Habitual Excessive Dietary Salt Intake and

Blood Pressure Levels in Renal Transplant Recipients

JOHN H. KALBFLEISCH, Ph.D.

LEE A. HEBERT, M.D.

Columbus, Ohio

JACOB LEMANN, JR., M.D.

WALTER F. PIERING, M.D.

JOSEPH A. BERES, M.D.

Milwaukee, Wisconsin

From the Department of Medicine and the VA Research Service, Medical College of Wisconsin, Milwaukee, Wisconsin and the Department of Medicine, Chii State University, Cotumbus, Ohio. Supported in part by Grants NIH RR-00058, AM- 15089. and Ht.-25404. Requests for reprints should be addressed to Dr. Lee A. Hebert, Ohio State University, 466 West 10th Ave., Columbus, Ohio 43210. Manuscript accepted on January 26, 1982.

We observed that renal transplant recipients with good graft function (mean serum creatinine level 1.5 mg/dl f 0.5 SD, N = 68) had di- etary salt intakes (estknated from serial measurements of 24-hour sodlum excretion rate) which averaged 43 percent hlgbsr than that of a comparable group of heatthy sub@cts. There was no correlation between Mood pressure levels and satt intake and, desptto the htgh dietary satt intake, hypertension was present in onty 29 pattents and was usuatty mtid; mean systolic and diastotic Mood pressures were 132f lOmmHgand8gf7mmH&r~~~whlkthepatients were receiving antihypertensive medication (median number of standard doses of antihypertensive medication was 1.0 doses/ patient per day). These observations suggest that high dietary satt intake does not exert a powerful blood pressure elevating effect, since any effect of high dietary salt intake to raise Mood pressure should have been magnified In the renal transplant rectpients be- cause of their reduced renal mass and their chronic glucocorticoid therapy.

During the serial follow-up of our patients with successful kidney grafts we observed that their urinary sodium excretion, a measure of salt intake, was consistently much higher than that observed in healthy subjects eating normal diets. Since it has been suggested that ex- cessive dietary salt intake may predispose a patient to the develop- ment of hypertension [I-3], we have examined the relationship be- tween dietary salt intake and blood pressure levels in our patients. Renal transplant recipients are apt subjjcts for such a study since their reduced renal mass [4] and long-term glucocorticoid therapy [5] promote sodium retention and should magnify the effect of a higher dietary salt intake to raise blood pressure. Another asset of the present study is that abundant data on blood pressure levels and dietary salt intake were available on each patient. In all previous studies of the relationship between salt intake and blood pressure level, salt intake was estimated from either the dietary history or relatively few mea- surements of urinary sodium excretion obtained over a relatively brief period. Such approaches to estimate salt intake are not optimal be- cause dietary history is often unreliable as a measure of salt intake, and urinary sodium excretion tends to vary widely from day to day rendering a small number of measurements unreliable as an estimate of the prevailing dietary salt intake [6-81. Thus, the present study represents a more detailed analysis of the relationship between dietary salt intake and blood pressure levels than has previously been re- ported.

August 1982 The Annrlcan Journal ef Medicine Volume 73 205

SALT INTAKE AND HYPERTENSION-KALEFLEISCH ET AL.

TABLE I Clinical Data on the Health Control Subjects and Transplant Patients of this Study

Healthy Adults

All Recipients

Recipients Not Taking

Antihypertensive Medication

Recipients Taking Antihypertensive

Medication

Number of subjects 80 68 Age (years) 20 to 40 35f 12 Weight (kg) 736 15 72f 15 Systolic pressure (mm Hg) NA 124 f 12 Diastolic pressure (mm Hg) NA 83 f 8 Urinary creatinine excretion 1800 f 480 1620 f 405’ l

(mgiday) Serum creatinine concentra- 1.1 f 0.2 1.5 f 0.3t

tion (mg/lOO ml) Creatinine clearance (ml/min) 1 IO f 24 80 f 2Ot Urinary sodium excretion 167f62 224 f 55t

