Renal resistive index in hypertensive patients

Abstract

Spectral Doppler ultrasonography provides the evaluation of renal resistive index (RRI), a noninvasive and reproducible measure to investigate arterial compliance and/or resistance. RRI seems to possess an important role in the evaluation of diverse cases of secondary hypertension. In essential hypertension, RRI is associated with subclinical markers of target organ damage and reflects renal disease progression beyond albuminuria and creatinine clearance. Also, RRI can estimate cardiovascular and renal risk. The evaluation of RRI may also help the therapeutic decisions. Given its simple assessment, RRI emerges as a simple method and a “multifunctional” tool that could help on the cardiovascular risk evaluation of the hypertensive patient.

1. INTRODUCTION

Spectral Doppler ultrasonography provides the evaluation of renal resistive index (RRI), a noninvasive and reproducible measure to investigate arterial compliance and/or resistance.1 It is calculated with the following formula: (peak systolic velocity‐end diastolic velocity)/peak systolic velocity (Figure 1). The mean value of three measurements at each kidney is usually taken under consideration. A RRI value 0.6 ± 01 (mean ± SD) has usually been considered as normal, with a value of 0.70 being regarded as the upper normal threshold.2 However, this threshold increases with age and a recent population‐based study has established reference values for RRI, according to age and sex.2 RRI is evaluated in segmental or interlobar arteries.3 However, it is generally recommended that sampling for RRI should be done at the level of the arcuate or interlobar arteries, as resistance to blood flow progressively increases from the hilar arteries toward peripheral parenchymal vessels.4 Several factors can influence the value of RRI (Table 1). A number of studies indicate that this index actually reflects systemic hemodynamics and depends on the aortic pulse pressure, which is affected by parameters like age, presence of hypertension (HTN), or diabetes.5 Other systemic factors such as severe aortic stenosis and increased blood volume can cause an increase in RRI.3, 6 Moreover, bradycardia increases RRI as there is more time for the diastolic flow to decrease. In patients with widespread atherosclerosis or reduced vascular compliance, RRI may be increased even with normal kidney function. Furthermore, increased interstitial and venous pressure representing the renal capillary wedge pressure result in high RRI.3, 6 An animal study provides evidence that RRI is dramatically affected by pulse pressure and by renal interstitial pressure.7

Figure 1.

RRI measurement. With the permission of Journal of Hypertension

Table 1.

Factors influencing RRI

Systemic	Renal
Pulse pressure	Interstitial pressure
Atherosclerosis	Renal venous pressure
Vascular compliance	Renal artery compliance
Heart rate
Blood volume
Aortic stenosis

The purpose of this review is to summarize recent evidence on the role of RRI in the evaluation of hypertensive disease, both secondary and essential, in order to support its wider use in the clinical management of a hypertensive patient.

2. RRI AND SECONDARY HYPERTENSION

In the setting of renal parenchymal disease, an important cause of secondary HTN, ultrasound findings do not allow for a specific etiological diagnosis. RRIs are typically elevated in advanced diabetic nephropathy, whereas they are often normal in the early stage of the disease.8 An elevated RRI (≥0.70) is usually associated with impaired renal function, increased proteinuria, and poor prognosis.9 Beyond the established use of B‐mode ultrasound imaging for detection of urinary obstruction, in hydronephrosis, which can be a cause of secondary HTN, a higher RRI in the hydronephrotic kidney may support the diagnosis. A mean RRI > 0.7 and a difference >0.06‐0.08 between the average RRI of the two kidneys strongly suggests obstruction.8 In final stages of chronic kidney disease (CKD), Doppler reveals globally reduced parenchymal perfusion and increased RRIs values, indicating disease irreversibility and poor prognosis.8

In case of suspected renovascular HTN, an important cause of secondary HTN, there is a difference in RRI between the two kidneys of at least 0.05.10 Duplex ultrasound can miss important vascular lesions, resulting in “false negatives,” especially in obese, hypoechogenic patients or when the operator is not very experienced.11 The imaging of the renal arteries in order to measure RRI may be difficult, even in experienced hands. Fibromuscular dysplasia, especially when distal renal artery branches are involved, may also cause “false‐negative” results. Finally, tachycardia, which is not rare in patients with various comorbidities does not allow the diastolic flow to fully decrease, thus lowering the RRI and providing “false‐negative” results.9 However, although highly operator‐dependent, Doppler assessment of RRI is a noninvasive method of low cost, widely available and suitable for outpatient use.

