Albuminuria in relation to the single and combined effects of systolic and diastolic blood pressure in Chinese

Abstract

We investigated the single and combined effects of systolic (SBP) and diastolic blood pressure (DBP) on albuminuria in Chinese. We measured blood pressure (BP), anthropometry and urinary excretions of albumin and creatinine, and defined albuminuria as a urinary albumin-to-creatinine ratio of at least 17 and 25 mg/g in men and women, respectively. The 1068 participants (mean age 56.3 years) included 407 (38.5%) men and 410 (38.4%) hypertensive patients. A J-shaped relationship between the risk of albuminuria and BP was observed for both SBP (mean ± SD, 126.1 ± 18.9 mmHg) and DBP (77.1 ± 9.4 mmHg) with nadir levels of 110 mmHg and 70 mmHg, respectively. The risk of albuminuria was significantly (p ≤ 0.01) associated with DBP in the subjects with a SBP of at least 130 mmHg and with SBP in subjects with a DBP of at least 80 mmHg, and inversely and significantly (p = 0.04) associated with SBP in subjects with a DBP below 70 mmHg. In conclusion, as far as albuminuria is concerned, there is indeed a J-shaped phenomenon. However, it has a nadir level far below the currently recommended target BP of 140/90 or 130/80 mmHg.

Key Words::

• Albuminuria
• blood pressure
• J-curve

Introduction

With the increasing longevity and obesity in most countries over the world, the prevalence of hypertension increases rapidly. On the other hand, the control rate of hypertension remains low, particularly in China. According to the 2002 National Nutrition and Health Survey, the prevalence of hypertension was 18.8%, with approximately 160 million people with a systolic/diastolic blood pressure (SBP/DBP) of at least 140/90 mmHg, and the control rate was only 6.2% (1). The high prevalence and low control rate of hypertension are the major contributing factors of the fast increasing incidence rate of stroke and coronary heart disease in the most populous country over the world.

By lowering blood pressure (BP), antihypertensive treatment may effectively prevent complications of hypertension such as stroke, coronary heart disease, congestive heart failure and renal dysfunction (2,3). In spite of the overwhelming evidence of the benefit of antihypertensive therapy, the J-curve phenomenon has been a major concern and hot topic and is recently revived in a high profile, because several trials that included normotensive subjects (4–7) and/or tested intensive BP lowering (8,9) did not show overall beneficial effects. The optimal BP level might be personalized, depending on the perfused organ, the function and structure of the vascular system, and probably also the combined effects of the peak systolic (SBP) and trough diastolic (DBP) level of BP within a cardiac cycle.

Albuminuria is an early marker of vascular and renal damage, and predicts cardiovascular events and mortality (10–12). In the present study, we investigated the single and combined effects of SBP and DBP on albuminuria in a Chinese population sample, with the focus on the J-curve phenomenon.

Methods

Study population

The present cross-sectional analysis was based on the recent follow-up data of an ongoing longitudinal population study on multiple cardiovascular risk factors in Shanghai, China (13). The Ethics Committee of Ruijin Hospital, Shanghai Jiaotong University School of Medicine approved the study protocol. All subjects gave written informed consent. The study subjects were recruited from a newly established residential area in the suburb of Shanghai, 30 kilometers from the city centre. Most residents were immigrants from the nearby villages since 2003, and previously doing farming or other agricultural work. In the period from 2004 to 2006, we invited all residents at least 12 years of age to take part. All residents were potentially eligible for inclusion into our study, as long as they could walk to the local examination centre on their own and had no clinically manifested disease. Of the 2275 invited, 1700 (74.7%) participated in the study.

At the time of the follow-up study in 2009, 1630 of the 1700 subjects were still alive, 1192 (73.1%) participated in the follow-up examinations. We excluded 124 subjects, because they did not have blood (n = 22) or spot urine sampling (n = 58) or anthropometric measurements (n = 18), or because they were younger than 30 years of age (n = 26). Thus, the total number of subjects included in the present analysis was 1068.

Field work

One experienced physician measured each participant's BP five times consecutively by mercury sphygmomanometry, after the subjects had rested for at least 5 min in the sitting position. These five BP readings were averaged for analysis. The same observer also administered a standardized questionnaire to collect information on medical history, smoking habit, alcohol intake and the use of medications. Hypertension was defined as a sitting BP of at least 140 mmHg SBP or 90 mmHg DBP, or as the use of antihypertensive drugs. A trained technician performed anthropometric measurements.

