0022-534 7/91/1453-0644$03.00/0 THE JOURNAL OF UROLOGY Copyright© 1991 by AMERICAN UROLOGICAL ASSOCIATION, INC.
Vol. 145, 644-646, March 1991
Printed in U.S.A.
RENAL ARTERIAL DUPLEX DOPPLER ULTRASOUND IN DOGS WITH URINARY OBSTRUCTION GERALD D. DODD, III,* PAUL N. KAUFMAN
AND
R. BRUCE BRACKEN
From the Departments of Radiology and Surgery, University of Cincinnati School of Medicine, Cincinnati, Ohio
ABSTRACT
Recent clinical studies using duplex Doppler sonography identified an alteration in renal arterial blood flow in obstructed hydronephrotic kidneys that reportedly can be used to distinguish obstructive from nonobstructive collecting system dilatation. We attempted to verify these clinical findings and establish the temporal relationship of the alteration in the Doppler spectrum to the onset of urinary obstruction by evaluating surgically induced urinary obstruction in dogs. We performed laparotomies on 11 dogs, with the left ureter isolated and ligated in five dogs, and left intact in six dogs (control group). Duplex Doppler examination of the left renal arteries performed nine times during the first postoperative month identified a statistically significant difference (p less than .05) in the Doppler resistive index calculation between the two groups on days 1, 2, 4, and week 4. A resistive index discriminatory threshold of 0.7 (greater than 0.7, obstructed; less than 0.7, nonobstructed) produced a test sensitivity of 74% and specificity of 77%. We conclude from our study that renal arterial duplex Doppler sonography can detect a change in renal perfusion as a result of urinary obstruction and that this change can be detected as early as 24 hours after obstruction. However, high false-positive and false-negative rates may limit the ability of this modality to reliably distinguish obstructive from nonobstructive collecting system dilatation. KEY WORDS:
Doppler sonography, urinary obstruction
Conventional sonography is a sensitive, noninvasive method for detecting dilation of the renal collecting system. However, it often cannot distinguish obstructive from nonobstructive dilatation. 1 Therefore, patients commonly undergo more invasive procedures, such as intravenous urography, radionuclide renography, retrograde pyelography, or a Whitaker test, to make this distinction. Duplex Doppler sonography, a relatively new ultrasound adaptation, has recently been reported by Platt and associates to be significantly better than conventional sonography in differentiating obstructive from nonobstructive collecting system dilatation. 2 •3 Duplex Doppler ultrasound scanners are capable not only of producing morphologic images but also are able to interrogate specified volumes for the presence and speed of moving blood by detecting a Doppler frequency shift from the transducer's insonating frequency. 4 The relative systolic and diastolic velocities of flowing blood within individual vessels can be selectively evaluated, These velocities (the Doppler signal) are usually displayed on a time/velocity graph. This graphic information (the Doppler spectrum) can be analyzed by several methods, of which the resistive index (RI) is the most common (fig. 1). RI is defined as follows: RI
MATERIALS AND METHODS
We performed laparotomies on 11 dogs (female hounds, 30 to 40 lb.). In five, the left ureter was isolated and ligated twice with silk suture. In the remaining six (the control group), the left ureter was isolated without ligation. Duplex Doppler sonography was performed with an ATL UltraMark 4 ultrasound machine (ATL, Bothell, Washington) equipped with a 3.5 MHz transducer. The Doppler was performed at 100 percent output power, a wall filter of 100 Hz and a gate size of 1.5 mm. The left kidneys were examined once preoperatively and nine times postoperatively. The postoperative scans were obtained over four weeks: one each day for the first four days, one every other day during the second week, and one at the end of the third and fourth weeks. Before scanning, the dogs were sedated with five mg. of acepromazine, administered subcutaneously. During each scanning session, four to six Rls were calculated from Doppler spectrums obtained from segmental, interlobar, or arcuate renal arteries. The daily mean RI was calculated for
= Peak Systolic Velocity - Lowest Diastolic Velocity Peak Systolic Velocity
Using the RI, Platt and associates evaluated the renal arterial Doppler shift in patients with obstructive and nonobstructive collecting system dilatation. 2 •3 They identified an increased RI in obstructed hydronephrotic human kidneys, and reported that an RI greater than 0.7 was diagnostic of obstructive hydronephrosis (sensitivity, 92%; specificity, 88%). We attempted to verify these clinical findings and establish the temporal relationship of the alteration in the Doppler spectrum to the onset of ureter al obstruction in a dog model. Accepted for publication October 19, 1990.
