Nephrol Dial Transplant ( 1997) 12: 1376–1380
Nephrology
Dialysis
Transplantation
Original Article
The pulsatility index and the resistive index in renal arteries.
Associations with long-term progression in chronic renal failure
L. J. Petersen1,2, J. R. Petersen2, U. Talleruphuus2, S. D. Ladefoged1, J. Mehlsen2 and H. Æ. Jensen1
1Department of Nephrology, Hvidovre Hospital, Hvidovre; 2Department of Clinical Physiology and Nuclear Medicine,
Frederiksberg Hospital, Frederiksberg, Denmark
Abstract
Background. The pulsatility index (PI ) and the resistive
index (RI ) are used as pulsed-wave Doppler measurements of downstream renal artery resistance. PI and
RI have been found to correlate with renal vascular
resistance, filtration fraction and eective renal plasma
flow in chronic renal failure. The aim of the present
study was to evalute the potential relationship between
these indices and the rate of decline in renal function,
as reflected by changes in dierent parameters of renal
function in patients with chronic renal failure.
Methods. Twenty-one patients (8 females, 13 males,
mean age 58 years (36–75)) with chronic renal failure
were enrolled in the study. Doppler examinations were
performed in the segmental arteries by an Acuson
128XP. The PI and the RI was calculated from the
blood flow velocities. Parameters of renal function
were measured every 3D months, and all patients were
followed for 18–21 months. Progression of renal dysfunction was estimated by linear regression of parameters of renal function versus time.
Results. In a multiple regression analysis both PI and
RI correlated significantly to the rate of decline in
reciprocal serum creatinine (PI: r=−0.48, P=0.03; RI:
r=−0.52, P=0.02 ). Furthermore, when separating the
patients in two groups by the median RI value, there
was a significant dierence between the groups in rate
of decline in reciprocal serum creatinine (P=0.02). For
PI this distinction was also present (P=0.04 ).
Conclusion. PI and RI correlated to the severity of the
renal disease, as reflected by the rate of decline in
reciprocal serum creatinine during antihypertensive
treatment. The median RI or PI value could separate
the patients into groups one of slow and another of
fast progression.
Key words: duplex scanning; doppler measurements;
progression of renal failure; renal artery; renal function; renal haemodynamics
Introduction
The pulsatility index (PI ) and the resistive index (RI )
are dierent indices calculated from the blood flow
velocities in vessels during the cardiac cycle. PI and
RI are used as pulsed-wave Doppler measurements of
downstream resistance in arteries. Originally both PI
and RI were introduced to detect peripheral vascular
diseases [1,2], but are rarely used in these conditions.
However, if the indices are measured in renal arteries
they are reported as reliable measurements of downstream renal resistance [3–5 ]. Furthermore, PI has
been reported to increase while increasing renal vascular resistance by angiotensin II infusion [6,7]. An
increasing amount of information is available about
the usefulness of PI and RI in the investigation and
monitoring of renal artery stenosis [8–12 ] and kidney
graft rejection [13–15], but only little information is
available about their usefulness in chronic renal failure.
RI has been reported to correlate to the level of serum
creatinine in chronic renal failure [16 ], but also to be
normal in patients with renal diseases limited to the
glomeruli [17 ]. We have previously reported that both
PI and RI correlate significantly to the eective renal
plasma flow (ERPF ), the renal vascular resistance
(RVR) and the filtration fraction (FF ) in patients with
chronic renal failure [5 ]. In that study an abnormally
high PI or RI was associated with a low ERPF, a high
RVR and a high FF. This points to the possibility that
the indices may be used to monitor the severity of the
renal disease.
Therefore the purpose of this study was to examine
possible correlations between PI and RI measured at
baseline and dierent measurements of long-term progression in patients with chronic renal failure.
Subjects and methods
Patients
Correspondence and oprint requests to: Lars Juhl Petersen MD
PhD, Department of Nephrology, Hvidovre Hospital, Kettegård
Allé 30, DK-2650 Hvidovre, Denmark.
Twenty-one patients, (8 females and 13 males, mean age 58
years (range 36–75 )), with chronic renal failure were enrolled
in the study. Renal insuciency was caused by glomerulo-
© 1997 European Renal Association–European Dialysis and Transplant Association
PI and RI correlates to long-term progression
nephritis in two patients, diabetic nephropathy in five,
nephrosclerosis in nine, polycystic kidney disease in three,
and contracted kidneys in two patients. All patients were
hypertensive and received as antihypertensive medication
either isradipine or spirapril, or a combination of the two.
