- Email: [email protected]
Computerized Tomography Attenuation Value of Renal Calculus: Can It Predict Successful Fragmentation of the Calculus by Extracorporeal Shock Wave Lithotripsy? A Preliminary Study
0022-5347/02/1675-1968/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 167, 1968 –1971, May 2002 Printed in U.S.A.
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS: CAN IT PREDICT SUCCESSFUL FRAGMENTATION OF THE CALCULUS BY EXTRACORPOREAL SHOCK WAVE LITHOTRIPSY? A PRELIMINARY STUDY PETER JOSEPH, A. K. MANDAL, S. K. SINGH, PURABI MANDAL, S. N. SANKHWAR S. K. SHARMA
AND
From the Department of Urology, Postgraduate Institute of Medical Education and Research, and MRI and CT Scan Centre, Chandigarh, India
ABSTRACT
Purpose: We evaluated the attenuation value of renal calculi on unenhanced axial computerized tomography (CT) images as a predictor of calculous fragmentation by extracorporeal shockwave lithotripsy (ESWL) (Dornier Medical Systems, Inc., Marietta, Georgia). Materials and Methods: We included 30 patients with renal calculi up to 20 mm. in this prospective study. Calculous attenuation value was measured in Hounsfield units on unenhanced CT sections through the calculi. Patients were subsequently treated with ESWL. Results: Patients were grouped according to calculous attenuation value as groups 1—less than 500, 2—500 to 1,000 and 3— greater than 1,000 Hounsfield units. Of the 30 patients 24 (80%) underwent successful treatment. The rate of stone clearance was 100% (12 of 12 cases) in group 1, 85.7% (6 of 7) in group 2 and 54.5% (6 of 11) in group 3. The success rate for stones with an attenuation value of greater than 1,000 Hounsfield units was significantly lower than that for stones with a value of less than 1,000 Hounsfield units (6 of 11 versus 18 of 19 cases, chi-square 7.07, p ⬍0.01). Patients in group 3 required a greater median number of shock waves for stone fragmentation than those in groups 1 and 2 (7,300, 2,500, and 3,390, respectively). The mean attenuation value and number of shock waves required for calculous fragmentation correlated significantly (r ⫽ 0.779, p ⬍0.001). Conclusions: The CT attenuation value of renal calculi can help to differentiate stones that are likely to fragment easily on ESWL from those that would require a greater number of shock waves for fragmentation or may fail to fragment on ESWL. KEY WORDS: kidney calculi; tomography, x-ray computed; kidney; lithotripsy
Currently extracorporeal shockwave lithotripsy (ESWL) is the treatment of choice for most renal calculi not exceeding 20 mm.1 Technological advances have made it more patient and physician friendly. Its success rate is 60% to 99% in various series.2– 4 However, failure to fragment by ESWL results in medical costs, requirement of an alternative treatment procedure and the undesirable exposure of renal parenchyma to shock waves.5 Hence, it is desirable to identify patients who would be better served by an alternate treatment modality beforehand. We determined whether the computerized tomography (CT) attenuation value of renal calculi would predict fragmentation by ESWL. PATIENTS AND METHODS
A total of 17 men and 13 women participated in this prospective study. The ethics committee at our institute approved the protocol. Patients with a symptomatic solitary renal calculus larger than 5 and up to 20 mm. in the largest dimension in satisfactorily functioning and unobstructed renal units were included in the study. All patients underwent hematological, biochemical and radiological evaluation. Those with a calculus less than 5 and greater than 20 mm. elevated serum creatinine (greater than 2 mg./dl.) and bleeding diathesis were excluded from study. The maximal linear diameter of the calculus was measured by plain x-ray of the kidneys, ureters and bladder. Unenhanced (no oral or intravenous contrast) CT using 2 Accepted for publication December 21, 2001.
