What do Doppler indices of renal perfusion tell us for the evaluation of renal disease?

In this issue of the Journal, Ohta et al. 1 report on Doppler indices of renal perfusion in patients with chronic renal disease and control subjects without clinical evidence of renal impairment. Patients with renal disease were stratified according to the underlying disease diabetic nephropathy, chronic glomerulonephritis and nephrosclerosis. The authors describe two intriguing findings. First, the resistance index (RI) and the pulsatility index (PI) of patients with diabetic nephropathy were higher than in patients with other types of renal disease and in control subjects. This was equally true for both Doppler indices measured in the interlobar arteries and for Doppler indices measured in the main renal arteries. The difference remained statistically significant after adjustment for age, gender, systolic blood pressure, total cholesterol, serum glucose, body mass index, hematocrit and creatinine clearance. However, it should be noted that hypertension was present in the majority of patients and controls, and that most subjects received antihypertensive drugs. Second, pulse wave velocity (PWV), an established marker of large artery compliance, correlated with the resistance and the pulsatility indices of the main renal arteries and the interlobar arteries. In multivariate analysis, and after adjustment for the above-mentioned factors, the correlation between the RI of the main renal artery and PWV remained statistically significant, and the correlation between the RI of the interlobar arteries and PWV almost reached statistical significance. Ultrasound examination of the kidneys with color-coded duplex and PW-Doppler is now established in the clinical evaluation of patients with hypertension and/or chronic renal disease 2. The Doppler indices of perfusion generally used are: (i) RI = peak (systolic) flow velocity – minimum (diastolic) flow velocity/peak (systolic) flow velocity and (ii) PI = peak (systolic) flow velocity – minimum (diastolic) flow velocity/mean flow velocity. The clinical relevance of renal Doppler indices has been discussed equivocally. Some authors relate these Doppler indices to renal vascular resistance and to damage of the renal vascular bed (nephrosclerosis). However, others relate the Doppler indices to alterations of the arterial system. In favour of the first hypothesis (i.e. RI as marker of renal vascular resistance and damage), there is growing evidence that renal resistive indices are prognostic markers for patients with renal artery stenosis and chronic renal disease. Their diagnostic value for the management of kidney transplant patients is well established 3. Radermacher et al. 4 demonstrated that patients with atherosclerotic renal artery stenosis and high resistance indices of intrarenal arteries did not benefit from renal artery angioplasty. The authors also suggested that these patients with an unfavourable prognosis had a higher degree of nephrosclerosis (i.e. irreversible renal vascular damage). Furthermore, the same group was able to demonstrate that patients with chronic renal disease and increased resistance indices of intrarenal arteries displayed a faster deterioration of renal function. Again, the authors related the increased renal resistance indices to fibroproliferative scarring of the kidneys 5. Increased renal resistive indices as a marker of accelerated decline in renal function in patients with chronic kidney disease were also reported earlier by Petersen et al. 6. Finally, Radermacher et al. 7 were able to show that increased resistance indices of the graft were associated with decreased long-term graft survival in kidney transplant patients, again suggesting that increased resistance indices of the kidney graft are a marker of advanced vascular damage of the graft. Progressive renal disease is associated with interstitial fibrosis and loss of capillaries and glomeruli, and the consequent progressive reduction in the number and cross-sectional area of renal vessels. This eventually leads to increased renal vascular resistance. Petersen et al. 8 showed that both RI and PI correlate significantly with effective renal plasma flow, renal vascular resistance and filtration fraction in patients with chronic renal failure. In addition, in healthy volunteers, acute changes in renal hemodynamics correlate with changes in renal resistive indices, as assessed by ultrasound 9. Therefore, it is compelling to assume that the resistance index is a measure of renal vascular resistance. However, in favour of the second hypothesis (i.e. RI as marker of systemic arterial disease), there is evidence that central hemodynamic alterations can affect renal Doppler indices. An elegant study performed in isolated perfused kidneys demonstrated that acute and substantial changes in renal vascular resistance hardly affect renal resistive indices 10. Instead, renal resistive indices demonstrated a linear correlation with pulse pressure. This was one of the first studies to relate proximal