- Email: [email protected]
HELICAL COMPUTERIZED TOMOGRAPHY ARTERIOGRAPHY FOR EVALUATION OF LIVE RENAL DONORS UNDERGOING LAPAROSCOPIC NEPHRECTOMY
0022-5347/99/1621-003UO THEJOIXNN.OF UROLOGY Copyright 0 1999 by AMERICAN UROLOCICAL ASSOCIATION, INC
Vol. 162. 3 1 3 4 . July
1999
Printed i n U.S.A.
HELICAL COMPUTERIZED TOMOGRAPHY ARTERIOGRAPHY FOR EVALUATION OF LIVE RENAL DONORS UNDERGOING LAF'AROSCOPIC NEPHRECTOMY JOSEPH J. D E L PIZZO,'* GEOFFREY N. SKLAR, J A D E WONG YOU-CHEONG, BRIAN LEVIN, THORSTEN KREBS Froin the Departnzeiits of Surgery and Radiolog,>!,Divisioii
of
S T E P H E N C. JACOBS
ANI)
Urology3 Uriiurrsity of M n r ~ l o i i dSchool
of' Medicine.
Baltiniore, Marslarid
ABSTRACT
Purpose: Traditionally, live r e n a l donors are evaluated w i t h excretory urography a n d renal arteriography. Helical computerized tomography ( C T ) arteriography offers a less invasive alternative for d e m o n s t r a t i n g necessary anatomical information before laparoscopic allograft harvest. W e evaluate the accuracy of helical C T arteriography in depicting renal vascular a n a t o m y with an e m p h a s i s o n the detection of arterial and venous anomalies. Materials and Methods: Imaging studies w e r e done o n 175 patients according to a s t a n d a r d CT arteriography protocol w i t h early arterial phase s c a n n i n g (14to 20-second delay), and 1 mm. axial and 3-dimensional m a x i m u m intensity projection reconstructions. R e n a l vascular a n a t o m y w a s mapped w i t h attention t o aberrant arterial and venous anatomy. Intraoperative findings were correlated at laparoscopic donor nephrectomy. Results: There w a s overall a g r e e m e n t between CT arteriography and laparoscopic findings in 163 cases (93%). S u p e r n u m e r a r y renal arteries w e r e identified in 40 cases (23%). Sensitivity, specificity and accuracy of CT arteriography for arterial a n a t o m y w e r e 91, 98 and 96'70, respectively. Cases w i t h less than 2 m m . accessory arteries or early branching single vessels simulating dual arteries w e r e misdiagnosed. Venous anomalies occurred in 11patients (6.3%). Sensitivity, specificity a n d accuracy of C T arteriography for venous a n a t o m y w e r e 65, 100, and 97%, respectively. Misdiagnoses included early venous bifurcations and s u p e r n u m e r a r y tributary veins, which w e r e poorly opacified. Conclusions: Helical C T is highly accurate and specific for the demonstration of renal arterial anatomy. Poor opacification resulted in a lower sensitivity for venous anatomy. Overall, helical C T provides essential anatomical information, and is an alternative to standard urography and arteriography . KEY WORDS:tomography, angiography, laparoscopy, ludney transplantation Potential live renal donors undergo extensive preoperative evaluation which includes medical history, laboratory testing and radiological imaging. 1-3 Traditionally, imaging has included excretory urography (IVP) and conventional renal a r t e r i ~ g r a p h y . ~N-p ~ is used to confirm 2 normally functioning kidneys and ureters, and identify parenchymal lesions, renal calculi and congenital anomalies. Renal arteriography is performed to delineate renal vascular anatomy, number of renal arteries, early arterial branching and renal vascular disease. The traditional role of these 2 imaging modalities in the evaluation of potential kidney donors has been challenged by helical computerized tomography (CT) arteri~graphy.~-" Helical CT arteriography can be used to image the entire region of the kidney and obtain relatively noninvasively arteriogram-like images of the renal arteries using peripheral intravenous injection of contrast material. In addition, these images can be reformatted t o provide the surgeon with 3-dimensional (D) display of the data and allow improved vascular imaging."." A delayed plain abdominal film is taken to demonstrate the collecting systems and ureters, simulating IVP. Thus, CT arteriography has the potential to be used as a single imaging modality in preoperative evaluation of potenAccepted for publication February 5 , 1999. * Requests for reprints: Division of Urology, University of Maryland School of Medicine, 22 S. Greene St., Room S8D18, Baltimore, Maryland 21201.