(mmol/day) UtiV/Weight 2.26 f 0.69 3.25 f O-83$ uN,v/uCrv 0.099 f 0.032 0.142 f 0.034t

39 32f 13 69f 15 118 f 9 78 f 7 1598 f 415”’

1.5 f 0.3$ 1.5 f 0.3x

78 f 2Ot 82 f 2Oi 217 f 57t 234 f 51x

3.25 f 0.82f 0.138 f 0.027$

3.24 f 0.855 0.148 f 0.041t

29 38f 11 74f 13 132 f 10’ 89 f 7’ 1650 f 396

UN,V = 24 hr urinary sodium excretion rate; Uc,V = 24 hr urinary creatinine excretion rate; NA = not available. l p <O.Ol indicates that the mean value is significantly higher than the corresponding mean value in re- cipients not taking antihypertensive medication. l l p <0.05 ( indicates that the mean value is significantly different from the corresponding value for $ b <O.Oi 1 normal adults.

METHODS

The data analyzed in this study were obtained from the clinical records of the renal transplant recipients of the Medical College of Wisconsin at the Milwaukee Regional Medical Center. The patients selected for study had stable renal

function with a serum creatinine level of 2.5 mg/dl or less and were free from diseases that cause renal sodium retention or wasting. The patients were not aware that data were being collected to examine the relationship between dietary salt intake and blood pressure levels. Serial measurements of body weight, blood pressure, serum creatinine concentration, and 24-hour urinary excretion of creatinine and sodium were obtained on two to 16 occasions (mean 7 f 3) in 68 patients. Individual 24-hour urine values were excluded from analysis if, at the time of collection of the urine specimen, there was evidence of a sudden weight change of more than 1 kg. There were 25 women and 43 men whose ages ranged from 14 to 58 years (mean age 34 f 13 years). The measurements were made every three to six months beginning three months after grafting. The period of observations ranged from three to 60 months. Ninetyane percent of the patients were followed for more than 24 months. The number and type of diuretics and other antihypertensive agents administered, if any, were recorded. A standard dose of an antihypertensive medication was defined as follows: hydrochlorothiazide, 50 mg; furo- semide, 40 mg; methyldopa, 250 mg; propanoloi, 40 mg; metoprolol, 50 mg; and hydralazine, 25 mg. All patients were eating an unrestricted diet and none were receiving sodium bicarbonate therapy. Ail patients received azathioprine and prednisone or methylprednisolone according to a dose schedule previously described [9]. Almost all of the patients

were receiving maintenance doses of azathioprine and ste- roids at the time of this study. Measurements of body weight, serum creatinine concentration, and 24-hour urinary excretion of creatinine and sodium were also made in 80 healthy, nonobese adults (59 men and 21 women) who were eating self-selected diets. These individuals ranged in age from 20 to 40 years, were free of cardiovascular renal diseases and were taking no drugs.

The data are statistically summarized by presentation of means and standard deviations for patient groups. For con- tinuous variates the independent sample t test was used to assess differences between the means of differing groups of patients or healthy control subjects. Simple linear re- gression and correlation analysis was used to assess and determine the straight-line relationship between two vari- ables. Ninety-five percent confidence zones about regression lines were obtained from the confidence interval formulas for future observations. Multiple regression analysis was used to assess the joint influence of several independent (but correlated) variates on systolic and diastolic blood pressures. levels of significance are indicated herein.

RESULTS

Sodium Excretion. As shown in Table I, there were

no significant differences in body weight between the patients and the control subjects, but the patients ex- creted slightly less creatinine in their urine. The mean serum creatinine concentration was higher in the pa- tients and, as a consequence, mean creatinine ciear- ante was lower. Daily urinary sodium excretion, whether expressed on an absolute basis or per unit

206 August 1992 The American Journal of Medicine Volume 73

SALT INTAKE AND HYF’ERTENSIDN-KALBFLEISCH ET AL.