Furthermore, a recent study in 181 hypertensive subjects without any evidence of renal artery stenosis has evaluated the utility of the difference between the resistive index in the segment arteries of the kidney and the spleen as an additional parameter in the Doppler evaluation of renal artery stenosis. The authors conclude that this parameter is easily determinable, with high specificity and sensitivity and improves the diagnostic accuracy in the detection of renal artery stenosis.12

Beyond the diagnosis of renovascular HTN, RRI has been also proposed as a marker of possible benefit from revascularization. According to a prospective study in patients with a >50% unilateral or bilateral renal artery stenosis who underwent renal angioplasty or surgery, a RRI value of at least 0.8 before revascularization was a strong predictor of renal function impairment and failure of blood pressure (BP) lowering, despite the correction of renal artery stenosis. Conversely, lower resistance‐index values were associated with an improvement in both renal function and BP after the intervention.13 However, subsequent studies have not supported this notion. As many “nonrenal” factors affect RRI, there is currently a controversy on whether RRI can predict the success of intervention.14 On the other hand, in hypertensives with unilateral atherosclerotic renal artery stenosis, 12 months after revascularization, RRI > 0.73 in the contralateral kidney was an independent predictor of renal failure outcome, whereas it did not predict BP outcome.15 The pathophysiological background is that long‐standing HTN may cause nephrosclerosis or glomerulosclerosis, thus increasing vascular resistance in both the affected and the unaffected kidney. These HTN‐induced structural alterations in smaller renal arteries or arterioles distal to the renal artery stenosis may represent a possible cause of poor response to interventional treatment.

Moreover, regarding the endocrine causes of secondary HTN, a recent study has shown that RRI was significantly higher in patients with aldosteronoma.16 The resistive indices of main, hilum, and interlobar arteries were significantly reduced 1 month after adrenalectomy and remained stable for 12 months. Of note, the initial RRI was significantly associated with long‐term postoperative resistant HTN.

3. RRI AND TARGET ORGAN DAMAGE IN ESSENTIAL HTN

An elevated RRI has been associated with subclinical signs of renal organ damage in untreated patients with primary HTN and normal renal function, demonstrating a direct relationship with urine albumin excretion. Moreover, the increased impedance to blood flow at the parenchymal level is often associated with a mild reduction in glomerular filtration rate (GFR), increased albuminuria or both.17 In 133 hypertensives, each 0.1 increase in RRI was associated with a 5.4‐fold increase in the relative risk of albuminuria.5 Apart from these, RRI has been associated with faster decline of renal function in patients with proteinuric CKD or in diabetics with microalbuminuria even when GFR is still within normal values.1, 18 It is worth mentioning that in cases of mild‐to‐moderate renal dysfunction, RRI is superior to renal function estimation alone in prediction of CKD progression and poor outcome.19 Another study in 66 patients with essential HTN demonstrated a significant relationship between high RRI and future increase in urinary albumin excretion.20 RRI was the only variable that significantly predicted a >50% increase in the urinary albumin‐to‐creatinine ratio over 2 years and the optimal cut‐off value of RRI that predicted this increase was 0.71 (sensitivity 52.4% and specificity 84.4%).21 This cut‐off value is in line with other studies reporting that RRI values >0.7 are more prevalent in patients with left ventricular hypertrophy or with advanced carotid atherosclerosis and is associated with higher mortality in hypertensive patients with CKD without clinically significant renal artery stenosis.20, 22, 23 In a recent prospective observational cohort study in diabetic and hypertensive subjects, dynamic RRI, a new ultrasound‐based biomarker consisting in the change in resistive index after sublingual nitrate administration and conceivably representing an index of renal vasodilating capacity, was found to be able to predict microalbuminuria onset and, most interestingly, was found early compromised in diabetics developing microalbuminuria.24 Thus, it seems that renal vasodilating capacity is reduced in diabetics before the onset of established renal damage and in the presence of normal RRI values, meaning that functional rather than structural alterations might be already present, indicating a subclinical stage of renal damage. These data suggest the benefit from RRI assessment in the diagnosis of subclinical vascular damage, even before the onset of albuminuria.25 Dynamic RRI, however, is not yet established and caution is needed in the interpretation of the results.

