Prognostic significance of the renal resistive index in the primary prevention of type II diabetes

Abstract

The renal resistive index has been demonstrated to predict the progression of renal disease and recurrence of major cardiac events in high‐risk cardiovascular patients, in addition to other comorbidities. We aimed to assess the prognostic significance of the renal resistive index in type 2 diabetic patients for primary prevention. From 2008 to 2011, patients with type 2 diabetes underwent cardiovascular evaluation, including renal resistive index assessment by renal Doppler ultrasound. The incidence of all‐cause death, cardiovascular events, dialysis requirement or a twofold increase in creatinine was recorded. Survival curves were estimated by the Kaplan‐Meier method. Two hundred sixty‐six patients were included; 50% of the patients were men, an HbA1C level of 8.1 ± 1.7% (65 ± 13.6 mmol/mol) and a serum creatinine level of 8 [7‐9] mg/L. The mean 24‐hour systolic blood pressure, 24‐hour diastolic blood pressure, and 24‐hour pulse pressure were 133.4 ± 16.7, 76.5 ± 9.4, and 56.9 ± 12.4 mm Hg, respectively. The median renal resistive index was 0.7 [0.6‐0.7] with a threshold of 0.7 predictive of monitored events. After adjustment of the 24‐hour pulse pressure, age and 24‐hour heart rate, a renal resistive index ≥0.70 remained associated with all‐cause death (hazard ratio: 3.23 (1.16‐8.98); P = .025) and the composite endpoint of major clinical events (hazard ratio: 2.37 (1.34‐4.18); P = .003). An elevated renal resistive index with a threshold of 0.7 is an independent predictor of a first cardiovascular or renal event in type 2 diabetic patients. This simple index should be implemented in the multiparametric staging of diabetes.

1. INTRODUCTION

The increasing burden of type 2 diabetes is mostly related to its cardiovascular and renal complications.1 The prevention strategy in this population should be based on multiparametric risk stratification that encompasses the different threatened organs. The renal resistive index (RRI) has been demonstrated to predict not only the progression of renal disease but also the recurrence of cardiac major events in chronic kidney disease, coronary, and heart failure patients,2, 3, 4 in addition to frequently associated comorbidities such as hypertension, diabetes mellitus, and aging.5, 6, 7 Although diabetes mellitus is associated with alterations in global hemodynamic parameters (eg, increased central aortic pressure and pulse wave velocity)8 that are directly expected to modify RRI,9, 10 the independent prognostic significance of diabetic‐related alterations in RRI remains to be investigated in primary prevention. Thus, we aimed to assess the prognostic significance of RRI regarding all‐cause death and major cardiovascular and renal events by exploring a cohort of patients with type 2 diabetes in primary prevention.

2. METHODS

From January 2008 to December 2011, we screened high‐risk asymptomatic diabetics for atherosclerotic lesions, as recommended by the French Society of Cardiology and the ALFEDIAM (Association de langue Française pour l’Etude du Diabète et des maladies Métaboliques).11 The patients had undergone scheduled cardiovascular evaluation, including the following:

1. Completion of a questionnaire regarding personal and family history of cardiovascular disease, lifestyle, habitus, and medical treatment. The different classes of antihypertensive medications [beta‐blockers, angiotensin‐converting enzyme inhibitors, angiotensin 2‐receptor antagonists, diuretics (eg, thiazides and spironolactone), calcium‐channel blockers, alpha‐blockers, and central acting agents] were recorded to create a treatment score, defined as the sum of the number of different antihypertensive agents taken by a patient each day.

2. Clinical and anthropometrical data recording.

3. Twenty‐four‐hour blood pressure (BP) monitoring. Twenty‐four‐hour BP monitoring was performed using the Spacelab 90207 device, with a BP measurement programmed every 15 minutes. The awake period was defined as 7 am to 10 pm, and the sleeping period as 10 pm to 7 am The daytime BP used in the analyses was the mean of the BP measurements performed during the awake period, and the nighttime BP was the mean of the BP measurements performed during the sleeping period. Nocturnal hypertension was defined as a systolic BP ≥120 mm Hg and/or a diastolic BP ≥70 mm Hg. The dipping status was determined as a fall of at least 10% of the mean nighttime BP value compared with the daytime value.

