Preventive Medicine 60 (2014) 48–54

Contents lists available at ScienceDirect

Preventive Medicine
journal homepage: www.elsevier.com/locate/ypmed

Associations of proteinuria and the estimated glomerular filtration rate

with incident hypertension in young to middle-aged
Japanese males☆,☆☆
Naoki Okumura a, Takahisa Kondo a,b,⁎, Kunihiro Matsushita a,c, Shigeki Osugi a, Keiko Shimokata a,
Kyoko Matsudaira a, Kentaro Yamashita a, Kengo Maeda a, Toyoaki Murohara a
a
Department of Cardiology, Nagoya University Graduate School of Medicine, Nagoya, Japan
b
Department of Advanced Medicine in Cardiopulmonary Disease, Nagoya University Graduate School of Medicine, Nagoya, Japan
c
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, USA

a r t i c l e i n f o a b s t r a c t

Available online 14 December 2013 Objective: To investigate the independent associations of proteinuria and the estimated glomerular filtration
rate (eGFR) with incident hypertension.
Keywords: Methods: We investigated 29,181 Japanese males 18–59 years old without hypertension in 2000 and exam-
Dipstick proteinuria ined whether proteinuria and the eGFR predicted incident hypertension independently over 10 years. Incident
eGFR
hypertension was defined as a newly detected blood pressure of ≥140/90 mm Hg and/or the initiation of anti-
Hypertension
Japanese males
hypertensive drugs. Proteinuria and the eGFR were categorized as dipstick negative (reference), trace or ≥1+
Prospective cohort and ≥ 60 (reference), 50–59.9 or b 50 ml/min/1.73 m2, respectively. Cox proportional hazards models were
used to estimate the hazard ratios (HRs) of incident hypertension.
Results: At baseline, 236 (0.8%) and 477 (1.6%) participants had trace and ≥1+ dipstick proteinuria, while 1416
(4.9%) and 129 (0.4%) participants had an eGFR of 50–59.9 and b 50 ml/min/1.73 m2, respectively. The adjusted
HRs were significant for proteinuria ≥1+ (HRs 1.20, 95% CI: 1.06–1.35) and an eGFR of b50 ml/min/1.73 m2
(1.29, 1.03–1.61). When two non-referent categories were combined (dipstick ≥ trace vs. negative and
eGFR b 60 vs. ≥60 ml/min/1.73 m2), the association was more significant for proteinuria (1.15, 1.04–1.27) than
for eGFR (0.99, 0.92–1.07).
Conclusions: Proteinuria and a reduced eGFR are independently associated with future hypertension in young to
middle-aged Japanese males.
© 2013 The Authors. Published by Elsevier Inc. All rights reserved.

Introduction with hypertension. However, approximately 90% of adults with hyper-

Hypertension is a highly prevalent disorder that affects more than
one quarter of the population worldwide (Kearney et al., 2005) and is
a major risk factor for stroke, cardiovascular disease and end-stage
renal disease (Arima et al., 2003; Gueyffier, 2003; Klag et al., 1996).
Hypertension is even more prevalent in Japan, with an estimated prev-
alence of ~40% (Kubo et al., 2008). Several factors, such as high sodium
intake (1988), obesity (Fox et al., 2007) and physical inactivity
(Dickinson et al., 2006), have been identified to be highly associated
☆ This is an open-access article distributed under the terms of the Creative Commons
Attribution-NonCommercial-No Derivative Works License, which permits non-
commercial use, distribution, and reproduction in any medium, provided the original
author and source are credited.
☆☆ Funding sources: None.
⁎ Corresponding author at: Department of Advanced Medicine in Cardiopulmonary
Disease, Nagoya University Graduate School of Medicine, 65, Tsurumai-cho, Showa-ku,
Nagoya 466-8550, Japan. Fax: +81 52 744 2138.
E-mail address: takahisa@med.nagoya-u.ac.jp (T. Kondo).
tension are considered to have essential hypertension, a condition with-
out an overt primary cause (Anderson et al., 1994; Carretero and Oparil,
2000; Nishikawa et al., 2007; Rossi et al., 2006).
The kidney plays a significant role in the regulation of blood pressure
(BP) by controlling blood volume, the levels of electrolytes and the sym-
pathetic nervous system and hormonal systems, such as the renin–an-
giotensin–aldosterone system (Brewster and Perazella, 2004; Komukai
et al., 2010). Therefore, kidney damage and dysfunction, such as pro-
teinuria and a reduced glomerular filtration rate (GFR), have attracted
attention as predictors of hypertension (Brantsma et al., 2006; Forman
et al., 2008; Gerber et al., 2006; Gueyffier, 2003; Jessani et al., 2012;
Kestenbaum et al., 2008; Palatini et al., 2005; Takase et al., 2012;
Wang et al., 2005). However, to the best of our knowledge, only a few
studies have investigated the associations of proteinuria and GFR simul-
taneously with the development of hypertension, and the results were
not consistent (Kestenbaum et al., 2008; Takase et al., 2012), leaving
uncertainty regarding the respective contributions of these factors to
the development of hypertension. Asians, a racial/ethnic group with a

