- Email: [email protected]
Hypertension is Associated With Increased Urinary Calcium Excretion in Patients With Nephrolithiasis
Hypertension is Associated With Increased Urinary Calcium Excretion in Patients With Nephrolithiasis Brian H. Eisner,*,† Sima P. Porten, Seth K. Bechis and Marshall L. Stoller‡ From the Departments of Urology, Massachusetts General Hospital, Harvard Medical School (BHE), Boston, Massachusetts, and Department of Urology, University of California-San Francisco (BHE, SPP, SKB, MLS), San Francisco, California
Abbreviations and Acronyms BMI ⫽ body mass index CaOx ⫽ calcium oxalate CaP ⫽ calcium phosphate SS ⫽ supersaturation Submitted for publication June 9, 2009. Study received institutional review board approval. * Correspondence: Department of Urology, GRB 1102, Massachusetts General Hospital, 55 Fruit St., Boston, Massachusetts 02114 (FAX: 617-726-6131; e-mail: [email protected]). † Financial interest and/or other relationship with Boston Scientific and Ravine Group. ‡ Financial interest and/or other relationship with Percsys and Ravine Group.
Purpose: The epidemiological relationship between nephrolithiasis and hypertension is well-known. Patients with hypertension are at increased risk for nephrolithiasis and those with nephrolithiasis are at risk for hypertension. Urine calcium or urine citrate may be related to hypertension status. We examined the relationship between hypertension and 24-hour urine composition in patients with nephrolithiasis. Materials and Methods: We retrospectively reviewed the database on 462 stone forming patients to examine the relationship between hypertension and 24-hour urine composition. Multivariate linear regression models were adjusted for age, race, gender, body mass index, diabetes mellitus and 24-hour urine constituents. Nominal logistic regression was also done to examine the hypertension prevalence by quintile of calcium and citrate excretion. Results: On adjusted multivariate analysis compared with normotensive stone formers those with hypertension excreted 25.6 mg per day more urine calcium, corresponding to a 12% increase in urinary calcium excretion. The relative risk of hypertension was significantly associated with quintile of calcium excretion but not with quintile of citrate excretion (1.29, 95% CI 1.02 to 1.61 vs 0.94, 95% CI 0.78 to 1.14). Conclusions: In stone formers hypertension was associated only with significantly increased urine calcium. This association is important when treating patients with nephrolithiasis since those with hypertension may require unique dietary and medical therapy. Key Words: kidney, nephrolithiasis, hypertension, calcium, citric acid
576
www.jurology.com
NEPHROLITHIASIS is a common source of morbidity with a lifetime prevalence in the United States of 10%, which appears to be increasing.1 The epidemiological relationship between nephrolithiasis and hypertension has been established for many years. Patients with hypertension are at significantly increased risk for urinary stone disease.2,3 More recent evidence shows that this relationship is reciprocal, that is patients with urinary stone disease are at risk for incident hypertension.4 – 6 Several groups have examined 24-hour urine chemistry to discover
possible differences in urine constituents between normotensive and hypertensive patients. Urine calcium and citrate abnormalities have been implicated as potential hypertension sequelae that could increase the risk of nephrolithiasis.3 We examined 24-hour urine composition in normotensive and hypertensive stone forming patients.
0022-5347/10/1832-0576/0 THE JOURNAL OF UROLOGY® Copyright © 2010 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 183, 576-579, February 2010 Printed in U.S.A. DOI:10.1016/j.juro.2009.10.011
MATERIALS AND METHODS Study Design We retrospectively reviewed a database on 24-hour urinalysis at a tertiary care
HYPERTENSION AND CALCIUM EXCRETION IN PATIENTS WITH NEPHROLITHIASIS
academic medical center metabolic stone clinic. Patients who presented for an initial metabolic stone assessment and were 18 years old or older were identified and included in the study. Outpatient clinic and hospital records, and 24-hour urine composition data were analyzed. BMI was calculated in kg/m2 from self-reported patient height and weight at 24-hour urine collection. Electronic medical records were reviewed to determine patient demographic information (self-reported race and gender), medical history (history of diabetes mellitus and hypertension) and medication use (thiazide diuretics and potassium citrate). Patients were categorized with hypertension if they met 2 criteria, including 1) hypertension was listed in the medical history of the electronic medical record and 2) they were on at least 1 anti-hypertensive medication, eg -blocker, calcium channel blocker, thiazide, angiotensin converting enzyme inhibitor or angiotensin II receptor blocker. Patients were excluded from study if BMI or medical history could not be obtained, or 24-hour urine collection was deemed inadequate, that is 24-hour urine creatinine less than 800 and less than 600 mg in men and women, respectively. Our standard practice with a new patient with stones is to do the initial clinical evaluation, provide dietary guidelines and perform 24-hour urine chemistry at least 1 month after the patient has adjusted to dietary recommendations. Our dietary guidelines include adequate fluid intake to ensure 1.5 to 2.0 l urine output per day, a low salt diet (potassium salt substitute, low sodium soy sauce substitute, limited salt when shopping and/or cooking, and limited out of house meals) and limited animal protein intake at individual meals.
