- Email: [email protected]
Magnetic resonance imaging and magnetic resonance angiography before postchemotherapy retroperitoneal lymph node dissection
ADULT UROLOGY
MAGNETIC RESONANCE IMAGING AND MAGNETIC RESONANCE ANGIOGRAPHY BEFORE POSTCHEMOTHERAPY RETROPERITONEAL LYMPH NODE DISSECTION DAVID A. CORRAL, DATLA G. VARMA, EDWARD F. JACKSON, ROBERT J. AMATO, S. MACHELE DONAT, AND LOUIS L. PISTERS
ABSTRACT Objectives. Retroperitoneal lymph node dissection (RPLND) after primary chemotherapy is an accepted therapeutic approach for metastatic nonseminomatous germ cell testicular cancer. Because of the intense desmoplastic reaction and adherence to venous and arterial walls, accurate imaging of the retroperitoneal vasculature and its relation to residual tumor is essential. We report our experience with magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), including the recently developed technique of bolus-contrast MRA, in patients undergoing postchemotherapy RPLND. Methods. Eighteen patients underwent MRI of the retroperitoneal region before RPLND. In addition to routine sequences, MRA was performed in 10 patients, including 8 with a three-dimensional technique using bolus intravenous MR contrast. Results were compared with intraoperative and pathologic findings. Results. MRI and MRA provided detailed information on retroperitoneal vasculature and its relation to tumor, including multiple renal vessels (n ⫽ 5), duplex inferior vena cava (n ⫽ 1), left retroaortic renal vein (n ⫽ 2), and common iliac vein thrombus (n ⫽ 1). In all cases, bolus-contrast MRA provided unique information on the location and number of renal and lumbar arteries, in addition to information on the aorta and the mesenteric and iliac vessels. The origin and number of renal arteries were accurately identified in all patients by bolus-contrast MRA; 2 patients had supernumerary renal arteries discovered at RPLND that had not been identified on non-bolus-contrast MRI. Conclusions. Bolus-contrast three-dimensional MRA provides unique information on renal and lumbar vessels. The potential benefit of avoiding vascular injury during dissection should be prospectively evaluated. UROLOGY 55: 262–266, 2000. © 2000, Elsevier Science Inc.
D
uring the platinum-based chemotherapy era, primary chemotherapy followed by retroperitoneal lymph node dissection (RPLND) became a widely accepted therapeutic strategy for metastatic nonseminomatous germ cell testicular cancer (NSGCT).1 After chemotherapy, residual retroperitoneal tumors commonly exhibit an intense fi-
From the Departments of Urology, Radiology , and Genitourinary Oncology, University of Texas M. D. Anderson Cancer Center, Houston, Texas; and Department of Urology, Memorial SloanKettering Cancer Center, New York, New York Reprint requests: David A. Corral, M.D., Department of Urologic Oncology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263 Submitted: July 9, 1999, accepted (with revisions): September 8, 1999
262
© 2000, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
brotic, desmoplastic reaction intimately involving and often encasing the walls of major vessels and resulting in treacherous conditions for dissection.2,3 Accurate preoperative imaging of residual tumor and its relation to retroperitoneal vasculature can provide critical information for intraoperative decision making during RPLND. Computed tomography (CT) scanning of patients before RPLND has been routinely used to visualize residual retroperitoneal tumor.4 Although CT may demonstrate patency or displacement of the aorta or vena cava, visualization of smaller vessels is difficult. In a preliminary report, magnetic resonance imaging (MRI) was used to evaluate the displacement, compression, and patency of the inferior vena cava in 5 patients with retroperitoneal metastatic NSGCT.5 Recently, ad0090-4295/00/$20.00 PII S0090-4295(99)00428-8
TABLE I. Patient characteristics Patients (n) Age (yr) Median Range Courses of chemotherapy (n) Median Range Pathologic stage (n) N0 N1 N2 N3 Histologic findings (n) No residual tumor Mature teratoma Mature teratoma and NSGCT
18 29 19–49 8 4–18 9 2 3 4 9 4 5
KEY: NSGCT ⫽ nonseminomatous germ cell testicular cancer.