0091-7435/$ – see front matter © 2013 The Authors. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ypmed.2013.12.009
 N. Okumura et al. / Preventive Medicine 60 (2014) 48–54 49

high prevalence of hypertension (Kearney et al., 2005; Kubo et al.,
2008), are particularly understudied regarding this issue.
Therefore, the purpose of the present study was to investigate the in-
dependent association of the presence of proteinuria and a reduced
eGFR with incident hypertension in a prospective cohort study of
young to middle-aged Japanese males with annual BP evaluation.
Materials & methods
Study population
The study subjects included Japanese males who underwent annual medical
checkups at their workplaces, all of which were blue-chip companies in Japan
(Kondo et al., 2013; Yamashita et al., 2012). Japanese males 16–59 years of
age (n = 33,914) were recruited in 2000. We excluded participants with
preexisting hypertension (systolic BP ≥ 140 mm Hg, diastolic BP ≥ 90 mm Hg
or the use of antihypertensive drugs; n = 4688 at baseline examination) and
excluded participants aged b18 years old (n = 45), with a final sample of
29,181 participants.
Data collection and measurements
Annual medical checkups including blood test and dipstick urine test were
conducted through 2010 or until retirement at around 60 years of age. All par-
ticipants were individually interviewed using a structured questionnaire in the
baseline and annual follow-up surveys. The following information was recorded
by trained observers: smoking status, alcohol intake, medical history and
medications. The smoking status and alcohol intake were classified as current
vs. former/never. Weight and height were measured while the subject was
wearing light clothing without shoes. The body mass index (BMI) was com-
puted as the weight in kilograms divided by the square of the height in meters.
Urine and blood samples were obtained in the morning with overnight fasting.
A urinalysis for proteinuria was conducted with Uropaper III (Eiken Chemical
Co., Ltd., Tokyo, Japan), and the results were measured using a US-2100 Auto-
mated Urine Analyzer (trace (±) corresponds to proteinuria ≥15 mg/dl, 1+ to
≥30 mg/dl, 2+ to ≥100 mg/dl, 3+ to ≥300 mg/dl and 4+ to ≥1000 mg/dl).
The blood analyses were conducted at a single laboratory. The GFR was estimated
using the three-variable equation proposed by the Japanese Society of Nephrology
(eGFR [ml/min/1.73 m2] = 194 × serum creatinine−1.094 × age−0.287 × 0.739
[if female]) (Matsuo et al., 2009). In this study, the proteinuria using a dipstick
and eGFR were measured at baseline (2000). Diabetes mellitus was defined as a
concentration of serum fasting glucose of ≥126 mg/dl or the use of glucose-
lowering medications.
Blood pressure assessment and incident hypertension
BP was measured annually with the participant in the sitting position after
5 min of rest using an automated sphygmomanometer (BP-203IIIB; Colin
Corporation, Tokyo, Japan). The BP was measured two times at intervals of
1 min on the right arm, and the average value was calculated as the baseline
BP. When a subject had frequent premature contractions or atrial fibrillation,
trained nurses confirmed the BP using a conventional mercury sphygmoma-
nometer. Incident hypertension was defined as a newly detected BP of ≥140/
90 mm Hg and/or the initiation of antihypertensive drugs during follow-up.
Statistical analysis
All analyses were performed using the STATA software program version 11
(Stata Corp. College Station, TX, USA). Continuous variables were presented as
the medians (interquartile ranges), and differences between the two/three
groups were evaluated using the Wilcoxon test/Kruskal–Wallis analysis because
not all continuous variables were normally distributed. Categorical variables
were presented as numbers (percentages), and comparisons across the groups
were made using the chi-square test. Survival curves were calculated according
to the Kaplan–Meier method and compared using the log-rank test. Cox propor-
tional hazards models were used to estimate the hazard ratios (HRs) of incident
hypertension according to the level of proteinuria and eGFR adjusted for
age (continuous), BMI (continuous), serum total cholesterol (continuous),
serum uric acid (continuous), diabetes mellitus (category), current smoking
(category), current alcohol intake (category) and proteinuria (category) or
eGFR (continuous), as appropriate. We used time from baseline as time variable
in the Cox models.
We assessed the independent associations of proteinuria and eGFR with in-
cident hypertension after dividing both kidney measures into three categories
(dipstick proteinuria: negative, trace and ≥1+; and eGFR: b50, 50–59.9 and
≥ 60 ml/min/1.73 m2). A dipstick negative status and eGFR of ≥ 60 ml/min/
1.73 m2 were used as reference groups. Due to the limited number of individ-
uals with an eGFR of b 50 ml/min/1.73 m2 and dipstick proteinuria ≥1+, we
also tested dichotomized proteinuria (positive [trace, and ≥1+] vs. negative)

Table 1
Baseline characteristics (dipstick negative, ±, ≥+ and eGFR ≥ 60, 50–59.9, b50 ml/min/1.73 m2): Aichi, Japan, 2000.

