Accurate assessment of the extent of local recurrence as well as of distant disease (if any) is of paramount importance in accurate restaging of the disease and in planning the optimal therapy in any cancer. PET and PET/CT have been effectively used in a variety of cancers for primary staging and restaging of disease after therapy. In tumors of the urinary tract, FDG-PET has been used to accurately assess the metastatic spread but has seldom been used to image the primary/recurrent lesions in the urinary tract, primarily due to interference by high FDG concentrations in the urine.
FDG undergoes glomerular filtration similar to glucose, but it is not reabsorbed in the tubules and is largely excreted in urine, producing high activity in the urinary tract.[9
] Therefore, for quite some time, FDG-PET has been considered to be of limited value in the detection of urinary tract cancers and perivesical lymph nodes.[10
] Washing out the excreted FDG is the key to overcoming this limitation of PET imaging.
Historically, attempts have been made to improve the sensitivity of PET by achieving complete bladder washout using furosemide injection before image acquisition or by retrograde bladder irrigation; however, these techniques have met with limited success.[14
] Investigators who used retrograde saline irrigation of the urinary bladder to remove FDG radioactivity failed to reduce tracer activity to background levels. Using retrograde saline irrigation of the bladder while imaging patients with FDG-PET, Kosuda et al
. reported a 40% false-negative rate for the detection of recurrent or residual tumor in the bladder.[14
] It was noted that continuous bladder irrigation and immediate postvoid images were not effective in reducing intravesical activity because the FDG-laden urine excreted by the kidneys kept trickling into the bladder. Cook et al
. opine that although catheterization and drainage reduces urinary bladder activity, it still leaves small amounts of concentrated urine, which may resemble hypermetabolic lesions and be misinterpreted as such.[16
] Also, bladder catheterization and retrograde filling of the bladder are invasive procedures and, in addition to increasing the risk of urinary infection in the patient, continuous bladder irrigation is known to increase radiation to the staff.[17
High uptake of FDG in cancerous lesions of the bladder was first demonstrated by Harney et al
. in rats.[19
] Drieskens et al
. used FDG-PET for preoperative staging of invasive bladder cancer and concluded that the metabolism-based information provided by FDG-PET when added to the information from CT yielded a high diagnostic and prognostic accuracy.[20
] In their study, nine of 40 patients had discordance between the findings of PET/CT and that of CT alone. Of these, PET was proved to be correct in two-thirds (six of nine). They opined that, based on the presence of nodal/distant metastases detected by PET, the median survival of patients with invasive bladder cancer could be predicted. Studies have demonstrated hypermetabolism in TCC of the upper urinary tract and suggest that FDG-PET imaging may be useful in detecting primary and metastatic lesions of renal pelvic TCC as well.[21
For follow-up and restaging of invasive bladder cancer, traditionally, a combination of cystoscopy and CT abdomen has been used. While the former determines the extent of intravesical disease, the latter is used to assess perivesical spread and detect nodal or distant metastases. It is a known fact that CT may be false-negative for early recurrence in the bladder wall as well as for metastases in normal-sized pelvic/perivesical nodes. Hence, we analyzed the utility of composite PET/CT (performed with forced diuresis and a dual-phase protocol), as compared with CT alone, for restaging of invasive bladder cancers.
The primary objective of an ideal staging modality is to achieve adequate visualization of the cancerous lesion at the primary site (the urinary tract, in the context of this study), ascertain the extent of locoregional disease and detect distant metastases if any. All these objectives were adequately achieved in our cases by the use of a dual-phase PET/CT technique, with forced diuresis and oral hydration. Initial whole-body PET/CT scan served the purpose of detecting (or ruling out) distant metastases and a postdiuretic delayed scan of the region of interest helped assess the bladder for local disease and the perivesical/pelvic region for regional nodal metastases.