FIG. 1. Duplex Doppler image of dog kidney with simultaneous display of gray-scale image (upper left corner) and Doppler spectrum. Peak systole = x; lowest diastole = +.
* Requests for reprints: Department of Radiology, Presbyterian-
University Hospital, DeSoto & O'Hara St., Pittsburgh, PA 15213.
644
645
DUPLEX DOPPLER EVALUATION OF URINARY OBSTRUCTION IN DOGS
each dog. The mean Rls from each day for obstructed and nonobstructed dog ureters were independently averaged, and standard deviations were calculated. These data were analyzed, using a two-way repeated measurement analysis of variance. Values were considered significantly different when the p value was less than .05. True-positive, false-positive, true-negative and false-negative rates were calculated. Serum blood urea nitrogen and creatinine levels were measured just before sacrifice. RESULTS
Table 1 shows the daily average RI for each dog, as well as the daily mean and standard deviation of the RI for the obstructed and nonobstructed dog groups. A two-way repeated measurement analysis of variance demonstrated a significant difference (p less than .05) in the RI between the two groups on postoperative days 1, 2, 4, and week 4, but not on postoperative days 3, 7, 9, 11, and week 3. Using a discriminatory RI level of 0. 7 (greater than 0. 7, obstructed; less than 0.7, nonobstructed), we obtained a true-positive rate of 74%, false-positive rate of 23%, true-negative rate of 77%, and a false-negative rate of 26%. The serum blood urea nitrogen and creatinine levels were normal in all dogs. DISCUSSION
Renal failure is a common clinical problem with a multitude of etiologies. One of the primary goals of imaging in renal failure is to identify readily treatable causes, such as obstructive hydronephrosis. There are many imaging methods used for the diagnosis of obstructive hydronephrosis, however, most are nonspecific, invasive, or contraindicated in the presence of renal failure. The development of a safe, noninvasive method that can reliably diagnose obstructive hydronephrosis would be of considerable clinical utility. Previous animal research has shown that urinary obstruction causes a decrease in renal perfusion that persists throughout the obstructive period. 5 •6 Based on this research, Platt and associates recently performed a prospective clinical study, which demonstrated that duplex Doppler sonography can consistently detect altered arterial flow in obstructed hydronephrotic human kidneys. By using RI to analyze renal arterial Doppler spectra, they found that a discriminatory RI level of 0.7 (greater than 0.7, obstructed; less than 0.7, nonobstructed) differentiated obstructed from nonobstructed collecting system dilatation with a sensitivity of 92% and specificity of 88% (sample size = 38 obstructed, 32 nonobstructed). Three falsenegative and four false-positive results were reported. The false-negative results were reduced from three to one when the Rls of the obstructed kidneys were compared to the Ris of the contralateral nonobstructed kidneys. In six cases (two with Ris TABLE
less than 0.7), the difference in the RI between the obstructed and contralateral nonobstructed kidneys was greater than 0.1. This difference was not identified between normal kidneys and was thus thought to be diagnostic of unilateral obstruction on the side with the elevated RI. Platt et al. also evaluated 159 nondilated kidneys, of which 50 were proved to have medical renal disease and 109 were normal. Twenty-seven (54%) of the kidneys with medical renal disease had elevated Rls greater than or equal to 0.7. All normal kidneys exhibited an RI less than 0.7. In our study, we were able to demonstrate a statistically significant elevation in the RI obtained from surgically obstructed dog kidneys. However, this significant difference, which was initially detected 24 hours after obstruction, occurred sporadically over the four-week test period. Using a discriminatory RI level of 0. 