If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Incidence of cardiovascular events was more frequent in low estimated glomerular filtration rate (eGFR) than high eGFR group.
•
A large decline in eGFR was more observed in high eGFR than low eGFR group.
•
A large decline in eGFR was at a 2-fold risk of cardiovascular events.
Abstract
Background
It has been reported that a large decline in estimated glomerular filtration rate (eGFR) over time is associated with increased incidence of cardiovascular disease. We investigated whether this association differs according to the baseline eGFR.
Methods
A total of 4666 patients (male 71%) with measurements of eGFR at both baseline and 1 year and that had no cardiovascular events at 1-year follow-up were retrieved from the Shinken Database between June 2004 and March 2015. The study population was divided into three groups by baseline eGFR (mL/min/1.73 m2): high (≥60, n = 1650), intermediate (45–59, n = 1947), and low (<45, n = 1069) eGFR groups. Each eGFR group was further divided into two groups by eGFR slope (change at 1 year, <-10 and ≥-10 mL/min/1.73 m2). The patient characteristics and the incidences of cardiovascular events within 3 years (after 1-year follow-up) were compared between the negatively large eGFR slope (<-10) and others (≥-10) in each eGFR group.
Results
A total of 187 cardiovascular events occurred during the mean follow-up of 2.8 ± 0.6 years. The adjusted hazard ratios of eGFR slope (<-10 with reference to ≥-10) were 2.37 (95% CI, 1.28–4.40), 3.10 (95% CI, 1.78–5.40), and 2.66 (95% CI, 1.15–6.13) in the high, middle, and low eGFR groups, respectively. Similar results were found in patients with structural heart disease, but not in those without.
Conclusions
Decline in eGFR was associated with an increase in cardiovascular events, and this effect was consistent regardless of the baseline eGFR.
]. Decreased estimated glomerular filtration rate (eGFR) is the main contributing factor for CKD, which is not only an indicator of renal dysfunction, but also a surrogate marker for the progression of atherosclerosis [
Although eGFR declines with age and disease progression, the degree of the decline in eGFR differs between individuals. The degree of decline in eGFR, which is likely to be affected by the accumulation of risk factors for CVD, could be associated with increased risk of the incidence of CVD [
]. However, as the degree of decline in eGFR would also be affected by the eGFR itself, this raises the question of whether the risk of decline in eGFR is constant regardless of eGFR. This study was performed to investigate the distribution of degree of decline in eGFR according to eGFR, and the impact of the decline in eGFR on the incidence of cardiovascular events according to eGFR in a single-center cohort from a cardiovascular hospital in Japan.
Methods
Study patients
The study population was retrieved from the Shinken Database, a single-hospital cohort database consisting of all new patients who visited the Cardiovascular Institute Hospital, Tokyo, Japan [
. This database is on-going, beginning in June 2004 and 19487 patients had been registered by March 2015. Using this database, we retrospectively analyzed the effect of decline in eGFR at 1-year follow-up from baseline on the incidence of cardiovascular events after the 1-year time point. We selected 5180 patients whose eGFR at baseline and 1-year follow-up were available. We excluded those with end-stage renal disease on dialysis (n = 71) and those with cardiovascular events within the first year (n = 443). The remaining 4666 patients comprised the present study population (Fig. 1).
The ethics committee of the Cardiovascular Institute approved the study protocol, and all patients registered in the database provided written informed consent.
Patient follow-up
The health status and the incidences of cardiovascular events and mortality are maintained in the database by linking to the medical records of the hospital, and by study documents of prognosis sent once per year for those who stopped hospital visits or were referred to other hospitals. In the present study, the follow-up data until March 2019 were included in the analysis.
Data collection at initial visit
The baseline patient characteristics, including age, sex, height, weight, CVD, and medications, were retrieved. The cardiovascular status was evaluated using the data of echocardiogram, exercise test, 24-h Holter recording, and blood laboratory data at the discretion of the attending physician. The diagnosis of structural heart diseases (SHD) was defined as follows: valvular heart disease, moderate or severe stenosis or regurgitation on echocardiogram; coronary artery disease, diagnosed on angiogram or scintigram; hypertrophic and dilated cardiomyopathy, diagnosed on echocardiography or magnetic resonance imaging (MRI). Heart failure was diagnosed when the patients had symptoms of New York Heart Association class ≥2.
