Effect of Hemodialysis on Oxygen-Hemoglobin Affinity in Chronic Uremics

Effect of Hemodialysis on OxygenHemoglobin Affinity in Chronic Uremics* James J. Szwed, M.D.;·· Friedrich C. Luft, M.D.; John R. Boykin, M.D.; Mark O. Farber, M.D.; and Stuart A. Kleit, M.D. P 60, plasma pH, red ceU pH, 2,3-DPG, ATP, lactate, pyruvate, and glycolytic rate were me.ured before and after hemodialysis in 25 patients on chronic hemodialysis. The prehemodialysi9--resting 2,J-DPG level and P 60 were elevated in accordance with the degree of anemia. The resting lactate and glycolytic rate were mildly elevated. Plasma and red ceU pH increased significantly with dia1ysis. The in vivo P 60 decre.ed significantly, whereM the

in vitro PM' 2,J-DPG, and ATP levels remained constant. The faD in the in vivo P 60 can be attributed entirely to the rise in red ceU pH or Bohr effect. The deCreMe in the in vivo P 60 during hemodialysis may transiently impair tissue oxygen delivery in these patients. Marked shifts in pH, especially into the alkaIemic range, should be minimized, particularly in patients undergoing frequent or prolonged hemodialysis.

The anemia of chronic renal disease is of paramount importance because it reduces the supply of oxygen to the tissues by lowering the oxygencarrying capacity of the blood. 1 One of the mechanisms utilized by anemic patients to compensate for this lowered oxygen carrying capacity is to reduce the affinity of hemoglobin for oxygen. It is now generally conceded that if MCHC, temperature, and pH are constant, this reduction is mediated by ATP and more importantly 2,3-DPG.2-5 Chronically ill patients on hemodialysis exist in an unique situation . because their survival is dependent upOn treatments which are operational only 12-18 hours per week in six hour sessions. This procedure subjects these patients to innumerable abrupt alterations in the internal environment, especially rapid shiftS in pH.6-8

commonly reported in chronic hemodialysis. 7•8 In addition, 2,3-DPG and ATP levels were investigated in order to assess their possible role in any acute, observed changes in hemoglobin-oxygen affinity.12.13 Finally, in order to delineate other possible alterations in red cell function, lactate, pyruvate, and glycolytic rate were measured.

The affinity of hemoglobin for oxygen in chron-

ic renal failure has been measured in both hemodialyzed and nonhemodialyzed patients.6 .9-11 Changes in the organophosphates 2,3-DPG and ATP, as well as alterations in pH, have been postulated as crucial factors governing oxygen delivery to tissues in these patients. Egger, Blumberg, and Marti6 studied the affinity of hemoglobin for oxygen before and after dialysis and found it to rise acutely, thus potentially decreasing the delivery of oxygen to the tissues. They attributed this rise to the sizable rise in pH engendered by the dialysis procedure. The purpose of our study was to better define the affinity of hemoglobin for oxygen before and after hemodialysis at the degree of pH change most °From the Deparonents of Medicine, Indiana University and Veterans Administration Hospital, Pulmonary Research Laboratories Indianapolis. OORecipient of grants-in-aid from Eli Lilly Company and Indiana Kidney Foundation. Presented in part at the 6th annual meeting, American Society of Nephrology, Washington, D.C., November, 1973. Supported in part by the Indiana Lung Association Grant 57-825-30. Manuscript received March 11; revision accepted April 23. Reprint requests: Dr. Szwed, Indiana University Medical School, Indianapolis 46202