7 to differentiate obstructive from nonobstructive dilatation, we achieved an overall sensitivity of 74% and specificity of 77%. Our false-positive and false-negative rates were fairly high, at 23% and 26%, respectively. These incorrect diagnoses occurred sporadically throughout the testing period. Because of technical difficulty in scanning our dogs, we did not use Doppler to examine the contralateral nonobstructed kidneys, and thus we were unable to determine whether a difference in the Rls between the two kidneys could have reduced our high false-negative rates. Platt and associates reported that their false-positive levels occurred in patients with nonobstructive hydronephrosis complicated by underlying medical renal disease. It is unlikely that our false-positive results were caused by underlying medical renal disease, since all dogs had normal blood urea nitrogen and creatinine levels, and the falsely elevated Ris occurred sporadically throughout the test period in all dogs. Our test results, which were inferior to Platt's, may have been caused by a failure of the dog model. Technical factors that could have degraded our results include a wide variation in the preoperative RI between dogs, marked variability in the amplitude of the Doppler signal in the same dog, pronounced tachycardia and tachypnea, and severe hydronephrosis. None of these entities were reported as problematic by Platt's group. We conclude from our study that renal arterial duplex Doppler sonography can detect a change in renal blood flow as a result of urinary obstruction, and that the change is detectable as early as 24 hours after obstruction. However, our results suggest that duplex Doppler sonography may not be as good at differentiating obstructive from nonobstructive collecting system dilatation as reported by Platt et al. Our high false-negative rate indicates a significant limitation in the use of this modality for screening for obstructive hydronephrosis. Because of the difference in results between our work and Platt's, further investigation will be required to clearly define the role of duplex Doppler sonography in urinary obstruction.
1. Renal RI of dogs with obstructed and nonobstructed ureters
pre op
POD 1
POD 2
POD3
POD4
POD?
POD9
POD 11
POwk3
POwk4
1 2 3 4 5 mean± ISD
.65 .61 .54 .67 .64 .62 ± .05
.79 .75 .73 .79 .71 .75 ± .04
.80 .72 .62 .69 .73 .71 ± .06
.76 .62 .68 .79 .73 .72 ± .07
.83 .69 .74 .88 .70 .77 ± .08
.86 .70 .77 .79 .76 .77 ± .06
.87 .72 .60 .80 .82 .76 ± .1
.87 .68 .60 .81 .73 .74 ± .1
.77 .81 .64 .76 .65 .73 ± .08
.86 .85 .70 .66 .78 .77 ± .09
6
.62 .75 .64 .67 .61 .63 .65 ± .05
.66 .65 .66 .63 .58 .66 .64 ± .03
.66 .65 .65 .68 .61 .61 .64 ± .03
.67 .74 .64 .63 .58 .63 .65 ± .05
.62 .69 .65 .67 .65 .65 .66 ± .02
.72 .81 .63 .64 .72 .64 .69 ± .07
.62 .76 .66 .78 .62 .73 .69 ± .07
.62 .81 .72 .66 .66 .71 .69 ± .07
.67 .87 .74 .56 .65 .58 .69 ± .1
.64 .73 .58 .69 .64 .56 .64 ± .06
no
yes
yes
no
yes
no
no
no
no
yes
Obstructed Dog Ureters
Nonobstructed Dog Ureters
7 8 9 10 11
mean± ISD Obstructed vs. Nonobstructed pvalue< .05
646
DODD, KAUFMAN AND BRACKEN
REFERENCES 1. Ellenbogen, P. H., Scheible, W. and Talner, L. B.: Sensitivity of
gray-scale ultrasound in detecting urinary tract obstruction. A.J.R., 130: 731, 1978. 2. Platt, J. F., Rubin, J. M. and Ellis, J. H.: Distinction between obstructive and nonobstructive pyelocaliectasis with duplex Doppler sonography. A.J.R., 153: 997, 1989. 3. Platt, J. F., Rubin, J. M., Ellis, J. H. and Dipietro, M. A.: Duplex Doppler US of the kidney: differentiation of obstructive from nonobstructive dilatation. Radiology, 171: 515, 1989.
4. Nelson, T. R. and Pretorius, D. H.: The Doppler signal: where does it come from and what does it mean? A.J.R., 151: 439, 1988. 5. Ryan, P. C., Maher, K. P., Murphy, B., Jurley, G.D. and Fitzpatrick, J. M.: Experimental partial ureteric obstruction: pathophysiologic changes in upper tract pressures and renal blood flow. J. Urol., 138: 674, 1987. 6. Vaughan, E. D., Sorenson, E. J. and Gillenwater, J. Y.: The renal hemodynamic response to chronic unilateral complete ureteral occlusion. Invest. Urol., 8: 78, 1970.
