Bette Sea monds, Javad Towfighi, and Dan A. Aryan1 ...... Quart. J. Med. 1, 247 (1932). 43. Saltzman,. H. A., Heyman,. A., and Sieker, H. 0., Correlation of clinical.
Determinationof IonizedCalciumin Serumby Useof an Ion-SelectiveElectrode I. Determination of Normal Values under Physiologic
Conditions, with
Comments on the Effects of Food Ingestion and Hyperventilation Bette Sea monds, Javad Towfighi, and Dan A. Aryan1
Ionized calcium was determined in serum from 84 normal individuals by using a flow-through calcium-selective electrode operated at 37#{176}C. The mean value was 1.08 ± 0.06 (2 SD) mmol/liter. The relationship of serum ionized calcium concentration to other serum constituents was examined. The effects of hyperventilation and food ingestion on ionized calcium were studied. Respiratory alkalosis, induced by hyperventilation, caused a decrease in ionized calcium of 0.05 ± 0.02 (2 SD) mmol/liter per 0.1 unit increase in pH. In contrast, the metabolic alkalosis induced by food ingestion caused a decrease in ionized calcium of 0.105 ± 0.025 (2 SD) mmol/liter per 0.1 unit increase in pH.
Additional Keyphrases alkalosis
#{149}
total calcium
genous and endogenous
correlations
factors
between states
respiratory
and
metabolic
AutoAnalyzer #{149} exoaffecting serum calcium
#{149}
of serum
calcium
#{149}
pH,
data regarding the effects of pH and P0, on the binding of calcium by protein and, therefore, on the concentration of ionized calcium (19-21), we have compared the changes in serum ionized calcium induced by respiratory alkalosis and by metabolic alkalosis.
Materials and Methods Blood Samples Venous blood from 84 fasting (overnight) individuals was obtained by completely filling “Vacutamer” tubes. The group comprised 49 men, 22 to 63 years old, and 25 women, 19 to 42 years old. All samples were allowed to clot at room temperature, centrifuged after 30 mm, and then equilibrated anaerobically to 38#{176}C (± 1#{176}C), in a temperature-controlled hood.
phosphate
Determination Since McLean and Hastings (1) demonstrated that ionic calcium is the physiologically active form of calcium in the body, numerous methods have been developed in an attempt to assay this fraction (2-7). The practicality of these methods for routine use, as well as their precision and relative accuracy, left much to be desired until the recent development of the improved calcium-specific electrode (8). Several reports have appeared in the literature describing ionized calcium concentrations in normal subjects (9-16), as well as in various pathologic conditions such as in primary disturbances of calcium metabolism and in various dysproteinemias (9, 13, 17, 18). The following study describes our experience with the calcium electrode in the determination of normal values for ionized calcium, at 37#{176}C, in an apparently normal population. Because a review of the literature yielded conflicting
From The William Pepper Laboratory,Department of Pathology, University of Pennsylvania, Philadelphia, ‘Address reprint requests to D. A. A.
Received
Pa.
June 25, 1971;acceptedOct.27, 1971.
19104.
of Ionized
Calcium
Calcium ion activity was determined within 1 to 2 h after blood collection. The method we followed closely resembled that described by several others (10, 13, 15, 22). Our system consisted of an Orion calcium electrode (Model 88-20) used in conjunction with a saturated KC1/calomel reference electrode and a digital pH meter (Model 801) for potential readings (Orion Research, Inc., Cambridge, Mass. 02139). The procedure for storage and maintenance of the electrode was that outlined in the Orion manual (22). Calcium standards containing 0.5, 1, and 2 mmol of calcium chloride and 150 mmol of sodium chloride per liter were used for calibration. The standard solutions were prepared to contain 0.02 g of trypsin and 20 Ml of triethanolamine (1 mol/ liter), pH 8.0, per 33.3 ml. Serum samples were transferred anaerobically to 1-ml tuberculin syringes equipped with 18-mm 20-gauge needles. Samples and standards were pumped at the rate of 50 zl/min and solutions of serum or the 1 mmol calcium standard were alternately passed through the system until millivolt readings became stable. To ensure maximal stability, the 1 mmol/liter standard was checked between each serum unCLINICAL CHEMISTRY, Vol.18,No.2,1972 155
known. In most instances, duplicate tubes of blood were examined or replicates from the same tube were assayed. The allowed variation of response for the 1 mmol/liter standard was 0.1 mY, ensuring an equally good response in the unknowns.