mm. contiguous sections through the renal calculus was performed using a soft tissue setting of a window width and level of 280 and 15 Hounsfield units, respectively, on a W450 scanner (Hitachi Medical Corporation, Tokyo, Japan) at 120 kV., 70 mA. and scan time 4.5 seconds. The image showing the calculus in largest dimension was selected and a pixel map of the largest possible region of interest within the calculus was generated. The pixel map consisted of a maximum of 100 attenuation values in a 10 ⫻ 10 matrix (fig. 1). Each value on the pixel map represented the attenuation value for mean of 4 pixels. The lowest, highest and most common attenuation values were recorded and the mean calculous attenuation value was calculated. All treatment was done by an electromagnetic Lithostar Multiline lithotriptor (Siemens, Erlangen, Germany). Calculi were fragmented under fluoroscopic guidance. A change in calculous size and/or outline, or separation of stone fragments indicated fragmentation. During 1 ESWL session a maximum of 3,500 shocks were delivered at the energy level of 2 to 4, corresponding to 14 to 15 kV. Treatment was terminated if satisfactory fragmentation was noted earlier. Another ESWL session was done after 3 weeks if followup plain x-ray of the kidneys, ureters and bladder showed significant residual fragments. When a calculus did not fragment even after 3 sessions, the patient was offered alternative therapy. The total number of shock waves delivered to any renal unit was limited to a maximum of 10,500 during the course of treatment.
1968
1969
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS
FIG. 1. Unenhanced axial CT section through calculus with calculous pixel map of 100 attenuation values in 10 ⫻ 10 matrix.
TABLE 1. Demographic data No. pts. No. men/No. women Mean age ⫾ SD No. calculi with largest linear diameter (mm.): Less than 10 11–20 Generator energy applied (kV.) No. calculous sites (%): Renal pelvis Upper calix Mid calix Lower calix
30 17:13 39.2 ⫾ 13.8 13 17 14–15 21 (70) 5 (16.6) 2 (6.6) 2 (6.6)
Plain x-ray of the kidneys, ureter and bladder was performed 6 weeks after the completion of treatment for assessing the outcome. Cases of residual fragments larger than 3 mm. or of a stone that did not fragment satisfactorily were considered failures. Statistical analysis was performed using the unpaired t and chi-square tests, contingency table analysis, Pearson’s coefficient of correlation analysis, the MannWhitney U test and Kruskal-Wallis analysis of variance. RESULTS
Table 1 lists patient characteristics and treatment data. The mean CT attenuation value of the calculi was between 136 and 1,585 Hounsfield units. Patients were stratified into groups based on calculous CT attenuation values. There were 12 patients in group 1 (less than 500 Hounsfield units), 7 in group 2 (500 to 1,000 Hounsfield units) and 11 in group 3 (greater than 1,000 Hounsfield units). Table 2 shows the treatment outcome in these patient groups. Of the patients 24 (80%) had adequate fragmentation and complete stone clearance, while 5 with partial fragmentation had significant residual fragments at the end of the study. These calculi were located in the renal pelvis in 3 cases and in the upper calix in 2. The only patient in whom the stone did not fragment even
after 3 ESWL sessions had a 20 mm. calculus in the renal pelvis with a CT attenuation value of 1,168 Hounsfield units. Patients in whom ESWL failed had calculi with a significantly higher mean CT attenuation value (p ⬍0.001) that required a greater number of shock waves (p ⬍0.001) than those with a successful outcome. The 2 groups were otherwise comparable (table 3). When tested for normality using a Q-Q plot and Levine test, data on the number of shock waves was not normal. Hence, the Mann-Whitney U test was done to test the significance of differences. The success rate was 92.3% (12 of 13 cases) for stones up to 10 mm. and 70% (12 of 17) for stones greater than 10 mm. However, the difference was not statistically significant (chi-square 2.16, p ⬎0.1). Maximum stone diameter in the 6 patients in the failure group was 8, 12, 12, 18, 20 and 20 mm., respectively. In the successful group 3 patients also had stones with a maximum diameter of 20 mm. each. The mean calculous attenuation value significantly correlated with the number of shock waves required for fragmentation (r ⫽ 0.779, p ⬍0.001, fig. 2). The chemical composition of urinary stones was studied by qualitative chemical analysis in 20 cases. Infrared spectroscopy of some of these stones was performed elsewhere. Table 4 lists the details of these cases, including chemical composition, CT attenuation value of stones and treatment outcome. The results of qualitative stone chemical analysis in 4 cases of failed ESWL showed calcium oxalate only in 3, and mixed calcium oxalate and phosphate in 1. Unfortunately infrared spectroscopy was not done on these stones and, therefore, we cannot comment on their predominant monohydrate or dihydrate composition. DISCUSSION
While ESWL continues to be the treatment of choice for calculi less than 20 mm., the effectiveness of this type of therapy depends on various factors, such as stone composition, stone location and pelvicaliceal anatomy.6 Of these factors the one that eludes physicians is the stone composition, which cannot be predicted accurately before retrieved stones are analyzed. The type of crystals and/or particles excreted in urine after ESWL may be an indicator of stone composition.7, 8 Urinalysis with scanning electron microscopy and x-ray energy dispersive spectroscopy for determining stone composition before ESWL9 has its own limitations. Plain radiography has been evaluated for identifying calculi susceptible to fragmentation by ESWL. Chaussy and Fuchs believed that if a stone has lesser radiodensity than the spine, it is likely to break readily, while stones more radiodense than spine would be more difficult to fragment.10 During ESWL of ureteral calculi Mattelaer et al observed that high radio-opaque stones needed an average of 1.7 times more shock waves than low radio-opaque stones of the same size.11 The disadvantage of plain x-ray for predicting stone characteristics is obvious. Calculi must have an appreciable diameter without any bowel gas interference and away from bony structures. Moreover, density measurements are not uniform but subjective and, hence, clinical applicability is limited. CT is a noninvasive technique that is easily available and provides greater density discrimination than conventional radiography.
