Effects of Calcium Infusion on Blood Pressure inHypertensive and Normotensive Humans

DAVID H. ELLISON, ROBERT SHNEIDMAN, CYNTHIA MORRIS, AND DAVID A. MCCARRON

SUMMARY Disorders of calcium and parathyroid hormone homeostasis have been reported insubjects with essential hypertension. In many of these studies, dietary intakes of sodium and calciumwere not carefully controlled. The present study was designed to compare calcium and parathyroidhormone homeostasis in normal and hypertensive subjects on controlled dietary sodium and calciumintakes and to examine the impact of dietary sodium loading on hemodynamic and metabolic re-sponses to infused calcium. Seven subjects with essential hypertension and seven age-matched and sex-matched controls were studied while consuming a standard diet containing 600 mg of elementalcalcium. Each subject was studied while consuming 10,160, and 510 mEq of sodium per day, before,during, and after a 3-hour calcium infusion (3.75 mg/kg/hr). Before calcium infusion, hypertensivesubjects had increased urinary cyclic adenosine 3,5-monophosphate excretion independent of sodi-um intake (p<0.05). Urinary potassium excretion was greater in normotensive than in hypertensivesubjects (p = 0.002). At baseline, dietary sodium intake had no effect on systolic, diastolic, or meanarterial pressure. During calcium infusion, systolic pressure increased in both groups, whereasdiastolic pressure increased only when dietary sodium content was high and mean arterial pressureincreased only in hypertensive subjects (p = 0.007). Together, these data provide evidence for interac-tions between dietary sodium intake and the cardiovascular response to calcium. They confirm thathypertensive subjects exhibit enhanced parathyroid gland function even when dietary factors arecontrolled, and they suggest that these subjects are more sensitive to the cardiovascular effects ofshort-term calcium infusion. (Hypertension 8: 497-505, 1986)

KEY WORDS • parathyroid hormone * blood pressure • cyclic adenosine monophosphate

CALCIUM metabolism may be abnormal in pa-tients with essential hypertension and in sev-eral experimental models of the disease. Dis-

turbances that have been described in humans includeincreased urinary calcium excretion,1-2 enhanced para-thyroid gland activity,1 2 and reduced levels of serumionized calcium.3•* In several experimental models ofhypertension, similar abnormalities have been identi-fied,5 6 including increased rates of parathyroid hor-

From the Division of Nephrology and Hypertension, OregonHealth Sciences University, Portland, Oregon. Dr. Ellison was therecipient of Tarter Trust and Oregon Heart Association Fellow-ships. Dr. Ellison's current address is Division of Nephrology,Department of Medicine, Yale University School of Medicine, 333Cedar Street, New Haven, CT 06510.

Portions of this research were presented in abstract form at theAmerican Society of Nephrology Meeting, December 1982 (KidneyInt I983;23:169).

Address for reprints: David H. Ellison, M.D., Division of Ne-phrology, Yale University School of Medicine, 2074 LMP, 333Cedar Street, New Haven, CT 06510.

Received April 26, 1985; accepted December 9, 1985.

mone (PTH) secretion5 6 and alterations of membranecalcium sensitivity,7 perhaps secondary to disorderedmembrane calcium binding.8

Most clinical reports of disordered calcium homeo-stasis have neglected careful control of several impor-tant determinants of calcium metabolism. Dietary in-take of calcium or sodium may influence renalexcretion of the other ion,6 9 and each may participatein blood pressure control, either directly or throughtheir diverse interactions.9"" Recent reports highlightthe major effects of changes in dietary sodium chlorideintake on renal calcium excretion9- " and suggest thatsome sodium chloride-dependent effects may be me-diated directly by changes in systemic calcium bal-ance. The present experiments were designed to com-pare the hemodynamic and metabolic responses toshort-term calcium infusion in normal subjects and insubjects with essential hypertension. Calcium infusionwas performed while subjects consumed diets that dif-fered in sodium content to investigate interactionsbetween dietary sodium intake and the response tocalcium.

497

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

498 HYPERTENSION VOL 8, No 6, JUNE 1986

Subjects and MethodsSeven subjects with essential hypertension and sev-

en age-matched and sex-matched control subjects par-ticipated in the study. All subjects were white. Criteriafor entry of hypertensive subjects included document-ed ambulatory blood pressures in excess of 140/90 mmHg on at least three occasions and the absence of otherchronic diseases requiring medical therapy. Hyperten-sive subjects had previously undergone specific inves-tigations designed to rule out secondary causes of hy-pertension, and each carried the diagnosis of essentialhypertension. At the time of entry into the study, 1)routine urinalysis results showed less than 1 + protein-uria and were negative for occult blood (Multistix,Ames, Elkhart, IN, USA) and 2) automated chemistryscreen results (SMAC, Technicon, Tarrytown, NY,USA) showed normal serum potassium, creatinine,total calcium, and phosphorus levels. Table 1 com-pares characteristics of the two groups. Although hy-pertensive subjects tended to weigh more than controlsubjects, this difference was not significant ( /= 1.16by unpaired t test).