Total Proteinuria p Value eGFR (ml/min/1.73 m2) p Value

(n = 29,181)
Negative Trace (±) (≥+) ≥60 50–59.9 b50
(n = 28,468) (n = 236) (n = 477) (n = 27,636) (n = 1416) (n = 129)

Age (years old) 35 (30–40) 36 (31–43) 39 (31–50) 41 (33–51) b0.001 36 (30–42) 46 (42–51) 52 (48–57) b0.001
Body mass index (kg/m2) 22.3 22.4 22.7 23.3 b0.001 22.3 23.6 23.6 b0.001
(20.6–24.3) (20.6–24.3) (20.8–25.1) (21.0–25.7) (20.6–24.3) (22.0–25.3) (21.8–25.0)
Systolic blood pressure (mm Hg) 114 114 117 118 b0.001 114 116 120 b0.001
(106–121) (106–122) (108–126) (110–128) (106–122) (108–124) (110–129)
Diastolic blood pressure (mm Hg) 68 69 72 74 b0.001 69 74 77 b0.001
(62–76) (63–77) (65–80) (66–81) (63–76) (66–81) (69–84)
Total cholesterol (mg/dl) 190 190 194 199 b0.001 189 206 210 b0.001
(169–214) (169–214) (168–219) (173–225) (168–213) (185–230) (184–231)
HDL cholesterol (mg/dl) 55 55 54 52 b0.001 55 54 52 b0.001
(47–64) (47–65) (45–63) (45–62) (47–65) (45–63) (44–62)
Triglyceride (mg/dl) 101 100 109 113 b0.001 100 115 121 b0.001
(70–148) (70–148) (78–175) (77–179) (69–147) (81–169) (92–169)
Uric acid (mg/dl) 5.8 5.8 6.0 6.1 b0.001 5.8 6.4 6.9 b0.001
(5.1–6.6) (5.1–6.6) (5.0–6.9) (5.3–7.0) (5.1–6.5) (5.6–7.3) (6.1–7.7)
2
eGFR (ml/min/1.73 m ) 75.5 75.0 73.1 70.8 b0.001 75.5 56.9 47.1 b0.001
(69.4–82.8) (68.3–82.0) (64.7–80.5) (63.0–78.8) (69.4–82.8) (55.1–58.3) (43.2–49.3)
Fasting blood glucose (mg/dl) 89 90 94 93 b0.001 89 93 94 b0.001
(83–96) (84–96) (86–103) (86–105) (83–96) (87–101) (88–102)
Diabetes (%) 2 2 9 13 b0.001 2 4 9 b0.001
Current smoker (%) 55 55 55 52 0.524 56 42 36 b0.001
Current alcohol intake (%) 72 72 70 66 0.027 72 75 60 0.001

The data are presented as medians (interquartile ranges) for continuous variables and percentages for categorical variables.
HDL, high-density lipoprotein; eGFR, estimated glomerular filtration rate.
50 N. Okumura et al. / Preventive Medicine 60 (2014) 48–54

and eGFR (reduced [b 60] vs. preserved [≥60 ml/min/1.73 m2]), particularly in A (%)
the subgroup analysis. 75
A subgroup analysis was conducted according to the baseline BP (optimal Log-rank (overall) P<0.001
[systolic b 120 mm Hg and diastolic b 80 mm Hg] vs. normal or high-normal

Cumulative incidence of
[systolic is 120–139 mm Hg or diastolic is 80–89 mm Hg]), age (b 40 vs.
≥40 years), BMI (b 25 vs. ≥25 kg/m2), dyslipidemia (serum total cholesterol

hypertension
50
b 220 vs. ≥220 mg/dl), diabetes mellitus, current smoking and current alcohol
intake. The interaction terms between proteinuria and each subgroup were
assessed using likelihood ratio tests in the individual analyses. All reported
p values were two-sided, and p b 0.05 was considered to be statistically 25
significant.

Results
0
0 2 4 6 8 10
Baseline characteristics
Analysis time (year)
Number at risk
The baseline characteristics of the participants according to the negative 28,468 26,077 22,056 18,794 16,279 13,819
level of dipstick proteinuria and eGFR are shown in Table 1. The median trace 236 202 149 117 96 83
≥+ 477 383 272 224 176 148
age was 35 (30–40) years, and the median eGFR was 75.5 (69.4–82.8)
ml/min/1.73 m2. There were 713 participants (2.4%) with proteinuria B
(dipstick trace: n = 236, proteinuria ≥+: n = 477). The positive (%)
proteinuria group was associated with an older age, lower eGFR and 75
Log-rank (overall) P<0.001
higher prevalence of traditional cardiovascular risk factors, except for

Cumulative incidence of
current smoking. There were 1545 participants (5.3%) with a reduced
eGFR (50–59.9 ml/min/1.73 m2: n = 1416, 45–49.9 ml/min/1.73 m2:

hypertension
50
n = 118, b45 ml/min/1.73 m2: n = 11). The reduced eGFR group
was associated with an older age and higher risk profile of traditional
cardiovascular risk factors.
25