Cardiovascular risk factors were defined as follows: hypertension, antihypertensive drug use, systolic blood pressure ≥140 mmHg, or diastolic blood pressure ≥90 mmHg; diabetes mellitus, hypoglycemic drug or insulin use, or glycosylated hemoglobin ≥6.5%; dyslipidemia, statin use, or low-density lipoprotein ≥140 mg/dL, high-density lipoprotein <40 mg/dL, or fasting triglyceride ≥150 mg/dL; and CKD, eGFR <60 mL/min/1.73 m2. The eGFR was calculated using the Japanese Society of Nephrology formulae for Japanese individuals: eGFR = 194 × (serum creatinine)−1.094 × (age)−0.287 for men and 194 × (serum creatinine)−1.094 × (age)−0.287 × 0.739 for women [
]. Anemia was defined as hemoglobin level <11 g/dL. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared.
Definition of study endpoint
The study endpoint was cardiovascular events, which was a composite of admission or death with cardiovascular diseases, including heart failure, acute coronary syndrome, ischemic stroke, or hemorrhagic stroke.
Statistical analysis
All analyses were performed using SPSS 19.0 (SPSS Inc., Chicago, IL, USA). In all analyses, p < 0.05 was taken to indicate statistical significance. Study patients were divided into three groups: lower (eGFR <45 mL/min/1.73 m2), intermediate (45 ≤ eGFR < 60 mL/min/1.73 m2), and higher (60 ≤ eGFR mL/min/1.73 m2) eGFR groups. We defined the eGFR slope as the gap between eGFR at baseline and at 1-year follow up: [eGFR slope] = [eGFR at 1-year follow up] − [eGFR at baseline]. The patients in each eGFR group were further divided into two groups at an eGFR slope of −10 mL/min/1.73 m2. For dichotomization by eGFR slope, the cut-off value varied in previous studies, but mostly were approximately −5 mL/min/1.73 m2 [
]. In the present study, we evaluated the incidence rate of cardiovascular events in categories with intervals of 5 by eGFR slope (≥0, −0.1~−5.0, −5.1~−10.0, −10.1~−15.0, and <−15.1) in total patients and found eGFR slope −10 mL/min/1.73 m2 was a good cut-off. We applied the same cut-off value for each eGFR level (Fig. 2).
Fig. 2The incidence rate of cardiovascular events according to the levels of eGFR slope in total patients and in patients with high, intermediate, and low eGFR. eGFR, estimated glomerular filtration rate.
First, baseline characteristics were compared between the baseline eGFR groups. In addition, the baseline characteristics were further compared between large negative (< −10) and small negative or positive (≥ −10) eGFR slope in each eGFR group. Continuous variables and categorical variables are expressed as the mean ± SD and absolute number (percentage), respectively. The differences in categorical variables among groups were tested by the χ2 test. The differences in continuous variables among two groups were tested by the unpaired Student's t test when normally distributed and the Mann–Whitney test when non-normally distributed. The differences in continuous variables among the three groups were tested by one-way analysis of variance. Second, the cumulative incidence rate of cardiovascular events was estimated by the Kaplan–Meier method, and the differences among groups were tested by the log-rank test. Third, univariable and multivariable Cox regression analyses were performed to estimate the hazard ratio (HR) of the category of eGFR slope (eGFR slope < −10 mL/min/1.73 m2) for cardiovascular events in the total patients. In the multivariable model, the category of eGFR slope and baseline eGFR levels, and covariables with p < 0.10 in the univariable models were forcedly entered. Fourth, univariable and multivariable Cox regression analyses were performed again in groups stratified according to eGFR level (low, intermediate, and high eGFR) to estimate the HR of the category of eGFR slope for cardiovascular events in each eGFR level. In the multivariable model, the category of eGFR slope and baseline eGFR levels, and covariables with p < 0.10 in the univariable models were forcedly entered.
