Abstract
ABSTRACT. Menopause is associated with increased urinary calcium excretion, which could increase the risk for the development of calcium-containing kidney stones. However, it is unknown whether menopause and postmenopausal hormone (PMH) use are independent risk factors for incident kidney stone disease in women. Data from 91,731 female Nurses’ Health Study participants who provided information on diet, menopause status, and kidney stone disease were used to examine the independent association between menopause and PMH use and risk of incident kidney stones. No association was found between menopause and incident kidney stones in age-adjusted (relative risk [RR], 1.07; 95% CI, 0.85 to 1.34) or multivariate models (RR, 1.12; 95% CI, 0.89 to 1.41). However, when the association between the type of menopause and risk of incident kidney stones was examined, surgical menopause was associated with an increased risk in both the age-adjusted (RR, 1.37; 95% CI, 1.05 to 1.77) and multivariate models (RR, 1.39; 95% CI, 1.07 to 1.81), whereas natural menopause was not. Compared with never-use, past or current PMH use (including duration of PMH use) was not associated with incident kidney stones among postmenopausal women. In conclusion, no association was found between menopause and PMH use and incident kidney stones. Surgical menopause, however, may be associated with an increased risk. E-mail: gcurhan@Partners.org
Approximately 5% of all women living in the United States will experience the passage of a kidney stone before the age of 70 yr (1). The passage of a kidney stone is very painful, and the costs incurred due to treatment, morbidity, and time lost from work are substantial (2). Increased urinary calcium excretion is an important risk factor for the development of calcium-containing kidney stones (3), which account for over 80% of all kidney stones in the United States (4). Compared with urinary calcium concentrations < 75 mg/L, the relative risk for kidney stones increased from twofold in subjects with urinary calcium 100 to 149 mg/L to over fourfold in subjects with ≥ 200 mg/L (5). Thus, even moderate increases in urinary calcium excretion may substantially increase the risk for stone formation.
Menopause is associated with an increase in urinary calcium excretion (6,7⇓), which may increase the risk for calcium-containing stone formation. In subjects matched for serum total and ionized calcium levels, urinary calcium excretion was 50% higher (P < 0.001) in postmenopausal women compared with premenopausal women (6). Thus, the onset of menopause may increase urinary calcium excretion and the risk for kidney stone formation. In contrast, postmenopausal hormone (PMH) use has been shown to decrease fasting urinary calcium by 50% compared with baseline in postmenopausal women (8). It is currently not known whether menopause and PMH use are independent risk factors for kidney stone formation in women. The aim of this study was to prospectively study the independent association among menopause, PMH use, and risk of incident kidney stones in women.
Materials and Methods
Subjects
The Nurses’ Health Study (NHS) is a prospective study of 121,700 female registered nurses who were 30 to 55 yr of age and residing in 1 of 11 US states when they who completed and returned the initial questionnaire in 1976. The cohort is followed by means of biennial questionnaires, which inquire about a wide range of lifestyle factors and medical conditions. We limited the study to the women who provided information on lifetime history of kidney stones on the 1992 questionnaire or a subsequent biennial questionnaire and who completed at least one of six dietary questionnaires since 1980. Subjects who reported a first stone before 1980 (when dietary information was first collected) were excluded. The majority (>95%) of the NHS cohort was white; 1.2% of the cohort was African-American, 0.7% was Asian, and 0.5% was Hispanic.
Menopause and PMH Use
Menopausal status was first determined with the initial questionnaire completed by the NHS subjects in 1976 and was subsequently updated with each biennial questionnaire. The questionnaires specifically asked whether the subject’s menstrual periods had ceased permanently, and if so, at what age and for what reason (e.g. natural, surgical, radiation, etc.). If menopause was due to surgery, the subject was asked to report the number of ovaries removed. Self-reported age at menopause and type of menopause has been validated in the NHS and shown to be highly accurate (9).
Women were defined as postmenopausal from the self-reported time of natural menopause or bilateral oophorectomy. We excluded women with indeterminable age at onset of menopause for the analysis of menopause and risk of incident kidney stones. Subjects with indeterminable menopausal status were women who underwent hysterectomy without bilateral oophorectomy, women with incomplete data on the extent of pelvic surgery, women using exogenous estrogen before the onset of menopause, women with permanent cessation of menses due to radiation therapy, and women who did not report the age of onset of menopause.