Vol. 145, 644-646, March 1991
Printed in U.S.A.
RENAL ARTERIAL DUPLEX DOPPLER ULTRASOUND IN DOGS WITH URINARY OBSTRUCTION GERALD D. DODD, III,* PAUL N. KAUFMAN
AND
R. BRUCE BRACKEN
From the Departments of Radiology and Surgery, University of Cincinnati School of Medicine, Cincinnati, Ohio
ABSTRACT
Recent clinical studies using duplex Doppler sonography identified an alteration in renal arterial blood flow in obstructed hydronephrotic kidneys that reportedly can be used to distinguish obstructive from nonobstructive collecting system dilatation. We attempted to verify these clinical findings and establish the temporal relationship of the alteration in the Doppler spectrum to the onset of urinary obstruction by evaluating surgically induced urinary obstruction in dogs. We performed laparotomies on 11 dogs, with the left ureter isolated and ligated in five dogs, and left intact in six dogs (control group). Duplex Doppler examination of the left renal arteries performed nine times during the first postoperative month identified a statistically significant difference (p less than .05) in the Doppler resistive index calculation between the two groups on days 1, 2, 4, and week 4. A resistive index discriminatory threshold of 0.7 (greater than 0.7, obstructed; less than 0.7, nonobstructed) produced a test sensitivity of 74% and specificity of 77%. We conclude from our study that renal arterial duplex Doppler sonography can detect a change in renal perfusion as a result of urinary obstruction and that this change can be detected as early as 24 hours after obstruction. However, high false-positive and false-negative rates may limit the ability of this modality to reliably distinguish obstructive from nonobstructive collecting system dilatation. KEY WORDS:
Doppler sonography, urinary obstruction
Conventional sonography is a sensitive, noninvasive method for detecting dilation of the renal collecting system. However, it often cannot distinguish obstructive from nonobstructive dilatation. 1 Therefore, patients commonly undergo more invasive procedures, such as intravenous urography, radionuclide renography, retrograde pyelography, or a Whitaker test, to make this distinction. Duplex Doppler sonography, a relatively new ultrasound adaptation, has recently been reported by Platt and associates to be significantly better than conventional sonography in differentiating obstructive from nonobstructive collecting system dilatation. 2 •3 Duplex Doppler ultrasound scanners are capable not only of producing morphologic images but also are able to interrogate specified volumes for the presence and speed of moving blood by detecting a Doppler frequency shift from the transducer's insonating frequency. 4 The relative systolic and diastolic velocities of flowing blood within individual vessels can be selectively evaluated, These velocities (the Doppler signal) are usually displayed on a time/velocity graph. This graphic information (the Doppler spectrum) can be analyzed by several methods, of which the resistive index (RI) is the most common (fig. 1). RI is defined as follows: RI
MATERIALS AND METHODS
We performed laparotomies on 11 dogs (female hounds, 30 to 40 lb.). In five, the left ureter was isolated and ligated twice with silk suture. In the remaining six (the control group), the left ureter was isolated without ligation. Duplex Doppler sonography was performed with an ATL UltraMark 4 ultrasound machine (ATL, Bothell, Washington) equipped with a 3.5 MHz transducer. The Doppler was performed at 100 percent output power, a wall filter of 100 Hz and a gate size of 1.5 mm. The left kidneys were examined once preoperatively and nine times postoperatively. The postoperative scans were obtained over four weeks: one each day for the first four days, one every other day during the second week, and one at the end of the third and fourth weeks. Before scanning, the dogs were sedated with five mg. of acepromazine, administered subcutaneously. During each scanning session, four to six Rls were calculated from Doppler spectrums obtained from segmental, interlobar, or arcuate renal arteries. The daily mean RI was calculated for
= Peak Systolic Velocity - Lowest Diastolic Velocity Peak Systolic Velocity
Using the RI, Platt and associates evaluated the renal arterial Doppler shift in patients with obstructive and nonobstructive collecting system dilatation. 2 •3 They identified an increased RI in obstructed hydronephrotic human kidneys, and reported that an RI greater than 0.7 was diagnostic of obstructive hydronephrosis (sensitivity, 92%; specificity, 88%). We attempted to verify these clinical findings and establish the temporal relationship of the alteration in the Doppler spectrum to the onset of ureter al obstruction in a dog model. Accepted for publication October 19, 1990.