Results
Patient characteristics
The total study population had an average age of 63 years, and included 3328 males (71%). Table 1 shows the baseline clinical characteristics according to baseline eGFR. The numbers of patients in the high, intermediate, and low eGFR groups were 1650 (male, 44%), 1947 (male, 84%), and 1069 (male, 91%), respectively. Compared with the high eGFR group, the patients in the low eGFR group were older, had a higher percentage of males, and higher prevalence rates of hypertension, diabetes mellitus, atrial fibrillation (AF), and SHD.
Table 1Baseline characteristics according to the baseline eGFR levels.
All n = 4666 (100%)
60≤ eGFR n = 1650 (35%)
45≤ eGFR< 60 n = 1947 (42%)
eGFR< 45 n = 1069 (23%)
p-value
Age (years)
63.3 ± 12.1
60.6 ± 12.9
62.9 ± 11.2
68.1 ± 10.7
<0.001
Sex, male, n (%)
3328 (71.3)
724 (43.9)
1631 (83.8)
973 (91.0)
<0.001
BMI, kg/m2
24.1 ± 3.6
23.4 ± 3.8
24.4 ± 3.4
24.6 ± 3.4
<0.001
Coexisting disorders
Hypertension, n (%)
2978 (63.8)
980 (59.4)
1222 (62.8)
776 (72.6)
<0.001
Diabetes mellitus, n (%)
1245 (26.7)
389 (23.6)
504 (25.9)
352 (32.9)
<0.001
Dyslipidemia, n (%)
2543 (54.5)
891 (54.0)
1074 (55.2)
578 (54.1)
0.744
Hyperuricemia, n (%)
1430 (30.6)
263 (15.9)
603 (30.1)
564 (52.8)
<0.001
Anemia, n (%)
205 (4.4)
61 (3.7)
66 (3.4)
78 (7.3)
<0.001
AF, n (%)
1020 (21.9)
271 (16.4)
471 (24.2)
278 (26.0)
<0.001
SHD, n (%)
2731 (58.5)
872 (52.8)
1143 (58.7)
716 (67.0)
<0.001
Ischemic heart diseases, n (%)
1910 (40.9)
579 (35.1)
845 (43.4)
486 (45.5)
<0.001
Valvular diseases, n (%)
729 (15.6)
255 (15.5)
252 (12.9)
222 (20.8)
<0.001
Cardiomyopathy, n (%)
424 (9.1)
116 (7.0)
169 (8.7)
139 (13.0)
<0.001
Laboratory data
Baseline eGFR, mL/min/1.73 m²
56.4 ± 16.1
73.3 ± 12.0
52.5 ± 4.1
37.5 ± 6.7
<0.001
eGFR at 1-year, mL/min/1.73 m²
55.0 ± 15.8
69.1 ± 13.9
52.0 ± 8.0
38.8 ± 9.3
<0.001
eGFR slope, mL/min/1.73 m²
-0.9 (-5.8, 3.5)
-3.5 (-9.8, 2.0)
-0.5 (-4.6, 3.8)
0.5 (-2.9, 4.9)
<0.001
BNP, pg/ml
66 (25, 188)
53 (21, 143)
59 (24, 163)
109 (44, 316)
<0.001
Medication at baseline
Ca blocker, n (%)
1636 (35.1)
537 (32.5)
668 (34.3)
431 (40.3)
<0.001
RAS inhibitor, n (%)
2190 (46.9)
709 (43.0)
886 (45.5)
595 (55.7)
<0.001
Statin, n (%)
1408 (30.2)
483 (29.3)
598 (30.7)
327 (30.6)
0.608
Loop diuretic, n (%)
714 (15.3)
211 (12.8)
231 (11.9)
272 (25.4)
<0.001
β blocker, n (%)
1751 (37.5)
534 (32.4)
739 (38.0)
478 (44.7)
<0.001
Antiplatelet, n (%)
2112 (45.3)
679 (41.2)
892 (45.8)
541 (50.6)
<0.001
Anticoagulant, n (%)
1184 (25.4)
340 (20.6)
490 (25.2)
354 (33.1)
<0.001
Echocardiographic parameters
LVEF, n (%)
62.9 ± 13.1
65.1 ± 12.0
62.6 ± 12.4
59.9 ± 15.2
<0.001
Continuous values are presented as mean ± standard deviation or median (inter-quartiles).