TABLE 2. Outcome after ESWL in various patient groups Group 1 2 3
No. Pts. 12 7 11
p Value
Median No. Shock Waves Required for Fragmentation (interquartile range)
No. Fragmented, Cleared (%)
No. Significant Residual Fragments
No. No Fragmentation
2,500 (2,000–3,000) 3,390 (2,500–4,501) 7,300 (5,500–8,500) 0.001 (Kruskall-Wallis ANOVA)
12 (100) 6 (85.7) 6 (54.5) 0.022 (contingency table analysis)
0 1 4
0 0 1
1970
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS TABLE 3. Comparison of variables in patients with and without a successful outcome of ESWL Variables
Success
Failure
No. pts. Mean age ⫾ SD Mean body mass index ⫾ SD Mean calculous diameter ⫾ SD (mm.) Mean calculous CT attenuation value ⫾ SD (Hounsfield units) Median no. shock waves required for fragmentation (interquartile range)
24 39.9 ⫾ 14.2 23.2 ⫾ 2.9 12.0 ⫾ 4.5 631.6 ⫾ 328.3
6 36.5 ⫾ 12.9 24.2 ⫾ 3.9 15.0 ⫾ 5.0 1,224.8 ⫾ 218.1
p Value
3,000 (2,500–4,505)
8,750 (8,300–10,100)
0.585 (unpaired 0.586 (unpaired 0.217 (unpaired ⬍0.001 (unpaired
Student’s Student’s Student’s Student’s
t t t t
test) test) test) test)
⬍0.001 (Mann-Whitney U test)
FIG. 2. Scatterplot demonstrates correlation of mean CT attenuation values of renal calculi with number of shock waves required for stone fragmentation.
CT can distinguish a density difference of 0.5%, while plain x-ray requires a density difference of about 5%.12 CT has been done to analyze the chemical composition of renal calculi in vitro.13, 14 Using computerized mapping techniques Hillman et al differentiated uric acid from struvite and calcium oxalate calculi, and suggesting its clinical usefulness.13 On a MEDLINE search of the literature we did not identify any clinical study correlating CT measured stone density with calculous fragmentation by ESWL. In our clinical study the range of mean CT attenuation values of urate stones was 136 to 402 Hounsfield units and that of calcium oxalate dihydrate stones was 371 to 588 Hounsfield units. These calculi are known to fragment easily on ESWL15 and we observed 100% success in such cases. Overall the CT attenuation value of renal calculi had an inverse relationship with the ESWL success rate. The higher the attenuation, the less chance of success. The mean calculous attenuation value significantly correlated positively with the number of shock waves required for fragmentation. Of the 6 patients in whom
ESWL was not successful 5 had stones with a mean attenuation value exceeding 1,000 Hounsfield units (1,078 to 1,585), while the remaining 1 had a stone with a mean value approaching 1,000 Hounsfield units (956). Our data suggest that when a stone has a mean attenuation value of greater than 950 Hounsfield units and 7,500 shocks have not successfully achieved fragmentation, percutaneous nephrolithotomy should be performed and a third session of ESWL should not be considered. Stone location has a bearing on stone clearance after ESWL. Lower caliceal stones are associated with a relatively lower ESWL success rate.16, 17 In our study 2 patients with a 9 and 10 mm. stone, respectively, in a lower pole calix had complete clearance. CONCLUSIONS
In this study the CT attenuation value of a renal calculus correlated well with the outcome after ESWL and number of
TABLE 4. Chemical composition, CT attenuation value and treatment outcome of renal calculi Chemical Composition
No. Pts.