Urine Collection Patients who presented to our clinic for metabolic stone evaluation underwent 24-hour urinalysis (Litholink®). Self-reported height and weight were recorded at urine collection. Standard urinary parameters were evaluated, including sodium, calcium, citrate, creatinine, uric acid, oxalate, potassium, phosphorus, magnesium, sulfate, pH and urine volume. The SS ratio of CaOx, CaP and uric acid was calculated using the iterative computer program EQUIL2.
95% CI was calculated for all regression coefficients. All analysis was done using JMP® 8.0.
RESULTS Of 462 patients analyzed 139 (30.1%) had a baseline diagnosis of hypertension and 323 (69.9%) did not. Hypertensive patients were older (mean age ⫾ SD 59.7 ⫾ 11.2 vs 48.6 ⴞ 12.7 years, p ⬍0.001), and had greater BMI (28.2 ⴞ 5.8 vs 26.4 ⴞ 5.9 kg/m2, p ⫽ 0.003), a greater incidence of diabetes mellitus (20.9% vs 5.3%, p ⬍0.001) and greater thiazide use (19.4% vs 6.5%, p ⫽ 0.02). Hypertensive patients also had a greater Asian/Pacific Islander-to-white race ratio (28.8%:71.2% vs 15.8%:84.2%, p ⫽ 0.001). There was no difference in gender (40.3% vs 40.9% female, p ⫽ 0.91) or potassium citrate use (10.1% vs 9.6%, p ⫽ 0.87) in hypertensive vs normotensive patients. On univariate analysis hypertensive patients had significantly lower urinary phosphate excretion (0.94 ⫾ 0.34 vs 1.01 ⫾ 0.37 gm per day, p ⫽ 0.04), and significantly lower urine pH (5.95 ⫾ 0.52 vs 6.01 ⫾ 0.54, p ⫽ 0.005) and SSCaP (0.88 ⫾ 0.83 vs 1.23 ⫾ 1.08, p ⬍0.001) than normotensive patients. There were no differences in any other 24-hour urine parameters between the 2 groups (table 1). On multivariate analysis adjusted for potential confounders, including age, race, gender, BMI, diabetes mellitus and 24-hour urine constituents (dietary intake), hypertensive patients excreted significantly more calcium than normotensive patients (25.6 mg per day, 95% CI 5.7– 45.5). There were no differences between the 2 groups for any other 24-hour urine parameters on multivariate analysis (table 2). Similarly adjusted multivariate nominal logistic regression was performed to determine whether the hypertension prevalence changed with increasing Table 1. Univariate analysis of 24-hour urine chemistry by hypertensive status
Statistical Analysis In our analysis we included only a single 24-hour urinalysis. In patients with more than 1, 24-hour urinalysis available only the first urinalysis was used. Univariate analysis of 24-hour urine constituents in patients with and without hypertension was done using the Student t test. Multivariate linear regression of the relationship between hypertension and 24-hour urine constituents was adjusted for possible confounders, including age, gender, race, BMI, diabetes mellitus, thiazide, potassium citrate and 24-hour urine composition (sodium, calcium, citrate, creatinine, uric acid, oxalate, potassium, phosphorus, magnesium, sulfate, pH and urine volume). Another similarly adjusted multivariate analysis was done, in which patients were divided into quintiles of calcium and citrate excretion to determine the prevalence of hypertension as the urine concentration of either one increased. All tests were 2-sided with significance considered at p ⬍0.05. The
577
Calcium (mg) Oxalate (mg) Citrate (mg) Uric acid (gm) Sodium (mmol) Potassium (mEq) Magnesium (mg) Phosphate (gm) Sulfate (mmol) Creatinine (mg) pH Vol (l) SSCaOx (mg) SSCaP Uric acid SS (gm)
Mean ⫾ SD No Hypertension
Mean ⫾ SD Hypertension
p Value
213.4 ⫾ 111.0 42.7 ⫾ 18.5 592.4 ⫾ 364.4 0.70 ⫾ 0.24 170.9 ⫾ 77.1 68.9 ⫾ 32.6 109.8 ⫾ 43.7 1.01 ⴞ 0.37 45.3 ⫾ 17.2 1,574.9 ⫾ 468.2 6.10 ⴞ 0.54 2.13 ⫾ 1.03 7.10 ⫾ 3.96 1.23 ⴞ 1.08 0.96 ⫾ 0.95
206.7 ⫾ 134.5 42.9 ⫾ 18.0 522.7 ⫾ 363.1 0.69 ⫾ 0.22 175.7 ⫾ 71.9 66.1 ⫾ 25.9 104.8 ⫾ 40.1 0.94 ⴞ 0.34 44.3 ⫾ 16.1 1,482.6 ⫾ 489.8 5.95 ⴞ 0.52 2.15 ⫾ 0.89 6.76 ⫾ 4.20 0.88 ⴞ 0.83 1.11 ⫾ 0.97
0.60 0.91 0.06 0.68 0.52 0.34 0.24 0.01 0.58 0.06 0.005 0.88 0.42 <0.001 0.12
Bold type indicates statistically significant at p ⬍0.05.
