Symposium on Response to Infection and Injury II
Total Body Potassium' and Nutritional Status Harry M. Shizgal, M.D. *
The body cell mass was defined by Moore as that component of body composition "containing -the oxygen-exchanging, potassium-rich, glucose-oxidizing, work-performing tissue."5The body cell mass is in effect the living component of body composition and therefore its measurement permits a quantitative assessment of the nutritional state of the individual. Unfortunately, the body cell mass cannot be measured directly. However, a linear relationship has been demonstrated between the body cell mass and total exchangeable potassium eKe), which is equal to total body potassium. s Moore et aJ.5 described an excellent correlation, with narrow confidence limits, between the intracellular water volume and Ke, in a large group of normal volunteers. Kinney et aP demonstrated that the resting metabolic expenditure is a function of Ke. Talso et al. 6 ,7, in a large group of patients with a variety of water and electrolyte abnormalities, reported a constant potassium to nitrogen ratio in skeletal muscle, despite marked variations in the muscle content of water, sodium, potassium, and nitrogen. Recently, by means of a whole body counter and neutron activation analysis, an excellent correlation was reported between the total body nitrogen and Ke in 164 patients. 2 There are thus considerable data in the literature supporting, the use of Ke as an indirect measure of the body cell mass. Isotope dilution with a radioactive isotope of potassium is the method of choice for measuring Ke. Unfortunately, the radioactive isotopes of potassium, which are commerically available, namely potassium-42 and potassium-43, decay rapidly with half lives of 12.5 and 22.4 hours respectively. Their use in clinical studies is therefore logistically inconvenient and expensive. The whole body counter offers an alternate method of measuring Ke. This method is based on the fact that 0.012 per cent of all naturally occurring potassium is in the form of potassium-40, which is radioactive. However, the whole body counter is an expensive installation. In addition its calibration remains a difficult problem. 9 A technique has therefore been developed which indirectly measures Ke. This method involves the use of sodium-22 and tritiated water. Both isotopes are safe and convenient to use in clinical investigaions. "Medical Research Council of Canada Scholar and Assistant Professor of Surgery, McGill University; ~ssistant Surgeon, Royal Victoria Hospital, Montreal, Quebec, Canada
Surgical Clinics of North America-Vol. 56, No.5, October 1976
1185
1186
HARRY
M.
SHIZGAL
THE INDIRECT MEASUREMENT OF K" The indirect measurement of K., was based on the hypothesis that the ratio (R), where R _ Na + K - H20
(1)
where Na = sodium content K = potassium content H 2 0 = water content is a constant for all tissues within any individual. This hypothesis was based on the absence of a significant osmotic gradient between the intracellular and extracellular compartments (except for several specialized tissues) and on the fact that the major source of osmotic pressure of total body water is the electrolytes, of which sodium and potassium are the two principal cations. If the ratio R is a constant for all the tissues within the individual, it follows that it must also be equal to the sum of all the tissues, i.e., R= Na.,+ K., TBW
(2)
where Na., = total exchangeable sodium TBW = total body water Rearranging equation 2 results in the following: K., = R (TBW) - Na.,
(3)
The latter relationship indicates that K., can be calculated if the remaining parameters are measured. TBW and Na., can both be easily measured by isotope dilution using tritiated water and sodium-22 respectively. In addition, the ratio R can be determined by measuring the sodium, potassium, and water content of a sample of whole blood. The validity of equation 3 has been demonstrated experimentally in both man and several laboratory animals. to The experiments involved both normal animals and animals with a variety of pathologic states. Because of space limitations only the results of the human studies will be summarized. In 20 patients, many of whom were in a terminal state because of neoplastic disease, K., was simultaneously determined indirectly and directly by potassium-42 dilution. The correlation between the two sets of measurements were excellent with a correlation coefficient of 0.99 (Fig. 1). In addition, in a group of 25 normal volunteers, where K., was expressed as a function of TBW (K./TBW), to normalize for the variation in body size, the mean K.,ITBW was 80.0 (SEM = 1.03) mEq per L. The
1187
TOTAL BODY POTASSIUM AND NUTRITIONAL STATUS
•
CT
w
E
4000
::;
::>
iii Ul
Figure 1. Total exchangeable potassium was simultaneously determined indirectly and directly by isotope dilution using potassium-42. The correlation between the two sets of measurements is excellent, with a correlation coefficient (r) of 0.988. The regression line (solid line) is almost identical to the line of identity (broken line).