tial renal donors. Advantages of using CT alone include reduced cost of imaging, and decreased patient discomfort and morbidity associated with arteriography. In addition, crosssectional images may detect a pathological condition that would preclude renal donation. However, to be satisfactory as a sole imaging technique CT arteriography must provide the surgeon with the necessary anatomical and vascular information to determine suitability for renal transplantation and allow the kidney to be harvested safely from the live renal donors. Recently a few small studies have indicated that CT may have an accuracy rate sufficient to replace IVP and renal a r t e r i ~ g r a p h y . ~ We - ~ evaluate the accuracy of helical CT arteriography in depicting renal vascular anatomy, especially arterial and venous anomalies, in a large series of live renal donors evaluated for laparoscopic donor nephrectomy. PATIENTS AND METHODS
A total of 103 female and 75 male potential kidney donors 20 to 62 years old (mean age 35)were evaluated between July 1996 and April 1998. Before imaging clinical and laboratory assessment was performed to exclude from donation candidates with renal dysfunction and hypertension. Donors had documented histocompatibility with recipients. Of 178 patients 175 underwent laparoscopic allograft harvest. Imaging was done according to a standard CT arteriography protocol using a high speed scanner. A standard CT abdominal scout image was obtained followed by contiguous
31
32
EVALUATING LIVE RENALDONORS UNDERGOING LAPAROSCOPIC NEPHRECTOMY
axial unenhanced images through the kidneys to localize the scanning area of interest, and assess the presence of 2 kidneys and abnormal renal or vascular calcifications. Then 150 ml. nonionic contrast medium at 4 ml. per second were injected into a peripheral vein via an 18 gauge angiographic catheter.During the arterial phase, 14 to 20 seconds after the start of iqjection, helical scanning was begun above the kidneys and continued to 3 cm. below the aortic bifurcation using 3 mm. collimation and a pitch varying from 1.5 to 1.8 depending on the required length of coverage. Immediately aRerwards a helical scan of the abdomen and pelvis was performed at 7 mm. collimation to evaluate the renal parenchyma for masses. A delayed abdominal scout film was also obtained to visualize the ureters and bladder. Helical CT images were reconstructed at 1 mm. thickness, and transferred to a workstation and reviewed before reformatting into 3-D coronal plane maximum intensity projection images. The 3-D axial and maximum intensity projection reconstructions were used for interpretation. Renal arterial anatomy was evaluated for the number and location of arteries, accessory arteries, early pre-hilar branching of the main renal artery (within 1.5 cm. of its origin) and arterial stenosis. Venous anatomy was evaluated and nonvascular abnormalities, including parenchymal and collecting system abnormalities, and calculi, were reported. All images were interpreted by the same 2 radiologists at our institution (J. J. W.and T.K.) and reviewed with the operating surgeon before laparoscopic allograft harvest. A total of 175 subjects underwent successful laparoscopic donor nephrectomy. The findings reported on CT arteriography were used to guide selection of the donor kidney and aided in the planned approach to isolate the vasculature of the selected kidney. At laparoscopic allograft harvest the surgeon noted the number of renal arteries and the presence of early branching, accessory vessels and renal vein anomalies. Preoperative helical CT arteriography results were correlated with intraoperative findings at laparoscopic donor nephrectomy. In cases of discrepancies between original CT arteriography interpretations and surgical findings the CT images were retrospectively reviewed and errors were classified as interpretive or technical.
renal arteries)in 3 and several small simple cysts in the right kidney in 1. CT arteriography findings in 175 donors. Supernumerary renal arteries were present in 40 cases (23%),and consisted of 2 arteries in 33 (19%)and 3 in 7 (4%).Early branching of the main renal artery within 1.5 cm. from its origin was seen in 26 cases (15%).Venous anomalies occurred in 11patients (6.3%),and consisted of retroaortic renal vein in 2, circumaortic renal vein in 3, multiple renal veins in 3 and early branching veins in 4. Nonvascular renal abnormalities were recorded in all 175 subjects. Renal calculi were seen in 5 and simple renal cysts were identified in 12 patients. A pelvic kidney was diagnosed via CT arteriography and harvested laparoscopically without difficulty. CT arteriography and surgery correlation. There was overall agreement between CT arteriography and laparoscopic findings in 163 cases (93%,figs. 1 and 2). CT arteriography was specific in the delineation of arterial anatomy. T w o cases were misdiagnosed when early branching single vessels found at laparoscopy were interpreted as dual arteries on initial CT arteriography. These surgical findings were identified on CT arteriography retrospectively. Under detection of accessory arteries resulted in slightly poorer sensitivity. CT arteriography did not prospectively reveal an accessory artery in 4 cases, of which 3 were identified on retrospective analysis. None of the CT arteriography errors affected surgical treatment of the renal donor. When the findings at surgery were used as the reference for determining renal arterial anatomy, sensitivity, specificity and accuracy of helical CT arteriography were 91, 98 and 96%,respectively. There were 7 cases of discordance between CT arteriography and surgery in the identification of venous anatomy. Preoperative CT arteriography missed supernumerary tributary veins in 4 cases and early venous bifurcations encountered at surgery in 3. On retrospective review 2 tributary veins and 1 early bifurcating left renal vein were seen. Poor opacification of the venous system appears to be the main factor contributing to the poorer sensitivity in delineating renal venous anatomy. Again, none of the CT arteriography errors led to significant bleeding or intraoperative morbidity. Sensitivity, specificity and accuracy of CT arteriography for venous anatomy were 65, 100, and 97%,respectively.