Relationship of proteinuria and eGFR with incident hypertension

During a mean follow-up period of 9.3 years (271,383 person- 0

years), 43.9% of the cohort (12,818 participants) developed hyperten- 0 2 4 6 8 10
sion. The number of incident hypertension cases determined by the Number at risk Analysis time (year)
use of antihypertensive drugs was 2.2% (292 participants) of all incident
≥60 27,636 25,373 21,504 18,369 15,949 13,565
hypertension cases. The cumulative incidence of hypertension was 50-59.9 1,416 1,202 921 729 576 467
higher in the positive proteinuria group than in the negative proteinuria <50 129 87 52 37 26 18
group in a Kaplan–Meier analysis (negative: 43.6%; trace: 54.2%; ≥1+:
Fig. 1. A. Cumulative incidence of hypertension according to the level of proteinuria esti-
61.0% in 10 years; log-rank test, p b 0.001) (Fig. 1A). Similarly, the cu-
mated using a Kaplan–Meier analysis. Footnote: The blue dotted line indicates subjects
mulative incidence of hypertension was higher in the reduced eGFR without proteinuria. The brown dotted line indicates subjects with dipstick trace protein-
group than in the preserved eGFR group (≥ 60 ml/min/1.73 m2: uria. The green dotted line indicates subjects with dipstick ≥+ proteinuria. B. Cumulative
43.4%; 50–59.9 ml/min/1.73 m2: 52.9%; b 50 ml/min/1.73 m2: 62.8% in incidence of hypertension according to the eGFR estimated using a Kaplan–Meier analysis.
10 years; log-rank test, p b 0.001) (Fig. 1B). The median duration Footnote: The blue dotted line indicates subjects with an eGFR of ≥60 ml/min/1.73 m2.
The brown dotted line indicates subjects with an eGFR of 50–59.9 ml/min/1.73 m2.
since test of proteinuria/reduced eGFR was 5 (2–10) years, and that of The green dotted line indicates subjects with an eGFR of b50 ml/min/1.73 m2. eGFR: esti-
reduced eGFR 5 (2–10) years. mated glomerular filtration rate.
The association between the two positive proteinuria categories
(trace and ≥1+) and incident hypertension remained significant even
after adjusting for age (Table 2). Further adjustment for other potential
confounders attenuated the associations; however, the association for
proteinuria ≥1+ remained significant, even in model 5, which included analyses divided by the following parameters: baseline BP, age, BMI, di-
eGFR (adjusted HR 1.19 [95% CI, 1.06 to 1.34], p b 0.001). Notably, when abetes mellitus, dyslipidemia, current smoking and current alcohol in-
we compared positive vs. negative proteinuria, the adjusted HR was sta- take. Positive associations between positive proteinuria and incident
tistically significant, even in model 5 (1.14 [95% CI, 1.03 to 1.26], hypertension were observed in several of the subgroups tested, with
p b 0.001). On the other hand, the association between a reduced few significant interactions. Of importance, the HR was significant
eGFR (≥60 ml/min/1.73 m2) and incident hypertension was more sub- among individuals with an optimal BP at baseline (b 120/80 mm Hg)
stantially attenuated by the adjustment for age. However, a significant (adjusted HR 1.31 [95% CI, 1.10 to 1.56], p = 0.003) (Fig. 2A). On the
association was observed for an eGFR of b50 ml/min/1.73 m2 only (vs. other hand, a reduced eGFR of b60 ml/min/1.73 m2 was not positively
≥ 60 ml/min/1.73 m2) after further adjustment (1.29 [95% CI, 1.03 to associated with the incidence of hypertension in nearly all of the
1.61] in model 5, p b 0.001). We did not observe any significant associ- subgroups tested (Fig. 2B). A reduced eGFR of b50 ml/min/1.73 m2
ations between a reduced eGFR (b 60 ml/min/1.73 m2) and incident hy- (vs. eGFR ≥60 ml/min/1.73 m2) was significantly associated with the
pertension in models 3–5 (HR 1.02 [0.95–1.10] in model 3). incidence of hypertension in several groups, with few interactions
(Fig. 2C). We conducted a sensitivity analysis BMI cut off of 23.0 kg/m2,
Subgroup analysis because the Regional Office for Western Pacific Region of WHO (WPRO
criteria) proposed a separate classification of obesity for Asia defining
We further evaluated the association between positive proteinuria adult overweight as a BMI ≥23.0 kg/m2, and got similar results (data
(vs. negative proteinuria) and incident hypertension in several subgroup not shown).
 N. Okumura et al. / Preventive Medicine 60 (2014) 48–54 51

Table 2
Multivariate-adjusted hazard ratios of incident hypertension according to the level of dipstick proteinuria/eGFR.

Hazard ratio (95% CI)

Model 1 Model 2 Model 3 Model 4 Model 5

Dipstick proteinuria
Negative (n = 28,468) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference)
Trace (n = 236) 1.48 (1.24–1.76) 1.42 (1.19–1.69) 1.35 (1.13–1.60) 1.05 (0.88–1.25) 1.04 (0.88–1.24)
≥1+ (n = 477) 1.79 (1.59–2.01) 1.68 (1.49–1.88) 1.49 (1.32–1.67) 1.20 (1.06–1.35) 1.19 (1.06–1.34)
Negative (n = 28,468) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference)
Positive (n = 713) 1.68 (1.52–1.85) 1.59 (1.44–1.75) 1.44 (1.30–1.59) 1.15 (1.04–1.27) 1.14 (1.03–1.26)
eGFR
≥60 (n = 27,636) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference)
50–59.9 (n = 1416) 1.46 (1.35–1.57) 1.14 (1.06–1.23) 1.00 (0.93–1.08) 0.97 (0.90–1.05) 0.97 (0.90–1.05)
b50 (n = 129) 2.47 (1.99–3.07) 1.71 (1.37–2.13) 1.55 (1.25–1.94) 1.32 (1.06–1.65) 1.29 (1.03–1.61)
≥60 (n = 27,636) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference)
≤60 (n = 1545) 1.58 (1.48–1.70) 1.15 (1.07–1.24) 1.04 (0.96–1.11) 1.00 (0.93–1.07) 0.99 (0.92–1.07)

Model 1: Crude model.