In the context of primary HTN, an increased RRI is also indicative of the presence of extrarenal hypertensive organ damage such as left ventricular hypertrophy and carotid intima‐media thickening. In 288 essential hypertensives, each standard deviation increase in RRI gave a 47% higher odds of having left ventricular hypertrophy.26 A recent study highlighted a close independent correlation between RRI and severity of carotid atherosclerotic disease in a population of hypertensive subjects.27 There was a stepwise increase in RRI corresponding to the groups of progressive severity of carotid atherosclerosis. These observations strengthen the concept that RRI is a marker of systemic vascular changes and an increased RRI could be considered as a marker of generalized atherosclerotic vessel damage beyond the kidney.27 In the same lines, an increased RRI in essential HTN has been associated with increased central pulse pressure and aortic pulse wave velocity, ambulatory arterial stiffness index, and nondipping profile.4, 5, 28 RRI varies directly with arterial stiffness or impedance.6 In a recent study conducted in 264 hypertensive patients with and without impaired renal function, intrarenal resistive index was significantly and positively associated with large arterial stiffness even after adjustment for several confounding factors. According to the authors, RRI may be elevated in hypertensive patients even without CKD and progressively increases as renal function deteriorates.29 The possible underlying pathophysiological mechanism of the association between RRI and arterial stiffness is that increased arterial stiffness might predispose the renal circulation to a greater hemodynamic pressure, mainly pulse pressure, leading to higher renal vascular resistance.30 Besides, it has been suggested that RRI is an expression of renal pulsatile flow.5 On the other hand higher RRI may in the long term contribute to systemic arterial stiffening, possibly through renal dysfunction.30 This pathophysiological linkage between macrovascular and microvascular impairment has also been observed in a recent cross‐sectional study of 202 hypertensive patients, where RRI was positively correlated with pulse wave velocity, ambulatory arterial stiffness index and 24‐hour pulse pressure, as well as with atherosclerotic burden and endothelial dysfunction measured by serum asymmetric dimethylarginine.31 More interestingly, in the multivariate analysis arterial stiffness was an independent determinant of increased RRI, suggesting an independent association between renal hemodynamics and arterial stiffness.31

Beyond this, it seems that there is a direct interaction between the heart and peripheral circulation in the kidney. According to another recent study in 171 unselected subjects, RRI was significantly and independently associated with central pulse pressure and left ventricular systolic and diastolic Doppler blood flow indexes. RRI was significantly increased with systolic and early diastolic intracardiac Doppler blood flow.32 These findings imply that in addition to the anthropometric characteristics, cardiac hemodynamic factors influence the intrarenal arterial Doppler waveform patterns. The authors suggest that exposure of small renal arteries to high blood flow in addition to high pulsatile pressure leads to increased RRI and might in long run result to microvascular damage and, therefore, renal insufficiency.32 In the same lines, in a total of 449 untreated hypertensives, left ventricular dilatation has been associated with subclinical renal damage, namely albuminuria and early intrarenal vascular changes, expressed as the renal volume to RRI ratio.33 This ratio of ultrasound determined renal volume to intra parenchymal resistive index has been proposed as a measure of nephroangiosclerosis and has been shown to be useful in identifying patients characterized by reduced kidney volume and increased renovascular stiffness.33

4. RRI AND CARDIOVASCULAR AND RENAL OUTCOMES

An increasing number of longitudinal studies demonstrate that an increased RRI, indicative of impaired renal hemodynamics, has a prognostic role for CV morbidity, mortality and renal outcomes in essential hypertensive and CKD patients, in addition to albuminuria and eGFR, independently of the traditional risk factors.1, 26, 34, 35 Interestingly, in a 3 years follow‐up study that included 426 essential hypertensives with no previous CV disease, the relationship between high RRI and CV and renal outcomes was significant, even in the subgroup with eGFR < 60 mL/min per 1.73 m². The combination of high RRI and low eGFR was a powerful independent predictor of worse outcome.36 In another recent longitudinal study including hypertensive CKD patients without renal artery stenosis, over two third of patients had RRI ≥ 0.70. Older age, female sex, and multiple comorbid conditions were associated with higher RRI. The analysis also showed that, during a median follow‐up of 2.2 years, RRI ≥ 0.70 was significantly associated with increased all‐cause mortality.23 The fact that this association was more pronounced among younger patients and those with stage 3 CKD makes this biomarker even a more exciting discovery. Of note, those with RRI ≥ 0.70 had lower eGFR compared to those with RRI < 0.70, but this relationship with eGFR was not seen in the multivariate analysis. Thus, an elevated RRI among those with CKD is not simply because of a lower eGFR.37 Interestingly, the Cox model for all‐cause mortality was simultaneously adjusted for systolic and diastolic BP. This suggests that RRI has prognostic value above and beyond pulse pressure, which is a representative of arterial stiffness.37 The background pathophysiology circle includes reduced vascular compliance, increased renal vascular resistance, arteriosclerosis, tubulointerstitial disease and a reduced capillary surface area that results in constrained area of the vascular bed and increased vascular resistance.37