In four subjects, we followed the effects of food ingestion on ionized and total calcium, phosphate, total protein, and pH for 6.5 to 7 h. Samples were taken before and 2 h after breakfast, before and 2 h after lunch.
Determination of Total and Diffusible pH, Phosphate, and Serum Proteins
Results
Calcium,
The pH of all sera was determined at 37#{176}C with a Radiometer pH meter (Copenhagen, Denmark). An ultrafiltrate containing diffusible calcium was prepared (at 25#{176}C) by the method of Farese et al. (23), by use of Centriflo membranes (Amicon Corp., Lexington, Mass, 02173). This as well as total calcium were measured by AutoAnalyzer (Technicon Instrument Corp., Tarrytown, N. Y. 10591) (24). Phosphate and total protein were also determined by AutoAnalyzer (25, 26). All sera were electrophoresed on cellulose acetate and the immunoglobulins quantitated (27) to ensure normalcy.
Comparison of Respiratory and Metabolic Alkalosis Produced by Hyperventilation and Food Ingestion The effects of hyperventilation were examined in three subjects after the following protocol: (a) The subjects reclined throughout the experiment. (b) They respired deeply at a rate of 30/mm for 15 mm to 25 mm. (c) The blood pressure and pulse were monitored at regular intervals. (d) All symptoms and the time of their occurrence were noted. (e) Total, diffusible, and ionized calcium, phosphate, and pH were measured in the resting state and at 5-mm intervals during hyperventilation. (f) After the period of hyperventilation the subject continued to rest for at least 35 mm, and at least two further blood samples were taken.
Table 1 lists the mean values for total, diffusible, and ionized calcium, pH, and phosphate for the entire group, as well as those for men and women. No sex-related differences were noted in the values for pH or total and ionized calcium, but statistically significant differences were found for diffusible calcium and for phosphate. The mean percentage of ionized calcium was 43.2% of the total, while the mean percentage of diffusible calcium was 55.3% (55.9% for men and 54.6% for women). Thus 11.4 to 12.7% (mean 12.1%) of the total calcium was accounted for by calcium complexes of low molecular weight. Linear correlations between selected pairs of variables were studied. The results of the correlation are summarized in Table 2. There appears to be a correlation in only one instance, that of total vs. protein-bound calcium (r = 0.74). In spite of this apparent relationship, however, a relatively large scatter was found when a linear plot of the values were constructed (Figure 1). The effects of hyperventilation on one of the subjects are summarized in Table 3 and Figure 2. An increase in total calcium and pH and a decrease in ionized calcium and phosphate are apparent. In all three subjects the magnitude of the change for the same variable was similar. When the data from all three subjects are combined, the mean decrease in ionized calcium per 0.1 unit increase in pH, calculated from a total of 10 differences, is 0.05 ± 0.02 mmol/liter (2 SD). Table 4 and Figure 3 summarize changes induced in one subject in total and ionized calcium. pH, and phosphate as a result of food ingestion, As occurred in the case of respiratory alkalosisthe total calcium and pH increased while the ion,
Table 1. Comparison in Men and Women of Serum Calcium and Phosphate Concentration and pH Men and women Mean ± 2 SD SE mean
Mean
Men (49) ± SD SE mean
Women Mean
± 2 SD
(35) SE mean
Totalcalciumc Diffusiblecalciumc Ionized calciumc
2.50 ±0.16 1.38 ±0.13 1.08 ± 0.06
0.009 0.008 0.003
2.50 ±0.19 1.40 ±0.12 1.08 ± 0.05
0.012 0.009 0.004
2.49 ±0.14 1.36 ±0.12 1.08 ± 0.06
0.012 0.011 0.005
Phosphatec pH
1.10 ±0.34 7.371±0.034
0.019 0.002
1.07 ±0.36 7.369±0.035
0.027 0.002
1.17 ±0.36 7.372±0.040
0.031 0.003
Student’s test. Snedecor F test. mmol/Iiter. d005 > P >0.01.
156
CLINICAL CHEMISTRY, Vol.18,No. 2,1972
Men vs. women F6
-0.44
-1.16 2.34 0.83
1.42 1.00 1.44 1.00 1.38
Table 2. Correlation Coefficients for Selected Pairs of Variables ProteinTotal
Ionized
bound
Plio.
calcium calcium calcium
Total calcium Ionized calcium Protein-bound calcium Low-molecular-weight calcium complexes
phate 6
0.11
..-
0.74
6
6
a 6
S
C
0.06
Phosphate
pH Albumin
Total globulin 7-Globulin
0.30
6
0.32
S
6
0.35 0.15
0.09 0.04
0.35 0.09
S
0.10
S
0.06 0.06 #{149} Reversecoefficient determined.
