Post Renal Transplant Surgical Complications; MRI Standard Applications and Diagnostic Outcomes

Objective: Fast MR imaging sequences together with paramagnetic contrast agents, offer multiple advantages in the assessment of renal function. It provides cross sectional and vascular information without the risk of ionizing radiation, iodinated contrast or arterial catheterization. Post transplantation complications can be grouped as surgical or medical. Immediate surgical complications include renal artery thrombosis or stenosis, urinary leak or lymphocele. Renal allograft frequently require repeated imaging studies during the immediate post-operative period and various times thereafter, when renal function is compromised. Background: End stage renal disease is common and can result from a variety of diseases. Kidney transplantation from living-related donors offered the best prognosis. Imaging modalities that are currently used to evaluate transplanted kidneys are ultrasound (US), computed tomography (CT), scintigraphy, intravenous urography (IVU), contrast angiography, and magnetic resonance imaging (MRI). Methods: This study was conducted on 181 renal transplant recipients. Recipients were 139 males and 42 females. Their age ranged from 20 to 58 years (mean age 39 years). The patients underwent clinical assessment, Laboratory investigations, and different Radiological imaging procedures as: IGray scale and color Doppler ultrasonography. IIMagnetic Resonance Imaging. 3D Gd-enhanced MRA. MR Urography. Selective IA-DSA of the graft artery. IIIPercutaneous catheter nephrostomy (PCN) and antegrade pyelography. IVRadio-isotope diuretic renogram using 99m Tc-MAG3. Results: 30 renal transplants were examined by MRI in the 1st 2 weeks after renal transplantation. At the end of 1st 2 weeks, MR examinations were carried out, as basal studies (including MRI, MRA and MRU) for 98 transplants. From this group, 64 transplants were subjected to other MR examinations. After the 1st 2 weeks, 53 transplants were subjected to MR examinations for the 1st time at variable post-transplant duration. Among the studied 181 renal transplants, MR examinations detected 3 cases with graft arteries thrombosis (1.6%), 10 with graft arteries stenosis (5.5%), 6 with segmental infarctions (3.3%), 3 cases with graft intrarenal arteries pseudo-aneurysms (1.6%) and 2 cases with arteriovenous fistulae (1.1%) after graft biopsies. Conclusion: MRI is highly recommended to evaluate intra-/extra-renal graft vascular lesions, urinary obstructive syndrome, compressive collections (urinoma, lymphocele), inflammatory and tumoral lesions of the renal graft.


Introduction
End stage renal disease is common and can result from a variety of diseases. The expense and morbidity of dialysis has made renal transplantation the preferred treatment whenever available (1). Kidney transplantation from livingrelated donors offered the best prognosis, a superior quality of life (as compared with hemodialysis or peritoneal dialysis) and improved rehabilitation (2). Imaging modalities that are currently used to evaluate transplanted kidneys are ultrasound (US), computed tomography (CT), scintigraphy, intravenous urography (IVU), contrast angiography, and magnetic resonance imaging (MRI) (3).
Fast MR imaging sequences together with paramagnetic contrast agents, offer multiple advantages in the assessment of renal function. It provides cross sectional and vascular information without the risk of ionizing radiation, iodinated contrast or arterial catheterization (4). Post transplantation complications can be grouped as surgical or medical. Immediate surgical complications include renal artery thrombosis or stenosis, urinary leak or lymphocele. Medical complications include rejection, cyclosporine toxicity and acute tubular necrosis (ATN) (5). Renal allografts frequently require repeated imaging studies during the immediate postoperative period and various times thereafter, when renal function is compromised (6).
This study aims to evaluate: the capability of MRI applications to diagnose and differentiate various post transplantation surgical complications.

A-Gray scale and color Doppler ultrasonography:
Routinely performed for all examined recipients before MR examinations. Equipment: Acuson sequoia 512 gray scale ultrasonography, with 5MHZ multi-frequency curved linear array transducer, was used to exclude obstruction and peri-graft collection. Cross section area of the graft was measured as an indicator for overall graft size. Post-processing techniques for evaluation of Gdenhanced MRA:-3D reconstruction with MIP images: It was routinely done for all examined (181) recipients and carried out on a second console (ultra. Spark /1/sun micro system).
MPR images: MPR images were performed for 71 cases in which multiple graft arteries were surgically reported, tortuous main graft artery and vascular complications were suspected from routine MIP images.
Axial and coronal post-contrast gradient images: Routinely performed for all examined patients immediately after MRA and just before excretory MRU. Imaging analysis: 1) Presence of non-enhancing parenchymal (nephrographic) defects.
2) Assessment of enhancement pattern of renal (parenchymal) and/or peri-renal masses as well as cystic fluid collections.
MRU:-For assessment of the graft size, pelvi-calyceal system configuration and ureter.
(a) Gd-enhanced excretory MRU: Routinely performed after MRA for all examined renal allografts.
Coronal oblique contrast enhanced MRU: 3D reconstruction with MIP is used to evaluate of the following: Renal allograft parenchyma for; shape, size, contour of the kidney and presence of parenchymal filling defects. Pelvi-calyceal system for its shape (e.g. compression, displacement and stretching), and presence of any filling defects (e.g. Blood clots and stones). Ureter and uretero-vesical anastomosis for detection of course and caliber of the ureter and presence of any filling defects. Urinary leaks and fistulae. (b) T2-weighted static MRU: -Heavy T2-weighted MRU: Performed for 98 cases with rising serum creatinine above 2mg/dl, dilated pelvi-calyceal system with or without dilated ureter, presence of perigraft collection (on routine US examination) and in suspected cases of urinary leaks.

