Mm. Hanasono et al., Uses and limitations of FDG positron emission tomography in patients with head and neck cancer, LARYNGOSCOP, 109(6), 1999, pp. 880-885
Objective: Numerous authors have reported the potential usefulness of posit
ron emission tomography (PET), These studies have had conflicting results,
at least partly owing to limited sample sizes. The objective of this study
is to define not only the uses, but also the limitations of PET in patients
with head and neck cancer. Study Design: Nonrandomized, retrospective anal
ysis of PET at an academic institution. Methods: The authors performed 146
PET scans on 133 patients with head and neck cancer. Eighteen patients (19
PET scans) with thyroid disorders were excluded. A minimum 1 year of follow
-up was available in 84 patients, who were separated into groups based on w
hether the PET was used to detect unknown primary cancers (n = 20), stage n
eck nodal and distant metastases (n = 8), monitor response to nonsurgical t
herapy (n = 22), or detect recurrent or residual cancers (n = 34), The resu
lts of PET were compared with results from computed tomography (CT) and mag
netic resonance imaging (MRI) performed in the same patients. Results: Of t
he unknown primary cancers, PET correctly identified 7 of 20 primary sites,
giving a sensitivity of 35%. When combined with CT or MRI, the sensitivity
increased to 40%. When used for detection of metastatic disease, PET demon
strated five of five nodal metastases (100%) and two of four distant metast
ases (50%). in evaluating the response to nonsurgical therapy, PET had a se
nsitivity of 50% and a specificity of 83% for detecting tumor at the primar
y site and a sensitivity of 86% and a specificity of 73% for detecting noda
l disease. When used for evaluation of recurrent/residual disease, PET iden
tified seven of seven cases of local recurrences/residual disease and had a
specificity of 85%. PET also detected seven of seven cases of nodal diseas
e and had a specificity of 89%. Conclusions: For staging purposes, PET is l
imited by its lack of anatomic detail. However, PET compares favorably with
CT and MRI in detecting recurrent/residual cancers, PET imaging complement
s the more traditional imaging modalities (CT or MRI), especially for an un
known primary cancer.