The patients were recruited from a longitudinal trial evaluating the influence of the above-mentioned antihypertensive
drugs on the progression in chronic renal failure (the IS
study). Before entering this Doppler study the antihypertensive treatment had been kept constant for at least 3 months.
The antihypertensive medication was kept constant during
the study, allowing only changes in diuretics during the
study. The patients received either 5 mg of isradipine, 6 mg
of spirapril, or a combination of 2.5 mg of isradipine and
3 mg of spirapril. Thirteen patients received diuretics, and
six patients also received labetalol to keep blood pressure
suciently low. In only four patients were the doses of
diuretics changed during the study.
The study was approved by the local ethics comittee, and
all participants gave informed consent.
Methods
No antihypertensive medication was given on the days of
the examinations. All patients underwent 99mTc-DTPA
(diethylenetriaminepentaacetate) renography and ultrasound
examination of the kidneys before entry to the study to
exclude subjects with renal artery stenosis or postrenal
obstructive disease [18]. The pulsed-wave Doppler examinations were performed by an Acuson 128XP, using a 3.5 MHz
transducer during a duplex scan. The Doppler frequency was
5 MHz. The patients were examined in the supine position
after at least 30 min of rest during the period of constant
infusion of 51Cr-EDTA. The kidneys were examined by
ultrasound from a lateral lumbar window. Measurements
were performed at the level of the segmental arteries in the
hilus area. At least three measurements were done in dierent
segmental arteries in each kidney. The PI and the RI were
computer estimated by the Acuson software. The PI and the
RI were calculated as:
PI=(PSV–MDV )/MV
1377
volume, P =PAH concentration in the plasma, and t=
PAH
time of the clearance period.
For both measurements, the first h of infusion was used
to reach a steady state followed by two 30 min clearance
periods. The patients were in a resting supine position during
infusion. Urine was collected by free voiding.
The renal blood flow (RBF ) and the filtration fraction
(FF ) were calculated as:
RBF=ERPF/(1–HCT)
and
FF=GFR/ERPF
where HCT=haematocrit.
Renal vascular resistance ( RVR) was calculated as:
RVR=79680×(MAP–12 mmHg)/RBF
where 79680 is a factor for converting mmHg to dyn/cm2,
and ml/min to cm3/s, MAP=mean arterial blood pressure.
A constant renal venous pressure of 12 mmHg was assumed.
Serum and urine level of creatinine and urea were measured
by an autoanalyser ( Technicon SMAC 3, Tarrytown, NY,
USA).
Clearance of creatinine (C ) and urea (C ) were calculated
Cr
Ur
as:
C =( U ×V )/(S ×t)
Cr
Cr
u
Cr
and
C =(U ×V )/(S ×t)
Ur
Ur
u
Ur
where U =concentration of creatinine in the urine, V =
Cr
u
urine volume, S =serum creatinine, t was 24 h, U =
Cr
Ur
concentration of urea in the urine, and S =serum urea.
ur
All parameters of renal function were measured every 3D
months. All patients were followed for 18–21 months.
Progression of renal disease was estimated by linear regression with a GFR estimate as the dependent variable and
time as the independent variable. Furthermore, progression
was also estimated by linear regression of reciprocal serum
creatinine versus time.
All parameters of renal function and haemodynamics were
corrected to a standard surface area of 1.73 m2.
and
RI=(PSV–MDV )/PSV
Statistics
where PSV=peak systolic velocity, MDV=minimum diastolic velocity, and MV=mean velocity (time averaged velocity). Doppler measurements were performed at baseline of
the study after at least 3 months of constant antihypertensive
treatment.
GFR was calculated from the clearance rate of 51Cr-EDTA
during constant infusion:
Covariation was expressed by a multiple regression analysis
by the least-squares method. Statistical comparisons between
groups were made with a two-sample t test. A two-sided 5%
level of significance was used. Results are given by the mean
values and ranges.