Most hypertensive subjects were taking oral antihy-pertensiye medications before entry into the study.Medications were withdrawn approximately 2 weeksbefore the study began. Although at least 10 drug-freedays were required before beginning the study, thepossibility that some of the observed differences weredue to residual drug effects cannot be excluded. Ageneral outline of the experimental protocol is shownin Figure 1. The study was divided into three parts andlasted 13 days. On Day 1, the subject received dietaryinstruction and was given a dietary log book in whichto record daily food consumption. The daily diet con-tained 10 mEq of sodium and approximately 600 mg ofelemental calcium, a portion of which was adminis-tered as calcium carbonate. The subject consumed thisdiet throughout the study. A 24-hour urine sample wascollected on the fifth day and was analyzed for sodium,calcium, and creatinine as a check on dietary adher-ence. On the fifth day, the subject entered the ClinicalResearch Center for Test Day 1. After the test dayprotocol (described below) had been completed, anadditional 150 mEq of sodium per day (in a pre-weighed package) was added to the initial diet. Thesubject then ingested the same diet with the additional

TABLE 1. Subject Characteristics at Time of Entry into Study

Characteristic

Sex (M/F)

Age (yr)

Blood pressure (mm Hg)

Systolic

Diastolic

Weight (kg)

Creatinine (mg/dl)

Hypertensive

4:3

42±4.5

156±2.8

1OO±3.5

80±6.6

1.0 ±0.05

Normotensive

4:3

42±4.5

118 + 4.2*

72±2.8*

70±4.8

0.9±0.07

Values are means ± SE.*p < 0.001, compared with values in hypertensive subjects.

lira 0835 0915

FIGURE 1. Experimental protocol. Each subject consumedthe same diet throughout the 13 study days. Sodium was sup-plemented in preweighed packages. On the test days, inulinandp-aminohippurate (PAH) clearance measurements were begunat 0915. CRC = Clinical Research Center.

sodium for the next 4 days. A 24-hour urine samplewas collected on the third day (Study Day 9) and ana-lyzed for sodium, creatinine, and calcium. The subjectentered the Clinical Research Center on Study Day 10for Test Day 2. Following completion of the test dayprotocol, dietary sodium was increased to 510 mEq/day. On the twelfth day, a 24-hour urine sample wascollected and analyzed for sodium, creatinine, and cal-cium. On Day 13, the subject entered the ClinicalResearch Center for the last of 3 test days.

The protocol during each of the 3 test days wasidentical. Two intravenous catheters were placed, onefor infusions and one for drawing blood. An oral waterload (750 ml) was administered to ensure adequateurine flow, and loading doses of p-aminohippurate(PAH), 6 mg/kg, and inulin, 50 mg/kg, were givenintravenously during a 5-minute period. Following thisloading dose, dextrose in water, 50 g/L, with PAH,2 g/L, and inulin, 5 g/L, were infused at 250 ml/hour.After 45 minutes had elapsed to allow the attainment ofsteady state conditions, clearance measurements werebegun. Urine samples were collected every 20 min-utes, blood samples were obtained at the midpoint ofeach 20-minute cycle. This procedure provided a con-tinuous series of 20-minute clearance periods. Bloodpressure, pulse, and respiratory rate were measuredevery 20 minutes during the infusions. Two hours afterclearance measurements had begun, an infusion of ele-mental calcium as gluconate was initiated at a rate of3.75 mg/kg/hr and continued for 3 hours. At the end ofthis period, calcium was removed and inulin and PAHinfusion continued for 2 more hours. Thus, clearanceswere measured continuously for 7 hours.

Blood pressures were measured by standard sphyg-momanometry. Systolic pressure was defined as the

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

CALCIUM INFUSION IN HYPERTENSIVE HVMANS/Ellison et al. 499

point at which the Korotkoff sounds first appeared,diastolic pressure was defined as the point at whichthese sounds disappeared. Serum samples were ana-lyzed for sodium, potassium (Autoanalyzer), ionizedcalcium (Clin ± Ion, Menlo Park, CA, USA), inulin,12

and PAH (by photometry at 540 nm following cou-pling with diazotized JV-[l-napthyl]ethylenediaminedihydrochloride). Urine was analyzed for sodium, po-tassium, calcium, PO4 (Autoanalyzer), cyclic adeno-sine 3',5'-monophosphate (cAMP; New England Nu-clear, Boston, MA, USA), and inulin and PAH (asabove). Clearances of inulin and PAH were calculatedas [U], x V7[P]X, where [U]x is the concentration ofinulin or PAH in the urine, V is the volume flow rate,and [P], is the concentration of inulin or PAH in theplasma measured at the midpoint of the urine collec-tion. Fractional ion excretion was calculated as

100[U],,

where [U], is the urinary ion concentration, [P], is theplasma ion concentration, and [P]IN and [U],N are theinulin concentrations in plasma and urine respectively.

Baseline data were analyzed using a two-way analy-sis of variance (ANOVA) and unpaired t test. Statisti-cal analysis of the calcium infusion data employed amultivariate analysis of variance and covariance withrepeated-measure design (BMDP and SAS programs).With the use of this approach, variances could be at-tributed to group effects" (hypertensive vs normal sub-jects), time effects (effects of calcium infusion), anddiet effects (effects of dietary sodium intake) as wellas to interactions between these variables. To assesschanges in blood pressure, baseline measurements (be-fore calcium infusion) were taken as covariates. Statis-tical significance was accepted at the/? less than 0.05level.

This protocol was approved by the Human Investi-gations Committee of the Oregon Health Sciences Uni-versity. Each subject gave voluntary informed consentbefore participation.

ResultsBaseline data obtained before entry into the study

are shown in Table 1. The normotensive and hyperten-sive subjects were well matched for age, sex, baselinecreatinine levels, and weight. Subjects were judged tohave adhered to the dietary regimens by two criteria.First, food diaries indicated that each subject remainedwithin the dietary guidelines. Second, baseline 24-hour urinary sodium excretion increased when dietarysodium intake was raised.