RESULTS
CT arteriography was performed on 356 kidneys in 178 potential renal donors. Image interpretations and subsequent intraoperative findings were correlated in the 175 kidneys harvested by laparoscopic nephrectomy. Of the 3 candidates excluded from donation based on preoperative CT arteriography 1 had a large renal arterial aneurysm, 1 had multiple bilateral renal stones and 1 had a 1.5 cm. solid left renal mass which proved to be renal cell carcinoma at resection. Nephrectomy was performed on the left side in 171 and on the right side in 4 donors after preoperative imaging showed complex vasculature of the left kidney (3 or more
DISCUSSION
Renal imaging is a crucial step in preoperative evaluation of potential renal donors.l4 Assessment of the kidneys and renal vasculature will identify some unsuitable donors and dictate which renal unit is chosen for transplantation in others. Traditionally, assessment has required 2 separate imaging modalities. An I W assesses renal size and pelvicaliceal anatomy, and detects renal or ureteral calcifications, and conventional renal arteriography assesses renal artery number and length, and underlying renal vascular disease. CT arteriography is a method for obtaining arteriogram-
RG.1. A, CT arteriogramshows 2 left renal arteries (arrows) off of aorta.B, intraoperativelyleft renal arteries (I and 2 )have been isolated and renal vein UIY) ie visible.
EVALUATING I, 1V E Ii E NA I, I ) ( ) N ( I
hl Y
33
FIG. 2. A, anterior and posterior 3-D reconstructed images demonstrate retroaortic left renal vein (arrowhead).B, retroaortic renal vein is isolated at laparoscopic donor nephrectomy (arrow) and is seen traveling behind aorta ( A ) .
I
,1 ~
like images using the volume acquisition capabilities inherent to helical CT. Recent advances allow these images to be reformatted and provide a 3-D display of the data. With this technology CT arteriography has been proposed as a single imaging test t o replace the urogram and arteriography in live renal donor A major advantage of CT over conventional arteriography is that these arteriogram-like images can be obtained relatively noninvasively using peripheral intravenous administration of contrast material. This method results in far less patient discomfort and morbidity than standard renal arteriography. At the same time CT imaging of the renal parenchyma and collecting system can be obtained, with information about other abdominal and pelvic structures. CT arteriography has an impact on decreasing the cost of preoperative testing for patients undergoing renal Precise delineation of vascular anatomy is crucial in patients undergoing donor nephrectomy, and especially a laparoscopic procedure. Laparoscopic dissection of the renal pedicle is technically challenging and maintenance of hemostasis is critical to complete successful laparoscopic harvest. Thus, preoperative identification of arterial and venous anatomy, particularly vascular branching and accessory vessels, is important. Our results indicate that CT arteriography is highly accurate and specific for the demonstration of renal arterial anatomy. Misdiagnoses occurred when early branching single vessels identified at surgery were interpreted as dual arteries on preoperative CT arteriography. In each case retrospective review of the 3-D images revealed a n early branch of the main renal artery within 1.5 cm. of the aorta. Under detection of accessory arteries on CT arteriography resulted in slightly poorer sensitivity. Accessory arteries missed prospectively by CT arteriography were visible on retrospective review of the images in 3 of 4 cases. Attention has been given to detection of these small (less than 2 mm.) accessory vessels to help surgeons avoid additional blood loss and possibly a focal renal infarct. However, most are sacrificed at surgery as they are too small t o transplant and do not affect patient outcome, since renal volume loss and bleeding are negligible. Although renal arteriography has long been standard for examination of renal arterial anatomy in renal donors, it is not perfect. When results of conventional renal arteriography are correlated with surgical findings, accessory vessels or early branches are missed in approximately 8%of patient^.''.'^ This discrepancy is often due to failure of arteriography to detect small arteries to the upper pole. Significant arteries can also be missed on conventional arteriography if they arise from the same level in the anteroposterior projection and overlap.12.l 3 Rotation of the 3-D data set with helical CT can separate overlapping arteries without additional injection of contrast medium, unlike arteriography.". '' CT arteriography was less sensitive for the detection of venous anatomy. After peripheral venous injection of contrast material the bolus becomes maximally diluted in the venous system, so that high levels of venous contrast enhancement are not as readily achieved as they are in the arterial system. In addition, a short scan delay (less than 15