GFR=1.1×((U
Results
×V )/( P
×t))
EDTA
u
EDTA
where U
=51Cr-EDTA concentration in the urine, V =
EDTA
u
urine volume, P
=51Cr-EDTA concentration in the
EDTA
plasma, and t=time of the clearance period. The factor 1.1
compensates for the underestimation of GFR by the EDTA
clearance method [19]. GFR was always measured at the
same time in the morning, as this parameter has a high
degree of diurnal variation [20,21].
Eective renal plasma flow (ERPF ) was calculated from
the clearance rate of paraaminohippuricacid (PAH ) during
constant infusion:
×V )/(P ×t)
PAH
u
PAH
where U =PAH concentration in the urine, V =urine
PAH
u
ERPF=(U
Doppler measurements were succesful in all patients
except one in whom only RI was possible to calculate.
None of the patients suered from cardiac arrythmias
that could influence the Doppler flow velocity measurements. The mean heart rate was 74 beats per minute
(range 60–90).
The mean baseline PI was 1.57 (median 1.55; range
1.02–2.15) and mean baseline RI was 0.76 (median
0.75; range 0.64–0.87).
Progression of renal failure was estimated by the
rate of decline in GFR, creatinine clearance, urea
1378
clearance and reciprocal serum creatinine. Covariation
between the Doppler indices and the dierent measures
of progression in renal failure was expressed in a
multiple regression analysis. In this analysis only progression estimated by reciprocal serum creatinine was
significantly correlated to PI and RI ( PI: r=−0.48;
P=0.03 and RI: r=−0.52; P=0.02) (Figure 1 ). RI
was significantly correlated to the age of the patients
(r=0.49; P=0.02), but PI was not (r=0.42; P=0.07).
The age was not correlated to progression estimated
by reciprocal serum creatinine (r=−0.09; P=0.70) or
to progression estimated by other methods. In a new
multiple regression analysis PI could by described from
progression estimated by reciprocal serum creatinine
(coecient −2.990) and age (coecient 0.011) (r=
0.61; P=0.02). RI could also be describe in a multiple
regression analysis by progression estimated by recip-
L. J. Petersen et al.
rocal serum creatinine (coecient −0.669) and age
(coecient 0.003) (r=0.68; P=0.004).
Furthermore, the patients were separated in two
groups by either the median baseline PI or RI.
Diagnoses of the dierent subgroups are shown in
Table 1. Baseline characteristics of the patients in the
dierent subgroups are listed in Table 2. Patients with
a high RI were older than patients with a low RI, only
a trend was present regarding PI (Table 2 ). No dierence was present in other baseline characteristics
(Table 2 ). The patients with high baseline PI had a
significantly faster progression measured by the rate
of decline in reciprocal serum creatinine than the
patients with PI lower than the median value (−0.063
l/(mmol×1.73 m2×month) versus −0.020 l/(mmol
×1.73 m2×month), P=0.04 ). When separating the
patients into two groups by the median baseline RI,
patients with high baseline RI also had a significantly
faster progression measured the reciprocal serum creatinine than patients with low RI (−0.065 l/(mmol
×1.73 m2×month) versus −0.017 l/(mmol×1.73 m2
×month), P=0.02) in patients with baseline RI lower
than median value.
Discussion
Studying PI and RI in patients with chronic renal
failure we found that a PI higher than 1.55 or a RI
higher than 0.75 was associated with a faster decline
in renal function evaluated by the reciprocal serum
creatinine. Although patients with high indices tended
to be older than patients with low indices, the faster
progression could not be explained from dierence in
age as no significant covariation between age and
progression was found. Furthermore, we found a significant correlation of PI and RI to the progression of
renal failure evaluated by reciprocal serum creatinine.
The indices also depended on the age of the patients,
and in the multiple regression analysis both indices
were significantly correlated to progression and age.
To our knowledge similar results have not been demonstrated previously.
We have previously demonstrated correlations
between renal haemodynamics and the Doppler indices, suggesting that PI and RI are associated with the
severity of renal disease [5 ]. The indices are calculated
from the blood flow velocity during systole and diaTable 1. Diagnoses in the subgroups
PI<1.55 PI1.55 RI<0.75 RI 0.75
n=10
n=10
n=10
n=11
Fig. 1. Pulsatility index and resistive index plotted against the rate
of decline in reciprocal serum creatinine in patients with chronic
renal failure.