Blood pressures are shown in Table 2. As expected,before calcium infusion, systolic and diastolic pressureand mean arterial pressure (MAP) were higher in thehypertensive subjects than in the normotensive controlgroup. During the study, dietary sodium intake did notalter baseline blood pressure significantly in eithergroup (diet effect, p = 0.827 by ANOVA; see Table2). Systolic pressure rose significantly in response tocalcium infusion in both groups (time effect, p< 0.001by ANOVA; Figure 2). Dietary sodium intake alteredthe relation between calcium infusion and systolicpressure in both groups (diet/time effects, p< 0.05 byANOVA); however, dietary sodium affected the pres-sor response to calcium in hypertensive subjects differ-ently than in normotensive subjects (diet/time/groupeffect, p<0.02 by ANOVA). Specifically, normoten-sive subjects increased systolic pressure in response tocalcium infusion only during the higher sodium dietperiods (160 and 510 mEq/day), whereas hypertensivesubjects were sensitive to the cardiovascular effects ofcalcium regardless of sodium intake. There was nocorrelation between changes in serum ionized calciumand systolic pressure in either group (Figure 3).

Diastolic pressure responses are shown in Table 2and Figure 4. Although calcium infusion did not alterdiastolic pressure significantly when all dietary condi-tions were considered together (time effect, p = 0.212by ANOVA), calcium infusion led to increases in dia-stolic pressure during certain dietary sodium periods.Calcium infusion did not alter diastolic pressure ineither group during low sodium intake, but diastolicpressure did increase (diet/time effect, p = 0.037 byANOVA) when dietary sodium was greater. The inter-

TABLE 2. Effects of Dietary Sodium Content on Systolic, Diastolic, and Mean Arterial Pressure Response to Infused Calcium

Variable

Systolic BP

Normotensive

Hypertensive

Diastolic BP

Normotensive

Hypertensive

MAP

Normotensive

Hypertensive

Low

B

117±5.9

141±3.1

73±5.1

91±1.5

88±4.2

108 ±1.4

Na+ (10

D

114 + 3

155±4

69±193±2

84±2

I14±2

mEq/day)

.3

.0

.6

.4

.0

.7

A

114±6.1

I44±5.7

69 ±3.1

91±4.3

84±2.8

109±5.4

Medium

B

113±5.O

139±5.4

69±2.8

87±2.8

83±2.8

104±3.5

Na+ (260

D

123±4.0

149±3.2

70 ±2.0

92±2.2

87 ±2.4

111+2.7

mEq/day)

A

120±6

150±6

71 ±3

95 ±5

87 ±3

114±5

.8

.6

.7

.6

.1

.7

High

B

111+4.5

I45±6.6

69 ±4.4

88 ±5.1

83±3.2

107±5.0

Na+ (510 mEq/day)

D

1I8±3.7

154±5.2

71 ±2.1

94±4.0

87±2.5113 ±4.7

A

I22±5.8

158±9.3

71 ±4.3

97 ±7.0

88±3.3

118±7.4

VaJues are means ± SE before (B), during (D), and after (A) calcium infusion. BP = blood pressure; MAP = mean arterial pressure.

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

500 HYPERTENSION VOL 8, No 6, JUNE 1986

FIGURE 2. Effects of calcium infusion on sys-tolic blood pressure. Measurements weremade immediately before, during (at 1, 2, and3 hours), and I hour after calcium infusion.Statistics are shown in Table 2.

165

155

145

Syitsllc PTMSUT*. 135mm HQ

125

115

105

95

•••

o

•

•o-

• * •

- l l r

Nui MMI~ LOW Sodium

HyperUralvt- Low Sodium

Norm»l- rtadium Sodium

HyptrUretvi tttdlum Sodium

Normt)- High Sodium

HyperUnsfvt High Sodium

B*Tort During During Dirlng AfUrMutton

ASytWk

so

20

• 10

•

••V

i, mm. -02 0.0-10

-20

-30

FIGURE 3. Comparison of changes in systolic pressure andserum ionized calcium concentration.

action between diet and calcium infusion could not beshown to be different between groups (diet/time/groupeffect, p = 0.935 by ANOVA).

The effects of calcium infusion on MAP are shownin Table 2 and Figure 5. During the infusion of calciumin hypertensive subjects, MAP rose significantly andindependently of dietary sodium intake, whereas innormal control subjects calcium infusion had no sig-nificant effect on MAP (group effect, p = 0.007 by

ANOVA). Of further note is the temporal dissociationbetween blood pressure and serum calcium concentra-tion. In the hypertensive subjects, MAP rose duringcalcium infusion as serum calcium levels increased butMAP remained elevated as serum calcium levels beganto fall during the recovery period.

Inulin and PAH clearances and fractional excretionrates for sodium and potassium are shown in Table 3.Note that all fractional excretion rates were measuredafter the start of diuresis. Inulin clearance (overallmean, 105 ± 7.7 vs 105 ± 7.4 ml/min) and PAH clear-ance (overall mean, 544 ± 45.6 vs473 ± 35.6 ml/min)were similar in normotensive and hypertensive sub-jects, respectively, and were not altered significantlyby sodium intake. Although, as expected, fractionalsodium excretion was influenced significantly by di-etary sodium intake (diet effect, p<0.001), overallmean fractional sodium excretion was similar in nor-motensive and hypertensive subjects (2 ± 0.4 vs2 ±0 .6%, respectively; group effect, p = 0.618).Fractional potassium excretion, however, was signifi-cantly higher in normal than in hypertensive subjects(overall mean before calcium infusion, 32 ± 0.4 vs21 ± 0 . 3 % , respectively; group effect, p = 0.002).Neither inulin nor PAH clearance was affected signifi-cantly by calcium infusion in either group. Increasingdietary sodium intake tended to increase PAH clear-

FIGURE 4. Effects of calcium infusion ondiastolic pressure. Measurements were madeimmediately before, during (at I, 2, and 3hours), and I hour after calcium infusion. Sta-tistics are shown in Table 2.