vances in MRI techniques have provided superior visualization of accessory renal and lumbar vessels. In magnetic resonance angiography (MRA), vascular contrast is maximized by enhancing the signal from moving spins in flowing blood and suppressing the signal from stationary spins in surrounding tissue. Various MRA techniques have been developed, including time-of-flight and phase-contrast MRA, which uses the phase shifts produced by velocity differences in flowing blood relative to stationary surrounding tissues to produce contrast. Since June 1997, we have used the technique of bolus-contrast MRA, in which intravenous gadolinium diethylenetriamine penta-acetic acid (GdDTPA) is injected during image acquisition. Prince et al.6 demonstrated that this technique provides excellent visualization of the renal, celiac, and superior mesenteric vessels and has several inherent advantages over CT angiography. In the present, retrospective study, we reviewed our experience with the bolus-contrast MRA technique in 8 consecutive patients who underwent postchemotherapy RPLND and with previous patients who underwent conventional MRI or phase-contrast MRA to evaluate the accuracy of these techniques for the imaging of retroperitoneal vessels and their relation to residual tumor masses. MATERIAL AND METHODS Eighteen patients who received systemic chemotherapy for metastatic NSGCT underwent MRI of the retroperitoneal region before postchemotherapy RPLND. Patient characteristics, including final pathologic stage and histologic findings, are given in Table I. The patients in this series had been heavily pretreated, having received a median of eight courses of chemotherapy before RPLND. In 10 consecutive patients undergoing postchemotherapy RPLND, MRA of the aorta and renal vessels was performed, in UROLOGY 55 (2), 2000
addition to routine spin-echo T1-weighted or fast spin-echo T2-weighted and breath-hold fast gradient-echo sequences. Routine MR images were evaluated independently from MRA images. In 8 of these patients, three-dimensional (3D) boluscontrast MRA was performed, and in two studies, 3D phasecontrast MRA was performed. All MRI was performed using GE Signa Horizon EchoSpeed or HiSpeed scanners (version 5.6, GE Medical Systems, Milwaukee, Wis). For the coronal, 3D, contrast-enhanced MRA studies, an enhanced fast spoiled gradient-echo sequence was used without flow compensation. The echo times were 1.4 ms, repetition times were 8 to 9 ms, and 26 to 32 2-mm slices were acquired and interpolated to 44 to 56 50%-overlapping slices (slice ZIP (zero interpolation) ⫽ 2). The images were acquired with a flip angle of 50°, 32-kHz sampling bandwidth, 256 ⫻ 128 matrix, 1 NEX (number of excitations), and field of views of 34 ⫻ 31 cm. The breath-hold duration in all cases was less than 27 seconds and was typically 21 to 25 seconds. A hand-injected bolus of 35 to 40 mL of GdDTPA (Magnevist, Berlex Laboratories, Wayne, NJ) was started approximately 15 seconds before scan initiation. Three sets of postinjection scans were acquired with minimal interscan delay. The 3D phase-contrast MRA images were acquired in the coronal plane, with 32 3-mm slices, flow-encoding in all directions, first-order flow compensation, a VENC (velocity encoded) of 60 cm/s, flip angle of 20°, echo time of 8 ms, repetition time of 40 ms, 256 ⫻ 128 matrix, 1 NEX, and a scan time of approximately 7 minutes. In each patient, the abdominal aorta and common iliac, renal, celiac, and superior and inferior mesenteric arteries were evaluated on preoperative imaging for their relationship to the adjacent tumor. Images were also evaluated for vascular anomalies, such as a multiplicity of arteries, and the location and number of lumbar arteries. Results of the MRI and MRA studies were compared with intraoperative and pathologic findings. The most recent CT scan during or after systemic chemotherapy was compared with the preoperative MRI or MRA study for its ability to demonstrate retroperitoneal vasculature. In general, CT scans were performed with both gastrointestinal and intravenous contrast enhancement using 10-mm axial sections. One patient underwent CT scanning with oral and rectal contrast only because of an elevated creatinine level. Anatomic findings noted at RPLND were considered the reference standard of retroperitoneal anatomy. The median interval from acquisition of the MRI or MRA studies to RPLND was 6 days (range 2 to 26).
RESULTS The bolus-contrast MRA procedure was well tolerated, with no allergic or other adverse reactions. Anatomic findings on preoperative imaging and at RPLND are given in Table II. Multiple renal arteries were detected in 5 patients by bolus-contrast MRA, and each of these findings was confirmed at RPLND. Two additional patients were found at RPLND to have multiple renal arteries that were not detected on preoperative imaging. Neither of these patients had undergone bolus-contrast MRA. One of these patients had undergone phase-contrast MRA, and the other had undergone conventional MRI. None of the accessory renal vessels were seen on routine preoperative CT scanning. In 1 patient, a second right renal vein was not identified on bolus-contrast MRA. A left retroaortic renal vein was correctly identified by conventional MRI in 1 patient and by bolus-contrast MRA in 263
TABLE II. Preoperative imaging results Study Conventional MRI Phase-contrast MRA Bolus-contrast MRA
Anatomic Findings Left retroaortic renal vein No vascular anomalies Duplex IVC No vascular anomalies Multiple renal arteries Left retroaortic renal vein Iliac vein thrombosis Lumbar artery adjacent to renal artery No vascular anomalies
Patients Identified (n) on CT 1 6 1 1 5* 1 1 2* 0
No — Yes — No No No No —
KEY: CT ⫽ computed tomography; MRI ⫽ magnetic resonance imaging; MRA ⫽ magnetic resonance angiography; IVC ⫽ inferior vena cava. * Includes 1 patient with both anatomic findings.
