This retrospective single-institution study was approved by the Committee on Human Research at the University of California, San Francisco, and was compliant with the Health Insurance Portability and Accountability Act. Informed consent was waived. We performed a computerized search of our radiology information system (IDXrad software, version 9.7.1, IDX Systems) to identify the first 100 consecutive patients who had undergone PET/CT at our institution during March 2006. Each patient underwent one study, for a total of 100 PET/CT examinations.

The study population consisted of 58 women and 42 men with a mean age of 55 years (range, 18-84 years). The indications for PET/CT were staging or surveillance of breast cancer (n = 29), lymphoma (n = 15), lung cancer (n = 14), melanoma (n =12), head and neck cancer (n = 8), ovarian cancer (n =7), colorectal cancer (n = 6), sarcoma (n = 2), pheochromocytoma (n = 3), esophageal cancer (n =2), bladder cancer (n = 1), and nerve sheath tumor (n =1).

All studies were performed on a PET/CT scanner (Biograph 16, Siemens Medical Solutions) that is a hybrid PET and 16-MDCT unit. CT was performed first at 152 mAs, 120 kV, and 0.75-mm collimation. Images were obtained from the top of the patient's head to the feet and were reconstructed as contiguous 5-mm slices. In 86 patients, 150 mL of IV iohexol (Omnipaque 350, Nycomed Amersham) was administered 70-80 seconds before CT. The contrast material was injected at a rate of 3 mL/s and followed by a 50-mL saline flush. PET was performed immediately after CT without repositioning the patient, and imaging commenced 60 ± 15 minutes (mean ± SD) after the injection of 12.5 ± 2.5 mCi (462.5 ± 92.5 MBq) of FDG. PET images were obtained at 7-10 stations per patient, with an acquisition time of 4 minutes per station, from the top of the patient's head to the feet. In other words, the PET images of the pelvis were obtained approximately 25-40 minutes after the injection of IV CT contrast material. CT data were used for attenuation correction, and results were reconstructed into sagittal, coronal, and axial planes. In the other 14 patients, no iodinated contrast material was given because of renal insufficiency (n = 11) or suspected pheochromocytoma (n = 3). Patients were instructed to remain still throughout the entire examination.

A single observer who was unaware of whether IV contrast material had been administered reviewed only the PET images on a dedicated image-processing workstation (Leonardo, Siemens Medical Solutions) and recorded the presence or absence of anterior layering of excreted FDG in the bladder. All studies were viewed with and without attenuation correction. Anterior layering of FDG in the urinary bladder was considered to be present when a fluid-fluid level was seen in the expected position of the urinary bladder with relative photopenia in the posterior bladder. Similar assessment was made for the renal pelvises. Any abnormal focal uptake in the urinary tract was also noted. Each patient's medical records were reviewed for the presence or absence of bladder abnormalities, and these findings were later correlated with the PET findings.

To further elucidate the cause of anterior layering of excreted FDG in the bladder, we performed in vitro PET/CT of specially constructed phantoms designed to mimic the content of the bladder during a PET/CT examination. The first phantom consisted of a 1-L radiolucent bottle containing an agitated mixture of 400 mL of diluted iodinated contrast material and 200 mL of diluted FDG. The diluted solution of FDG consisted of 250-μCi (9,250 kBq) FDG in 200 mL of isotonic saline. The diluted solution of iodinated contrast material consisted of 360 mL of isotonic saline mixed with 40 mL of iohexol (Omnipaque 350).

For the second phantom, a separate 1,000-mL bottle was filled with 100 mL of 125-μCi (4,625 kBq) FDG diluted in saline and then 100 mL of the same diluted iodinated contrast material was manually injected, using a catheter-tip syringe, into the dependent portion of the bottle over a period of 1 minute. We chose to inject the iodinated contrast material into the dependent portion of the bottle because at our institution CT contrast material is given after FDG and because the ureters insert into the posterior (dependent) portion of the bladder. PET/CT images of both phantoms were obtained at 0, 5, and 10 minutes after agitation using the same imaging parameters as described for our clinical in vivo studies.

Anterior layering of excreted FDG in the bladder was seen on 61 of 86 studies (71%) performed with IV iodinated contrast material but was not seen on any of the 14 studies performed without IV contrast material (Fig. 2A, 2B). This difference was highly significant (p < 0.001). Attenuation correction did not affect the appearance of anterior layering of FDG in the urinary bladder in any patient. In one patient, FDG uptake in a malignant posteriorly located bladder mass was unmasked by the anterior layering phenomenon (Fig. 3A, 3B, 3C). This mass was later biopsied at cystoscopy, with a final pathologic diagnosis of intestinal-type adenocarcinoma.

Imaging of the agitated phantom containing diluted FDG and iodinated contrast material showed the gradual development of a separation gradient, with the FDG activity becoming gradually localized to the nondependent portion of the phantom (Fig. 4A, 4B). Images of the nonagitated phantom revealed a sharp fluid-fluid level with no evidence of mixing on delayed imaging (Fig. 5).