seconds), while allowing clearer delineation of the arterial anatomy, will result in poorer venous opacification. The optimal interval between onset of injection and scan initiation depends on several patient factors including cardiac function, and may be difficult to gauge. A test bolus of contrast medium may be used t o estimate transit time and optimum scan delay but none was used in our study. Venous anomalies, such as multiple renal veins, a retroaortic left renal vein and a circumaortic left renal vein, were easily identified by helical CT with 3-D reconstruction. Our lower sensitivity resulted from missed early venous bifurcations and supernumerary tributary veins which were poorly opacified on preoperative CT. On retrospective review of 7 surgically confirmed cases 3 missed veins were seen and 4 supernumerary veins were still not identified. Venous opacification and anatomical detection are better on CT than conventional arteriography. Although delineation is not as crucial for venous as for arterial anatomy, the information can aid the surgeon. Preoperative identification of venous anomalies can result in shorter operative times and minimize morbidity from unexpected vascular injury. At our institution the CT arteriography protocol was modified slightly to increase the scan delay after contrast medium administration from 14 to 20 seconds in an attempt to opacify and characterize the venous system better. Helical CT arteriography also confers several advantages over urography and conventional arteriography in the detection of nonvascular abnormalitie~.~.' Benign and malignant masses, renal and ureteral calculi, parenchymal scarring, congenital anomalies, ureteral duplication and caliceal diverticula can be identified on unenhanced and enhanced CT, which are more sensitive for these abnormalities than arteri~graphy.~.'*'"In addition, CT enables evaluation of other abdominal and pelvic organs, and may identify patients not suitable for kidney donation secondary to abnormalities outside the kidneys. CT arteriography can be performed as a safe outpatient procedure. Patients do not need to be monitored and placed on bedrest as with renal arteriography. There are fewer complications inherent to a peripheral venous puncture for contrast medium administration compared with the more invasive arterial access required for arteriography. In addition, multiple anatomical views can be obtained with CT arteriography from the same contrast medium injection, compared to arteriography which requires repeat injection when turning the patient to obtain additional images. Also, there is a 40 to 50% reduction in cost for imaging the potential donor with CT arteriography versus standard renal arteriography and CONCLUSIONS
Helical CT arteriography is accurate and highly specific for arterial and venous anatomy. It provides all information needed by the surgical team before open or laparoscopic live donor nephrectomy. At our institution this imaging modality provides the surgeon with a detailed picture of renal vascular anatomy in preparation for laparoscopic allograft harvest. Helical CT arteriography can become the primary imaging
34
EVALUATING LIVE RENAL DONORS UNDERGOING LAPAROSCOPIC NEPHRECTOMY
modality for preoperative assessment of potential renal donors in lieu of arteriography and I W . Continued experience with interpretation and identification of venous anatomy should result in continued accurate preoperative assessment of live renal donors while reducing morbidity and overall imaging cost. REFERENCES
Fauchald, P.: Influence of renal angiography in living potential renal donors. Acta Rad., 3 2 368,1991. 7. Rubin, G. D., Alfrey, E. J., Dake, M. D., Semba, C. P., Sommer, F. G., Kuo, P. C., Dafoe, D. C., Waskerwitz, J. A., Bloch, D. A. and Jeffrey, R. B.: Assessment of living renal donors with spiral CT. Radiology, 195 457, 1995. 8. Cochran, S. T., Kransy, R. M., Danovitch, R. M., Rajfer, J., Barbaric, Z. M., Wilkinson, A. and Rosenthal, J . T.: Helical CT angiography for examination of living renal donors. AJR, 168:
1569,1997. 1. Lowell, J. A. and Taylor, R. J.: The evaluation of the living renal 9. Bluemke, D. A. and Chambers, T. P.: Spiral CT angiography: an donor, surgical techniques and results. Sem. Urol., 1 2 102. alternative to conventional angiography (Editorial). Radiol1994. ogy, 195: 317,1995. 2. Riehle, R. A., Jr., Steckler, R.. Naslund, E. B.. Riggio. R.. Cheigh, 10. Platt, J. F.,Ellis, J. H., Korobkin, M. and Reige, K.: Helical CT J. and Stubenbord, W.: Selection criteria for the evaluation of evaluation of potential kidney donors: findings in 154 subjects. living related renal donors. J . Urol.. 144. 845,1990. AJR, 1 6 9 1325,1997. 3. Spanos, P. K.,Simmons, R. L., Kjellstrand, C. M., Buselmeier, T. J. and Najarian, J . S.: Screening potential related trans- 11. Dachman, A. H., Newmark, G. M.. Mitchell, M. T. and Woodle. E. S.: Helical CT examination of potential kidney donors. AJR: plant donors for renal disease. Lancet, 1: 645,1974. 171: 193, 1998. 4. Derauf. B. and Goldberg. M. E.: Angiographic assessment of potential renal transplant donors. Rad. Clin. N. Amer., 2 5 12. Sussman, S.K., Weinerth, J . L., Braun, S. D., Saeed, M., Illescas, 261, 1987. F. F., Cohan, R. H., Newman, G. E., Perlmutt, L. M. and 5. Walker. T. G., Geller. S. C., Delmonico, F. L., Waltman, A. C. and Dunnick, N. R.: Intravenous digital subtraction angiography Anthanasoulis, C. A.: Donor renal angiography: its influence in the evaluation of potential renal donors. J. Urol., 138 28, on the decision to use the right or left kidney. AJR, 151: 1149, 1987. 1988. 13. Sherwood, T., Ruutu, M. and Chisholm, G. D.: Renal angiogra6. Kjellevand, T. O., Kolmannskog, F., Pfeffer, P., Scholz, T. and phy problems in live kidney donors. Brit. J . Rad., 51: 99,1978.
