The urinary fistula frequency ranges from 1 to 5 %, the most common early complication, occurring during the first 2 weeks after transplantation. The majority of urinary leaks are attributed to ureteral ischemia. The other causes of urinary leaks are failure of vesicoureteral anastomosis or a missed ureteral duplication. Leakage may be suspected when increasing volumes of a clear liquid are collected by drains, while diuresis tends to decline during the days following surgery. Determination of fluid electrolytes, ingestion of methylene blue thereafter found in the fluid collected and US showing a perirenal fluid collection can confirm the diagnosis. US usually finds a well-defined anechoic collection at the lower pole of the kidney (Fig. 6). This fluid appears hypointense on unenhanced T1-weighted MR images and hyperintense on T2W sequences. T2W MR urography shows the collection and the entire dilated excretory system. The differential diagnosis includes lymphoceles, which usually occur later. Chemical analysis of the fluid may suggest the diagnosis if the creatinine concentration is higher than that in the blood. However, the definitive diagnosis of urinoma is based on the demonstration of an extravasation of contrast medium into the collection after intravenous injection. This can be obtained with Gd-enhanced MRI, iodine-enhanced CT, or radionuclides. Urine collection contamination by the contrast agent or the radiotracer often requires delayed imaging (5–15 min after injection) (Fig. 6d, e).

Whereas the treatment of most of these leaks is surgical, small ones can be treated by nephrostomy and/or ureter stenting. Open surgery consists of reimplanting the ureter, with the technique to be used chosen as a function of its remaining healthy length.

This complication develops later (several months), and its frequency tends to increase with time, 5 % at 1 year and 10 % at 5 years. In 80 % of the cases, it is the consequence of progressive fibrosis of the vesicoureteral anastomosis, but it can also result from inflammatory infiltration of the ureter wall during rejection. The differential diagnosis includes intraluminal causes of obstruction, such as lithiasis or blood clot, and extraluminal compression by perigraft fluid collections.

The diagnosis is suspected when decreased renal function is associated with dilatation of the collecting system on US (Fig. 7a, b). Differentiation between obstructive and nonobstructive pyelocaliectasis remains difficult, and Doppler RI measurement is neither sensitive nor specific in transplanted kidneys (Platt et al. 1991). Renal scintigraphy may characterize the obstruction when the tracer accumulates within the collecting system on delayed images and by measuring increased clearance time after furosemide injection.

When obstruction is suspected, visualization of the entire upper urinary tract is mandatory to determine its exact location and cause. The least invasive method for that purpose is MR urography because most of these patients have decreased renal function (Fig. 7). Both types of sequences (T2W and Gd-enhanced T1-weighted) have to be obtained. When dilatation is sufficient, T2W sequences demonstrate the entire collecting system up to the anastomosis or the site of obstruction. If not, T1-weighted Gd-enhanced sequences, with delayed acquisitions, usually do.

Urinary infections are common during the first month following transplantation. They are usually nosocomial bacterial infections, sometimes facilitated by the presence of catheters, which can lead to real pyelonephritis of the graft or the development of renal or perirenal abscesses. Sometimes, perigraft abscesses may be secondary to bacterial contamination of a preexisting fluid collection (hematoma, lymphocele, or urinoma).

US, using B-mode and Doppler techniques, should be performed first and will show a perigraft collection (Fig. 8), often extending toward superficial planes, which, in this context, must always be considered infected. Thin or coarse echoes, sediment, and septa and a thickened hypervascularized peripheral wall are evocative of infection. When infection of a perigraft collection is suspected, contrast-enhanced MRI or CT will help to assess their exact extension before treatment. Fluid aspiration from these collections, under US control, may be necessary for confirmation of the diagnosis, and optimal treatment combines percutaneous or surgical drainage and systemic antibiotic therapy.

Renal parenchyma infection may be seen as an increased graft volume and/or areas of decreased or increased echogenicity with decreased flow on color Doppler examination (Fig. 9). Gd-enhanced MRI and enhanced CT are able to distinguish between infection and infarction in most of cases, because enhancement is observed in the former and not in the latter, except for a thin peripheral capsular rim.

Urinary infections with Corynebacterium urealyticum is uncommon, but can expose the transplant to complications, such as ureteral obstruction, renal abscess formation, or progressive destruction of the graft (Dominguez-Gil et al. 1999). Color Doppler US detects these urothelial calcifications that produce a twinkling artifact within the bladder and/or the upper urinary tract (Fig. 10).

In addition to those associated with fistulas and perigraft abscesses discussed above, these fluid buildups can be constituted of blood and/or lymph. Postsurgical lymphorrhea causing a lymphocele remains the most frequent cause of perigraft collections, occurring in 1–20 % of renal graft recipients, usually after the fourth week posttransplantation. Sometimes, their volume may cause ureteral or venous compression. On ultrasonograms, a lymphocele appears as a well-defined anechoic collection (Fig. 11a). CT density values and SI on MR sequences (Fig. 11) are typical of simple fluids, without any enhancement after contrast injection. When a lymphocele is responsible for ureteral or venous compression, radical therapy is essential. Simple aspiration and drainage are ineffective because they do not prevent lymph leakage. Therefore, percutaneous drainage must be combined with sclerosis of the cavity by repeated instillations of doxycycline, tetracycline, acetic acid, alcohol, or povidone–iodine. Multiple sessions until daily drainage falls below 10 mL are necessary for larger ones (Karcaaltincaba and Akhan 2005).

