- Email: [email protected]
Fluoroscopic Landmarks for Optimal Visualization of the Proximal Renal Arteries
Vascular Diagnosis
Fluoroscopic Landmarks for Optimal Visualization of the Proximal Renal Arteries1 -
Patricia A. Kim, M D Neil M. Khilnani, M D David W. Trost, M D Thomas A. Sos, M D Lisa Lee, M D
Index terms: Renal angiography Renal arteries, anatomy * Renal arteries, stents and prostheses
JVIR 1999; 10:37-39 Abbreviations: LISP = L-1 spinous process, LAO = left anterior oblique, SD = standard deviation, SMA = superior mesenteric artery
PURPOSE: To accurately determine the in vivo orientation of the origin of the renal arteries from the aorta relative to a fluoroscopic bony landmark for optimal diagnostic arteriography and renal artery stent placement. MATERIALS AND METHODS: One hundred sixty abdominal computed tomography (CT) scans of patients in eight age groups (20-90 years) were reviewed to determine the angle of the origins of the renal arteries from the aorta relative to the long axis of the L-1 spinous process (LISP). RESULTS: The right renal artery arises ventrally at an angle of 30" (standard deviation [SD] = 15")from a plane orthogonal to the long axis of the LISP. The left renal artery arises dorsally at an angle of 7" (SD = 13")relative to the same plane. CONCLUSIONS: The optimal initial angle for angiographic evaluation of the origin of the renal artery and for renal artery stent placement is 30" left anterior oblique (LAO) relative to the LlSP for the right renal artery and 7" LAO for the left renal artery. Unfortunately, there is variability in the angle of the renal artery origins from the aorta which cannot be controlled for using this technique. In some patients, additional views will be necessary to optimally depict the origins of the renal arteries.
From the Department of Radiology (P.A.K., D.W.T., N.M.K., T.A.S.), The New York Hospital-Cornell Medical Cen. ter, 525 E. 68th St., New York, NY, 10021, and the Department of Radiology (L.L.), North Shore Hospital, New York, NY. Received June 27, 1997; revision requested August 29; revision received June 15, 1998; accepted June 16. Address correspondence to N.M.K. O SCVIR, 1999
TO accurately demonstrate the proximal renal artery with x-ray angiography, the plane of the xray beam must be orthogonal to the plane of the renal artery origin from the aorta. This concept is important when investigating patients with suspected atheromatous renal artery stenoses, as the lesions are most frequently found a t the renal artery origin. Moreover, as percutaneous treatment of ostial renal artery stenoses requires stent placement to eliminate elastic recoil (l,2), accurate depiction of the renal artery ostium as it originates from the aorta is crucial to adequately place an endovascular stent. We recently undertook a review
of the angle of the origin of the renal arteries from the aorta as measured by axial computed tomography (CT) images. The goal of our study was to determine whether correcting for subtle variations in the rotation of the patient on the examination table by using the long axis of the spinous process of the L-1 vertebral body, a fluoroscopically visible landmark, could lead to the development of a reliable angiographic projection for the depiction of the renal artery ostia. We also evaluated the location of the superior mesenteric artery (SMA) orifice in relation to the same fluoroscopic landmark to allow comparison of our data with previous studies.
38
a
Fluoroscopic Landmarks of the Proximal Renal Arteries
January 1999 JVIR
I MATERIALS AND METHODS The CT scans of the upper abdomen of 160 patients, obtained between October 1, 1995, and February 29, 1996, were retrospectively reviewed. The CT scans that were examined were collected to include 20 patients in each decade between 20 and 80 years of age as well as 20 patients each who were younger than 20 and who were older than 80 years old. The age range of the patients examined was 1-92 years old with a mean age of 50 years and the gender distribution was 84 females and 76 males (Table). All scanning was performed on a HiSpeed Advantage scanner (General Electric, Waukesha, WI) at this institution utilizing helical technique, intravenous contrast material, and 5-mm contiguous sections. The angle of origin from the aorta of the dominant renal artery and the SMA was measured on the Advantage workstation by one author (P.A.K.) after the data had been reloaded to the workstation from an optical disk. The center of the aorta was located by approximating the circumference of the aorta with either a circular or ovoid tracing and collapsing it to its center using the preloaded software. The center of the renal or mesenteric artery was located by measuring the diameter of the vessel at its origin and determining its midpoint by dividing it in half using preloaded software. A line was then drawn connecting the center of the aorta to the center of the branch artery origin, as depicted in Figure 1. The angle of the branch artery origin from the aorta was measured relative to the plane defined by the L-1 spinous process (LISP). This plane was calculated by drawing a line through the center of the LlSP and a point midway between the inner cortex of each pedicle of the L-2 vertebral body, as depicted in Figure 2. The angle of the origin of the branch artery relative to this plane was calculated using preloaded software, as depicted in Figure 3. With use of this technique, the angle of the origin of each dominant renal artery and the SMA (+I-
Position of the Arterial Origins Dorsal (-) or Ventral (+) to the Patient's Coronal Plane (in degrees) Age Group
Right RA
(Y)
<20 21-30 31 4 0 41-50 51-60 61-70 71-80 >80
Total
No.