~
3000
/
i Cl
~
2000
I U
X
W ...J
g....
y = -145 + 1.08x
, = 0.988
1000
.....
..•....... .••.. 1000
2000
3000
4000
TOTAL EXCHANGEA8LE POTASSIUM (mEq) k-42 DILUTION
latter was not significantly different from the mean of 81.9 (SEM 1.13) mEq per L calculated from the data of Moore et al.,5 who measured K., by K-42 dilution in 33 normal individuals. The indirect determination of K., involves the intravenous injection of 500 microcuries of tritiated water and 8 microuries of sodium-22. A 15 ml whole blood sample is obtained at 24 hours. In addition, the urine excreted during this period is collected to determine the amount of isotope excreted during the 24-hour equilibration period. The TBW volume and N~ are determined by mixing equal volumes of the 24-hour plasma sample and a 10 per cent trichloroacetic acid solution to precipitate the plasma proteins. One milliliter of the resultant supernatant is added to 10 ml of a liquid scintillation solution (Aquasol, New England Nuclear) and counted in a liquid scintillation counter. Differential counting is employed to determine the plasma concentration of each isotope. This involves counting each sample with two different window settings on the pulse height analyser. Each window corresponds to the major energy peak of the radiation given off by each isotope. The amount of isotope excreted during the 24-hour equilibration period is determined by measuring, in a similar manner, the concentration of the two isotopes in the urine. The TBW volume is given by TBW = counts/min injected - counts/min excreted plasma tritium concentration at 24 hours and the N~
N~
is determined from
= (counts/min injected - counts/min excreted) (plasma Na cone.) plasma Na-22 concentration at 24 hours
1188
HARRY
M.
SHIZGAL
To determine the ratio R, a 1 ml aliquot of the 24-hour whole blood sample is desiccated to constant weight to obtain the water content. The whole blood sodium and potassium concentrations are determined by flame photometry and corrected for water content. The ratio R is calculated using equation 1. With the direct measurement of Na", TBW, and the ratio R, it is possible to calculate K" using equation 3. The indirect measurement of K e , described above, is accurate, safe, and simple to perform. The isotopes employed are inexpensive and relatively stable. The half-life of tritium and sodium-22 is 12.4 and 2.6 years respectively. The total radiation exposure to patients resulting from the intravenous injection of these isotopes is 257 mRem, which is much less than the radiation received from either a barium meal or an intravenous pyelogram. In addition, both TBW and Na" are important body composition parameters. The TBW is important as it serves as the independent variable, Le., it is used to correct for variation in body size. Moore etal.5 have demonstrated that the various parameters of body composition correlate .better with TBW than body weight. Thus, when normalizing for body size, both K" and Na" can either be plotted against the TBW volume or expressed as a function of TBW, i.e., Na,,/TBW and K,,/TBW (Fig. 2). Na., is also an important body composition parameter as it is a measure of the extracellular supporting component of body composition. With malnutrition, as the body cell mass contracts, indicated by a decrease inK", Nae increases, Le., the extracellular component of body composition expands. Thus the ratio of Na" to K" (Na,,/K,,) becomes a sensitive index of the state of nutrition.
4000
3
.§. 3000 ~
~ 2000
w
~
~
z 1000
~
*
20
40
60
TOTAL BODY WATER ILl
Figure 2. The K., is plotted as a function of TBW for 18 patients who are in a chronic catabolic state. Included are the regression line and 95 per cent confidence limits determined from the data obtained in 25 normal volunteers. The normal range is therefore defined by the area bounded by the two outer confidence limits.