4. Oscillometric clinical BP measurement using the Accutorr plus MINDRAY* device after 5 minutes of rest, according to European guidelines.12

5. Biological evaluation, including the plasma creatinine level, fasting glucose level, HBA1C serum level, and serum lipid profile. The glomerular filtration rate (eGFR) was estimated using the four‐component Modification of Diet in Renal Disease equation incorporating age, race, sex, and serum creatinine concentration.13 Natriuresis evaluation and urinary protein excretion were measured in urine samples.

6. Conventional resting electrocardiography in all the patients. Electric left ventricular hypertrophy was defined by the Sokolow‐Lyon formula (SV1 + RV5 or RV6 > 35 mm), Cornell product and size of the r wave in the AVR lead.

7. Comprehensive transthoracic echocardiography performed by experienced physicians. The left ventricular ejection fraction was estimated according to Simpson's rules. The left ventricular mass was measured according to the American Society of Echocardiography and was normalized against height^2.7.14

8. Echo Doppler ultrasound of the supra‐aortic trunks, legs, and kidneys was performed in all the patients. RRI was calculated using the following formula: peak systolic velocity‐end diastolic velocity/peak systolic velocity in an interlobar artery. The mean RRI of the right and left kidneys was used for statistical analysis. At this time, the renal Doppler ultrasound was used to screen for significant renal artery stenosis.

9. All of the patients benefited from myocardial screening. In the case of a positive test by coronary angiography revealing significant coronary narrowing, the patient was excluded from the study (Figure 1).

Figure 1.

Flowchart of the study population

The patients had undergone yearly clinical follow‐up by different specialists (cardiologists, diabetologists, and nephrologists) at our institution. The clinical outcomes were determined at the end of the follow‐up period between June and November 2018 based on medical recordings. The primary endpoint was all‐cause death. A secondary major clinical endpoint was defined as the composite criterion of major cardiovascular events (MACE) or renal progression. The MACE endpoint included cardiovascular death, stroke, acute coronary syndrome, acute limb ischemia, and hospitalization for cardiac failure. The renal progression endpoint was defined as dialysis requirement or a doubling of the serum creatinine level, which was based on the annual routine creatine measurement—that is, the last creatinine measured at the time of the most recent clinical consultation. MACE and renal progression were analyzed separately. Adjudication was made for each outcome by analyzing the medical recording by two cardiologists not involved in the data analysis.

2.1. Statistical analysis

Quantitative variables were described using means ± standard deviation or medians and interquartile range. Distribution normality was verified using graphs and the Shapiro‐Wilk test. The RRI cutoff of 0.70 was validated for the different clinical events as the optimal cutoff value by maximizing the hazard ratio. We compared the population with an RRI < 0.70 with the population with an RRI ≥ 0.70. The categorical variables were described using frequencies and percentages. In the case of a sufficient sample size, the quantitative variables were compared between RRI groups using Student's t test. In the case of non‐normality of data, nonparametric Wilcoxon tests were used.

Comparisons of the categorical variables were performed using chi‐squared test when the sample size was sufficient. In the case of nonvalidity of the tests (expected counts < 5), Fisher's exact test was used.

Survival curves were estimated using the Kaplan‐Meier method.

The effect of the RRI groups on the different studied events (all‐cause death, 1 combined endpoint of the major clinical events including MACE or renal progression, MACE and renal progression) was determined using hazard ratios (unadjusted and adjusted on prespecified confusing factors: 24‐hour pulse pressure, age and 24‐hour heart rate) calculated using the Cox proportional hazards model. The optimum threshold for RRI was selected for each event using maximization of the hazard ratio algorithm.

The threshold of significance was 0.05. The analyses were performed using SAS software version 9.4 (SAS Institute).

3. RESULTS

3.1. Global characteristics

Two hundred sixty‐six patients were enrolled in our cohort. The study flowchart is summarized in Figure 1, and the patient characteristics are presented in Table 1. The patients were middle aged and comprised 50% men; most had diabetes longer than 10 years. The median of the average RRI value of both kidneys was 0.7 [0.6‐0.7]: 166 patients had a low RRI (<0.7), and 100 patients had an RRI ≥ 0.70. Patients with a lower RRI were younger, with less history of hypertension, a shorter duration of diabetes mellitus, and a higher proportion of current smokers (P = .013). The body mass index and abdominal circumference were similar between the high and low RRI groups. The patients with an elevated RRI displayed a lower hemoglobin level and glomerular filtration rate.

Table 1.