Model 2: Adjusted for age (continuous).
Model 3: Model 2 + adjusted for BMI (continuous), total cholesterol (continuous), uric acid (continuous), diabetes mellitus (yes/no), current smoking (yes/no) and current alcohol intake
(yes/no).
Model 4: Model 3 + adjusted for baseline mean BP (continuous).
Model 5: Model 4 + adjusted for eGFR (continuous) or proteinuria (positive/negative), as appropriate.
Incident hypertension was defined as a newly detected BP of ≥140/90 mm Hg and/or the initiation of antihypertensive drugs.
eGFR: estimated glomerular filtration rate, BMI: body mass index, BP: blood pressure.

Discussion
The present study, which employed annual blood pressure measure-
ment for 10 years, demonstrated that dipstick proteinuria and a re-
duced eGFR are associated with incident hypertension independently
of each other and act as potential confounders in young to middle-
aged Japanese males. The observed positive associations were consis-
tent for proteinuria in various clinical subgroups. Similarly, a significant
association between the eGFR and the incidence of hypertension was
observed in the participants with an eGFR of b 50 ml/min/1.73 m2.
When eGFR values of b60 or ≥ 60 ml/min/1.73 m2 were compared,
the associations were not significant after adjusting for age and other
potential confounders.
Our results showing a positive association between proteinuria and
incident hypertension are in line with those of previous studies
(Brantsma et al., 2006; Forman et al., 2008; Gerber et al., 2006; Inoue
et al., 2006; Jessani et al., 2012; Wang et al., 2005; Wang et al., 2007)
and extend the literature in several aspects. First, we confirmed the
presence of this association among a large cohort of Asian males.
Second, the association was independent of eGFR. Third, the association
remained significant, even in the participants with an optimal BP at
baseline. This means that our findings did not change after excluding in-
dividuals with latently elevated BP associated with proteinuria, who are
likely to develop hypertension. Fourth, we observed a consistent associ-
ation across several subgroups according to clinical risk factors, such as
age, diabetes mellitus and dyslipidemia. Finally, we were able to evalu-
ate the association for a long term of over 10 years.
There are several potential mechanisms linking proteinuria to inci-
dent hypertension. Proteinuria exerts a toxic effect on proximal tubular
epithelial cells, generating chemotactic factors, such as monocyte che-
motactic protein-1 (MCP-1) and reactive oxygen species (ROS) (Morigi
et al., 2002; Wang et al., 1999). These factors damage the renal microvas-
culature and tubulointerstitium, resulting in the impairment of salt
excretion and thus salt-sensitive hypertension (Johnson et al., 2002).
Additionally, protein overload in proximal tubular cells leads to the
secretion of endothelin-1, which can constrict systemic blood vessels
(Dhaun et al., 2012). Furthermore, proteinuria reflects systemic
endothelial dysfunction, which is known to be associated with vasocon-
striction due to a decreased level of endothelial nitric oxide (NO)
(Gkaliagkousi et al., 2009; Higashi and Chayama, 2002; Quyyumi and
Patel, 2010; Sander et al., 1999). Therefore, the presence of proteinuria
may be a harbinger of future hypertension.
The law stipulates annual medical health examinations for all
workers in Japan. Dipstick urine tests have the advantage of being inex-
pensive, quick and easy to perform therefore, it can be carried out dur-
ing screening in any countries. Also, to evaluate kidney measures and
follow these markers may encourage individuals at risk for hyperten-
sion to modify their life style such as sodium intake or physical activity
at an early stage of pre-hypertension. Previous studies in Japan have
clarified that the detection of proteinuria using dipstick tests in mass
screening settings is a strong, independent predictor of end-stage
renal disease (Iseki et al., 2003; Iseki et al., 2008). Measuring the level
of urinary proteins is important not only for assessing the prognosis
and diagnosis of kidney diseases (Matsushita et al., 2010; Herget-
Rosenthal et al., 2013), but also managing hypertension and diabetes
mellitus, both of which can induce nephropathy (Araki et al., 2007,
Ibsen et al., 2005). Our results suggest that the early detection of pro-
teinuria with a simple urine dipstick test may allow clinicians to identify
individuals at high risk for developing hypertension. In addition,
obtaining information regarding proteinuria may be useful for encour-
aging persons at high risk of hypertension to modify their lifestyle.
However, further studies are needed to evaluate whether these ap-
proaches are actually effective, particularly given the modest effect of
positive proteinuria and incident hypertension observed in our study.
In contrast to the many studies investigating the association be-
tween proteinuria and incident hypertension, the number of epidemio-
logical studies reporting an association between a reduced eGFR and
future hypertension is limited (Brantsma et al., 2006; Kestenbaum
et al., 2008; Takase et al., 2012). Two studies have reported a significant
association between a reduced kidney function and the incidence of hy-
pertension (Kestenbaum et al., 2008; Takase et al., 2012). On the other
hand, a weaker association with incident hypertension for eGFR than
for proteinuria has been reported in the PREVEND (Prevention of
REnal and Vascular End stage Disease) Study (Brantsma et al., 2006).
Similarly, in our study, the association between an eGFR of b60
compared to ≥ 60 ml/min/1.73 m2 and incident hypertension was
weaker than that for positive proteinuria (vs. negative proteinuria). In
this study, the eGFR was associated with incident hypertension only
when it was lower than 50 ml/min/1.73 m2, a level recommended for
referral to a nephrologist by the Japanese Society of Nephrology (Imai
et al., 2008). However, the number of participants with an eGFR of
b60 ml/min/1.73 m2 in our study was quite small; thus, these results
should be interpreted carefully. Further investigations are needed to de-
termine what level of GFR deterioration begins to affect blood pressure.
52 N. Okumura et al. / Preventive Medicine 60 (2014) 48–54