On the other hand, in a recent study, the routine Duplex examinations among stage 2‐4 CKD patients did not improve risk prediction for progression to end‐stage renal disease beyond a validated equation.38 There is the notion that as RRI is highly influenced by extrarenal factors such as cardiac function and systemic arterial compliance, with pulse pressure being the principal determinant, its prognostic value has been related to vascular disease that is often systemic.6 However, in a 7 years follow‐up study of 726 Americans from the Cardiovascular Health Study Renal, Doppler signals from the renal parenchyma showed significant associations with subsequent CV events after controlling for other significant factors, such as renal function, HTN, diabetes and history of CV events.25 RRI has been also examined in the setting of renal transplantation. In a prospective study, RRI was related to recipient age and central hemodynamic factors but not to graft function; thus, the RRI was associated with recipient survival but not with graft survival.39

In addition, RRI has been also correlated with diastolic dysfunction and heart failure with preserved ejection fraction (HFpEF). In a large sample of hypertensives, a positive correlation was found between RRI and left ventricular mass index and a negative correlation with tissue Doppler imaging diastolic parameters.22 Taking it a step further, another study showed that RRI was independently associated with HFpEF. Increased RRI was an independent predictor of poor outcome even after adjustment for mean BP, hemoglobin concentration and GFR.34 These results reinforce the notion that RRI is a marker of extrarenal pathology.

Regarding the notably interesting issue of resistant HTN (RHTN), a recent study demonstrated that RRI is an independent associate of RHTN, after controlling for clinical characteristics, hemodynamic load and target organ damage indices.40 Of note, it was shown that a RRI cutpoint of 0.648 has a sensitivity of 78% and a specificity of 72% for prediction of RHTN and that the odds of an increased RRI (>0.7) is 6.72 times higher in patients with RHTN compared with controlled hypertensive patients, regardless of BP levels. The fact that eGFR failed to stand as an independent predictor of RHTN suggests that changes in renal vascular resistance may precede overt renal functional impairment. Hence, the assessment of renal hemodynamics with RRI may be helpful in the clinical setting and, when available, in predicting hard‐to‐control BP in hypertensives well managed with antihypertensive drugs. In terms of pathophysiology, the strong correlation of RRI with BP levels can be explained mathematically considering the formula that expresses the index. An increase in systolic BP increases and promotes a greater peak renal velocity and correspondingly a decrease in diastolic arterial pressure results in a lower end‐diastolic velocity.40 The interrelationship between hemodynamic load and RRI is reflected by the fact that RRI is highly dependent on the aortic pulse pressure. Chronically elevated BP levels, as in RHTN, may lead to arterial remodelling, increased arterial stiffness and eventually higher renal vascular resistance and an increase in RRI.41 Moreover, sympathetic nervous system activity is implicated in RHTN, while a significant relationship between RRI and sympathetic tone has been documented, indicating a BP‐independent effect on RRI. Of note in 88 patients with RHTN, RRI decreased significantly after renal denervation and the decrease did not correlate with BP reduction.42

5. RRI AND ANTIHYPERTENSIVE TREATMENT

The evaluation of RRI may also have therapeutic implications. There are data suggesting that during chronic antihypertensive treatment, changes in RRI keep up with changes in urine albumin excretion.4, 43 Also, an increase in RRI may be a signal of intrarenal stiffness and suggests caution in titrating renin‐angiotensin system inhibitors to avoid renal function deterioration, especially in CKD patients, diabetics or the elderly. Moreover, previous reports have shown that some antihypertensive agents affect RRI.43 More specifically, in patients with essential HTN and especially in those with microalbuminuria, RAS inhibitors like valsartan and lisinopril are able to improve renal function by reducing renal vascular resistance and thus preventing future renal failure.44 Hence, the observation of an increased RRI may favor the prescription of nephroprotective drugs or avoid the prescription of nephrotoxic agents (ie, contrast media, nonsteroidal anti‐inflammatory drugs). Whether the effect of antihypertensive medications on RRI is due to effect on systemic or local renal hemodynamics remains to be elucidated.

6. CONCLUSION

In the setting of hypertensive disease, the assessment of renal hemodynamics by RRI seems to possess an important role in the evaluation of diverse cases of secondary HTN. Beyond this, in essential HTN, RRI reflects renal disease progression beyond the established measures of glomerulopathy, as well as it can estimate CV and renal prognosis. Large, prospective population‐based studies are needed to confirm these observations. RRI is also recently implicated in the challenging evaluation of resistant HTN. Thus, given that the Doppler assessment of RRI is simple, needs a limited training and is reproducible, RRI emerges as a simple and a “multifunctional” tool that could provide more insight on the global CV and renal risk of the hypertensive patient.

CONFLICT OF INTEREST

None.

Andrikou I, Tsioufis C, Konstantinidis D, et al. Renal resistive index in hypertensive patients. J Clin Hypertens. 2018;20:1739–1744. 10.1111/jch.13410