Baseline characteristics of the study population

Parameter	Overall population N = 266	Renal resistive index < 0.70 N = 166	Renal resistive index ≥ 0.70 N = 100	P
Clinical data
Age, years	65.4 ± 9.8	62.1 ± 8.7	71 ± 8.9	<.001
Men	133 (50)	89 (53.6)	44 (44)	.13
Current smoker	71 (26.7)	53 (31.9)	18 (18)	.013
Familial history of artery disease	48 (18)	33 (19.9)	15 (15)	.32
History of hypertension	247 (92.9)	150 (90.4)	97 (97)	.042
Treatment for sleep apnea	19(7.1)	15(9)	4(4)	.12
Body Mass Index, Kg/m2	31.85 ± 6.51	31.4 ± 6.66	32.1 ± 6.27	.75
Abdominal circumference, cm	111.4 ± 15	111 ± 15.1	112.1 ± 14.7	.58
Duration of diabetes, months	140.2 ± 90.5	118.6 ± 74.7	175 ± 102.5	<.001
Ankle brachial index	1.1 (1; 1.2)	1.1 (1; 1.2)	1.1 (1; 1.2)	.67
Biological data
Hemoglobin level, g/dL	13.6 ± 1.5	14 ± 1.5	13 ± 1.4	<.001
Us‐C‐reactive protein, mg/L	2.4 (1; 5.4)	2.2 (1; 4.5)	2.6 (0.9; 6.4)	.31
Creatinine level, mg/L	8 (7; 9)	8 (7; 9)	8 (7; 10)	.047
eGFR, mL/mn/1.73 m2	90 (77; 108)	92 (81; 110)	82 (67; 104)	<.001
Creatininuria, mg/L	741 (532; 1084)	793 (593.5; 1106)	668 (460; 935)	.018
Natriuresis, mEq/L	79 (54; 109)	83 (52; 115)	78.5 (56.5; 99.5)	.34

Continuous variables are presented using mean ± standard deviation or median (Q1; Q3) otherwise. Categorical variables are presented using N (%).

3.2. Medical treatment

The prescription of the different classes of drugs is presented in Table 2. Overall, the median number of antihypertensive drugs was 2 [1‐3] drugs per day, with a higher rate of antihypertensive drug intake in the ≥0.70 RRI group (P < .001). The prescription of the renin‐angiotensin system inhibitor was high (88.7% of the entire population) regardless of RRI. The proportions of calcium antagonists (P = .005) and beta‐blockers (P = .039) were higher in the ≥0.70 RRI group. Conversely, no difference was found between groups in the lipid‐lowering agents or antihyperglycemic drug prescription.

Table 2.

Cardiovascular and diabetes treatment at baseline

Parameter	Overall population N = 266	Renal resistive index < 0.70 N = 166	Renal resistive index ≥ 0.70 N = 100	P
Antihypertensive drugs
Angiotensin‐converting enzyme inhibitors or angiotensin II receptor antagonists	236 (88.7)	145 (87.3)	91 (91)	.36
Angiotensin‐converting enzyme inhibitors	123 (53.5)	80 (56.7)	43 (48.3)	.21
Angiotensin II receptor antagonists	107 (46.5)	61 (43.3)	46 (51.7)
Calcium antagonists	131 (49.4)	71 (42.8)	60 (60.6)	.005
Diuretics	100 (37.6)	56 (33.7)	44 (44)	.094
Aldosterone antagonists	21 (7.9)	11 (6.6)	10 (10)	.32
Beta‐blockers	52 (19.5)	26 (15.7)	26 (26)	.039
Treatment score	2 (1; 3)	2 (1; 3)	2 (1.5; 3)	<.001
Lipid lowering
Statins	226 (95.4)	137 (94.5)	89 (96.7)	.54
Fibrates	11 (4.6)	8 (5.5)	3 (3.3)
Antiplatelets
Aspirin	186 (69.9)	109 (65.7)	77 (77)	.051
Antihyperglycemic
Metformin	169 (63.5)	111 (66.9)	58 (58)	.15
Sulfonylureas	92 (34.6)	52 (31.3)	40 (40)	.15
Thiazolidinedione	21 (7.9)	13 (7.8)	8 (8)	.96
Repaglinide	71 (26.7)	46 (27.7)	25 (25)	.63
Incretin	77 (28.9)	49 (29.5)	28 (28)	.79
Insulin	121 (45.5)	72 (43.4)	49 (49)	.37

Continuous variables are presented using mean ± standard deviation or median (Q1; Q3) otherwise. Categorical variables are presented using N (%).

