Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2211001558
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Paul, J.-F.
Right arrow Articles by Gaux, J.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Paul, J.-F.
Right arrow Articles by Gaux, J.-C.
(Radiology. 2001;221:261-265.)
© RSNA, 2001

Technical Developments

Unilateral Renal Artery Stenosis: Perfusion Patterns with Electron-Beam Dynamic CT—Preliminary Experience1

Jean-François Paul, MD, Patricia Ugolini, MD, Marc Sapoval, MD, PhD, Elie Mousseaux, MD, PhD and Jean-Claude Gaux, MD

1 From the Department of Radiology, Hôpital Marie-Lannelongue, 133 Avenue de la Résistance, 92350 Le Plessis-Robinson, France (J.F.P.); Department of Vascular Radiology, Hôpital Européen Georges Pompidou, Paris, France (J.F.P., M.S., E.M., J.C.G.); and Department of Radiology, Montreal Heart Institute, Montreal, Quebec, Canada (P.U.). From the 2000 RSNA scientific assembly. Received September 20, 2000; revision requested November 12; revision received February 23, 2001; accepted March 23. Address correspondence to J.F.P. (e-mail: pauljf@ccml.com).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and Methods
Results
 Discussion
 REFERENCES
 
Renal dynamic computed tomographic (CT) data for 16 patients with unilateral renal artery stenosis were compared with those for 12 control subjects. Three patterns of perfusion were distinguished in stenotic kidneys: pattern A, symmetric time-attenuation curves; pattern B, asymmetric time-attenuation curve with similar perfusion; and pattern C, asymmetric time-attenuation curve with impaired perfusion. Additional functional data can be obtained from the initial timing scan in a CT study of unilateral renal artery stenosis.

Index terms: Computed tomography (CT), angiography, 961.12915, 961.12916 • Kidney, perfusion, 81.12119 • Renal angiography, 961.12915, 961.12916 • Renal arteries, stenosis or obstruction, 961.721


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and Methods
Results
 Discussion
 REFERENCES
 
Noninvasive imaging techniques currently used for diagnosis of renal artery (RA) stenosis include Doppler ultrasonography (US) and magnetic resonance (MR) or computed tomographic (CT) angiography. These techniques are valuable for the detection of stenosis but are poorly informative about renal function. Renal scintigraphy is a functional technique to assess renal perfusion (first-pass measurements) and split renal function. It is much more informative about renal function than about perfusion, however, and separate perfusion values for the kidneys are only estimated as relative indexes. The relationship between RA stenosis and the cortical perfusion of stenotic kidneys remains poorly documented, to our knowledge.

Tissue perfusion may be estimated on a segmental basis by calculating time-attenuation curves from dynamic CT acquisitions (1). The gradient method, which is based on compartmental analysis, is an easily applicable method, as demonstrated at renal scintigraphy (2). It has been used with dynamic CT to determine an index of perfusion for other organs, including the brain, liver, and pancreas (3–5). Few studies address the perfusion of stenotic kidneys (6). Electron-beam CT technology allows functional imaging with high temporal and spatial resolution and multilevel acquisitions, which makes accurate first-pass measurements possible. Dynamic acquisition with a small amount of contrast medium is currently used to determine the delay before imaging of the aorta or RAs (7). By using dynamic electron-beam CT as a timing test before a spiral CT acquisition, the goal of this study was to describe time-attenuation curves observed in cases of unilateral RA stenosis and to compare them with normal curves.


   Materials and Methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
 Discussion
 REFERENCES
 
Patients
All patients with RA stenosis and all control subjects were referred for CT angiography for clinical reasons. In the control group were 12 consecutive subjects (11 men and one woman; age range, 42–79 years; mean age, 64 years) without RA stenosis who were referred to undergo CT angiography to evaluate aortic aneurysm or aortoiliac disease. RA stenosis was excluded in control subjects on the basis of findings at intraarterial digital subtraction angiography (DSA). Creatinine levels were normal for all control subjects (creatinine level, <115 µmol/L).

In the group with RA stenosis were 16 consecutive patients (nine women, seven men; age range, 19–77 years; mean age, 55 years) with hypertension in whom unilateral RA stenosis was diagnosed at intravenous DSA and who underwent CT angiography of the aorta, kidneys, or adrenal glands. Patients with bilateral RA stenosis, cardiac insufficiency, or severe renal insufficiency (creatinine level, >200 µmol/L) were excluded. Stenosis involved the main trunk in 15 cases and the accessory (polar) RA in one case.