2) Percutaneous catheter nephrostomy (PCN) and antegrade pyelography:
Performed for 15 cases (in presence of urinary leakage, renal stones and rising serum creatinine secondary to graft obstruction).

3) Selective IA-DSA of the graft artery:
Performed for 8 patients as a preliminary step of angioplasty. Classification of post renal transplant surgical complications (7)

Results
This study was carried out on 181 renal transplant recipients, their age distribution ranged from 20 to 58 years (mean age 39 years). They were 139 males and 42 females.
Among the studied 181 renal transplants, MR examinations detected 3 cases with graft arteries thrombosis (1.6%), 10 with graft arteries stenosis (5.5%), 6 with segmental infarctions (3.3%), 3 cases with graft intrarenal arteries pseudo-aneurysms (1.6%) and 2 cases with arterio-venous fistulae (1.1%) after graft biopsies. No surgical interference or interventional procedures were performed for patients with segmental infarctions, those with graft arteries pseudo-aneurysms and arterio-venous fistulae, as the infracted areas were small and the follow up color-Doppler US of the grafts showed spontaneous thrombosis of the pseudo-aneurysms and obliteration of the fistulae with no need for embolization.
In our study, the 3 cases with graft arteries thrombosis as diagnosed by Gd-enhanced 3D MRA were surgically explored where 2 cases were managed by graft nephrectomy and the 3 rd case with incomplete arterial thrombosis was managed by thrombus removal and graft survival. Among the diagnosed cases with graft arteries stenosis (10 cases), surgical re-anastomosis of the graft arteries to the internal iliac arteries were performed in 2 patients with arterial atherosclerosis and calcified arterial wall in one patient and failure of previous angioplasty in the other one. IA-DSA for the remaining 8 patients were performed as a preliminary step for angioplasty and revealed 6 cases with significant arterial stenosis who were subjected to balloon dilatation while the remaining 2 cases showed no evidence of stenosis.
The MRA findings for graft arteries thrombosis, as compared with surgical findings, were 100% as regarding sensitivity, specificity and overall accuracy while those for graft arteries stenosis as compared to the IA-DSA and surgical findings were 100%, 98.8%, 98.9% and 80% as regarding sensitivity, specificity, overall accuracy and positive predictive value. The sensitivity, specificity, overall accuracy and positive predictive values for detection of graft arteries stenosis were calculated and were 100% 90.9%, 93.3%, 80%, for 3D-FSPGR and 100%, 95.4%, 96.7% and 88.9% for 3D-PC respectively.
The sensitivity, specificity and overall accuracy of MIP images and MPR images, as post-processing techniques, performed for 71 patients in evaluation of surgically and interventionally suspected vascular complications were; 66.6%, 100%, 98.6%, 100%, 96.8% and 97.2% for Graft arteries thrombosis and graft arteries stenosis detected by MIP images while they were 100% for MPR in detection of graft artery thrombosis and 100%, 98.4% and 98.6% in detection of graft arteries stenosis.
We had 100% sensitivity, specificity, overall accuracy and positive predictive value for MRA in detection of graft artery thrombosis and 100%, 98.8%, 98.9% and 80% for graft artery stenosis respectively. MRA detected 3 cases with graft artery thrombosis (1.6%), 10 with graft artery stenosis (5.5%), 6 with segmental infarction (3.3%) and 3 cases with post biopsy graft pseudo-aneurysms (1.6%). The sensitivity and specificity of detecting graft artery stenosis in our study were comparable to those reported by Aneesh Srivastava et al. (9) and were 100% and 98% respectively.
The sensitivity, specificity, overall accuracy and positive predictive values for 3D-Gd enhanced FSPGR MRA and 3D-PC MRA, performed for 30 patients, for the detection of graft arteries stenosis were: These data are comparable to those reported by Liu X et al. (8) concerning the sensitivity and specificity for 3D-TOF and 3D-PC MRA in evaluation of post renal transplant artery stenosis which were: However our data are incomparable to those reported by Huber et al. (10) which were: The sensitivity, specificity an overall accuracy of MIP and MPR images were 66.6%, 100%, 98.6% and 100% for detection of graft arteries thrombosis while they were 100%, 96.8%, 97.2%, 100%, 98.4% and 98.6% for detection of graft arteries stenosis.
These results are comparable to those of Hany et al. (11) which were 96%, 91% and 92% for MIP, regarding sensitivity, specificity and overall accuracy in detection of renal artery stenosis while for MPR they were 96%, 97% and 96%. In spite of the short post-processing times for MIP images, they are handicapped by projection -related limitations in a manner identical to conventional catheter angiogram.
Eccentrically located stenosis and superimposition of structures may simulate the presence of stenosis; therefore MIP images are generally acquired in two or more projections.
The sensitivity, specificity and overall accuracy for Gdenhanced MRU in detection of urinary leakage were 66.6%, 100% and 97.9% while both heavy T 2 W and single-shot MRU failed to detect any urine leak. Our results, concerning the superiority of Gd-enhanced MRU in detection of posttransplant urinary leakage agree with Claus et al. (12) who reported that Gd-enhanced MRU is a very accurate and promising imaging technique for the detection of urinary leaks and fistula.
The misdiagnosed one patient with ureteral obstruction was due to marked hydrouretronephrosis with subsequent delayed excretion of Gd-DTPA and inaccurate localization of the obstruction level in both source images and MIP image. The sensitivity, specificity and overall accuracy for Gdenhanced MRU in detection of ureteral obstruction were 80%, 100% and 98.9% and for both heavy T 2 W and single-shot MRU they were 100%. These results concerning static heavy T 2 W MRU were comparable to those reported by Schubert et al. (13) and were 80%, 100% and 98.8%.
As regarding the peri-transplant fluid collections; hematoma is common in the immediate postoperative period. It is usually small and resolves spontaneously while large haematoma may displace the transplanted kidney producing hydronephrosis or rupture intraperitoneally and may produce shock. In such cases diagnostic aspiration with or without percutaneous drainage may be performed (2).
Urinoma is usually found between the transplanted kidney and the bladder in the first one or two weeks post-operatively. It occurs because of continued, slow extravasation of urine from the renal pelvis, ureter or uretero-vesical anastomosis (14). Lymphocele occurs either in early postoperative period or in late postoperative period. If large, it increases progressively in size causing hydronephrosis and requires drainage (15). Abscess formation can arise de novo, or it may be due to superimposition of a peri-transplant fluid collection. Needle aspiration with either surgical or sonographically guided percutaneous drainage is performed (14).
In our study; MRI detected six of seven patients with perigraft hematomas (85.7%), 5 of 6 patients with urinomas (83.3%), 8 of 10 patients with lymphoceles (80%) and 3 cases with abscesses formation (100%). The remaining one patient with peri-graft hematoma was interpreted as lymphocele; this was due to central liquefaction of the haematoma and higher SI than that of urine (U.B).
The misdiagnosed one patient with urinoma was diagnosed as lymphocele; this may be attributed to the presence of small associated peri-graft hematoma altering the signal intensity in both T1 and T2-W MR images. While the remaining misdiagnosed two patients with lymphoceles were interpretated as perigraft hematomas due to presence of small associated perirenal hematoma and masking of peri-renal fat which gave high signals in both T 1 and T 2 WIs.
Our results, concerning the distribution of posttransplantation fluid collections, are comparable to Fang et al. (8) in that the patients with peri-graft hematomas and urinomas were detected within the 1 st postoperative two weeks while those with lymphoceles and abscesses formation were detected at variable post-operative periods.