278 SZWED ET At

METHODS AND MATERIALS

Twenty-five in-center hemodialysis patients were investigated. Their duration of dialysis varied from two months to three years. The etiology of renal failure in this group included chronic glomerulonephritis, chronic pyelonephritis, polycystic disease, and acute tubular necrosis without recovery. The patients ranged in age from 25 to 61 years. The dialysate acetate concentrations were 36.6 mEq/L in the case of the Tmvenol dialysate and 38.0 mEq/L with the Cobe solutions respectiv~ly. Twelve patients were using the Travenol RSP delivery system with ultra-flow 2 coil dialyzers for six hours twice a week. These patients were in training for home dialysis. Thirteen patients used the Cobe Centry delivery systems with two Gambro-Lundia dialyzers in series six hours twice a week. Although the latter group of patients was dialyzed on a greater surface area, there was no significant difference in the results from the two groups, and the entire patient population is treated as a whole. Predialysis blood samples were drawn from the arterial cannula or fistula needle prior to the beginning of dialysis. Post dialysis samples were drawn within ten minutes of completion of dialysis from the same site. Blood for P w was drawn anaerobically and measured immediately. The P w corrected to pH 7.4 (in vitro Pw) was calculated, after tonometry, using the Severinghaus nomagram. 14 The in vivo hemoglobin oxygen affinity (in vivo PliO) was determined by the method of Bellingham and his colleagues. 11i The 2,3-DPG and ATP samples were preserved in trichloroacetic acid. The 2,3-DPG determinations were done by the automated method of Atkinson and are expressed in p. moles/g/Hb,16 The ATP levels were determined with the Sigma analytic kit, 366-uv. The P02 and pH were measured on an Instrumentation Lab 313 blood gas analyzer. The red cell pH was measured by the method of Bromberg. I7 Red cell lactate and pyruvate levels were determined by the method of Prins and LooS.I8 The glycolytic rate was measured by the method described by Brewer et al. 19 The BUN, creatinine, hemoglobin, hematocrit, plasma

CHEST, 66: 3, SEPTEMBER, 1974

sodium, potassium, chloride, phosphorus and calcium were measured by standard clinical laboratory and auto analyzer methods. The mean corpuscular hemoglobin concentration was obtained from the ratio of the hemoglobin to the hematocrit. Student's T test was employed for statistical analysis. RESULTS

The six hour hemodialysis sessions resulted in the anticipated serum changes in the routine parameters used to determine the adequacy and efficiency of hemodialysis. These patients appeared clinically well dialyzed. Their degree of anemia was similar to the rest of our dialysis population. The interdialytic serum phosphorus levels were maintained near normal limits with aluminum hydroxide binding therapy. Table 1 lists the effect of hemodialysis on P50 and the factors known to affect the affinity of hemoglobin for oxygen. Baseline 2,3-DPG levels and P50 values were elevated in this patient population. The ATP and MCHC levels were normal. The PC02 and bicarbonate values were depressed, whereas the plasma pH was normal. The red cell pH indicated an initial red cell acidosis. After dialysis, the following statistically significant changes were found: fall in the in vivo P50, rise in plasma and red cell pH, and rise in bicarbonate. The in vitro P50, the PC02, and the organophosphates failed to change significantly after hemodialysis. The MCHC changed only slightly, but remained within normal limits, whereas the serum phosphorus concentrations fell as anticipated. Table 2 shows the effect of hemodialysis on red cell metabolism. The blood glucose determinations were not uniformly obtained in the fasting state. The whole blood lactate was elevated and fell slightly, but not significantly after dialysis. The already elevated glycolytic rate rose to a slightly higher level. The pyruvate level, on the other hand, increased but remained essentially within normal limits. This rise is reflected in the decreased lactate/ pyruvate ratio after dialysis. Table I-Parameter. Affecting Hb02 Affinity in 25 Hemodialy';. Patien" p

Normal

0 Hour

6 Hour

(0-6 Hrs)

Table 2-The Effect 01 Hemodialy.i. on Red Cell Metaboli.m in 25 Patient. Normal· 80-120 0.So1.2

1.3± .3

l.6± .3

>.11

Lactate m molee/L whole blood

O.SOU

1.36 ± .16

1.11±.14

>.2

Pyruvate m molee/L whole blood

O.OS-O.lO

.07±.006

L/p-ratio

10:1-16:1

19:1

SA~~

28.0±0.6

<.06 >.11

pH (pl811ma)

7.41±.03

7.38±0.03 7.46±0.11

<.001

P5G pH 7.4

·pHto

7.18±.04

7.10±0.06 7.16±0.06

<.001

28mmHg

22.6± 1.5

<.011

2.3-DPG " molee/gm Hg

14.0±2

19.8±2.8

2O.4±3.6

>.4

ATP mmoleefLRBC

1.00-1.50

1.16±0.06 1.22 ±.41

MCHC %

34.0±2.0 32.2±.O3

Serum phoephorus mg/lOO ml 2.4-4.7

31 ±2.0

CHEST, 66: 3, SEPTEMBER, 1974

>.06 <.06

3.6±.1

<.001

6.3±.3

11:1

<.01 <.01

80

3'I"C

'II =~ ._

($ ~ SUPPLIED)

25

---- ------·40 40

>.11

33.4±O.3

·pH ,e indicates intracellular pH of the red blood cell.