Intraarterial DSA was the standard of reference for evaluation of the degree of stenosis. The range of stenosis was 25%–100% (mean, 69%). Creatinine levels were normal in all but one patient with mild renal insufficiency (creatinine level, 154 µmol/L). Arteriosclerosis caused RA stenosis in 14 patients and arteritis in two. Renovascular hypertension was diagnosed in patients with cured or improved hypertension after angioplasty (n = 3) or in patients with a high degree of stenosis (>70%) and a positive captopril-enhanced scintigraphy test (n = 2). In the other 11 patients, RA stenosis was not considered to be the cause of hypertension.

Data Acquisition
Dynamic CT was performed as an initial timing test, which is routine for vascular studies in our institution. With this timing test, the time of peak enhancement for the aorta was calculated, and this value was used as the starting delay for the subsequent spiral CT acquisition. We currently use a multilevel protocol available with electron-beam CT that allows as short an exposure time per section as possible (50 msec); thus, the radiation dose per acquisition is minimal. We could study renal perfusion patterns with this same routine protocol simply by centering the dynamic acquisition on the kidneys. Neither the patients nor the control subjects received any additional radiation from this study. Time-attenuation curves for the patients were compared with those for the control subjects. In addition, dynamic data for two patients were available after RA dilation.

In all patients, 20 sequential 8-mm-thick transverse images were obtained at two targets (four levels) with a matrix of 360 x 360, a 35-cm field of view, and an acquisition time of 50 msec per image. The four levels were centered on the renal hilum, and section levels were selected on initial nonenhanced scans. A 20-gauge needle made of a synthetic fluorine-containing resin (Insyte-autoguard; Becton-Dickinson, Le Pont-de-Claix, France) was inserted into the antebrachial vein of each patient, and 40 mL of a nonionic iodinated contrast medium (iohexol, Omnipaque 350; Nycomed-Amersham, Oslo, Norway) was injected at a flow rate of 5 mL/sec. The scanning delay was 10–15 seconds depending on the patient’s age (<60 years, 10 seconds; >60 years, 15 seconds). The acquisition sequence was 15 scans at 1.5-second intervals followed by five scans at 2-second intervals, which accounted for a total acquisition time of 32.5 seconds. Patients were asked to hold their breath for the duration of the entire sequence, if possible.

Data Analysis
Immediately after the first acquisition, region-of-interest analysis was performed for the aorta of each patient, and time-to-peak enhancement for the aorta was calculated to determine the scanning delay for subsequent spiral CT acquisitions.

Cortical renal perfusion was calculated off-line from the computer-generated curves. Each region of interest was carefully drawn with a trackball device, with use of zooming if necessary. The aortic region of interest was selected on the upper level in the suprarenal aorta. As much as possible of the aortic lumen was included in a round region of interest. The region of interest in the renal cortex was drawn at maximal enhancement, to optimize the signal-to-noise ratio. The inner outline of the cortex was drawn manually, and the external outline was determined with the help of built-in software. As much as possible of the cortex was included to minimize the SDs of measurements. The regions of interest in the kidneys were drawn on each of the four levels available if possible.

Time-attenuation curves were fitted with a gamma-variate algorithm, and the maximum gradient of the first moment of the curves was calculated (Fig 1). Regional cortical perfusion was then calculated (in milliliters per minute per cubic millimeter of tissue) for each kidney, as the ratio of cortical gradient to aortic peak enhancement (1). Cortical peak height and cortical peak time values were obtained at each of the four levels and then averaged. In the case of stenosis of a right upper polar artery, only the values derived from the right upper level were considered. Region-of-interest placement and calculations were performed independently by two radiologists (J.F.P., P.U.), and the SDs of both measurements were used to assess interobserver variability.



View larger version (72K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Time-attenuation curve (bottom) was calculated from the transverse dynamic CT scan (top), which was obtained after injection of a contrast material bolus, on the basis of regions of interest drawn on the left renal cortex. The maximum gradient of the upslope (arrow) was calculated by using the equation of the gamma curve and was divided by the aortic peak enhancement value to obtain the value of the regional perfusion of the left kidney (in milliliters per minute per cubic millimeter of tissue).

 
In control subjects, cortical peak time, cortical peak height, and cortical perfusion were recorded for the right and left kidneys, and the SDs of the differences were calculated for each parameter. A difference of 2 SDs for at least two of the three parameters was used to classify the perfusion patterns. The association between perfusion patterns and renovascular hypertension was evaluated with the Fisher exact test.