578
HYPERTENSION AND CALCIUM EXCRETION IN PATIENTS WITH NEPHROLITHIASIS
Table 2. Multivariate adjusted differences in 24-hour urine composition in patients with vs without hypertension Parameter
Difference (95% CI)*
Calcium (mg) Oxalate (mg) Citrate (mg) Uric acid (gm) Sodium (mmol) Potassium (mEq) Magnesium (mg) Phosphate (gm) Sulfate (mmol) Creatinine (mg) pH Vol (l) SSCaOx (mg) SSCaP (mg) Uric acid SS (gm)
25.6 (5.7–45.5) 1.7 (⫺1.7–5.1) ⫺46.3 (⫺117.3–24.6) ⫺0.001 (⫺0.03–0.03) 3.9 (⫺9.3–16.0) 2.1 (⫺2.6–6.8) ⫺2.7 (⫺10.0–4.5) ⫺0.04 (⫺0.09–0.005) ⫺0.7 (⫺3.0–1.6) ⫺17.9 (⫺70.2–34.2) ⫺0.08 (⫺0.2–0.009) ⫺0.03 (⫺0.2–0.1) 0.47 (⫺0.16–1.11) 0.06 (⫺0.09–0.21) ⫺0.005 (⫺0.12–0.11)
* Normotensive is referent and differences are adjusted for age, gender, BMI, diabetes mellitus, thiazide, potassium citrate and 24-hour urine constituents. Bold type indicates statistically significant at p ⬍0.05.
quintile of urinary excretion of calcium or citrate. Analysis revealed that increasing calcium excretion quintile was associated with significant increases in the hypertension prevalence (RR 1.29, 95% CI 1.02– 1.61, p ⫽ 0.03). There was no association between citrate excretion quintile and the hypertension prevalence (RR 0.94, 95% CI 0.78 –1.14, p ⫽ 0.56).
DISCUSSION The relationship between hypertension and nephrolithiasis has been investigated in detail in the last 25 years. Several studies show an increased risk of incident nephrolithiasis in patients with hypertension while others reveal an increased risk of hypertension in those with nephrolithiasis.2,3,5,6 These findings suggest a potential common physiological pathway linking the 2 diseases. Since 24-hour urinalysis is the mainstay of metabolic assessment for nephrolithiasis,7 investigators have attempted to define specific urinary abnormalities associated with hypertension that may be treated with dietary or medical intervention to decrease recurrent stone episodes. In several studies patients with hypertension excreted increased amounts of calcium. Borghi et al compared urine composition and incident urinary stone disease in patients with vs without hypertension.3 No patients had a history of nephrolithiasis. When adjusted for BMI, hypertensive men had increased urine oxalate, SSCaOx and SSCaP, and hypertensive women excreted increased urine calcium, oxalate and SSCaOx compared with normotensive patients. Also, hypertensive patients were at 5-fold increased risk for incident nephrolithiasis during almost 8 years of followup. McCarron et al noted compensatory hyperparathyroidism in hypertensive patients
that resulted from increased urine calcium.8 Animal data corroborate these findings, in that studies have consistently shown hypercalciuria in hypertensive rodent models.9 Our findings are consistent with these reports. On multivariate analysis compared with normotensive patients those with hypertension excreted 25 mg per day more calcium, corresponding to about a 12% increase in calcium excretion. Our findings are underscored by the fact that the hypertensive group had 3-fold greater thiazide diuretic use (19% vs 6%), which decreases urine calcium. Urine calcium was the only significantly different variable in normotensive vs hypertensive cases on multivariate analysis. We cannot assign causality to our findings because we did not study the temporal relationship between hypertension and urinary stone disease. Several plausible explanations exist. Hypertension may be associated with enhanced parathyroid function.8 On the other hand, patients with nephrolithiasis and hypercalciuria may experience renal damage due to urinary obstruction from stone disease, which could in turn be a risk factor for hypertension.4 Patients with hypertension may also experience a renal calcium leak.10 Other groups have examined the possible link between disturbances in renal acid-base handling, which may manifest as hypocitraturia, and hypertension. In rat models hypertension is associated with metabolic acidosis.11,12 Acidosis in turn can lead to hypocitraturia, a common finding in patients with nephrolithiasis.7,13 Taylor et al examined almost 3,000 men and women, including stone formers and nonstone formers, and concluded that hypertension is related to urinary citrate (prevalent hypertension increases as urinary citrate quartile decreases) and hypertension is not consistently related to urine calcium excretion.14 We performed similar analysis in our study, that is multivariate logistic regression of the relationship between calcium or citrate excretion quintile and prevalent hypertension. We found the opposite of what Taylor et al concluded.14 Prevalent hypertension in our study increased with increasing quintile of urine calcium excretion (RR 1.29, p ⫽ 0.03) but was not related to citrate excretion (RR 0.94, p ⫽ 0.56). There are several important distinctions between the 2 series. Their analysis combined stone formers and nonstone formers, which may have yielded different results from our study, in which we examined stone formers only. They performed analysis in younger women, older women and men, respectively, while we included men and women of all ages in analysis, and adjusted for age and gender on multivariate analysis. Also, close examination of their results reveals that prevalent hypertension was significantly associated with increasing quartile