Mean Attenuation Value (range) (Hounsfield units)
Av. No. Shock Waves Required for Fragmentation
No. Success
No. Failure
Total No.
Urate, calcium oxalate traces Calcium oxalate dihydrate Calcium oxalate, unknown predominant composition Calcium oxalate, phosphate
3 4 10
136–402 (276) 371–588 (476) 467–1,330 (984)
1,860 3,010 5,912
3 4 7
0 0 3
3 4 10
3
412–1,585 (955)
5,500
2
1
3
16
4
20
Totals
20
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS
shock waves required during therapy. In a subset of patients with renal calculi otherwise suitable for ESWL a greater number of shock waves was required for fragmentation or ESWL was not successful. In these cases the mean calculous CT attenuation value exceeded 1,000 Hounsfield units. The possible outcome of ESWL may be explained to these patients before treatment. However, a larger trial is necessary. REFERENCES
1. Wolf, J. S. and Clayman, R. V.: Percutaneous nephrostolithotomy. Urol Clin N Am, 24: 43, 1997 2. Lingeman, J. E., Newman, D., Mertz, J. H. O. et al: Extracorporeal shock wave lithotripsy: the Methodist Indiana experience. J Urol, 135: 1134, 1996 3. Ehreth, J. T., Drach, G. W., Arnett, M. L. et al: Extracorporeal shock wave lithotripsy: multicenter study of kidney and upper ureter versus middle and lower ureter treatments. J Urol, 152: 1379, 1994 4. Cass, A. S.: Comparison of first generation (Dornier HM3) and second generation (Medstone STS) lithotriptors: treatment results with 13,864 renal and ureteral calculi. J Urol, 153: 588, 1995 5. Lingeman, J. E., Woods, J. R. and Toth, P. D.: Blood pressure changes following extracorporeal shock wave lithotripsy and other forms of treatment for nephrolithiasis. JAMA, 263: 1789, 1990 6. Bon, D., Dore, B., Irani, J. et al: Radiographic prognostic criteria for extracorporeal shock wave lithotripsy. Urology, 48: 556, 1996 7. Khan, S. R., Hackett, R. L. and Finlayson, B.: Morphology of
8. 9.
10. 11. 12. 13. 14. 15. 16. 17.
1971
urinary stone particles resulting from ESWL treatment. J Urol, 136: 1367, 1986 Bowsher, W. G., Crocker, P., Ramsay, J. W. A. et al: Single urine sample diagnosis. A new concept in stone analysis. Br J Urol, 65: 236, 1990 Cohen, N. P., Parkhouse, H., Scott, M. L. et al: Prediction of response to lithotripsy: the use of scanning electron microscopy and X-ray energy dispersive spectroscopy. Br J Urol, 70: 469, 1992 Chaussy, C. G. and Fuchs, G. J.: Extracorporeal shockwave lithotripsy. Monogr Urol, 4: 80, 1987a Mattelaer, P., Schroder, T., Fischer, N. et al: In situ extracorporeal shock wave lithotripsy of distal ureteral stones: parameters for therapeutic success. Urol Int, 53: 87, 1994 Federle, M. P., McAninch, J. W., Kaiser, J. A. et al: Computed tomography of urinary calculi. AJR Am J Roentgenol, 136: 255, 1981 Hillman, B. J., Drach, G. W., Tracey, P. et al: Computed tomographic analysis of renal calculi. AJR Am J Roentgenol, 142: 549, 1984 Mostafavi, M. R., Ernst, R. D. and Saltzman, B.: Accurate determination of chemical composition of urinary calculi by spiral computerized tomography. J Urol, 159: 673, 1998 Dretler, S. P.: Stone fragility: a new therapeutic distinction. J Urol, 139: 1124, 1988 Sabnis, R. B., Naik, K., Patel, S. H. et al: Extracorporeal shock wave lithotripsy for lower calyceal stones: can clearance be predicted? Br J Urol, 80: 853, 1997 Madbouly, K., Sheir, K. Z. and Elsobsky, E.: Impact of lower pole renal anatomy on stone clearance after shock wave lithotripsy: fact or fiction? J Urol, 165: 1415, 2001
Vol. 167, 1968 –1971, May 2002 Printed in U.S.A.