   Results
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
Discussion
 REFERENCES
 
The SDs of the differences in measurements between the two radiologists were 4 HU, 0.6 second, and 0.4 mL/min/mm3 for cortical peak height, cortical peak time, and cortical perfusion, respectively.

In the control subjects, the mean differences in cortical peak height, cortical peak time, and cortical perfusion between the right and left kidneys averaged 3 HU ± 6, 0 second ± 0.6, and 0.1 mL/min/mm3 ± 0.3, respectively.

In the patients with RA stenosis, the mean index of perfusion was 3.1 mL/min/mm3 ± 1.1 (range, 1.2–5.2 mL/min/mm3) versus 3.9 mL/min/mm3 ± 0.8 (range, 2.7–5.3 mL/min/mm3) in the contralateral kidney (Table 1).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Dynamic Data for Patients with Unilateral RA Stenosis

 
With use of a cutoff based on the means of the differences in cortical peak height, cortical perfusion, and cortical peak time within ±2 SDs of values for the control subjects, the cortical peak height was lower than that of control subjects in six patients with RA stenosis and higher in two. Cortical perfusion was lower than that of control subjects in six patients, and cortical peak time was delayed in 11.

Three perfusion patterns were defined on the basis of these results: pattern A, symmetric curves (Fig 2); pattern B, similar cortical perfusion with delayed cortical peak time and higher cortical peak height (Fig 3); pattern C, lower cortical perfusion with delayed cortical peak time and lower cortical peak height (Fig 4).



View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Pattern A time-attenuation curve in a 77-year-old male patient with 80% left RA stenosis. The time-attenuation curves were symmetric. CP = cortical perfusion, CPH = cortical peak height, CPT = cortical peak time.

 


View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Pattern B time-attenuation curve in a 54-year-old female patient with 75% right RA stenosis and renovascular hypertension. The time-attenuation curves revealed a higher and delayed peak of contrast medium enhancement in the stenotic kidney, but perfusion was identical in the two kidneys. Both patients with pattern B time-attenuation curves had renovascular hypertension.

 


View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Pattern C time-attenuation curve in a 36-year-old female patient with 95% left RA stenosis. The time-attenuation curves revealed a lower and delayed peak of contrast medium enhancement in the stenotic kidney, which had perfusion lower than that in the nonstenotic kidney.

 
Classification of the curves into these three patterns was identical for the two radiologists. Pattern A was seen in all control subjects. Pattern A was seen in eight patients with RA stenosis, pattern B in two, and pattern C in six. In the two patients for whom data were available before and after RA dilation, the patterns were B and C before RA dilation and normalized (pattern A) after.

None of the patients with pattern A time-attenuation curves had renovascular hypertension. In contrast, both of the patients with pattern B time-attenuation curves and three of the six patients with pattern C time-attenuation curves had renovascular hypertension. The association between asymmetric curves (pattern B or C) and renovascular hypertension was statistically significant (P = .02) (Table 2).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Association between Renovascular Hypertension and Cortical Time-Attenuation Curves in Patients With Unilateral RA Stenosis

 

   Discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
 Discussion
REFERENCES
 
In this study, three different first-pass patterns were identified in stenotic kidneys, and two of these three patterns were associated with renovascular hypertension.

In control subjects, cortical perfusion, cortical peak height, and cortical peak time were very similar in the two kidneys. The SDs were low, which means that good temporal and spatial resolution were required to detect any difference. High temporal resolution (ideally less than 1 second) was required to detect a delay as short as 1.2 seconds between the right and left renal cortical peak times. The 1.5-second interscan delay used in this study may be too large, but the fitting of curves increased accuracy for the determination of cortical peak time and cortical peak height, and the averaging of data from four levels reduced error.

The minimum difference in cortical peak height (12 HU) required for pattern classification corresponded to about 15% of the mean cortical enhancement. This value depended on bolus characteristics, and high injection rates were required to assess any difference. A similar value (16 HU) was found in a previous study (8) for distinguishing substantial differences in RA stenosis, but the interscan delay in this previous study was long at 5 seconds.

Interobserver variability was similar to the variability between right and left kidney values in control subjects, and measurement errors between the two observers were not large enough to affect pattern classification.

Perfusion was high, with symmetric values, in 10 stenotic kidneys, including those of four patients with high-grade (>=70%) stenosis of the RA. As previously reported with scintigraphy (9), levels of perfusion may be high in stenotic kidneys in patients with renovascular hypertension.

