Imaging of the renal donor and transplant recipient

R A D I O L O G I CC L I N I C S

O F N O R T H A M E R I C A

Radiol Clin N Am 46 (2008) 79–93

79

Imaging of the Renal Donorand Transplant RecipientAnand K. Singh, MDa,b, Dushyant V. Sahani, MDa,b,*

- Renal donor evaluationSurgical considerations for donor

nephrectomyImaging in donorsMultidetector CT versus MR imaging for

evaluation of renal donorsMultidetector CT techniqueMR imaging in evaluation of renal donorsRole of image postprocessing

- Renal transplant recipient evaluationSurgical considerations in recipientsRole of ultrasound and DopplerComplications and imaging considerations

in recipients- Summary- References

Renal transplant remains the mainstay of thetreatment of end-stage renal disease. With improve-ment in management strategies and the diverse im-aging options, the yearly survival of recipients withfunctional kidneys has improved significantly. Thisimproved survival is attributed to factors such as im-munosuppressive therapy planning in recipients,human leukocyte antigen matching, surgeon experi-ence, and recipient’s age. Transplantation offers theclosest thing to a normal state if the transplantedkidney can replace the failed kidneys. Living-donorkidney transplants are playing a vital role in bridgingthe gap between decreased supply of, and increaseddemand for, kidneys for transplant. Early detectionand characterization of complications in the recipi-ent are of immense clinical relevance, allowingtimely intervention to prevent graft failure.

With the advances in several imaging options likemultidetector CT (MDCT), MR imaging, and ultra-sound, along with recent technical upgrades in

a Department of Radiology, Division of AbdominalHospital, White-2-270, 55 Fruit Street, Boston, MA 0211b Harvard Medical School, Boston, MA, USA* Corresponding author. Department of Radiology, DMassachusetts General Hospital, White 2-270, 55 Fruit SE-mail address: [email protected] (D.V. Sahani).

0033-8389/08/$ – see front matter ª 2008 Elsevier Inc. All rightsradiologic.theclinics.com

image postprocessing, meticulous donor selectionand early detection of transplant complication arenow possible [1,2].

Renal donor evaluation

Surgical considerations for donornephrectomy

It is important that the healthy kidney be left in thedonor and the other harvested for the recipient. Af-ter confirmation of normal size, location, and func-tion of donor kidneys, a precise delineation of thedonor kidney’s vascular and collecting system anat-omy is a vital prerequisite for presurgical planning.Although the kidney with the less complex vascularanatomy is preferred for harvesting, it is equally im-portant to ensure that the single donor kidney to beleft out is free of pathologies such as stones, cysts,and so forth. Such details would also help thesurgeon decide if the preferred and less invasive

Imaging and Intervention, Massachusetts General4, USA

ivision of Abdominal Imaging and Intervention,treet, Boston, MA 02114.

reserved. doi:10.1016/j.rcl.2008.01.009

Fig. 1. A console-generated coronal maximum inten-sity projection in a 56-year-old female donor showingthree arteries (thin arrows) supplying the right kid-ney, branching of the right main renal artery (thickarrow), and two renal veins (asterisks).

Singh & Sahani80

laparoscopic approach to donor nephrectomy ispossible or whether an open approach should beadopted [3]. In view of this, the presence of a tinysingle calculus (<5 mm diameter) or a cyst ofsame size in one of the donor kidneys is not a con-traindication for its retrieval for a recipient, irre-spective of the complexities in its vascularvariants. If both donor kidneys are normal, thenthe kidney with the less complex vascular anatomyis preferred, making a less invasive laparoscopic ap-proach for nephrectomy feasible. The left kidney isusually preferred for a recipient because it providesa longer segment of the renal vein, which joins theinferior vena cava (IVC), and thus provides moremaneuverability to the surgeon to suture the donorvessel patch to the recipient’s iliac vein.

In addition to defining the vascular anatomy andvariants, imaging should clearly depict pathologicconditions like renal artery atherosclerosis, fibro-muscular dysplasia, aneurysm, and thrombosis. Ac-cessory renal arteries are seen in up to 30% of cases,and they usually originate from the aorta. Occa-sionally, these arteries may arise from the iliac ar-teries and rarely, from the mesenteric and lumbararteries [4]. Delineation and clear outlining ofsmall accessory arteries, which can be as small as1 to 2mm in diameter, are important imaging pre-requisites from a surgical standpoint. Furthermore,a clear differentiation between two separate acces-sory arteries from prehilar branching (renal arterybranching within 20 mm of renal artery origin) isextremely helpful and can sometimes help avoidtorrential bleeding complications [5].

Similarly, multiple renal veins are seen in up to30% of patients. An important presurgical imagingcommunication is confirmation of the presence orabsence of venous variants such as the circumaorticrenal vein (a single renal vein that is split or two renalveins encircling the aorta before joining the IVC),an isolated retroaortic left renal vein, and abnorma-lities such as venous thrombosis and varices [4].

Recently, the assessing of kidney volume beforetransplant has also gained importance, becausetransplant of the larger donor kidney has a more fa-vorable posttransplant outcome rate.

Imaging in donors

With the advent of MDCT and advances in the MRscanner, current donor evaluation protocols are im-proving rapidly. Both these imaging modalitieshave proven promising in detecting vascular andcollecting system variants with an established in-crease in readers’ confidence [6]. With this develop-ment, the use of catheter angiography for mappingrenal vasculature has virtually faded. Furthermore,the value of image postprocessing has added to in-creased acceptability of the CT and MR images to

referring physicians because postprocessed imagesprovide a close simulation to the operative findingsduring surgery [7]. The high–resolution, thin-sliceacquisitions provided by the newer CT and MR im-aging scanners make it now possible to detect thinaccessory renal arteries (Fig. 1) [8]. CT and MR ur-ography also provide a clear delineation of the pye-loureteral anatomy, with added benefits providedby three-dimensional (3D) postprocessing.

Multidetector CT versus MR imagingfor evaluation of renal donors

The better spatial resolution, faster speed, andgreater cost effectiveness of CT have led to a wide ac-ceptance of CT over MR imaging in most centers. Al-though CT and MR angiography have demonstratedsubstantial agreement in the preoperative evalua-tion of renal donors [9], more published researchdata on the integrity of CT technique, contrast vol-ume, and injection rates, and various revolutionaryCT protocol techniques, have definitely tilted thebalance toward MDCT, leading to its widespread ac-ceptance for imaging renal donors. The interob-server disagreement in the interpretation of CTand MR angiography is related to overreading andunderreading of small vessels (1–2 mm in diame-ter) (Fig. 2) [10,11]. With the similarity of CT andMR imaging accuracies, the potential advantagesand disadvantages associated with each modalityhave been widely discussed recently.

MR angiography is a safe and noninvasive tech-nique for comprehensive evaluation of renaldonors. It is radiation free and particularly advanta-geous in patients who are prone to allergic reactionfrom iodinated contrast media. The limitations of

Fig. 2. MDCT coronal maximum intensity projection images in a 48-year-old female donor (A) and a 54-year-oldmale donor (B), showing early branching (arrows) of the main left renal arteries.

Renal Donor and Transplant Recipient 81

MR imaging include decreased spatial resolutionand restrictions in slab length, which can lead tofailure to image large volumes [10]. These limita-tions can lead to misinterpretations, especially incases where the accessory artery arises from the iliacvessel or where the lower ureter has a tiny stone[11]. Such considerations have led to increased ac-ceptance of CT angiography for the preoperativeevaluation of renal donors.

Multidetector CT technique

Earlier four-phase CT acquisition for donor evalua-tion usually used precontrast, arterial, venous, and4- to 10-minute delayed phases. However, becauseof concerns about radiation dose, particularly inhealthy renal donors, various studies have sug-gested a reduction in phase acquisitions and an al-teration in scan parameters. From these, it is nowgenerally accepted that the use of a dedicated ve-nous phase acquisition can be avoided by increas-ing the delay time for the arterial phase, whichproduces contrast opacification of renal veins andthe adjoining IVC, along with the renal arteriesand aorta. For arterial phase acquisitions, most au-thorities have recommended an empiric delay timeof 25 seconds [12]. However, use of automated bo-lus tracking software is favored because accuracy inthe timing of image acquisition is so important. Ac-curate depiction of lumbar and adrenal veins maynecessitate a dedicated venous phase acquisitionat 55 to 65 seconds delay. However, in the authors’experience, obtaining a delayed arterial phase ac-quisition with 25 to 30 seconds’ delay and usingpostprocessing algorithms on these CT datasets ofdelayed arterial phase axial sections, differentiation

among lumbar vessels and accessory renal arteriesand identification of adrenal and gonadal veinscan easily be achieved. Acquisitions with such de-lays should be thin (1–2 mm) to ensure good qual-ity of postprocessed images for detection of arterialand venous variants (Fig. 3).

Some groups have also gone to the extent ofomitting the excretory phase, believing that the lo-calizer radiograph or CT excretory phase scoutimages are sufficient to outline the pyeloureteralanatomy (Figs. 4 and 5) [13]. However, becauseof the possibility of overlooking some rare urothe-lial neoplasms, this proposal has still not becomeuniversally accepted.

Excretory phase acquisition is usually acquiredafter 4 to 8 minutes’ delay time of contrast injec-tion. In the authors’ institution, they usually obtainan 80-kilovolt (peak) (kV[p]) excretory phase scoutimage before the excretory phase acquisition be-cause such a kilovoltage value provides closer ap-proximation to the k-edge value of iodinatedcontrast in the collecting system and enhances thevisibility and delineation on scout images [14]. Inview of dose savings for renal donors, the excretoryphase acquisitions should be thicker (eg, 10 mmthickness obtained at 10 mm intervals) so that theproximal two thirds of the pelvicalyceal system iscovered by the scan, with exclusion of the pelvicportion. Although some authorities still use excre-tory phase acquisitions of lesser thickness coveringthe entire pelvicalyceal system up to the pelvis, theauthors feel that knowledge of the proximal twothirds of the collecting system, which can easilybe revealed by scout images, is generally sufficient.Thicker 10 mm CT acquisitions of the restricted

Fig. 3. (A) CT angiography axial image of a 45-year-old male donor with a retroaortic renal vein (arrow). (B) Thickaxial MDCT volume maximum intensity projection showing a retroaortic renal vein in a 47-year-old femaledonor. (C) Thick coronal MDCT volume maximum intensity projection showing a circumaortic renal vein ina 51-year-old male donor. Anterior (arrowhead) and retroaortic (arrow) portions of left renal vein are seen.

Singh & Sahani82

portion can also assess for distal urothelial pathol-ogies, which can lead to a significant reduction inradiation dose for these healthy subjects.

Recent papers have advocated omitting the pre-contrast phase for the CTevaluation of these healthysubjects and have pointed out the efficacy of con-trast-enhanced arterial phase images for detectionof small calculus up to 8 mm [15,16]. However, be-cause of the prevalence of some false-positive detec-tions, further validation of this work is needed.

Given these considerations, the protocol at theauthors’ institution is tailored to include a precon-trast phase, an arterial phase (25-second delay) ofthin section acquisition, and thicker (10 mm) excre-tory phase acquisition of the upper two thirds of thecollecting system, in addition to the delayed-phaseCT scout images (Table 1). A further increase indose savings and cost effectiveness can be achievedby using a higher concentration of contrast mediain smaller volumes with higher injection rates andby optimizing scan parameters, such as tube current(milliamperes) and tube voltage (kilovolts).

Some believe that before scanning the patient, itis advisable to obtain good distension of the pelvi-calyceal system by ensuring adequate intake of oralcontrast. However, fluid-distended bowel loopsmay interfere with the quality of postprocessed im-ages; as a result, most discourage the use of oralcontrast in their CT angiography protocols.

The administration of intravenous contrast mediashould be designed so that it coincides with andcomplements the speed of phase acquisition. Inthe authors’ institution, nonionic contrast (300 or370 mg iodine/mL concentration) is administeredat a rate of 4 to 5 mL/s. The contrast media dose isbased on the patient’s body weight (550–600 mgiodine/kg weight). The maximum permissible limitwith respect to injectable volume of contrast is 120mL of 300 mg iodine and 100 mL of 370 mg Iodine.

MR imaging in evaluation of renal donors

Certain MR imaging sequences pertinent to preop-erative evaluation of renal donors are illustrated

Fig. 4. Incomplete duplication of the left collectingsystem seen in thick-volume coronal maximum inten-sity projection image generated from the delayed-phase MDCT acquisitions in a 44-year-male donor.

Renal Donor and Transplant Recipient 83

in Table 2. The postcontrast sequence after a timingbolus is usually a 3D fast-gradient echo sequence(repetition time 5.00 ms, echo time 2.3 ms, flip an-gle 30�) using 25 to 30 mL intravenous gadoliniumwith immediate postinjection acquisition followedby three scan series acquisitions per breath hold.Lastly, coronal 3D fast-gradient echo with fat sup-pression is obtained immediately after the dynamicseries of delayed contrast-enhanced images. Coro-nal T2-weighted images can serve as a localizer se-quence, whereas characterization of fluid-filled

Fig. 5. (A) Low-dose (80 kilovolt [peak]) CT scout (localizer)47-year-old male donor showing normal collecting systemabdomen. (B) Low-dose CT scout shows presence of incommal left collecting system anatomy in a 54-year-old fema

lesions like cysts and abscesses can be better appre-ciated on the axial T2-weighted fat-suppressed se-quence. In- and out-of-phase sequences are usefulfor confirmation of small amounts of fat, whereasthe postcontrast gradient-echo fat-suppressed se-quence can be useful for delineating renal venousanatomy and detecting thrombus.

In imaging of renal donors, a special emphasisneeds to be placed on sequences that optimally de-lineate the vascular anatomy. Use of a larger flipangle (up to 45�) can be applied to the MR arteriog-raphy sequence to reduce the background noisearound the bright-appearing contrast-enhanced re-nal arteries [17]. The venous acquisition shouldbe acquired immediately after the arterial and it isadvisable to use a lower flip angle (15–25�) to bet-ter visualize the renal veins, which contain less con-trast volume than renal arteries because of thenephrographic phase washout [17]. The acquireddata can then be sent for postprocessing, wheretwo-dimensional and 3D volume maximum inten-sity projections (MIPs) and volume-rendered (VR)images are created at dedicated workstations.

With the advent of 3-tesla and higher-generationMR scanners, a better signal-to-noise ratio is nowachievable, with shorter sequence acquisitiontimes, which greatly adds to the resolution of theimages and enhances the quality of postprocessedimages. However, the full potential and advantagesof these scanners are still under evaluation.

Role of image postprocessing

These recent advances in CT and MR imaging haveemphasized the benefits and quality of the

images obtained 6 minutes after contrast injection in. Surgical staples from prior surgery are seen in the leftplete duplication of right collecting system with nor-

le donor.

Table 1: Sixteen- and 64-slice multidetector CT protocol for renal donors

Phase Scan range Slice thickness (mm) Interval (mm)

Precontrast Dome of diaphragmto iliac crest

10 10

Automated bolus triggering used to obtain a delay of 25 seconds after contrast injectionArterial phase Dome of diaphragm

to iliac crest1.250 (16-slice MDCT) 0.6250.625 (64-slice MDCT)

6 minute delayed scout images of the abdomen and pelvis: 80 kV(p), 10 mA (anteroposterior scout)6 minute delayed phase Dome of diaphragm

to iliac crest10 10

Contrast injection: up to 120 mL of 370 mg iodine/mL or up to 150 mL of 300 mg iodine/mL at 4 mL/s; 40 mL saline chase.Noise index is 10–5, depending on body weight.

Singh & Sahani84

postprocessed 3D images (Figs. 6, 7). Image postpro-cessing provides not only a simplified display of in-formation of entire MDCT or MR data sets but alsoa close simulation of the operative field of view tothe surgeon. A stack of MIP reformats obtained ina plane parallel to the renal vessel axis outlines the re-nal vascular anatomy better (Fig. 8) [7]. VR and MIPscreen-capture images in various desired angles suchas the coronal or oblique coronal planes are becom-ing increasingly popular with surgeons for preopera-tive mapping of donor kidney anatomy [7,18,19].From recent study results and experience, it is now be-lieved that 3D images obtained at various desired an-gles alone can provide all the useful information tothe readers, and independent viewing of axial imagesis essential only for increasing diagnostic confidencein the visualization of smaller accessory arteries orthe differentiation of two thin accessory arteriesfrom an early arterial branching [7]. Furthermore, be-cause of increased clinical acceptability of the 3D im-ages, image postprocessing has evolved to become anintegral part of donor imaging. MDCT image post-processing protocol is given in Table 3.

An enhanced realization of postprocessing bene-fits is particularly evident in renal donor imaging

Table 2: MR protocol for imaging renal donors

MR sequence Repetition time (ms) Echo

Axial T2W FSE 2500 90

Axial T1W GE(in-phase andout-of-phase)

170 4.7/2.

Coronal T2W FSE 1500 105

Postcontrast 3Daxial fat-saturated(bolus timing 120–30 mL intravenousgadolinium)a

5.0 2.3

Abbreviations: FSE, fast spin-echo; GE, gradient echo; T1W, Ta Immediate acquisition after injection followed by acquisitio

with the ready availability of donor kidney volumesbecause the larger kidney is usually preferred fromthe donor, provided the other kidney is functionallynormal and free of pathologies.

Renal transplant recipient evaluation

Assessment for graft dysfunction immediately aftersurgery is imperative because it can channel themedical management appropriately in cases withcomplications. Ultrasound, Doppler, and nuclearmedicine are the main imaging modalities usedfor immediate assessment of transplanted kidneys.Ultrasound plays a crucial role in the imaging of re-cipients immediately after surgery and has provenuseful for follow-up in such patients [20,21]. More-over, ultrasound has emerged as a simple-to-use anda cost-effective modality. An added vital advantageis its usefulness in image-guided percutaneous inter-ventions, whereas Doppler ultrasound remains themainstay for detecting most vascular complications.

Radionuclide imaging is a good tool to demon-strate graft function [22] and is usually used in caseswhere rejection is suspected. For the evaluation ofa renal transplant recipient, MDCT and MR imaging

time (ms) Fat suppression (±) Breath hold (±)

1 1

3 – 1

– –

– 1

1-weighted; T2W, T2-weighted.ns at 30, 70, and 120 seconds.

Fig. 6. The left adrenal vein (arrow) and left gonadalvein (arrowhead) on a coronal MDCT reformat ofa 49-year-old male donor.

Renal Donor and Transplant Recipient 85

are usually used as adjunct modalities either to con-firm the findings of the ultrasound or to image largepatients because of unsatisfactory ultrasound im-ages. Catheter angiography still retains its gold stan-dard value in the diagnosis and treatment ofpostsurgery vascular complications in a recipient,but its use has diminished greatly because of thepopularity gained by MR imaging in the delineationof vascular pathologies such as renal artery stenosis(RAS) and renal vein thrombosis.

Surgical considerations in recipients

Renal allografts are usually placed in the iliac fossa,extraperitoneally. The cadaver kidneys are harvestedwith an intact main renal artery and an adjoiningcut portion of the aorta. The aortic portion is thentrimmed in such a way that it can be exactly suturedto the recipient’s external iliac artery (end to side

Fig. 7. Delineation of left gonadal vein (arrow) in a 44-yeMDCT reformat (A) than a coronal postcontrast T1-weighfrom MDCT.

anastomosis) (Fig. 9) [23,24]. However, in somecases, the internal iliac artery is preferred. Ina case with multiple arteries, if the origin is presenton a single aortic patch, the above procedure is fol-lowed. However, in a case of arteries with a distantorigin or polar arteries, anastomosis to the main re-nal artery is usually done.

Similarly, the renal vein is harvested with the ad-joining portion of the IVC and is anastomosed tothe external iliac vein. The left renal vein is longerthan the right, which usually makes the left kidneypreferable for harvesting in donors, and allowsdirect suturing of the left renal vein on the nativeiliacs. However, in cases where the right kidneyis harvested because of the vascular complexitiesof the left kidney, a large segment of the donor’sIVC is usually retrieved, making the anastomoticprocedure in the recipient more cumbersome.

Ureteral anastomosis is the most difficult part ofthe surgery, which involves implantation of the cutureteral end from the donor, directly into the recip-ient’s bladder muscle [23]. The double ureter can betransplanted separately or each one partially anas-tomosed to the other beforehand and then trans-planted as a single unit [25].

Usually, the venous anastomosis is performedfirst, followed by arterial anastomosis, and finally,the ureteral anastomosis. In the case of iliac arteryatherosclerosis in the recipient, the less affectedside is usually chosen for a transplant.

Role of ultrasound and Doppler

Ultrasound still remains the most preferred andeconomic modality for imaging recipients aftertransplant. It is noninvasive, can be easily done atthe bedside, and is conveniently repeatable to assessresponse or monitor treatment in such patients.

ar-old male donor is better appreciated on a coronalted MR (B) because of improved resolution provided

Fig. 8. CT angiography images of left kidney in a 32-year-old female donor. (A) Reconstruction design used at theworkstation for generation of 6-mm oblique coronal MIPs planned parallel to the obliquity of renal vasculature.(B) Oblique coronal MIP showing accessory artery (thin arrow) and vein (thick arrow) at the lower pole. Acces-sory venous origin is from the left common iliac vein.

Singh & Sahani86

A transplant kidney can easily be visualized in ei-ther iliac fossa and is usually placed a few centime-ters below the skin surface. However, certain factorsmay alter the image resolution, including postoper-ative edema in the skin and soft tissue, the patient’sbuild, and the distance of the kidney below the skinsurface.

A transplant kidney usually appears morpholog-ically similar to a native kidney, except for somesubtle differences. Usually, the peripheral paren-chyma is well defined and hypoechoic, comparedwith the bright appearance of the echogenic renalsinus. Also, in the transplant kidney, the pyramidsare usually well delineated and easily differentiatedfrom the adjacent parenchyma (Fig. 10). They usu-ally appear more hypoechoic, compared with theremaining parenchyma. A transplanted kidney

Table 3: Postprocessing protocol forgeneration of maximum intensity projectionsusing multidetector CT data sets atworkstations

ViewAlgorithmused

Slicethickness(mm)

Overlap(mm)

Axial MIP 6 3.0Coronal MIP 6 3.0Obliquecoronalsa

MIP — 3.0

Delayed(single)coronalobliqueb

MIP 7 3.5

a Obtained for both kidneys.b Including both kidneys.

may also display a mild degree of hydronephrosis,which is usually due to edema at the site wherethe transplanted ureter is anastomosed to the na-tive’s bladder. This hydronephrosis usually resolvesin time and may act as a baseline for subsequent ul-trasounds of the recipient.

Color Doppler provides a satisfactory and imme-diate assessment of the blood perfusion in thetransplanted kidney. However, spectral Doppler isalways used for quantification. Usually, the flowin the interlobar artery of the transplant kidneyshould demonstrate a waveform similar to the

Fig. 9. The most commonly used technique for sur-gery in recipient: end to side anastomosis of vesselsof donor kidney to the recipient’s external iliac ves-sels. The cut end of donor’s ureter is directly suturedto the bladder wall.

Fig. 10. (A) Transplanted right kidney showing a well-defined and hypoechoic peripheral parenchyma comparedwith the bright appearance of the echogenic renal sinus with easy differentiation of pyramids, as comparedwith the kidney appearance in a normal individual (B).

Renal Donor and Transplant Recipient 87

one in the native kidney or the recipient hepatic ar-tery. Normally, it is a low-resistance waveform withthe diastolic flow contributing up to nearly one halfof the peak systolic value. Reduction in the diastolicflow is usually indicative of a pathologic process. Ina case with adverse findings, it is usually advisableto perform Doppler studies every 24 to 48 hoursin the early transplant period until the results aresatisfactory.

The most frequently relied on Doppler indicesare the pulsatility index (PI) and the resistive index(RI). With regard to the renal artery, peak systolicvelocity up to 2.5m/s is considered normal [26–28]. The renal vein has no predefined value and itis simply assessed by the presence or absence offlow.

Complications and imaging considerationsin recipients

Complications after renal transplant can be broadlyclassified into renal, urologic, vascular, and sys-temic. A list of preferred investigations is given inTable 4.

Early complicationsAcute tubular necrosis versus acute rejection

Differentiating between acute tubular necrosis(ATN) and acute rejection is often difficult, consider-ing the similarity of symptom presentation in theearly posttransplant period. Ultrasound offers a lim-ited role in diagnosis and usually, a histologic spec-imen of the transplanted kidney is required fordiagnostic confirmation. Ultrasound is, however,the modality of choice for monitoring transplantdysfunction and assessing response to dialysis. Acute

rejection usually presents within 1 to 3 weeks andmay be asymptomatic or associated with fever or up-per respiratory symptoms (Fig. 11). The presence ofacute rejection is, however, a poor prognostic indica-tor [29]. On the other hand, in ATN, the presence ofother associated symptoms is usually rare.

B-mode ultrasound may reveal findings like re-duction in corticomedullary differentiation and re-nal sinus echoes, increased cortical echogenicity,and initial increase in renal length and cross-sec-tional area. However, these findings are nonspecificand no longer relied on for prediction of acute re-jection. Color Doppler, in combination with spec-tral Doppler, is more specific for assessment of thevessels and detection of RAS (Fig. 12). A PI greaterthan 1.8 and an RI greater than 0.7 are regarded asabnormal. However, Doppler has also failed to dif-ferentiate ATN from acute rejection; but the modal-ity does provide a reasonable qualitative andquantitative evaluation of renal perfusion. Loss ofdiastolic flow with alteration in PI and RI can occurin ATN and acute rejection. Certain studies have re-vealed some encouraging results; for example,short acceleration time on day 1 was associatedwith longer duration of delayed function, whereasan acceleration time of less than 90 millisecondsat day 5 was associated with a high risk of rejection[30].

Ultrasound features, when correlated with clini-cal symptoms, are usually sufficient for a convincingdiagnosis. Radionuclide study is the next in order; itmay also be helpful in the diagnosis of acute rejec-tion where an early normal postoperative phasestudy changes subsequently into an abnormalstudy. In cases of ATN, the renal excretion is

Table 4: Common posttransplant complications and diagnostic investigations for recipient workup

Complication Ultrasonography/Doppler Confirmatory investigation

A. Early complicationsAcute tubular necrosis vs acuterejection

Nonconfirmatory: PI>1.8,RI>0.7, unable to differentiate

Radionuclide imaging/biopsy

Vascular:Renal vein thrombosis (rare) Confirmatory: dilatation

of renal vein with thrombuswith absent flow

—

Hemorrhage Confirmatory: guidedaspiration needed

(Re-exploration if significant)

Perirenal collection:Hematoma, abscess Confirmatory: guided

aspiration needed—

Ureteric obstruction (extrinsiccompression by collection orintraluminal clots)

Detection of hydronephrosisby ultrasonography

Ultrasound-guided percutaneousnephrostomy if persistent

B. Late complicationsChronic rejection Nonconfirmatory: decrease

in kidney size and corticalechogenicity

Ultrasound-guided biopsy

RAS Nonconfirmatory: PSV>2.5 m/s,RI>0.7 (confirmatory if RI>0.9)

Angiography (not needed if R.I>0.9), (MR angiography done inselect places)

Arteriovenous fistula Nonconfirmatory: arterial andvenous flow on Doppler (smallfistulae resolve on their own)

Angiography for confirmation iflarge and persistent

Abbreviation: PSV, peak systolic velocity.

Singh & Sahani88

abnormal but the perfusion phases are relativelywell maintained (Fig. 13). MR imaging may showa diffuse increase in cortical intensity and loss of cor-ticomedullary differentiation on T1-weighted scans,whereas an angiogram may show prolonged arterialphase, poor washout, and a patchy nephrogram.

Thrombosis Renalvein thrombosis is rare, althoughmore common than renal arterial thrombosis, andusually occurs between the third and eighth day ofthe posttransplant period. Characteristic featuresof renal vein thrombosis include a dilated

Fig. 11. (A) Acute rejection of transplanted right kidney sold male recipient. (B) Marked reduction of renal perfusi

transplanted renal vein containing a thrombuswith absent venous flow, reversed diastolic flow,or a low-amplitude parvus tardus within the intra-renal arterial system and the transplant renal arteryon Doppler imaging [31,32]. Arterial thrombosis,although rare, occurs in the early transplant periodand shows complete absence of flow in the maintransplant renal artery and intrarenal vasculatureon color flow and spectral analyses [33].

Infections, obstruction, and perirenal collections

Indwelling urinary catheters are usually responsible

howing increase in cortical echogenicity in a 46-year-on is seen on power Doppler.

Fig. 12. Spectral Doppler analysis ofa transplanted renal artery showingcolor aliasing, peak systolic (PS) velocityof 4.4m/s, and RI of 0.87, consistentwith transplant artery stenosis.

Renal Donor and Transplant Recipient 89

for infections. The ultrasound picture may revealechoes in the bladder and an associated thickeningof the bladder wall but mostly the kidneys appearnormal in early infection. Obstruction can be dueto intraluminal blood clots secondary to surgeryor extraluminal compression of the ureter by collec-tions such as a lymphocele or hematoma. Hema-toma mostly appears as a collection withseptations and echoes, and is usually formed as

Fig. 13. Radionuclide scintigraphy study showing normalplanted right kidney, consistent with acute tubular necro

a result of bleeding from the skin excision of thevascular anastomosis itself (Fig. 14). The presenceof a clearly anechoic collection or fluid with mini-mal echoes around the bladder, with reduction ofurine output and increasing abdominal pain, is sug-gestive of an urinoma. In such cases, performing anisotope scan, cystogram, or an antegrade pyelogra-phy may prove useful for confirmation becausecommunication between the extravesical and

perfusion but delayed dye excretion from the trans-sis.

Fig. 14. (A) A small perinephric collection seen around the transplanted right kidney (arrow) in a 49-year-old fe-male recipient. (B) A complex collection with echoes (thin arrow) seen in a 42-year-old male recipient above thetransplanted right kidney (thick arrow), which shows an increase in cortical echogenicity consistent with acuterejection.

Singh & Sahani90

intravesical fluid can then be established. However,in such cases, ultrasound still remains the modalityof choice for providing guidance to aspirate anddrain these collections.

Early strictures are managed by balloon dilata-tion and stent placement for 1 to 12 weeks, whichgives transient or permanent relief of obstructionin some patients.

However, in a case with presenting symptoms ofurinary obstruction, ultrasound may fail to revealthe presence of a tiny calculus as its cause, in whichcase an unenhanced thin-section helical CT scan isthe modality of choice. In a case with lymphoceles,scintigraphy demonstrates a photopenic area,

Fig. 15. MR angiogram showing kinking (arrows) ofthe renal artery of the transplanted left kidney ina 50-year-old male recipient.

which does not fill up with tracer on delayed images[34], whereas CT usually reveals a well-definedround or oval collection of 0 to 20 HU with no vis-ible communication with the bladder lumen. Ultra-sound usually serves as the initial modality ofchoice except in cases where visualization becomesdifficult because of patient build or considerablepostoperative skin edema, in which case CT or MRimaging is useful.

Late complicationsChronic rejection and drug toxicity Cyclosporineand tacrolimus have well-documented side effectson renal function and may lead to gradual failureof the transplanted kidney. At times it is difficultto distinguish gradual failure from chronic rejec-tion. The strongest predisposing factor for chronicrejection is a history of repeated episodes of acuterejection. A generalized decrease in size of the trans-planted kidney and increased echogenicity aresome features of chronic rejection on ultrasound.However, the biopsy specimen provides the onlymeans of confirmation, by revealing interstitialfibrosis and tubular atrophy.

Radionuclide studies show rapid uptake andwashout. MR shows loss of corticomedullary differ-entiation, whereas MR spectroscopy shows alteredlevels of phosphate metabolites.

Vascular complications: renal artery stenosis and

arteriovenous fistula The most common site ofstenosis is at the site of anastomosis, followed bythe postanastomotic transplant artery, mostlybecause of injury by the perfusion cannula. It is im-portant to assess the recipient’s iliac artery proximalto the anastomosis and ensure normal bloodflow before the anastomotic region is assessed.

Fig. 16. (A) Angiogram showing RAS (arrow) in a 47-year-old female recipient. (B) RAS (arrow) in the region ofthe graft with immediate appearance of contrast in renal vein (arrowhead), consistent with arteriovenous fis-tula. An area of arteriovenous communication is also depicted (curved arrow).

Renal Donor and Transplant Recipient 91

Uncontrolled long-standing hypertension and dete-rioration of renal function following angiotensinconverting enzyme inhibitor therapy are alarmingsigns and should raise suspicion for RAS. Early diag-nosis before the setting of irreversible ischemia iscrucial; angioplasty may salvage the transplant kid-ney and prevent host death. Stenosis can occurearly, within a few months, most often caused bytrauma to the donor’s or recipient’s vessel duringclamping, or it may be delayed for few years, inwhich case atherosclerosis of the adjacent iliac ar-tery is usually the cause. Kinking of the renal arterycan occur when the transplant arterial length is lon-ger than the vein, which usually occurs with theright-sided graft, raising clinical suspicion of RAS.

Doppler is preferred as the initial diagnostic mo-dality. Most stenotic lesions, however, occur at, orclose to, the anastomotic site and appear as focalaliasing on color flow imaging with peak systolic ve-locity greater than 2.5m/s on spectral flow [35]. Nu-merous twists and turns of the transplanted renalartery may contribute to defective Doppler anglecorrection on color flow imaging, in which casepower Doppler may prove useful in providinga complete display of the renal vessel in question.The stenosis can be confirmed by angiography,which also provides a good estimate of the vesselextent and helps in the planning of percutaneoustransluminal angioplasty.

Conventional angiography is invasive and posesan increased risk for thromboembolism leading tograft loss, groin hematoma pseudoaneurysms, andarteriovenous fistula. Contrast-induced nephropa-thy in the setting of a transplanted kidney andradiation dose issues are some important

considerations that may make MR angiographypreferable to CT and conventional angiography inthe future. Nevertheless, conventional angiographyhas still retained its crucial role in diagnosticworkup in most institutions. Doppler imaging canestablish signs of RAS or venous thrombosis inthe graft and is usually the first modality in theworkup list because of its cost effectiveness, butMR imaging is usually done to confirm the findingsof Doppler. MR imaging may be particularly helpfulwhen a patient presents with signs of RAS despitenormal Doppler findings; arterial kinks can be read-ily identified on MR imaging (Fig. 15). Surgical-clips, unless made of titanium, limit MR imagingbecause they may lead to artifacts, which can mimicRAS, in which case conventional angiography isdone.

Formation of arteriovenous fistula is often a post-renal biopsy phenomenon and usually resolves onits own (Fig. 16) [36]. Spectral analysis revealsa focal pool of color flow containing arterial and ve-nous components on spectral analysis; visualiza-tion of the pathologically increased flow of thefistula can be achieved by increasing the pulse rep-etition frequency.

Summary

MDCT provides greater accuracy and spatial resolu-tion, a faster scan time, and better cost effectivenesswhen compared with MR imaging, which makes it,for most, the modality of choice for preoperativeevaluation of renal donors. Although MR imagingshows accuracy similar to CT for evaluation of therenal vascular anatomy in renal donors, it has

Singh & Sahani92

potential disadvantages, such as a longer scan time,higher cost, and inferior resolution, compared withMDCT. However, it is advisable to tailor the MDCTprotocol for renal donors so that the protocol in-cludes a precontrast phase, a late arterial phase ofthin section acquisition, and thicker (8–10 mm) ex-cretory phase acquisitions of the upper two thirdsof the collecting system, which can reduce theoverall radiation dose to these healthy patientscompared with former protocols. Image postpro-cessing plays a vital role in CT and MR, wherethick-volume MIPs and VR images can clearly dem-onstrate the renal vascular variants. Surgeons havewelcomed these reconstructions because they helpthem choose the donor kidney and plan their lapa-roscopic versus open surgical approach.

Ultrasound and Doppler still remain the modal-ities of choice for assessment of posttransplantcomplications in recipients. They are useful as pre-dictors of immediate complications like ATN andgraft rejection and for diagnosing renal vascular pa-thologies. Ultrasound also facilitates the diagnosisof postoperative collections like hemorrhage, ab-scess, and urinoma. It is a preferred modality forguided diagnostic and therapeutic interventionspertaining to these complications. CT and MR im-aging are preferred in patients where ultrasoundimaging is difficult because of factors like increasedskin thickness due to postoperative edema, andlarge patient size, but MR imaging is more acceptedin a posttransplant setting because CT and conven-tional angiography pose risks of contrast-inducednephropathy. Radionuclide imaging plays an im-portant role in demonstrating perfusion of the renalgraft posttransplant and is usually preferred to con-firm findings of ultrasound in the case of graft rejec-tion and ATN. Angiography still retains its goldstandard value in the diagnosis of vascular compli-cations such as RAS and arteriovenous fistula butMR angiography has benefited from significanttechnologic advances.

References

[1] Rubin GD. MDCT imaging of the aorta and pe-ripheral vessels [review]. Eur J Radiol 2003;45(Suppl 1):S42–9.

[2] Flohr T, Stierstorfer K, Bruder H, et al. Newtechnical developments in multislice CT–Part 1:approaching isotropic resolution with sub-milli-meter 16-slice scanning. Rofo 2002;174(7):839–45.

[3] Flowers JL, Jacobs S, Cho E, et al. Comparison ofopen and laparoscopic live donor nephrectomy.Ann Surg 1997;226(4):483–9 [discussion:489–90].

[4] Urban BA, Ratner LE, Fishman EK. Three-dimen-sional volume-rendered CT angiography of the

renal arteries and veins: normal anatomy, vari-ants, and clinical applications. [review]. Radio-graphics 2001;21(2):373–86 [questionnaire:549–55].

[5] Kawamoto S, Lawler LP, Fishman EK. Evaluationof the renal venous system on late arterial andvenous phase images with MDCT angiographyin potential living laparoscopic renal donors.AJR Am J Roentgenol 2005;184(2):539–45.

[6] Sahani DV, Rastogi N, Greenfield AC, et al. Multi-detector row CT in evaluation of 94 living renaldonors by readers with varied experience. Radiol-ogy 2005;235(3):905–10.

[7] Singh AK, Sahani DV, Kagay CR, et al. Semiauto-mated MIP images created directly on 16-sectionmultidetector CT console for evaluation of livingrenal donors. Radiology 2007;244(2):583–90.

[8] Moritz M, Halpern E, Mitchell D, et al. Compar-ison of CT and MR angiography for evaluation ofliving renal donors. Transplant Proc 2001;33(1–2):831–2 No abstract available.

[9] Toki K, Takahara S, Kokado Y, et al. Comparisonof CT angiography with MR angiography in theliving renal donor. Transplant Proc 1998;30(7):2998–3000.

[10] Neimatallah MA, Dong Q, Schoenberg SO, et al.Magnetic resonance imaging in renal transplanta-tion. J Magn Reson Imaging 1999;10(3):357–68.

[11] Israel GM, Lee VS, Edye M, et al. ComprehensiveMR imaging in the preoperative evaluation of liv-ing donor candidates for laparoscopic nephrec-tomy: initial experience. Radiology 2002;225(2):427–32.

[12] Kawamoto S, Montgomery RA, Lawler LP, et al.Multi-detector row CT evaluation of living renaldonors prior to laparoscopic nephrectomy [re-view]. Radiographics 2004;24(2):453–66.

[13] Foley WD. Renal MDCT. Eur J Radiol 2003;45(Suppl 1):S73–8.

[14] Hussain SM, Kock MC, IJzermans JN, et al. MRimaging: a ‘‘one-stop shop’’ modality for preop-erative evaluation of potential living kidney do-nors. Radiographics 2003;23(2):505–20.

[15] Kim JK, Park SY, Kim HJ, et al. Living donor kid-neys: usefulness of multi-detector row CT forcomprehensive evaluation. Radiology 2003;229(3):869–76.

[16] Singh AH, Sahani DV, Hahn PF, et al. Is a low dose(80 kVp) excretory phase scout image a viable al-ternative to thin section CT urography for accuratedelineation of pyeloureteric anatomy of a kidneydonor? [abstract 279]. In: Abstract book of Amer-ican Roentgen Ray Society 106th annual meeting.Leesburg (VA): American Roentgen Ray Society;2006. p. 73–4.

[17] Kalva SP, Sahani DV, Hahn PF, et al. Using theK-edge to improve contrast conspicuity and tolower radiation dose with a 16-MDCT: a phan-tom and human study. J Comput Assist Tomogr2006;30(3):391–7.

[18] Rastogi N, Sahani DV, Blake MA, et al. Eva-luation of living renal donors: accuracy of

Renal Donor and Transplant Recipient 93

three-dimensional 16-section CT. Radiology2006;240(1):136–44 [Epub May 23, 2006].

[19] Rubin GD. 3-D imaging with MDCT [review].Eur J Radiol 2003;45(Suppl 1):S37–41.

[20] Baxter GM. Ultrasound of renal transplantation[review]. Clin Radiol 2001;56(10):802–18.

[21] Pozniak MA, Dodd GD III, Kelcz F. Ultrasono-graphic evaluation of renal transplantation [re-view]. Radiol Clin North Am 1992;30(5):1053–66.

[22] Brown ED, Chen MY, Wolfman NT, et al. Com-plications of renal transplantation: evaluationwith US and radionuclide imaging [review]. Ra-diographics 2000;20(3):607–22.

[23] Hanto DW, Simmons RL. Renal transplantation:clinical considerations [review]. Radiol ClinNorth Am 1987;25(2):239–48.

[24] Shapiro R, Simmons RL. Renal transplantation.Atlas of organ transplantation. New York: GowerMedical; 1992.

[25] Oliver JH III. Clinical indications, recipient eval-uation, surgical considerations and role of CTand MR in renal transplantation. Radiol ClinNorth Am 1995;33(3):435–46.

[26] Duda SH, Erley CM, Wakat J, et al. Posttrans-plant renal artery stenosis–outpatient intraarterialDSA versus color aided duplex Doppler sonogra-phy. Eur J Radiol 1993;16(2):95–101.

[27] Snider JF, Hunter DW, Moradian GP, et al. Trans-plant renal artery stenosis: evaluation with du-plex sonography. Radiology 1989;172(3 Pt 2):1027–30.

[28] Rigsby CM, Burns PN, Weltin GG, et al. Vascularcomplications in renal allografts: detection withduplex Doppler US. Radiology 1987;162(1 Pt 1):31–8.

[29] Pirsch JD, Ploeg RJ, Gange S, et al. Determinantsof graft survival after renal transplantation.Transplantation 1996;61(11):1581–6.

[30] Merkus JW, Hoitsma AJ, van Asten WN, et al.Doppler spectrum analysis to diagnose rejectionduring post transplant acute renal failure. Trans-plantation 1994;58(5):570–6.

[31] Baxter GM, Morley P, Dall B. Acute renal veinthrombosis in renal allografts: new Doppler ul-trasonic findings. Clin Radiol 1991;43(2):125–7.

[32] Reuther G, Wanjura D, Bauer H. Acute renal veinthrombosis in renal allografts: detection withduplex Doppler US. Radiology 1989;170(2):557–8.

[33] Sidhu PS, Baxter GM. Ultrasound of abdominaltransplantation. Stuttgart (Germany): Thieme;2002. p. 27–42.

[34] Kumar R, Bharathi Dasan J, Choudhury S, et al.Scintigraphic patterns of lymphocele in post-renal transplant. Nucl Med Commun 2003;24(5):531–5.

[35] Baxter GM, Ireland H, Moss JG, et al. ColourDoppler ultrasound in renal artery stenosis: in-trarenal waveform analysis. Br J Radiol 1996;69(825):810–5.

[36] Merkus JW, Zeebregts CJ, Hoitsma AJ, et al. Highincidence of arteriovenous fistula after biopsy ofkidney allografts. Br J Surg 1993;80(3):310–2.