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS: CAN IT PREDICT SUCCESSFUL FRAGMENTATION OF THE CALCULUS BY EXTRACORPOREAL SHOCK WAVE LITHOTRIPSY? A PRELIMINARY STUDY PETER JOSEPH, A. K. MANDAL, S. K. SINGH, PURABI MANDAL, S. N. SANKHWAR S. K. SHARMA
AND
From the Department of Urology, Postgraduate Institute of Medical Education and Research, and MRI and CT Scan Centre, Chandigarh, India
ABSTRACT
Purpose: We evaluated the attenuation value of renal calculi on unenhanced axial computerized tomography (CT) images as a predictor of calculous fragmentation by extracorporeal shockwave lithotripsy (ESWL) (Dornier Medical Systems, Inc., Marietta, Georgia). Materials and Methods: We included 30 patients with renal calculi up to 20 mm. in this prospective study. Calculous attenuation value was measured in Hounsfield units on unenhanced CT sections through the calculi. Patients were subsequently treated with ESWL. Results: Patients were grouped according to calculous attenuation value as groups 1—less than 500, 2—500 to 1,000 and 3— greater than 1,000 Hounsfield units. Of the 30 patients 24 (80%) underwent successful treatment. The rate of stone clearance was 100% (12 of 12 cases) in group 1, 85.7% (6 of 7) in group 2 and 54.5% (6 of 11) in group 3. The success rate for stones with an attenuation value of greater than 1,000 Hounsfield units was significantly lower than that for stones with a value of less than 1,000 Hounsfield units (6 of 11 versus 18 of 19 cases, chi-square 7.07, p ⬍0.01). Patients in group 3 required a greater median number of shock waves for stone fragmentation than those in groups 1 and 2 (7,300, 2,500, and 3,390, respectively). The mean attenuation value and number of shock waves required for calculous fragmentation correlated significantly (r ⫽ 0.779, p ⬍0.001). Conclusions: The CT attenuation value of renal calculi can help to differentiate stones that are likely to fragment easily on ESWL from those that would require a greater number of shock waves for fragmentation or may fail to fragment on ESWL. KEY WORDS: kidney calculi; tomography, x-ray computed; kidney; lithotripsy
Currently extracorporeal shockwave lithotripsy (ESWL) is the treatment of choice for most renal calculi not exceeding 20 mm.1 Technological advances have made it more patient and physician friendly. Its success rate is 60% to 99% in various series.2– 4 However, failure to fragment by ESWL results in medical costs, requirement of an alternative treatment procedure and the undesirable exposure of renal parenchyma to shock waves.5 Hence, it is desirable to identify patients who would be better served by an alternate treatment modality beforehand. We determined whether the computerized tomography (CT) attenuation value of renal calculi would predict fragmentation by ESWL. PATIENTS AND METHODS
A total of 17 men and 13 women participated in this prospective study. The ethics committee at our institute approved the protocol. Patients with a symptomatic solitary renal calculus larger than 5 and up to 20 mm. in the largest dimension in satisfactorily functioning and unobstructed renal units were included in the study. All patients underwent hematological, biochemical and radiological evaluation. Those with a calculus less than 5 and greater than 20 mm. elevated serum creatinine (greater than 2 mg./dl.) and bleeding diathesis were excluded from study. The maximal linear diameter of the calculus was measured by plain x-ray of the kidneys, ureters and bladder. Unenhanced (no oral or intravenous contrast) CT using 2 Accepted for publication December 21, 2001.
mm. contiguous sections through the renal calculus was performed using a soft tissue setting of a window width and level of 280 and 15 Hounsfield units, respectively, on a W450 scanner (Hitachi Medical Corporation, Tokyo, Japan) at 120 kV., 70 mA. and scan time 4.5 seconds. The image showing the calculus in largest dimension was selected and a pixel map of the largest possible region of interest within the calculus was generated. The pixel map consisted of a maximum of 100 attenuation values in a 10 ⫻ 10 matrix (fig. 1). Each value on the pixel map represented the attenuation value for mean of 4 pixels. The lowest, highest and most common attenuation values were recorded and the mean calculous attenuation value was calculated. All treatment was done by an electromagnetic Lithostar Multiline lithotriptor (Siemens, Erlangen, Germany). Calculi were fragmented under fluoroscopic guidance. A change in calculous size and/or outline, or separation of stone fragments indicated fragmentation. During 1 ESWL session a maximum of 3,500 shocks were delivered at the energy level of 2 to 4, corresponding to 14 to 15 kV. Treatment was terminated if satisfactory fragmentation was noted earlier. Another ESWL session was done after 3 weeks if followup plain x-ray of the kidneys, ureters and bladder showed significant residual fragments. When a calculus did not fragment even after 3 sessions, the patient was offered alternative therapy. The total number of shock waves delivered to any renal unit was limited to a maximum of 10,500 during the course of treatment.
