Calcium Requirements During Total Parentera/ Nutrition in Well-nourished lndividuals Marion F. Wittine, B.A. Joel B. Freeman, M.D., F.A.C.S., F.R.C.S. (C)

Front the Departnients of Surgery, University of Iona and Veterans Adniinistration Hospitals. The University of Iowa, Io1c.a City, Iowa. This study i w s supported by Gram RR-S9front the General Clinical Research Centers Prograni. Division of Research Resources, National Institutes of Health. A n abstract of this paper was presented at the First hfeeting of the Association for Paren feral and Enteral Nutrition.. Chicago. Illinois, January 1977. Requests f o r reprints should be addressed to Joel Freeman. hf.D.. Chiej; Departnient of Surgery. 0ttatc.a General Hospital. University of Ottawa. 0ttaic.a. Ontario K I N SC8. Canada. ,

ABSTRACT: Eleven patients, receiving all nutrition intravenously, were given varying doses of calcium (0-20 mg/ kg/day) to determine an optimal level for calcium administration during postoperative parenteral nutrition. During each study period, nitrogen, phosphorus, vitamin, and caloric intakes were constant. Negative calcium balance resulted when less than 2 mgCa++/kg body weight was given daily. During excessive urinary c5lcium losses, serum calcium concentration remained in the normal range. Increasing calcium intake to approximately 5 mg/kg/day (500 mg/day) yielded an apparent retention of calcium, as did higher doses. Serum calcium did not rise a t this time. Urinary calcium excretion was directly proportional to calcium intake. T h e preliminary data suggest that a minimum dose of 5 mg Ca++/kg/ day is necessary to attain equilibrium between intake and urinary output. This value is higher than recent suggestions for calcium replacement duringintravenous feeding.

The necessity for calcium in bone formation, neuromuscular function, blood clotting, and cell permeability has led t o recommendations for adequate calcium intake, depending on age and physical condition.52 Metabolic studies indicate that calcium absorption and excretion are affected not only by the amount of calcium ingested but also by many dietary components (protein, vitamin D, phosphorus, and magnesium), as well as diuretics.34 Most calcium balance data to date are derived from oral balance studies and extrapolation of these data has been used to formulate the intravenous dose. The required dietary allowance of calcium is 800 mg/day for adults.9 Values based on weight indicate that equilibrium is attained with intakes of 9 to 10 mg/kg/day.l0-12 The net absorption of dietary calcium ranges from 5% to 40%.5"-'3 There are few studies on the calcium needs of parenterally-fed subjects. As a result, the amount of calcium to be infused daily is not well established. The dosages vary, and are lower than those for oral intake because the absorption factor is eliminated during intravenous feeding. During eucaloric, total parenteral nutrition (TPN) there are two approaches t o calcium administration for adults: either a fixed amount of calcium (100-400 mg/2500 kcal) is added daily, or calcium is omitted, unless its serum concentration is low. 14J5 During hypercaloric alimentation of undernourished patients, Rudman and associates recommended a calcium dose of 2.56 mgfkg ideal body weight.16 '

The purpose of the current study was t o assess the effect of different calcium infusions (0-20 mg/ kg/day) on calcium excretion in adult surgical patients receiving all nutrition intravenously, either eucalorically as T P N o r hypocalorically as amino acids with and without dextrose. Our intention was t o determine an optimal level for calcium administration during parenteral alimentation. TABLE I CLINICAL SUhlMARY OF SUBJECTS Sex: 9 females and 2 males f 5 years Weight (Preoperative): 101 f 13 kg Problem: Gastric bypass (7) Sphincteroplasty (2) Left hemicolectomy ( I ) ldioaathic diarrhea [ I )

vitamin infusion3 containing 1000 international, units (I.U.) of vitamin D were given daily. The source of added calcium was calcium gluconate in seven patients and calcium chloride dihydrate in three patients. Ten milliliters of 10% calcium gluconate .solution contain approximately 100 mg C a o r 5 mEq; 10 milliliters of 10% CaC12 2H:O solution is roughly three times more concentrated. T h e protein hydrolysate contained 100 mg Ca++/l.This was accounted forin the balancestudies. A11 solutions were infused through the subclavian vein using a constant infusion pump. T h e nitrogen, phosphorus, and caloric intakes were maintained constant during each study period.