another patient. A duplex inferior vena cava in 1 patient was accurately identified on MRI, and in 1 patient, a common iliac vein thrombosis was correctly identified by bolus-contrast MRI. In 2 patients, a lumbar artery was identified by MRA immediately adjacent to a renal artery. In all patients, MRI and MRA provided information on patency, displacement, and adherence to retroperitoneal vessels by residual tumor that was essential for optimal intraoperative decision making. For 9 patients, MRI or MRA provided novel information on the retroperitoneal vasculature that was not provided by preoperative CT. All patients underwent full bilateral RPLND with complete removal of the para-aortic, interaortocaval, and pericaval lymph nodes, except for 1 patient, who underwent a modified template dissection. Figure 1 demonstrates the preoperative bolus-contrast MRA (Fig. 1A, upper panel) and intraoperative photograph after removal of the tumor (Fig. 1B, upper panel) in a 32-year-old man with extensive residual para-aortic adenopathy after 10 courses of chemotherapy. Two right renal arteries in close proximity that were identified by preoperative bolus-contrast MRA were isolated and preserved by meticulous dissection in the narrow angle between the inferior vena cava, left renal vein, and aorta. In another case, preoperative bolus-contrast MRA in a 24-year-old man with residual para-aortic adenopathy at the level of the renal vessels after four courses of chemotherapy demonstrated a lumbar artery immediately adjacent to a right upper pole renal artery (Fig. 1A, lower panel). At RPLND, the smaller upper pole artery was also preserved (Fig. 1B, lower panel). Dissection in these patients was difficult because of intense fibrosis and desmoplasia. Preoperative knowledge of the existence of the supernumerary renal arteries and their relation to the lumbar arteries greatly facilitated their preservation. 264
COMMENT Advances in platinum-based chemotherapy for metastatic testicular cancer and the technique of RPLND have resulted in an overall long-term survival rate of approximately 85% to 90% for patients with Stage II disease.1,7 Although controversy exists regarding the optimal treatment of patients who present with Stage II NSGCT (primary chemotherapy with RPLND reserved for patients with residual mass versus RPLND with or without adjuvant chemotherapy8), there is general agreement that patients with bulky retroperitoneal disease should be treated with primary chemotherapy followed by full bilateral RPLND for residual tumor.9 –13 The split and roll technique with division of the lumbar arteries and veins is used to achieve circumferential removal of the tumor and lymphatic tissue surrounding the great vessels down to the anterior longitudinal spinal ligament. Complete resection of bulky disease may rarely require en bloc nephrectomy if the residual mass is inseparable from the kidney or hilar structures. Because of the fibrotic reaction after chemotherapy, injury to the aorta, major retroperitoneal arteries, or inferior vena cava resulting from subadventitial dissection is possible, occasionally requiring resection or repair of the inferior vena cava2,14,15 or aortic or aortoiliac graft placement.2 The incidence of recognized injury to the renal and splanchnic arteries during RPLND is greater in postchemotherapy patients.3 Furthermore, the incidence of anomalies of the renal vessels is not insignificant. In a review of the anatomic findings in the lumbar and renal arteries in 104 consecutive patients undergoing RPLND at Indiana University, Baniel et al.16 documented accessory renal arteries in 24 patients (23%), including 6 patients with bilateral supernumerary renal arteries. A retroaortic left renal vein was also noted in 3 patients (3%). Because our study was not performed in a randomized prospecUROLOGY 55 (2), 2000
FIGURE 1. (Upper Panel, A) Bolus-contrast MRA of a 32-year-old man with residual para-aortic adenopathy after 10 courses of chemotherapy demonstrating accessory right lower pole renal artery in close proximity to the main renal artery (arrows) that was (B) identified and preserved at RPLND (arrow). (Lower Panel, A) MRA of a 24-year-old man with a lumbar artery adjacent to an accessory right upper pole renal artery reveals the accessory renal artery and the adjacent lumbar artery (arrows). (B) Intraoperative photograph revealing the preserved accessory renal artery (arrow).
tive fashion, we cannot accurately assess whether the use of bolus-contrast MRA resulted in a decrease in complications. The advantage of MRA in the visualization of stenotic lesions of the renal arteries has been preUROLOGY 55 (2), 2000
viously documented.17,18 For patients undergoing RPLND after chemotherapy, accurate radiographic imaging, including visualization of the retroperitoneal vasculature, is essential for restaging and preoperative planning. The 3D bolus-enhanced MRA 265
technique offers unique information on the retroperitoneal vasculature and appears to be superior to phase-contrast MRA for the detection of accessory renal arteries.19 Helical CT reconstruction techniques may fail to identify these vessels.20 Although both helical CT and MRA are less invasive than conventional angiography, MRA holds several advantages over helical CT or 3D CT angiography.21,22 Gadolinium is a safer contrast medium than iodine-based contrast, resulting in fewer idiosyncratic reactions and low nephrotoxicity. This is particularly advantageous in a patient population heavily treated with platinum-based chemotherapy for whom maximal preservation of renal function is paramount. Furthermore, MRA may provide higher resolution, avoids the risk of ionizing radiation, and allows imaging in any plane for optimal coverage of the retroperitoneal vasculature. Also, calcified tissue such as arterial plaques or bone can be confused with contrast-filled vessels on CT angiography but are easily distinguished on MRA. Previously, the necessity for a longer MRA imaging time resulted in degradation of the image because of respiratory motion. The development of the rapid sequence bolus-contrast technique has allowed image acquisition during a single breath hold, thus eliminating motion artifacts.6 In our series, images were typically acquired within 21 to 25 seconds and all breath-hold durations were less than 27 seconds. Although the present study was not a direct comparison of CT and MRI, unique information on the retroperitoneal vasculature was provided by MRI or MRA in 9 of 18 patients. CONCLUSIONS MRI and MRA provide unique information about the relationship of residual retroperitoneal tumor to adjacent vasculature. Bolus-enhanced 3D MRA allows accurate visualization of the location and number of renal and lumbar vessels in patients undergoing postchemotherapy RPLND. REFERENCES 1. Donohue JP, Thornhill JA, Foster RS, et al: The role of retroperitoneal lymphadenectomy in clinical stage B testis cancer. The Indiana University Experience (1965 to 1989). J Urol 153: 85– 89, 1995. 2. Donohue JP, Thornhill JA, Foster RS, et al: Vascular considerations in postchemotherapy retroperitoneal lymph node dissection: part I, vena cava: part II, aorta. World J Urol 12: 182–186, 1994. 3. Baniel J, Foster RS, Rowland RG, et al: Complications of post-chemotherapy retroperitoneal lymph node dissection. J Urol 153: 976 –980, 1995. 4. Fernandez EB, Moul JW, Foley JP, et al: Retroperitoneal imaging with third and fourth generation computed axial tomography in clinical stage I nonseminomatous germ cell tumors. Urology 44: 548 –552, 1994.
