Figure 1. Hypermetabolic lesions are present in the liver and mid-abdomen. The kidneys are malpositioned inferiorly, with their lower poles directed medially.

Figure 2. There is a horseshoe kidney. At the isthmus, there is a heterogeneous hypermetabolic lesion (SUV_max 7.2). In the left moiety, there is a large, well-defined non-hypermetabolic cystic lesion, with simple features.

Figure 3. In segment VI of the liver, there is a rim-enhancing lesion with central low density that causes minor capsular retraction. The lesion has rim-hypermetabolic activity (SUV_max 17.4). The background liver metabolic activity is SUV_max 1.4.

Figure 4. The hypermetabolic lesion in the horseshoe kidney is seen, as is the hypermetabolic hepatic lesion.

Renal cell carcinomas (RCC) arise from tubular epithelium and are thus adenocarcinomas. The classic clinical triad of costovertebral pain, palpable mass and hematuria, is only seen in 10%, and did not occur in this case because of the occurrence in a horseshoe kidney. Nevertheless, hematuria is the most reliable sign, as occurred in our patient. Tumours may otherwise be diagnosed due to presentation with constitutional symptoms, incidental finding on a radiologic investigation, or due to a paraneoplastic syndrome ascribed to abnormal hormone production, including polycythemia, hypercalcemia, hypertension, hepatic dysfunction, feminization or masculinization, Cushing syndrome, eosinophilia, leukemoid reactions and amyloidosis. Polycythemia was indeed present in this patient, though was only retrospectively attributed to the RCC.

A horseshoe kidney is the most common fusion anomaly of kidneys, and is found in about 1 in 500 to 1000 autopsies. Ninety per cent of such kidneys are fused at the lower pole- as in our case- with a parenchymal or fibrous isthmus, and 10% are fused at the upper pole. The incidence of RCC amongst horseshoe kidneys has not been definitely proven to be greater than amongst the normal population.

A relatively high false negative rate has been reported for RCC using FDG-PET, likely secondary to variability of glucometabolism in RCC due to the heterogeneous expression of GLUT-1 (glucose transporter-1) in these tumours. No significant correlation has been shown between expression for GLUT-1 and tumour grade or extent. As FDG is excreted in the urine, urine activity may mask the appearance of renal pelvis RCCs. In this patient however, FDG uptake was observed, with morphologic clarification of occurrence in a horseshoe kidney made possible with the fused CT.

The simultaneous diagnosis of cholangiocarcinoma in this patient is likely coincidental, with no association reported. Indeed in this patient, the liver tumour was initially considered radiologically to be metastatic from the RCC. Microscopically, cholangiocarcinomas resemble adenocarcinomas arising in other parts of the body, though are usually markedly desmoplastic, retrospectively seen in this patient as a minor degree of capsular retraction.