Does SPECT Provide Incremental Value to CT or MRI in Assessing Mandibular Invasion by Oral Squamous Cell Carcinoma?

PURPOSE

Single photon emission computed tomography (SPECT) has been commonly used as an additional method to assess mandibular bone invasion in patients with oral squamous cell carcinoma (OSCC). In the present study, we measured the diagnostic validity of SPECT findings compared with the histologic findings.

PATIENTS AND METHODS

We implemented a retrospective cross-sectional study and enrolled a sample of patients with OSCC adjacent to the mandible. The staging examinations included magnetic resonance imaging (MRI) and/or computed tomography (CT) and additional SPECT. The patients' medical records and imaging data were reviewed by 2 readers, and bone invasion was classified as positive or negative for each diagnostic method. The predictor was bone invasion found on CT and/or MRI compared with the combination of CT and/or MRI with either positive or negative SPECT results. The primary outcome variable was histologic bone invasion. Other variables of interest were clinicopathologic data, type of mandibular resection, and resection margin status. Bivariate tests were used to compare the diagnostic specificity, sensitivity, and accuracy of each imaging modality.

RESULTS

The study enrolled 122 patients, with a mean age of 67 years (55% male; study period, January 2010 to December 2017). In 60 patients (49%), segmental mandibular resection was performed. The sensitivity and specificity for bone invasion were 100% and 39% for SPECT and 84% and 75% for CT and/or MRI. The negative results for bone invasion using SPECT did not match the false-positive CT and/or MRI results in 5% (n = 6 of 122). Positive SPECT findings, in addition to positive CT and/or MRI findings increased the sensitivity to 100% but decreased the specificity to 29% (P < .001). Considering only negative SPECT findings increased the specificity to 85% (P = .03), sensitivity remaining unaffected. The accuracy of CT and/or MRI alone was 80% and was 84% when combining CT and/or MRI and negative SPECT findings (P = .03).

CONCLUSIONS

In oncologic staging with CT and/or MRI, the addition of SPECT provided only small benefits. Only negative SPECT results allowed for greater specificity and accuracy. The use of SPECT could be considered to rule out bone invasion in cases of radiologic uncertainty of positive CT or MRI findings.