Age: 43

SUBJECTS Table I su'mmarizes the clinical data. Eleven patiknts were studied in the Clinical Research Center for approximately two weeks following surgery. Their ages ranged from 27 t o '77 with a n average of 43 (k 5) years. The mean preoperative weight of 101 (+ 13) kg was high due t o seven patients with morbid obesity who had gastric bypass operations performed. All preoperative serum calcium 'levels were normal. T h e subjects participated in 15 b8ance studies, rangingfrom4 t o 27 days, with a mean of fd (* 2) days. Due t o the controversy regarding the time needed for Ca++adaptation, and concern for the patients' wishes, no adaptation period was utilized. Balance data were collected beginning at 8 A M on the first postoperative day with the initiation of T P N or other IV designated in the protocol. In three crossover studies, each patient received two levels of calcium. One subject on T P N was later infused with'normal saline t o compare calcium excretion during eucaloricandacaloric parenteral infusions.

MATERIALS Oral intake was restficted t o distilled, ,water. Nine patients received TPN 'consisting of 25% dextrose with the source of nitrogen either 4%crystalline amino acids* or 5% protein hydrolysate.7 'Thiave;ag;'e nifrogen intake was 17.2 (& 0.5) gm per day, J,wo patients*.received crystal- , isonitrogenous, but hypocaloric infusions of line amino acids*,* ,with 'and without 5% dextrose?. Only " one patient 'received blood. Ten, ,m,$iliters,. 2f a ,multi,,, ,'I.'. . . . .. :

4%

-

'Abborr Anrinosyn (7%). ::.tBa-vgi A_n?igen(10%). :,,:; *'hfcGaw FreArnine I I (8.S%)

,;

I,

.

,. .,j,I,

'3 -

>:I--

.. c:

)

METHODS All urine was collected in 24 hour containers. Aliquots were analyzed daily for creatinine t o assure adequacy of collection and for calcium using a n automated spectrophotometric procedure.17P Serum calcium determinations were measured on the S M A 12/60. No fecal markers o r enemas- were administered. Random stool samples were analyzed in an atomic absorption flame photometer.19 RESULTS Analysis of 24 hour stool collections indicated a low range of 0.26 t o 0.45 mg of C a / k g actual body weight. Taking into account that urinary contamination was at times unavoidable, and that endogenous fecal values occasionally represented more than one day's losses, fecal calcium excretion was considered t o be negligible. In only one patient with idiopathic diarrhea, 5.3 mg/kg was excreted daily in stool. The difference between calcium intake and u-rinary calcium excretion was, therefore, taken to be the balance. The individual urinary calcium excretion varied considerably from day t o day, even though the calcium infused remained constant. Twenty-four hour urine volumes did nqt vary significantly. The daily urinary calcium excretion also varied widely in thosesubjects who received T P N and who were maintained on the same nitrogen, calcium, and' caloric intakes (Table, 11). For example, at a calcium intake of 0.78 mg/ kg/day, average urinary losses in each, subject were 0.86, 1.18, and 2.03 mg/kg/day, with a mean daily loss of 1.36 rngfkg. In two of.the .crossover studies, there was a n increase of .., . -fhree urjnary calcium with a higher intake. These results were not.,ivfluenced by the order of infusion of the respective . . .. doses. < A t zero calcium ..intake, the average calc,ium urina,ry calcium excretion was 2.48 mg/ kglday. AS the .of calcium increased, urinary calcium excretion intake . .)I,-.

)!

L

'>*],>,

,, :USV Laborarories, Tuckahoe. Neiv York

1

1

..

t

CALClU hl REQUl REMENTS

TABLE I I

EFFECT O F VARIOUS CA++ INTAKES ON URINARY CA++ EXCRETION A N D SERUhI CALCIUhf COSCENTRATIOS Solution 4% Aminosyn

+ 25% dextrose

4% Aminosyn -!-25% dextrose 4% Aminosyn

+ 25% dextrose

5% Arnigen f 25% dextrose 5% Amigen i25% dextrose 5% Arnigen i25% dextrose Normal saline 4% Fre-Amine 4% Fre-Amine i- 5% dextrose

Infused Calcium (mg/kg/day)

Urinary Calcium (mg/kg/day)

Balance (mg/kg/day)

Serum Calcium (mg/dl)

0

0.99.1.54t.4.92Y (2.48 f 1.23) 0.86*, I.l8t, 2.03 (1.36 f 0.35)1 1.40*, 1.96.5.46" (2.93 f 1.27)A 4.02 8.83 13.5010 3.65 2.25 1.48

-2.48

9.1 f 0.1

-0.58

8.6 f 0.1

- I .27

8.5 f 0.1

$0.86 4-0.95 +7.00= -3.65 -0.81

8.8 f 0.1 8.8 f 0.1 9.4 f 0.0 7.9 f 0.3 9.9 0.2 8.2 f 0.1

0.68*, 0.92t, 0.74 (0.78 f 0.07)A 1.39.. 1.63, 1.98Y (1.67 f 0.17)A 4.88 9.78 20.50 0 1.44 1.66

i-0.18

*

t Y *: Crossover srudies. Each sjnibol represents one individual who

IVOS srudied at two dflerent levels of infused calcium This porienr had idiopathic diarrhea with fecal losses of 5.3 mg Ca++/kg/day so rhar the true balance is octualb +1.7.