HYPERTENSION AND CALCIUM EXCRETION IN PATIENTS WITH NEPHROLITHIASIS
of calcium excretion in men. While in older women there was no statistically significant relationship between hypertension and calcium excretion, in the highest calcium excretion quartile there was a trend (although it was not statistically significant) toward increased prevalent hypertension in older women (RR 1.3, p ⫽ 0.27). Finally, their analysis of younger women showed a significant decrease in the hypertension prevalence, which is different than our findings. Our study has several inherent weaknesses. It is retrospective and, thus, subject to the shortcomings of a nonprospective study design. Height and weight were self-reported so that reporting errors could potentially have confounded our results. However, the accuracy of self-reported height and weight was previously validated in the adult population.15 Also, hypertensive patients were classified as such by retrospective review of medical records. Although they had a history of hypertension and some were on anti-hypertensive medication, we did not measure blood pressure at urine collection and some patients may have been normotensive at that time. Previous groups have taken a similar approach and the validity of self-reported hypertension, as ascertained from the longitudinal medical record, has been described previously.16,17 We did not report stone analysis in nondiabetic patients. Previous studies from our
579
clinic show a stone distribution similar to that in the literature for stone disease in the United States.1,18 Patients analyzed were stone formers only and the same study in nonstone forming patients may yield different results. Finally, although we used a large number of variables on multivariate analysis,19 the ratio of total patients (462) to the number of variables is within the limits of what is acceptable in common statistical analysis.19 Our finding that certain differences between the groups were not evident on univariate analysis but were present on multivariate analysis suggests that the adjustment for confounders on multivariate analysis may more accurately reflect the relationship between blood pressure and urine composition than simple univariate analysis.19
CONCLUSIONS In patients with nephrolithiasis hypertension appears to be related to increased urinary calcium excretion but not to urinary citrate excretion. This association is important when treating patients with nephrolithiasis since those with hypertension may require unique dietary and medical therapy.
ACKNOWLEDGMENTS Dr. David S. Goldfarb provided advice.
REFERENCES 1. Pearle MS, Calhoun EA, Curhan GC et al: Urologic Diseases in America Project: urolithiasis. Urologic Diseases of America Project. J Urol 2005; 173: 848. 2. Cappuccio FP, Siani A, Barba G et al: A prospective study of hypertension and the incidence of kidney stones in men. J Hypertens 1999; 17: 1017. 3. Borghi L, Meschi T, Guerra A et al: Essential arterial hypertension and stone disease. Kidney Int 1999; 55: 2397. 4. Gillen DL, Coe FL and Worcester EM: Nephrolithiasis and increased blood pressure among females with high body mass index. Am J Kidney Dis 2005; 46: 263. 5. Madore F, Stampfer MJ, Rimm EB et al: Nephrolithiasis and risk of hypertension. Am J Hypertens 1998; 11: 46. 6. Madore F, Stampfer MJ, Willett WC et al: Nephrolithiasis and risk of hypertension in women. Am J Kidney Dis 1998; 32: 802. 7. Pak CY, Britton F, Peterson R et al: Ambulatory evaluation of nephrolithiasis. Classification, clin-
ical presentation and diagnostic criteria. Am J Med 1980; 69: 19. 8. McCarron DA, Pingree PA, Rubin RJ et al: Enhanced parathyroid function in essential hypertension: a homeostatic response to a urinary calcium leak. Hypertension 1980; 2: 162. 9. McCarron DA, Yung NN, Ugoretz BA et al: Disturbances of calcium metabolism in the spontaneously hypertensive rat. Hypertension 1981; 3: I162. 10. Sakhaee K: Nephrolithiasis as a systemic disorder. Curr Opin Nephrol Hypertens 2008; 17: 304. 11. Batlle DC, Sharma AM, Alsheikha MW et al: Renal acid excretion and intracellular pH in saltsensitive genetic hypertension. J Clin Invest 1993; 91: 2178. 12. Lucas PA, Lacour B, McCarron DA et al: Disturbance of acid-base balance in the young spontaneously hypertensive rat. Clin Sci (Lond) 1987; 73: 211. 13. Pak CY, Skurla C and Harvey J: Graphic display of urinary risk factors for renal stone formation. J Urol 1985; 134: 867.
14. Taylor EN, Mount DB, Forman JP et al: Association of prevalent hypertension with 24-hour urinary excretion of calcium, citrate, and other factors. Am J Kidney Dis 2006; 47: 780. 15. Rimm EB, Stampfer MJ, Colditz GA et al: Validity of self-reported waist and hip circumferences in men and women. Epidemiology 1990; 1: 466. 16. Ascherio A, Rimm EB, Giovannucci EL et al: A prospective study of nutritional factors and hypertension among US men. Circulation 1992; 86: 1475. 17. Colditz GA, Martin P, Stampfer MJ et al: Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am J Epidemiol 1986; 123: 894. 18. Low RK, Stoller ML and Schreiber CK: Metabolic and urinary risk factors associated with Randall’s papillary plaques. J Endourol 2000; 14: 507. 19. Vittinghoff E, Glidden DV, Shiboski SC et al: Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models. New York: Springer 2007; p 344.
