You are on page 1of 11

A study of persistent unilateral middle

ear effusion caused by occult skull base


lesions
| Reprints

April 18, 2013


by John P. Leonetti, MD

Abstract
The goal of this article is to review a series of patients with persistent
unilateral middle ear effusion (MEE) and to suggest a more contemporary
diagnostic algorithm. The author conducted a retrospective chart review
of adults with persistent unilateral MEE and normal findings on physical
and nasopharyngoscopic examinations whose MEE was eventually found
to be caused by a variety of occult skull base lesions. The study
population was made up of 79 patients-52 women and 27 men, aged 21 to
83 (mean: 54.8) at presentation-who had been referred to an academic
tertiary care medical center between July 1, 1988, and June 30, 2008.
Follow-up ranged from 9 months to 19.5 years (mean: 8.7 yr). Of this
group, 50 patients (63.3%) had a malignant tumor, 26 (32.9%) had a
benign tumor, and 3 (3.8%) had an internal carotid artery aneurysm.
Eustachian tube occlusion had been caused by diffuse invasion in 33
patients (41.8%), by intracranial pathology in 24 (30.4%), and by
extracranial-infratemporal lesions in 22 (27.8%). Nasopharyngoscopy
cannot identify a variety of rare skull base lesions that cause eustachian
tube compression or tissue invasion that ultimately leads to MEE.
Therefore, patients with unexplained persistent unilateral MEE should
undergo coronal magnetic resonance imaging or computed tomography to
look for any intra- or extracranial lesions before undergoing ventilation
tube placement.

Introduction
Secretory otitis media is a common cause of aural fullness and hearing
loss secondary to eustachian tube inflammation, edema, or
obstruction.1 Some cases of middle ear effusion (MEE) are self-limited
while others require medical management in the form of a topical steroid
nasal spray, an oral steroid, a decongestant, an antihistamine, an
antibiotic, or allergy therapy.2-4
In addition to a complete head and neck examination, all patients who
present with persistent unilateral MEE should undergo nasopharyngoscopy
to look for congenital narrowing, acquired stenosis, or neoplastic
obstruction of the eustachian tube orifice.5 When findings on both
examinations are normal, many adult patients undergo additional medical
management or ventilation tube insertion for presumed MEE induced by
eustachian tube dysfunction.6 Although such a treatment strategy is safe

and successful in patients with typical recurrent MEE, it might not detect
occult skull base lesions that compress or invade the eustachian tube.7
In this article, the author describes his review of a series of MEE patients
with occult skull base lesions.

Patients and Methods


Upon obtaining Institutional Review Board approval, the author
retrospectively reviewed the charts of 79 adults (19 yr) with persistent
unilateral MEE whose findings on initial head and neck examination and
nasopharyngoscopy were normal. The 79 patients were made up of 52
women and 27 men, aged 21 to 83 (mean: 54.8) at presentation.
All 79 patients had been referred to the author from community
otolaryngologists or primary care physicians 5 to 51 months after their
initial presentation (mean: 37.2). They were evaluated at the author's
tertiary care academic medical center between July 1, 1988, and June 30,
2008.
All patients were subsequently found to have had an occult skull base
lesion that was believed to have caused the MEE by obstructing or
invading the eustachian tube. Follow-up ranged from 9 months to 19.5
years (mean: 8.7 yr).
Between 1988 and 1996 at the author's institution, the diagnostic
algorithm for unilateral MEE with a normal head and neck examination
included contrast-enhanced paranasal sinus computed tomography (CT).
Since 1996, we have employed a paranasal sinus magnetic resonance
imaging (MRI) series with and without contrast; in the past 6 years, we
have minimized the MRI to only coronal sections with and without
contrast. For patients who cannot undergo MRI, we obtain a coronal sinus
CT. Our overall diagnostic algorithm is shown in figure 1.
Figure 1. This algorithm provides a diagnostic and treatment guide for
the treatment of adults who do not respond to the medical
management of unilateral ME

Results
Presenting symptoms and signs. All 79 patients had complained of
aural fullness and hearing loss. Other presenting otologic symptoms
included tinnitus, otalgia, and dizziness (table 1).
Table 1. Presenting symptoms and signs (N = 79)

Finding

n (%)

Symptoms
Aural fullness

79 (100)

Hearing loss

79 (100)

Tinnitus

37 (46.8)

Otalgia

24 (30.4)

Dizziness

5 (6.3)

Signs
Middle ear fluid

79 (100)

Pars tensa retraction

26 (32.9)

Trismus

19 (24.1)

Finding

n (%)

Facial hypoesthesia

12 (15.2)

Cranial nerve VI palsy

3 (3.8)

Vocal fold paralysis

1 (1.3)