Glomerulonephritis
Diabetic nephropathy
Nephrosclerosis
Polycystic kidney disease
Contracted kidneys
2
2
5
0
1
0
3
3
3
1
PI, pulsatility index; RI, resistive index.
2
3
4
0
1
0
2
5
3
1
PI and RI correlates to long-term progression
1379
Table 2. Patient characteristics (mean values) at baseline in the groups of high and low pulsatility index ( PI) and resistive index ( RI )
S
(mmol/l )
Creatinine
Albuminuria (mmol/24 h)
Mean arterial blood pressure (mmHg)
Heart rate (beats/min)
GFR (ml/min*1.73 m2)
ERPF (ml/min*1.73 m2)
Filtration fraction (%)
RVR (dyn×s×cm–5)
Age (years)
PI<1.55
n=10
PI1.55
n=10
P
RI<0.75
n=10
RI0.75
n=11
P
205
20.9
111
76
32
297
14
29951
54
224
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
0.08
213
241
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
0.04
12.8
105
72
28
250
12
27108
63
21.1
19.6
112
107
76
29
284
14
32325
53
71
29
247
13
30501
63
GFR, glomerular filtration rate; ERPF, eective renal plasma flow; RVR, renal vascular resistance.
stole, and a high index is associated with a high
dierence in velocity between the systole and the
diastole. Dierences in flow velocities reflect downstream resistance, which could at least in part depend
on the degree of arterial stiness. But also functional
changes like angiotensin II infusion [6,7] or vasoconstriction due to hepatic failure [22 ] may raise the
indices without a morphological pathology.
Other studies evaluating RI in chronic renal failure
have demonstrated a correlation of this index to the
level of serum creatinine [16,17]. Furthermore, RI has
been found to be normal in renal disease limited to
the glomeruli [17], which is not in concordance with
the present study and a previous study from our group
suggesting a relation between the indices and renal
haemodynamics [5 ]. In patients with acute renal failure
PI has been shown to be higher in oliguric patients
than in non-oliguric patients [23]. This result also
indicates a dependency of the indices upon the severity
of the renal disease.
The Doppler derived indices have been used as a
marker of acute changes in renal plasma flow and
renal vascular resistance in patients with essential
hypertension and in healthy volunteers [6–8 ]. These
results underline that the indices also depend upon
functional changes. To our knowledge the indices have
not been used as markers of chronic changes in renal
haemodynamics.
The rate of decline in chronic renal failure is normally monitored by repeated determinations of GFR or
simply by the slope of the reciprocal value of serum
creatinine. No single-standing parameter has been
found useful as a marker of future progression in renal
disease. Such a marker would be of great value, as
therapy could be intensified before the renal function
is lost.
The use of antihypertensive drugs could be a confounder in the present study, as ACE inhibition has
been shown to reduce PI in normotensives [6 ], and to
increase RI in hypertensives [7]. The potential influence
of other antihypertensive drugs including diuretics has
not been reported. However, the antihypertensive treatment was kept constant during the investigation.
Furthermore, the dierent renal diagnoses could confound the results.
PI and RI are new parameters reflecting renal
haemodynamics. Although the exact information given
by these indices is still unknown, the indices in the
present study correlated to the severity of the renal
disease, measured as the rate of decline in reciprocal
serum creatinine during antihypertensive treatment. Of
great interest is the question of whether changes in
progression will be reflected by similar changes in the
Doppler indices. To answer this question further intervention studies are needed.
References
1. Gosling RG, Dunbar G, King DH et al. The quantitative
analysis of occlusive peripheral arterial disease by a non-invasive
ultrasonic technique. Angiology 1971; 22: 52–55
2. Pourcelot L. Applications clinique de l∞examen Doppler transcutane. In: Peronneau P (ed). Symposium: Velocimetric ultrasonnor
Doppler. Paris: Inserm 1974; 34: 213–240
3. White EM, Choyke PL. Duplex sonography in the abdomen.
In: Grant EG, White EM, ed. Duplex Sonography. SpringerVerlag, New York., 1987; 129–190