3.3. Target organ damage and cardiovascular risk factors

The mean clinical systolic and diastolic BP values were 145.6 ± 21.4 and 85.1 ± 13.0 mm Hg at inclusion, respectively. Table 3 presents the 24‐hour BP monitoring data and target organ damage data. The different 24‐hour BP parameters were higher in the RRI ≥ 0.70 group, which displayed a lower heart rate and higher median height of the r wave size in the AVL lead. Conversely, left ventricular hypertrophy evaluation by echocardiography, left ventricular ejection fraction, level of proteinuria, and biological cardiovascular risk factors (LDL cholesterol and HBA1C) were similar between the groups.

Table 3.

Cardiovascular risk factor control and target organ damage at baseline

Parameter	Overall population N = 266	Renal resistive index < 0.70 N = 166	Renal resistive index ≥ 0.70 N = 100	P
Blood pressure data
24‐h systolic BP, mm Hg	133.4 ± 16.7	130.9 ± 15.9	137.5 ± 17.4	.002
24‐h diastolic BP, mm Hg	76.5 ± 9.4	78.2 ± 8.9	73.7 ± 9.4	<.001
24‐h pulse pressure, mm Hg	56.9 ± 12.4	52.7 ± 10.4	63.8 ± 12.4	<.001
Daytime systolic BP, mm Hg	137.1 ± 16.6	134.6 ± 16	141.1 ± 17	.002
Daytime diastolic BP, mm Hg	79.6 ± 9.8	81.3 ± 9.4	76.6 ± 9.6	<.001
Nighttime systolic BP, mm Hg	125.9 ± 18.2	123.8 ± 17.1	129.7 ± 19.4	.012
Nighttime diastolic BP, mm Hg	70.5 ± 9.8	72.2 ± 9.4	67.5 ± 9.7	<.001
Biological data
LDL cholesterol, g/L	0.9 ± 0.3	1 ± 0.3	0.9 ± 0.3	.094
HDL cholesterol, g/L	0.5 ± 0.2	0.5 ± 0.2	0.5 ± 0.2	.52
HbA1C, %	8.1 ± 1.7	8 ± 1.6	8.3 ± 1.9	.23
HBA1C, mmol/mol	65 ± 13.6	64 ± 12.8	67 ± 15.3	.23
Proteinuria, g/L	0.1 (0; 0.1)	0.1 (0; 0.1)	0.1 (0; 0.2)	.63
Left ventricular hypertrophy data
Sokolow index, mm	13 (10; 17)	13 (10; 17)	13 (10; 16)	.96
Cornell product, mm*ms	1080 (800; 1501)	1073 (783; 1410)	1113 (855; 1598)	.19
r wave AVL lead, mm	5.5 (4; 8)	5 (3; 7)	6 (4; 9)	.010
Left ventricular mass, g/m2.7	38.7 ± 11	37.3 ± 9.7	41 ± 12.5	.068

Continuous variables are presented using mean ± standard deviation or median (Q1; Q3) otherwise. Categorical variables are presented using N (%).

3.4. Doppler ultrasound data

No hemodynamically renal artery stenosis was diagnosed at inclusion. The mean sizes of the right and left kidneys were 112.9 ± 11.1 and 115.6 ± 11.3 mm, respectively, and the median RRI values of each group were 0.7 [0.7‐0.8] and 0.6 [0.6‐0.7], respectively. We noted a higher rate of lower limb stenosis >50% (7.4% vs 15.2%, P = .046) in the ≥0.70 RRI group. Conversely, the rate of carotid stenosis more than 50% and mean aortic diameter were similar between the groups.

