183 / 232
were almost all after initial enucleation (10/14). Median time to
recurrence in this group was 9.5 months (range 3
e
540 months).
Two had a recurrence in the cervical lymph nodes that was treated
with selective neck dissection. Neither of these patients had a neck
dissection as part of their initial surgery. One patient with rhab-
domyosarcoma developed lung metastases that were treated
curatively with chemotherapy and radiation but ultimately died of
treatment associated acute myeloid leukemia.
Mean length of follow up for patients with major salivary gland
malignancies was 13.5 years (range: 0.2
e
62.3). Two patients died of
major salivary malignancy making the overall disease speci
fi
c
survival 96%. Two patients died of acute myeloid leukemia thought
to be secondary to chemoradiation treatment received for their
salivary malignancy and two patients died of other causes.
4.4. Factors associated with recurrence
Information on factors associated with recurrence is presented in
Table 2
. There was evidence that adverse pathologic factors (extrac-
apsular spread, vascular invasion and or perineural spread), enucle-
ation or super
fi
cial versus total parotidectomy, and no neck dissection
increased the risk of recurrence. However, statistical signi
fi
cance was
only seen in patients who underwent enucleation versus total paro-
tidectomy (p value
¼
0.005) (
Fig. 1
). There was insuf
fi
cient evidence
that low versus high grade pathologic types, T stage, and addition of
adjuvant radiation was related to risk of recurrence.
4.5. Complications
Based on retrospective chart review, the most common
complication involved the facial nerve with 4 patients having
complete facial paralysis, and another 4 patients with limited
branch facial paralysis. Gustatory sweating (Frey's syndrome) was
recorded in 5 patients. Other complications related to surgery
included hypertrophic scar and major depression associated with
appearance from facial paralysis. Complications associated with
radiation included facial lymphedema, xerostomia, paresthesias,
external auditory canal stenosis and arrested mandibular growth
requiring reconstructive surgery. Two patients developed treat-
ment related acute myeloid leukemia.
4.6. Long term follow up survey data
An attempt was made to contact all 52 patients still alive at last
follow up. Ultimately, 13 patients or parents of patients could be
reached for a phone survey. Average follow up time for this cohort
was 28.7 years (range 2.1
e
62.3 years). One hundred percent re-
ported normal facial movement with no eye problems. One patient
reported facial twitching or spasm despite not having any facial
weakness after initial treatment. Over half (54%) described symp-
toms of Frey's syndrome. All of these patients reported that their
gustatory sweating symptoms never resolved and stated that the
effect on their quality of life was a
“
1
”
on a scale of 1
e
10.
Other reported long term side effects of treatment included
facial numbness, change in ear position, speech impairment, dif
fi
-
culty eating, chronic facial pain, need for long term feeding tube,
dif
fi
culty whistling/blowing, excessive scarring and drooling (n
¼
1
for all). Four patients that were treated with surgery alone reported
excessively dry mouth. One patient reported a recurrence 45 years
after being treated for mucoepidermoid carcinoma with surgery,
radiation and chemotherapy.
5. Minor gland
There were 4 cases of minor salivary gland cancer (2 low grade
mucoepidermoid, 1 high grade mucoepidermoid, 1 low grade
adenocarcinoma). Three out of 4 patients suffered a local recur-
rence and one of these had a cervical lymph node recurrence 3.5
years later. Two patients (high grade mucoepidermoid and
adenocarcinoma) developed metastases and both died of their
disease, which made an overall disease speci
fi
c survival of 50% for
this group. Average length of follow up in this cohort was 6.9 years.
6. Discussion
We present our single institution experience treating pediatric
salivary gland malignancies over a 62 year time period. Our series
con
fi
rms that mucoepidermoid carcinoma is the most common
histologic type in pediatric patients followed by acinic cell and
adenoid cystic carcinoma
[1,4,7]
. An average age at presentation of
14
e
15 years also appears to be consistent across studies
[1,7]
.
Table 2
Factors associated with locoregional recurrence of parotid and submandibular gland tumors.
Locoregional recurrence n/total n (%)
p-value
Low grade pathology
a
8/32 (25%)
High grade pathology
b
5/18 (28%)
T1/T2
10/35 (29%)
T3/T4
3/10 (30%)
Adverse pathologic factors
2/5 (40%)
0.6
No adverse pathologic factors
13/46 (28%)
Positive intraparotid LN
0/4 (0%)
No positive intraparotid LN
14/45 (31%)
Enucleation
7/11 (64%)
Super
fi
cial parotidectomy (SP)
2/5 (40%)
Total parotidectomy (TP)
5/32 (16%)
De
fi
nitive surgery
Enucleation vs. TP
0.005
Enucleation vs. SP
0.15
SP vs. TP
0.10
Neck dissection
3/22 (14%)
0.09
No neck dissection
12/30 (40%)
Adjuvant radiation
1/6 (17%)
0.57
No adjuvant radiation
14/46 (30%)
LN: lymph nodes.
a
Low grade mucoepidermoid, acinic cell, lymphoma.
b
Intermediate and high grade mucoepidermoid, adenoid cystic, rhabdomyosarcoma, high grade synovial cell sarcoma.
C.C. Cockerill et al. / International Journal of Pediatric Otorhinolaryngology 88 (2016) 1
e
6
161