,, ,

'_~=~

"

"

29.4±O.7

32 ± 1.4

"',.~.:,'

80

29.2±O.7

40mm Hg

.10±.009

The amount of oxygen delivered to the tissues at any arterial P02 cardiac output, and hemoglobin level are dependent upon the affinity of hemoglobin for oxygen. This affinity is conventionally expressed as the P50, the partial pressure of oxygen at which hemoglobin is 50 percent saturated. P50 is a convenient term to describe the position of the oxyhemoglobin dissociation curve. Figure 1 illustrates the oxyhemoglobin dissociation curve and the important parameters which affect it. The examples in Figure 1 are displayed at an arterial P02 of 80 mm Hg and a mixed venous P02 of 40 mm Hg. A leftward shift of the curve, or a fall in the P50 theoretically decreases the oxygen supply to the tissues. A rightward shift, or increase in the P50, should increase the oxygen supply to the tissues. These unique properties of the oxyhemoglobin dissociation curve are the result of an increase or decrease in the affinity of hemoglobin for oxygen respectively. At a given temperature, changes in P50 caused by factors other than pH can be appreciated by correcting the measured, or in vivo P50, for the Bohr effect. Any remaining change in the corrected, or in vitro P50 would therefore be due to organophosphates or MCHC. In our patient population, the in vitro PliO was elevated before dialysis. This reflects an elevated 2,3DPG. Elevated 2,3-DPGs in dialysis populations have been reported on several previous occasions by

29.8±0.7

24 mEq/L 18.6 ± .8

110

DISCUSSION

26.6±.6

Pco.mm HI!

132

·For our laboratory.

26.II±.1I

HCO,mEq/L

(0-6 Hrs)

Plasma glucoee mg/100 ml

P .. (in

mm HI!

6 Hour

Glycolytic rate mg glucoee/gm Hb/hr

P,. (in vivo) mm Hg 1>"'0)

p

o Hour

20

o~""",-,-""""-+.....,...~~--l--.---.

o

20

40

80

80

100

~(mmllg)

FIGURE 1. The effects of various factors on the position of the oxyhemoglobin dissociation curve and on oxygen delivery.

HEMODIALYSIS EFFECT ON OXYGEN·HEMOGLOBIN AFFINITY IN UREMIA 279

Egger, Blumberg, Marti, and their colleagues.6 •9•1o Lichtman and associates20 have also noted elevated 2,3-DPG levels in dialysis patients, although to not as marked a level. The elevation in 2,3-DPG in our patients is compatible with the degree of anemia. 1 At the end of dialysis there was no significant change in the in vitro P:;o. The in vivo P:;o, on the other hand, dropped significantly from 29.8 to 28.0 mm Hg. This decline was associated with both an extra and intracellular pH change of approximately 10 nanomoles of hydrogen ion per liter. Thus, the pH-mediated increase in the affinity of hemoglobin for oxygen or Bohr effect, was responsible for the fall in the in vivo P:;0.21 Egger, Blumberg, and Marti6 recorded a fall in the in vivo P:;o from 29.5 to 26.3 mm Hg in six dialysis patients. Correspondingly, they noted a rise in plasma pH from 7.39 to 7.52, approximately twice as great a change as occurred in our patients. The reason for the more pronounced pH change in the patients of Egger et al is not entirely clear. It may be partially explained by the fact that their dialysis sessions lasted seven to eight hours rather than six hours as employed in our patients. The PC02 in our patient population was 8 mm Hg below normal and was not affected by dialysis. Rosenbaum and colleagues8 also noted a low PC02 in hemodialysis patients which was unchanged by the procedure. The mechanism underlying this continued hyperventilation may be partially explained by the patients' severe anemia. Sproule et al22 studied the cardiopulmonary physiologic responses in a variety of anemic patients, and found that in their subjects with a mean Hb of 6.5, the CO2 tension was 35 mm Hg in the face of a normal pH. In addition, the dialysis patients have accumulation of fixed acids in the interdialytic period with corresponding, mild, compensated, metabolic acidosis. The bicarbonate rose after dialysis as would be expected with the removal of hydrogen ion and the addition of acetate. Neither ATP nor 2,3-DPG levels showed a significant change after hemodialysis, findings which are in agreement with those of Egger et al. 6 The decrease in hydrogen ion concentration is probably the most important stimulus for 2,3-DPG synthesis; however, the time span required for the increased synthesis to occur in significant amounts is greater than the dialysis period. Bellingham, Detter, and Lenfant1:; induced metabolic alkalosis in normal volunteers with sodium bicarbonate. The onset of alkalosis caused no change in 2,3-DPG for about eight hours. The levels gradually rose and achieved a maximum at 48 hours. The decrease in plasma phosphate induced by dialysis did not alter the organophosphate levels, a fact also noted by Lichtman and his colleagues. 2o 280 SZWED ET AL