DiulolJcPr§uurt, tn

Ho,

105

95

73

65

55fofort During During Duing Atltr

infwion

•• Norm*!- Low Sodium

°- Hypertensive- LowSodtum

• - Nonrxl- rhdtum Sodium

°- Hyptrtanstvt MtdlumSodium

A- Normil- High Sodium

* - Hypert«njivt HighSodium

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

CALCIUM INFUSION IN HYPERTENSIVE HVMANS/Ellison et al. 501

112

rtan ArtirttlPrmurtbtm '02

92

02B«fort Duing CXrino, During AJUr

C«lclum Infusion

All Ot»U

All Di«t3

FIGURE 5. Effects of calcium infusion onmean arterial pressure (calculated as diastolicpressure + I'13(systolic —diastolic pressure).Measurements were made immediately before,during (at I, 2, and 3 hours), and I hourafter calcium infusion. Statistics are shown inTable 2.

ance in normal but not in hypertensive subjects, al-though this difference did not reach statistical sig-nificance (diet/group effect, p = 0.064). Fractionalsodium excretion was increased by calcium infusion(time effectp<0.001 by ANOVA) and by increaseddietary sodium intake (diet effect, /?<0.001 byANOVA). Calcium infusion had ho effect on fraction-al excretion of potassium (p = 0.925 by ANOVA),although' fractional potassium excretion remainedhigher in riormotensive than in hypertensive subjects(group effect, p = 0.023 by ANOVA).

Serurn ionized calcium, fractional calcium excre-tion, and urinary cAMP excretion are shown in Table4. Serum ionized calcium concentration was similar innormotensive and hypertensive subjects (overall meanbefore calcium infusion, 2.1 ±0.38 vs 2.1 ±0.34mEq/dl, respectively; group effect, p = 0.661). Frac-tional calcium excretion tended to be elevated in hy-pertensive subjects (6 ± 1.1 vs 4 ± 0.7 in normoten-sive subjects), but this effect was not significant.Urinary cAMP excretion at baseline (Figure 6) washigher in hypertensive than in normotensive subjects

(5.2 ± 0.52 vs 3.0 ± 0.64 nmol/dl of glomerular filtra-tion rate; p<0.05). Calcium infusion in both groupssignificantly increased serum calcium levels frombaseline regardless of the sodium content of the diet(time effect, p<0.001 by ANOVA). Fractional excre-tion of calcium also increased significantly during theinfusion (time effect, p<0.001 by ANOVA). Con-versely, calcium infusion significantly reduced urinarycAMP excretion (time effect, p = 0.003 by ANOVA)in both groups.

DiscussionThe present study was designed to compare hemo-

dynamic and metabolic responses to short-term cal-cium infusion in hypertensive and normotensive sub-jects and to assess the influence of dietary sodiumintake on the response to administered calcium. Theresults show that subjects with essential hypertensionexcreted more urinary cAMP than did normal subjectseven when dietary sodium and calcium intake werecontrolled and that dietary sodium chloride intake al-

TABLE 3. Effects of Calcium Infusion on Fractional Sodium and Potassium Excretion and lnulin and p-Aminohippurate Clearance

Variable

Normotensive

Hypertensive

Normotensive

Hypertensive

Normotensive

Hypertensive

CPAH

Normotensive

Hypertensive

Low

B

1 ±0.21 ±0.2

37±6.224±6.2

100±5.196±9.3

445 ±53.7

479 + 59.4

Na+ (10 mEq/day)

D

2 ±0.5

2 ±0.4

31 ±7.8

16 + 2.6

88±2.9

I13±16.1

494 + 34.2

614± 125.4

A

1 ±0.2

1 ±0.4

35 ±4.9

18±3.O

88 + 7.3

103 ±16.3

481+42.7

479 ±68.0

Medium

B

2±0.6

2 ±0.9

26±6.1

16±5.7

98 ± 11.0

111 + 16.3

6I7± 105.0

514 ±70.7

Na+ (260 mEq/day)

D .

6±0.95±1.1

32±6.319±2.9

96±9.3

92± 11.9

596 ±72.6

431 ±75.6'

A

5±1.7

3±0.8

34±5.4

25 + 3.7

86±11.5

93 ±9.6

502 ±42.4

425 ±68.5

High

B

• 3±0.7

5± 1.2

32 ±7.4

22±3.6

101 ±10.4

93±12.6

571 ±65.1

426 ±59.0

Na+ (510 mEq/day)

D

6± 1.6

7±2.1

40 + 9.9

18±5.6

95 ±18.0

98±15.l

625 + 98.1

483 ±100.0

A

4±0.9

3 + 0.9

32±9.4

18±3.1

85 ±12.7

108± 14.7

492 ±88.8

487±71.0

Values are means ± SE before (B), during (D), and after (A) calcium infusion. FEN, = fractional excretion of sodium; FEK = fractionalexcretion of potassium; Q N = inulin clearance; CPAH = p-aminohippurate clearance.