All patients had presented with serous MEE, and almost one-third of them
demonstrated a pars tensa retraction. Other presenting signs included
trismus, facial hypoesthesia, cranial nerve VI palsy, and vocal fold
paralysis (table 1).
History. Of the 79 patients, 61 (77.2%) had undergone at least one
myringotomy with ventilation tube insertion, which had delayed their
diagnosis by a mean of more than 3 years; roughly half of these patients
had undergone more than one such procedure. Only 23 of the 79 patients
(29.1%) had undergone any radiographic studies prior to referral. These
investigations included CT of the sinuses in 6 patients, CT of the temporal
bones in 5, CT of the temporomandibular joint in 5, MRI of the head in 4,
and a panoramic x-ray of the mandible in 3.
Site of Origin. Of the 79 previously undetected skull base lesions, 42
were located on the right and 37 on the left. The most common sites of
origin were the infratemporal fossa and the middle cranial fossa. The site
of origin could not be determined in 19 patients because of the extent of
the lesion (table 2).
Table 2. Site of lesion origin (N = 79)

Site of origin

n (%)

Infratemporal fossa

21 (26.6)

Middle cranial fossa

19 (24.1)

Clivus

12 (15.2)

Petrous apex

8 (10.1)

Not determined (diffuse)

19 (24.1)

Site of origin

n (%)

Total

79 (100)

Eustachian tube occlusion had been caused by diffuse invasion in 33


patients (41.8%), by intracranial pathology in 24 patients (30.4%), and by
extracranial-infratemporal lesions in 22 patients (27.8%).

Treatment approach. A total of 76 patients (96.2%) had either a


malignant (n = 50 [63.3%]) or benign (n = 26 [32.9%]) tumor, and they
underwent a surgical approach for either biopsy or resection (table 3). The
remaining 3 patients (3.8%) had a giant aneurysm of the petrous segment
of the internal carotid artery (figure 2), and they underwent cerebral
angiography with intraluminal stenting.
Table 3. Diagnostic (biopsy) or treatment approach (N = 79)

Approach

n (%)

* ITF = infratemporal fossa.


Surgical
Preauricular ITF*

37 (46.8)

Orbitocranial zygomatic

22 (27.8)

Pterional

12 (15.2)

Middle cranial fossa

5 (6.3)

Endovascular
Angiography w/ stenting

3 (3.8)

Approach
Total

n (%)
79 (100)

Figure 2. Coronal T1-weighted MRI shows a giant aneurysm of the


petrous segment of the internal carotid artery.

Tumor histology. Of the 50 malignant tumors, 20 were undifferentiated


carcinomas that had likely originated in the submucosal lateral
nasopharynx (figure 3). Twelve patients had a metastasis from the kidney,
prostate, lung, or esophagus. The most common of the 26 benign tumors
were trigeminal neuromas (figure 4, A) and meningiomas (figure 4, B).
Figure 3. Axial MRI shows a diffuse, undifferentiated carcinoma of the
skull base

Figure 4. These coronal MRIs show an infratemporal trigeminal neuroma


(A) and a transcranial meningioma (B) of the skull base.

Additional treatment. Of the 50 patients with a malignant tumor, 19


underwent wide-field skull base radiotherapy, and 30 received
chemotherapy with radiotherapy; 1 patient refused additional treatment.
Sixteen of the 19 patients who underwent radiotherapy had a parotid
gland malignancy, and the other 3 had an undifferentiated carcinoma.
Of the 26 patients with a benign tumor, stereotactic radiotherapy was
delivered to 2 who had a trigeminal neuroma and to 4 who had a
subtotally resected meningioma. The remaining 20 patients underwent
surgical resection or drainage alone.

All 3 patients with an internal carotid artery aneurysm were managed with
cerebral angiography and endovascular stenting.

Outcomes. Thirty-four of the 50 patients with a malignant tumor (68.0%)


died of their disease, and 10 (20.0%) were alive but with active disease at
study's end. None of the 26 patients with a benign tumor had died, and 17
of them (65.4%) had no evidence of disease. Likewise, all 3 patients with
an aneurysm were alive and well after treatment (table 4).
Table 4. Outcomes according to the type of lesion (N = 79) during
follow-up of 9 months to 19.5 years (mean: 8.7 yr)

n (%)
Type of lesion

Alive
w/o
disease

Alive
with
disease

Died of
disease

Died
other
causes

Total

2 (2.5)

17
(21.5)

1 (1.3)

20
(25.3
)

Key: Ca = carcinoma;
ICA = internal carotid
artery.
Malignant tumors (n =
50)
Undifferentiated Ca

n (%)
Type of lesion

Alive
w/o
disease

Alive
with
disease

Died of
disease

Died
other
causes

Total

Metastasis

1 (1.3)