Laboratory work and definitions of disease conditions

Venous blood samples were drawn after overnight fasting for the measurement of plasma glucose concentration and for measurements of serum concentrations of uric acid and creatinine. Diabetes mellitus was defined as a plasma glucose of at least 7.0 mmol/l fasting or 11.1 mmol/l at any time, or as the use of antidiabetic agents (14). Estimated glomerular filtration rate (eGFR) was calculated according to the Cockcroft–Gault formula (15). Body surface area was calculated using the DuBois formula (16). Renal dysfunction was defined as an eGFR lower than 60 ml/min/1.73 m². Fresh spot urine samples were collected for the measurement of urinary albumin and creatinine using immunoturbidimetry and enzymatic methods, respectively. Urinary albumin excretion was expressed as the ratio of albumin-to-creatinine ratio (mg/g).

We defined microalbuminuria, according to a sex-specific criterion, as a urinary albumin-to- creatinine ratio in the range from 17 to 299 mg/g in men and from 25 to 299 mg/g in women (17,18), and defined macroalbuminuria as a urinary albumin-to-creatinine ratio more than 300 mg/g in men as well as women. Albuminuria included microalbuminuria and macroalbuminuria.

Statistical analysis

For database management and statistical analysis, we used SAS software (version 9.13, SAS Institute, Cary, NC, USA). Means and proportions were compared with the standard normal z-test and Fisher's exact test, respectively. We performed multiple logistic regression analyses to study the association between albuminuria and BP, while controlling for covariates including sex, age, body mass index, current smoking, alcohol intake, use of antihypertensive drugs and diabetes mellitus. We considered individuals with SBP of 110–119 mmHg or DBP of 70–74 mmHg the reference group. Dummy variables were used to compute odds ratios (95% confidence intervals) for each subgroup against the reference group. For the analysis on the joint effect of SBP and DBP, we first subdivided the study subjects into three subgroups according to one component of BP and then studied the association of another component with the risk of albuminuria.

Results

The 1068 participants included 407 (38.1%) men, 66 (6.2%) diabetic patients and 410 (38.4%) hypertensive patients, of whom 331 (31.0%) took antihypertensive drugs. Men and women had similar SBP (± SD; 125.1 ± 18.7 mmHg), pulse rate (68.5 ± 8.4 beats/min), and fasting plasma glucose (4.55 ± 1.34 mmol/l, Table I). However, men, compared with women, were slightly older (+ 1.7 years, p = 0.03), had a significantly (p ≤ 0.05) greater body mass index (+ 0.4 kg/m²), and higher DBP (+ 3.2 mmHg), and reported higher (p < 0.0001) proportions of current smoking (47.8% vs 0.5%) and alcohol intake (35.4% vs 1.2%).

Table I. Characteristics of the study population.

Characteristic	Men (n = 407)	Women (n = 661)	p-value
Age, years	57.4 ± 12.2	55.7 ± 12.2	0.03
Body mass index, kg/m^2	24.3 ± 3.4	23.9 ± 3.2	0.03
Blood pressure, mmHg
Systolic	126.3 ± 17.6	125.9 ± 19.5	0.69
Diastolic	79.0 ± 9.6	75.8 ± 9.1	< 0.0001
Pulse rate, beats/min	68.1 ± 8.9	68.7 ± 8.0	0.24
Taking antihypertensive drugs, n (%)	139 (33.8)	192 (28.8)	0.11
Hypertension, n (%)^a	179 (43.6)	231 (34.7)	0.01
Diabetes mellitus, n (%)^a	28 (6.8)	38 (5.7)	0.50
Plasma fasting glucose, mmol/l	4.52 ± 1.56	4.63 ± 1.31	0.21
Serum creatinine, μmol/l	70.0 ± 12.5	54.3 ± 17.9	< 0.0001
Serum uric acid, μmol/l	363.8 ± 85.5	291.9 ± 83.0	< 0.0001
Microalbuminuria, n (%)^a	28 (6.9)	40 (6.1)	0.63
Macroalbuminuria, n (%)^a	3 (0.7)	12 (1.8)	0.14
Renal dysfunction^a	25 (6.1)	33 (5.0)	0.43

Values are mean ± SD, or number of subjects (%). ^aFor definitions of hypertension, diabetes mellitus, microalbuminuria, macroalbuminuria and renal dysfunction, see Methods.

