Objectives: To compare different surgical interventions for the treatment o
f extensive cervicofacial lymphangiomas and to define the minimal extent of
surgery necessary to control disease.
Design: Retrospective study. Mean +/- SD follow-up was 31 +/- 4 months afte
r surgery. Surgical procedures were grouped as follows: (1) total removal,
(2) subtotal removal (all cystic structures removed, small plaques of cyst
walls left attached to vital structures), (3) partial removal (major cysts
removed, some partially resected cystic structures left in place), and (4)
incision and aspiration with subsequent compression bandage. Control of dis
ease was defined as no recurrent or residual tumor or as recurrent or resid
ual tumor less than 10% of initial tumor size without evidence of growth on
several postoperative examinations and without clinical symptoms or aesthe
tic disfigurement.
Patients: Twenty-one patients with cervicofacial lymphangiomas (>3 cm in ma
ximum diameter) without thoracic involvement were evaluated. Fifteen patien
ts were 6 years or younger and 6 were older than 6 years. No surgery was ye
t performed in 3 patients, for a total of 24 surgical interventions in 18 p
atients.
Setting: Hospitalized care in 2 referral centers.
Results: After total removal, disease was controlled in 5 of 5 cases; after
subtotal removal, in 8 of 9 cases; after partial removal, in 1 of 7 cases;
and after incision and aspiration with subsequent compression bandage, in
0 of 3 cases. Two complications were encountered-1 fully reversible paresis
of the marginal branch of the facial nerve and 1 secondary healing.
Conclusions: Surgical removal of cervicofacial lymphangiomas is a safe trea
tment modality. Disease control can be achieved if all cystic structures ar
e removed. Small plaques of cyst walls attached to vital structures may be
left in place. If small cystic extensions of lymphangiomas are only opened
and left in place or if lymphangiomas are only drained following compressio
n bandage, symptomatic residual tumor or recurrence is frequent.