Vol. 162. 3 1 3 4 . July
1999
Printed i n U.S.A.
HELICAL COMPUTERIZED TOMOGRAPHY ARTERIOGRAPHY FOR EVALUATION OF LIVE RENAL DONORS UNDERGOING LAF'AROSCOPIC NEPHRECTOMY JOSEPH J. D E L PIZZO,'* GEOFFREY N. SKLAR, J A D E WONG YOU-CHEONG, BRIAN LEVIN, THORSTEN KREBS Froin the Departnzeiits of Surgery and Radiolog,>!,Divisioii
of
S T E P H E N C. JACOBS
ANI)
Urology3 Uriiurrsity of M n r ~ l o i i dSchool
of' Medicine.
Baltiniore, Marslarid
ABSTRACT
Purpose: Traditionally, live r e n a l donors are evaluated w i t h excretory urography a n d renal arteriography. Helical computerized tomography ( C T ) arteriography offers a less invasive alternative for d e m o n s t r a t i n g necessary anatomical information before laparoscopic allograft harvest. W e evaluate the accuracy of helical C T arteriography in depicting renal vascular a n a t o m y with an e m p h a s i s o n the detection of arterial and venous anomalies. Materials and Methods: Imaging studies w e r e done o n 175 patients according to a s t a n d a r d CT arteriography protocol w i t h early arterial phase s c a n n i n g (14to 20-second delay), and 1 mm. axial and 3-dimensional m a x i m u m intensity projection reconstructions. R e n a l vascular a n a t o m y w a s mapped w i t h attention t o aberrant arterial and venous anatomy. Intraoperative findings were correlated at laparoscopic donor nephrectomy. Results: There w a s overall a g r e e m e n t between CT arteriography and laparoscopic findings in 163 cases (93%). S u p e r n u m e r a r y renal arteries w e r e identified in 40 cases (23%). Sensitivity, specificity and accuracy of CT arteriography for arterial a n a t o m y w e r e 91, 98 and 96'70, respectively. Cases w i t h less than 2 m m . accessory arteries or early branching single vessels simulating dual arteries w e r e misdiagnosed. Venous anomalies occurred in 11patients (6.3%). Sensitivity, specificity a n d accuracy of C T arteriography for venous a n a t o m y w e r e 65, 100, and 97%, respectively. Misdiagnoses included early venous bifurcations and s u p e r n u m e r a r y tributary veins, which w e r e poorly opacified. Conclusions: Helical C T is highly accurate and specific for the demonstration of renal arterial anatomy. Poor opacification resulted in a lower sensitivity for venous anatomy. Overall, helical C T provides essential anatomical information, and is an alternative to standard urography and arteriography . KEY WORDS:tomography, angiography, laparoscopy, ludney transplantation Potential live renal donors undergo extensive preoperative evaluation which includes medical history, laboratory testing and radiological imaging. 1-3 Traditionally, imaging has included excretory urography (IVP) and conventional renal a r t e r i ~ g r a p h y . ~N-p ~ is used to confirm 2 normally functioning kidneys and ureters, and identify parenchymal lesions, renal calculi and congenital anomalies. Renal arteriography is performed to delineate renal vascular anatomy, number of renal arteries, early arterial branching and renal vascular disease. The traditional role of these 2 imaging modalities in the evaluation of potential kidney donors has been challenged by helical computerized tomography (CT) arteri~graphy.~-" Helical CT arteriography can be used to image the entire region of the kidney and obtain relatively noninvasively arteriogram-like images of the renal arteries using peripheral intravenous injection of contrast material. In addition, these images can be reformatted t o provide the surgeon with 3-dimensional (D) display of the data and allow improved vascular imaging."." A delayed plain abdominal film is taken to demonstrate the collecting systems and ureters, simulating IVP. Thus, CT arteriography has the potential to be used as a single imaging modality in preoperative evaluation of potenAccepted for publication February 5 , 1999. * Requests for reprints: Division of Urology, University of Maryland School of Medicine, 22 S. Greene St., Room S8D18, Baltimore, Maryland 21201.