1968
1969
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS
FIG. 1. Unenhanced axial CT section through calculus with calculous pixel map of 100 attenuation values in 10 ⫻ 10 matrix.
TABLE 1. Demographic data No. pts. No. men/No. women Mean age ⫾ SD No. calculi with largest linear diameter (mm.): Less than 10 11–20 Generator energy applied (kV.) No. calculous sites (%): Renal pelvis Upper calix Mid calix Lower calix
30 17:13 39.2 ⫾ 13.8 13 17 14–15 21 (70) 5 (16.6) 2 (6.6) 2 (6.6)
Plain x-ray of the kidneys, ureter and bladder was performed 6 weeks after the completion of treatment for assessing the outcome. Cases of residual fragments larger than 3 mm. or of a stone that did not fragment satisfactorily were considered failures. Statistical analysis was performed using the unpaired t and chi-square tests, contingency table analysis, Pearson’s coefficient of correlation analysis, the MannWhitney U test and Kruskal-Wallis analysis of variance. RESULTS
Table 1 lists patient characteristics and treatment data. The mean CT attenuation value of the calculi was between 136 and 1,585 Hounsfield units. Patients were stratified into groups based on calculous CT attenuation values. There were 12 patients in group 1 (less than 500 Hounsfield units), 7 in group 2 (500 to 1,000 Hounsfield units) and 11 in group 3 (greater than 1,000 Hounsfield units). Table 2 shows the treatment outcome in these patient groups. Of the patients 24 (80%) had adequate fragmentation and complete stone clearance, while 5 with partial fragmentation had significant residual fragments at the end of the study. These calculi were located in the renal pelvis in 3 cases and in the upper calix in 2. The only patient in whom the stone did not fragment even
after 3 ESWL sessions had a 20 mm. calculus in the renal pelvis with a CT attenuation value of 1,168 Hounsfield units. Patients in whom ESWL failed had calculi with a significantly higher mean CT attenuation value (p ⬍0.001) that required a greater number of shock waves (p ⬍0.001) than those with a successful outcome. The 2 groups were otherwise comparable (table 3). When tested for normality using a Q-Q plot and Levine test, data on the number of shock waves was not normal. Hence, the Mann-Whitney U test was done to test the significance of differences. The success rate was 92.3% (12 of 13 cases) for stones up to 10 mm. and 70% (12 of 17) for stones greater than 10 mm. However, the difference was not statistically significant (chi-square 2.16, p ⬎0.1). Maximum stone diameter in the 6 patients in the failure group was 8, 12, 12, 18, 20 and 20 mm., respectively. In the successful group 3 patients also had stones with a maximum diameter of 20 mm. each. The mean calculous attenuation value significantly correlated with the number of shock waves required for fragmentation (r ⫽ 0.779, p ⬍0.001, fig. 2). The chemical composition of urinary stones was studied by qualitative chemical analysis in 20 cases. Infrared spectroscopy of some of these stones was performed elsewhere. Table 4 lists the details of these cases, including chemical composition, CT attenuation value of stones and treatment outcome. The results of qualitative stone chemical analysis in 4 cases of failed ESWL showed calcium oxalate only in 3, and mixed calcium oxalate and phosphate in 1. Unfortunately infrared spectroscopy was not done on these stones and, therefore, we cannot comment on their predominant monohydrate or dihydrate composition. DISCUSSION
While ESWL continues to be the treatment of choice for calculi less than 20 mm., the effectiveness of this type of therapy depends on various factors, such as stone composition, stone location and pelvicaliceal anatomy.6 Of these factors the one that eludes physicians is the stone composition, which cannot be predicted accurately before retrieved stones are analyzed. The type of crystals and/or particles excreted in urine after ESWL may be an indicator of stone composition.7, 8 Urinalysis with scanning electron microscopy and x-ray energy dispersive spectroscopy for determining stone composition before ESWL9 has its own limitations. Plain radiography has been evaluated for identifying calculi susceptible to fragmentation by ESWL. Chaussy and Fuchs believed that if a stone has lesser radiodensity than the spine, it is likely to break readily, while stones more radiodense than spine would be more difficult to fragment.10 During ESWL of ureteral calculi Mattelaer et al observed that high radio-opaque stones needed an average of 1.7 times more shock waves than low radio-opaque stones of the same size.11 The disadvantage of plain x-ray for predicting stone characteristics is obvious. Calculi must have an appreciable diameter without any bowel gas interference and away from bony structures. Moreover, density measurements are not uniform but subjective and, hence, clinical applicability is limited. CT is a noninvasive technique that is easily available and provides greater density discrimination than conventional radiography.