Q):

also increased. In each study with intakes equal to or less than 1.67 mg/ kg/day, urinary losses exceeded input. This resulted in a negative balance. There was apparent calcium -retention when 4.88 mg/kg/day or more was infused. The balance data for these higher calcium intakes were therefore positive. Urinary calcium losses and balances were similar in the subjects who received saline or hypocaloric amino acid solutions and in those who received T P N (Table 11). Hence, the calcium requirements of patients receiving hypocaloric solutions are similar to those receiving all caloric and nitrogen requirements. The mean serum calcium values obtained during the respective study periods were generally in the normal range (8.5-10/5 mg/dl), but overall these calcium levels were slightly but not significantly lower than values obtained preoperatively, irrespective of the amount of calcium infused. Albumin levels were lower than the preoperative values in 12 of the 15 studies. That may in part explain the slight fall in serum calcium. However, there was no correlation between serum calcium and albumin values obtained during the. study periods. Hypercalcemia was not observed at any infusion level. Significant hypocalcemia (less than 8.5 mg/dl) was observed in two patients. One of these received calciumfree normal saline infusions for nine days postoperatively. The other patient was on a protein sparing regimen consisting of 120 gm of crystalline amino acids and 150 gm of dextrose daily. Her calcium dose was 1.66 mg/kg/day. O n the fifth day after sphincteroplasty and common duct exploration she had several episodes of tetany, but without clinical evidence of pancreatitis. Hyperventilation was the presumed etiology.

DISCUSSION Calcium requirements depend upon several factors, including age, body size,, nutritional status, and renal function.12 Consideration must also begiven to the extent of immobilization and phosphorus and vitamin D intakes in determining calcium homeostasis. In the current study, all patients were encouraged to ambulate t o prevent hypercalciuria.2 Because of the possibility of calcium precipitation in the presence of large amounts of phosphates, Ca and P were added to separate IV bottles. No precipitation occurred with the addition of 27'mEq of Ca to 1 liter of 5%Amigen. The ratio of daily Ca:P intakes ranged from 0.04 to 0.31, except during periods when calcium and phosphorus were eliminated. In the current studies there was a direct relationship between the Ca:P infusion ratio and the urinary calcium losses. As the ratio increased, so did urinary calcium losses (r = 0.679). Positive calcium balance was attained with a Ca:P ratio of 0.2 and greater because the calcium increments exceeded urinary losses in each study period. Similar correlations between the Ca:P ratio and urinary calcium excretion have also been noted during oral calcium balance ~tudies.~P Unlike the high correlation between calcium intake and fecal calcium excretion found on oral diets,l3 our infusion studies showed that the urinary excretion of calcium was linearly related to calcium intake. The preliminary data suggest that a theoretical minimum dose of 4.5 mg Ca++/kg/day is needed to attain equilibrium. This value does not take into account calcium losses via sweat (15 mglday), feces (significant in one subject with idiopathic diarrhea), or prolonged gastrointestinal drainage.*$*? A positive balance was obtained at a calcium intake equal to

or greater than 4.88 mg/kg/day. Although urinary calcium excretion increased with higher intakes, calcium hoses greater than 4.5 mg/ kg more than compensated for \hese increased urinary losses and resulted in apparent I,alcium retention. Serum calcium concentrations d o not appear to be an acceptable measurement for determining calcium requirements. Serum levels remained within the normal range during consecutive days of excessive urinary caIcium excretion and negative calcium balance (Table 11). When calcium was withdrawn during TPN, daily urinary calcium losses averaged 250 mg. Yet the serum concentration was 9.1 mg/dl. Higher intakes did not produce a concomitant rise in serum calcium levels. Serum calcium concentrations were unrelated to calcium intake or to the Ca:P ratio in contradistinction to calcium balance data obtained during oral feeding.5 Low serum calcium values have been associated with magnesium deficiency, but each of our patients received maintenance doses of this metal.23 On an average oral diet, urinary calcium excretion is normally less than 150 mg/day.zJ However, this norm was exceeded in 10 out of 15 ofour intravenousstudies. When calcium was eliminated from the solutions, calcium conserving mechanisms were not apparent. Our results show that daily calcium infusions of 10 mEq (200 mg) or less did not maintain calcium equilibrium. A daily dose of 5 to 10 mg/kg actual weight was required to fulfill the needs of parenterally-fed adults and maintain a positive calcium balance. These data apply both to eucaloricTPN therapy and to hypocaloric protein sparing infusions.

