Diuretic Doppler ultrasonography in chronic unilateral partial ureteric obstruction in dogs



Objective  To compare the effects of diuresis induced by a loop diuretic (frusemide), an osmotic diuretic (mannitol) and a high-osmolar radio-contrast medium (sodium iothalamate) on the intrarenal resistive index (RI) in dogs with chronic unilateral partial ureteric obstruction (UPUO).

Materials and methods  The split renal clearance and intrarenal RI were estimated in 11 dogs with chronic UPUO. Doppler ultrasonography measurements of the interlobar arteries were obtained before and 10 min after the intravenous infusion of 1 mg/kg frusemide, 1 g/kg mannitol or 20 mL 60% sodium iothalamate. The same experimental protocol was repeated with another drug at 1-week intervals.

Results  There was a significant difference between the intrarenal RI of obstructed and unobstructed kidneys in the chronic phase of UPUO. While mannitol and sodium iothalamate significantly increased the RI in both kidneys, differences in RI between the kidneys decreased after infusing the two drugs. However, while frusemide insignificantly increased the RI in the obstructed kidney, it decreased the RI in unobstructed one. Consequently, the difference in RI between the kidneys increased significantly after administering frusemide. There were no significant differences in urinary volume after administering each of the drugs.

Conclusions  As there were no significant differences in the diuretic effects of the drugs, frusemide may have additional effects on the RI of unobstructed kidneys other than diuresis. Frusemide increased the difference between the intrarenal RI of the kidneys and therefore may improve the detection of unilateral urinary obstruction in humans.


Urinary tract obstruction is the most common urological cause of renal parenchymal damage and, if untreated, may lead to permanent renal loss. The early diagnosis and release of obstruction can preserve renal function. As the recovery of renal function is inversely related to the duration and extent of the obstruction, the precise diagnosis of urinary tract obstruction is important for the appropriate management of this disease. Excretory urography and renal ultrasonography (US) have routinely been used to evaluate the collecting system. However, the functional significance of upper tract dilatation cannot be assessed by these imaging studies alone. Similar degrees of urinary tract dilatation have frequently been noted in both obstructed and unobstructed systems in infants and young children. However, functionally significant ureteric obstruction can occur with no apparent upper tract dilatation. Physiological tests that assess the functional significance of a ureteric obstruction may better predict a patient's response to reconstruction than an anatomical test. However, tests such as diuretic renal scintigraphy and pressure-flow study (Whitaker test) cannot provide anatomical information. Until recently, no imaging technique could provide both an anatomical and a physiological evaluation of urinary tract obstruction.

Although conventional grey-scale US provides only an anatomical depiction of the changes in ureteric obstruction, certain physiological aspects of obstruction can be investigated by adding Doppler techniques. Substantial changes in the intrarenal arterial Doppler waveform are associated with various pathological conditions. The intrarenal resistive index (RI) calculated as ((peak systolic velocity − end diastolic velocity)/peak systolic velocity), is the most commonly used variable to quantify changes in intrarenal Doppler waveforms. A reduction in diastolic flow relative to systolic flow occurs in hydronephrotic kidneys, and this reduction is reflected in an increased intrarenal RI. In normal individuals the intrarenal RI is leqslant R: less-than-or-eq, slant0.70 [1] but there are many patients with mild but functionally significant hydronephrosis in whom the intrarenal RI is <0.70 [2]. To enhance the diagnostic accuracy of the RI in these patients, the diuretic intrarenal RI measurement (diuretic Doppler US) has recently been introduced [3–6]. Diuresis induced by frusemide and normal saline increases the intrarenal RI in obstructed kidneys, whereas it significantly decreases the RI in contralateral unobstructed kidneys [6]. This diverse response to diuresis increases the difference in intrarenal RI between the types of kidney and enhances the accuracy of diagnosis.