Abbreviations and Acronyms BMI ⫽ body mass index CaOx ⫽ calcium oxalate CaP ⫽ calcium phosphate SS ⫽ supersaturation Submitted for publication June 9, 2009. Study received institutional review board approval. * Correspondence: Department of Urology, GRB 1102, Massachusetts General Hospital, 55 Fruit St., Boston, Massachusetts 02114 (FAX: 617-726-6131; e-mail: [email protected]). † Financial interest and/or other relationship with Boston Scientific and Ravine Group. ‡ Financial interest and/or other relationship with Percsys and Ravine Group.
Purpose: The epidemiological relationship between nephrolithiasis and hypertension is well-known. Patients with hypertension are at increased risk for nephrolithiasis and those with nephrolithiasis are at risk for hypertension. Urine calcium or urine citrate may be related to hypertension status. We examined the relationship between hypertension and 24-hour urine composition in patients with nephrolithiasis. Materials and Methods: We retrospectively reviewed the database on 462 stone forming patients to examine the relationship between hypertension and 24-hour urine composition. Multivariate linear regression models were adjusted for age, race, gender, body mass index, diabetes mellitus and 24-hour urine constituents. Nominal logistic regression was also done to examine the hypertension prevalence by quintile of calcium and citrate excretion. Results: On adjusted multivariate analysis compared with normotensive stone formers those with hypertension excreted 25.6 mg per day more urine calcium, corresponding to a 12% increase in urinary calcium excretion. The relative risk of hypertension was significantly associated with quintile of calcium excretion but not with quintile of citrate excretion (1.29, 95% CI 1.02 to 1.61 vs 0.94, 95% CI 0.78 to 1.14). Conclusions: In stone formers hypertension was associated only with significantly increased urine calcium. This association is important when treating patients with nephrolithiasis since those with hypertension may require unique dietary and medical therapy. Key Words: kidney, nephrolithiasis, hypertension, calcium, citric acid
576
www.jurology.com
NEPHROLITHIASIS is a common source of morbidity with a lifetime prevalence in the United States of 10%, which appears to be increasing.1 The epidemiological relationship between nephrolithiasis and hypertension has been established for many years. Patients with hypertension are at significantly increased risk for urinary stone disease.2,3 More recent evidence shows that this relationship is reciprocal, that is patients with urinary stone disease are at risk for incident hypertension.4 – 6 Several groups have examined 24-hour urine chemistry to discover
possible differences in urine constituents between normotensive and hypertensive patients. Urine calcium and citrate abnormalities have been implicated as potential hypertension sequelae that could increase the risk of nephrolithiasis.3 We examined 24-hour urine composition in normotensive and hypertensive stone forming patients.
0022-5347/10/1832-0576/0 THE JOURNAL OF UROLOGY® Copyright © 2010 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 183, 576-579, February 2010 Printed in U.S.A. DOI:10.1016/j.juro.2009.10.011
MATERIALS AND METHODS Study Design We retrospectively reviewed a database on 24-hour urinalysis at a tertiary care
HYPERTENSION AND CALCIUM EXCRETION IN PATIENTS WITH NEPHROLITHIASIS
academic medical center metabolic stone clinic. Patients who presented for an initial metabolic stone assessment and were 18 years old or older were identified and included in the study. Outpatient clinic and hospital records, and 24-hour urine composition data were analyzed. BMI was calculated in kg/m2 from self-reported patient height and weight at 24-hour urine collection. Electronic medical records were reviewed to determine patient demographic information (self-reported race and gender), medical history (history of diabetes mellitus and hypertension) and medication use (thiazide diuretics and potassium citrate). Patients were categorized with hypertension if they met 2 criteria, including 1) hypertension was listed in the medical history of the electronic medical record and 2) they were on at least 1 anti-hypertensive medication, eg -blocker, calcium channel blocker, thiazide, angiotensin converting enzyme inhibitor or angiotensin II receptor blocker. Patients were excluded from study if BMI or medical history could not be obtained, or 24-hour urine collection was deemed inadequate, that is 24-hour urine creatinine less than 800 and less than 600 mg in men and women, respectively. Our standard practice with a new patient with stones is to do the initial clinical evaluation, provide dietary guidelines and perform 24-hour urine chemistry at least 1 month after the patient has adjusted to dietary recommendations. Our dietary guidelines include adequate fluid intake to ensure 1.5 to 2.0 l urine output per day, a low salt diet (potassium salt substitute, low sodium soy sauce substitute, limited salt when shopping and/or cooking, and limited out of house meals) and limited animal protein intake at individual meals.
Urine Collection Patients who presented to our clinic for metabolic stone evaluation underwent 24-hour urinalysis (Litholink®). Self-reported height and weight were recorded at urine collection. Standard urinary parameters were evaluated, including sodium, calcium, citrate, creatinine, uric acid, oxalate, potassium, phosphorus, magnesium, sulfate, pH and urine volume. The SS ratio of CaOx, CaP and uric acid was calculated using the iterative computer program EQUIL2.