11
(13.9)

12
(15.2
)

Adenoid cystic Ca

4 (5.1)

3 (3.8)

2 (2.5)

9
(11.4
)

Mucoepidermoid Ca

1 (1.3)

4 (5.1)

2 (2.5)

7
(8.9)

Osteosarcoma

2 (2.5)

0 2 (2.5)

Trigeminal neuroma

8 (10.1)

2 (2.5)

1 (1.3)

11
(13.9
)

Meningioma

3 (3.8)

4 (5.1)

1 (1.3)

8
(10.1
)

Chordoma

1 (1.3)

1 (1.3)

2
(2.5)

Cholesterol granuloma

2 (2.5)

2
(2.5)

Pleomorphic adenoma

2 (2.5)

2
(2.5)

Benign tumors (n =
26)

n (%)
Type of lesion

Alive
w/o
disease

Alive
with
disease

Died of
disease

Died
other
causes

Total

Cholesteatoma

1 (1.3)

1
(1.3)

ICA aneurysm (n = 3)

3 (3.8)

3
(3.8)

Total

25
(31.6)

17
(21.5)

34
(43.0)

3 (3.8)

79
(100)

Discussion
Aural fullness secondary to MEE is most often associated with eustachian
tube dysfunction.1 Long-term eustachian tube failure may lead to
negative middle ear pressure, tympanic membrane atelectasis, adhesive
middle ear disease, or cholesteatoma.8 In the adult population, unilateral
MEE may result from eustachian tube compression or invasion rather than
from physiologic dysfunction. Gacek identified the four levels of potential
eustachian tube compromise as being (1) the lumen of the eustachian
tube, (2) the nasopharynx, (3) the infratemporal fossa, and (4) the petrous
apex.9 Of these four locations, only the nasopharynx is accessible to
physical examination.
The initial evaluation of any patient with MEE begins with a complete
history of symptoms, including the duration of aural fullness, similar
episodes in the past, recent airplane travel or other causes of barotrauma,
blunt temporal bone injury, sinus disease or surgery, radiotherapy to the
head and neck, tobacco use, cleft palate, gastroesophageal reflux, or
sinonasal allergic disease.5

References

1.

Silverstein H, Miller GF Jr., Lindeman RC. Eustachian tube dysfunction as a


cause for chronic secretory otitis in children. (Correction by pressureequalization). Laryngoscope 1966; 76 (2): 259-73.

2.

Cantekin EI, Bluestone CD, Rockette HE, Beery QC. Effect of decongestant
with or without antihistamine on eustachian tube function. Ann Otol Rhinol
Laryngol Suppl 1980; 89 (3 Pt 2): 290-5.

3.

Tracy JM, Demain JG, Hoffman KM, Goetz DW. Intranasal bec-lomethasone as
an adjunct to treatment of chronic middle ear effusion. Ann Allergy Asthma
Immunol 1998; 80 (2): 198-206.

4.

Silverstein H, Light JP, Jackson LE, et al. Direct application of dexamethasone


for the treatment of chronic eustachian tube dysfunction. Ear Nose Throat J 2003;
82 (1): 28-32.

5.

Seibert JW, Danner CJ. Eustachian tube function and the middle ear.
Otolaryngol Clin North Am 2006; 39 (6): 1221-35.

6.

Knight LC, Hilger A. The effects of grommet insertion on Eustachian tube


function. Clin Otolaryngol Allied Sci 1993; 18 (6): 459-61.

7.

Uhlmann C, Krueger GR, Sesterhenn K, et al. Nasopharyngeal and adjacent


neoplasms: A clinic-pathologic and immunologic study. Arch Otorhinolaryngol
1978; 218 (3-4): 163-77.

8.

Bluestone CD, Cantekin EI, Beery QC, Stool SE. Function of the Eustachian
tube related to surgical management of acquired aural cholesteatoma in children.
Laryngoscope 1978; 88 (7 Pt 1): 1155-64.

9.

Gacek RR. A differential diagnosis of unilateral serous otitis media.


Laryngoscope 1992; 102 (4): 461-8.

10.

van Heerbeek N, Ingels KJ, Snik AF, Zielhuis GA. Eustachian tube function in
children after insertion of ventilation tubes. Ann Otol Rhinol Laryngol 2001; 110
(12): 1141-6.

From the Department of Otolaryngology-Head and Neck Surgery, Loyola University


Stritch School of Medicine, Maywood, Ill.
Correspondence: John P. Leonetti, MD, Professor and Vice-Chairman, Department of
Otolaryngology-Head and Neck Surgery, Loyola University Medical Center, 2160 S.
First Ave., Maywood, IL 60153. Email: jleonet@lumc.edu
Ear Nose Throat J. 2013 April-May;92(4-5):195-200