266
5. Ng CS, Husband JES, Padhani AR, et al: Evaluation by magnetic resonance imaging of the inferior vena cava in patients with nonseminomatous germ cell tumors of the testis metastatic to the retroperitoneum. Br J Urol 79: 942–951, 1997. 6. Prince MR, Narasimham DL, Stanley JC, et al: Breathhold gadolinium-enhanced MR angiography of the abdominal aorta and its major branches. Radiology 197: 785–792, 1995. 7. Einhorn LH: Treatment of testicular cancer: a new and improved model. J Clin Oncol 8: 1777–1781, 1990. 8. Motzer RJ, and Bosl GJ: Role of adjuvant chemotherapy in patients with stage II nonseminomatous germ-cell tumors. Urol Clin North Am 20: 111–116, 1993. 9. Harding MJ, Brown IL, MacPherson SG, et al: Excision of residual masses after platinum based chemotherapy for nonseminomatous germ cell tumors. Eur J Cancer Clin Oncol 23: 1689 –1694, 1990. 10. Toner GC, Panicek DM, Heelan RT, et al: Adjunctive surgery after chemotherapy for nonseminomatous germ cell tumors: recommendations for patient selection. J Clin Oncol 8: 1683–1694, 1990. 11. Bajorin DF, Herr H, Motzer RJ, et al: Current perspectives on the role of adjunctive surgery in combined modality treatment for patients with germ cell tumors. Semin Oncol 19: 148 –158, 1990. 12. Hendry WF, A’Hern RP, Hetherington JW, et al: Paraaortic lymphadenectomy after chemotherapy for metastatic non-seminomatous germ cell tumours: prognostic value and therapeutic benefit. Br J Cancer 71: 208 –213, 1993. 13. Fox EP, Weathers TD, Williams SD, et al: Outcome analysis for patients with persistent nonteratomatous germ cell tumor in postchemotherapy retroperitoneal lymph node dissection. J Clin Oncol 11: 1294 –1299, 1993. 14. Donohue JP, Thornhill JA, Foster RD, et al: Resection of the inferior vena cava during RPLND for metastatic germ cell cancer: indications and results. J Urol 146: 346 –349, 1991. 15. Ahlering TE, and Skinner DG: Vena caval resection in bulky metastatic germ cell tumors. J Urol 142: 1497–1499, 1989. 16. Baniel J, Foster RS, and Donohue JP: Surgical anatomy of the lumbar vessels: implications for retroperitoneal surgery. J Urol 153: 1422–1425, 1995. 17. Kent KC, Edelman RR, Kim D, et al: Magnetic resonance imaging: a reliable test for the evaluation of proximal atherosclerotic renal arterial stenosis. J Vasc Surg 13: 311– 318, 1991. 18. Meyers SP, Talagala SL, Totterman S, et al: Evaluation of the renal arteries in kidney donors: value of three-dimensional phase-contrast MR angiography with maximum-intensity-projection or surface rendering. AJR Am J Roentgenol 164: 117–121, 1995. 19. De Cobelli F, Vanzulli A, Sironi S, et al: Renal artery stenosis evaluation with breath-hold, three-dimensional, dynamic gadolinium-enhanced versus three-dimensional, phase contrast MR angiography. Radiology 205: 689 – 695, 1997. 20. Dachman AH, Newmark GM, Mitchell MT, et al: Helical CT examination of potential kidney donors. AJR Am J Roentgenol 171: 193–200, 1998. 21. Prince MR: Gadolinium-enhanced MR aortography. Radiology 191: 155–164, 1994. 22. Bakker J, Beek FJA, Beutler JJ, et al: Renal artery stenosis and accessory renal arteries: accuracy of detection and visualization with gadolinium-enhanced breath-hold MR angiography. Radiology 207: 497–504, 1998.