3.5. Prognostic data

The four major outcomes were studied with a 10‐year follow‐up. Overall, there were 23 (8.6% of the population) deaths, among which 7 (2.6%) were cardiovascular, 15 (Z5.6%) were hospitalizations for acute cardiac failure, 24 (9.0%) were myocardial infarctions, 22 (8.3%) were strokes, 12 (4.5%) were limb ischemia, and 25 (9.4%) patients showed renal progression. The ≥0.70 RRI group was associated with a higher risk of all‐cause death [HR (95% CI): 2.83 (1.22‐6.54); P = .015] in univariate analysis, and the association persists after adjustment for confounding factors (Table 4, Figure 2)—that is, 24‐hour pulse pressure, age and 24‐hour heart rate. The incidence of major clinical events was higher in patients with an elevated RRI [HR (95% CI): 2.07 (1.29‐3.31); P = .003], even after adjustment for confounding factors—that is, 24‐hour pulse pressure, age and 24‐hour heart rate (Table 4, Figure 3). This higher risk with elevated RRI was consistently observed for events comprising the composite criterion of major cardiovascular events—that is, MACE (Table 4, Figure 4) and renal progression (Table 4, Figure 5). After adjustment of the 24‐hour pulse pressure, diabetes duration, and baseline glomerular filtration rate, the higher risk with elevated RRI was only observed for renal progression and composite criterion of MACE and renal progression (Table 4). After adjustment of the 24‐hour pulse pressure, hemoglobin, and baseline glomerular filtration rate, the higher risk with elevated RRI persisted only for major clinical events and the composite criterion of MACE (Table 4). The univariate association with all‐cause mortality of traditional cardiovascular risk factors and principal variables used in our prognostic analysis are presented in Table S1. The 24‐hour blood pressure monitoring parameters were the only parameters associated with the risk of all‐cause mortality.

Table 4.

Unadjusted and adjusted hazard ratios (HR) for different outcomes of patients with a renal resistive index ≥ 0.7 (reference group)

	Unadjusted analysis		Adjusted analysisa		Adjusted analysisb		Adjusted analysisc
	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P
All‐cause death	2.83 (1.22‐6.54)	.015	3.23 (1.16‐8.98)	.025	2.32 (0.84‐6.37)	.10	1.98 (0.77‐5.06)	.16
Major clinical event	2.07 (1.29‐3.31)	.003	2.37 (1.34‐4.18)	.003	1.84 (1.04‐3.24)	.036	1.97 (1.13‐3.43)	.017
MACE	2.12 (1.24‐3.62)	.006	2.07 (1.08‐3.95)	.028	1.60 (0.85‐3.00)	.14	2.46 (1.30‐4.68)	.006
Renal progression	3.35 (1.48‐7.60)	.004	4.82 (1.89‐12.32)	.001	3.85 (1.39‐10.67)	.009	2.27 (0.92‐5.59)	.074

Abbreviations: HR, hazard ratio; MACE, major adverse cardiovascular event.

^aAdjustment on 24‐h pulse pressure, age and 24‐hour heart rate.

^bAdjustment on 24‐h pulse pressure, diabetes duration, and glomerular filtration rate at baseline.

^cAdjustment on 24‐h pulse pressure, hemoglobin, and glomerular filtration rate at baseline.

Figure 2.

Prognostic significance of the renal resistive index in the risk of death

Figure 3.

Prognostic significance of the renal resistive index in the risk of major clinical events (major cardiovascular event or renal progression)

Figure 4.

Prognostic significance of the renal resistive index in the risk of major cardiovascular events

Figure 5.

Prognostic significance of the renal resistive index in the risk of renal progression

4. DISCUSSION

Exploring the clinical prognostic significance of the RRI in a primary prevention diabetic population, we demonstrated that RRI is a strong and independent marker of long‐term major cardiovascular events and renal function impairment. Interestingly, RRI showed its independent association with outcome even after adjustment of its main confounder—that is, the pulse pressure.

RRI is dependent on systemic hemodynamic factors8, 9, 10, 11, 12, 13, 14, 15 and varies with the heart rate.16 A significant decrease in RRI was observed with an increasing heart rate. In our work, the heart rate was significantly lower (clinical and 24‐hour measurement) in the ≥0.7 RRI group based on a higher prescription rate of beta blocker in the ≥0.70 group. RRI is also dependent on left ventricular outflow. In a pathophysiological situation with decreased left ventricular outflow, RRI is low.17 In this work, the complete data of the systolic and diastolic Doppler blood flow indexes were not available, but the left ventricular ejection fraction was the same between the groups. Pulse pressure is linked to stroke volume18 and is the principal determinant of RRI.5 A higher pulse pressure in the RRI ≥ 0.7 group was not dependent on the stroke volume but probably reflected higher vascular stiffness. Systemic atherosclerosis severity is influenced by several cardiovascular risk factors. In our work, the patients were older in the group with high RRI, but the risk of clinical events persists after the adjustment of age. The prescription of renin‐angiotensin system inhibitors benefits cardiovascular prevention and influences the RRI level.19, 20 The prescription of this class of antihypertensive drug was equivalent between the groups.