eGFR, estimated glomerular filtration rate; BMI, body mass index; AF, atrial fibrillation; SHD, structural heart disease; BNP, brain natriuretic peptide; Ca blocker, calcium channel blocker; RAS inhibitor, renin–angiotensin system inhibitor; LVEF, left ventricular ejection fraction.
In Table 2, the patient characteristics were compared between the dichotomized categories of eGFR slope in each baseline eGFR group. Regardless of baseline eGFR, patients with eGFR slope < −10 mL/min/1.73 m2 were older and had consistently higher prevalence rates of hypertension, diabetes mellitus, and anemia than those with eGFR slope ≥ −10 mL/min/1.73 m2. Moreover, the large eGFR slope group had higher rates of prescription of loop diuretics and antihypertensive drugs, including calcium channel blockers, β-blockers, and renin-angiotensin system inhibitors.
Table 2Baseline characteristics of two groups according to eGFR slope ranging in each baseline eGFR group.
60≤ eGFR, n = 1650 (100%)
45≤ eGFR<60, n = 1947 (100%)
eGFR <45, n = 1069 (100%)
−10 ≤ eGFR slope n = 1255 (76%)
eGFR slope < −10 n = 395 (24%)
p-value
−10 ≤ eGFR slope n = 1788 (92%)
eGFR slope < −10 n = 159 (8%)
p-value
−10 ≤ eGFR slope n = 1029 (96%)
eGFR slope < −10 n = 40 (4%)
p-value
Age (years)
59.8 ± 13.2
63.1 ± 11.4
<0.001
62.3 ± 11.1
69.5 ± 9.9
<0.001
67.9 ± 10.7
73.3 ± 10.5
0.544
Sex, male, n (%)
522 (41.6)
202 (51.1)
0.001
1511 (84.5)
120 (75.5)
0.003
942 (91.5)
31 (77.5)
0.007
BMI, kg/m2
23.3 ± 3.8
23.6 ± 3.8
0.191
24.4 ± 3.4
24.2 ± 3.8
0.417
24.6 ± 3.5
24.5 ± 3.3
0.934
Coexisting disorders
Hypertension, n (%)
717 (57.1)
263 (66.6)
<0.001
1089 (60.9)
133 (83.6)
<0.001
743 (72.2)
33 (82.5)
0.102
Diabetes mellitus, n (%)
262 (20.9)
127 (32.2)
<0.001
442 (24.7)
62 (39.0)
<0.001
334 (32.5)
18 (45.0)
0.071
Dyslipidemia, n (%)
651 (51.9)
240 (60.8)
0.001
975 (54.5)
99 (62.3)
0.036
563 (54.7)
15 (37.5)
0.024
Hyperuricemia, n (%)
196 (15.6)
67 (17.0)
0.286
548 (30.6)
55 (34.6)
0.173
545 (53.0)
19 (47.5)
0.302
Anemia, n (%)
32 (2.5)
29 (7.3)
<0.001
48 (2.7)
18 (11.3)
<0.001
68 (6.6)
10 (25.0)
<0.001
AF, n (%)
214 (17.1)
57 (14.4)
0.125
442 (24.7)
29 (18.2)
0.039
273 (26.5)
5 (12.5)
0.030
SHD, n (%)
603 (48.0)
269 (68.1)
<0.001
1013 (56.7)
130 (81.8)
<0.001
686 (66.7)
30 (75.0)
0.177
Ischemic heart diseases, n (%)
375(29.9)
204(51.6)
<0.001
741(41.4)
104(65.4)
<0.001
465(45.2)
21(52.5)
0.362
Valvular diseases, n (%)
187(14.9)
68(17.2)
0.267
219(12.2)
33(20.8)
0.002
210(20.4)
12(30.0)
0.142