UROLOGY 55 (2), 2000
MAGNETIC RESONANCE IMAGING AND MAGNETIC RESONANCE ANGIOGRAPHY BEFORE POSTCHEMOTHERAPY RETROPERITONEAL LYMPH NODE DISSECTION DAVID A. CORRAL, DATLA G. VARMA, EDWARD F. JACKSON, ROBERT J. AMATO, S. MACHELE DONAT, AND LOUIS L. PISTERS
ABSTRACT Objectives. Retroperitoneal lymph node dissection (RPLND) after primary chemotherapy is an accepted therapeutic approach for metastatic nonseminomatous germ cell testicular cancer. Because of the intense desmoplastic reaction and adherence to venous and arterial walls, accurate imaging of the retroperitoneal vasculature and its relation to residual tumor is essential. We report our experience with magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), including the recently developed technique of bolus-contrast MRA, in patients undergoing postchemotherapy RPLND. Methods. Eighteen patients underwent MRI of the retroperitoneal region before RPLND. In addition to routine sequences, MRA was performed in 10 patients, including 8 with a three-dimensional technique using bolus intravenous MR contrast. Results were compared with intraoperative and pathologic findings. Results. MRI and MRA provided detailed information on retroperitoneal vasculature and its relation to tumor, including multiple renal vessels (n ⫽ 5), duplex inferior vena cava (n ⫽ 1), left retroaortic renal vein (n ⫽ 2), and common iliac vein thrombus (n ⫽ 1). In all cases, bolus-contrast MRA provided unique information on the location and number of renal and lumbar arteries, in addition to information on the aorta and the mesenteric and iliac vessels. The origin and number of renal arteries were accurately identified in all patients by bolus-contrast MRA; 2 patients had supernumerary renal arteries discovered at RPLND that had not been identified on non-bolus-contrast MRI. Conclusions. Bolus-contrast three-dimensional MRA provides unique information on renal and lumbar vessels. The potential benefit of avoiding vascular injury during dissection should be prospectively evaluated. UROLOGY 55: 262–266, 2000. © 2000, Elsevier Science Inc.
D
uring the platinum-based chemotherapy era, primary chemotherapy followed by retroperitoneal lymph node dissection (RPLND) became a widely accepted therapeutic strategy for metastatic nonseminomatous germ cell testicular cancer (NSGCT).1 After chemotherapy, residual retroperitoneal tumors commonly exhibit an intense fi-
From the Departments of Urology, Radiology , and Genitourinary Oncology, University of Texas M. D. Anderson Cancer Center, Houston, Texas; and Department of Urology, Memorial SloanKettering Cancer Center, New York, New York Reprint requests: David A. Corral, M.D., Department of Urologic Oncology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263 Submitted: July 9, 1999, accepted (with revisions): September 8, 1999
262
© 2000, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
brotic, desmoplastic reaction intimately involving and often encasing the walls of major vessels and resulting in treacherous conditions for dissection.2,3 Accurate preoperative imaging of residual tumor and its relation to retroperitoneal vasculature can provide critical information for intraoperative decision making during RPLND. Computed tomography (CT) scanning of patients before RPLND has been routinely used to visualize residual retroperitoneal tumor.4 Although CT may demonstrate patency or displacement of the aorta or vena cava, visualization of smaller vessels is difficult. In a preliminary report, magnetic resonance imaging (MRI) was used to evaluate the displacement, compression, and patency of the inferior vena cava in 5 patients with retroperitoneal metastatic NSGCT.5 Recently, ad0090-4295/00/$20.00 PII S0090-4295(99)00428-8
TABLE I. Patient characteristics Patients (n) Age (yr) Median Range Courses of chemotherapy (n) Median Range Pathologic stage (n) N0 N1 N2 N3 Histologic findings (n) No residual tumor Mature teratoma Mature teratoma and NSGCT
18 29 19–49 8 4–18 9 2 3 4 9 4 5
KEY: NSGCT ⫽ nonseminomatous germ cell testicular cancer.