The analysis of PMH use and incident kidney stones was limited to postmenopausal women. In these analyses, women with indeterminable menopausal status were considered menopausal when they reached the age at which natural menopause occurred in 90% of the cohort (54 yr for smokers and 56 yr for nonsmokers).
In 1976, the Nurses Health Study participants were first queried on the use of PMH therapy, including duration of use. Information on the types of PMH use was subsequently collected beginning in 1978. All of this information was updated with each biennial questionnaire. Duration of PMH use was calculated from information provided on the questionnaires. Each biennial questionnaire asked whether the subject was currently using PMH (within the last month) and the number of months PMH had been used during the past 24 mo since completion of the last biennial questionnaire. Women with missing data on PMH use for a given 2-yr period were assigned to a missing category for that time period.
Kidney Stones
Beginning in 1992, the biennial questionnaire inquired whether the participants had ever been diagnosed with a kidney stone, and new reports of kidney stones were updated with each subsequent questionnaire. A supplementary questionnaire was sent to subjects who reported kidney stones to confirm the diagnosis and ascertain the date of occurrence and other relevant medical information. Subjects had to report associated pain or hematuria on the supplementary questionnaire to be considered cases. To confirm the validity of the participants’ reports, we obtained the medical records from a random sample of 90 women who reported a kidney stone. The records confirmed the diagnosis for all but one woman (98%) (10).
Assessment of Diet
Participants were asked in 1980, 1984, 1986, 1990, and 1994 to complete semiquantitative food-frequency questionnaires on which they reported the average use of specified foods and beverages during the past year. The 1980 dietary questionnaire contained a list of 61 items, and subsequent questionnaires contained approximately 130 items. Nutrient intake was computed from the reported frequency of consumption of each specified unit of food or beverage and from published data on the nutrient content of the specified portions (11). The reproducibility and validity of the questionnaires in this cohort have been previously published (11).
Nutrient values were adjusted for total energy intake using a linear regression model with total caloric intake as the independent variable and absolute nutrient intake as the dependent variable (11). Total energy intake for a person tends to be fixed within a narrow range. Thus, variations in nutrient intake are largely due to changes in composition of the diet and not the total amount of food consumed. Energy-adjusted values reflect the nutrient composition of the diet independent of the total amount of food consumed.
Statistical Analyses
The study design was prospective, and information on menopause and PMH use was collected before the onset of the kidney stone. Person-years of follow-up were calculated from the date of the return of the 1980 questionnaire to the date of the first kidney stone, or May 31, 1998. Person-time was allocated to menopausal status or PMH use status according to the 1980 questionnaire and then updated with each subsequent biennial questionnaire. If information on menopausal status, PMH use, or diet was missing at the start of a time period, the subjects were assigned to the missing category for that time period.
Categorical variables were compared using the χ2 test, and continuous variables were compared using the t test. We adjusted for age by using 1-yr intervals and computed age-adjusted risks for incident kidney stones according to menopausal status (natural or surgical) with premenopausal status as the reference group. Postmenopausal women were then stratified by type of menopause (surgical versus natural) and age-adjusted risks for incident kidney stones and PMH use (current or past) and duration of current or past PMH use (<1 yr, 1 to 1.9 yr, 2 to 4.9 yr, 5 to 9.9 yr, ≥10 yr) were calculated with never-use as the reference group.
We used pooled logistic regression to determine the independent association between menopausal status and risk for incident kidney stones while simultaneously adjusting for multiple risk factors (12,13⇓). The following covariates were included in the multivariate pooled logistic regression models: age (1-yr intervals), body mass index (five categories), presence of hypertension (yes/no), supplemental calcium (0 mg/d, 1 to 100 mg/d, 101 to 500 mg/d, and >500 mg/d), alcohol consumption (eight categories), and quintiles of dietary intake of calcium, vitamin B6, animal protein, potassium, sodium, sucrose, magnesium, and total fluid. In the analysis of PMH use and risk of incident kidney stones, we controlled for age at menopause (1-yr intervals) in addition to the demographic and nutritional covariates included in the model of menopausal status and risk of incident kidney stones. All P values are two-tailed and 95% confidence intervals (CI) were calculated for all relative risks (RR).