tial renal donors. Advantages of using CT alone include reduced cost of imaging, and decreased patient discomfort and morbidity associated with arteriography. In addition, crosssectional images may detect a pathological condition that would preclude renal donation. However, to be satisfactory as a sole imaging technique CT arteriography must provide the surgeon with the necessary anatomical and vascular information to determine suitability for renal transplantation and allow the kidney to be harvested safely from the live renal donors. Recently a few small studies have indicated that CT may have an accuracy rate sufficient to replace IVP and renal a r t e r i ~ g r a p h y . ~ We - ~ evaluate the accuracy of helical CT arteriography in depicting renal vascular anatomy, especially arterial and venous anomalies, in a large series of live renal donors evaluated for laparoscopic donor nephrectomy. PATIENTS AND METHODS
A total of 103 female and 75 male potential kidney donors 20 to 62 years old (mean age 35)were evaluated between July 1996 and April 1998. Before imaging clinical and laboratory assessment was performed to exclude from donation candidates with renal dysfunction and hypertension. Donors had documented histocompatibility with recipients. Of 178 patients 175 underwent laparoscopic allograft harvest. Imaging was done according to a standard CT arteriography protocol using a high speed scanner. A standard CT abdominal scout image was obtained followed by contiguous
31
32
EVALUATING LIVE RENALDONORS UNDERGOING LAPAROSCOPIC NEPHRECTOMY
axial unenhanced images through the kidneys to localize the scanning area of interest, and assess the presence of 2 kidneys and abnormal renal or vascular calcifications. Then 150 ml. nonionic contrast medium at 4 ml. per second were injected into a peripheral vein via an 18 gauge angiographic catheter.During the arterial phase, 14 to 20 seconds after the start of iqjection, helical scanning was begun above the kidneys and continued to 3 cm. below the aortic bifurcation using 3 mm. collimation and a pitch varying from 1.5 to 1.8 depending on the required length of coverage. Immediately aRerwards a helical scan of the abdomen and pelvis was performed at 7 mm. collimation to evaluate the renal parenchyma for masses. A delayed abdominal scout film was also obtained to visualize the ureters and bladder. Helical CT images were reconstructed at 1 mm. thickness, and transferred to a workstation and reviewed before reformatting into 3-D coronal plane maximum intensity projection images. The 3-D axial and maximum intensity projection reconstructions were used for interpretation. Renal arterial anatomy was evaluated for the number and location of arteries, accessory arteries, early pre-hilar branching of the main renal artery (within 1.5 cm. of its origin) and arterial stenosis. Venous anatomy was evaluated and nonvascular abnormalities, including parenchymal and collecting system abnormalities, and calculi, were reported. All images were interpreted by the same 2 radiologists at our institution (J. J. W.and T.K.) and reviewed with the operating surgeon before laparoscopic allograft harvest. A total of 175 subjects underwent successful laparoscopic donor nephrectomy. The findings reported on CT arteriography were used to guide selection of the donor kidney and aided in the planned approach to isolate the vasculature of the selected kidney. At laparoscopic allograft harvest the surgeon noted the number of renal arteries and the presence of early branching, accessory vessels and renal vein anomalies. Preoperative helical CT arteriography results were correlated with intraoperative findings at laparoscopic donor nephrectomy. In cases of discrepancies between original CT arteriography interpretations and surgical findings the CT images were retrospectively reviewed and errors were classified as interpretive or technical.
renal arteries)in 3 and several small simple cysts in the right kidney in 1. CT arteriography findings in 175 donors. Supernumerary renal arteries were present in 40 cases (23%),and consisted of 2 arteries in 33 (19%)and 3 in 7 (4%).Early branching of the main renal artery within 1.5 cm. from its origin was seen in 26 cases (15%).Venous anomalies occurred in 11patients (6.3%),and consisted of retroaortic renal vein in 2, circumaortic renal vein in 3, multiple renal veins in 3 and early branching veins in 4. Nonvascular renal abnormalities were recorded in all 175 subjects. Renal calculi were seen in 5 and simple renal cysts were identified in 12 patients. A pelvic kidney was diagnosed via CT arteriography and harvested laparoscopically without difficulty. CT arteriography and surgery correlation. There was overall agreement between CT arteriography and laparoscopic findings in 163 cases (93%,figs. 1 and 2). CT arteriography was specific in the delineation of arterial anatomy. T w o cases were misdiagnosed when early branching single vessels found at laparoscopy were interpreted as dual arteries on initial CT arteriography. These surgical findings were identified on CT arteriography retrospectively. Under detection of accessory arteries resulted in slightly poorer sensitivity. CT arteriography did not prospectively reveal an accessory artery in 4 cases, of which 3 were identified on retrospective analysis. None of the CT arteriography errors affected surgical treatment of the renal donor. When the findings at surgery were used as the reference for determining renal arterial anatomy, sensitivity, specificity and accuracy of helical CT arteriography were 91, 98 and 96%,respectively. There were 7 cases of discordance between CT arteriography and surgery in the identification of venous anatomy. Preoperative CT arteriography missed supernumerary tributary veins in 4 cases and early venous bifurcations encountered at surgery in 3. On retrospective review 2 tributary veins and 1 early bifurcating left renal vein were seen. Poor opacification of the venous system appears to be the main factor contributing to the poorer sensitivity in delineating renal venous anatomy. Again, none of the CT arteriography errors led to significant bleeding or intraoperative morbidity. Sensitivity, specificity and accuracy of CT arteriography for venous anatomy were 65, 100, and 97%,respectively.