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

502 HYPERTENSION VOL 8, No 6, JUNE 1986

TABLE 4. Effects of Calcium Infusion on Serum Ionized Calcium Concentration, Fractional Calcium Excretion, and Urinary cAMPExcretion

Variable

Serum Ca2+ (mEq/L)

Normotensive

Hypertensive

Normotensive

Hypertensive

UcAMP (nmol/dl GFR)

Normotensive

Hypertensive

2

2

3

5

Low

B

.1 ±0.14

. l ± 0 . 0 6

4 + 0.9

4±1.4

.0±0.64

.2±0.52

Na+ (10 mEq/day)

D

2.7±0.13

2.8±0.1l

17±3.9

17±2.9

2.1 ±0.75

2.5 ±0.70

A

2.5 ±0.08

2.6±0.06

20±7.8

13±1.9

—

—

2

2

3

4

Medium

B

.2±0.

. l±0 .

5±1.

7±2.

.6±l .

.7+1.

.04

.05

.6

.3

31

.25

Na+ (260

D

2.7±O.1O

2.8 + 0.07

25±6.019 + 64

1.7 + 0.81

2 4 ± I . 1 4

mEq/day)

A

2.5±O.1O

2.5 + 0.08

23±4.6

19±3.7

—

—

2,

High

B

I ±0.05

2.1 ±0.07

2.

4

3±0.8

5±2.l

3±0.98

.2 ±1.02

Na+ (510 mEq/day)

D

2.8±O.IO

2.8±0.09

26±5 5

28 ±8.0

l.7±0.46

2.2±0.97

A

2.5 ±0 07

2.6±O.O7

24 ±8.2

I6±3.O

—

—

Values are means ± SE before (B), during (D), and after (A) calcium infusion.cAMP excretion; GFR = glomerular filtration rate.

= fractional excretion of calcium; UcAMP = urinary

cArt> tar»Oon.nmoi/IOOmieFR

6

5

4

5

2

I

0

ItTT

Blood Pressure

FIGURE 6. Urinary excretion rate of cAMP in normal andhypertensive subjects before calcium infusion (p<0.05).GFR = glomerular filtration rate.

tered the cardiovascular response to infused calcium.Hypertensive subjects were also more sensitive to theacute effects of calcium on blood pressure than werenormal subjects. Together, these and other abnormali-ties of sodium and calcium handling may contribute tothe pathogenesis of essential hypertension.

Several clear differences in calcium and PTH ho-meostasis between normal and hypertensive subjectswere evident before calcium infusion. Hypertensivesubjects excreted more urinary cAMP per deciliter ofglomerular filtration rate at baseline (see Table 4).Urinary cAMP excretion reflects the level of parathy-roid activity13; the rapid suppression of cAMP excre-tion observed during calcium infusion supports thevalue of this measure in the present study. Enhancedparathyroid activity has been described in patients withessential hypertension1 2 M and in rats with spontane-ous hypertension.3 6 Although the cause is unknown, itmay be related to increased urinary calcium excretionin this group. We1 and others2 have observed increasedrates of urinary calcium excretion in patients with es-sential hypertension under nonstimulated conditions.Hypertensive subjects tended to excrete more calcium

before infusion than did controls in the present study,but this did not reach statistical significance.

Whereas some larger studies have shown lower lev-els of serum ionized calcium in subjects with essentialhypertension compared with normotensive controls,the present results showed no clear difference. A possi-ble cause for this lack of significant difference is ouruse of controlled dietary sodium and calcium intake.The controlled dietary conditions, although of relative-ly brief duration, may have minimized differences inserum ionized calcium, since hypertensive subjectshave been reported to consume less dietary calciumthan do normal subjects15 on ad libitum dietary intake.It also seems likely, however, that sample size limitedour ability to demonstrate previously described abnor-malities. Whether longer term control of dietary sodi-um and calcium intake would have attenuated or abol-ished the observed differences in urinary cAMPexcretion cannot be addressed by the current data.

As expected, calcium infusion markedly increasedserum ionized calcium concentration and urinary cal-cium excretion and decreased urinary cAMP excre-tion. Responses in both groups were quite similar inthe magnitude of the increases in serum calcium and inurinary responses. Strazzullo et al.2 reported that uri-nary calcium excretion remained higher in patientswith essential hypertension than in normal subjectsduring short-term calcium infusion. In that study, hy-pertensive subjects excreted more calcium than didnormal subjects at every level of serum calcium. In thepresent experiments, we did not find that hypertensivesubjects excreted more calcium during the infusion.The reason for the differences between studies is notclear. Fractional sodium excretion also increased dur-ing the calcium infusion in the present study. Whenadministered acutely, calcium is natriuretic,16 whichmight cause marked volume depletion during a 3-hourinfusion.17 We could not, however, find significantchanges in either inulin or PAH clearance during orafter the calcium infusion period.