400

URINE 300

Na

200

mmol/day 100

0

HEALTHY SUBJECTS

0

/??TJZ 22

I I 1 I 1 I I

50 60 70 80 90 100 110

BODY WEIGHT Kg

. _ . . . . . . . . . . Figure 1. Daily urinary sodium (Na) excretion as a function of OoUy weignt N) neanny aams eating their usual self-selected diets. The mean regression relationship: Uinaty Ne, mmoUday = 15 + 2.5 kg body weight; r = 0.59, is shown by the heavy line whereas the lighter lines show the 95 percent confidence zone for the relationship.

400

URINE 300

Na 200

mmoliday

100

0

KIDNEY TRANSPLANT PATIENTS

. l . .

. l .

. . . . l 8 ” :*

. .

0 0. ’ . 0. : e :

. y. 0.

. ;_. l .

.

0. l .

.

60 60 70 80 90 100 li0

BODY WEIGHT Kg

Figure 2. Daily urinary sodium (Na) excretion as a functkm of body weight in kkfney transplant @en&s eatm thsir usual self-se&teddMs. The data points are vim on the 95mnt confidence zone for the sang reWonship in h&thy a&&s. For all the kMney transplant patients the mean regression relationship is urinary Na, mmoilday = 97 -t 1.8 kg body weight; r = 0.47.

August 1982 The Amwlcan JourMl d Modkkw vohlma 73 207

SALT INTAKE AND HYPERTENSION-KALBFLEISCH ET AL.

TABLE II Simple Linear Correlation Coefficients

Patients Not Taking Patients Taking

All Antihypertensive Antihypertensive Patients Medication Medication

Systolic blood pressure with

Age 0.34’ 0.23 0.22 Weight 0.35” 0.40+ 0.18 Ccr -0.10 -0.22 -0.15 UNav 0.04 0.10 -0.32

Diastolic blood pressure with

Age 0.29+ 0.31 0.06 Weight 0.40’ 0.46’ 0.27 Ccr -0.05 0.01 -0.30

UNav 0.15 0.14 -0.04

Ccr = creatinine clearance; U&V = 24 hr urinary sodium excretion rate. l p <O.Ol. + p <0.05.

creatinine or per unit body weight, was significantly greater among the transplant patients than among the healthy subjects.

Figure 1 shows that urinary sodium excretion among the healthy adults was significantly related to body weight, an expected result, since in the absence of sweating or abnormal gastrointestinal fluid losses, urinary sodium excretion closely matches dietary salt intake, and subjects with higher lean body mass can be expected to eat greater quantities of food and, thus more salt. A similar relationship between sodium ex- cretion and urinary creatinine excretion was also ob- served: urinary sodium, mmol/day = 37 i- 0.072 X urinary creatinine, mg/day, r = 0.55.

Figure 2 shows the data points for daily urinary so- dium excretion as a function of body weight for each of the transplant patients, superimposed upon the same relationship in normal subjects (Figure 1). As can be seen, the data points for all but five of the patients were on or above the mean regression relationship for normal subjects, and the data points for 13 of the patients were above the 95 percent confidence zone. The differences in urinary sodium excretion between the patients and the normal subjects was even more striking when the urinary sodium excretion rate was expressed per gram of urinary creatinine. This analysis showed that 67 of 68 patients had urinary sodium excretion rates above the mean regression line for normal subjects and 20 of the patients had urinary sodium excretion rates above the 95 percent confidence zone established for healthy individuals. Blood Pressure Levels. For each patient the mean systolic and mean diastolic blood pressures during the