1189
TOTAL BODY POTASSIUM AND NUTRITIONAL STATUS
CHRONIC CATABOLIC STATE The measurements described above were performed in 18 patients who were in a chronic catabolic state. The majority of these patients were in our surgical intensive care unit for a prolonged period of time with a variety of problems (sepsis, gastrointestinal fistulae, pancreatitis, inflammatory bowel disease, etc.). Intravenous hyperalimentation was indicated in all of these patients, but was not administered for a variety of reasons. The normal range of N a" and Ke was defined by similar measurements in a group of 25 normal volunteers (see Fig. 2). In both the normal and catabolic patients the data were normalized for variation in body size by expressing Na" and Ke as a function of TBW (Fig. 3). As expected the body cell mass was markedly contracted in the catabolic patients as indicated by a Ke/TBW of 49.2 (SEM = 2.3) mEq per L. This was significantly lower (p < .05) than the mean of 80.0 (SEM = 1.0) mEq per L observed in the normal volunteers. In contrast, the extracellular component of body composition was expanded in the catabolic group as indicated by a mean Na,,/TBW of 96.9 (SEM = 4.8) mEq per L, compared to a mean, in the normal group, of 77.5 (SEM ~ 0.9) mEq per L. Because in the catabolic patients the Na" increased while the ke decreased, the ratio Na,,/Ke was a sensitive index of the individual's state of nutrition. The latter ratio is in effect a measure of the extracellular component of body composition, expressed as a function of the body cell mas·s. In the normal group the mean ratio was 0.98 (SEM = 0.2), compared to 2.1 (SEM = 0.14) in the catabolic group.
120
o
Normal
•
Catabolic
2.0
• '"
10 --.
'"
Z
*
Figure 3. The mean and standard error of the mean for the data obtained from the 25 normal volunteers and the 18 patients who were in a chronic catabolic state. The K./TBW is a measure of the body cell mass while the N a,ITBW is a measure of the extracellular supporting component of body composition.
1190
HARRY
M.
SHIZGAL
THE PROTEIN SPARING EFFECT OF AMINO ACID SOLUTIONS FOLLOWING MAJOR SURGERY Following a major surgical procedure the normal patient experiences a short but intensive period of catabolism, which is directly related to the severity of surgical trauma. 4 In most instances this catabolic state persists for 2 to 5 days and is characterized by positive water and sodium balance, and negative nitrogen and potassium balance. Recently Blackburn and Flatt' proposed the hypothesis that significant protein sparing can be achieved by avoiding the intravenous infusion of glucose containing solutions in the postoperative period and using an amino-acid solution to infuse the required fluids and electrolytes. This hypothesis is based on the inhibition of fat mobilization by the rise in plasma insulin secondary to the intravenous infusion of glucose. The decreased availability of lipid, as an endogenous fuel, in the semistarved patients, necessitates increased protein breakdown to provide endogenous calories. The total calories infused intravenously, in the majority of postoperative patients, is insufficient to meet their daily requirements. In contrast, the infusion of amino acid containing solutions does not increase the plasma insulin concentration, permitting lipid mobilization to meet the daily energy requirements. In addition, the infusion of amino acid solutions supplies the daily protein requirement. Thus, according to this hypothesis, body protein is spared. To test this hypothesis, the measurements described above were performed preoperatively and on the fifth postoperative day, in two groups of 19 patients each who were undergoing major abdominal surgery.u In the majority either a gastric or colon resection was performed. The patients were divided into the two groups depending on the type of fluid infused postoperatively. The first group received all their intravenous fluids and electrolytes as a 5 per cent glucose solution. In the second group, glucose containing solutions were excluded, and were replaced by a 5 per cent casein hydrolysate (Amigen, Baxter Laboratories). In the patients receiving glucose, there was a significant (p < .001) decrease in the mean body weight of 2.6 kg on the fifth postoperative day while the TBW volume decreased by 0.58 liters. Since the lean body mass is equal to TBW(0.73 and body fat is the difference between the body weight and lean body mass,5 the observed postoperative decrease in body weight was due to a decrease in both the lean body mass and body fat. In addition, in these patients there was a significant decrease (p < .001) in K., of 391 mEq, indicating a loss of body cell mass. This was accompanied by an increase in the Na" (p < .01), Le., an expansion of the extracellular component of body composition. In the group of patients receiving amino acids postoperatively, the mean body weight decreased by 2.0 kg (p < .001) by the fifth postoperative day. The mean TBW, on the other hand, increased by 1.3 liters (p < .05), implying an increase in the lean body mass. Thus the postoperative decrease of body weight in the patients receiving amino acid was due only to a loss of body fat. In contrast, as described above, in the glucose
1191
TOTAL BODY POTASSIUM AND NUTRITIONAL STATUS
~ PRE-OP • POST-OP
p-
'"
• •
•
0 "Cl >£
0 0
80 60
0
Figure 6. The same as Figure 5 except that the total cal/kg infused per day was plotted against the mean daily change in the N a.,/Ke ratio.