95% CI was calculated for all regression coefficients. All analysis was done using JMP® 8.0.
RESULTS Of 462 patients analyzed 139 (30.1%) had a baseline diagnosis of hypertension and 323 (69.9%) did not. Hypertensive patients were older (mean age ⫾ SD 59.7 ⫾ 11.2 vs 48.6 ⴞ 12.7 years, p ⬍0.001), and had greater BMI (28.2 ⴞ 5.8 vs 26.4 ⴞ 5.9 kg/m2, p ⫽ 0.003), a greater incidence of diabetes mellitus (20.9% vs 5.3%, p ⬍0.001) and greater thiazide use (19.4% vs 6.5%, p ⫽ 0.02). Hypertensive patients also had a greater Asian/Pacific Islander-to-white race ratio (28.8%:71.2% vs 15.8%:84.2%, p ⫽ 0.001). There was no difference in gender (40.3% vs 40.9% female, p ⫽ 0.91) or potassium citrate use (10.1% vs 9.6%, p ⫽ 0.87) in hypertensive vs normotensive patients. On univariate analysis hypertensive patients had significantly lower urinary phosphate excretion (0.94 ⫾ 0.34 vs 1.01 ⫾ 0.37 gm per day, p ⫽ 0.04), and significantly lower urine pH (5.95 ⫾ 0.52 vs 6.01 ⫾ 0.54, p ⫽ 0.005) and SSCaP (0.88 ⫾ 0.83 vs 1.23 ⫾ 1.08, p ⬍0.001) than normotensive patients. There were no differences in any other 24-hour urine parameters between the 2 groups (table 1). On multivariate analysis adjusted for potential confounders, including age, race, gender, BMI, diabetes mellitus and 24-hour urine constituents (dietary intake), hypertensive patients excreted significantly more calcium than normotensive patients (25.6 mg per day, 95% CI 5.7– 45.5). There were no differences between the 2 groups for any other 24-hour urine parameters on multivariate analysis (table 2). Similarly adjusted multivariate nominal logistic regression was performed to determine whether the hypertension prevalence changed with increasing Table 1. Univariate analysis of 24-hour urine chemistry by hypertensive status
Statistical Analysis In our analysis we included only a single 24-hour urinalysis. In patients with more than 1, 24-hour urinalysis available only the first urinalysis was used. Univariate analysis of 24-hour urine constituents in patients with and without hypertension was done using the Student t test. Multivariate linear regression of the relationship between hypertension and 24-hour urine constituents was adjusted for possible confounders, including age, gender, race, BMI, diabetes mellitus, thiazide, potassium citrate and 24-hour urine composition (sodium, calcium, citrate, creatinine, uric acid, oxalate, potassium, phosphorus, magnesium, sulfate, pH and urine volume). Another similarly adjusted multivariate analysis was done, in which patients were divided into quintiles of calcium and citrate excretion to determine the prevalence of hypertension as the urine concentration of either one increased. All tests were 2-sided with significance considered at p ⬍0.05. The
577
Calcium (mg) Oxalate (mg) Citrate (mg) Uric acid (gm) Sodium (mmol) Potassium (mEq) Magnesium (mg) Phosphate (gm) Sulfate (mmol) Creatinine (mg) pH Vol (l) SSCaOx (mg) SSCaP Uric acid SS (gm)
Mean ⫾ SD No Hypertension
Mean ⫾ SD Hypertension
p Value
213.4 ⫾ 111.0 42.7 ⫾ 18.5 592.4 ⫾ 364.4 0.70 ⫾ 0.24 170.9 ⫾ 77.1 68.9 ⫾ 32.6 109.8 ⫾ 43.7 1.01 ⴞ 0.37 45.3 ⫾ 17.2 1,574.9 ⫾ 468.2 6.10 ⴞ 0.54 2.13 ⫾ 1.03 7.10 ⫾ 3.96 1.23 ⴞ 1.08 0.96 ⫾ 0.95
206.7 ⫾ 134.5 42.9 ⫾ 18.0 522.7 ⫾ 363.1 0.69 ⫾ 0.22 175.7 ⫾ 71.9 66.1 ⫾ 25.9 104.8 ⫾ 40.1 0.94 ⴞ 0.34 44.3 ⫾ 16.1 1,482.6 ⫾ 489.8 5.95 ⴞ 0.52 2.15 ⫾ 0.89 6.76 ⫾ 4.20 0.88 ⴞ 0.83 1.11 ⫾ 0.97
0.60 0.91 0.06 0.68 0.52 0.34 0.24 0.01 0.58 0.06 0.005 0.88 0.42 <0.001 0.12
Bold type indicates statistically significant at p ⬍0.05.