The mean predialysis red cell pH of 7.10 in our patients was below the accepted value of 7.18 + .04 for normal individuals. This finding is at variance with the results of Egger et al6 who reported a predialysis red cell pH of 7.16. The reason for the initial red cell acidosis in our patient group is not clear. Concomitant plasma pH was normal. A mild degree of initial lactic acidemia may have contributed to the low red cell pH. The lactate level tended to fall with dialysis, at least in part attributable to this procedure. 23 The lactate/pyruvate ratio decreased. Calculations based on the lactic dehydrogenase reaction indicate that a rise in pH may decrease the lactate/pyruvate ratio. 24 The elevated resting glycolytic rate is appropriate in the face of the severe anemia seen in these patients. A further rise in the glycolytic rate is expected with the fall in hydrogen ion concentration produced by dialysis. 2:; It has been shown that if the hemoglobin level, cardiac output, arterial and venous pH, and P02 are constant, a shift of 4 mm Hg in the P:;o results in a 7 percent change in oxygen delivery.26 In Our patients, if the above mentioned variables remained constant, a 2 mm Hg decrease in P:;o would theoretically result in a 3.5 percent decrease in tissue oxygen delivery ie, as much as 35 ml O2 per minute. Cardiac output and arteriovenous oxygen difference in the face of PM measurements have not yet been determined in the dialysis population. Until these additional variables are known, decreases in tissue oxygen delivery remain imperfectly defined. Pending the results of definitive studies on these critical parameters of oxygen transport, it appears reasonable to avoid alkalemia during dialysis and thus minimize the fall in PM. REFERENCES

1 Torrance J, Jacobs P, Restrepo A, et al: Intraerythrocytic adaptation to anemia. N Engl J Moo 283:156-169,1970 2 Brewer GJ, Eaton JW: Erythrocyte metabolism: interaction with oxygen transport. Science 171:1205-1211, 1971 3 Cameron BF, Lian CY, CarvajaIino OJ, et al: Regulation of oxygen dissociation by 2,3-diphosphoglycerate in the human erythrocyte. Pan-Am Assn of Biochem Societies, 1, New York, Academic Press, 1972, pp 169-196 4 Klocke RA: Oxygen transport and 2,3-diphosphoglycerate. Chest 62:79S-85S, 1972 (supplement) 5 Shappell SD, Lenfant CJM: Adaptive, genetic, and iatrogenic alterations of the oxyhemoglobin-dissociation curve. Anesthesiology 37:127-139,1972 6 Egger V, Blumberg A, Marti HR: Acid-base balance and oxygen affinity of hemoglobin in patients on maintenance dialysis. Clin Nephrol 1:70-75, 1973 7 Goss JE, Alfrey AC, Vogel JHK, et al: Hemodynamic changes during hemodialysis. Trans Amer Soc lot Organs 13:68-74, 1967 8 Rosenbaum BJ, Coburn JW, Shinaberger JH, et al: Acidbase status during the interdialytic period in patients maintained with ch~onic hemodialysis. Ann intern Med 71:1105-1111,1969