...• y~511
•
(5
fo-
0
-267x
50 p < 0 001
-.06
ro ro
u
• ... •
,~O
-08
"-
20
-.04
-.02
0
.02
.04
/:,. Nae/Ke/Day
of intravenous hyperalimentation was 1.54 (SEM = .08). Thus any improvement in nutritional status results in a decrease in this ratio. Thus the inverse relationship observed between the daily mean change of N~/I(" and the total calories infused (Fig. 6). Because of the large scatter in the data, the correlation coefficient was only 0.50. Nevertheless, it was statistically significant (p < .001). The intercept of the regression curve was 51.1 cal per kg per day. Thus the data indicate that in this group of patients, nutritional balance was achieved by infusing approximately 50 cal per kg per day (see Figs. 5 and 6). The infusion of calories in excess of this amount should therefore result in an expansion of the body cell mass, while the opposite should occur when less calories are infused.
SUMMARY A method for the indirect measurement of I(" is described. It is accurate, safe, and simple to perform in the clinical setting, and provides quantitative information regarding both the extracellular supporting component of body composition and the body cell. As a result, this technique was applied to quantitatively assess the nutritional status of several groups of surgical patients. A marked contraction of the body cell mass, as indicated by a significant decrease in 1(", was recorded in a group of patients in a chronic catabolic state. The decrease in I(" was accompanied by a reciprocal expansion of the extracellular supporting component of body composition. Thus, in these patients, there was a large increase in the N~/I(" ratio. Similar changes, which however were much smaller in magnitude, were also observed following a major abdominal operation in a group of patients whose postoperative intravenous fluids contained glucose. In contrast, in a second similar group of patients, the above changes in N~/I(" were avoided, and postoperative protein sparing was achieved, by avoiding intravenous glucose in the postoperative period and infusing
HARRY
1194
M.
SHIZGAL
instead solutions containing amino acids. Finally, data were presented which indicates that the effect of a chronic catabolic state on both Na" and K. can be prevented by administering intravenous hyperalimentation, provided that the total calories infused exceeded 50 cal per kg per day.
REFERENCES 1. Blackburn, G. L., and Flatt, J. P.: Protein sparing therapy during period of starvation with sepsis or trauma. Ann. Surg., 177 :588, 1973. 2. Harvey, T. C., et al.: Measurement of whole body nitrogen by neutron activation analysis. Lancet, 2:395,1973. 3. Kinney, J. M., Lister, J., and Moore, F. D.: Relationship of energy expenditure to total exchangeable potassium. Ann. N.Y. Acad. Sci.,ll0:711, 1963. 4. Moore, F. D.: Metabolic Care of the Surgical Patient. Philadelphia, W. B. Saunders Co., 1959. 5. Moore, F. D., Olesen, K. H., McMurray, J. D., et al.: The Body Cell Mass and Its Supporting Environment. Body Composition in Health and Disease. Philadelphia, W. B. Saunders Co., 1963. 6. Talso, P. J., Spafford, M., and Blaw, M.: The metabolism of water and electrolytes in congestive heart failure. The electrolyte and water content of normal human skeletal muscle. J. Lab. Clin. Med., 41 :405, 1953. 7. Talso, P. J., Spafford, N. S., and Blaw, M.: The metabolism of water and electrolyte in congestive heart failure: The distribution of water and electrolytes in edematous patients with congestive heart failure before and after treatment. J. Lab. Clin. Med., 41: 405,1953. 8. Talso, P. J., Miller, C. E., Carballo, A. J., and Vasquez, I.: Exchangeable potassium as a parameter of body composition. Metabolism, 9:456, 1960. 9. Tyson, I., Sebastian, G., Jones, R. L., et al.: Body potassium measurements with a total body counter. J. Nucl. Med., 2:255,1970. 10. Shizgal, H. M., Spanier, A. H., Humes, J., and Wood, C. D.: The indirect measurement of total body potassium. In press. 11. Spanier, A. H., Carmody, P., Milne, C. A., and Shizgal, H. M.: Preservation of body cell mass following major abdominal surgery. Surg. Forum, 26:3, 1975. Royal Victoria Hospital 687 Pine Avenue West Montreal, Quebec Canada
Total Body Potassium' and Nutritional Status Harry M. Shizgal, M.D. *