FIG. 1. Duplex Doppler image of dog kidney with simultaneous display of gray-scale image (upper left corner) and Doppler spectrum. Peak systole = x; lowest diastole = +.
* Requests for reprints: Department of Radiology, Presbyterian-
University Hospital, DeSoto & O'Hara St., Pittsburgh, PA 15213.
644
645
DUPLEX DOPPLER EVALUATION OF URINARY OBSTRUCTION IN DOGS
each dog. The mean Rls from each day for obstructed and nonobstructed dog ureters were independently averaged, and standard deviations were calculated. These data were analyzed, using a two-way repeated measurement analysis of variance. Values were considered significantly different when the p value was less than .05. True-positive, false-positive, true-negative and false-negative rates were calculated. Serum blood urea nitrogen and creatinine levels were measured just before sacrifice. RESULTS
Table 1 shows the daily average RI for each dog, as well as the daily mean and standard deviation of the RI for the obstructed and nonobstructed dog groups. A two-way repeated measurement analysis of variance demonstrated a significant difference (p less than .05) in the RI between the two groups on postoperative days 1, 2, 4, and week 4, but not on postoperative days 3, 7, 9, 11, and week 3. Using a discriminatory RI level of 0. 7 (greater than 0. 7, obstructed; less than 0.7, nonobstructed), we obtained a true-positive rate of 74%, false-positive rate of 23%, true-negative rate of 77%, and a false-negative rate of 26%. The serum blood urea nitrogen and creatinine levels were normal in all dogs. DISCUSSION
Renal failure is a common clinical problem with a multitude of etiologies. One of the primary goals of imaging in renal failure is to identify readily treatable causes, such as obstructive hydronephrosis. There are many imaging methods used for the diagnosis of obstructive hydronephrosis, however, most are nonspecific, invasive, or contraindicated in the presence of renal failure. The development of a safe, noninvasive method that can reliably diagnose obstructive hydronephrosis would be of considerable clinical utility. Previous animal research has shown that urinary obstruction causes a decrease in renal perfusion that persists throughout the obstructive period. 5 •6 Based on this research, Platt and associates recently performed a prospective clinical study, which demonstrated that duplex Doppler sonography can consistently detect altered arterial flow in obstructed hydronephrotic human kidneys. By using RI to analyze renal arterial Doppler spectra, they found that a discriminatory RI level of 0.7 (greater than 0.7, obstructed; less than 0.7, nonobstructed) differentiated obstructed from nonobstructed collecting system dilatation with a sensitivity of 92% and specificity of 88% (sample size = 38 obstructed, 32 nonobstructed). Three falsenegative and four false-positive results were reported. The false-negative results were reduced from three to one when the Rls of the obstructed kidneys were compared to the Ris of the contralateral nonobstructed kidneys. In six cases (two with Ris TABLE
less than 0.7), the difference in the RI between the obstructed and contralateral nonobstructed kidneys was greater than 0.1. This difference was not identified between normal kidneys and was thus thought to be diagnostic of unilateral obstruction on the side with the elevated RI. Platt et al. also evaluated 159 nondilated kidneys, of which 50 were proved to have medical renal disease and 109 were normal. Twenty-seven (54%) of the kidneys with medical renal disease had elevated Rls greater than or equal to 0.7. All normal kidneys exhibited an RI less than 0.7. In our study, we were able to demonstrate a statistically significant elevation in the RI obtained from surgically obstructed dog kidneys. However, this significant difference, which was initially detected 24 hours after obstruction, occurred sporadically over the four-week test period. Using a discriminatory RI level of 0. 7 to differentiate obstructive from nonobstructive dilatation, we achieved an overall sensitivity of 74% and specificity of 77%. Our false-positive and false-negative rates were fairly high, at 23% and 26%, respectively. These incorrect diagnoses occurred sporadically throughout the testing period. Because of technical difficulty in scanning our dogs, we did not use Doppler to examine the contralateral nonobstructed kidneys, and thus we were unable to determine whether a difference in the Rls between the two kidneys could have reduced our high false-negative rates. Platt and associates reported that their false-positive levels occurred in patients with nonobstructive hydronephrosis complicated by underlying medical renal disease. It is unlikely that our false-positive results were caused by underlying medical renal disease, since all dogs had normal blood urea nitrogen and creatinine levels, and the falsely elevated Ris occurred sporadically throughout the test period in all dogs. Our test results, which were inferior to Platt's, may have been caused by a failure of the dog model. Technical factors that could have degraded our results include a wide variation in the preoperative RI between dogs, marked variability in the amplitude of the Doppler signal in the same dog, pronounced tachycardia and tachypnea, and severe hydronephrosis. None of these entities were reported as problematic by Platt's group. We conclude from our study that renal arterial duplex Doppler sonography can detect a change in renal blood flow as a result of urinary obstruction, and that the change is detectable as early as 24 hours after obstruction. However, our results suggest that duplex Doppler sonography may not be as good at differentiating obstructive from nonobstructive collecting system dilatation as reported by Platt et al. Our high false-negative rate indicates a significant limitation in the use of this modality for screening for obstructive hydronephrosis. Because of the difference in results between our work and Platt's, further investigation will be required to clearly define the role of duplex Doppler sonography in urinary obstruction.