CHEST, 66: 3, SEPTEMBER, 1974

9 Blumberg A, Marti HR: Adaptation to anemia by decreased oxygen affinity of hemoglobin in patients on dialysis. Kidney International 1:263-270, 1972 10 Blumberg A, Scherrer M, Marti HR, et al: Red cell organic phosphate levels and in vivo oxygen-hemoglobin dissociation in patients on maintenance dialysis with anemia. Blut 22:109-115,1970 11 Mitchell TR, Pegrum GD: The oxygen affinity of hemoglobin in chronic renal failure. Br J Haem 21:463-472, 1971 12 Lichtman MA, Miller DR: Erythrocyte glycolysis, 2,3diphosphoglycerate and adenosine triphosphate concentration in uremic subjects: Relationship to extracellular phosphate concentration. J Lab Clin Med 76:267-279, 1970 13 Travis SF, Sugennan HJ, Ruberg RL, et al: Alterations of red cell glycolytic intennediates and oxygen transport as a consequence of hypophosphatemia in patients receiving intravenous hyperalimentation. N Engl J Med 285:763768, 1971 14 Edwards MJ, Martin RJ: Mixing techniques for oxygen hemoglobin equilibrium and Bohr Effect. J App Physiol 21: 1898-1902, 1966 15 Bellingham AJ, Detter JC, Lenfant C: Regulatory mechanisms of hemoglobin oxygen affinity in acidosis and alkalosis. J Clin Invest 50:700-706, 1971 16 Atkinson KF: Modified automated detennination of 2,3diphosphoglycerate in whole blood. Clin Chern 18:10011004, 1972 17 Bromberg PA, Theodore J, Robin ED, et al: Anion and hydrogen ion distribution in human blood. J Lab Clin

Med 66:464-475,1965 18 Prins HK, Loos JA: Detennination of energy-rich phosphate 2,3-diphosphoglycerate, lactate, and glutathione in small amounts of blood cells. Advances in Automated Analysis, Technicon International Congress, 1969, White Plains, New York, mediad, 1970, p285 19 Brewer GJ, Eaton JW, Weil JV, et al: Studies of red cell glycolysis and interactions with carbon monoxide, smoking and altitude. Red Cell Metabolism and Function, 1; (Brewer GJ, ed), New York, Plenum Press, 1970, p 95 20 Lichtman MA, Murphy MS, Byer BJ, et al: Hemoglobinoxygen affinity in chronic renal disease: Effect of hemodialysis. Clin Res 21:696,1973 (abstract) 21 Wranne B, Woodson RD, Detter JC: Bohr effect: Interaction between H + C02, and 2,3-DPG in fresh stored blood. J Appl PhysioI32:749-754, 1972 22 Sproule BJ, Mitchell JH, Miller WF: Cardiopulmonary physiological responses to heavy exercise in patients with anemia. J Clin Invest 39:378-388, 1960 23 Selroos 0, Pasternack A, Kuhlbaeck B: Laktatacidos och fenfonnin. Nordisk Medicin 80: 1658-1661, 1968 24 Cohen PJ: The metabolic function of oxygen and biochemical lesions of hypoxia. Anesthesiology 37: 148-177, 1972 25 Williamson JR: General features of metabolic control as applied to the erythrocyte. Red Cell Metabolism and Function, 6; (Brewer GJ, ed), New York, Plenum Press, 1970, p 117 26 Edwards MJ, Novy MJ, Walters CL, et al: Improved oxygen release: an adaption of mature red cells to hypoxia. J Clin Invest 47: 1851-1857, 1968

Maya Art Of all ancient cultures of the Pre-Columbian America the Maya is considered the most brilliant for a number of reasons. It spanned a vast period of time-the early Maya developmental phase began about 1000 B C and the whole culture only declined finally as late as the seventeenth century A D. It extended over a huge geographical area, including the States of Yucatan, Quintana Roo, Campeche, Tabasco, Chiapas, and beyond the limits of Mexico itself, to Guatemala, Honduras and EI Salvador. Its achievements include the development of hieroglyphic system of writing and a mathematically computed calendar of great accuracy. Their architecture has

CHEST, 66: 3, SEPTEMBER, 1974

a basic form probably derived from the primitive rectangular hut with low walls and high sloping roof. A characteristic element is the corbelled arch elliptical in shape and constructed by placing each successive block of stone so that it projected beyond the one below, and was held in position by the weight of the superstructure. In general, the fresco is the most unusual form of Maya painting, and was used either to decorate buildings, for ceramic ornamentation or to illustrate manuscripts. Abbate F (ed): Pre-Columbian Art of North America and Mexico (translated by Evans, E) . London, Octopus, 1972

HEMODIALYSIS EFFECT ON OXYGEN·HEMOGLOBIN AFFINITY IN UREMIA 281