6 Coefficient not considered pertinent, therefore lated. Linear plot failed to reveal any correlation.
ized calcium decreased. However, under the conditions of this study, the serum phosphate concentrations showed no significant change and the magnitude of the pH increase was less then in the case of respiratory alkalosis. In this instance, however, the decrease in ionized calcium resulting from metabolic alkalosis was significantly greater. The mean decrease in ionized calcium per 0.1 unit increase in pH calculated from 12 differences in four subjects was 0.105 ± 0.025 mmol/liter (2 SD). Discussion
S
not calcu-
Direct determination of ionized calcium in biological fluids has long been the goal of many investigators (28-31). Since the recent development by Ross (8) of the calcium ion-selective electrode, the measurement of ionized calcium has been
C
E
a
U
E
V
E
2
C
U
0
770
C) 0
C
C
0
7.60
E C
C
,J
7.50
0
V
a-
7.40
0
730
1.5
ao
2.5
Fig. 2. Effect of respiratory alkalosis on serum ionized and pH in one subject
Total Calcium (mmWl)
calcium
Fig. 1. Correlation between serum protein-bound calcium and total calcium in 84 normal individuals
Table 3. The Effect of Hyperventilation on Serum Ionized calcium Time,
mm
(a) Relation of ionized calcium and pH. (b) Simultaneous changes in ionized calcium and pH with time
pH, Calcium, Phosphate, and Total Protein Total calcium
Diffusible calcium
Phosphate
mmol/liter
pH
Total protein, g/dl
Hyperventilation
0 5 10 15 25
7.38
40 60
7.4 7.40
7.48
7.59 7.59 7.61
1.08 1.03 0.95
2.38 2.40
0.96 0.95
2.50 2.50 2.55
1.04 1.09
2.38 2.40
1.33 1.25 1.28 1.25
0.89 0.67 0.54
. . -
0.42
0.96
7.2 . . ... . -.
7.5
After recovery
1.28 1.30
0.32 0.42
CLINICAL CHEMISTRY, Vol.18,No. 2,1972 157
Table 4. The Effect of Food Ingestion on Serum pH, Calcium, Phosphate, and Total Protein ionized calcium Time,
Fast 2 h p.p.”
h
pH
0 2 4
7.35 7.40
1.10
(0)
I .05
a
U
1.00 C 0
0.95 735
7.40
pH
7:45
7:50
7.50 V
3 Time
4 (hours)
Fig. 3. Effect of metabolic alkalosis on serum ionized calcium and pH in one subject
(a) Relation of ionized calcium and pH. (b) Simultaneous changes in ionized calcium and pH with time facilitated, and several reports describing normal values and changes owing to pathological conditions have appeared (9-18). Our normal values for ionized calcium have a considerably narrower range than do those determined by several others (9-16). The standard deviation of the differences in the duplicates is 0.007 mmol/liter. This degree of control was achieved by processing the 1.00 mmol/liter standard between each serum unknown and allowing only ±0.1 mY drift rather than ±0.2 mV. It has been reported by others that trypsin and triethanolamine stabilize, though alter, readings of potential (10, 14, 15). We found that the amount of triethanolamine added to the standards influenced the results considerably. One set of experiments in which we followed the protocol described by Schwartz et al. (14) and Lindgarde and Zettervall (16) [i.e., using three times the amount of tnethanolamine] gave results that were 0.075 to 0.100 mmol/liter greater than those obtained on 158 CLINICAL CHEMISTRY, Vol. 18, No. 2, 1972
Total protein, g/dl
1.10 1.05 1.04
2.43 2.45 2.45
1.31 1.28 1.28
7.1 7.4 7.6
0.99
2.50
1.28
7.2
a true fasting sample because the postprandial
C
7
Phosphate
mmol/liter
7.41 “Fast” 6.5 7.44 2 h p.p.” #{149} 2 h postprandial. The midday fasting specimen cannot be considered to last several hours (38).
