- Top of page
- Conflict of interest/financial disclosure
Persistent elevation of urinary albumin excretion (UAE) is an independent predictor for cardiovascular disease (CVD) in many prospective studies.[1-3]
Elevation of UAE (albuminuria) is traditionally defined as levels of urinary albumin creatinine ratio (UACR) above 30 mg/g. Recent meta-analysis studies showed that higher levels of albumin excretion within normal range (high normal albuminuria, UACR 10–30 mg/g), under the definition for albuminuria, are associated with increased cardiovascular mortality additive to conventional risk factors.[5, 6]
In subjects without kidney disease, elevation of UAE is recognized as a signal of systemic endothelial dysfunction in kidney. However, the causes of UAE elevation or pathophysiology by which elevated UAE increases cardiovascular risk are unknown.
Vitamin D and parathyroid hormone (PTH) are both responsible for maintaining extracellular calcium homoeostasis, and for that purpose, both vitamin D and PTH operate within a well-controlled feedback system. For example, in a circumstance of low-serum calcium level, PTH release is triggered and PTH increases calcium reabsorption at skeletal sites at the expense of an increased risk of fracture. PTH also increases active vitamin D level by 1-alpha hydroxylase, and in turn, active vitamin D increases intestinal absorption of calcium. However, in addition to these traditional roles, many prospective studies showed that low levels of vitamin D increase CVD risk independent of other established risk factors. Although the evidence was far less than for vitamin D, a community-based prospective study demonstrated that elevated PTH level predicted cardiovascular mortality.
Low levels of vitamin D and high levels of PTH may be related to the elevation of UAE and the pathophysiology that elevated UAE increases CVD risk. Chronic vitamin D deficiency and secondary hyperparathyroidism are associated with increased insulin resistance,[10, 11] activation of renin–angiotensin–aldosterone (RAA) system[12, 13] and endothelial dysfunction.[14, 15] Also, both decreased vitamin D level and increased PTH level are associated with hypertension,[16-18] obesity[19, 20] and metabolic syndrome,[21, 22] all of which are risk factors for elevated UAE.
With vitamin D, cross-sectional studies showed that decreased vitamin D levels result in higher relative risk of UAE level above UACR 30 mg/g. However, to the best of our knowledge, there is no study that investigated the relationship between PTH level and UAE. Therefore, we examined the relationship between PTH, vitamin D and UAE in persons older than 50. Also, we checked the association between these two hormones to high normal albuminuria within normal range of UAE (UACR 10–30 mg/g), which has been noticed as a novel risk factor for CVD risk.[10, 11]
- Top of page
- Conflict of interest/financial disclosure
Proportions of persons with vitamin D <10 and 30 ng/ml (suggested definition of vitamin D deficiency and insufficiency) were 4·37% and 95·27%, respectively, in Korean adults above 50 years of age. Percentages of subjects with PTH level <10 and 60 pg/ml (normal range of PTH: 10–60 pg/ml) were 0·04% and 55·44%, and proportions of persons with UAE: optimum, high normal albuminuria, high albuminuria and very high albuminuria were 74·78%, 13·89%, 9·50% and 1·84%, respectively.
In Table 1, variables related to elevated UAE were age (P < 0·001), waist circumference (P < 0·001), diabetes (P < 0·001), hypertension (P < 0·001) and PTH quartiles (P = 0·001), whereas sex (P = 0·949), BMI (P = 0·429), lipid status (P = 0·470), eGFR (P = 0·246), smoking (P = 0·174), physical activity (P = 0·117), alcohol use (P = 0·852) and vitamin D (P = 0·899) did not show significant associations.