5. McBean LD, Speckmann EW: A recognition of the interrelation-

6.

7.

8. 9.

10.

I I.

12. 13. 14.

15.

16.

17.

18.

19.

REFERENCES I. Duggin GG, Dale NE, Lyneham R C et al: Calcium balance in pregnancy. Lancet 2:926-927, 1974. 2. Walker AR: Letter: calcium balance in pregnancy. Lancet 1:107, 1975. 3. Chu J-Y, Margen S, Costa FM: Studies incalcium metabolism. 11: Effects of low calcium and variable protein intake o n human calcium metabolism. Am J Clin Nutr 28:1028-1035, 1975. 4. hfargen S, Chu J-Y, Kaufmann NA et al: Studies in calcium metabolism. I: The calciuretic effect of dietary protein. Am J Clin Nutr 27:584-589, 1974.

T H E JOURNAL O F PARENTERAL A N D ENTERAL XUTRITION

20.

21. 22. 23.

24.

ship of calcium with various dietary components. A m J Clin Nutr 27:603-609, 1974. Schwartz R, Woodcock NA. BlakelyJD et al: Metabolic responses of adolescent boys t o two levels of dietary magnesium and protein. 11: Effect of magnesium and protein level o n calcium balance. Am J Clin Nutr 26519-523, 1973. Briscoe AM, Ragan C Effects of magnesium o n calcium metabolism in man. Am J Clin Nutr 19~296-306,1966. Calcium, magnesium, and diuretics. Br M e d J 5951:170-171,1975. Food and Nutrition Board-National Research Council: Recommended dietary allowances. Washington, DC, National Academy of Sciences, 1974. Ewe K: Calcium absorption in health and disease. Ergeb Inn hled Kinderheilkd 33:23 1-269; 1972. Hegsted DM: Major minerals. I n Goodhart RS, Shils M E (eds): Modern Nutrition in Health and Disease. Philadelphia, Lea & Febiger, 1973. Nordin BEC, Smith DA: Diagnostic Procedures in Disorders of Calcium Metabolism. Boston, Little, Brown & Co. 1965. Harper HA: Review of Physiological Chemistry. Los Altos, California, Lange Medical Publications, 1975. Shils ME: Total parenteral nutrition. I n Goodhart RS, Shils M E (eds): Modern Nutrition in Health and Disease. Philadelphia, Lea & Febiger, 1973. Shils M E Minerals. I n White PL, Nagy ME, Fletcher DC: Total Parenteral Nutrition. Acton, Massachusettes, Publishing Sciences Group, Inc, 1974. Rudman D, Millikan WJ, Richardson TJ et al: Elemental balances during intravenous hyperalimentation of underweight adult subjects; J Clin Invest 55:94-104, 1975. Kessler G, Wolfman M: An automated procedure for the simultaneous determination of calcium and phosphorus. Clin Chem 10:686-703, 1964. Gitelman H: An improved automated procedure for the determination of calcium in biological specimens. Anal Biochem I8:521-53 I, 1967. Perkin-Elmer Corp: Analytical Methods of Atomic Absorption Spectrophotornetry. Norwalk, Connecticut, 1973. Lutwak L, Whedon GH. Lachance PA et al: Mineral, electrolyte and nitrogen balance studies of the Gemini-VII fourteen-day orbital space flight. J Clin Endocrinol Metab 29:1140-1156, 1969. Davidson S. Passmore R, Brock JF et al: Minerals, In Human Nutrition and Dietetics. Edinburgh, Churchill Livingstone, 1975. Lentner C, Lauffenburger T, Guncaga J et a 1 The metabolic balance technique. Metabolism 24:461-471, 1975. Shils h E : Guidelines for total parenteral nutrition. J Am Med ASOC 220:1721-1729, 1972. Goodale RH, Widmann FK: Clinical Interpretation of Laboratory Tests. Philadelphia, FA Davis Co, 1969.

I55