E
Total
calcium
alkaline tide has been shown
the identical sera by the aforementioned procedure. This at least partially explains some of the discrepancies in the literature values. On the other hand, the total calcium compares well with the values reported elsewhere, as do the values for diffusible calcium and for phosphate (9, 32). The greater phosphate values observed for women than for men are consistent with the observation of other investigators (32). The reason for the apparently increased values of diffusible calcium in the males is not known. The values obtained for serum pH are in reasonable agreement with those described by Gambino (33). The correlation coefficients determined for the pairs of parameters described in Table 2 agree in part with those found by others (9, 34, 35). No significant correlations were found between ionized and total calcium or ionized, total, or proteinbound calcium and proteins. However, an apparently significant relationship between total and protein-bound calcium was found. Whereas a correlation between protein-bound (or total) calcium and albumin has been found by others to be significant (9, 35, 36), we did not find such a relationship. We did not expect a significant correlation between protein-bound (or total) calcium and the concentration of serum proteins. Held and Freeman (37) examined the binding of calcium by human plasma proteins under simulated physiologic conditions and found that the concentration of calcium ions bound per gram of protein was not significantly different for albumin, a- (a, + a,), and fl-globulins, whereas 7-globulins bound considerably less. Because the actual concentrations of the a- and fl-globulins under normal conditions could comprise from 10% to 30% of the total protein, the suggestion of a simple relationship between protein-bound (or total) calcium and serum proteins, most notably albumin, is an oversimplification. Whether or not correlations between calcium fractions and proteins in pathologic states (e.g., dysproteinemias) exist is currently under study in our laboratory. The lack of correlation between ionized calcium and pH and phosphate suggests that other factors contributing to these relationships may be involved.
The apparent differences in the effects of metabolic and respiratory alkalosis on calcium fractions and phosphate are interesting. The relationship between pH and ionized calcium in the presence of a respiratory alkalosis agrees favorably with the results described by Moore (9) and others (14, 21), namely, that a rise in pH of 0.1 unit is accompanied by a decrease in ionized calcium of 0.05 (± 0.002) mmol/liter. On the other hand, we find the effect of food ingestion is more profound. Our data indicate that the induced alkalosis in this instance, although not of simple metabolic origin, causes the concentration of ionized calcium to decrease below the lower limits of normal whenever the pH exceeds the upper limit of normal range (i.e., >7.41). The fact that Moore (9) chose his samples near the peak of alkalosis (38), i.e., approximately 2 h postprandially, may explain why his range includes values that we believe to be below the fasting normal range. Our results differ somewhat from those of Pedersen (21) and of Lindgarde and Zettervall (16). Pedersen found no difference between the effects of simulated respiratory and metabolic alkalosis on concentration of ionized calcium. Since his studies were dQne in vitro, however, additional factors are probably involved in bringing about the effects on ionized calcium of metabolic alkalosis produced by food ingestion. These influences obviously could not be exerted in an in vitro system. Lindgarde and Zettervall (16) studied the diurnal variation of ionized calcium and found no change in values; however, it appears that none of their samples was drawn at the 2-h postprandial peak. Because we have noted inverse changes in ionized calcium corresponding to published data concerning the variation in acidbase balance as a result of food ingestion (38), we stress the importance of controlling the time of sampling of serum in addition to conforming to analytical stringency. We believe that the changes in ionized calcium associated with food ingestion are due to exogenous factors and not to endogenous diurnal variations. Wills (39) has studied the diurnal variation in total calcium concentration in fasting as well as in nonfasting subjects and has noted the lack of change in values in the former. Similarly the acid-base status in fasting individuals shows no diurnal change (38). The apparent increase in serum total calcium after food ingestion agrees in part with the trend shown by Wills (39). There appears to be no simple explanation for differences in response to the two types of alkalosis. It has been reported that hyperventilation causes mobilization of serum phosphate into the erythrocytes, leading to an increase in 2,3-diphosphoglycerate, hexose diphosphate, and adenosine diphosphate (40). Thus, we speculate that the lesser effect of the respiratory alkalosis on the ionized calcium could in some way represent a homeostatic
mechanism to prevent ionized calcium from decreasing to concentrations one might expect to induce tetany. Increased serum total calcium after hyperventilation has been described by several workers (40-42); others have found little or no change in its concentration (43). Considering the apparent reciprocal relation between phosphate and calcium ions, we believe that the increases observed may result in part as a response to the lowered phosphate concentrations. The inverse of this situation has been described by Reiss et al. (44) in response to phosphate feeding. In addition, the decreased ionized calcium concentration would likely increase secretion of parathormone, which could produce a rise in concentration of serum total calcium.