The intrarenal RI can be influenced by many factors other than urinary obstruction. In the clinical setting therefore an understanding of the factors that affect RI is extremely important for interpreting small differences in intrarenal RI. Shokeir et al.[7] showed that NSAIDs significantly affect the intrarenal RI in acutely obstructed kidneys, but to our knowledge, the effects of diuresis induced by drugs other than frusemide and normal saline on intrarenal RI have not yet been evaluated. In the present study we compared the effects of diuresis induced by a loop diuretic (frusemide), an osmotic diuretic (mannitol) and a hyperosmolar radio-contrast medium (sodium iothalamate) on the intrarenal RI in dogs with chronic unilateral partial ureteric obstruction (UPUO).

Materials and methods

This study was approved by the Experimental Animal Care and Use Committee of our institution. Eleven female mongrel dogs (10–17 kg) were used; all procedures were carried out under general anaesthesia using nitrous oxide, oxygen and halothane (Takeda Chemical Industries, Osaka, Japan) with endotracheal intubation and mechanical ventilation after intravenous thiopental sodium (25 mg/kg, Dainippon Pharmaceutical, Osaka, Japan).

In these animals the UPUO was produced using a technique developed previously [8,9]; briefly, a unilateral catheter-less cutaneous ureterostomy was created first and geqslant R: gt-or-equal, slanted2 weeks after surgery the left distal ureter was partially obstructed. This procedure causes a form of mild hydronephrosis that persists for many months. Throughout the study, cefaclor (100 mg/day, Shionogi Pharmaceutical, Osaka, Japan) was given orally to prevent UTI; after each procedure, urine analysis and urine culture excluded UTI. The degree of hydronephrosis was monitored by excretory urography every 4 weeks for 6 months after establishing UPUO.

At least 6 months after establishing UPUO, the animals were placed supine and given an intravenous infusion of 0.45% sodium chloride and 2.5% glucose in water, at 0.2 mL/kg/min during the experiment. The mean arterial BP (MAP) was monitored with catheters placed in a femoral artery. The urine volume from each kidney was measured throughout the experiment. Split renal clearance was assessed, and followed by Doppler US measurements before and 10 min after the intravenous infusion of each of the drugs. The drugs tested were frusemide (1 mg/kg, Aventis Pharma, Tokyo, Japan), 20% mannitol (5 mL/kg, Nikken Chemicals, Tokyo, Japan), and 66.8% sodium iothalamate (20 mL, Daiichi Pharmaceutical, Tokyo, Japan). Each experimental protocol was carried out at 1-week intervals.

To assess split renal clearance, a 10 F balloon catheter was inserted into the bladder to collect urine drained from the right kidney, and a polyethylene urine-sampling bag attached with adhesive around the ureteric stoma to obtain urine from the left kidney. For renal clearance studies, inulin (Wako Junyaku, Osaka, Japan) and para-aminohippuric acid (PAH; Daiichi) were infused at 15 mg/min and 3 mg/min, respectively. These infusions followed a priming dose of 750 mg inulin and 120 mg PAH. After a 1-h equilibration period, two consecutive 1-h clearance periods were obtained in each dog. Samples of arterial blood were collected at the midpoint of each clearance period for measurement, and inulin (Cin) and PAH (CPAH) clearance in each kidney was calculated.

All US examinations were undertaken using a real-time/colour-coded duplex Doppler US unit (Aloka SSD-2000, Tokyo, Japan) with a 5-MHz sector transducer using the lowest wall filter setting and a minimum sample gate size. The renal interlobar artery, at an angle of insonation of <30°, was identified on colour Doppler imaging. Sequential Doppler waveforms were obtained from at least three regions in each kidney. The RI was calculated from the Doppler waveforms using the integral software (Fig. 1).

Figure 1.

The spectral Doppler US waveform from an interlobar artery of the dog kidney.

All data are presented as the median (95% CI); an anova with Bonferroni correction was used to assess differences, with P<0.05 considered to indicate significance.