In the current study, symmetric curves (pattern A) were seen in all five patients with low-grade (<60%) stenosis. This result was expected, because mild stenosis is known to have no hemodynamic effect. The higher and delayed peak time in the stenotic kidney that was responsible for pattern B was possibly due to the retention of contrast medium in the renal cortex, owing to constriction of the efferent arterioles. In both pattern B cases, the index of perfusion was high (>4 mL/min/mm3) despite the high degree of stenosis. In five (83%) of the six pattern C cases, the kidney was small (<10 cm long). Evidence of abnormally low perfusion in a stenotic kidney may have predictive value for renal atrophy and may constitute an indication for revascularization. A low perfusion level was observed despite occlusion of the RA in one patient; this persistent flow may be due to anastomosis between cortical arteries and other pelvic arteries.

Normalization of patterns B and C after RA dilation demonstrated that RA stenosis was responsible for asymmetric curves. Although there appeared to be an association between patterns B and C and renovascular hypertension, the design of our study and the small sample prevented any assessment of the value of these patterns for predicting renal angioplasty outcomes. In cases of bilateral stenosis, the bilateral effect of the stenoses may result in symmetric curves, which would reduce the value of the technique for the detection of RA stenosis. Interestingly, dynamic CT can help detect segmental hypoperfusion, owing to multilevel acquisition, which may help assess the hemodynamic significance of stenosis of an accessory RA.

The major drawback of dynamic CT is the increased radiation burden. With use of a very short acquisition time (50 msec), however, the additional dose of radiation is small, since the radiation dose is proportional to the exposure time. Ten 50-msec x-ray exposures are roughly equivalent to the exposure during acquisition of one conventional abdominal electron-beam CT scan with 500-msec exposure time. We chose the multilevel mode with electron-beam CT, because it is the only mode that permits use of an acquisition time as short as 50 msec. Two levels are acquired with the same exposure on two different targets. In addition, segmental hypoperfusion may be revealed, as in one patient in this study. Multilevel acquisition is now available with multi–detector row CT, which allows multilevel analysis of perfusion in an individual kidney.

Dynamic CT facilitates optimal timing for the spiral CT acquisition, which is essential for the imaging of RAs (7). We compensated for the initial iodine dose (40 mL) by reducing the dose for the subsequent spiral CT acquisition to 90 mL; thus, the total amount or contrast medium (130 mL) was similar to the standard dose used in vascular studies. Iodine injection should be avoided in patients with renal insufficiency, however, and this is an important limitation of CT angiography. Dynamic CT acquisition requires a long breath hold to allow correct gamma fitting, and this may be difficult for some patients. Kidney motion was substantial in some patients, and calculation of cortex-attenuation values was required for each acquisition time.

First-pass imaging is also feasible with MR imaging (10) or Doppler US by injecting contrast material. At MR imaging, however, renal perfusion can not be directly estimated by using the gradient technique, because the relationship between the MR signal intensity and the concentration of gadolinium chelate is not linear. Previous first-pass scintigraphic studies did not show perfusion to be of prognostic value for angioplasty outcomes (11,12), but the spatial resolution of scintigraphy is poor. Measurements may not be sufficiently accurate, especially for detection of segmental involvement of the kidney.

In conclusion, findings in this preliminary study show that CT time-attenuation curves may provide additional information about kidney perfusion in patients with unilateral RA stenosis. Time-attenuation curves may help distinguish between RA stenosis with and that without preserved perfusion. Different patterns of perfusion may be identified on a regional basis, which would allow detection of segmental hypoperfusion. We found an association between two patterns and renovascular hypertension. Further prospective studies are needed to assess the potential of CT in the evaluation of renal perfusion and as a diagnostic tool for renovascular hypertension.


   FOOTNOTES
 
Abbreviations: DSA = digital subtraction angiography, RA = renal artery

Author contributions: Guarantor of integrity of entire study, J.F.P.; study concepts, J.F.P., E.M., M.S.; study design, J.F.P.; literature research, J.F.P.; clinical studies, all authors; data acquisition and analysis/interpretation, J.F.P., P.U., E.M., M.S.; statistical analysis, J.F.P.; manuscript preparation, J.F.P.; manuscript definition of intellectual content, all authors; manuscript editing, J.F.P.; manuscript revision/review, J.F.P., P.U., E.M., M.S.; manuscript final version approval, all authors.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
 Discussion
 REFERENCES
 