Increased dietary sodium chloride intake expandedextracellular fluid volume in both normal and hyper-

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

CALCIUM INFUSION IN HYPERTENSIVE HUMANS/Ellison et al. 503

tensive subjects, as indicated by increased urinary so-dium excretion. There was no significant effect of di-etary sodium chloride on blood pressure beforecalcium infusion in either group. Some patients withessential hypertension are especially sensitive to theeffects of dietary salt.18 Many others, and most sub-jects with normal blood pressure, demonstrate smallincreases in systolic and diastolic pressure when theyingest extremely large quantities of sodium (as muchas twice the maximal sodium intake employed in thepresent study)." The minimal effects of dietary sodiumobserved in the present study are compatible with theminimal effects of dietary sodium loading previouslydescribed,19 although a longer duration of reduced orelevated sodium consumption might have affectedblood pressure significantly. Increasing dietary sodi-um intake, when consumed as sodium chloride, caninduce a state of negative calcium balance by increas-ing urinary calcium excretion.9

Some hemodynamic changes often directly attribut-ed to altered sodium chloride intake may be mediatedby alterations in systemic calcium homeostasis." Thefact that there were no significant differences in sodi-um excretion between normal and hypertensive sub-jects before calcium infusion reflects similar adher-ence to prescribed dietary regimens. Hypertensivesubjects, however, excreted less potassium than didnormal controls, independent of the level of sodiumintake. Dietary potassium intake has been reported tobe deficient in hypertensive patients,20 and its supple-mentation may attenuate increases in blood pressureduring dietary sodium loading.19 Because dietary po-tassium intake was not rigidly controlled in the presentstudy, the reduced fractional potassium excretion mayreflect differences in potassium intake20 or intrinsicdefects in renal potassium handling. Any alteration inpotassium balance in hypertensive subjects may havecontributed to the observed differences in blood pres-sure response to infused calcium.

Systolic and diastolic pressure increased in both nor-motensive and hypertensive subjects in response toinfused calcium under certain dietary conditions.Short-term calcium infusion increases blood pressureprimarily because it increases systemic vascular resis-tance21-22 in both normotensive and hypertensive sub-jects. Cardiac output may increase during the first sev-eral minutes of calcium infusion,23 but it returns tobaseline within 30 minutes of exposure, and the in-creased pressure is maintained by increased vasculartone.21-n There is strong evidence that direct vasculareffects of calcium play important roles in the increasedvascular resistance. Resistance of isolated perfusedvascular beds varies directly, in most species, with thecalcium concentration of the perfusate near physio-logical levels.24^27 In humans, local calcium infusions,designed to prevent changes in systemic calcium con-centration and the resultant changes in circulating hor-mone levels and central hemodynamics, were shownto increase vascular resistance.28 In rats, however,vasodilation occurs when ambient calcium concentra-tions are raised.29 Increasing calcium concentrations

near physiological levels in vitro increases vasculartone, while higher concentrations lead to vasorelaxa-tion.30 Together, these studies suggest a role for directlocal effects of calcium on vascular smooth muscle inthe development of blood pressure changes during cal-cium infusion.24"27

Weidmann and co-workers17 21-22 have studied pos-sible hormonal consequences of calcium infusion andtheir contribution to the hemodynamic responses. Cal-cium infusion significantly increased peripheral levelsof norepinephrine but had no effect on renin, aldoste-rone, or dopamine levels. They suggested that en-hanced catecholamine release may participate in thepressor response to infused calcium. Another possiblecontributor to blood pressure homeostasis during cal-cium infusion is PTH. Short-term infusion of PTHreduces systemic pressure.31 32 In subjects with secon-dary hyperparathyroidism, elevated PTH levels mayreduce pressure homeostatically, since the increase insystemic pressure during short-term calcium infusioncorrelates best with the fall in serum PTH.33 In moststudies, the pressor response to infused calcium is notclearly correlated with the increase of serum calcium.This variability of the hemodynamic response suggeststhat changes in blood pressure during exposure to cal-cium may be multifactorial in origin and not simplydirect effects of calcium itself.

The present results suggest that hypertensive sub-jects are more sensitive than normal subjects to theacute vascular effects of calcium, at least under certaindietary conditions. Bianchetti et al.22 found no differ-ence in the slope of blood pressure versus serum cal-cium level during calcium infusion in normal subjectsand a small number of hypertensive subjects; however,blood pressure rose in response to a low dose calciuminfusion (2 mg/kg/hr) in hypertensive, but not in nor-motensive, subjects. Vascular reactivity to infusedpressors is enhanced nonspecifically in hypertensivesubjects, possibly because of structural changes in ves-sel walls.7 Overbeck et al.28 found that local infusionsof calcium produced similar increases in vascular re-sistance in both normal subjects and those with essen-tial hypertension when differences in baseline vasculartone were considered. Since systemic and humoraleffects were prevented by infusing calcium locally,28

vascular effects were assumed to be direct. If the directeffects of calcium in normal and hypertensive subjectsare similar, then the increased response of hyperten-sive subjects observed in the present study and sug-gested by others22 may reflect reduced compensatorymechanisms, altered secondary responses, or differingresponses to experimental stress. Weidmann et al.17

found that subjects with mild chronic renal insufficien-cy, some of whom were hypertensive, were more sen-sitive to the hemodynamic effects of calcium than werenormal subjects. Mori23 reported that total peripheralresistance decreased during very short-term calciuminfusion in normotensive subjects, whereas resistancedid not change in subjects with essential hypertension.Although mean arterial pressure and cardiac index in-creased in both groups during the infusion described by

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

504 HYPERTENSION VOL 8, No 6, JUNE 1986

Mori,23 these data suggest that vasodilation early dur-ing calcium infusion, either as a compensatory re-sponse or as a direct one, is greater in normotensivethan in hypertensive subjects. As discussed previous-ly, PTH has been shown to be vasodilative when ad-ministered acutely in vivo and in vitro.31-32 If baselinePTH concentrations are higher in hypertensive sub-jects, then acute suppression of PTH release duringcalcium infusion might reduce the vasodilative effectsof this hormone to a greater extent in hypertensive thanin normal subjects.33

The present results indicate an interaction betweensodium intake and the effects of calcium on bloodpressure. Diastolic pressure did not increase duringcalcium infusion when subjects consumed little dietarysodium, whereas it rose during calcium infusion whensodium intake was greater. In normotensive subjects, arise in systolic pressure was also dependent on dietarysodium intake. When dietary sodium intake was low,systolic pressure actually fell during the calcium infu-sion. On the other hand, systolic pressure rose in hy-pertensive subjects when calcium was infused regard-less of dietary sodium intake. Thus, hypertensive andnormal subjects exhibited frankly different responsesto calcium infusion only when dietary sodium wasreduced. Under conditions of more typical dietary in-take, no qualitative differences were found. These dif-ferences suggest disordered interactions between sodi-um and calcium metabolism in hypertensive subjects.