Men and women had similar prevalence of albuminuria (7.8%), and mild renal dysfunction (5.4%). The prevalence of albuminuria increased with SBP and DBP in men as well as women (Figure 1). However, we noticed a J-shape relationship for both BP components with a nadir level at 110–119 mmHg SBP and 70–74 mmHg DBP (Table II). After adjustment for sex, age, body mass index, current smoking, alcohol intake, use of antihypertensive drugs and diabetes mellitus, and additional mutual adjustment for the BP components, the J-shape relationship remained, particularly with DBP (Table II). Indeed, the smallest odds ratio was observed at 110–119 mmHg SBP and 70–74 mmHg DBP.

Figure 1. Prevalence of albuminuria by sex and by systolic (left panel) and diastolic (right panel) blood pressure. Closed and open circles represent men and women, respectively. The number of subjects per group in men and women is given.

Table II. Albuminuria in relation to systolic and diastolic blood pressures.

	Number of participants	Number of patients with albuminuria (prevalence, %)	Odds ratio (95% CI)^a	p-value
Systolic blood pressure (mmHg)
< 110	204	6 (2.9)	1.36 (0.45–4.16)	0.59
110–119	234	9 (3.8)	1.0^b	–
120–129	228	11 (4.8)	1.06 (0.41–2.76)	0.91
130–139	172	15 (8.7)	1.53 (0.58–4.04)	0.39
≥ 140	230	42 (18.3)	3.01 (1.12–8.06)	0.03
Diastolic blood pressure (mmHg)
< 70	271	11 (4.1)	1.66 (0.55–5.06)	0.37
70–74	200	5 (2.5)	1.0^b	–
75–79	200	11 (5.5)	2.05 (0.70–6.06)	0.19
80–84	182	20 (11.0)	3.33 (1.18–9.38)	0.03
≥ 85	215	36 (16.7)	4.68 (1.62–13.52)	0.004

^aOdds ratios (95% confidence interval [CI]) were calculated for each subgroup against the risk of the reference group in a single model, adjusted for sex, age, body mass index, current smoking, alcohol intake, use of antihypertensive drugs and diabetes mellitus, and mutually adjusted for diastolic or systolic blood pressure. ^bReference group.

In further analysis, we studied the combined effects of SBP and DBPs on albuminuria. After subdivision of all participants into three subgroups according to SBP (< 120, 120–129 and ≥ 130 mmHg) and DBP (< 70, 70–79 and ≥ 80 mmHg) BPs, we studied the risk of albuminuria in relation to the tertile distributions of DBP and SBPs, respectively. Both before (Figure 2) and after (Table III) adjustment for sex, age, body mass index, current smoking, alcohol intake, use of antihypertensive drugs and diabetes mellitus, the risk of albuminuria was positively and significantly (p ≤ 0.01) associated with DBP in the subjects with a SBP of at least 130 mmHg and with SBP in subjects with a DBP of at least 80 mmHg, and inversely and significantly (p = 0.04) associated with SBP in subjects with a DBP below 70 mmHg.

Figure 2. Prevalence of albuminuria by systolic and diastolic blood pressures. Values are prevalence of albuminuria in each tertile of systolic (right panel) and diastolic blood pressures (left panel) after subdivision by diastolic and systolic blood pressures, respectively. Tertiles 1, 2 and 3 are shown by open and closed circles, squares and diamonds, respectively.

Table III. Albuminuria in relation to the combined effects of systolic and diastolic blood pressures.