vances in MRI techniques have provided superior visualization of accessory renal and lumbar vessels. In magnetic resonance angiography (MRA), vascular contrast is maximized by enhancing the signal from moving spins in flowing blood and suppressing the signal from stationary spins in surrounding tissue. Various MRA techniques have been developed, including time-of-flight and phase-contrast MRA, which uses the phase shifts produced by velocity differences in flowing blood relative to stationary surrounding tissues to produce contrast. Since June 1997, we have used the technique of bolus-contrast MRA, in which intravenous gadolinium diethylenetriamine penta-acetic acid (GdDTPA) is injected during image acquisition. Prince et al.6 demonstrated that this technique provides excellent visualization of the renal, celiac, and superior mesenteric vessels and has several inherent advantages over CT angiography. In the present, retrospective study, we reviewed our experience with the bolus-contrast MRA technique in 8 consecutive patients who underwent postchemotherapy RPLND and with previous patients who underwent conventional MRI or phase-contrast MRA to evaluate the accuracy of these techniques for the imaging of retroperitoneal vessels and their relation to residual tumor masses. MATERIAL AND METHODS Eighteen patients who received systemic chemotherapy for metastatic NSGCT underwent MRI of the retroperitoneal region before postchemotherapy RPLND. Patient characteristics, including final pathologic stage and histologic findings, are given in Table I. The patients in this series had been heavily pretreated, having received a median of eight courses of chemotherapy before RPLND. In 10 consecutive patients undergoing postchemotherapy RPLND, MRA of the aorta and renal vessels was performed, in UROLOGY 55 (2), 2000
addition to routine spin-echo T1-weighted or fast spin-echo T2-weighted and breath-hold fast gradient-echo sequences. Routine MR images were evaluated independently from MRA images. In 8 of these patients, three-dimensional (3D) boluscontrast MRA was performed, and in two studies, 3D phasecontrast MRA was performed. All MRI was performed using GE Signa Horizon EchoSpeed or HiSpeed scanners (version 5.6, GE Medical Systems, Milwaukee, Wis). For the coronal, 3D, contrast-enhanced MRA studies, an enhanced fast spoiled gradient-echo sequence was used without flow compensation. The echo times were 1.4 ms, repetition times were 8 to 9 ms, and 26 to 32 2-mm slices were acquired and interpolated to 44 to 56 50%-overlapping slices (slice ZIP (zero interpolation) ⫽ 2). The images were acquired with a flip angle of 50°, 32-kHz sampling bandwidth, 256 ⫻ 128 matrix, 1 NEX (number of excitations), and field of views of 34 ⫻ 31 cm. The breath-hold duration in all cases was less than 27 seconds and was typically 21 to 25 seconds. A hand-injected bolus of 35 to 40 mL of GdDTPA (Magnevist, Berlex Laboratories, Wayne, NJ) was started approximately 15 seconds before scan initiation. Three sets of postinjection scans were acquired with minimal interscan delay. The 3D phase-contrast MRA images were acquired in the coronal plane, with 32 3-mm slices, flow-encoding in all directions, first-order flow compensation, a VENC (velocity encoded) of 60 cm/s, flip angle of 20°, echo time of 8 ms, repetition time of 40 ms, 256 ⫻ 128 matrix, 1 NEX, and a scan time of approximately 7 minutes. In each patient, the abdominal aorta and common iliac, renal, celiac, and superior and inferior mesenteric arteries were evaluated on preoperative imaging for their relationship to the adjacent tumor. Images were also evaluated for vascular anomalies, such as a multiplicity of arteries, and the location and number of lumbar arteries. Results of the MRI and MRA studies were compared with intraoperative and pathologic findings. The most recent CT scan during or after systemic chemotherapy was compared with the preoperative MRI or MRA study for its ability to demonstrate retroperitoneal vasculature. In general, CT scans were performed with both gastrointestinal and intravenous contrast enhancement using 10-mm axial sections. One patient underwent CT scanning with oral and rectal contrast only because of an elevated creatinine level. Anatomic findings noted at RPLND were considered the reference standard of retroperitoneal anatomy. The median interval from acquisition of the MRI or MRA studies to RPLND was 6 days (range 2 to 26).
RESULTS The bolus-contrast MRA procedure was well tolerated, with no allergic or other adverse reactions. Anatomic findings on preoperative imaging and at RPLND are given in Table II. Multiple renal arteries were detected in 5 patients by bolus-contrast MRA, and each of these findings was confirmed at RPLND. Two additional patients were found at RPLND to have multiple renal arteries that were not detected on preoperative imaging. Neither of these patients had undergone bolus-contrast MRA. One of these patients had undergone phase-contrast MRA, and the other had undergone conventional MRI. None of the accessory renal vessels were seen on routine preoperative CT scanning. In 1 patient, a second right renal vein was not identified on bolus-contrast MRA. A left retroaortic renal vein was correctly identified by conventional MRI in 1 patient and by bolus-contrast MRA in 263
TABLE II. Preoperative imaging results Study Conventional MRI Phase-contrast MRA Bolus-contrast MRA
Anatomic Findings Left retroaortic renal vein No vascular anomalies Duplex IVC No vascular anomalies Multiple renal arteries Left retroaortic renal vein Iliac vein thrombosis Lumbar artery adjacent to renal artery No vascular anomalies
Patients Identified (n) on CT 1 6 1 1 5* 1 1 2* 0
No — Yes — No No No No —
KEY: CT ⫽ computed tomography; MRI ⫽ magnetic resonance imaging; MRA ⫽ magnetic resonance angiography; IVC ⫽ inferior vena cava. * Includes 1 patient with both anatomic findings.