1. Renal RI of dogs with obstructed and nonobstructed ureters
pre op
POD 1
POD 2
POD3
POD4
POD?
POD9
POD 11
POwk3
POwk4
1 2 3 4 5 mean± ISD
.65 .61 .54 .67 .64 .62 ± .05
.79 .75 .73 .79 .71 .75 ± .04
.80 .72 .62 .69 .73 .71 ± .06
.76 .62 .68 .79 .73 .72 ± .07
.83 .69 .74 .88 .70 .77 ± .08
.86 .70 .77 .79 .76 .77 ± .06
.87 .72 .60 .80 .82 .76 ± .1
.87 .68 .60 .81 .73 .74 ± .1
.77 .81 .64 .76 .65 .73 ± .08
.86 .85 .70 .66 .78 .77 ± .09
6
.62 .75 .64 .67 .61 .63 .65 ± .05
.66 .65 .66 .63 .58 .66 .64 ± .03
.66 .65 .65 .68 .61 .61 .64 ± .03
.67 .74 .64 .63 .58 .63 .65 ± .05
.62 .69 .65 .67 .65 .65 .66 ± .02
.72 .81 .63 .64 .72 .64 .69 ± .07
.62 .76 .66 .78 .62 .73 .69 ± .07
.62 .81 .72 .66 .66 .71 .69 ± .07
.67 .87 .74 .56 .65 .58 .69 ± .1
.64 .73 .58 .69 .64 .56 .64 ± .06
no
yes
yes
no
yes
no
no
no
no
yes
Obstructed Dog Ureters
Nonobstructed Dog Ureters
7 8 9 10 11
mean± ISD Obstructed vs. Nonobstructed pvalue< .05
646
DODD, KAUFMAN AND BRACKEN
REFERENCES 1. Ellenbogen, P. H., Scheible, W. and Talner, L. B.: Sensitivity of
gray-scale ultrasound in detecting urinary tract obstruction. A.J.R., 130: 731, 1978. 2. Platt, J. F., Rubin, J. M. and Ellis, J. H.: Distinction between obstructive and nonobstructive pyelocaliectasis with duplex Doppler sonography. A.J.R., 153: 997, 1989. 3. Platt, J. F., Rubin, J. M., Ellis, J. H. and Dipietro, M. A.: Duplex Doppler US of the kidney: differentiation of obstructive from nonobstructive dilatation. Radiology, 171: 515, 1989.
4. Nelson, T. R. and Pretorius, D. H.: The Doppler signal: where does it come from and what does it mean? A.J.R., 151: 439, 1988. 5. Ryan, P. C., Maher, K. P., Murphy, B., Jurley, G.D. and Fitzpatrick, J. M.: Experimental partial ureteric obstruction: pathophysiologic changes in upper tract pressures and renal blood flow. J. Urol., 138: 674, 1987. 6. Vaughan, E. D., Sorenson, E. J. and Gillenwater, J. Y.: The renal hemodynamic response to chronic unilateral complete ureteral occlusion. Invest. Urol., 8: 78, 1970.