+26 2 +32 2 +28 ? +33 ? +32 ? +33 2 +30 ? +26 ?
12 14 16 12 19 13 18 17
17 19 18 18 20 19 18 19
+30
15
148
?
Left RA -3 -3 -5 -7 -10 -9 -6 -14
No.
SMA*
No.
2
9 2 11 2 12 29 ? 10 2 13 2 16 5 15
17 19 19 18 20 19 19 19
+8426 +82?8 +79?9 +7628 +75 2 10 +78 2 12 +77 ? 15 +75 2 13
19 20 19 20 20 20 20 20
-7 2 13
150
+78 2 11
158
Note.-RA = renal artery. * To the left.
standard deviation LSD]) relative to the LlSP was calculated for each patient. The mean angle of origin of each dominant renal artery and the SMA was analyzed for the entire population and for each age group.
SMA
1
I
1 RESULTS Of the 160 patients and 320 real arteries examined, the origin of the dominant renal artery could not be visualized in eight patients on the right and 10 patients on the left kidneys due to artifact caused by adjacent surgical clips, congenital absence, previous nephrectomy, or because the artery origin was not in the plane of any of the CT sections obtained in that particular patient. The origin of the SMA could not be visualized in two patients due to artifact caused by adjacent surgical clips. The mean angle of the origin of the renal artery origins from the aorta, relative to the LISP, was directed 30" (SD 15") dorsally on the right and 7" (SD 13") ventrally on the left for the entire population studied. The mean angle of origin of the SMA was 12" (SD 11") to the left of the LlSP when looking at all of the patients studied. No significant age related trends were identified for all of the arteries studied.
1 DISCUSSION Localization of the site of origin of the renal artery is critical in os-
Figure 1. Diagram illustrating the drawing of the line from the center of the aorta to the center of the vessel origin where B = Al2.
tial renal artery stent placement. Ideally the renal artery stent should cover the entire stenotic segment but extend into the aorta by less than 3 mm (3). This guideline is based on the technical difficulties of catheterizing the stent for additional manipulations as well as the theoretical risk of peripheral embolism if it extends further into the aorta. In addition, if the stent is placed too far into the renal artery, the ostial lesion will be inadequately treated. If this should be recognized during the procedure, an additional stent would be needed to cover the ostium, increasing the cost and risk of the procedure (2,4). Obtaining the appropriate projection to visualize the renal artery ostia in profile is critical to optimal imaging and stent placement. Inadequate images have proven to be a potential source of error in diagnosis (5) as well as in treatment (2). If the stent is deployed in a projection not perpen-
Kim et a1
39
Volume 10 Number 1
cant respects: we based our angle measurements on a fluoroscopically visible landmark (ie, the LISP) whereas they based their findings relative to the horizontal plane defined by the CT table top, and our methodology utilized the software of the Advantage workstation and their methodology predominantly utilized visual estimation.
8.
Figures 2, 3. (2) Diagram illustrating how the plane of the L l S P was determined where B = A/2. (3) Diagram illustrating the results of our study in cross-sectional anatomy.
Figure 4. Axial CT image at the level of the right renal artery origin demonstrates a renal artery stent. An anteroposterior projection (black lines) overestimates the actual length of the proximal stent protruding into the aortic lumen (white lines).
dicular to the long axis of the renal artery origin and the contiguous aorta, then i t may be placed either too proximally or too distally (Fig 4). One way to choose the best angiographic angle for imaging the renal artery ostium is to empirically perform angiography in sev-
eral different oblique projections a n d then choose the best view. However, since there is a high incidence of renal insufficiency in patients with atheromatous renal artery disease, this approach is not practical. Prior studies have addressed the issue of the angle " of origin of the renal artery from the aorta. The two most cited studies were based on cadaver dissections and defined the angle relative to the anterior surface of the aorta or relative to the origin of the SMA (6,7). Unfortunately, these landmarks a r e not fluoroscopically visible and therefore a r e not useful for guiding angiography. A previously published review of the angle of origin of the renal arteries a s depicted with CT suggested t h a t the right renal artery origin was 24" anterolateral and the left renal artery 5" posterolateral (8). However, they pointed out t h a t the variability in the angle of origin was great. Likewise, our study arrived a t similar conclusions, demonstrating a n average right renal artery origin 30" anterolatera1 and a n average left renal artery origin 7" posterolateral, with SDs of 15" and 13", respectively. Our study, however, differed from that of Verschuyl et a1 in two signifi-
SURlMARY Optimal diagnosis of ostial renal artery lesions and stent placement are facilitated by a knowledge of the renal artery anatomy relative to osseous landmarks that are easily identified on fluoroscopy. When the L l S P and the pedicles of L-2 are used a s a reference plane, the optimal the origins of the right and left renal arteries are 30" LAO (left anterior oblique) and 7" LAO (practically a vertical beam), respectively. References 1. Cicuto KP, McLean GK, Oleaga JA, Freiman DB, Grossman RA, Ring EJ. Renal artery stenosis: anatomic classification for percutaneous transluminal angioplasty. AJR 1981; 137599-601. 2. Rees CR. Palmaz stent in atherosclerotic stenoses involving the ostia of the renal arteries: preliminary report of a multicenter study. Radiology 1991; 181:507-514. 3. Raynaud AC, Beyssen BM, TurmelRodrigues LE, et al. Renal artery stent placement: immediate and midterm technical and clinical results. JVIR 1994; 5:849-858. 4. Hennequin LM, Joffre FG, Rousseau HP, et al. Renal artery stent placement: long-term results with the Wallstent endoprosthesis. Radiology 1994; 191:713-719. 5. Foster JH, Klatte EC, Burko H. Arteriographic pitfalls in the diagnosis of renovascular hypertension. Arch Surg 1969; 99:792-801. 6. Keen EN. Origin of the renal arteries from the aorta. Acta Anat 1981; 110:285-286. 7. Odman P, Ranninger K. The location of the renal arteries: an angiographic and postmortem study. Am J Roentgen01 1968; 104:283-288. 8. Verschuyl EJ, Kaatee R, Beck FJ, et al. Renal artery origins: location and distribution in the transverse plane at CT. Radiology 1997; 203:7175.
Fluoroscopic Landmarks for Optimal Visualization of the Proximal Renal Arteries1 -
Patricia A. Kim, M D Neil M. Khilnani, M D David W. Trost, M D Thomas A. Sos, M D Lisa Lee, M D
Index terms: Renal angiography Renal arteries, anatomy * Renal arteries, stents and prostheses
JVIR 1999; 10:37-39 Abbreviations: LISP = L-1 spinous process, LAO = left anterior oblique, SD = standard deviation, SMA = superior mesenteric artery
PURPOSE: To accurately determine the in vivo orientation of the origin of the renal arteries from the aorta relative to a fluoroscopic bony landmark for optimal diagnostic arteriography and renal artery stent placement. MATERIALS AND METHODS: One hundred sixty abdominal computed tomography (CT) scans of patients in eight age groups (20-90 years) were reviewed to determine the angle of the origins of the renal arteries from the aorta relative to the long axis of the L-1 spinous process (LISP). RESULTS: The right renal artery arises ventrally at an angle of 30" (standard deviation [SD] = 15")from a plane orthogonal to the long axis of the LISP. The left renal artery arises dorsally at an angle of 7" (SD = 13")relative to the same plane. CONCLUSIONS: The optimal initial angle for angiographic evaluation of the origin of the renal artery and for renal artery stent placement is 30" left anterior oblique (LAO) relative to the LlSP for the right renal artery and 7" LAO for the left renal artery. Unfortunately, there is variability in the angle of the renal artery origins from the aorta which cannot be controlled for using this technique. In some patients, additional views will be necessary to optimally depict the origins of the renal arteries.