4. Norris CS, Barnes RW. Renal artery flow velocity analysis: a
sensitive measure of experimental and clinical renovascular
resistance. J Surg Res 1984; 36: 230–236
5. Petersen LJ, Petersen JR, Ladefoged SD, Mehlsen J, Jensen
HÆ. The pulsatility index and the resistive index in renal arteries
in patients with hypertension and chronic renal failure. Nephrol
Dial Transplant 1995; 10: 2060–2064
6. Bardelli M, Jensen G, Volkmann R, Caidahl K, Aurell M.
Experimental variations in renal vascular resistance in normal
man as detected by means of ultrasound. Eur J Clin Invest 1992;
22: 619–624
7. Jensen G, Bardelli M, Volkmann R, Caidahl K, Rose G, Aurell
M. Renovascular resistance in primary hypertension: experimental variations detected by means of Doppler ultrasound.
J Hypertens 1994; 12: 959–964
8. Bardelli M, Jensen G, Volkmann R, Aurell M. Non-invasive
ultrasound assessment of renal artery stenosis by means of the
Gosling pulsatility index. J Hypertens 1992; 10: 985–989
9. Veglio F, Provera E, Pinna G et al. Renal resistive index after
captopril test by echo-Doppler in essential hypertension. Am
J Hypertens 1992; 5: 431–436
10. Avasthi PS, Voyles WF, Greene ER. Noninvasive diagnosis of
renal artery stenosis by echo-Doppler velocimetry. Kidney Int
1984; 25: 824–829
11. Homann U, Edwards JM, Carter S et al. Role of duplex
scanning for the detection of atherosclerotic renal artery disease.
Kidney Int 1991; 39: 1232–1239
12. Handa N, Fukanaga R, Ogawa S, Matsumoto M, Kimura K,
Kamada T. A new accurate and non-invasive screening method
1380
13.
14.
15.
16.
17.
for renovascular hypertension: the renal artery Doppler technique. J Hypertens 1988; 6 [ Suppl 4]: S458–S460
Stevens PE, Gwyther SJ, Hanson ME, Woodrow DF, Phillips
ME, Boultbee JE. Interpretation of duplex Doppler ultrasound
in renal transplants in the early postoperative period. Nephrol
Dial Transplant 1993; 8: 255–258
Rasmussen K, Pedersen E. Monitoring of renal allografts by
Doppler ultrasound: precision and refence values. Scand J Clin
Invest 1990; 50: 51–55
Rasmussen K, Pedersen E. Doppler ultrasound in the diagnosis
of renal allograft rejection and in monitoring the eect of
treatment. Scand J Clin Invest 1990; 50: 56–61
Kim SH, Kim WH, Choi BI, Kim CW. Duplex Doppler
Ultrasound in patients with medical renal disease: resistive index
vs serum creatinine level. Clin Radiol 1992; 45: 85–87
Platt JF, Ellis JH, Rubin JM, DiPietro MA, Sedman AB.
Intrarenal arterial Doppler sonography in patients with nonobstructive renal disease: correlation of resistive index with biopsy
findings. Am J Radiol 1990; 154: 1223–1227
L. J. Petersen et al.
18. Platt JF. Duplex Doppler evaluation of native kidney dysfunction: obstructive and non-obstructive disease. Am J Radiol 1992;
158: 1035–1042
19. Brøchner-Mortensen J. Routine methods and their reliablility
for assessment of glomerular filtration rate in adults. [ Thesis].
Copenhagen, Denmark, Lœgeforeningens forlag, 1978
20. Koopman JP, Koomaen GCM, Krediet RT, de Moor EAM,
Hoek FJ, Arisz L. Circadian rhythm of glomerular filtration
rate in normal individuals. Clin Sci 1989; 77: 105–111
21. van Acker BAC, Kooman GCM, Koopman MG, Krediet RT,
Ariesz L. Discrepancy between circadian rhythms of inulin and
creatinine clearance. J Lab Clin Med 1992; 120: 400–410
22. Sacerdoti D, Bolognesi M, Merkel C, Angeli P, Gatta A. Renal
vasoconstriction in cirrhosis evaluated by duplex Doppler ultrasonography. Hepatology 1993; 17: 219–224
23. Stevens PE, Gwyther SJ, Hanson ME, Boultbee JE, Kox WJ,
Phillips ME. Non-invasive monitoring of renal blood flow
characteristics during acute renal failure in man. Intens Care
Med 1990; 16: 153–158
Received for publication: 20.3.96
Accepted in revised form: 3.3.97