period of observation were calculated. Table I shows the average of the individual means for blood pressure measurements in the 68 patients; the 39-patient subgroup not taking antihypertensive medication and the 29-patient subgroup taking antihypertensive med- ications. Both systolic and diastolic blood pressures were significantly higher in the subgroup taking the medication than they were in the subgroup not taking the medication (p <O.OOl). Among the patients re- ceiving antihypertensive medication, the median number of standard daily doses of antihypertensive medication was 1 .O doses/patient per day. All hyper- tensive patients received a diuretic (either hydrochlo- rothiazide or furosemide). Thus, half of the hypertensive patients received hydrochlorothiazide, 50 mg/day or furosemide, 40 mg/day or less. The mean number of standard doses of antihypertensive medication for the entire group taking medication was 2.1 f 2.3 doses/ patient per day. However, only six patients received more than two standard doses of antihypertensive medications per day. Thus, in almost all of the hyper- tensive patients, blood pressure elevations were mild and/or readily controlled by small amounts of antihy- pertensive medications. Correlates of Blood Pressure Levels. Multiple and simple linear regression analyses were applied to the data to examine for the correlates of systolic and dia- stolic blood pressure levels. Independent variables used were age, body weight, daily urinary sodium excretion, and creatinine clearance. The simple linear correlation coefficients are shown in Table II. The analyses indi- cated that, for the entire group of transplant patients, systolic and diastolic blood pressure levels were sig- nificantly correlated only to age and body weight.

COMMENTS

The role of dietary salt intake in the pathogenesis of hypertension is controversial [lo]. However, the pre- vailing view is that in many individuals, a high-salt intake will lead eventually to elevated blood pressure and, in established hypertension, a high-salt intake will further elevate blood pressure [ l-3, lo- 151.

The evidence that a high dietary salt intake leads to hypertension in man is based largely on epidemiologic studies which have compared dietary salt intake and blood pressure levels in different societies. In general, it has been found that societies with a high dietary salt intake have a greater incidence of hypertension than do societies with a low dietary salt intake [2, lo]. These studies, however, have been criticized on the grounds that, in addition to differences in dietary salt intake, these societies have cultural and genetic differences which could account for the differences in blood pres- sure levels. Furthermore, most intrasocietal evaluations

208 August 1982 The American Journal of Medicine Volume 73

SALT INTAKE AND HYPERTENSION-KALBFLEISCH ET AL

of the relationship between salt intake and blood pressure have failed to demonstrate a significant cor- relation between the two [6, lo]. The latter studies, however, have been criticized on the grounds that salt intake is already excessive in the societies studied and is at a level that can support hypertension. Thus, vari- ations within that excessive level of salt intake may have little further effect on blood pressure [lo]. A fur- ther criticism of these studies is that the assessment of blood pressure levels and salt intake were based on relatively few measurements per patient and these measurements can be expected to vary widely from day to day in each patient [ 6-81. Thus, such data may yield inaccurate estimates of the prevailing level of salt intake and the blood pressure level.

This study shows that, despite habitual excessive dietary salt intake and an apparent predisposition to dietary salt-related hypertension, the incidence of hy- pertension in the transplant population was only slightly higher than that expected in the general Milwaukee population * Furthermore, when hypertension was present in the transplant patient, it was usually mild and well controlled with a single daily dose of a diuretic.

It is unlikely that the modest incidence and the mildness of hypertension in the patients with high di- etary salt intake are the result of renal salt wasting by the allograft or by the patient’s native kidneys. Most of the patients had negligible or no urine output from their

native kidneys at tha tima of renal transplantation, and significant renal salt wasting by renal allografts with

adequate and stable renal function has not previously

been reported nor have we observed such a phenom-

l In the age range of 35 to 44 years, approximately 30 percent of the white men and 40 percent of the black men in the general Mil- waukee population have systolic blood pressures greeter than 140 mm H9 and/or diastolic blood pressures greater than 90 mm Hg 1161.

enon. Diuretic therapy itself also appears not to be the cause of the chronic increase in urinary sodium ex- cretion in the group taking the medication since di- uretics do not increase the appetite for salt [ 171. Thus, it seems likely that the high 24-hour urinary sodium excretion rates observed in our patients are the result of a primary increase in dietary salt intake. Interestingly, when patients were questioned at the conclusion of the study regarding their level of salt intake, most perceived their salt intake to be less than normal. This misper- ception could be the result of a change in the taste threshold for salt or it could be an attempt on the part of the patients to conceal their dietary “rebound he- donism” which so commonly occurs in renal transplant recipients who have received a successful graft after months or years of dietary restrictions. In favor of the former hypothesis is the recent demonstration by Ma- hajan et al. [ 181 of decreased perception of saltiness by renal transplant recipients. Regardless of these considerations, the present data indicate that among renal transplant recipients with adequate and stable levels of renal function, remarkably high dietary salt intakes are well tolerated. Nevertheless, high dietary salt intake should not be encouraged in renal transplant patients since there is no evidence that a high salt intake is of benefit to the patient.