The body cell mass was defined by Moore as that component of body composition "containing -the oxygen-exchanging, potassium-rich, glucose-oxidizing, work-performing tissue."5The body cell mass is in effect the living component of body composition and therefore its measurement permits a quantitative assessment of the nutritional state of the individual. Unfortunately, the body cell mass cannot be measured directly. However, a linear relationship has been demonstrated between the body cell mass and total exchangeable potassium eKe), which is equal to total body potassium. s Moore et aJ.5 described an excellent correlation, with narrow confidence limits, between the intracellular water volume and Ke, in a large group of normal volunteers. Kinney et aP demonstrated that the resting metabolic expenditure is a function of Ke. Talso et al. 6 ,7, in a large group of patients with a variety of water and electrolyte abnormalities, reported a constant potassium to nitrogen ratio in skeletal muscle, despite marked variations in the muscle content of water, sodium, potassium, and nitrogen. Recently, by means of a whole body counter and neutron activation analysis, an excellent correlation was reported between the total body nitrogen and Ke in 164 patients. 2 There are thus considerable data in the literature supporting, the use of Ke as an indirect measure of the body cell mass. Isotope dilution with a radioactive isotope of potassium is the method of choice for measuring Ke. Unfortunately, the radioactive isotopes of potassium, which are commerically available, namely potassium-42 and potassium-43, decay rapidly with half lives of 12.5 and 22.4 hours respectively. Their use in clinical studies is therefore logistically inconvenient and expensive. The whole body counter offers an alternate method of measuring Ke. This method is based on the fact that 0.012 per cent of all naturally occurring potassium is in the form of potassium-40, which is radioactive. However, the whole body counter is an expensive installation. In addition its calibration remains a difficult problem. 9 A technique has therefore been developed which indirectly measures Ke. This method involves the use of sodium-22 and tritiated water. Both isotopes are safe and convenient to use in clinical investigaions. "Medical Research Council of Canada Scholar and Assistant Professor of Surgery, McGill University; ~ssistant Surgeon, Royal Victoria Hospital, Montreal, Quebec, Canada
Surgical Clinics of North America-Vol. 56, No.5, October 1976
1185
1186
HARRY
M.
SHIZGAL
THE INDIRECT MEASUREMENT OF K" The indirect measurement of K., was based on the hypothesis that the ratio (R), where R _ Na + K - H20
(1)
where Na = sodium content K = potassium content H 2 0 = water content is a constant for all tissues within any individual. This hypothesis was based on the absence of a significant osmotic gradient between the intracellular and extracellular compartments (except for several specialized tissues) and on the fact that the major source of osmotic pressure of total body water is the electrolytes, of which sodium and potassium are the two principal cations. If the ratio R is a constant for all the tissues within the individual, it follows that it must also be equal to the sum of all the tissues, i.e., R= Na.,+ K., TBW
(2)
where Na., = total exchangeable sodium TBW = total body water Rearranging equation 2 results in the following: K., = R (TBW) - Na.,
(3)
The latter relationship indicates that K., can be calculated if the remaining parameters are measured. TBW and Na., can both be easily measured by isotope dilution using tritiated water and sodium-22 respectively. In addition, the ratio R can be determined by measuring the sodium, potassium, and water content of a sample of whole blood. The validity of equation 3 has been demonstrated experimentally in both man and several laboratory animals. to The experiments involved both normal animals and animals with a variety of pathologic states. Because of space limitations only the results of the human studies will be summarized. In 20 patients, many of whom were in a terminal state because of neoplastic disease, K., was simultaneously determined indirectly and directly by potassium-42 dilution. The correlation between the two sets of measurements were excellent with a correlation coefficient of 0.99 (Fig. 1). In addition, in a group of 25 normal volunteers, where K., was expressed as a function of TBW (K./TBW), to normalize for the variation in body size, the mean K.,ITBW was 80.0 (SEM = 1.03) mEq per L. The
1187
TOTAL BODY POTASSIUM AND NUTRITIONAL STATUS
•
CT
w
E
4000
::;
::>
iii Ul
Figure 1. Total exchangeable potassium was simultaneously determined indirectly and directly by isotope dilution using potassium-42. The correlation between the two sets of measurements is excellent, with a correlation coefficient (r) of 0.988. The regression line (solid line) is almost identical to the line of identity (broken line).
