Hemoglobin (Hb) is responsible for this dramatic increase in blood’s O₂-carrying capacity. Hb consists of four polypeptide chains, each with a heme (iron-containing) protein that can bind O₂ with iron in the ferrous (Fe²⁺) form. Methemoglobin results when iron is in the ferric form (Fe³⁺) and cannot bind O₂. Small amounts of methemoglobin normally occur in blood and slightly reduce the amount of O₂ that can be bound to Hb. One gram of pure adult Hb can bind 1.39 mL of O₂ when fully saturated, but methemoglobin reduces this value to 1.34-1.36. The cellular packaging of Hb is important for the biophysics of the microcirculation, and it provides physiologic control of O₂ binding through cellular changes in the Hb microenvironment.

The four subunits of Hb include two α– and two β-chains, and variations in the amino acid sequence of these polypeptides explain the differences in Hb-O₂ affinity between species and at different stages of development. The threedimensional shape of an Hb molecule, which is determined by the allosteric interactions of its four subunits, causes the O₂-equilibrium curve to be S-shaped, or sigmoidal (Fig. 43.2). O₂-equilibrium curves for individual α– and β-chains are not sigmoidal but simple convex curves similar to the O₂-equilibrium curve for myoglobin. Myoglobin occurs in muscle and has only a single polypeptide chain with one heme group. The sigmoidal shape of the O₂-Hb equilibrium curve facilitates O₂ loading into blood in the lungs and O₂ unloading from blood in the tissues.

The two forms of the O₂ equilibrium curve are shown in Figure 43.2: (a) saturation of Hb with O₂ (SO₂) versus PO₂, and (b) O₂ concentration in blood (CO₂) versus PO₂. Saturation quantifies the amount of O₂ in blood as the percentage of the total Hb sites available for binding O₂ that actually bind O₂ at a given PO₂. Therefore, saturation equilibrium curves are independent of Hb concentration in blood. In contrast, concentration curves quantify the absolute amount of O₂ in a volume of blood with a given PO₂, and they depend on the amount of Hb available.

O₂ capacity (O_2cap) is defined as the O₂ concentration in blood when Hb is 100% saturated with O₂. Pure Hb binds 1.39 mL O₂/g Hb. Figure 43.2 illustrates that the O_2cap for normal blood with Hb concentration of 15 g/dL is 20.85 mL/ dL (1.39 × 15). Physically dissolved also contributes a small amount to O₂ concentration. Therefore, total O₂ concentration in blood (in mL O₂/dL blood) is calculated as:

Co₂ = (O_2cap × [So₂/100]) + (0.003 × PO₂)

where O_2cap is the O₂ capacity, So₂ the saturation, and 0.003 is the physical solubility for O₂ in blood in (mL/dL)/mm Hg.

The shape of the O₂-Hb equilibrium curves is complex and can be generated only experimentally or by sophisticated mathematical algorithms. However, remembering only four points on the normal adult curve allows one to solve many common problems of O₂ transport:

a. PO₂ = 0 mm Hg, SO₂ = 0% (the origin of the curve)

b. PO₂ = 100 mm Hg, SO₂ = 98% (normal arterial blood, which is almost fully saturated)

c. PO₂ = 40 mm Hg, SO₂ = 75% (normal mixed-venous blood)

d. PO₂ = 26 mm Hg, SO₂ = 50% (i.e., P₅₀)

The P₅₀ quantifies the affinity of Hb for O₂ as the PO₂ at 50% saturation under standard conditions of partial pressure of CO₂ (PCO₂) = 40 mm Hg, pH = 7.4, and 37°C. P₅₀ is only 20 mm Hg in the human fetus, as discussed below. A decrease in P₅₀ indicates an increase in O₂ affinity because SO₂ or CO₂ is greater for a given PO₂.

The O₂-equilibrium curve can be physiologically modulated in three ways: (a) the vertical height of the concentration curve (but not the saturation curve) can change, indicating a change in O_2cap, (b) the horizontal position of saturation and concentration curves can change, indicating a change in Hb-O₂ affinity, and (c) the shape of saturation and concentration curves can change, indicating a change in the chemical reaction between O₂ and Hb. The maximum height of the saturation curve cannot change by definition; the maximum is always 100% when O₂ is bound to all available Hb sites. However, changes in Hb concentration [Hb] will change the maximum height of the concentration curve, according to the relationship between O₂cap and [Hb] as described previously. Mean corpuscular Hb concentration (MCHC) quantifies [Hb] in red blood cells and hematocrit (Hct) quantifies the percentage of blood volume that is red blood cells. Therefore, [Hb], in g/dL of blood, depends on both of these factors:

[Hb] = MCHC. Hct

Typical adult values of MCHC = 0.33, Hct = 45%, and [Hb] = 15 g/dL are used in Figure 43.2, which shows normal CaO₂ = 20 mL/dL and that CaO₂ = 15 mL/dL in mixed-venous blood (C[V with bar above]_oxygen). If [Hb] decreases, for example with decreased Hct in anemia, O_2cap and concentration decrease at any given PO₂. The O₂cap increases when [Hb] increases
by the stimulation of red blood cell production in bone marrow by the hormone erythropoietin or by transfusion. Erythropoietin transcription is regulated by hypoxia-inducible factors (HIF-1 and 2) released from cells in the kidneys in response to decreases in arterial O₂ levels. Polycythemia (increased Hct) occurs with chronic hypoxemia in healthy people (e.g., during acclimatization to altitude) and with disease.

The horizontal position of the Hb-O₂ equilibrium curves reflects the affinity of Hb for O₂, and changes in horizontal position are quantified as changes in P₅₀. A decrease in P₅₀ is referred to as a left shift of the equilibrium curve and indicates increased Hb-O₂ affinity; SO₂ or CO₂ is increased for a given PO₂. Similarly, increased P₅₀, or a right shift, reflects decreased Hb-O₂ affinity. The three most important physiologic variables that can modulate P₅₀—pH, PCO₂, and temperature—are shown in Figure 43.3.

The Bohr effect describes changes in P₅₀ with changes in blood PCO₂ and pH. Decreased PCO₂ causes Hb-O₂ affinity to increase (decreased P₅₀), and increased PCO₂ causes Hb-O₂ affinity to decrease (increased P₅₀). As described later, pH decreases when PCO₂ increases and vice versa, and pH changes explain most of the Bohr effect with PCO₂ changes in blood. Hydrogen (H⁺) binds to histidine residues in Hb molecules, thereby changing the conformation of Hb and the ability of heme sites to bind O₂. However, CO₂ also has a small, independent effect on Hb-O₂ affinity if pH is held constant. The physiologic advantage of the Bohr effect is that it facilitates loading in the lungs, where CO₂ is low and pH is high. In the tissues, the opposite occurs, and increased CO₂ causes pH to decrease and facilitates O₂ unloading from Hb to the tissues. The effect of temperature on Hb-O₂ affinity also has physiologic advantages. Warm temperatures in intensely exercising muscles will increase P₅₀ and decrease Hb-O₂ affinity to facilitate O₂ unloading to tissues.

A physiologic O₂-blood equilibrium curve can be defined as the curve that shows the change in blood O₂ concentration when PO₂ decreases from arterial to venous levels in the tissues or increases in the opposite direction in the lungs. The increase in PCO₂, decrease in pH, and potential increase in temperature between arterial and venous points make the physiologic curve steeper than individual curves (Fig. 43.3). This is an advantage for gas exchange because it increases the change in O₂ concentration for a given change in PO₂, thereby increasing O₂ uptake or delivery (see the section, Cardiovascular and Tissue Oxygen Transport).

Hb-O₂ affinity is also affected by organic phosphates, with 2,3-diphosphoglycerate (2,3-DPG) being most important in humans. 2,3-DPG is produced during glycolysis in red blood cells and increases P₅₀ by interacting with Hb β-chains to decrease their O₂-binding affinity. Physiologic stimuli that lead to enhanced O₂ delivery (e.g., chronic decreases in blood PO₂ levels) typically increase the concentration of 2,3-DPG and promote O₂ delivery to tissues. In blood stored in blood banks, 2,3-DPG is generally decreased, and the increased Hb-O₂ affinity can lead to problems in O₂ delivery after blood transfusion.

Carbon monoxide (CO) is a deadly gas that also modulates the Hb-O₂ dissociation curve by changing the shape and position of concentration or saturation curves. The affinity of Hb for CO is 240 times greater than it is for O₂, so even very small amounts of CO greatly reduce the capacity for Hb to bind O₂—that is, the O_2cap. However, CO also decreases the P₅₀ and makes the Hb-O₂ curve less sigmoidal. CO causes a left shift of the curve by altering the ability of the Hb molecule to bind O₂; therefore, blood O₂ concentration remains high until Po₂ decreases to very low levels, which impairs O₂ unloading from blood to tissues. CO poisoning also has direct effects on cellular cytochromes, which contribute to its deadliness. CO is particularly dangerous because it is colorless and odorless, and the decrease it causes in arterial O₂ concentration is not sensed by respiratory control systems, which respond only to O₂ partial pressure, as explained in the Respiratory Control section. Hyperbaric O₂ exposure is used to treat CO poisoning because only very high Po₂ levels are effective at competing with CO for Hb-binding sites and driving CO out of the blood.

The normal human fetus is exposed to a level of O₂ that would be considered severe hypoxia in adults, similar to the levels experienced on the summit of Mt. Everest! This is possible not only because the fetus is less sensitive to hypoxia than adults, but also because of fetal Hb (HbF). HbF has a P₅₀ of 20 torr, compared to 26 torr for adults, thus facilitating O₂ transfer to the fetus in the placenta (Fig. 43.4). HbF is gradually replaced by adult Hb in the first year. O₂ delivery across the placenta is ˜8 mL O₂/min/kg of fetal mass, which is approximately twice the rate for an adult, but blood O₂ stores in the fetus are sufficient for only a few minutes of metabolism. O₂ delivery across the placenta is limited by blood flow and PO₂ levels in the mother and fetus, but not by diffusion. For example, decreasing maternal PaO₂ below 70 torr can reduce placental O₂ transfer. Increasing maternal PaO₂ to 600 torr, with O₂ breathing, can increase fetal umbilical O₂ tensions by 3-5 torr, which can be important if the fetus is suffering any hypoxic stress. Both the left shift of fetal versus adult O₂-Hb equilibrium curves and the high fetal O₂ capacity facilitate O₂ unloading from the mother to the fetus (Fig. 43.4). It is also significant that O₂ capacity is decreased in the mother near term (11.5 vs. 14 g/dL Hb). The Bohr effect also contributes to placental O₂ transport, as pH decreases from 7.42 in the maternal artery to 7.35 in the maternal vein.

Physically dissolved CO₂ is a function of CO₂ solubility in plasma, which is 0.067 mL/(dL mm Hg) and 20 times more soluble than O₂. Still, dissolved CO₂ contributes only ˜5% of total CO₂ concentration in arterial blood.

Carbamino compounds comprise the second form of blood CO₂. These compounds occur when CO₂ combines with amine groups in blood proteins, especially with the globin of Hb. However, this chemical combination between CO₂ and Hb is much less important than Hb-O₂ binding; therefore carbamino compounds comprise only 5% of the total CO₂ in arterial blood.

Bicarbonate ion HCO^–₃ is the most important form of CO₂ carriage in blood. CO₂ combines with water to form carbonic acid, and this dissociates to HCO^–₃ and H⁺:

CO₂ + H₂O ↔ H₂CO₃ + HCO^–₃ + H⁺

Carbonic anhydrase is the enzyme that catalyzes this reaction, making it almost instantaneous. Carbonic anhydrase occurs mainly in red blood cells, but it also occurs on pulmonary capillary endothelial cells and accelerates the reaction in plasma in the lungs. The uncatalyzed reaction will occur in any aqueous medium, but at a much slower rate, requiring >4 minutes for equilibrium. The rapid conversion of CO₂ to bicarbonate results in ˜90% of the CO₂ in arterial blood that is carried in that form and has important implications for acid-base balance.

Figure 43.6 shows the carbonic acid reactions in plasma and red blood cells and illustrates important ion fluxes that occur with CO₂ transport in blood. CO₂ rapidly enters red blood cells from the plasma because it is soluble in cell membranes. Carbonic anhydrase catalyzes the rapid formation of HCO.^–₃ and H⁺ in the cells and an electrically neutral bicarbonate-chloride exchanger moves some of this HCO^–₃ out of
the cell. The chloride shift (Hamburger shift) is an increased intracellular chloride level with increased CO₂, or vice versa. The H⁺ produced from CO₂ reacts with Hb and affects both the O₂ equilibrium curve (Bohr effect) and CO₂ equilibrium curve, as described next.

Hb-O₂ saturation is the major factor affecting the position of the CO₂ equilibrium curve. The Haldane effect increases CO₂ concentration when blood is deoxygenated or decreases CO₂ concentration when blood is oxygenated at any given PCO₂ (Fig. 43.5). The Haldane effect is actually another view of the same molecular mechanism that causes the Bohr effect on the O₂ equilibrium curve (described previously). H+ ions from CO₂ can be thought of as competing with O₂ for Hb binding. Hence, increasing O₂ decreases the affinity of Hb for H+ and blood CO₂ concentration (Haldane effect), and increased [H+] decreases the affinity of Hb for O₂ (Bohr effect). These interactions are summarized in Figure 43.6.

The Haldane effect promotes unloading of CO₂ in the lungs when blood is oxygenated and CO₂ loading in the blood when O₂ is released to tissues. The Haldane effect also results in a steeper physiologic CO₂-blood equilibrium curve (see Fig. 43.5), which has the physiologic advantage of increasing CO₂ concentration differences for a given PCO₂ difference.

Rearranging the alveolar ventilation equation shows how [V with dot above]A and PACO₂ (or PACO₂) are inversely related for any given metabolic rate:

PACO₂ = ([V with dot above]CO₂/[V with dot above]A)K

Hence, if [V with dot above]A is doubled, PACO₂ is halved, regardless of the exact values for either variable. Hyperventilation is defined by a decrease in PaCO₂ from the normal value, implying excess [V with dot above]A for a given [V with dot above]CO₂. Hypoventilation is defined by an increase in PaCO₂, and this occurs when [V with dot above]A is lower than normal for a given [V with dot above]CO₂.

As explained previously, [V with dot above]A is reduced from total ventilation by the amount of the dead space:

[V with dot above]A = fR ([V with dot above]T – [V with dot above]DS)

Although anatomic dead space can be measured with gas analyzers and flow meters, it is more clinically relevant to estimate physiologic dead space, which is also called Bohr dead space. Physiologic dead space includes all “wasted ventilation” and can exceed anatomic dead space, as described next. Physiologic dead space can be calculated from another rearrangement of the Fick principle applied to CO₂ elimination by the lungs as:

VDS/VT = (PACO₂ – PĒCO₂)/PACO₂

where PeĒCO₂ = mixed-expired PCO₂ that is measured by collecting all expired gas in a bag or a spirometer and includes gas exhaled from the alveoli and dead space. In practice, arterial PCO₂ is substituted for alveolar PCO₂ because it is easily measured.

Alveolar PO₂ (PAO₂) can be predicted from the alveolar gas equation that models an “ideal lung” with only physiologic dead space:

PAO₂ = PIO₂ – (PACO₂/R) + F

where PIO₂ is inspired PO₂, PACO₂ is alveolar PCO₂, R is the respiratory exchange ratio, and F is a constant that can be ignored under normal conditions (F = PACO₂ × FIO₂ [1 – R]/R and increases PAO₂ only 2 mm Hg at normal O₂ and CO₂ levels).

In practice, arterial PCO₂ is substituted for alveolar PCO₂ because PaCO₂ = PaCO₂ in normal lungs and arterial samples are easily obtained. The alveolar gas equation is only valid if inspired PCO₂ = 0, which is a reasonable assumption for room air breathing.

The respiratory exchange ratio (R) is the ratio of uptake to CO₂ elimination by the lungs:

R = [V with dot above]CO₂/[V with dot above]O₂

Under steady-state conditions, R equals the respiratory quotient (RQ), which is the ratio of CO₂ production to O₂ consumption in metabolizing tissues. RQ averages 0.8 on a normal, mixed, adult diet, but it can range from 0.67 to 1, depending on the relative amounts of fat, protein, and carbohydrate being metabolized. R can exceed this range in nonsteady states, for example, when R exceeds 1 during hyperventilation. CO₂ stores in the body are much greater than O₂ stores because of bicarbonate in blood and tissues. R can increase because it takes longer to wash out the CO₂ stores than it does to charge up the much smaller O₂ stores in the body.

Substituting normal adult values in the alveolar gas equation predicts that PAO₂ = 100 mm Hg in breathing room air (PAO₂ = 150 – 40/0.8). Increases in [V with dot above]A (hyperventilation) increases PAO₂ by decreasing PACO₂, whereas decreases in [V with dot above]A (hypoventilation) decreases PAO₂. Why ideal alveolar PO₂ is greater than measured arterial PO₂ is explained in the next section.

Respiratory disease, and especially hypoventilation, is characterized by changes in blood gas homeostasis, namely, hypoxia and hypercapnia. Often, the only treatment that can be used is mechanical ventilation until the disease is resolved. While mechanical ventilation may be a lifesaver in certain conditions, this treatment can also injure the lungs with barotrauma. Epidemiologic data in the 1980s and 1990s showed that “permitting” an increase in PaCO₂ without imposing mechanical ventilation, or delaying mechanical ventilation, might have beneficial effects. Both animal and human studies have shown that therapeutic hypercapnia attenuates various measures of lung injury. For example, 8% CO₂ increases expression of genes encoding surfactant-associated proteins A and B, which contribute to host defense mechanisms in the lung.

In pediatrics, questions about this strategy have focused on bronchopulmonary dysplasia (BPD) in infants after prematurity. Higher PaCO₂ values are associated with lower BPD rates and large population-based studies in premature neonates show decreased BPD and death with permissive hypercapnia used before or during intubation and mechanical ventilation. Permissive hypercapnia also increased survival in neonates who have congenital diaphragmatic hernia. However, experiments in animal and human cell cultures also suggest some undesirable effects of elevated CO₂, such as suppressed immune function, increased oxidative lung injury, and impaired alveolar repair with potentiated tissue nitration. Summarizing, permissive hypercapnia is a ventilatory strategy that may reduce injury to the developing lung through a variety of mechanisms. Considering all of the evidence to date, permissive hypercapnia appears to be safe and beneficial in neonates subject to barotraumas.