Hematomas account for approximately 9 % of peritransplant collections and usually occur during the immediate postoperative period, due to surgery. The other main causes are as follows: (1) early renal biopsy complicated by a cortical pseudoaneurysm, which may increase in size and subsequently rupture in the perirenal space, occurring immediately (24–48 h) or several weeks after biopsy, and (2) graft rupture secondary to severe acute rejection, which occurs in 3–6 % of renal transplants and during the first 2 weeks after transplantation. During the early postoperative period, hematomas are echogenic collections without flow on US images and are hyperattenuated on unenhanced CT images (Fig. 12). MRI is more specific, showing high intensity on both T1-weighted and T2W sequences (Neimatallah et al. 1999).

Arterial thrombosis is very unusual, occurring in <1 % of renal graft recipients, but is extremely severe, leading in most cases to graft loss. It occurs early, caused by a hypercoagulable state, hypotension, hyperacute rejection, immunosuppressive therapy, or a surgical complication: anastomotic occlusion, arterial dissection, renal artery kinking when it is too long, or torsion of the renal artery when implanted intraperitoneally. Its diagnosis is suspected soon after transplantation, when severe renal impairment is associated with anuria. Confirmation is easily obtained with color flow Doppler US (Fig. 13) showing arterial flow in the iliac artery; the absence of flow within the entire graft, which is swollen and hypoechoic; and a persistent “to-and-fro” flow pattern within the renal veins (Grenier et al. 1991, 1997). If confirmation is necessary, Gd-enhanced MRI examination can demonstrate the complete devascularization of the graft (Hélénon et al. 1992). Only rapid reintervention, within 12 h for surgical thrombectomy, can save the graft and does so in half of the cases. Percutaneous endovascular revascularization has been described for allograft salvage, but only for late thrombosis (Juvenois et al. 1999).

Segmental infarcts can also be due to segmental or reimplanted accessory renal artery thrombosis or be associated with acute rejection. They are usually asymptomatic and renal function impairment depends on their size. Doppler US is not specific, showing wedge-shaped areas of decreased or increased echogenicity without flow (Dodd et al. 1991; Grenier et al. 1997) (Fig. 14). Injection of ultrasound contrast agents may help differentiate with infection (Lefevre et al. 2002). Similarly, contrast-enhanced MR images demonstrate the absence of enhancement within the infarcted segments, except for the subcapsular cortex corticis (Neimatallah et al. 1999; Sebastia et al. 2001).

The frequency of renal artery stenosis (RAS) in transplanted kidneys varies from 1 to 23 %, and they may represent around 75 % of all posttransplant vascular complications (Bruno et al. 2004). It develops most often during the first year following surgery. Their origin is multifactorial: atherosclerotic plaque in the donor’s renal artery or in the recipient’s iliac artery; dissection, kinking, or twisting of the renal artery (due to excessive vessel length); malpositioning of the graft; flow turbulences generating intimal hyperplasia; graft perfusion catheter during cannulation causing intimal damage; and wall ischemia due to excessive dissection with destruction of the vasa vasorum. It is also possible that prolonged cold ischemia may play a role through ischemia–reperfusion injury (Halimi et al. 1999; Patel et al. 2001).

Severe hypertension with or without allograft dysfunction is the most frequent clinical symptom. Hypertension is a common feature in transplant recipients (up to 80 %). Therefore, RAS is suspected when hypertension develops suddenly, rapidly becomes more severe and resistant to medical therapy, and is associated with graft dysfunction without any other cause or when associated with an audible bruit over the graft (Palleschi et al. 1980; Rijksen et al. 1982). It may account for around 1–5 % of posttransplant hypertension.

US is able to detect renal artery stenoses (Figs. 15 and 16). Both velocity-profile changes, responsible for spectral broadening and perivascular color artifact, and systolic acceleration must be observed to make this diagnosis. A systolic velocity threshold of 190–200 cm/s (Grenier et al. 1991; Loubeyre et al. 1997) or a systolic velocity ratio between the renal and external iliac arteries of 1.5 (Loubeyre et al. 1997) has been proposed for significant stenoses. When the renal artery is too long, kinking is easily demonstrated on the color display, but only spectral sampling is able to confirm the presence of a stenosis (Fig. 17). As described for the native kidneys, intrarenal sampling of interlobar arteries and looking for dampened waveforms may help detect severe proximal stenosis (Gottlieb et al. 1995). However, these intrarenal features are less useful in transplanted kidneys because the proximal changes are more easily accessible than in the native kidneys.