~
3000
/
i Cl
~
2000
I U
X
W ...J
g....
y = -145 + 1.08x
, = 0.988
1000
.....
..•....... .••.. 1000
2000
3000
4000
TOTAL EXCHANGEA8LE POTASSIUM (mEq) k-42 DILUTION
latter was not significantly different from the mean of 81.9 (SEM 1.13) mEq per L calculated from the data of Moore et al.,5 who measured K., by K-42 dilution in 33 normal individuals. The indirect determination of K., involves the intravenous injection of 500 microcuries of tritiated water and 8 microuries of sodium-22. A 15 ml whole blood sample is obtained at 24 hours. In addition, the urine excreted during this period is collected to determine the amount of isotope excreted during the 24-hour equilibration period. The TBW volume and N~ are determined by mixing equal volumes of the 24-hour plasma sample and a 10 per cent trichloroacetic acid solution to precipitate the plasma proteins. One milliliter of the resultant supernatant is added to 10 ml of a liquid scintillation solution (Aquasol, New England Nuclear) and counted in a liquid scintillation counter. Differential counting is employed to determine the plasma concentration of each isotope. This involves counting each sample with two different window settings on the pulse height analyser. Each window corresponds to the major energy peak of the radiation given off by each isotope. The amount of isotope excreted during the 24-hour equilibration period is determined by measuring, in a similar manner, the concentration of the two isotopes in the urine. The TBW volume is given by TBW = counts/min injected - counts/min excreted plasma tritium concentration at 24 hours and the N~
N~
is determined from
= (counts/min injected - counts/min excreted) (plasma Na cone.) plasma Na-22 concentration at 24 hours
1188
HARRY
M.
SHIZGAL
To determine the ratio R, a 1 ml aliquot of the 24-hour whole blood sample is desiccated to constant weight to obtain the water content. The whole blood sodium and potassium concentrations are determined by flame photometry and corrected for water content. The ratio R is calculated using equation 1. With the direct measurement of Na", TBW, and the ratio R, it is possible to calculate K" using equation 3. The indirect measurement of K e , described above, is accurate, safe, and simple to perform. The isotopes employed are inexpensive and relatively stable. The half-life of tritium and sodium-22 is 12.4 and 2.6 years respectively. The total radiation exposure to patients resulting from the intravenous injection of these isotopes is 257 mRem, which is much less than the radiation received from either a barium meal or an intravenous pyelogram. In addition, both TBW and Na" are important body composition parameters. The TBW is important as it serves as the independent variable, Le., it is used to correct for variation in body size. Moore etal.5 have demonstrated that the various parameters of body composition correlate .better with TBW than body weight. Thus, when normalizing for body size, both K" and Na" can either be plotted against the TBW volume or expressed as a function of TBW, i.e., Na,,/TBW and K,,/TBW (Fig. 2). Na., is also an important body composition parameter as it is a measure of the extracellular supporting component of body composition. With malnutrition, as the body cell mass contracts, indicated by a decrease inK", Nae increases, Le., the extracellular component of body composition expands. Thus the ratio of Na" to K" (Na,,/K,,) becomes a sensitive index of the state of nutrition.
4000
3
.§. 3000 ~
~ 2000
w
~
~
z 1000
~
*
20
40
60
TOTAL BODY WATER ILl
Figure 2. The K., is plotted as a function of TBW for 18 patients who are in a chronic catabolic state. Included are the regression line and 95 per cent confidence limits determined from the data obtained in 25 normal volunteers. The normal range is therefore defined by the area bounded by the two outer confidence limits.