TABLE 2. Outcome after ESWL in various patient groups Group 1 2 3
No. Pts. 12 7 11
p Value
Median No. Shock Waves Required for Fragmentation (interquartile range)
No. Fragmented, Cleared (%)
No. Significant Residual Fragments
No. No Fragmentation
2,500 (2,000–3,000) 3,390 (2,500–4,501) 7,300 (5,500–8,500) 0.001 (Kruskall-Wallis ANOVA)
12 (100) 6 (85.7) 6 (54.5) 0.022 (contingency table analysis)
0 1 4
0 0 1
1970
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS TABLE 3. Comparison of variables in patients with and without a successful outcome of ESWL Variables
Success
Failure
No. pts. Mean age ⫾ SD Mean body mass index ⫾ SD Mean calculous diameter ⫾ SD (mm.) Mean calculous CT attenuation value ⫾ SD (Hounsfield units) Median no. shock waves required for fragmentation (interquartile range)
24 39.9 ⫾ 14.2 23.2 ⫾ 2.9 12.0 ⫾ 4.5 631.6 ⫾ 328.3
6 36.5 ⫾ 12.9 24.2 ⫾ 3.9 15.0 ⫾ 5.0 1,224.8 ⫾ 218.1
p Value
3,000 (2,500–4,505)
8,750 (8,300–10,100)
0.585 (unpaired 0.586 (unpaired 0.217 (unpaired ⬍0.001 (unpaired
Student’s Student’s Student’s Student’s
t t t t
test) test) test) test)
⬍0.001 (Mann-Whitney U test)
FIG. 2. Scatterplot demonstrates correlation of mean CT attenuation values of renal calculi with number of shock waves required for stone fragmentation.
CT can distinguish a density difference of 0.5%, while plain x-ray requires a density difference of about 5%.12 CT has been done to analyze the chemical composition of renal calculi in vitro.13, 14 Using computerized mapping techniques Hillman et al differentiated uric acid from struvite and calcium oxalate calculi, and suggesting its clinical usefulness.13 On a MEDLINE search of the literature we did not identify any clinical study correlating CT measured stone density with calculous fragmentation by ESWL. In our clinical study the range of mean CT attenuation values of urate stones was 136 to 402 Hounsfield units and that of calcium oxalate dihydrate stones was 371 to 588 Hounsfield units. These calculi are known to fragment easily on ESWL15 and we observed 100% success in such cases. Overall the CT attenuation value of renal calculi had an inverse relationship with the ESWL success rate. The higher the attenuation, the less chance of success. The mean calculous attenuation value significantly correlated positively with the number of shock waves required for fragmentation. Of the 6 patients in whom
ESWL was not successful 5 had stones with a mean attenuation value exceeding 1,000 Hounsfield units (1,078 to 1,585), while the remaining 1 had a stone with a mean value approaching 1,000 Hounsfield units (956). Our data suggest that when a stone has a mean attenuation value of greater than 950 Hounsfield units and 7,500 shocks have not successfully achieved fragmentation, percutaneous nephrolithotomy should be performed and a third session of ESWL should not be considered. Stone location has a bearing on stone clearance after ESWL. Lower caliceal stones are associated with a relatively lower ESWL success rate.16, 17 In our study 2 patients with a 9 and 10 mm. stone, respectively, in a lower pole calix had complete clearance. CONCLUSIONS
In this study the CT attenuation value of a renal calculus correlated well with the outcome after ESWL and number of
TABLE 4. Chemical composition, CT attenuation value and treatment outcome of renal calculi Chemical Composition
No. Pts.