During the chronic phase of UPUO, excretory urography showed mild hydronephrosis with a normal contralateral collecting system in all of the dogs (Fig. 2). However, in five of the 11 animals, B-mode grey-scale US failed to detect dilatation of the collecting system in the obstructed kidneys before radio-contrast infusion. Serial excretory urography showed that the degree of hydronephrosis did not change in any of the dogs throughout the study.

Figure 2.

Excretory urography showing mild to moderate hydronephroureter with a normal contralateral upper tract 6 months after establishing UPUO in dogs.

During the chronic phase of UPUO, the median Cin and CPAH in the obstructive kidneys were 0.9 (0.8–1.0) and 2.6 (2.2–2.8) mL/min/kg, respectively. The median values in the contralateral unobstructed kidneys were 1.2 (1.1–1.3) and 3.3 (3.1–3.6) mL/min/kg, respectively. Comparing the median clearance in the two kidneys showed that the renal function in the mildly hydronephrotic kidneys had decreased significantly (Cin and CPAH both P<0.001) 6 months after imposing UPUO.

The respective urine volumes after administering frusemide, mannitol and iothalamate were 3.3 (2.5–4.0), 3.0 (2.6–3.6) and 3.0 (2.6–3.3) mL/min in the obstructed, and 2.8 (1.9–3.5), 3.0 (2.4–3.3) and 3.0 (2.2–3.0) mL/min in the contralateral unobstructed kidneys. The differences in the diuretic effects of these three drugs were not statistically significant.

Throughout the experiment, systemic BP was very stable in all the groups; the median MAP in the frusemide, mannitol and iothalamate groups was 123 (115–134), 120 (117–134) and 120 (114–126)  mmHg, respectively, with no significant differences between the groups.

The intrarenal RIs for both kidneys are shown in Table 1; in all groups the intrarenal RIs were significantly higher than in the contralateral kidneys during the chronic phase of UPUO. After drug administration, the median RI of both kidneys was significantly greater in the mannitol and iothalamate groups, but the differences in median RI between the kidneys became insignificant after infusing mannitol and iothalamate. Frusemide moderately increased the median RI of the interlobar arteries in obstructed kidneys but insignificantly decreased the median RI in the contralateral kidneys. Consequently, the differences in the median RI between the kidneys were significantly greater after frusemide infusion (Table 1).

Table 1.  The intrarenal RI of obstructed and contralateral unobstructed kidneys before and after administering frusemide, mannitol and sodium iothalamate
Median (95% CI) RIP
  • *

    P <0.05 vs contralateral.

obstructed0.46 (0.43–0.50)*0.49 (0.45–0.54)*0.211
contralateral0.40 (0.38–0.43)0.38 (0.37–0.44)0.902
ΔRI0.06 (0.05–0.07)0.09 (0.08–0.11)0.002
obstructed0.45 (0.41–0.49)*0.51 (0.49–0.61)0.005
contralateral0.41 (0.38–0.44)0.49 (0.43–0.57)0.003
ΔRI0.04 (0.03–0.05)0.05 (0.03–0.07)0.666
obstructed0.46 (0.42–0.49)*0.51 (0.46–0.56)0.039
contralateral0.40 (0.37–0.45)0.48 (0.44–0.52)0.009
ΔRI0.05 (0.03–0.06)0.03 (0.02–0.04)0.039


In the present study, the intrarenal RI in kidneys with mild hydronephrosis was significantly higher than in the contralateral unobstructed kidneys during the chronic phase of UPUO. Previous studies showed that the RI of hydronephrotic obstructed kidneys is significantly higher than that in kidneys with unobstructive pelvicalyceal dilatation [1]. A high intrarenal RI in obstructed kidneys is mainly a result of the decrease in diastolic renal blood flow velocity. Platt et al.[1] reported that the RI in normal adult humans is leqslant R: less-than-or-eq, slant0.7; using this value, Doppler US has clinically excellent sensitivity and specificity in determining the presence or absence of obstruction in patients with calyceal dilatation. However, there are cases of mild ureteric obstruction with an intrarenal RI of leqslant R: less-than-or-eq, slant0.70 [2] and the present UPUO dog model closely represents such chronic mild unilateral ureteric obstruction in humans.