We
thank
Miss K. Bayer for technicalassistance
with
the
immunological quantitations, the technical staff of the William Pepper Laboratory for their assistance with the described determinations, and Dr. P. J. Pegg and Mr. I. Schreibman for their comments and criticism of this manuscript. The continued encouragement of Dr. Rawnsley is gratefully acknowledged. This study was supported in part by grant No. GM-O 1462, NIH, USPHS.
References 1. McLean, F. C., and Hastings, A. B., A biological method for the estimation of calcium ion concentration. J. Bi.ol. Chem. 107, 337 (1934). E, R., Connov, T. B., and Howard, J. E., In vitro calcification of rachitic rat cartilage in normal and pathological human sera,with some observations on the pathogenesis. Bull. Johns Hopkins Hosp. 96, 1 (1955). 2. Yendt,
3. Dale,N. E.,and Kellerman,G. M., The bindingof calciumby the plasma proteinsin hyperparathyroidism. Clin. Sd. 32, 433 (1967). 4. Walser,
M., Ion
association
VI. Interactions
between
calcium,
magnesium, inorganic phosphate, citrate and proteinin normal human plasma. J. Clin. Invest. 40, 723 (1961). 5. Ettori, J., and Scoggan, J. M., Ionized calcium in biological media. Nature, 184, 1315 (1959). 6. Tendeloo, H. J. C., A new and easy method for the potentiometric determinations of calcium concentrations in solutions. J. Biol. Chem. 113, 333 (1936). 7. Greenberg, D. M., and Larson, C. E., Evidence of adsorption experiments on the forms of calcium and inorganic phosphorus in blood serum. J. Biol. Chem. 109, 105 (1935). 8. Ross,
J. W.,
Calcium
selective
electrode
with
liquid
ion
ex-
changer.Science 156, 1378 (1967). 9. Moore,
E. W., Studies
with
ion-exchange
calcium
electrodes
in
biological fluids: Some applications in biomedical research and clinical medicine. In Ion-Selective Electrodes, R. A. Durst, Ed. Nat. Bur. Stand. Spec. PubI. 314, U. S. Government Printing Office, Washington, D. C., 1969,p 215. 10. Hattner,R. S.,Johnson, R. W., Bernstein, D. S., Wachman, A., and Brachman, J., Electrochemical determination of apparent ionized serum calcium using a calcium-selective electrode: The method and values in normal humans and a comparison of total serum calcium. Cliii. Chim. Acta 28, 67 (1970). 11. Sachs, C., Bourdeau, A. M., and Balsan, S., Determination du calcium ionis#{233} dans le serum avec une’electrode sp#{233}cifiqueh membrane liquide. Ann. Biol. Clin. (Paris) 27, 487 (1969). 12. Raman, A., Determination of ionizedcalcium in serum with a calcium electrode. Biochem. Med. 3, 369 (1970). 13. Hansen,
S. 0.,
and Theodorsen,
L., The
CLINICAL CHEMISTRY,
usefulness
of an im-
Vol. 18, No. 2, 1972 159
proved calcium electrode in the measurement in serum. C/in. Chim. Ac/a 31, 119 (1971). 14. Schwartz, H. 1)., McConville, E. F., Serum ionized calcium by Chim. Ac/a 31,97 (1971).
B. C., specific
of ionized
calcium
teins,a.s determined by a new type calcium amalgam 126, 389 (1938).
electrode.
J. Biol. C/tern. and ion
Christ.opherson, electrode. C/in.
30. Sollner, K., A new potentiometric method to determine cations and anions with collodion and protamine-collodion “membrane electrode.” J. Amer. Che,n. Soc. 65, 2260 (1943).
15. Li, T. K., and Piechocki, J. T., Determination of serum ionic calcium with an ion-selective electrode: Evaluation of methodology and normal values. CLIN. CuEM. 17, 411 (1971).
31. Gregor, H. P., and Schonborn, H. J., Multilayer membrane electrodes III. Activity of alkaline earth salts in mixed electrolytes. J. Amer. Chem. Soc. 83, 3576 (1961).
16. Lindgarde, F. and Zettervall,0., Serum ionized calcium in a normal population studied with a calcium ion-sensitive electrode. Isr. J. Med. Sd. 7, 510 (1971).
32. Keating, F. it, Jones, J. 1).,Elveback, L. It, and Randall, R.. V., The relation of age and sex to distribution of values in healthy adult serum calcium, inorganic phosphorus, magnesium, alkaline phosphatase, total proteins, albumin, and blood urea. J. Lab. C/in. Med. 73, 825 (1969).