Cardiomyopathy, n (%)
91(7.3)
25(6.3)
0.532
152(8.5)
17(10.7)
0.347
130(12.6)
9(22.5)
0.069
Laboratory data
Baseline eGFR, mL/min/1.73 m²
71.8 ± 10.7
78.2 ± 14.7
<0.001
52.4 ± 4.1
53.4 ± 4.0
0.002
37.5 ± 6.7
38.3 ± 4.9
0.085
eGFR at 1-year, mL/min/1.73 m²
71.9 ± 12.9
60.1 ± 13.2
0.995
53.1 ± 7.1
39.1 ± 5.6
0.001
39.4 ± 9.0
24.5 ± 4.8
0.006
eGFR slope, mL/min/1.73 m²
−1.0 (−5.0, 3.9)
−14.9 (−20.1, −12.0)
<0.001
0.0 (−3.4, 4.2)
−13.2 (16.6, −11.2)
<0.001
0.81 (−2.4, 5.1)
−13.6 (−15.6, −1.6)
<0.001
BNP, pg/ml
48 (18, 121)
72 (29, 187)
0.057
57 (23, 361)
138 (44, 361)
<0.001
109 (44, 301)
130 (42, 549)
0.735
Medication at baseline
Ca blocker, n (%)
387 (30.8)
150 (38.0)
0.005
587 (32.8)
81 (50.9)
<0.001
408 (39.7)
23 (57.5)
0.019
RAS inhibitor, n (%)
485 (38.6)
224 (56.7)
<0.001
767 (42.9)
119 (74.8)
<0.001
562 (54.6)
33 (82.5)
<0.001
Statin, n (%)
331 (26.4)
152 (38.5)
<0.001
528 (29.5)
70 (44.0)
<0.001
313 (30.4)
14 (35.0)
0.323
Loop diuretic, n (%)
135 (10.8)
76 (19.2)
<0.001
185 (10.3)
46 (28.7)
<0.001
253 (24.6)
19 (47.5)
0.002
β blocker, n (%)
360 (28.7)
174 (44.1)
<0.001
641 (35.9)
98 (61.3)
<0.001
454 (44.1)
24 (60.0)
0.035
Antiplatelet, n (%)
443 (35.3)
236 (59.7)
<0.001
788 (44.1)
105 (65.6)
<0.001
511 (49.7)
30 (75.0)
0.001
Anticoagulant, n (%)
264 (21.0)
76 (19.2)
0.244
438 (24.5)
52 (32.5)
0.016
344 (33.4)
10 (25.0)
0.174
Echocardiographic Parameters
LVEF, n (%)
65.5 ± 11.9
64.2 ± 12.4
0.076
62.8 ± 15.5
60.0 ± 15.6
0.003
59.9 ± 15.2
53.4 ± 19.8
<0.001
Consecutive values are presented as mean ± standard deviation or median (inter-quartiles).
eGFR, estimated glomerular filtration rate; BMI, body mass index; AF, atrial fibrillation; SHD, structural heart disease; BNP, brain natriuretic peptide; Ca blocker, calcium channel blocker; RAS inhibitor, renin–angiotensin system inhibitor; LVEF, left ventricular ejection fraction.
During the follow-up period of 2.7 ± 0.6 years, 187 cardiovascular events occurred in total study patients (1.5 per 100 patient-years), which included 81 acute coronary syndrome, 42 heart failure, 35 stroke, and 29 cardiovascular death (Table 3). The HRs of eGFR slope < −10 mL/min/1.73 m2 for cardiovascular events with reference to eGFR slope ≥ −10 mL/min/1.73 m2 was 1.83 (95% CI, 1.29 – 2.59; p = 0.001) in univariable Cox regression analysis and 2.12 (95% CI, 1.45 – 3.10; p < 0.001) in multivariable Cox regression analysis (Table 4).
Table 3Details of the study endpoint according to the levels of baseline eGFR and eGFR slope.