The potential limitations of our study include the single measure- presence of dipstick proteinuria has been shown to predict the future
ment of eGFR and the use of dipstick proteinuria as a measure of kidney risk of albuminuria and is considered useful for screening (Matsushita
damage. Although the use of the urinary albumin-to-creatinine ratio et al., 2010). Also, we do not have data on causes of proteinuria or kid-
(UACR) is preferable, as recommended in clinical guidelines, the ney dysfunction, although the recent CKD guidelines emphasize the

A Hazard ratio p for interaction

Optimal BP (n=17,592) 1.31 (1.10-1.56)

0.079
Normal-high normal BP (n=11,589) 1.09 (0.97-1.23)

Age ≥40 (n=10,566) 1.19 (1.04-1.35)

0.042
Age <40 (n=18,615) 1.01 (0.87-1.19)

BMI ≥25 (n=5,478) 1.04 (0.89-1.23)

0.004
BMI <25 (n=23,703) 1.28 (1.13-1.45)

Diabetes (+) (n=690) 1.18 (0.88-1.59)

0.842
Diabetes (-) (n=28,491) 1.13 (1.02-1.26)

Dyslipidemia (+) (n=6,060) 0.99 (0.82-1.19)

0.040
Dyslipidemia (-) (n=23,121) 1.22 (1.09-1.38)

Smoking (+) (n=15,981) 1.12 (0.97-1.29)

0.211
Smoking (-) (n=13,200) 1.15 (1.00-1.33)

Alcohol (+) (n=20,911) 1.18 (1.05-1.32)

0.263
Alcohol (-) (n=8,270) 1.05 (0.87-1.26)

Hazard ratio p for interaction

B
Optimal BP (n=17,592) 0.91(0.80-1.04)
0.059
Normal-high normal BP (n=11,589) 1.03(0.94-1.13)

Age ≥40 (n=10,566) 0.94(0.87-1.02)

0.910
Age <40 (n=18,615) 1.06(0.88-1.27)

BMI ≥25 (n=5,478) 1.04(0.92-1.17)

0.330
BMI <25 (n=23,703) 0.96(0.88-1.05)

Diabetes (+) (n=690) 1.26(0.91-1.75)

0.136
Diabetes (-) (n=28,491) 0.98(0.91-1.06)

Dyslipidemia (+) (n=6,060) 0.93(0.82-1.05)

0.459
Dyslipidemia (-) (n=23,121) 1.03(0.94-1.13)

Smoking (+) (n=15,981) 0.95(0.85-1.07)

0.035
Smoking (-) (n=13,200) 1.00(0.91-1.10)
Alcohol (+) (n=20,911) 0.95(0.90-1.07)
0.428
Alcohol (-) (n=8,270) 1.05(0.91-1.22)

Fig. 2. A. Multivariate-adjusted hazard ratios (95% CIs) of incident hypertension in the subgroups (dipstick proteinuria positive vs. negative).Footnote:The HRs were adjusted for age (con-
tinuous), BMI (continuous), total cholesterol (continuous), uric acid (continuous), diabetes (yes/no), current smoking (yes/no), current alcohol intake (yes/no), baseline mean BP (con-
tinuous) and eGFR (continuous).eGFR: estimated glomerular filtration rate, BMI: body mass index, BP: blood pressure.B. Multivariate-adjusted hazard ratios (95% CIs) of incident
hypertension in the subgroups (eGFR b 60 vs. ≥60 ml/min/1.73 m2).Footnote:The HRs were adjusted for age (continuous), BMI (continuous), total cholesterol (continuous), uric acid
(continuous), diabetes (yes/no), current smoking (yes/no), current alcohol intake (yes/no), baseline mean BP (continuous) and proteinuria (positive/negative).eGFR: estimated glomer-
ular filtration rate, BMI: body mass index, BP: blood pressure.C. Multivariate-adjusted hazard ratios (95% CIs) of incident hypertension in the subgroups (eGFR b 50 vs. ≥60 ml/min/
1.73 m2).Footnote:The HRs were adjusted for age (continuous), BMI (continuous), total cholesterol (continuous), uric acid (continuous), diabetes (yes/no), current smoking (yes/no), cur-
rent alcohol intake (yes/no), baseline mean BP (continuous) and proteinuria (positive/negative).eGFR: estimated glomerular filtration rate, BMI: body mass index, BP: blood pressure.
 N. Okumura et al. / Preventive Medicine 60 (2014) 48–54 53

C Hazard ratio p for interaction

Optimal BP (n=16,865) 1.08(0.85-1.37)

0.371
Normal-high normal BP (n=10,900) 1.15(1.01-1.30)

Age ≥40 (n=9,387) 1.04(0.93-1.17)

0.554
Age <40 (n=18,378) 1.47(0.90-2.40)

BMI ≥25 (n=5,073) 1.24(1.02-1.52)

0.946
BMI <25 (n=22,692) 1.07(0.94-1.23)

Diabetes (+) (n=629) 1.64(1.16-2.33)