We conclude that a high dietary salt intake does not exert a powerful blood pressure elevating effect in renal transplant recipients who have adequate renal function. Furthermore, these data imply that a high dietary salt intake has even less of a blood pressure elevating effect among individuals with normal renal function.

ACKNOWLEDGMENT

The authors gratefully acknowledge the assistance of Dr. Nancy Adams and Elizabeth and Paul Hebert in the collection of data, and the secretarial assistance of Carmela Price and Sandy Taylor.

REFERENCES

1. Dahl LK: Salt and hypertension. Am J Clin Nub 1972; 25: 231-244.

2. Freis ED: Salt, volume and the penetration of hypertension. Circulation 1976; 53(4): 569-595.

3. Finn R, McCoonocbie K, Box MO, et al.: Blood pressure and salt intake: an intra-population study. Lencet 1961; I: 1097.

4. Haddy FJ: Mechanism. prevention and therapy of sodium- dependent hypertension. Am J h&d 1960; 69: 746-756.

5. Siegel RR, Luke RG, Hellebusch AA: Reduction of toxicity of corbieroid therapy after renal transplentetion. Am J Med 1972: 53: 159-169.

6. Liu K, Cooper R, McKeever J, et al.: Assessment of the as- sociation between habitual salt intake and high blood

pressure: methodological problems. Am J Epidemioll979; llO(2): 219-226.

7. Pickering G: Salt intake and essential hypertension. Cardic- vascular Reviews and Reports 1960; l(1): 13-17.

6. Cummins RO. Sheper AG, Walker M: Methodologicel problems with estimation of salt intake. Lencet 1961; I: 1373- 1374.

9. Breitenfield RV, Hebert LA, Lernenn J Jr, et al.: Stability of renal transplant function with alternate-day corticosteroid therapy. JAMA 1960; 224(2): 151-156.

10. Swales JD: Dietary salt and hypertension. Lancet 1980; I: 1177-1179.

11. Kempoer W: Treatment of hypertensive vascular disease with rice diet. Am J Med 1948; 4: 545-577.

Au@ 1982 The Amerkan Journal of Medkke Volume 73 209

SALT INTAKE AND HYPERTENSION-KALBFLEISCH ET AL.

12. Dole VP, Dahl LK, Cotzias GC, et al.: Dietary treatment of hypertension. Clinical and metabolic studies of patients on the rice-fruit diet. The Hospital of the Rockefeller Institute for Medical Research, New York: 1950.

13. Corcoran AC, Taylor RD, Page IH: Controlled observations on the effect of low sodium diet therapy in essential hyper- tension. Circulation 1951; 3(l): 1-16.

14. Parijs J, Joossens JV, Van der Linden L, et al.: Moderate so- dium restriction and diuretics in the treatment of hyper- tension. Am Heart J 1973; 85(l): 22-34.

15. Morgan T, Gillies A, Morgan G, et al.: Hypertension treated by salt restriction, Lancet 1978; 1: 227-230.

16. ltskovitz l-Q Kochar MS, Anderson AJ, et al.: Patterns of blood pressure in Milwaukee. JAMA 1977; 238(8): 864-868.

17. Bing RF, Thurston H, Swales JD: Salt intake and diuretic treatment of hypertension. Lancet, 1979; 2: 12 1 - 122.

18. Mahajan S, Bornstein K, Abraham J, et al.: Hypogeusia in renal transplant patients (abstract) 14th Annual Meeting, Amer- ican Society of Nephrology, November 22-24, 1981, Washington, DC.: 171A.

210 August 1982 The American Journal of Medlclne Volume 73