	Number of participants	Number of patients with albuminuria (prevalence, %)	Odds ratio (95% CI)^a	P
Systolic blood pressure < 120 mmHg
DBP tertile 1, < 68 mmHg	128	5 (3.9)	1.0^b	–
DBP tertile 2, 68–74 mmHg	149	6 (4.0)	1.89 (0.43–8.39)	0.40
DBP tertile 3, > 74 mmHg	151	4 (2.6)	1.09 (0.20–5.93)	0.92
Systolic blood pressure, 120–129 mmHg
DBP tertile 1, < 75 mmHg	76	1 (1.3)	1.0^b	–
DBP tertile 2, 75–82 mmHg	78	4 (5.1)	4.43 (0.44–44.77)	0.21
DBP tertile 3, > 82 mmHg	74	6 (8.1)	8.35 (0.73–95.88)	0.09
Systolic blood pressure ≥ 130 mmHg
DBP tertile 1, < 80 mmHg	127	9 (7.1)	1.0^b	–
DBP tertile 2, 80–86 mmHg	132	20 (15.1)>	2.33 (0.97–5.62)	0.06
DBP tertile 3, > 86 mmHg	143	28 (19.6)	3.57 (1.33–9.59)	0.01
Diastolic blood pressure < 70 mmHg
SBP tertile 1, < 105 mmHg	89	3 (3.3)	1.0^b	–
SBP tertile 2, 105–116 mmHg	92	5 (5.4)	0.46 (0.07–2.91)	0.41
SBP tertile 3, > 116 mmHg	90	3 (3.3)	0.10 (0.01–0.96)	0.04
Diastolic blood pressure, 70–79 (mmHg)
SBP tertile 1, < 114 mmHg	133	4 (3.0)	1.0^b	–
SBP tertile 2, 114–123 mmHg	117	3 (2.6)	0.61 (0.13–2.96)	0.54
SBP tertile 3, > 123 mmHg	150	9 (6.0)	0.89 (0.19–4.12)	0.88
Diastolic blood pressure ≥ 80 (mmHg)
SBP tertile 1, < 130 mmHg	130	6 (4.6)	1.0^b	–
SBP tertile 2, 130–144 mmHg	134	9 (6.7)	2.52 (0.97–6.56)	0.06
SBP tertile 3, > 144 mmHg	133	31 (23.3)	3.95 (1.34–11.69)	0.01

SBP, systolic blood pressure; DBP, diastolic blood pressure. ^aOdds ratios (95% confidence interval [CI]) were calculated for each of the three tertile subgroups against the risk of the reference group in a single model, after subdivision of the study subjects into tertiles of SBP and DBP, respectively, and adjusted for sex, age, body mass index, current smoking, alcohol intake, use of antihypertensive drugs and diabetes mellitus. ^bReference group.

Discussion

Our finding, obtained in a general population with relatively low levels of cardiovascular risk, showed a J-shaped relationship between BP and target organ damage, such as, for instance, albuminuria. However, this phenomenon became apparent only if both SBP and DBPs were at low levels. One of the main clinical implications of our finding therefore is that SBP and DBPs that exceed the current threshold for the diagnosis of hypertension either alone or in combination confer cardiovascular risk. Of course, too low BP might also confer risks instead of benefits.

We studied single as well as combined effects of SBP and DBP, and found that the coexistence of too low levels in both BP components is the driving force of the J-shaped relationship. Previous studies often studied single effects of SBP or DBP. In the post hoc analyses of the INVEST (the International Verapamil SR-Trandolapril Study) (19), TNT (the Treating to New Targets trial) (20) and ONTARGET (the Ongoing Telmisartan Alone and in combination with Ramipril Global Endpoint Trial study) trials (21), the relationship between BP and cardiovascular events was analyzed for SBP and DBP separately, and concluded that low DBP was the major cause of the J-shaped relationship. Mainly based on these findings, the scientific statement on antihypertensive therapy in patients with coronary artery disease recommended that DBP should not be lowered to a level below 60 mmHg (22). This recommendation is important, but may also be confusing. Some patients may have a DBP below 60 mmHg before antihypertensive drug treatment is initiated. Some others may have such a low DBP before SBP is lowed to the goal. Physicians might hesitate to initiate or intensify antihypertensive therapy in these patients. It is therefore imperative to consider SBP and DBP together.