0.177
Diabetes (-) (n=27,136) 1.12(0.99-1.26)

Dyslipidemia (+) (n=5,556) 1.09(0.92-1.29)

0.587
Dyslipidemia (-) (n=22,209) 1.18(1.02-1.37)

Smoking (+) (n=15,391) 1.01(0.84-1.22)

0.015
Smoking (-) (n=12,374) 1.23(1.07-1.42)
Alcohol (+) (n=19,853) 1.20(0.90-1.60)
0.441
Alcohol (-) (n=7,912) 1.57(1.10-2.25)

Fig. 2 (continued).

importance of causes (KDIGO guideline, 2013). Other potential limita-
tions of this study include the following: our study population consisted
of a single race and males only. With a healthy study population, the
study might be underpowered to detect an association between re-
duced eGFR (b60 ml/min/1.73 m2) and incident hypertension. Addi-
tionally, as with any observational study, we cannot rule out the
possibility of residual unmeasured and unknown confounding factors.
Conclusion
Both proteinuria, as assessed using a dipstick strip, and a reduced
eGFR (b50 ml/min/1.73 m2) are associated with incident hypertension
independently of each other and known potential confounders. These
findings suggest that both kidney damage and kidney dysfunction
play important roles in the development of hypertension in young to
middle-aged Japanese males.
Conflicts of interest statement
The authors declare that there are no conflicts of interest.
Acknowledgments
The authors thank the health care providers for their hard work and
excellent assistance with this study.
References
Anderson Jr., G.H., Blakeman, N., Streeten, D.H., 1994. The effect of age on prevalence of
secondary forms of hypertension in 4429 consecutively referred patients.
J. Hypertens. 12, 609–615.
Araki, S., Haneda, M., Koya, D., et al., 2007. Reduction in microalbuminuria as an integrated
indicator for renal and cardiovascular risk reduction in patients with type 2 diabetes.
Diabetes 56, 1727–1730.
Arima, H., Tanizaki, Y., Kiyohara, Y., et al., 2003. Validity of the JNC VI recommendations
for the management of hypertension in a general population of Japanese elderly:
the Hisayama study. Arch. Intern. Med. 163, 361–366.
Brantsma, A.H., Bakker, S.J., de Zeeuw, D., de Jong, P.E., Gansevoort, R.T., 2006. Urinary
albumin excretion as a predictor of the development of hypertension in the general
population. J. Am. Soc. Nephrol. 17, 331–335.
Brewster, U.C., Perazella, M.A., 2004. The renin–angiotensin–aldosterone system and the
kidney: effects on kidney disease. Am. J. Med. 116, 263–272.
Carretero, O.A., Oparil, S., 2000. Essential hypertension. Part I: definition and etiology.
Circulation 101, 329–335.
Dhaun, N., Webb, D.J., Kluth, D.C., 2012. Endothelin-1 & the kidney—beyond blood pres-
sure. Br. J. Pharmacol. 167, 720–731.
Dickinson, H.O., Mason, J.M., Nicolson, D.J., et al., 2006. Lifestyle interventions to reduce
raised blood pressure: a systematic review of randomized controlled trials.
J. Hypertens. 24, 215–233.
Forman, J.P., Fisher, N.D., Schopick, E.L., Curhan, G.C., 2008. Higher levels of albuminuria
within the normal range predict incident hypertension. J. Am. Soc. Nephrol. 19,
1983–1988.
Fox, C.S., Massaro, J.M., Hoffmann, U., et al., 2007. Abdominal visceral and subcutaneous ad-
ipose tissue compartments: association with metabolic risk factors in the Framingham
Heart Study. Circulation 116, 39–48.
Gerber, L.M., Schwartz, J.E., Pickering, T.G., 2006. Albumin-to-creatinine ratio predicts
change in ambulatory blood pressure in normotensive persons: a 7.5-year prospec-
tive study. Am. J. Hypertens. 19, 220–226.
Gkaliagkousi, E., Douma, S., Zamboulis, C., Ferro, A., 2009. Nitric oxide dysfunction in
vascular endothelium and platelets: role in essential hypertension. J. Hypertens.
27, 2310–2320.
Gueyffier, F., 2003. Observational epidemiological studies: values and limitations.
J. Hypertens. 21, 673–675.
Herget-Rosenthal, S., Dehnen, D., Kribben, A., 2013. Quellmann T. Progressive chronic kid-
ney disease in primary care: modifiable risk factors and predictive model. 57, 357-62.
Higashi, Y., Chayama, K., 2002. Renal endothelial dysfunction and hypertension.
J. Diabetes Complications 16, 103–107.
Ibsen, H., Olsen, M.H., Wachtell, K., et al., 2005. Reduction in albuminuria translates to re-
duction in cardiovascular events in hypertensive patients: losartan intervention for
endpoint reduction in hypertension study. Hypertension. 45, 198–202.
Imai, E., Horio, M., Yamagata, K., et al., 2008. Slower decline of glomerular filtration rate in
the Japanese general population: a longitudinal 10-year follow-up study. Hypertens.
Res. 31, 433–441.
Inoue, T., Iseki, K., Higashiuesato, Y., et al., 2006. Proteinuria as a significant determinant of
hypertension in a normotensive screened cohort in Okinawa. Jpn. Hypertens. Res. 29,