1189
TOTAL BODY POTASSIUM AND NUTRITIONAL STATUS
CHRONIC CATABOLIC STATE The measurements described above were performed in 18 patients who were in a chronic catabolic state. The majority of these patients were in our surgical intensive care unit for a prolonged period of time with a variety of problems (sepsis, gastrointestinal fistulae, pancreatitis, inflammatory bowel disease, etc.). Intravenous hyperalimentation was indicated in all of these patients, but was not administered for a variety of reasons. The normal range of N a" and Ke was defined by similar measurements in a group of 25 normal volunteers (see Fig. 2). In both the normal and catabolic patients the data were normalized for variation in body size by expressing Na" and Ke as a function of TBW (Fig. 3). As expected the body cell mass was markedly contracted in the catabolic patients as indicated by a Ke/TBW of 49.2 (SEM = 2.3) mEq per L. This was significantly lower (p < .05) than the mean of 80.0 (SEM = 1.0) mEq per L observed in the normal volunteers. In contrast, the extracellular component of body composition was expanded in the catabolic group as indicated by a mean Na,,/TBW of 96.9 (SEM = 4.8) mEq per L, compared to a mean, in the normal group, of 77.5 (SEM ~ 0.9) mEq per L. Because in the catabolic patients the Na" increased while the ke decreased, the ratio Na,,/Ke was a sensitive index of the individual's state of nutrition. The latter ratio is in effect a measure of the extracellular component of body composition, expressed as a function of the body cell mas·s. In the normal group the mean ratio was 0.98 (SEM = 0.2), compared to 2.1 (SEM = 0.14) in the catabolic group.
120
o
Normal
•
Catabolic
2.0
• '"
10 --.
'"
Z
*
Figure 3. The mean and standard error of the mean for the data obtained from the 25 normal volunteers and the 18 patients who were in a chronic catabolic state. The K./TBW is a measure of the body cell mass while the N a,ITBW is a measure of the extracellular supporting component of body composition.
1190
HARRY
M.
SHIZGAL
THE PROTEIN SPARING EFFECT OF AMINO ACID SOLUTIONS FOLLOWING MAJOR SURGERY Following a major surgical procedure the normal patient experiences a short but intensive period of catabolism, which is directly related to the severity of surgical trauma. 4 In most instances this catabolic state persists for 2 to 5 days and is characterized by positive water and sodium balance, and negative nitrogen and potassium balance. Recently Blackburn and Flatt' proposed the hypothesis that significant protein sparing can be achieved by avoiding the intravenous infusion of glucose containing solutions in the postoperative period and using an amino-acid solution to infuse the required fluids and electrolytes. This hypothesis is based on the inhibition of fat mobilization by the rise in plasma insulin secondary to the intravenous infusion of glucose. The decreased availability of lipid, as an endogenous fuel, in the semistarved patients, necessitates increased protein breakdown to provide endogenous calories. The total calories infused intravenously, in the majority of postoperative patients, is insufficient to meet their daily requirements. In contrast, the infusion of amino acid containing solutions does not increase the plasma insulin concentration, permitting lipid mobilization to meet the daily energy requirements. In addition, the infusion of amino acid solutions supplies the daily protein requirement. Thus, according to this hypothesis, body protein is spared. To test this hypothesis, the measurements described above were performed preoperatively and on the fifth postoperative day, in two groups of 19 patients each who were undergoing major abdominal surgery.u In the majority either a gastric or colon resection was performed. The patients were divided into the two groups depending on the type of fluid infused postoperatively. The first group received all their intravenous fluids and electrolytes as a 5 per cent glucose solution. In the second group, glucose containing solutions were excluded, and were replaced by a 5 per cent casein hydrolysate (Amigen, Baxter Laboratories). In the patients receiving glucose, there was a significant (p < .001) decrease in the mean body weight of 2.6 kg on the fifth postoperative day while the TBW volume decreased by 0.58 liters. Since the lean body mass is equal to TBW(0.73 and body fat is the difference between the body weight and lean body mass,5 the observed postoperative decrease in body weight was due to a decrease in both the lean body mass and body fat. In addition, in these patients there was a significant decrease (p < .001) in K., of 391 mEq, indicating a loss of body cell mass. This was accompanied by an increase in the Na" (p < .01), Le., an expansion of the extracellular component of body composition. In the group of patients receiving amino acids postoperatively, the mean body weight decreased by 2.0 kg (p < .001) by the fifth postoperative day. The mean TBW, on the other hand, increased by 1.3 liters (p < .05), implying an increase in the lean body mass. Thus the postoperative decrease of body weight in the patients receiving amino acid was due only to a loss of body fat. In contrast, as described above, in the glucose
1191
TOTAL BODY POTASSIUM AND NUTRITIONAL STATUS
~ PRE-OP • POST-OP
p-
'"
• •
•
0 "Cl >£
0 0
80 60
0
Figure 6. The same as Figure 5 except that the total cal/kg infused per day was plotted against the mean daily change in the N a.,/Ke ratio.