578
HYPERTENSION AND CALCIUM EXCRETION IN PATIENTS WITH NEPHROLITHIASIS
Table 2. Multivariate adjusted differences in 24-hour urine composition in patients with vs without hypertension Parameter
Difference (95% CI)*
Calcium (mg) Oxalate (mg) Citrate (mg) Uric acid (gm) Sodium (mmol) Potassium (mEq) Magnesium (mg) Phosphate (gm) Sulfate (mmol) Creatinine (mg) pH Vol (l) SSCaOx (mg) SSCaP (mg) Uric acid SS (gm)
25.6 (5.7–45.5) 1.7 (⫺1.7–5.1) ⫺46.3 (⫺117.3–24.6) ⫺0.001 (⫺0.03–0.03) 3.9 (⫺9.3–16.0) 2.1 (⫺2.6–6.8) ⫺2.7 (⫺10.0–4.5) ⫺0.04 (⫺0.09–0.005) ⫺0.7 (⫺3.0–1.6) ⫺17.9 (⫺70.2–34.2) ⫺0.08 (⫺0.2–0.009) ⫺0.03 (⫺0.2–0.1) 0.47 (⫺0.16–1.11) 0.06 (⫺0.09–0.21) ⫺0.005 (⫺0.12–0.11)
* Normotensive is referent and differences are adjusted for age, gender, BMI, diabetes mellitus, thiazide, potassium citrate and 24-hour urine constituents. Bold type indicates statistically significant at p ⬍0.05.
quintile of urinary excretion of calcium or citrate. Analysis revealed that increasing calcium excretion quintile was associated with significant increases in the hypertension prevalence (RR 1.29, 95% CI 1.02– 1.61, p ⫽ 0.03). There was no association between citrate excretion quintile and the hypertension prevalence (RR 0.94, 95% CI 0.78 –1.14, p ⫽ 0.56).
DISCUSSION The relationship between hypertension and nephrolithiasis has been investigated in detail in the last 25 years. Several studies show an increased risk of incident nephrolithiasis in patients with hypertension while others reveal an increased risk of hypertension in those with nephrolithiasis.2,3,5,6 These findings suggest a potential common physiological pathway linking the 2 diseases. Since 24-hour urinalysis is the mainstay of metabolic assessment for nephrolithiasis,7 investigators have attempted to define specific urinary abnormalities associated with hypertension that may be treated with dietary or medical intervention to decrease recurrent stone episodes. In several studies patients with hypertension excreted increased amounts of calcium. Borghi et al compared urine composition and incident urinary stone disease in patients with vs without hypertension.3 No patients had a history of nephrolithiasis. When adjusted for BMI, hypertensive men had increased urine oxalate, SSCaOx and SSCaP, and hypertensive women excreted increased urine calcium, oxalate and SSCaOx compared with normotensive patients. Also, hypertensive patients were at 5-fold increased risk for incident nephrolithiasis during almost 8 years of followup. McCarron et al noted compensatory hyperparathyroidism in hypertensive patients
that resulted from increased urine calcium.8 Animal data corroborate these findings, in that studies have consistently shown hypercalciuria in hypertensive rodent models.9 Our findings are consistent with these reports. On multivariate analysis compared with normotensive patients those with hypertension excreted 25 mg per day more calcium, corresponding to about a 12% increase in calcium excretion. Our findings are underscored by the fact that the hypertensive group had 3-fold greater thiazide diuretic use (19% vs 6%), which decreases urine calcium. Urine calcium was the only significantly different variable in normotensive vs hypertensive cases on multivariate analysis. We cannot assign causality to our findings because we did not study the temporal relationship between hypertension and urinary stone disease. Several plausible explanations exist. Hypertension may be associated with enhanced parathyroid function.8 On the other hand, patients with nephrolithiasis and hypercalciuria may experience renal damage due to urinary obstruction from stone disease, which could in turn be a risk factor for hypertension.4 Patients with hypertension may also experience a renal calcium leak.10 Other groups have examined the possible link between disturbances in renal acid-base handling, which may manifest as hypocitraturia, and hypertension. In rat models hypertension is associated with metabolic acidosis.11,12 Acidosis in turn can lead to hypocitraturia, a common finding in patients with nephrolithiasis.7,13 Taylor et al examined almost 3,000 men and women, including stone formers and nonstone formers, and concluded that hypertension is related to urinary citrate (prevalent hypertension increases as urinary citrate quartile decreases) and hypertension is not consistently related to urine calcium excretion.14 We performed similar analysis in our study, that is multivariate logistic regression of the relationship between calcium or citrate excretion quintile and prevalent hypertension. We found the opposite of what Taylor et al concluded.14 Prevalent hypertension in our study increased with increasing quintile of urine calcium excretion (RR 1.29, p ⫽ 0.03) but was not related to citrate excretion (RR 0.94, p ⫽ 0.56). There are several important distinctions between the 2 series. Their analysis combined stone formers and nonstone formers, which may have yielded different results from our study, in which we examined stone formers only. They performed analysis in younger women, older women and men, respectively, while we included men and women of all ages in analysis, and adjusted for age and gender on multivariate analysis. Also, close examination of their results reveals that prevalent hypertension was significantly associated with increasing quartile