Results
Overall, there were 91,731 women who did not die before 1980 and provided information on history of kidney stone disease and answered at least one of six dietary questionnaires since 1980. After excluding women with indeterminable menopausal status or kidney stones before 1980, there were 70,174 eligible for the analysis of menopause and risk of incident kidney stones. During 18 yr of follow-up (737,669 person-years), there were 747 incident kidney stones confirmed. The demographic and dietary characteristics of the cohort by their reported menopausal status in 1990 are shown in Table 1. There were no substantial differences in the mean body mass index between the two groups, but postmenopausal women had slightly higher intake of supplemental calcium. No substantial differences were noted in the mean dietary intake of calcium, protein, sodium, sucrose, magnesium, or fluid between the two groups.
Table 1. Demographic and dietary characteristics of the cohort by reported menopausal status in 1990a
The crude age and multivariate RR for menopausal status and risk of incident kidney stones are shown in Table 2. In the unadjusted model, menopausal status was associated with an increased risk of kidney stones (RR, 1.45; 95% CI, 1.24 to 1.68). However, we noted no association between menopause and incident kidney stones in the age-adjusted (RR, 1.07; 95% CI, 0.85 to 1.34) or multivariate models (RR, 1.12; 95% CI, 0.89 to 1.41). When we examined the type of menopause, surgical menopause was associated with a significantly increased risk in the age-adjusted model (RR, 1.37; 95% CI, 1.05 to 1.77), whereas natural menopause was not (RR, 0.95; 95% CI, 0.75 to 1.20). Further adjustment for dietary factors, fluid intake, history of hypertension, and body mass index revealed similar results for both surgical menopause (RR, 1.39; 95% CI, 1.07 to 1.81) and natural menopause (RR, 1.01; 95% CI, 0.79 to 1.28).
Table 2. Age and multivariate adjusted relative risk (RR) for incident kidney stones by menopausal statusa
We also examined the association between age at menopause and risk of incident kidney stones. After adjustment for age, dietary factors, fluid intake, history of hypertension, and body mass index, no association between age at menopause and risk of incident kidney stones was noted in women with natural menopause (RR, 1.00; 95% CI, 0.98 to 1.02) or surgical menopause (RR, 0.99; 95% CI, 0.97 to 1.00).
No significant association between past or current PMH use and risk of incident kidney stones was noted among postmenopausal women with natural or surgical menopause in the age-adjusted or multivariate models (Tables 3 and 4). Furthermore, there was no significant association between the duration of past (data not shown) or current PMH use and the risk of incident kidney stones among postmenopausal women with either natural (Table 3) or surgical menopause (Table 4).
Table 3. Crude, age, and multivariate-adjusted RR for incident kidney stones by postmenopausal hormone usea in women with natural menopause
Table 4. Crude, age, and multivariate-adjusted RR for incident kidney stones by postmenopausal hormone usea in women with surgical menopause
Discussion
The unadjusted RR for incident kidney stones in postmenopausal women was 45% higher compared with premenopausal women. After controlling for age, we noted no association between menopause and incident kidney stones. There was, however, a 39% increase in the multivariate-adjusted risk with surgical menopause. Menopause has been associated with an increase in urinary calcium excretion due to estrogen deficiency. In women matched for serum ionized calcium, Nordin et al. (6) found that fasting urinary calcium excretion was 50% higher (P < 0.001) in postmenopausal women compared with premenopausal women. Similar differences in urinary calcium excretion between premenopausal and postmenopausal women were found in an unmatched study (7). Estrogen deficiency increases the sensitivity of bone to parathyroid hormone, leading to a net increase in bone resorption and increased urinary calcium excretion (14). The menopausal changes in urinary calcium excretion may prevent hypercalcemia, but they could theoretically also yield an increased risk for bone loss and possibly kidney stone formation. However, with natural menopause, the loss of estrogen leading to increased calcium excretion may be too gradual to noticeably elevate the risk of kidney stone formation. Moreover, natural menopause may be associated with beneficial changes in other urinary factors, such as uric acid, leading to an overall null effect on risk of kidney stones, but data in this area are very limited.