17. Walser, M., The separate effects of hyperparathyroidism, hypercalcemia of malignancy, renal failure and acidosis on the state of calcium, phosphate and other ions in plasma. J. Cliii. Invest. 41, 1454 (1962).
33. Gambino,
S. R., p11 and
Pco,.
Stand.
Methods
C/in.
Chem.
18. Moore, E. W., Studie.s with ion exchange calcium electrodes III. The state of serum calcium in patients with cirrhosis.Gastroenterology 60, 43 (1971).
5, 169 (1965). 34. Smarz, C., and Adamski, A., The relationship between plasma proteins and calcium in various diseases. Pol. Med. J. 7, 358 (1967).
19. Prasad, A. S., and Funk, E. B., Effect of carbon dioxide on concentration of calcium in an ultrafiltrate of serum obtained by centrifugation. J. App!. Physiol. 10, 103 (1957).
35. Prasad, A. S.,and Flink, E. B., The base binding property of the serum proteins with respect to calcium. J. Lab. Gun. Med. 51, 345 (1958).
20. l)anesk, I. N. Z.,Walker, C. H. M., and Mathers, N. P., The relation of post-dialysispla.sma calcium and magnesium to the dialysate levels and to changes in blood pH. New Engi. J. Med. 282, 771 (1970).
36. Gutman, A. B., and Gutman, E. B., Relation of serum calcium to serum albumin and globulins. J. C/in. Invest. 16, 903 (1937).
21. Pedersen, K. 0., The effectof bicarbonate P0, and pH on serum calcium fractions. Scand. J. Gun. Lab. Invest. 27, 145 (1971).
37. Held, I. R., and Freeman, S., Binding of calcium by human plasma proteins under simulated physiologic conditions. J. App!. Physiol. 19, 292 (1964).
22. Instruction Manual, Model 88-20 Serum Calcium System, Orion Research Inc.,Cambridge, Mass.
38. Rune, S. J., and Lessen, N. A., Diurnal variations in the acidbase balance of blood. Scand. J. Cliii. Lab. Invest. 22, 151 (1968).
Flow-Thru
23. Farese, G., Mager, M., and Blatt, W. F., A membrane ultrafiltrationprocedure for determining diffusible calcium in serum. CLIN. CHEM. 16, 226 (1970).
39.
24. Kessler, G., and Wolfman, the simultaneous determination CLIN. CHEM. 10, 686 (1964).
40. George, W. K., George, W. 1)., Smith, J. P., Gordon, W. T., Baird, E. E., and Mills, G. C., Changes in serum calcium, serum phosphate and red cellphosphate during hyperventilation. New Engi. J. Med. 270, 726 (1964).
M., An automated procedure for of calcium and phosphorus.
25, Yee, H. Y., and Blackwell, L., A simplified mated phosphorus analysis of serum and urine. 898 (1968). 26. Technicon town, N. Y.
AutoAnalyzer
27. Fahey, J. L., and McKelvey, tion of serum immunoglobulins Immunol. 94, 84 (1965).
Method
N-14b,
method
for auto14,
CLN.
CHEM.
Techriicon,
Tarry-
determinaplates. J.
28. Fosbinder, R. 1., A study of the calcium amalgam electrode in dilute aqueous solutions. J. Amer. Client. Soc. 51, 1345 (1929). 29. Joseph, N. R., Heterogeneous equilibrium of protein solution. II. The interaction of calcium chloride and other salts with pro-
160
CLINICAL CHEMISTRY, Vol. 18, No. 2, 1972
M. R., The
concentration
effect of diurnal variation on total in normal subjects. J. Gun. Patho!.
plasma 23, 772
(1970).
41. Grant, Experimental (1920). 42.
E. M., Quantitative in antibody-agar
Wills,
calcium
S. B., and Goldman, A., Study of forced respiration: production of tetany. Amer. J. Physiol. 52, 209
McCance, It. A., Spontaneous 1, 247 (1932).
overbreathing
tetany.
Quart.
J. Med.
43. Saltzman, H. A., Heyman, A., and Sieker, H. 0., Correlation of clinical and physiologic manifestations of sustained hyperventilation. New Engl. J. Med. 268, 1431 (1963). 44. Reiss, E., Canterbury, J. M., Bercovitz, M. A., and Kaplan, E. L., The role of phosphate in the secretion of parathyroid hormone in man. J. Cliii. Invest. 49, 2146 (1970).