An elevated RRI is associated with a higher risk of death in patients with severe atherosclerotic disease or cardiac failure.3, 4 Although RRI is altered in renal microcirculation dysfunction, RRI is also altered by general hemodynamic parameters such as pulse pressure and heart rate. Pearce et al showed in elderly patients of the Cardiovascular Health Study that RRI optimizes cardiovascular risk stratification.22 Thus, we showed, in a younger population of diabetic patients without a cardiovascular medical history, a significant association of RRI with the clinical outcomes even after adjustment of the pulse pressure.23, 24 Due to the observational design of our study, we could not explore the potential causal link between renal resistive index and mortality.

The data on the association with albuminuria and RRI are controversial in the diabetic population. RRI is high in populations with both decreased GFR and increased albumin excretion.25 In our work, we found no association between high RRI and albuminuria, probably because of preserved GFR. In our work, the threshold of 0.70 was found to be associated with a poor prognosis. In a population with chronic kidney disease, Toledo et al had already shown that this threshold identifies a population at high risk.26 We confirmed in a population with a higher glomerular filtration rate that this higher normal value of 0.70 is a strong predictor of events.

4.1. Limitations

The BP and heart rate data were not available at the time of the renal Doppler assessment. Thus, we used the parameters measured within the same consultation. The data on BP and glycemic control were not available during the follow‐up. Central hemodynamic parameters and aortic pulse wave velocity data were not available in our study. Although these parameters are expected to be substantially correlated with RRI, they should be integrated in further studies evaluating the prognostic influence of the renal resistive index. The number of events was quite low despite the long‐term follow‐up of our study, hampering multiple adjustments in Cox analysis. This low rate of events in type 2 diabetic patients in primary prevention might reflect the high rates of statin and antihypertensive drug prescriptions. The statistical power of the main analysis (effect of RRI ≥ 0.70 on overall survival with a 10‐year follow‐up) was 89.9% (overall survival at year 10 = 85.5% for RRI < 0.7 (N = 166) and 72.8% for RRI ≥ 0.7 (N = 100)). The number of covariates in our Cox model was based on an appropriate ratio between the number of events and covariates to prevent our model from overfitting. We chose to focus on the independent association between the renal resistive index and prognosis regardless of its pathophysiological cofounders, notably pulse pressure and age.

5. CONCLUSION

An elevated RRI with a threshold of 0.7 is an independent predictor of the first cardiovascular or renal event in type 2 diabetic patients. This simple index should be implemented in the multiparametric staging of diabetic patients.

AUTHOR CONTRIBUTIONS

Pr Claire MOUNIER‐VEHIER is the head of the vascular medicine and hypertension department. Pr Pierre FONTAINE is the head of the endocrinology, diabetology, and metabolism department. Pr Anne VAMBERGUE is also a responsible of the endocrinology, diabetology, and metabolism department. They were all at the origin of the creation of a clinical collaboration between cardiologists and diabetologists with building up this local registry. They are also involved in the clinical management of patients and have given methodological advices to elaborate this study. Dr Benedicte LELIEVRE has collected the data and has carried out the clinical follow‐up of the patients. Mr François MACHURON is a statistician specialized in medical studies. He realized the statistical analysis and given several advices for the method. Dr Guillaume LEDIEU and Dr Sandro NINNI are cardiologists and have insured the clinical follow‐up of the population and the realization of echocardiography. Dr Emilie MERLEN is a diabetologist and has insured the clinical follow‐up of the population. Dr Marie FRIMAT is a nephrologist and has insured the clinical follow‐up of the population. Pr François GLOWACKI is a nephrologist, and he gave methodological advices regarding the renal prognosis evaluation of our analysis. He is also involved in the clinical and biological follow‐up of the population. Pr David MONTAIGNE is the head of the clinical physiology and echocardiography department. He has supervised the realization of the Doppler ultrasound, the echocardiography, and the myocardial ischemia screening. He has given several advices for the method. He also takes part in the writing of this manuscript. He also deeply reviewed the final version of the manuscript. Dr Pascal DELSART is a cardiologist. He takes part in the realization of the renal Doppler ultrasound, echocardiography and in the clinical follow‐up of this cohort. He has written the manuscript.

Supporting information

 

Delsart P, Vambergue A, Ninni S, et al. Prognostic significance of the renal resistive index in the primary prevention of type II diabetes. J Clin Hypertens. 2020;22:223–230. 10.1111/jch.13819