...• y~511
•
(5
fo-
0
-267x
50 p < 0 001
-.06
ro ro
u
• ... •
,~O
-08
"-
20
-.04
-.02
0
.02
.04
/:,. Nae/Ke/Day
of intravenous hyperalimentation was 1.54 (SEM = .08). Thus any improvement in nutritional status results in a decrease in this ratio. Thus the inverse relationship observed between the daily mean change of N~/I(" and the total calories infused (Fig. 6). Because of the large scatter in the data, the correlation coefficient was only 0.50. Nevertheless, it was statistically significant (p < .001). The intercept of the regression curve was 51.1 cal per kg per day. Thus the data indicate that in this group of patients, nutritional balance was achieved by infusing approximately 50 cal per kg per day (see Figs. 5 and 6). The infusion of calories in excess of this amount should therefore result in an expansion of the body cell mass, while the opposite should occur when less calories are infused.
SUMMARY A method for the indirect measurement of I(" is described. It is accurate, safe, and simple to perform in the clinical setting, and provides quantitative information regarding both the extracellular supporting component of body composition and the body cell. As a result, this technique was applied to quantitatively assess the nutritional status of several groups of surgical patients. A marked contraction of the body cell mass, as indicated by a significant decrease in 1(", was recorded in a group of patients in a chronic catabolic state. The decrease in I(" was accompanied by a reciprocal expansion of the extracellular supporting component of body composition. Thus, in these patients, there was a large increase in the N~/I(" ratio. Similar changes, which however were much smaller in magnitude, were also observed following a major abdominal operation in a group of patients whose postoperative intravenous fluids contained glucose. In contrast, in a second similar group of patients, the above changes in N~/I(" were avoided, and postoperative protein sparing was achieved, by avoiding intravenous glucose in the postoperative period and infusing
HARRY
1194
M.
SHIZGAL
instead solutions containing amino acids. Finally, data were presented which indicates that the effect of a chronic catabolic state on both Na" and K. can be prevented by administering intravenous hyperalimentation, provided that the total calories infused exceeded 50 cal per kg per day.
REFERENCES 1. Blackburn, G. L., and Flatt, J. P.: Protein sparing therapy during period of starvation with sepsis or trauma. Ann. Surg., 177 :588, 1973. 2. Harvey, T. C., et al.: Measurement of whole body nitrogen by neutron activation analysis. Lancet, 2:395,1973. 3. Kinney, J. M., Lister, J., and Moore, F. D.: Relationship of energy expenditure to total exchangeable potassium. Ann. N.Y. Acad. Sci.,ll0:711, 1963. 4. Moore, F. D.: Metabolic Care of the Surgical Patient. Philadelphia, W. B. Saunders Co., 1959. 5. Moore, F. D., Olesen, K. H., McMurray, J. D., et al.: The Body Cell Mass and Its Supporting Environment. Body Composition in Health and Disease. Philadelphia, W. B. Saunders Co., 1963. 6. Talso, P. J., Spafford, M., and Blaw, M.: The metabolism of water and electrolytes in congestive heart failure. The electrolyte and water content of normal human skeletal muscle. J. Lab. Clin. Med., 41 :405, 1953. 7. Talso, P. J., Spafford, N. S., and Blaw, M.: The metabolism of water and electrolyte in congestive heart failure: The distribution of water and electrolytes in edematous patients with congestive heart failure before and after treatment. J. Lab. Clin. Med., 41: 405,1953. 8. Talso, P. J., Miller, C. E., Carballo, A. J., and Vasquez, I.: Exchangeable potassium as a parameter of body composition. Metabolism, 9:456, 1960. 9. Tyson, I., Sebastian, G., Jones, R. L., et al.: Body potassium measurements with a total body counter. J. Nucl. Med., 2:255,1970. 10. Shizgal, H. M., Spanier, A. H., Humes, J., and Wood, C. D.: The indirect measurement of total body potassium. In press. 11. Spanier, A. H., Carmody, P., Milne, C. A., and Shizgal, H. M.: Preservation of body cell mass following major abdominal surgery. Surg. Forum, 26:3, 1975. Royal Victoria Hospital 687 Pine Avenue West Montreal, Quebec Canada