Conclusions
The study demonstrated that Gd-enhanced MRA (utilizing both FSPGR and PC) of the transplant artery with MPR alone /or with MIP images as post-processing techniques assessed graft artery stenosis with high accuracy, complementary Gdenhanced FSPGR T 1 weighted MR images and Gd-enhanced MRU allowed rapid Global assessment of the renal parenchyma, pelvi-calyceal system and ureter together with the peri-transplant region for enhancing masses or fluid collections. MRA can replace IA-DSA in patients with impaired renal function while conventional IA-DSA of the graft artery is reserved for those with positive MR angiographic findings as a preliminary step for interventional technique (percutaneous balloon dilatation).
MRU is a valuable non invasive, non nephrotoxic technique for the assessment of renal transplants in cases with suspicion of complication in the excretory system. Gdenhanced MRU is a promising alternative in evaluation of post-transplant urinary leakage, non dilated urinary tract with no hazards of radiation exposure or iodinated CM. static T 2weighted MRU is restricted to cases with dilated pelvicalyceal system and impaired excretory function of the graft.
The addition of diffusion MRI is helpful in many situations of post-transplant complications.
Case No. (1) A 40 year old male patient transplanted since 3 days presented with accidentally discovered hypo-echoic fluid collection and mild hydronephrosis on routine post-operative US.