Table 1. Characteristics of study participants by elevated UAE
|Variables||Optimal UAE (UACR <10 mg/g)||Elevated UAE (UACR ≥10 mg/g)||P valuea|
|Unweighted, n||2169||728|| |
|Male||46·9 (1·1)||47·1 (2·4)||0·949|
|Female||53·1 (1·1)||52·9 (2·4)|
|50–59||52·8 (1·4)||37·3 (2·4)||<0·001|
|60–69||27·6 (1·1)||31·2 (2·0)|
|70–79||16·7 (0·9)||24·3 (1·8)|
|Above 80||3·0 (0·4)||7·3 (1·0)|
|Normal||37·6 (1·5)||35·5 (2·2)||0·429|
|Overweight||28·3 (1·1)||26·9 (2·1)|
|Obesity||34·1 (1·4)||37·6 (2·2)|
|Waist circumference (cm)|
|Normal||67·5 (1·5)||59·0 (2·2)||<0·001|
|Abdominal obesity||32·5 (1·5)||41·0 (2·2)|
|Normal||66·1 (1·3)||48·7 (2·1)||<0·001|
|Prediabetes||27·0 (1·2)||29·7 (2·0)|
|Diabetes||6·9 (0·7)||21·6 (1·9)|
|Normal||44·1 (1·4)||26·4 (2·1)||<0·001|
|Prehypertension||40·7 (1·3)||43·3 (1·2)|
|Hypertension||15·3 (1·0)||19·1 (1·1)|
|Normal||56·7 (1·4)||58·2 (2·3)||0·470|
|Borderline||32·9 (1·3)||30·3 (2·2)|
|High||10·3 (0·9)||11·6 (1·5)|
|Quartiles of eGFR (ml/min)|
|1st (≥95·8)||27·2 (1·5)||24·6 (1·9)||0·246|
|2nd (85·7–95·8)||25·5 (1·3)||24·4 (2·0)|
|3rd (76·3–85·7)||25·8 (1·3)||25·2 (2·0)|
|4th (<76·3)||21·5 (1·2)||25·8 (2·0)|
|None||55·3 (1·2)||53·2 (2·4)||0·174|
|Former||25·5 (1·2)||23·6 (2·1)|
|Current||19·0 (1·1)||23·2 (1·9)|
|None||59·3 (1·2)||59·8 (2·3)||0·852|
|Regular drinking||40·7 (1·2)||40·2 (2·3)|
|None||47·7 (1·4)||53·1 (2·6)||0·117|
|Mild||31·8 (1·3)||31·1 (2·2)|
|Moderate||7·5 (0·7)||5·6 (1·0)|
|Vigorous||13·0 (1·1)||10·2 (1·4)|
|Vitamin D (ng/ml)||18·8 (0·3)||18·9 (0·4)||0·899|
|PTH (pg/ml)||64·8 (0·7)||71·5 (1·5)||<0·001|
When UACR was examined as a continuous outcome variable, PTH level was positively associated with UACR (in fully adjusted model: beta coefficient = 0·336, P = 0·003; data not shown). However, vitamin D level did not show a significant association with UACR (in fully adjusted model: beta coefficient = 0·001, P = 0·852; data not shown).
A stepwise increase in proportion of elevated UAE (21·8%, 23·2%, 23·2%, 31·8%; P for trend = 0·002) was observed with increasing quartiles of PTH (Fig. 1). In logistic regression analysis, after adjusting for potential confounders (sex, age, BMI, waist circumference, diabetes, hypertension, lipid, smoking, alcohol, physical activity status), a similar stepwise increase in odds ratio was observed with increasing PTH quartiles (Table 2, model 1, P for trend = 0·001). This association remained significant even after further adjustment was made for eGFR and vitamin D (Table 2, model 2, P for trend = 0·001; model 3, P for trend = 0·001). Subjects in the highest quartile of PTH levels had an increased likelihood of elevated UAE compared to those with lowest PTH levels in adjusted model (Table 2, model 3, odds ratio (OR) 1·72, 95% confidence interval (CI) 1·26–2·33, P for trend = 0·001). In contrast, there were no significant relationships between vitamin D and elevated UAE in any models (Table 3).
Table 2. Adjusted odds ratios of elevated UAE by quartiles of PTH
| ||Quartiles of PTH (pg/ml)||P for trenda|
|1st (<50·2)||2nd (50·2–62·1)||3rd (62·1–77·9)||4th (≥77·9)|
|Model 1||1·00 (reference)||1·07 (0·77–1·49)||1·11 (0·82–1·48)||1·67 (1·24–2·30)||0·001|
|Model 2||1·00 (reference)||1·07 (0·77–1·48)||1·10 (0·82–1·48)||1·69 (1·25–2·30)||0·001|
|Model 3||1·00 (reference)||1·06 (0·76–1·48)||1·11 (0·83–1·48)||1·72 (1·26–2·33)||0·001|
|Model 4||1·00 (reference)||1·28 (0·82–2·00)||1·47 (0·95–2·28)||1·94 (1·31–2·87)||0·001|
Table 3. Adjusted odds ratios of elevated UAE by quartiles of vitamin D levels
| ||Quartiles of vitamin D (ng/ml)||P for trenda|
|1st (<14·1)||2nd (14·1–18·1)||3rd (18·1–22·7)||4th (≥22·7)|
|Model 1||0·88 (0·63–1·23)||1·08 (0·80–1·46)||0·88 (0·62–1·26)||1·00 (reference)||0·700|
|Model 2||0·88 (0·62–1·26)||1·08 (0·80–1·46)||0·88 (0·62–1·26)||1·00 (reference)||0·695|
|Model 3||0·99 (0·70–1·39)||1·19 (0·88–1·61)||0·94 (0·66–1·34)||1·00 (reference)||0·708|
|Model 4||1·08 (0·72–1·63)||1·33 (0·90–1·95)||1·10 (0·74–1·64)||1·00 (reference)||0·485|
The association of PTH level with high normal albuminuria, assessed by excluding participants with high and very high albuminuria from analysis, showed the same pattern (Table 2, model 4, P for trend = 0·001). In participants with normal kidney function but vitamin D insufficiency or deficiency (90·23% of the participants), there was also a significant relationship between PTH and elevated UAE after adjusting for all potential confounders (model 3: P for trend = 0·002, model 4: P for trend = 0·002; data not shown).