A major limitation of our study is its cross- sectional design, which does not allow us to draw any casual inference. However, because albuminuria is unlikely to cause low BP, it is plausible that albuminuria in some persons is indeed the consequence of too low BP. Low BP in untreated subjects is different from that induced by antihypertensive drug treatment. In the latter case, BP may fluctuate from high to low levels depending on the use of antihypertensive drugs. Recent studies have demonstrated that increased visit-to-visit BP variability may increase cardiovascular risk independent of BP level (23). Nonetheless, too low BP on both untreated and treated conditions would decrease perfusion of important organs, such as the brain, heart and kidneys, and increase the risk of cardiovascular events. Administration of drugs with BP lowering action in patients without high BP, regardless of high or low cardiovascular risk, should be closely monitored, and started from lower dosages.

In addition, although albuminuria is a predictor of cardiovascular events and mortality, our research cannot be directly compared with studies on hard outcomes. Albuminuria is a measure of long-term cardiovascular and renal risk of high BP or other risk factors, such as diabetes mellitus. However, cardiovascular events or mortality can be the consequence of long-term risk as well as of acute accidents, such as rupture of plaques, vasospasm, etc.

We only studied albuminuria, which is usually considered a chronic kidney disease. Our finding probably cannot be extrapolated to other target organs, such as the brain and heart. The J-curve phenomenon can be organ-specific. Indeed, in the ACCORD (Action to Control Cardiovascular Risk in Diabetes) (8) and NAVIGATOR (the Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Research) trials (7), a J-shaped relationship between intensive BP lowering and cardiovascular events was observed for coronary events but not for stroke, the risk of which was actually reduced in both trials. In contrast, in the ADVANCE (the Action in Diabetes and Vascular Disease:Preterax and Diamicron Modified Release Controlled Evaluation) trial (24), intensive BP lowering significantly reduced the incidence of coronary events and albuminuria, but did not influence the risk of stroke, despite 6/3 mmHg of SBP and DBP reductions, respectively. The discrepancy between these trials is not entirely understood, but clearly suggested the complexity of the J-curve issue.

The results of our study should be interpreted within the context of several other limitations. First, the sample size of our study was relatively small. For this reason and also because of the difference in the distributions of BP between the three subgroups in our analyses on the combined effects of SBP and DBPs, we had to subdivide our study subjects according to tertile distributions instead of absolute values of BP. Second, our study subjects were recruited from a single town in the suburb of Shanghai. Our study might be less representative of the Chinese population than multicentre studies, such as the 2002 National Nutrition and Health Survey (1) and the 2008 National Diabetes Survey (25), but might not necessarily be less appropriate for research such as the present analysis. Finally, the low SBP and DBPs could also be a consequence of left ventricular dysfunction. We excluded patients with overt heart failure according to the questionnaire data on medical history and the use of drugs. However, we did not perform echocardiographic measurements, and therefore cannot entirely rule out the possibility that a small proportion of subjects with compensatory left ventricular dysfunction were included in the present analysis.

In conclusion, there is indeed a J-shaped relationship between BP and albuminuria. However, it has a nadir level far below the currently recommended target BP of 140/90 mmHg or 130/80 mmHg, and is apparent only if both SBP and DBPs are at low levels. Nonetheless, the J-curve issue is important, requires accurate measurement of BP, and should be investigated in future studies.

Acknowledgments

The authors gratefully acknowledge the voluntary participation of all study subjects and the technical assistance of the physicians and nurses of Zhaoxiang Community Health Centre (Qingpu District, Shanghai). The authors also appreciate the expert assistance of Jie Wang, and Wei-Zhong Zhang (The Shanghai Institute of Hypertension, Shanghai, China). The present study was financially supported by grants from the National Natural Science Foundation of China (grants 30871360, 30871081, and 81170245), the Ministry of Science and Technology (2006BAI01A03 and a grant for China-European Union collaborations [1012]), and the Ministry of Education (NCET-09-0544), Beijing China, the Shanghai Commissions of Science and Technology (grant 07JC14047, the “Rising Star” program 06QA14043, and 11QH1402000) and Education (grant 07ZZ32 and the “Dawn” project 08SG20), the Shanghai Bureau of Health (2009111 and XBR2011004), the Shanghai Shenkang Hospital Development Centre (SHDC12007318), Shanghai Jiaotong University School of Medicine (a grant of Distinguished Young Investigators to Yan Li), and the European Union (grants LSHM-CT-2006-037093 and HEALTH-F4-2007-201550).

Conflict of interest: None.