another patient. A duplex inferior vena cava in 1 patient was accurately identified on MRI, and in 1 patient, a common iliac vein thrombosis was correctly identified by bolus-contrast MRI. In 2 patients, a lumbar artery was identified by MRA immediately adjacent to a renal artery. In all patients, MRI and MRA provided information on patency, displacement, and adherence to retroperitoneal vessels by residual tumor that was essential for optimal intraoperative decision making. For 9 patients, MRI or MRA provided novel information on the retroperitoneal vasculature that was not provided by preoperative CT. All patients underwent full bilateral RPLND with complete removal of the para-aortic, interaortocaval, and pericaval lymph nodes, except for 1 patient, who underwent a modified template dissection. Figure 1 demonstrates the preoperative bolus-contrast MRA (Fig. 1A, upper panel) and intraoperative photograph after removal of the tumor (Fig. 1B, upper panel) in a 32-year-old man with extensive residual para-aortic adenopathy after 10 courses of chemotherapy. Two right renal arteries in close proximity that were identified by preoperative bolus-contrast MRA were isolated and preserved by meticulous dissection in the narrow angle between the inferior vena cava, left renal vein, and aorta. In another case, preoperative bolus-contrast MRA in a 24-year-old man with residual para-aortic adenopathy at the level of the renal vessels after four courses of chemotherapy demonstrated a lumbar artery immediately adjacent to a right upper pole renal artery (Fig. 1A, lower panel). At RPLND, the smaller upper pole artery was also preserved (Fig. 1B, lower panel). Dissection in these patients was difficult because of intense fibrosis and desmoplasia. Preoperative knowledge of the existence of the supernumerary renal arteries and their relation to the lumbar arteries greatly facilitated their preservation. 264
COMMENT Advances in platinum-based chemotherapy for metastatic testicular cancer and the technique of RPLND have resulted in an overall long-term survival rate of approximately 85% to 90% for patients with Stage II disease.1,7 Although controversy exists regarding the optimal treatment of patients who present with Stage II NSGCT (primary chemotherapy with RPLND reserved for patients with residual mass versus RPLND with or without adjuvant chemotherapy8), there is general agreement that patients with bulky retroperitoneal disease should be treated with primary chemotherapy followed by full bilateral RPLND for residual tumor.9 –13 The split and roll technique with division of the lumbar arteries and veins is used to achieve circumferential removal of the tumor and lymphatic tissue surrounding the great vessels down to the anterior longitudinal spinal ligament. Complete resection of bulky disease may rarely require en bloc nephrectomy if the residual mass is inseparable from the kidney or hilar structures. Because of the fibrotic reaction after chemotherapy, injury to the aorta, major retroperitoneal arteries, or inferior vena cava resulting from subadventitial dissection is possible, occasionally requiring resection or repair of the inferior vena cava2,14,15 or aortic or aortoiliac graft placement.2 The incidence of recognized injury to the renal and splanchnic arteries during RPLND is greater in postchemotherapy patients.3 Furthermore, the incidence of anomalies of the renal vessels is not insignificant. In a review of the anatomic findings in the lumbar and renal arteries in 104 consecutive patients undergoing RPLND at Indiana University, Baniel et al.16 documented accessory renal arteries in 24 patients (23%), including 6 patients with bilateral supernumerary renal arteries. A retroaortic left renal vein was also noted in 3 patients (3%). Because our study was not performed in a randomized prospecUROLOGY 55 (2), 2000
FIGURE 1. (Upper Panel, A) Bolus-contrast MRA of a 32-year-old man with residual para-aortic adenopathy after 10 courses of chemotherapy demonstrating accessory right lower pole renal artery in close proximity to the main renal artery (arrows) that was (B) identified and preserved at RPLND (arrow). (Lower Panel, A) MRA of a 24-year-old man with a lumbar artery adjacent to an accessory right upper pole renal artery reveals the accessory renal artery and the adjacent lumbar artery (arrows). (B) Intraoperative photograph revealing the preserved accessory renal artery (arrow).
tive fashion, we cannot accurately assess whether the use of bolus-contrast MRA resulted in a decrease in complications. The advantage of MRA in the visualization of stenotic lesions of the renal arteries has been preUROLOGY 55 (2), 2000
viously documented.17,18 For patients undergoing RPLND after chemotherapy, accurate radiographic imaging, including visualization of the retroperitoneal vasculature, is essential for restaging and preoperative planning. The 3D bolus-enhanced MRA 265
technique offers unique information on the retroperitoneal vasculature and appears to be superior to phase-contrast MRA for the detection of accessory renal arteries.19 Helical CT reconstruction techniques may fail to identify these vessels.20 Although both helical CT and MRA are less invasive than conventional angiography, MRA holds several advantages over helical CT or 3D CT angiography.21,22 Gadolinium is a safer contrast medium than iodine-based contrast, resulting in fewer idiosyncratic reactions and low nephrotoxicity. This is particularly advantageous in a patient population heavily treated with platinum-based chemotherapy for whom maximal preservation of renal function is paramount. Furthermore, MRA may provide higher resolution, avoids the risk of ionizing radiation, and allows imaging in any plane for optimal coverage of the retroperitoneal vasculature. Also, calcified tissue such as arterial plaques or bone can be confused with contrast-filled vessels on CT angiography but are easily distinguished on MRA. Previously, the necessity for a longer MRA imaging time resulted in degradation of the image because of respiratory motion. The development of the rapid sequence bolus-contrast technique has allowed image acquisition during a single breath hold, thus eliminating motion artifacts.6 In our series, images were typically acquired within 21 to 25 seconds and all breath-hold durations were less than 27 seconds. Although the present study was not a direct comparison of CT and MRI, unique information on the retroperitoneal vasculature was provided by MRI or MRA in 9 of 18 patients. CONCLUSIONS MRI and MRA provide unique information about the relationship of residual retroperitoneal tumor to adjacent vasculature. Bolus-enhanced 3D MRA allows accurate visualization of the location and number of renal and lumbar vessels in patients undergoing postchemotherapy RPLND. REFERENCES 1. Donohue JP, Thornhill JA, Foster RS, et al: The role of retroperitoneal lymphadenectomy in clinical stage B testis cancer. The Indiana University Experience (1965 to 1989). J Urol 153: 85– 89, 1995. 2. Donohue JP, Thornhill JA, Foster RS, et al: Vascular considerations in postchemotherapy retroperitoneal lymph node dissection: part I, vena cava: part II, aorta. World J Urol 12: 182–186, 1994. 3. Baniel J, Foster RS, Rowland RG, et al: Complications of post-chemotherapy retroperitoneal lymph node dissection. J Urol 153: 976 –980, 1995. 4. Fernandez EB, Moul JW, Foley JP, et al: Retroperitoneal imaging with third and fourth generation computed axial tomography in clinical stage I nonseminomatous germ cell tumors. Urology 44: 548 –552, 1994.
266
5. Ng CS, Husband JES, Padhani AR, et al: Evaluation by magnetic resonance imaging of the inferior vena cava in patients with nonseminomatous germ cell tumors of the testis metastatic to the retroperitoneum. Br J Urol 79: 942–951, 1997. 6. Prince MR, Narasimham DL, Stanley JC, et al: Breathhold gadolinium-enhanced MR angiography of the abdominal aorta and its major branches. Radiology 197: 785–792, 1995. 7. Einhorn LH: Treatment of testicular cancer: a new and improved model. J Clin Oncol 8: 1777–1781, 1990. 8. Motzer RJ, and Bosl GJ: Role of adjuvant chemotherapy in patients with stage II nonseminomatous germ-cell tumors. Urol Clin North Am 20: 111–116, 1993. 9. Harding MJ, Brown IL, MacPherson SG, et al: Excision of residual masses after platinum based chemotherapy for nonseminomatous germ cell tumors. Eur J Cancer Clin Oncol 23: 1689 –1694, 1990. 10. Toner GC, Panicek DM, Heelan RT, et al: Adjunctive surgery after chemotherapy for nonseminomatous germ cell tumors: recommendations for patient selection. J Clin Oncol 8: 1683–1694, 1990. 11. Bajorin DF, Herr H, Motzer RJ, et al: Current perspectives on the role of adjunctive surgery in combined modality treatment for patients with germ cell tumors. Semin Oncol 19: 148 –158, 1990. 12. Hendry WF, A’Hern RP, Hetherington JW, et al: Paraaortic lymphadenectomy after chemotherapy for metastatic non-seminomatous germ cell tumours: prognostic value and therapeutic benefit. Br J Cancer 71: 208 –213, 1993. 13. Fox EP, Weathers TD, Williams SD, et al: Outcome analysis for patients with persistent nonteratomatous germ cell tumor in postchemotherapy retroperitoneal lymph node dissection. J Clin Oncol 11: 1294 –1299, 1993. 14. Donohue JP, Thornhill JA, Foster RD, et al: Resection of the inferior vena cava during RPLND for metastatic germ cell cancer: indications and results. J Urol 146: 346 –349, 1991. 15. Ahlering TE, and Skinner DG: Vena caval resection in bulky metastatic germ cell tumors. J Urol 142: 1497–1499, 1989. 16. Baniel J, Foster RS, and Donohue JP: Surgical anatomy of the lumbar vessels: implications for retroperitoneal surgery. J Urol 153: 1422–1425, 1995. 17. Kent KC, Edelman RR, Kim D, et al: Magnetic resonance imaging: a reliable test for the evaluation of proximal atherosclerotic renal arterial stenosis. J Vasc Surg 13: 311– 318, 1991. 18. Meyers SP, Talagala SL, Totterman S, et al: Evaluation of the renal arteries in kidney donors: value of three-dimensional phase-contrast MR angiography with maximum-intensity-projection or surface rendering. AJR Am J Roentgenol 164: 117–121, 1995. 19. De Cobelli F, Vanzulli A, Sironi S, et al: Renal artery stenosis evaluation with breath-hold, three-dimensional, dynamic gadolinium-enhanced versus three-dimensional, phase contrast MR angiography. Radiology 205: 689 – 695, 1997. 20. Dachman AH, Newmark GM, Mitchell MT, et al: Helical CT examination of potential kidney donors. AJR Am J Roentgenol 171: 193–200, 1998. 21. Prince MR: Gadolinium-enhanced MR aortography. Radiology 191: 155–164, 1994. 22. Bakker J, Beek FJA, Beutler JJ, et al: Renal artery stenosis and accessory renal arteries: accuracy of detection and visualization with gadolinium-enhanced breath-hold MR angiography. Radiology 207: 497–504, 1998.
UROLOGY 55 (2), 2000