In women with surgical menopause, loss of ovarian estrogen production is sudden and complete (15). The sudden loss of estrogen is compounded by a substantial decrease in circulating testosterone levels (15). Even after menopause, the ovaries normally continue to produce testosterone, and removal of both ovaries after menopause decreases circulating testosterone by 40 to 50% (16). Peripheral tissues metabolize testosterone to estradiol, thus the continued production of testosterone in the ovaries after menopause probably helps to mitigate bone resorption in postmenopausal women (17,18⇓). The sudden loss of ovarian production of estrogens and androgens in women with surgical menopause leads to a more rapid bone loss compared with women with natural menopause (19–21⇓⇓). Genant et al. (24) noted a 7 to 9% decrease in vertebral bone mineral density during the first 2 yr after bilateral oophorectomy, a substantially higher rate than the 1 to 2% loss per year noted after natural menopause (22). Although several studies have noted increased urinary excretion of bone resorption markers after surgical menopause (23,24⇓), to our knowledge, no published studies have shown that women with surgical menopause have higher urinary calcium excretion compared with women with natural menopause. However, the higher rate of bone loss after surgical menopause likely reflects a higher urinary calcium excretion and could account for the increased risk of incident kidney stones we noted in this study. Data on the association between surgical menopause and the excretion of other urinary factors that may influence kidney stone formation such as oxalate, uric acid, or citrate in addition to calcium are scant. More studies are needed to determine whether the type of menopause (natural versus surgical) influences the excretion of these urinary factors.
The increase in urinary calcium excretion associated with menopause may be inhibited by estrogen use. After three weeks of orally administered estrogen in postmenopausal women, fasting urinary calcium decreased by 50% compared to baseline (8). In addition, animal studies have shown that estrogen influences the urinary excretion of oxalate. In oophorectomized rats, the administration of estrogen decreased urinary oxalate excretion by over 50% (25). However, we found no association between past or current PMH use or duration of past or current PMH use and incident kidney stones in postmenopausal women. This lack of an association may be partly explained by an increase in urinary uric acid excretion (26,27⇓), which may promote calcium stone formation (28). After the daily administration of oral conjugated estrogen (1.25 mg) for 4 d while being maintained on a purine-restricted diet, urinary uric acid excretion increased by 23% (P < 0.05) compared with baseline in seven postmenopausal women (26). Similar results were also noted in a study of 22 adult transsexual men treated with estrogen where mean uric acid excretion increased by 28% (P < 0.001) (27). Thus the beneficial effects of estrogen use on urinary calcium excretion may be offset by an increase in urinary uric acid excretion leading to an overall null effect on kidney stone formation.
In our study, information on menopause and PMH use was collected before the passage of a kidney stone. In addition, the high level of education and the strong health interest of the subjects are major strengths of this cohort and favorably influence the quality and accuracy of information reported on the questionnaires. However, one cannot determine when the kidney stones were formed, so it is possible that surgical menopause may lead to the movement of preexisting stones, which then become symptomatic, rather than the new formation of stones. It is also possible that the positive association we found between surgical menopause and incident kidney stones could be due to chance.
Self-reported age at onset of menopause may be subject to error (29,30⇓); however, we excluded all women with indeterminate age at menopause (hysterectomy without bilateral oophorectomy, incomplete data on the extent of pelvic surgery, exogenous estrogen use before the onset of menopause, and permanent cessation of menses due to radiation therapy) from all analyses. Age at onset of menopause is, however, highly accurate in women with bilateral oophorectomy (9). Moreover, we noted no association between age at onset of menopause and risk of incident kidney stones among women with either natural or surgical menopause. In addition, the association between PMH use and incident kidney stones did not change substantially when we controlled for age at menopause in women with natural or surgical menopause. It is thus unlikely that the results of this study were biased by possible misclassification of age at onset of menopause.
In summary, we found no association among natural menopause and PMH use and incident kidney stones. However, there may be a modestly increased risk for incident kidney stone disease with surgical menopause. This increased risk may be due to higher excretion rates of urinary calcium after surgical menopause.
Acknowledgments
Dr. Holly Mattix was an American Kidney Fund Clinical Scientist in Nephrology Fellow during completion of this project. This work was supported by grants from the National Institute of Health (DK59583 and CA87969).
- © 2003 American Society of Nephrology
References