From the Department of Radiology (P.A.K., D.W.T., N.M.K., T.A.S.), The New York Hospital-Cornell Medical Cen. ter, 525 E. 68th St., New York, NY, 10021, and the Department of Radiology (L.L.), North Shore Hospital, New York, NY. Received June 27, 1997; revision requested August 29; revision received June 15, 1998; accepted June 16. Address correspondence to N.M.K. O SCVIR, 1999
TO accurately demonstrate the proximal renal artery with x-ray angiography, the plane of the xray beam must be orthogonal to the plane of the renal artery origin from the aorta. This concept is important when investigating patients with suspected atheromatous renal artery stenoses, as the lesions are most frequently found a t the renal artery origin. Moreover, as percutaneous treatment of ostial renal artery stenoses requires stent placement to eliminate elastic recoil (l,2), accurate depiction of the renal artery ostium as it originates from the aorta is crucial to adequately place an endovascular stent. We recently undertook a review
of the angle of the origin of the renal arteries from the aorta as measured by axial computed tomography (CT) images. The goal of our study was to determine whether correcting for subtle variations in the rotation of the patient on the examination table by using the long axis of the spinous process of the L-1 vertebral body, a fluoroscopically visible landmark, could lead to the development of a reliable angiographic projection for the depiction of the renal artery ostia. We also evaluated the location of the superior mesenteric artery (SMA) orifice in relation to the same fluoroscopic landmark to allow comparison of our data with previous studies.
38
a
Fluoroscopic Landmarks of the Proximal Renal Arteries
January 1999 JVIR
I MATERIALS AND METHODS The CT scans of the upper abdomen of 160 patients, obtained between October 1, 1995, and February 29, 1996, were retrospectively reviewed. The CT scans that were examined were collected to include 20 patients in each decade between 20 and 80 years of age as well as 20 patients each who were younger than 20 and who were older than 80 years old. The age range of the patients examined was 1-92 years old with a mean age of 50 years and the gender distribution was 84 females and 76 males (Table). All scanning was performed on a HiSpeed Advantage scanner (General Electric, Waukesha, WI) at this institution utilizing helical technique, intravenous contrast material, and 5-mm contiguous sections. The angle of origin from the aorta of the dominant renal artery and the SMA was measured on the Advantage workstation by one author (P.A.K.) after the data had been reloaded to the workstation from an optical disk. The center of the aorta was located by approximating the circumference of the aorta with either a circular or ovoid tracing and collapsing it to its center using the preloaded software. The center of the renal or mesenteric artery was located by measuring the diameter of the vessel at its origin and determining its midpoint by dividing it in half using preloaded software. A line was then drawn connecting the center of the aorta to the center of the branch artery origin, as depicted in Figure 1. The angle of the branch artery origin from the aorta was measured relative to the plane defined by the L-1 spinous process (LISP). This plane was calculated by drawing a line through the center of the LlSP and a point midway between the inner cortex of each pedicle of the L-2 vertebral body, as depicted in Figure 2. The angle of the origin of the branch artery relative to this plane was calculated using preloaded software, as depicted in Figure 3. With use of this technique, the angle of the origin of each dominant renal artery and the SMA (+I-
Position of the Arterial Origins Dorsal (-) or Ventral (+) to the Patient's Coronal Plane (in degrees) Age Group
Right RA
(Y)
<20 21-30 31 4 0 41-50 51-60 61-70 71-80 >80
Total
No.
+26 2 +32 2 +28 ? +33 ? +32 ? +33 2 +30 ? +26 ?
12 14 16 12 19 13 18 17
17 19 18 18 20 19 18 19
+30
15
148
?
Left RA -3 -3 -5 -7 -10 -9 -6 -14
No.
SMA*
No.
2
9 2 11 2 12 29 ? 10 2 13 2 16 5 15
17 19 19 18 20 19 19 19
+8426 +82?8 +79?9 +7628 +75 2 10 +78 2 12 +77 ? 15 +75 2 13
19 20 19 20 20 20 20 20
-7 2 13
150
+78 2 11
158
Note.-RA = renal artery. * To the left.
standard deviation LSD]) relative to the LlSP was calculated for each patient. The mean angle of origin of each dominant renal artery and the SMA was analyzed for the entire population and for each age group.
SMA
1
I
1 RESULTS Of the 160 patients and 320 real arteries examined, the origin of the dominant renal artery could not be visualized in eight patients on the right and 10 patients on the left kidneys due to artifact caused by adjacent surgical clips, congenital absence, previous nephrectomy, or because the artery origin was not in the plane of any of the CT sections obtained in that particular patient. The origin of the SMA could not be visualized in two patients due to artifact caused by adjacent surgical clips. The mean angle of the origin of the renal artery origins from the aorta, relative to the LISP, was directed 30" (SD 15") dorsally on the right and 7" (SD 13") ventrally on the left for the entire population studied. The mean angle of origin of the SMA was 12" (SD 11") to the left of the LlSP when looking at all of the patients studied. No significant age related trends were identified for all of the arteries studied.
