Page 136 - Sentinel lymph node biopsy in oral cavity cancer - Inne J. den Toom

Chapter 7 The study of Haerle et al. showed all their additional SLNs in the same or adjacent levels as hotspots detected by planar lymphoscintigraphy alone and they suggest that even necks without hotpots should be explored with the gamma probe intraoperatively, based on the fact that the gamma probe identified SLNs in patients without hotpots on imaging.12 In contrast to their study we found 7 additional SLNs in a non-adjacent level or even in the other neck side compared to planar lymphoscintigraphy. However, we still found the (dynamic) planar lymphoscintigraphy of additional value in differentiating SLNs and second echelon nodes, especially using the criterion of rapidly emerging hot spots. Therefore we recommend a combination of planar static and dynamic imaging followed by SPECT-CT as the currently best imaging procedure for SLNB. We hypothesized that we could find additional SLNs due to SPECT-CT especially in patients with SLNs in close proximity of the primary tumour as is the case for SLNs in level I with a primary tumour in the floor of mouth. Indeed in 5 patients additional SLNs had been identified in level I, however 4 of these patients had a tongue tumour and only 1 a floor of mouth tumour. In addition, in 4 patients (2 tongue tumours, 2 floor of mouth tumours) a hot spot considered to be a SLN could be identified as injection spot rather than SLN in level I by SPECT-CT. We found a trend for more additional SLNs in floor of mouth tumours compared with tongue tumours, also resulting in a lower number needed to SPECT-CT (not presented). Despite our experience with SLNB in oral cancer, we report a relatively high number of false negative patients in this study. In this small cohort of 19 floor of mouth tumours 2 false negatives were present, compared to 2 false negatives in 39 tongue tumours. In 1 patient with a left-sided floor of mouth tumour the initially found SLN was located in level I on the right side, then this patient returned with a metastasis in level I on the left side 6 months after SLNB, which had been probably missed on the planar lymphoscintigraphy and SPECT-CT. The other false negative patient with a floor of mouth tumour had a regional metastasis in level III 13 months after SLNB. Both patients are alive with no evidence of disease for more than 2.5 years. Both patients with a tongue tumour and false negative SLNB had regional metastasis in level II ipsilateral, approximately 1 year after SLNB. One patient is alive with no evidence of disease for 3 years, one patient had been lost to follow-up. In our opinion SPECT-CT did not solve the problems of the lower accuracy in patients with floor of mouth tumours, despite the higher number of additional identified SLNs due to SPECT-CT. The finding that additional SLNs were mainly found in other 134