RESULTS
CT arteriography was performed on 356 kidneys in 178 potential renal donors. Image interpretations and subsequent intraoperative findings were correlated in the 175 kidneys harvested by laparoscopic nephrectomy. Of the 3 candidates excluded from donation based on preoperative CT arteriography 1 had a large renal arterial aneurysm, 1 had multiple bilateral renal stones and 1 had a 1.5 cm. solid left renal mass which proved to be renal cell carcinoma at resection. Nephrectomy was performed on the left side in 171 and on the right side in 4 donors after preoperative imaging showed complex vasculature of the left kidney (3 or more
DISCUSSION
Renal imaging is a crucial step in preoperative evaluation of potential renal donors.l4 Assessment of the kidneys and renal vasculature will identify some unsuitable donors and dictate which renal unit is chosen for transplantation in others. Traditionally, assessment has required 2 separate imaging modalities. An I W assesses renal size and pelvicaliceal anatomy, and detects renal or ureteral calcifications, and conventional renal arteriography assesses renal artery number and length, and underlying renal vascular disease. CT arteriography is a method for obtaining arteriogram-
RG.1. A, CT arteriogramshows 2 left renal arteries (arrows) off of aorta.B, intraoperativelyleft renal arteries (I and 2 )have been isolated and renal vein UIY) ie visible.
EVALUATING I, 1V E Ii E NA I, I ) ( ) N ( I
hl Y
33
FIG. 2. A, anterior and posterior 3-D reconstructed images demonstrate retroaortic left renal vein (arrowhead).B, retroaortic renal vein is isolated at laparoscopic donor nephrectomy (arrow) and is seen traveling behind aorta ( A ) .
I
,1 ~
like images using the volume acquisition capabilities inherent to helical CT. Recent advances allow these images to be reformatted and provide a 3-D display of the data. With this technology CT arteriography has been proposed as a single imaging test t o replace the urogram and arteriography in live renal donor A major advantage of CT over conventional arteriography is that these arteriogram-like images can be obtained relatively noninvasively using peripheral intravenous administration of contrast material. This method results in far less patient discomfort and morbidity than standard renal arteriography. At the same time CT imaging of the renal parenchyma and collecting system can be obtained, with information about other abdominal and pelvic structures. CT arteriography has an impact on decreasing the cost of preoperative testing for patients undergoing renal Precise delineation of vascular anatomy is crucial in patients undergoing donor nephrectomy, and especially a laparoscopic procedure. Laparoscopic dissection of the renal pedicle is technically challenging and maintenance of hemostasis is critical to complete successful laparoscopic harvest. Thus, preoperative identification of arterial and venous anatomy, particularly vascular branching and accessory vessels, is important. Our results indicate that CT arteriography is highly accurate and specific for the demonstration of renal arterial anatomy. Misdiagnoses occurred when early branching single vessels identified at surgery were interpreted as dual arteries on preoperative CT arteriography. In each case retrospective review of the 3-D images revealed a n early branch of the main renal artery within 1.5 cm. of the aorta. Under detection of accessory arteries on CT arteriography resulted in slightly poorer sensitivity. Accessory arteries missed prospectively by CT arteriography were visible on retrospective review of the images in 3 of 4 cases. Attention has been given to detection of these small (less than 2 mm.) accessory vessels to help surgeons avoid additional blood loss and possibly a focal renal infarct. However, most are sacrificed at surgery as they are too small t o transplant and do not affect patient outcome, since renal volume loss and bleeding are negligible. Although renal arteriography has long been standard for examination of renal arterial anatomy in renal donors, it is not perfect. When results of conventional renal arteriography are correlated with surgical findings, accessory vessels or early branches are missed in approximately 8%of patient^.''.'^ This discrepancy is often due to failure of arteriography to detect small arteries to the upper pole. Significant arteries can also be missed on conventional arteriography if they arise from the same level in the anteroposterior projection and overlap.12.l 3 Rotation of the 3-D data set with helical CT can separate overlapping arteries without additional injection of contrast medium, unlike arteriography.". '' CT arteriography was less sensitive for the detection of venous anatomy. After peripheral venous injection of contrast material the bolus becomes maximally diluted in the venous system, so that high levels of venous contrast enhancement are not as readily achieved as they are in the arterial system. In addition, a short scan delay (less than 15