1 DISCUSSION Localization of the site of origin of the renal artery is critical in os-
Figure 1. Diagram illustrating the drawing of the line from the center of the aorta to the center of the vessel origin where B = Al2.
tial renal artery stent placement. Ideally the renal artery stent should cover the entire stenotic segment but extend into the aorta by less than 3 mm (3). This guideline is based on the technical difficulties of catheterizing the stent for additional manipulations as well as the theoretical risk of peripheral embolism if it extends further into the aorta. In addition, if the stent is placed too far into the renal artery, the ostial lesion will be inadequately treated. If this should be recognized during the procedure, an additional stent would be needed to cover the ostium, increasing the cost and risk of the procedure (2,4). Obtaining the appropriate projection to visualize the renal artery ostia in profile is critical to optimal imaging and stent placement. Inadequate images have proven to be a potential source of error in diagnosis (5) as well as in treatment (2). If the stent is deployed in a projection not perpen-
Kim et a1
39
Volume 10 Number 1
cant respects: we based our angle measurements on a fluoroscopically visible landmark (ie, the LISP) whereas they based their findings relative to the horizontal plane defined by the CT table top, and our methodology utilized the software of the Advantage workstation and their methodology predominantly utilized visual estimation.
8.
Figures 2, 3. (2) Diagram illustrating how the plane of the L l S P was determined where B = A/2. (3) Diagram illustrating the results of our study in cross-sectional anatomy.
Figure 4. Axial CT image at the level of the right renal artery origin demonstrates a renal artery stent. An anteroposterior projection (black lines) overestimates the actual length of the proximal stent protruding into the aortic lumen (white lines).
dicular to the long axis of the renal artery origin and the contiguous aorta, then i t may be placed either too proximally or too distally (Fig 4). One way to choose the best angiographic angle for imaging the renal artery ostium is to empirically perform angiography in sev-
eral different oblique projections a n d then choose the best view. However, since there is a high incidence of renal insufficiency in patients with atheromatous renal artery disease, this approach is not practical. Prior studies have addressed the issue of the angle " of origin of the renal artery from the aorta. The two most cited studies were based on cadaver dissections and defined the angle relative to the anterior surface of the aorta or relative to the origin of the SMA (6,7). Unfortunately, these landmarks a r e not fluoroscopically visible and therefore a r e not useful for guiding angiography. A previously published review of the angle of origin of the renal arteries a s depicted with CT suggested t h a t the right renal artery origin was 24" anterolateral and the left renal artery 5" posterolateral (8). However, they pointed out t h a t the variability in the angle of origin was great. Likewise, our study arrived a t similar conclusions, demonstrating a n average right renal artery origin 30" anterolatera1 and a n average left renal artery origin 7" posterolateral, with SDs of 15" and 13", respectively. Our study, however, differed from that of Verschuyl et a1 in two signifi-
SURlMARY Optimal diagnosis of ostial renal artery lesions and stent placement are facilitated by a knowledge of the renal artery anatomy relative to osseous landmarks that are easily identified on fluoroscopy. When the L l S P and the pedicles of L-2 are used a s a reference plane, the optimal the origins of the right and left renal arteries are 30" LAO (left anterior oblique) and 7" LAO (practically a vertical beam), respectively. References 1. Cicuto KP, McLean GK, Oleaga JA, Freiman DB, Grossman RA, Ring EJ. Renal artery stenosis: anatomic classification for percutaneous transluminal angioplasty. AJR 1981; 137599-601. 2. Rees CR. Palmaz stent in atherosclerotic stenoses involving the ostia of the renal arteries: preliminary report of a multicenter study. Radiology 1991; 181:507-514. 3. Raynaud AC, Beyssen BM, TurmelRodrigues LE, et al. Renal artery stent placement: immediate and midterm technical and clinical results. JVIR 1994; 5:849-858. 4. Hennequin LM, Joffre FG, Rousseau HP, et al. Renal artery stent placement: long-term results with the Wallstent endoprosthesis. Radiology 1994; 191:713-719. 5. Foster JH, Klatte EC, Burko H. Arteriographic pitfalls in the diagnosis of renovascular hypertension. Arch Surg 1969; 99:792-801. 6. Keen EN. Origin of the renal arteries from the aorta. Acta Anat 1981; 110:285-286. 7. Odman P, Ranninger K. The location of the renal arteries: an angiographic and postmortem study. Am J Roentgen01 1968; 104:283-288. 8. Verschuyl EJ, Kaatee R, Beck FJ, et al. Renal artery origins: location and distribution in the transverse plane at CT. Radiology 1997; 203:7175.