Positron Emission Tomography/Computed Tomography Influences on the Management of Resectable Pancreatic Cancer and Its Cost-Effectiveness

To the Editor:

I would like to commend Heinrich and colleagues on their very fine article on the influence of positron emission tomography/computed tomography (PET/CT) on the management of pancreatic cancer.¹

However, I would like to raise several concerns regarding the reporting of the results of the paper, which I feel is biased in favor of the use of PET/CT. First, the authors suggest that the addition of PET/CT to the staging process is superior to contrast enhanced CT (ceCT) alone in the detection of distant metastasis as 5 patients missed by ceCT were detected via PET/CT. However, careful analysis of the study revealed that only 2 of these metastases were abdominal (1 hepatic and retroperitoneal nodes, 1 abdominal wall). The other metastases were in the lungs (n = 2) and cervical lymph nodes, which were not within the field of the abdominal ceCT. Hence, this raises the question of whether the addition of ceCT of the thorax (instead of abdomen alone) would be more cost-effective than PET/CT. Additionally, it is important to note that the abdominal wall metastasis was actually seen clearly on ceCT retrospectively, although it was missed initially. Second, the authors suggest that PET/CT can detect synchronous neoplasms (2 colorectal cancers in the study). Once again, this point is debatable as both cancers were actually demonstrated but missed on ceCT. These neoplasms would probably also have been detected intraoperatively if a complete and thorough laparotomy was performed during pancreatic resection.

Finally and most importantly, regarding the cost-effectiveness of PET/CT, the authors calculated the cost savings based on the assumption that the 5 patients with distant metastasis would have avoided the operative and hospitalization costs of a pancreatic resection. However, this is probably an overestimation. For example, the patient with intra-abdominal metastasis would probably not have undergone a pancreatic resection, even without a preoperative PET/CT, as the liver metastasis would have been detected during laparoscopy and the high costs and long hospitalization associated with pancreatic resection would than be avoided (diagnostic laparoscopy is probably more sensitive that PET/CT, and this is supported by the author's findings whereby 2 patients with small liver metastasis were missed by PET/CT but detected at laparoscopy).

Hence, although I agree with the authors that the addition of PET/CT to standard staging tests will probably have a positive impact on the accurate diagnosis² and staging of pancreatic carcinoma, the cost/benefit for the routine use of this modality remains unproven.

Brian K. P. Goh, MBBS, MRCS, MMed
Department of Surgery
Singapore General Hospital
Singapore
bsgkp@hotmail.com