pulse pressure to intrarenal Doppler indices of perfusion. Pulse pressure is one surrogate parameter of systemic arterial stiffness 11. Age-related increases in pulse pressure may, at least in part, explain why renal resistive indices increase with age. In accordance with these findings, Krumme et al. 12 found that the resistive indices of kidney allografts correlate much better with the age and pulse pressure of the recipients than with the donor parameters. All of these observations suggest that systemic arterial stiffness is a major determinant of renal resistive indices. The resistance index appears to be determined more by the interaction of pulsatile flow with vascular elasticity than by vascular resistance 13. The study by Ohta et al. 1 is one of the first to systematically assess the relationship between pulse wave velocity and renal resistive indices. The correlation between the increased resistive indices and pulse wave velocity in patients with diabetic nephropathy may indicate advanced systemic arterial disease in diabetic patients. Several authors have observed elevated renal resistive indices in patients with diabetes and diabetic nephropathy 14–16, whereas others have not 17. Interestingly, Taniwaki et al. 18 reported that renal resistance indices in patients with diabetic nephropathy were significantly correlated with age, duration of diabetes and systolic and diastolic blood pressure. These data indicate that renal resistance indices reflect systemic vascular damage in diabetic patients. The data provided by Ohta et al. 1 may thus be attributable to the more advanced arterial damage in patients with diabetic nephropathy. In further support of this, Ishimura et al. 14 reported a close correlation between intrarenal resistance indices and carotid and femoral artery intima–media thickness in patients with diabetic nephropathy. Moreover Ohta et al. 19 were able to show that renal resistive indices correlated with carotid artery structure and elasticity in patients with essential hypertension. Therefore, the prognostic value of increased renal resistive indices in patients with renal disease and hypertension may not only be explained by a damaged renal vascular bed, but also by systemic arterial disease with severe impairment of the mechanical properties of large artery walls. In patients with renal disease 20, as well as in patients with essential hypertension 21, increased large artery stiffness has been established as a major prognostic marker of increased cardiovascular morbidity and mortality. The same may hold true for renal resistive indices. However, a correlation between measures of large artery stiffness and Doppler indices of renal perfusion does not allow a causal relationship to be established between the two. Both renal resistance indices and measurements of large artery stiffness tend to increase with deteriorating kidney function 22. Impaired mechanical properties of large artery walls may contribute to the progression of renal disease, which in turn causes further deterioration of large artery elasticity 23. The observation by Ohta et al. 1 that RI and PI are increased in patients with diabetic nephropathy to a greater extent than in patients with other types of renal disease has to be considered with caution. Substantial differences in antihypertensive treatment, creatinine clearance and gender distribution between the patient groups were present, although these differences did not reach statistical significance, probably due to the limited sample size. Antihypertensive treatment affects pulse wave velocity as well as Doppler indices of renal perfusion. The response of renal resistance indices to antihypertensive drugs may differ between patient groups. For example, Taniwaki et al. 24 observed a decrease in RI after captopril in diabetic patients that was not observed in healthy control subjects. Any changes in renal resistance indices induced by antihypertensive therapy and their potential prognostic significance have to be studied in more detail. In patients with advanced chronic renal disease, a reduction in aortic pulse wave velocity induced by antihypertensive therapy was found to be associated with better survival 25. A similar finding may emerge for renal resistance indices. In conclusion, several studies suggest a prognostic relevance of renal resistance indices for the progression of kidney disease in various patient cohorts (chronic renal failure, kidney allograft recipients, patients with renal artery stenosis). The present study by Ohta et al. 1 provides a systematic assessment of the relationship between renal resistance indices and measures of systemic arterial stiffness in different patient cohorts with chronic renal disease. The data suggest that Doppler indices of renal perfusion may also emerge as prognostic indicators of cardiovascular morbidity and mortality in patients with chronic kidney disease, particularly in high-risk groups such as patients with diabetic nephropathy. Further studies are warranted to investigate this issue in more detail.