Total
High eGFR
Intermediate eGFR
Low eGFR
eGFR slope (mL/min/1.73 m²)
≥−10
< −10
≥−10
< −10
≥−10
< −10
Cardiovascular events, n (%)
187 (100)
23 (100)
18 (100)
58 (100)
16 (100)
66 (100)
6 (100)
Acute coronary syndrome, n (%)
81 (43)
5 (22)
8 (44)
23 (40)
12 (76)
33 (50)
0 (0)
Heart failure, n (%)
42 (22)
3 (13)
5 (28)
13 (22)
2 (12)
15 (23)
4 (67)
Stroke, n (%)
35 (19)
11 (48)
4 (22)
14 (24)
0 (0)
6 (9)
0 (0)
Cardiovascular death, n (%)
29 (16)
4 (17)
1 (6)
8 (14)
2 (12)
12 (18)
2 (33)
eGFR, estimated glomerular filtration rate.
n (%) indicates the number of events and the ratio to the number of cardiovascular events in each category.
In the multivariable model, the category of eGFR slope <-10 mL/min/1.73 m2 and baseline eGFR levels, and covariables with p-value <0.10 in the univariable models (age, hypertension, diabetes mellitus, and SHD) were forcedly entered.
The incidence rates of cardiovascular events (per 100 patient-years) in eGFR slope ≥ −10 mL/min/1.73 m2 and < −10 mL/min/1.73 m2 were 0.7 and 1.6 for high eGFR group, 1.2 and 3.6 for intermediate eGFR group, and 2.4 and 6.1 for low eGFR group. Fig. 3 shows the Kaplan-Meier curves of cardiovascular events where the difference between eGFR slope ≥ -10 mL/min/1.73 m2 and < −10 mL/min/1.73 m2 were significant in the high (log rank test, p = 0.005), intermediate (log rank test, p < 0.001), and low (log rank test, p = 0.017) eGFR groups. Tables 5 and 6 show the results of Cox regression analysis for cardiovascular events separately analyzed in each eGFR level. In univariable Cox regression analysis, the HRs of eGFR slope < −10 mL/min/1.73 m2 with reference to eGFR slope ≥ −10 mL/min/1.73 m2 were 2.37 (95% CI, 1.28 – 4.40; p = 0.006) in the high eGFR group, 3.10 (95% CI, 1.78 – 5.40; p < 0.001) in the intermediate eGFR group, and 2.66 (95% CI, 1.15 – 6.13; p = 0.022) in the low eGFR group (Table 5). In multivariable Cox regression analysis, the HRs of eGFR slope < −10 mL/min/1.73 m2 with reference to eGFR slope ≥ −10 mL/min/1.73 m2 were 1.86 (95% CI, 0.99 – 3.50; p = 0.052) in the high eGFR group, 2.15 (95% CI, 1.20 – 3.85; p = 0.010) in the intermediate eGFR group, and 2.22 (95% CI, 0.95 – 5.15; p = 0.064) in the low eGFR group (Table 6).
Fig. 3Kaplan-Meier curves for cardiovascular events in high, intermediate, and low eGFR group stratified by eGFR slope.
In the multivariable models, the category of eGFR slope <-10 mL/min/1.73 m2, and covariables with p-value <0.10 in the univariable models in each eGFR level (sex, diabetes mellitus, SHD, and AF for high eGFR; age, sex, hypertension, diabetes mellitus, and SHD for intermediate eGFR; and age and SHD for low eGFR, respectively) were forcedly entered.
Sensitivity analysis for existence or absence of SHD
As a sensitivity analysis, the impact of eGFR slope on cardiovascular events according to the eGFR levels was further evaluated in patients with and without SHD. In patients with SHD (Online Fig. 1), the difference in the incidence of cardiovascular events between eGFR slope ≥ −10 mL/min/1.73 m2 and <−10 mL/min/1.73 m2 were evident in all eGFR groups, where the impact of eGFR slope <−10 mL/min/1.73 m2 was mostly similar as shown in the total patients (Fig. 3). However, in patients without SHD (Online Fig. 2), the incidence of cardiovascular events was not significantly different in any of eGFR groups.