HYPERTENSION AND CALCIUM EXCRETION IN PATIENTS WITH NEPHROLITHIASIS
of calcium excretion in men. While in older women there was no statistically significant relationship between hypertension and calcium excretion, in the highest calcium excretion quartile there was a trend (although it was not statistically significant) toward increased prevalent hypertension in older women (RR 1.3, p ⫽ 0.27). Finally, their analysis of younger women showed a significant decrease in the hypertension prevalence, which is different than our findings. Our study has several inherent weaknesses. It is retrospective and, thus, subject to the shortcomings of a nonprospective study design. Height and weight were self-reported so that reporting errors could potentially have confounded our results. However, the accuracy of self-reported height and weight was previously validated in the adult population.15 Also, hypertensive patients were classified as such by retrospective review of medical records. Although they had a history of hypertension and some were on anti-hypertensive medication, we did not measure blood pressure at urine collection and some patients may have been normotensive at that time. Previous groups have taken a similar approach and the validity of self-reported hypertension, as ascertained from the longitudinal medical record, has been described previously.16,17 We did not report stone analysis in nondiabetic patients. Previous studies from our
579
clinic show a stone distribution similar to that in the literature for stone disease in the United States.1,18 Patients analyzed were stone formers only and the same study in nonstone forming patients may yield different results. Finally, although we used a large number of variables on multivariate analysis,19 the ratio of total patients (462) to the number of variables is within the limits of what is acceptable in common statistical analysis.19 Our finding that certain differences between the groups were not evident on univariate analysis but were present on multivariate analysis suggests that the adjustment for confounders on multivariate analysis may more accurately reflect the relationship between blood pressure and urine composition than simple univariate analysis.19
CONCLUSIONS In patients with nephrolithiasis hypertension appears to be related to increased urinary calcium excretion but not to urinary citrate excretion. This association is important when treating patients with nephrolithiasis since those with hypertension may require unique dietary and medical therapy.
ACKNOWLEDGMENTS Dr. David S. Goldfarb provided advice.
REFERENCES 1. Pearle MS, Calhoun EA, Curhan GC et al: Urologic Diseases in America Project: urolithiasis. Urologic Diseases of America Project. J Urol 2005; 173: 848. 2. Cappuccio FP, Siani A, Barba G et al: A prospective study of hypertension and the incidence of kidney stones in men. J Hypertens 1999; 17: 1017. 3. Borghi L, Meschi T, Guerra A et al: Essential arterial hypertension and stone disease. Kidney Int 1999; 55: 2397. 4. Gillen DL, Coe FL and Worcester EM: Nephrolithiasis and increased blood pressure among females with high body mass index. Am J Kidney Dis 2005; 46: 263. 5. Madore F, Stampfer MJ, Rimm EB et al: Nephrolithiasis and risk of hypertension. Am J Hypertens 1998; 11: 46. 6. Madore F, Stampfer MJ, Willett WC et al: Nephrolithiasis and risk of hypertension in women. Am J Kidney Dis 1998; 32: 802. 7. Pak CY, Britton F, Peterson R et al: Ambulatory evaluation of nephrolithiasis. Classification, clin-
ical presentation and diagnostic criteria. Am J Med 1980; 69: 19. 8. McCarron DA, Pingree PA, Rubin RJ et al: Enhanced parathyroid function in essential hypertension: a homeostatic response to a urinary calcium leak. Hypertension 1980; 2: 162. 9. McCarron DA, Yung NN, Ugoretz BA et al: Disturbances of calcium metabolism in the spontaneously hypertensive rat. Hypertension 1981; 3: I162. 10. Sakhaee K: Nephrolithiasis as a systemic disorder. Curr Opin Nephrol Hypertens 2008; 17: 304. 11. Batlle DC, Sharma AM, Alsheikha MW et al: Renal acid excretion and intracellular pH in saltsensitive genetic hypertension. J Clin Invest 1993; 91: 2178. 12. Lucas PA, Lacour B, McCarron DA et al: Disturbance of acid-base balance in the young spontaneously hypertensive rat. Clin Sci (Lond) 1987; 73: 211. 13. Pak CY, Skurla C and Harvey J: Graphic display of urinary risk factors for renal stone formation. J Urol 1985; 134: 867.
14. Taylor EN, Mount DB, Forman JP et al: Association of prevalent hypertension with 24-hour urinary excretion of calcium, citrate, and other factors. Am J Kidney Dis 2006; 47: 780. 15. Rimm EB, Stampfer MJ, Colditz GA et al: Validity of self-reported waist and hip circumferences in men and women. Epidemiology 1990; 1: 466. 16. Ascherio A, Rimm EB, Giovannucci EL et al: A prospective study of nutritional factors and hypertension among US men. Circulation 1992; 86: 1475. 17. Colditz GA, Martin P, Stampfer MJ et al: Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am J Epidemiol 1986; 123: 894. 18. Low RK, Stoller ML and Schreiber CK: Metabolic and urinary risk factors associated with Randall’s papillary plaques. J Endourol 2000; 14: 507. 19. Vittinghoff E, Glidden DV, Shiboski SC et al: Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models. New York: Springer 2007; p 344.