  1. Miles KA. Measurement of tissue perfusion by dynamic computed tomography. Br J Radiol 1991; 64:409-412.[Abstract]
  2. Peters AM, Brown J, Hartnell GG, Myers MJ, Haskell C, Lavender JP. Non-invasive measurement of renal blood flow with 99mTc DTPA: comparison with radiolabelled microspheres. Cardiovasc Res 1987; 21:830-834.[Medline]
  3. Koenig M, Klotz E, Luka B, Venderink DJ, Spittler JF, Heuser L. Perfusion CT of the brain: diagnostic approach for early detection of ischemic stroke. Radiology 1998; 209:85-93.[Abstract/Free Full Text]
  4. Miles KA, Hayball MP, Dixon AK. Functional images of hepatic perfusion obtained with dynamic CT. Radiology 1993; 188:405-411.[Abstract/Free Full Text]
  5. Miles KA, Hayball MP, Dixon AK. Measurement of human pancreatic perfusion using dynamic computed tomography with perfusion imaging. Br J Radiol 1995; 68:471-475.[Abstract]
  6. Lerman LO, Taler SJ, Textor SC, Sheedy PF, Stanson AW, Romero JC. Computed tomography-derived intrarenal blood flow in renovascular and essential hypertension. Kidney Int 1996; 49:846-854.[Medline]
  7. Kaatee R, Beek FJ, de Lange E, et al. Renal artery stenosis: detection and quantification with spiral CT angiography versus optimized digital subtraction angiography. Radiology 1997; 205:121-127.[Abstract/Free Full Text]
  8. Probst P, Mahler F, Roesler H, Fuchs WA. Renal artery stenosis and evaluation of the effect of endoluminal dilatation: comparison of dynamic CT scanning and I-131-OIHA renogram. Invest Radiol 1983; 18:264-271.[CrossRef][Medline]
  9. Arlart I. Predictive diagnostic value of renal blood flow measurement in renovascular hypertension. ROFO Fortschr Geb Rontgenstr Nuklearmed 1979; 131:1-8[German].
  10. Miller S, Schick F, Duda S, et al. Gd-enhanced 3D phase-contrast MR angiography and dynamic perfusion imaging in the diagnosis of renal artery stenosis. Magn Reson Imaging 1998; 16:1005-1012.[CrossRef][Medline]
  11. Peters AM, Brown J, Crossman D, et al. Noninvasive measurement of renal blood flow with technetium-99m-DTPA in the evaluation of patients with suspected renovascular hypertension. J Nucl Med 1990; 31:1980-1985.[Abstract/Free Full Text]
  12. Schreij G, van Es PN, van Kroonenburgh MJ, Kemerink GJ, Heidendal GA, de Leeuw PW. Baseline and postcaptopril renal blood flow measurements in hypertensives suspected of renal artery stenosis. J Nucl Med 1996; 37:1652-1655.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 RadiologyHome page
H. J. Michaely, S. O. Schoenberg, N. Oesingmann, C. Ittrich, C. Buhlig, D. Friedrich, A. Struwe, J. Rieger, C. Reininger, W. Samtleben, et al.
Renal Artery Stenosis: Functional Assessment with Dynamic MR Perfusion Measurements--Feasibility Study
Radiology, February 1, 2006; 238(2): 586 - 596.
[Abstract] [Full Text] [PDF]

Home page
 HypertensionHome page
D. C. Kem, D. F. Lyons, J. Wenzl, D. Halverstadt, and X. Yu
Renin-Dependent Hypertension Caused by Nonfocal Stenotic Aberrant Renal Arteries: Proof of a New Syndrome
Hypertension, August 1, 2005; 46(2): 380 - 385.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Radiol.Home page
K A Miles and M R Griffiths
Perfusion CT: a worthwhile enhancement?
Br. J. Radiol., April 1, 2003; 76(904): 220 - 231.
[Full Text] [PDF]

Home page
 RadiologyHome page
D. A. Gervais, F. J. McGovern, R. S. Arellano, W. S. McDougal, and P. R. Mueller
Renal Cell Carcinoma: Clinical Experience and Technical Success with Radio-frequency Ablation of 42 Tumors
Radiology, February 1, 2003; 226(2): 417 - 424.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
Y. Tsushima, J. Aoki, K. Endo, and J.-F. Paul
Underestimation of Renal Cortical Perfusion Calculated from Dynamic CT Data * Dr Paul responds:
Radiology, August 1, 2002; 224(2): 613 - 614.
[Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Paul, J.-F.
Right arrow Articles by Gaux, J.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Paul, J.-F.
Right arrow Articles by Gaux, J.-C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE