
Case 1
A 49-year-old insurance executive was scheduled for an exercise tolerance test. The patient had no significant past medical history and was in excellent health. He jogged two miles regularly three times a week and played golf on weekends. Screening for hypercholesterolemia in the preceding two years yielded values consistently under 200 mg/ dl. He had been a heavy cigarette smoker for over 20 years, but stopped in the preceding month. Physical examination of the heart and lungs was unremarkable.
Central cyanosis (blueness of skin, lips, mucous membranes), as opposed to peripheral cyanosis, is always a manifestation of hypoxemia. Except for the relatively uncommon causes of methemoglobinemia, sulfhemoglobinemia, and some hemoglobinopathies, central cyanosis is always accompanied by a low arterial Po2. As a result of hypoxemia an excess amount of hemoglobin is not saturated with oxygen; in currently accepted terminology this unsaturated hemoglobin is said to be reduced. It is the quantity of reduced hemoglobin (RHB) per deciliter of capillary blood, not the relative lack of oxygenated hemoglobin, that accounts for the bluish color of cyanosis.
In their monograph Lundsgaard and Van Slyke noted that:
since it is impossible to know what condition prevails [in the capillaries], we have, in relating cyanotic color to the content of reduced hemoglobin in the capillary blood, assumed . . . the average unsaturation of capillary blood is midway between that of the arterial and venous bloods respectively . . . and:
The effect of. . . modifying factors is to cause the mean capillary concentration of reduced hemoglobin at which cyanosis becomes perceptible to vary from 4 to 6 grams of reduced hemoglobin per 100 c.c. of blood, and perhaps sometimes even more widely, although it appears usually to lie near 5. Since, to our knowledge, no other investigators independently arrived at 5 g/dl RHB as the value necessary for central cyanosis, it is safe to assume that this widely quoted number originated with the 1923 Lundsgaard and Van Slyke article.
Of the ten original articles published after 1923, all but two reference the Lundsgaard and Van Slyke paper by footnote. Two of the ten articles erroneously compare an arterial value of RHB to the capillary value of Lundsgaard and Van Slyke.
One review article also makes this erroneous comparison and as a result states that the value of 5 g/dl is “incorrect.” Two other review articles err by placing the value of 5 g/dl RHB in the arterial circulation. None of the three errant review articles references the 1923 Lundsgaard and Van Slyke article.

“EVer since the discovery that incomplete oxygenation of arterial blood is the cause of central cyanosis, the question has been asked: How much reduced (deoxygenated or unsaturated) hemoglobin (RHB) is needed to generate central cyanosis? The amount commonly quoted, 5 g of RHB/dl of blood, comes from the classic 1923 monograph by Lunds- gaard and Van Slyke (emphasis added):

Many of the patients in this series required vasopressor therapy during the course of septic shock. It could be argued that the hemodynamic and myocardial changes found are related in part to the use of vasopressors. There are no differences in the data and subsequent conclusions, however, in the mean initial hemodynamic findings, the right ventricular or left ventricular ejection fractions, or the right ventricular or left ventricular volumes if one analyzes only the subset of patients not receiving vasopressor therapy (n = 22). Therefore, the entire patient group studied is presented herein.
Iii this study, 39 patients with blood culture-positive septic shock and a characteristic hyperdynamic hemodynamic profile were found to have biventricular cardiac dysfunction. This was characterized by depression of both the right and left ventricular ejection fractions with simultaneous dilatation of both ventricles. In the survivors, these abnormalities were severe at shock onset, but returned toward normal (for both ventricles) as the patients recovered. The septic shock-induced serial changes for the right ventricle usually paralleled those of the left ventricle, demonstrating that the myocardial depression and ventricular dilatation seen in humans with septic shock is a biventricular phenomenon. It is not clear why a small number of patients had changes in right ventricular size or function in a direction that was opposite the changes in left ventricular size or function. All of these patients were critically ill, and it is possible that a number of different factors acted together to produce different responses of the ventricles in individual patients.
Radionuclide Determined Ejection Fractions and Ventricular Volumes
Table 2 shows the initial and final right and left ventricular ejection fractions and the right and left ventricular end-diastolic volume indices for the survivors and nonsurvivors of septic shock. Survivors had a substantial depression of initial left ventricular ejection fraction at 0.31 and a depressed initial right ventricular ejection fraction of 0.35. With recovery these values increased significantly to 0.47 (p<0.001) and 0.51 (p<0.001), respectively. The initial left and right ventricular end-diastolic volume indices were substantially increased (145 ml/m2 and 124 ml/m2, respectively) and decreased toward normal with recovery (to 106 ml/m2, p = 0.012; and to 88 ml/m2, p = 0.03, respectively). Thus, survivors showed a pattern of transient biventricular depression of ejection fraction associated with transient biventricular dilatation.