Mean Attenuation Value (range) (Hounsfield units)
Av. No. Shock Waves Required for Fragmentation
No. Success
No. Failure
Total No.
Urate, calcium oxalate traces Calcium oxalate dihydrate Calcium oxalate, unknown predominant composition Calcium oxalate, phosphate
3 4 10
136–402 (276) 371–588 (476) 467–1,330 (984)
1,860 3,010 5,912
3 4 7
0 0 3
3 4 10
3
412–1,585 (955)
5,500
2
1
3
16
4
20
Totals
20
COMPUTERIZED TOMOGRAPHY ATTENUATION VALUE OF RENAL CALCULUS
shock waves required during therapy. In a subset of patients with renal calculi otherwise suitable for ESWL a greater number of shock waves was required for fragmentation or ESWL was not successful. In these cases the mean calculous CT attenuation value exceeded 1,000 Hounsfield units. The possible outcome of ESWL may be explained to these patients before treatment. However, a larger trial is necessary. REFERENCES
1. Wolf, J. S. and Clayman, R. V.: Percutaneous nephrostolithotomy. Urol Clin N Am, 24: 43, 1997 2. Lingeman, J. E., Newman, D., Mertz, J. H. O. et al: Extracorporeal shock wave lithotripsy: the Methodist Indiana experience. J Urol, 135: 1134, 1996 3. Ehreth, J. T., Drach, G. W., Arnett, M. L. et al: Extracorporeal shock wave lithotripsy: multicenter study of kidney and upper ureter versus middle and lower ureter treatments. J Urol, 152: 1379, 1994 4. Cass, A. S.: Comparison of first generation (Dornier HM3) and second generation (Medstone STS) lithotriptors: treatment results with 13,864 renal and ureteral calculi. J Urol, 153: 588, 1995 5. Lingeman, J. E., Woods, J. R. and Toth, P. D.: Blood pressure changes following extracorporeal shock wave lithotripsy and other forms of treatment for nephrolithiasis. JAMA, 263: 1789, 1990 6. Bon, D., Dore, B., Irani, J. et al: Radiographic prognostic criteria for extracorporeal shock wave lithotripsy. Urology, 48: 556, 1996 7. Khan, S. R., Hackett, R. L. and Finlayson, B.: Morphology of
8. 9.
10. 11. 12. 13. 14. 15. 16. 17.
1971
urinary stone particles resulting from ESWL treatment. J Urol, 136: 1367, 1986 Bowsher, W. G., Crocker, P., Ramsay, J. W. A. et al: Single urine sample diagnosis. A new concept in stone analysis. Br J Urol, 65: 236, 1990 Cohen, N. P., Parkhouse, H., Scott, M. L. et al: Prediction of response to lithotripsy: the use of scanning electron microscopy and X-ray energy dispersive spectroscopy. Br J Urol, 70: 469, 1992 Chaussy, C. G. and Fuchs, G. J.: Extracorporeal shockwave lithotripsy. Monogr Urol, 4: 80, 1987a Mattelaer, P., Schroder, T., Fischer, N. et al: In situ extracorporeal shock wave lithotripsy of distal ureteral stones: parameters for therapeutic success. Urol Int, 53: 87, 1994 Federle, M. P., McAninch, J. W., Kaiser, J. A. et al: Computed tomography of urinary calculi. AJR Am J Roentgenol, 136: 255, 1981 Hillman, B. J., Drach, G. W., Tracey, P. et al: Computed tomographic analysis of renal calculi. AJR Am J Roentgenol, 142: 549, 1984 Mostafavi, M. R., Ernst, R. D. and Saltzman, B.: Accurate determination of chemical composition of urinary calculi by spiral computerized tomography. J Urol, 159: 673, 1998 Dretler, S. P.: Stone fragility: a new therapeutic distinction. J Urol, 139: 1124, 1988 Sabnis, R. B., Naik, K., Patel, S. H. et al: Extracorporeal shock wave lithotripsy for lower calyceal stones: can clearance be predicted? Br J Urol, 80: 853, 1997 Madbouly, K., Sheir, K. Z. and Elsobsky, E.: Impact of lower pole renal anatomy on stone clearance after shock wave lithotripsy: fact or fiction? J Urol, 165: 1415, 2001