In the dual-phase technique, the timing of the furosemide injection is important for obtaining satisfactory urinary dilution. Injecting a diuretic before the standard PET/CT acquisition (i.e., during the tracer uptake phase of the first 45-60 min) gives poor results as the FDG circulating in blood in high concentrations gets excreted in the urine instead of getting concentrated in the tissues. This prevents adequate dilution of urine, negating the effect of furosemide and, also, the rapid filling of the bladder leads to patient discomfort and probable patient movement during scanning. In our cases, we injected the diuretic 60-90 min after radiotracer injection and then rescanned 1-1½ h later. This technique provided excellent urinary radiotracer washout and was successful in demonstrating recurrent tumors in the bladder wall, with good lesion-to-background contrast. Bladder activity was reduced to background levels in 21 of 22 bladder-preserved patients. In patients with cystectomy and urinary diversions, adequate urinary dilution could not be obtained because increased capacity and incomplete evacuation of bladder diversions (ileal conduits) led to retention of FDG-laden urine. However, this did not appear to affect the overall assessment of the patient by FDG-PET/CT for recurrent local disease, nodal metastases or metachronous upper tract lesions.
It has been reported that iterative reconstruction algorithms provide better image quality than filtered backprojection algorithms, which can have streak artefacts.[22
] In the present study, all images were constructed iteratively and no streak artefacts were observed.
Our results show that delayed FDG-PET/CT images after diuretic administration and oral hydration can demonstrate hypermetabolic lesions (representing cancer) in the urinary bladder as well as in the perivesical nodes with reasonably high clarity. Detection of locally recurrent or residual bladder tumors was significantly improved with this technique. Up to 59% of patients without cystectomy had more accurate local staging with postdiuretic dual-phase FDG-PET/CT. Delayed pelvic images after furosemide and oral hydration helped to detect pelvic lymph node metastases in two cases. These lesions would have otherwise been missed. We did not come across false-positive PET/ CT results (i.e., bladder wall uptake due to inflammatory reaction) in any of the cases who had undergone cystoscopic biopsy or transurethral resection before PET/CT. Probably the 3-month interval after the primary tumor resection allowed healing of any possible inflammatory reaction.
As a composite metabolic and anatomic diagnostic tool, PET/CT with FDG has the ability to overcome the limitations of CT and stand-alone PET.[23
] We found fused PET/CT images useful for precisely locating extravesical hypermetabolic lesions. This eliminated possible false-positives that could arise from focal FDG concentration in the bowel or ureters. CT images accurately demonstrated true wall thickening in the (distended) bladder when this was present. Noncontrast imaging with a distended bladder was helpful for avoiding artifactual thickening of the walls. On the other hand, PET images accurately detected early recurrence at seven bladder sites, where the corresponding CT images did not reveal any wall thickening.
However, false-negative PET/CT results in patients with superficial recurrent lesions (confirmed on cystoscopic biopsy) were still encountered in two patients. This finding re-emphasizes the pivotal and indispensable role of cystoscopy in the follow-up of bladder cancer patients with preserved bladders. Through this study, we intend to highlight that once local recurrence in the bladder is detected on a follow-up cystoscopy, FDG-PET/CT (with additional postdiuretic images), instead of CT alone, may be the investigation of choice for restaging because it can accurately detect perivesical and distant disease as well as local recurrence in the bladder wall. PET/CT can often depict multifocality of bladder lesions and thus help the surgeon provide optimal therapy.
Our results are concordant with the findings of other studies conducted on similar lines. Anjos et al
. have reported that PET/CT images after furosemide and oral hydration were able to overcome the difficulties posed by the urinary excretion of FDG and that detection of locally recurrent or residual bladder tumors was dramatically improved by this technique.[24
] Kamel et al
. documented improved accuracy of FDG-PET imaging in abdominopelvic malignancies when it was combined with forced diuresis.[25
The limitation of the study is its retrospective nature. Some amount of selection bias may have been present as it is likely that only those patients of bladder cancer suspected to have recurrence were referred for PET/CT.