Discussion
The major findings of the present study were as follows: 1) the rates of cardiovascular events after 1-year follow-up were higher in the group with low than high eGFR at baseline; 2) the eGFR had a larger negative slope in the high eGFR group than the low eGFR group; 3) a large negative eGFR slope (< -10) was associated with approximately 2-fold higher risk of CVD, which was mostly consistent regardless of eGFR level.
Decline of eGFR over time and incidence of CVD
A number of studies have reported that low eGFR is associated with CVD or all-cause mortality. For example, Weiner et al. reported that patients with eGFR <60 mL/min/1.73 m2 had higher risks of all-cause mortality and CVD compared to those with eGFR ≥60 mL/min/1.73 m2 (HR 1.19; 95% CI 1.07 – 1.32) [
]. On the other hand, the decline in renal function over time in CKD patients has also been reported to be associated with increased risks of all-cause mortality [
]. In the present study, eGFR slope < −10 mL/min/1.73 m2 was associated with increased risk of cardiovascular events (HR 1.83; 95% CI 1.29 – 2.59), consistent with the results of previous studies.
The association between the decline in eGFR and the incidence of CVD can be explained by several mechanisms. First, decline in eGFR would occur in patients without renal diseases via the progression of atherosclerosis. Therefore, the risk factors for CKD include various risk factors for atherosclerosis, such as age, hypertension, diabetes, and dyslipidemia. These are understandably risk factors of CVD [
]. Reversely, decline in eGFR could be affected by the treatment of CVD. For example, in patients with heart failure, decline in eGFR would occur as a result of treatment by diuretics [
Decline in eGFR and incidence of CVD: difference according to baseline eGFR
In the present study, the eGFR slope (decline in eGFR over time) was larger and more negative in patients with high eGFR than in those with low eGFR at baseline. This relationship suggests that the magnitude of variability depends on the size of the value of eGFR itself [
]. Matsushita et al. showed that, although decline in eGFR over time was a risk factor for all-cause mortality in patients with mildly impaired renal function, there was no association in those with normal renal function (eGFR ≥90 mL/min/1.73 m2; HR 0.96, 95%CI 0.80–1.16) [
. The controversial results may derive from the differences in the clinical characteristics in patients with normal eGFR.
In our study, approximately two-thirds had hypertension and a half had SHD including ischemic heart diseases even in high eGFR group. When we divided the study patients into those with and without SHD, the impact of a large negative eGFR slope (< −10) was extremely high in patients with SHD and no remarkable difference was observed in patients without SHD. Thus, it should be noted that the impact of a large negative eGFR slope (< −10) on cardiovascular events would be a matter irrespective of eGFR levels especially when the patients have SHD.
Limitations
This study had several limitations. First, the sample size of the present study was limited because of the lack of follow-up eGFR at 1 year. Second, our population was hospital-based and therefore was not representative of the general population. Third, eGFR could vary during the time course and therefore the measurement of eGFR slope with sampling only twice may have caused over- or underestimation [
]. Fourth, the impact of eGFR slope would differ according to the patient characteristics which would be one of the reasons of the controversy in the relationship between eGFR slope and cardiovascular events [
. Therefore, our data should be carefully interpreted especially taking into account the prevalence of SHD.
Conclusions
Decline in eGFR at 1-year follow-up was associated with incidence of CVD after 1-year follow-up regardless of baseline eGFR in patients who visited a cardiovascular hospital. Even in patients with high eGFR at baseline, large decline in eGFR showed a similar risk to those with intermediate or low eGFR at baseline, which were especially evident in patients with SHD.
Acknowledgments
We thank Shiro Ueda and Nobuko Ueda of Medical Edge Co Ltd, for assembling the database by Clinical Study Supporting System (CliSSS), and Yurika Hashiguchi, Takashi Osada, Hiroaki Arai, and Hiroshi Nakai for data management and system administration.
30411-1&id=gr1.jpg){kind=link}
30411-1&id=gr2.jpg){kind=link}
30411-1&id=gr3.jpg){kind=link}