To improve the sensitivity of intrarenal RI measurement for detecting mild but functionally significant ureteric obstruction, specialized techniques using diuretic stimulation have been developed [3–6]. Ordorica et al.[4] showed that frusemide significantly increased the intrarenal RI in obstructed kidneys in children. In animal studies, the effect of frusemide and normal saline loading on intrarenal RI differs in obstructed and unobstructed kidneys. While the RI in obstructed kidneys increased after diuresis, that in unobstructed kidneys decreased [6]. In the present experiment, frusemide increased the intrarenal RI in obstructed kidneys but had no significant effects on that of the unobstructed kidneys. The differences in RI between the kidneys significantly increased only after administering frusemide. We confirmed the possible improvement in the diagnostic accuracy by comparing the contralateral unobstructed kidneys after provocation with frusemide.

Previous Doppler US studies have shown a strong correlation between the intrarenal RI and ureteric pressure [10]. It had been assumed that the difference in the changes in intrapelvic pressure between obstructed and unobstructed kidneys after diuretic stimulation with frusemide results in the divergent physiological responses between them. However, the present results show that this assumption is incorrect. Diuresis induced by drugs caused an increase in intrarenal RI in obstructed kidneys, but frusemide did not cause an increase in intrarenal RI in unobstructed kidneys, whereas the intrarenal RI in these kidneys significantly increased after administering the osmolar diuretic mannitol and iothalamate. Intrapelvic pressure was not measured in the present study, but there were no significant differences in the diuretic effects of the drugs. Therefore, frusemide may have some additional effects on the intrarenal RI of unobstructed kidneys other than diuresis.

The intrarenal RI has been assumed to reflect changes in renal vascular resistance [11]. Both frusemide and mannitol reduce renal vascular resistance and increase renal blood flow [12,13], while hyperosmolar radio-contrast agents induce renal vasoconstriction and a decrease in renal blood flow [14]. However, mannitol increased the intrarenal RI in unobstructed kidneys. Therefore, the results of the present study cannot be explained by the effects of the drugs on renal vascular resistance. We previously showed that an acute reduction in renal blood flow results in a decrease in the intrarenal RI of autotransplanted dog kidneys [15]. Furthermore, Tublin et al.[16] showed in perfused rabbit kidneys that intrarenal RI values were not readily affected by acute changes in renal vascular resistance. The intrarenal RI is influenced by many factors and the present results do not support any conclusive explanation of the divergent physiological responses after administering frusemide.

In the present study, a high osmolar radio-contrast agent, sodium iothalamate, had diuretic effects similar to those of an osmotic diuretic, mannitol, and both drugs increased the intrarenal RI in both kidneys. In addition, Hetzel et al.[17] reported that intravenous injection with a low osmolar radio-contrast agent, iopamidol, increased the intrarenal RI in normal kidneys. Excretory urography using radio-contrast agents is still an essential diagnostic test for urinary obstruction in most institutions. The effects of these agents should not be interpreted as significant changes in Doppler measurements within a kidney or between individuals. Furthermore, Shokeir et al.[7] showed that NSAIDs significantly affected the intrarenal RI of acutely obstructed kidneys. Such drugs are routinely used as analgesics for ureteric colic. Many factors affect the intrarenal RI and may confuse the diagnosis. To interpret small differences in intrarenal RI it is necessary to understand the factors that affect the RI.

In conclusion, the intrarenal RI in chronically obstructed kidneys was significantly higher than those in normal kidneys. Diuresis induced by frusemide may improve the accuracy of detecting functionally significant ureteric obstruction. However, osmotic diuresis may cause misleading results. Thus, frusemide and Doppler US should be used instead of general diuretic Doppler US.

H. Yokoyama, Department of Urology, Fukuoka University School of Medicine, 7-45-1 Nanakuma, Jonan-ku, Fukuoka, 814–0180, Japan.
e-mail: yokoyama@minf.med.fukuoka-u.ac.jp