dissociation curve is essential for practicing clinicians when using pulse oximetry to guide clinical therapy (Figure 16-1). Most notably, arterial oxygen content declines precipitously when the oxygen saturation falls below 92%. Studies of pulse oximetry typically use a finger-applied probe as the measurement site. Alternative locations such as the ear or forehead are used when it is difficult to obtain a reliable tracing from the patient’s finger. To date there is no definitive literature describing the accuracy of such measurements, although limited results suggest that probe locations on the head may be preferable to the extremities for patients with compromised perfusion. CAPNOGRAPHY Noninvasive techniques, direct transcutaneous measurement, cutane ous photometric measurement, or sampling of expired gases can easily estimate the carbon dioxide content of arterial blood. Of these, only sampling of expired gases is in common in the ED. The carbon dioxide content of expired gases at the end of the expiration phase of respiration (Etco 2) proportionally approximates the Paco2 content in healthy individuals, because carbon dioxide rapidly diffuses into the alveolar space, and arterial levels are typically only about 5 mm Hg higher than alveolar samples when pulmonary ventilation and perfusion are normal. The carbon dioxide content of the entire volume of expired gas can be photometrically assessed by means of a mainstream detector, or a sample of the expired gases can be aspirated from the airway via a side stream detector. Mainstream detectors are larger and must be inserted into the breathing circuit, either in a ventilator circuit or with the patient breathing through an occlusive mask or other tube. Side stream detectors are smaller and have the advantage of ease of use in nonintubated patients. Interpretation of the capnogram is not necessarily intuitive. A summary of typical capnogram waveforms is described in Figure 16-2. End-tidal carbon dioxide measurements (Petco 2) do not accurately represent Paco2.

r tube. Side stream detectors are smaller and have the advantage of ease of use in nonintubated patients. Interpretation of the capnogram is not necessarily intuitive. A summary of typical capnogram waveforms is described in Figure 16-2. End-tidal carbon dioxide measurements (Petco 2) do not accurately represent Paco2. Mainstream and side stream sampling may each include air that has not been involved in the gas exchange process, either from physiologic dead space or from environmental air that enters the sample. An arterial blood sample is the reference standard for pH, oxygen, carbon dioxide, and lactate content providing a description of the oxygen and carbon dioxide content of the blood after leaving the pulmonary circulation and before any gas exchange in the peripheral tissues has occurred. In scenarios that require precise determination of these vari ables, an arterial sample is necessary. Arterial blood gas samples are sometimes used for evaluation of serum hemoglobin and electrolytes. Blood gas analyzers typically have good concordance with the reference laboratory autoanalyzer; however, there may be clinically significant variances depending on the equipment involved, especially for sodium, hemoglobin, and chloride. Results should be interpreted with caution when they are significantly abnormal. 4-8 VENOUS BLOOD GAS ANALYSIS Venous samples may be either centrally or peripherally obtained. Mixed venous blood samples from the central circulation provide a source of information regarding the systemic uptake of oxygen and can be used to calculate cardiac output. The ideal sampling site for a systemic venous sample is at the pulmonary artery because blood from all body sites is equally represented, but such a sample is rarely practical. More com monly, blood is sampled from the superior vena cava or right atrium after placement of a central venous catheter. Blood from the superior vena cava disproportionally represents cerebral and upper body blood flow, but is generally useful for determining systemic oxygen uptake and lactic acid production. Peripherally obtained venous samples are widely used in emergency medicine, and they correlate with arterial values closely enough to be clinically useful. Significantly abnormal values should be confirmed with an arterial sample. 9-11 Venous blood gases are commonly used to monitor serum pH and, in certain situations, may be an appropriate substitute to monitor Pco2 and serum lactate. When compared to arterial blood gases, the venous pH correlates closely such that most differences are not clinically significant (+/– 0.05 pH units). 4-6,12,13 Venous carbon dioxide values trend along with arterial carbon dioxide, although they vary somewhat (up to +/– 20 mm Hg). Normal venous carbon dioxide is predictive of normal Paco 2; however, the clinical outcomes of substituting venous carbon dioxide for evaluation of hypercarbia have not been described in the literature. 11,14 Venous Pao2 values do not correlate with arterial oxygen content and cannot be used for evaluation of oxygenation. Regular monitoring of the venous oxygen saturation at the superior vena cava (S cvo2) or right atrium (S rao2) had been recommended as part of the Surviving Sepsis Campaign, but a failure to demonstrate a benefit from this approach led to a removal of that recommendation in the 2016 guidelines. 15,16 Measurement of the oxygen saturation from the superior vena cava may not accurately reflect the S rao2, and neither the S cvo2 nor the Srao2 may accurately reflect the mixed venous oxygen saturation at the pulmonary artery (Svo2) for critically ill patients.17 Venous measurements of serum lactate are widely used for patient care, but most studies describing the clinical utility of serum lactate reference an arterial sample.

neither the S cvo2 nor the Srao2 may accurately reflect the mixed venous oxygen saturation at the pulmonary artery (Svo2) for critically ill patients.17 Venous measurements of serum lactate are widely used for patient care, but most studies describing the clinical utility of serum lactate reference an arterial sample. Normal and markedly abnormal venous lac tate values correlate with the arterial lactate, but mildly elevated venous lactate may not correlate. In the latter case, if necessary, it may be wise to confirm the result with an arterial sample. 18-22 PULSE OXIMETRY It is not possible to directly measure the arterial oxygen concentration by noninvasive means, and the standard bedside tool for estimating arterial oxygen saturation is the photometric pulse oximeter. Pulse oximetry measures the relative absorption of oxygenated hemoglobin and deoxygenated hemoglobin and reports the percent age of oxygenated hemoglobin compared to the total. Pulse oximetry is fairly accurate, typically within 2% to 5% of the directly measured value performed at blood gas analysis. Importantly, the discrepancy between peripheral pulse oxygenation results and arterial gas saturation increases with hypoxia and poor circulation. Pulse oxygenation values greater than 92% are highly predictive of arterial saturations greater than 90%. Borderline normal pulse oxygenation values (between 88% and 92%) do Tintinalli_Sec03_p0053-0142.indd 80 8/2/19 2:57 PM