Increased membrane calcium permeability has beendescribed in tissues from hypertensive animals.7 Sodi-um loading may potentiate the short-term effects ofcalcium in normal subjects, perhaps by inducing a stateof negative calcium balance." In hypertensive sub-jects, however, because of preexisting disorders ofcalcium and PTH homeostasis,'2-34 vascular effectsmay occur without volume expansion. Rats with spon-taneous hypertension exhibit impaired vasodilative re-sponses to infused calcium compared with normoten-sive controls.29 In human hypertension as well, thebalance between vasoconstrictive and vasodilative in-fluences of calcium may be altered.

The effects of acute infusion of calcium on bloodpressure, such as those observed in the current study,are quite different from more prolonged exposure tocalcium. Dietary calcium loading attenuates the devel-opment of hypertension in the spontaneously hyperten-sive rat as well as its normotensive control, the Wistar-Kyoto rat.5 35 Chronic dietary calcium loading alsoreduces blood pressure in hypertensive humans.9-M

Chronic elevations of ambient calcium may act toblock slow calcium channels and decrease calciumpermeability of vascular smooth muscle membranes.36

In fact, such down-regulation of calcium responsive-ness by calcium itself may be consistent with the pres-ent results. If hypertensive subjects exhibit chronicrelative calcium deficiency, as evidenced by dietarysurveys,15 reduced serum calcium levels,3 4 and elevat-ed urinary cAMP excretion,'-2 then a sudden increasein serum level consequent to calcium infusion mightstimulate greater vasoconstriction. According to such

a scheme dietary calcium loading should blunt the ef-fects of infused calcium on blood pressure in this groupof subjects.

In summary, we have shown that abnormalities ofcalcium and PTH homeostasis are present in subjectswith essential hypertension even when carefullymatched with a control population and studied whileingesting similar diets. These abnormalities includeevidence of parathyroid gland stimulation and a trendtoward increased urinary calcium excretion. We havealso shown that calcium infusion causes a greater risein blood pressure in subjects with essential hyperten-sion than in normal subjects when dietary sodium in-take is low. These findings are compatible with thepresence of a state of relative calcium deficiency insubjects with essential hypertension, a state that maycontribute to the enhanced calcium sensitivity demon-strated in the present study.

AcknowledgmentsThe authors acknowledge the staff of the Clinical Research Cen-

ter, the statistical assistance given by Dr. John Goffinet, the techni-cal assistance of Sandford Plant, Janet Dorow, and Sally Morton,and the dietary assistance of Holly Henry. Dr. Frank Seney pro-vided constructive criticism of the manuscript.

References1. McCarron DA, Pingree PA, Rubin RJ, Gaucher SM, Molitch

N, Krutzik S. Enhanced parathyroid function in essential hy-pertension: a homeostatic response to a urinary calcium leak.Hypertension 1980;2:162-168

2. Strazzullo P, Nunziata V, Cirillo M, et aJ. Abnormalities ofcalcium metabolism in essential hypertension. Clin Sci 1983;65:137-141

3. McCarron DA. Low serum concentrations of ionized calciumin patients with hypertension. N Engl J Med 1982;307:226-228

4. Resnik LM, Laragh JH, Sealey JH, Alderman MH. Divalentcations in essential hypertension: relations between serum ion-ized calcium, magnesium, and plasma renin activity. N Engl JMed 1983;309:888-891

5. McCarron DA, Yung NN, Ugoretz BA, Krutzik S. Distur-bances of calcium metabolism in the spontaneously hyperten-sive rat. Hypertension 1981;3(suppl I): I-162-1-167

6. Stem N, Lee DBN, Silis V, et al. Effects of high calciumintake on blood pressure and calcium metabolism in youngSHR. Hypertension 1984;6:639-646

7. Webb RC, Bohr DF. Recent advances in the pathogenesis ofhypertension: consideration of structural, functional and meta-bolic vascular abnormalities resulting in elevated vascular re-sistance. Am Heart J 1981;102:251-264

8. Postnov YV, Orlov SN, Pokudin NL. Decrease of calciumbinding by the red blood cell membrane in spontaneously hy-pertensive rats and in essential hypertension. Pflugers ArchI979;379:I91-I95

9. Castenmiller JJM, Mensink RP, van der Heijden L, et al. Theeffect of dietary sodium on urinary calcium and potassiumexcretion in normotensive men with different calcium intakes.Am J Clin Nutr 1985;41:52-60

10. McCarron DA. Calcium and magnesium in human hyperten-sion. Ann Intern Med 1983;98(Pt 2):8OO-805

11. Kurtz TW, Morris RC. Dietary NaCI as a determinant of disor-dered calcium metabolism in the Dahl salt-sensitive rat [Ab-stract]. Kidney Int 1985;27:194

12. Roe JH, Epstein JH, Goldstein NP. A photometric method forthe determination of inulin in plasma and urine. J Biol Chem1949;178:839-845