seconds), while allowing clearer delineation of the arterial anatomy, will result in poorer venous opacification. The optimal interval between onset of injection and scan initiation depends on several patient factors including cardiac function, and may be difficult to gauge. A test bolus of contrast medium may be used t o estimate transit time and optimum scan delay but none was used in our study. Venous anomalies, such as multiple renal veins, a retroaortic left renal vein and a circumaortic left renal vein, were easily identified by helical CT with 3-D reconstruction. Our lower sensitivity resulted from missed early venous bifurcations and supernumerary tributary veins which were poorly opacified on preoperative CT. On retrospective review of 7 surgically confirmed cases 3 missed veins were seen and 4 supernumerary veins were still not identified. Venous opacification and anatomical detection are better on CT than conventional arteriography. Although delineation is not as crucial for venous as for arterial anatomy, the information can aid the surgeon. Preoperative identification of venous anomalies can result in shorter operative times and minimize morbidity from unexpected vascular injury. At our institution the CT arteriography protocol was modified slightly to increase the scan delay after contrast medium administration from 14 to 20 seconds in an attempt to opacify and characterize the venous system better. Helical CT arteriography also confers several advantages over urography and conventional arteriography in the detection of nonvascular abnormalitie~.~.' Benign and malignant masses, renal and ureteral calculi, parenchymal scarring, congenital anomalies, ureteral duplication and caliceal diverticula can be identified on unenhanced and enhanced CT, which are more sensitive for these abnormalities than arteri~graphy.~.'*'"In addition, CT enables evaluation of other abdominal and pelvic organs, and may identify patients not suitable for kidney donation secondary to abnormalities outside the kidneys. CT arteriography can be performed as a safe outpatient procedure. Patients do not need to be monitored and placed on bedrest as with renal arteriography. There are fewer complications inherent to a peripheral venous puncture for contrast medium administration compared with the more invasive arterial access required for arteriography. In addition, multiple anatomical views can be obtained with CT arteriography from the same contrast medium injection, compared to arteriography which requires repeat injection when turning the patient to obtain additional images. Also, there is a 40 to 50% reduction in cost for imaging the potential donor with CT arteriography versus standard renal arteriography and CONCLUSIONS
Helical CT arteriography is accurate and highly specific for arterial and venous anatomy. It provides all information needed by the surgical team before open or laparoscopic live donor nephrectomy. At our institution this imaging modality provides the surgeon with a detailed picture of renal vascular anatomy in preparation for laparoscopic allograft harvest. Helical CT arteriography can become the primary imaging
34
EVALUATING LIVE RENAL DONORS UNDERGOING LAPAROSCOPIC NEPHRECTOMY
modality for preoperative assessment of potential renal donors in lieu of arteriography and I W . Continued experience with interpretation and identification of venous anatomy should result in continued accurate preoperative assessment of live renal donors while reducing morbidity and overall imaging cost. REFERENCES
Fauchald, P.: Influence of renal angiography in living potential renal donors. Acta Rad., 3 2 368,1991. 7. Rubin, G. D., Alfrey, E. J., Dake, M. D., Semba, C. P., Sommer, F. G., Kuo, P. C., Dafoe, D. C., Waskerwitz, J. A., Bloch, D. A. and Jeffrey, R. B.: Assessment of living renal donors with spiral CT. Radiology, 195 457, 1995. 8. Cochran, S. T., Kransy, R. M., Danovitch, R. M., Rajfer, J., Barbaric, Z. M., Wilkinson, A. and Rosenthal, J . T.: Helical CT angiography for examination of living renal donors. AJR, 168:
1569,1997. 1. Lowell, J. A. and Taylor, R. J.: The evaluation of the living renal 9. Bluemke, D. A. and Chambers, T. P.: Spiral CT angiography: an donor, surgical techniques and results. Sem. Urol., 1 2 102. alternative to conventional angiography (Editorial). Radiol1994. ogy, 195: 317,1995. 2. Riehle, R. A., Jr., Steckler, R.. Naslund, E. B.. Riggio. R.. Cheigh, 10. Platt, J. F.,Ellis, J. H., Korobkin, M. and Reige, K.: Helical CT J. and Stubenbord, W.: Selection criteria for the evaluation of evaluation of potential kidney donors: findings in 154 subjects. living related renal donors. J . Urol.. 144. 845,1990. AJR, 1 6 9 1325,1997. 3. Spanos, P. K.,Simmons, R. L., Kjellstrand, C. M., Buselmeier, T. J. and Najarian, J . S.: Screening potential related trans- 11. Dachman, A. H., Newmark, G. M.. Mitchell, M. T. and Woodle. E. S.: Helical CT examination of potential kidney donors. AJR: plant donors for renal disease. Lancet, 1: 645,1974. 171: 193, 1998. 4. Derauf. B. and Goldberg. M. E.: Angiographic assessment of potential renal transplant donors. Rad. Clin. N. Amer., 2 5 12. Sussman, S.K., Weinerth, J . L., Braun, S. D., Saeed, M., Illescas, 261, 1987. F. F., Cohan, R. H., Newman, G. E., Perlmutt, L. M. and 5. Walker. T. G., Geller. S. C., Delmonico, F. L., Waltman, A. C. and Dunnick, N. R.: Intravenous digital subtraction angiography Anthanasoulis, C. A.: Donor renal angiography: its influence in the evaluation of potential renal donors. J. Urol., 138 28, on the decision to use the right or left kidney. AJR, 151: 1149, 1987. 1988. 13. Sherwood, T., Ruutu, M. and Chisholm, G. D.: Renal angiogra6. Kjellevand, T. O., Kolmannskog, F., Pfeffer, P., Scholz, T. and phy problems in live kidney donors. Brit. J . Rad., 51: 99,1978.