- Top of page
- Conflict of interest/financial disclosure
The present study is the first report that shows high PTH levels are associated with an increased prevalence of elevated UAE in general population aged 50 and above. This association persisted after adjusting for confounders such as age, sex, obesity, abdominal obesity, diabetes, hypertension, dyslipidaemia, smoking, alcohol, physical activity and kidney function. Moreover, a similar trend between PTH levels and UAE was observed within normal range of albuminuria (UACR <30 mg/g). These findings suggest that increased PTH level may be a predictor for UAE elevation.
Our findings could not be interpreted as mild elevation of PTH level is a risk factor for deterioration in renal function. Although kidney failure is one of the long-term complications of primary hyperparathyroidism (PHPT), the evidence is inadequate that mild PHPT is associated with the development of renal insufficiency. In recent epidemiologic study, however, it was observed that patients with mild PHPT had high risk of renal failure. During the period of medical observation (median follow-up of 2·9 year) for mild asymptomatic PHPT cohort, the relative risk of developing renal failure was 19·30 compared with the healthy controls. The result of this study could reflect our findings that high PTH levels are associated with UAE elevation.
Because of the limitations of observational studies, we could not know the underlying mechanisms that explain the associations between PTH level and UAE. However, there is some research that points at an association between those. Above all, evidence has accumulated that aldosterone, a major hormone of the RAA system, is related to PTH. Several animal and human studies showed that PTH stimulated aldosterone secretion from adrenal glands directly or indirectly and aldosterone levels were decreased after surgical parathyroidectomy for primary hyperparathyroidism.[13, 28, 29] Aldosterone had no effect on intraglomerular haemodynamics, but could cause inflammation and fibrosis within kidney. Aldosterone may be related to general vascular function, and many studies showed that increased levels of aldosterone are associated with endothelial dysfunction.[31, 32] Although long-term effects on renal outcomes were largely unknown, many experimental studies showed that mineralocorticoid receptor antagonist such as spironolactone and eplerenone reduced albuminuria regardless of diabetes. Therefore, increased secretion of aldosterone due to elevated PTH may modulate the development of albuminuria.
In addition, the link between elevated PTH and elevated UAE may be explained by hypertension[18, 34] and metabolic syndrome, established risk factors for albuminuria and possibly related to PTH.
In the present study, there were no significant relationships between vitamin D levels and UAE. The results were inconsistent with previous studies that showed a significant relationship between vitamin D concentrations and elevated UAE (UACR ≥30 mg/g). De Boer et al. analysed the data from the US Third National Health and Nutritional Examination Survey and showed that the group with the lowest quartile of vitamin D (3·5–17·6 ng/ml) had a higher risk of microalbuminuria (relative risk 1·28, 95% confidence intervals 1·03–1·59) than the group in the highest quartile (32·0–97·6 ng/ml). In Korea, the prevalence of vitamin D deficiency has been reported to be very high, and in this study, most participants (95·3% of total participants) were categorized into vitamin D insufficiency or deficiency. Therefore, a skewed distribution of vitamin D levels and lack of adequate number of sufficient level of vitamin D may affect the results. Moreover, the relationship between vitamin D and albuminuria might be mediated by PTH level. Namely, secondarily elevated PTH level due to decreased vitamin D might be a cause of development of albuminuria.
As expected, in the present study, there was a significant but weak negative correlation between vitamin D level and PTH (Pearson correlation, r = −0·19, P < 0·001). This reflects the fact that PTH secretion is stimulated by decreased vitamin D level. Although we could not confirm the correlation between vitamin D and UAE, we verified the relationship of PTH and UAE in participants with normal kidney function and vitamin D insufficiency or deficiency.
This study has several limitations. Subjects were limited to adults aged 50 and above, and therefore, the result of the study could not be applicable to younger adults. With vitamin D levels, most subjects in this study were in a state of vitamin D deficiency or insufficiency, and the narrow range of vitamin D concentrations might affect the result. We did not evaluate the time of measurement of vitamin D levels and use of vitamin D supplements. However, because all participants in KNHANES were randomly selected through all seasons and effect of use of vitamin supplements may be reflected by serum vitamin D levels, these factors are not likely to confound the association we found. With UAE, we did not evaluate information about antihypertensive medication such as angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, although such medications are not known to affect serum vitamin D or PTH levels. The use of a single, untimed urine sample might affect the accuracy of UACR although this method has been recommended. The cross-sectional nature of this study does not allow the inference of casualty or temporal relationship between high PTH level and elevated UAE, and there may be unknown confounders between those two factors.
In conclusion, we found that elevated PTH levels are associated with elevated UAE in Korean adults aged 50 and above after adjusting other confounders. Our study calls for additional studies to confirm the relationship of PTH with elevated UAE, extending to whole age group and various populations. Prospective studies are also needed to evaluate the aetiologic roles of vitamin D and PTH in the development of albuminuria.