687–693.
Iseki, K., Ikemiya, Y., Iseki, C., Takishita, S., 2003. Proteinuria and the risk of developing
end-stage renal disease. Kidney Int. 63, 1468–1474.
Iseki, K., Tokashiki, K., Iseki, C., Kohagura, K., Kinjo, K., Takishita, S., 2008. Proteinuria and
decreased body mass index as a significant risk factor in developing end-stage renal
disease. Clin. Exp. Nephrol. 12, 363–369.
54 N. Okumura et al. / Preventive Medicine 60 (2014) 48–54
Jessani, S., Levey, A.S., Chaturvedi, N., Jafar, T.H., 2012. High normal levels of albuminuria
and risk of hypertension in Indo-Asian population. Nephrol. Dial. Transplant. 27
(suppl 3), |||58-64.
Johnson, R.J., Herrera-Acosta, J., Schreiner, G.F., Rodriguez-Iturbe, B., 2002. Subtle acquired
renal injury as a mechanism of salt-sensitive hypertension. N. Engl. J. Med. 346, 913–923.
KDIGO guideline, 2013. Intersalt: an international study of electrolyte excretion and blood
pressure. Results for 24 hour urinary sodium and potassium excretion. Intersalt Co-
operative Research Group 1988. (http://kdigo.org/home/guidelines/ckd-evaluation-
management/) BMJ 297, 319–328.
Kearney, P.M., Whelton, M., Reynolds, K., Muntner, P., Whelton, P.K., He, J., 2005. Global
burden of hypertension: analysis of worldwide data. Lancet 365, 217–223.
Kestenbaum, B., Rudser, K.D., de Boer, I.H., et al., 2008. Differences in kidney function and
incident hypertension: the multi-ethnic study of atherosclerosis. Ann. Intern. Med.
148, 501–508.
Klag, M.J., Whelton, P.K., Randall, B.L., et al., 1996. Blood pressure and end-stage renal
disease in men. N. Engl. J. Med. 334, 13–18.
Komukai, K., Mochizuki, S., Yoshimura, M., 2010. Gender and the renin–angiotensin–
aldosterone system. Fundam. Clin. Pharmacol. 24, 687–698.
Kondo, T., Osugi, S., Shimokata, K., et al., 2013. Cardiovascular events increased at normal
and high-normal blood pressure in young and middle-aged Japanese male smokers
but not in nonsmokers. J. Hypertens. 31, 263–270.
Kubo, M., Hata, J., Doi, Y., Tanizaki, Y., Iida, M., Kiyohara, Y., 2008. Secular trends in the in-
cidence of and risk factors for ischemic stroke and its subtypes in Japanese popula-
tion. Circulation 118, 2672–2678.
Matsuo, S., Imai, E., Horio, M., et al., 2009. Revised equations for estimated GFR from
serum creatinine in Japan. Am. J. Kidney Dis. 53, 982–992.
Matsushita, K., van der Velde, M., Astor, B.C., et al., 2010. Association of estimated glomer-
ular filtration rate and albuminuria with all-cause and cardiovascular mortality in
general population cohorts: a collaborative meta-analysis. Lancet 375, 2073–2081.
Morigi, M., Macconi, D., Zoja, C., et al., 2002. Protein overload-induced NF-kappaB activa-
tion in proximal tubular cells requires H(2)O(2) through a PKC-dependent pathway.
J. Am. Soc. Nephrol. 13, 1179–1189.
Nishikawa, T., Saito, J., Omura, M., 2007. Prevalence of primary aldosteronism: should we
screen for primary aldosteronism before treating hypertensive patients with medica-
tion? Endocr. J. 54, 487–495.
Palatini, P., Mormino, P., Mos, L., et al., 2005. Microalbuminuria, renal function and devel-
opment of sustained hypertension: a longitudinal study in the early stage of hyper-
tension. J. Hypertens. 23, 175–182.
Quyyumi, A.A., Patel, R.S., 2010. Endothelial dysfunction and hypertension: cause or
effect? Hypertension 55, 1092–1094.
Rossi, G.P., Bernini, G., Caliumi, C., et al., 2006. A prospective study of the prevalence of
primary aldosteronism in 1,125 hypertensive patients. J. Am. Coll. Cardiol. 48,
2293–2300.
Sander, M., Chavoshan, B., Victor, R.G., 1999. A large blood pressure-raising effect of nitric
oxide synthase inhibition in humans. Hypertension 33, 937–942.
Takase, H., Dohi, Y., Toriyama, T., et al., 2012. Evaluation of risk for incident hypertension
using glomerular filtration rate in the normotensive general population. J. Hypertens.
30, 505–512.
Wang, Y., Rangan, G.K., Tay, Y.C., Harris, D.C., 1999. Induction of monocyte
chemoattractant protein-1 by albumin is mediated by nuclear factor kappaB in
proximal tubule cells. J. Am. Soc. Nephrol. 10, 1204–1213.
Wang, T.J., Evans, J.C., Meigs, J.B., et al., 2005. Low-grade albuminuria and the risks of
hypertension and blood pressure progression. Circulation 111, 1370–1376.
Wang, T.J., Gona, P., Larson, M.G., et al., 2007. Multiple biomarkers and the risk of incident
hypertension. Hypertension 49, 432–438.
Yamashita, K., Kondo, T., Osugi, S., et al., 2012. The significance of measuring body fat per-
centage determined by bioelectrical impedance analysis for detecting subjects with
cardiovascular disease risk factors. Circ J. 76, 2435–2442.