Purpose: This retrospective study was undertaken to assess the clinical fea
tures and results of treatment of carcinomas of the ethmoid sinus.
Materials and methods: The records of 34 patients with ethmoid sinus carcin
omas treated with curative intent at the U.T.M.D. Anderson Cancer Center (U
TMDACC) between January 1969 and December 1993 were reviewed. The age of th
e patients ranged from 28 to 73 years with a median of 57 years. There were
28 Whites, four Hispanics, one Black and one Asian. A simple staging based
on anatomical criteria was used to describe the extent of the disease. Six
patients had T1, 13 patients had T2 and 15 patients had T3 disease. Twenty
-one patients were treated with surgery plus radiation and 13 patients were
treated with radiotherapy alone; nine patients received adjuvant chemother
apy. Radiation was given at -2 Gy per fraction to total doses of 50 Gy preo
peratively, 52-66 Gy (median 60 Gy) postoperatively and 50-70 Gy (median 63
Gy) when no surgery was performed.
Results: The actuarial 5-year overall, disease-free and disease-specific su
rvival rates were 55%, 58% and 63%, respectively. The actuarial 5-year loca
l control rate was 71% for the whole group (74% for surgery plus radiation
and 64% for radiation alone). Local recurrence occurred in nine patients, n
odal relapse occurred in three patients and distant metastases occurred in
four patients. Histologically proven dura mater invasion was associated wit
h a poorer local control rate in patients undergoing surgery and radiation.
The simple T-staging system used in this study was a good discriminator fo
r local control. Of nine patients receiving chemotherapy, three had complet
e responses and four had partial responses; six of the seven responders had
undifferentiated carcinoma. Severe complications of therapy occurred in pa
tients treated between 1969 and 1984 and consisted mainly of visual impairm
ent and brain necrosis.
Conclusions: This retrospective review of a large single institutional expe
rience showed that ethmoid sinus carcinomas have a tendency for extensive l
ocal invasion but a low propensity for lymphatic and hematogenous spread. H
ence, local recurrence was the main cause of cancer-related death. Combined
treatment with surgery and postoperative irradiation yielded the highest l
ocal control rate. However, radiotherapy alone eradicated two-thirds of pri
mary tumors and, consequently, is a reasonable alternative treatment for pa
tients with medical contraindications to surgery. For patients who underwen
t surgery and radiotherapy, the presence of histologically proven dura mate
r invasion was associated with a higher local recurrence rate. Severe radia
tion complications have been rare with the contemporary radiotherapy techni
que. Chemotherapy induced excellent responses in undifferentiated carcinoma
but its impact on overall disease control is unclear in this small series
of patients. (C) 1998 Elsevier Science Ireland Ltd. All rights reserved.