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

CALCIUM INFUSION IN HYPERTENSIVE HVMANS/Ellison et al. 505

13. Broadus AE, Mahaffey JE, Bartter FC, Neer RM. Nephroge-nous cyclic adenosine monophosphate as a parathyroid func-tion test. J Clin Invest 1977;6O:771-783

14. Orillo M, Strazzullo P, Galletti F, Siani A, Nunziata V. Theeffect of an intravenous calcium load on serum total and ion-ized calcium in normotensive and hypertensive subjects. J ClinHypertens 1985; 1:30-34

15. McCarron DA, Morris CD, Henry HJ, Stanton JL. Blood pres-sure and nutrient intake in the United States. Science 1984;224:1392-1398

16. Suki WN, Eknoyan G, Rector FC Jr, Seldin DW. The renaldiluting and concentrating mechanism in hypercalcemia.Nephron 1969;6:50-61

17. Weidmann P, Massry SG, Cobum JW, Maxwell MH, AtlesonJ, KleemanCR. Blood pressure effects of acute hypercalcemia:studies in patients with chronic renal failure. Ann Intern Med!972;76:741-745

18. Kawasaki T, Delea CS, Bartter FC, Smith H. The effect ofhigh-sodium and low-sodium intakes on blood pressure andother related variables in human subjects with idiopathic hyper-tension Am J Med 1978;64:193-198

19. Luft FC, Weinberger MH. Sodium intake and essential hyper-tension. Hypertension 1982;4(suppl III):III-14-III-19

20. Langford HG. Dietary potassium and hypertension: epidemio-logic data. Ann Intern Med 1983;98(Pt 2):77O-772

21. Marone C, Beretta-Piccoli C, Weidmann P. Acute hypercalce-mic hypertension in man: role of hemodynamics, catechol-amines and renin. Kidney Int 1980;20:92-%

22. Bianchetti MG, Beretta-Piccoli C, Weidmann P, et al. Calciumand blood pressure regulation in normal and hypertensive sub-jects. Hypertension 1983;5(suppl II):II-57-II-65

23. Mori K. The effects of infusion of calcium and magnesium ionson the cardiovascular system in man. Jpn Heart J 1978; 19:226-235

24. Frolich ED, Scott JB, Haddy FJ. Effect of cations on resistanceand responsiveness of renal and forelimb vascular beds. Am JPhysiol 1962;203:583-587

25. Farmer JB, Campbell IK. Calcium and magnesium ions: influ-

ence on the response of an isolated artery to sympathetic nervestimulation, noradrenaline and tyramine. Br J Pharmacol Che-mother l%7;29:319-328

26. Sialer S, McKenna DH, Corliss RJ, Rowe GR. Systemic andcoronary hemodynamic effects of intravenous administrationof calcium chloride. Arch Int Pharmacodyn 1967;169:177—184

27. Overbeck HW, Molnar JI, Haddy FJ. Resistance of blood flowthrough the vascular bed of the dog forelimb: local effects ofsodium, potassium, calcium, magnesium, acetate, hypertonic-ity and hypotonicity. Am J Cardiol 1961 ;8:533—541

28. Overbeck HW, Pamnani MB, Derifield RS. Similar vasocon-strictor responses to calcium in normotensive and essentialhypertensive men. Proc Soc Exp Biol Med 1975; 149:519—525

29. Overbeck HW. Attenuated arteriolar dilator responses to cal-cium in genetically hypertensive rats. Hypertension 1984;6:647-653

30. WebbRC, BohrDF. Mechanism of membrane stabilization bycalcium in vascular smooth muscle. Am J Physiol 1978;235:C227-C232

31. McCarron DA, Ellison DH, Anderson S. Vasodilation mediat-ed by human PTH 1-34 in the spontaneously hypertensive rat.Am J Physiol 1984;246:F96-F100

32. Ellison DH, McCarron DA. Structural prerequisites of para-thyroid hormone's hypotensive action. Am J Physiol 1984;246:F551-F556

33 McCarron DA, Muther RS, Plant SB, Krutzik S. Parathyroidhormone: a determinant of post-transplant blood pressure regu-lation. Am J Kidney Dis 1981;l:38-44

34. Morris CD, Henry HJ, McCarron DA. Randomized, placebo-controlled trial of oral C a + + in human hypertension [Ab-stract]. Kidney Int 1984;25:204

35. Ayachi S. Increased dietary calcium lowers blood pressure inthe spontaneously hypertensive rat. Metabolism 1979;28:1234-1238

36. Hurwitz L, McGuffe LJ, Smith PM, Little SA. Specific inhibi-tion of calcium channels by calcium ions in smooth muscle.J Pharmacol Exp Ther 1982;220:382-388

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from

D H Ellison, R Shneidman, C Morris and D A McCarronEffects of calcium infusion on blood pressure in hypertensive and normotensive humans.

Print ISSN: 0194-911X. Online ISSN: 1524-4563 Copyright © 1986 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/01.HYP.8.6.497

1986;8:497-505Hypertension. 

http://hyper.ahajournals.org/content/8/6/497World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://hyper.ahajournals.org//subscriptions/

is online at: Hypertension Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer information about this process is available in the

is located, click Request Permissions in the middle column of the Web page under Services. Furtherthe Editorial Office. Once the online version of the published article for which permission is being requested

can be obtained via RightsLink, a service of the Copyright Clearance Center, notHypertensionpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on June 2, 2018http://hyper.ahajournals.org/

Dow

nloaded from