85 / 232
US would accurately predict the presence of purulence
in children being evaluated for possible PTA.
MATERIALS AND METHODS
After obtaining institutional review board approval, a con-
venience sample of children and adolescents with suspected
PTA per the evaluating provider were prospectively enrolled in
the study. The diagnosis of PTA was based on a history of sore
throat and/or fever, neck pain, trouble swallowing, and voice
changes in conjunction with physical exam findings such as
asymmetric tonsils, palatal edema, uvular deviation, trismus,
and peritonsillar fullness or erythema. Children
<
2 years old,
those with significant airway compromise, and those being eval-
uated for retropharyngeal processes were excluded from the
study. A physician board certified in pediatrics or pediatric
emergency medicine evaluated each patient and then obtained
consultation from the otolaryngology service. Patients were
managed based on clinical impression alone, but all enrolled
patients underwent a transcervical US to evaluate for PTA.
The transcervical technique utilizes a high-frequency probe
placed below the inferior border of the mandible to visualize the
submandibular gland, deep to which the tonsil and peritonsillar
space can be assessed. A blinded radiologist (
A
.
B
.) viewed and
analyzed all final US images.
The results of the US were compared to the results of pro-
cedural interventions and clinical patient outcomes. A positive
US was defined as an anechoic or hypoechoic pocket in the peri-
tonsillar plane suggestive of abscess. A true-positive PTA was
defined as purulence discovered during surgical intervention in
the setting of a positive US. A false-positive PTA was defined as
the absence of purulence during surgical intervention or suc-
cessful medical management in the setting of a positive US. A
true-negative PTA was defined as no purulence noted during
procedural intervention or clinical improvement with medical
management alone in the setting of a negative US. Last, a
false-negative PTA was defined as purulence discovered during
surgery in the setting of a negative US.
Following discharge, patients’ medical records were retro-
spectively reviewed for treatment failures, defined as those
patients managed initially medically who ultimately underwent
drainage of a PTA. Statistical analysis with Fisher exact test
was performed using GraphPad Prism software GraphPad Soft-
ware, Inc., La Jolla, CA), and sensitivity, specificity, and posi-
tive and negative predictive values were calculated based on
our results. Multivariate regression analysis was conducted
analyzing any correlations between age, sex, otolaryngologist
clinical diagnosis, and US findings with the presence/absence of
PTA.
RESULTS
Forty-three patients were enrolled in this study
from May 2013 to April 2014. The demographic and age
distribution of these patients can be seen in Table I.
Using the definitions described earlier, we compared the
US findings to procedural findings and/or clinical man-
agement outcomes (Table II). The US was positive for
PTA in 17 (39.5%) patients. Of these patients, nine were
found to have had true-positive PTA by our definition.
The greatest measurable dimension of these abscesses
ranged from 7 mm to 32 mm, with a mean of 25 mm. Of
the eight false-positive ultrasounds, the diameter of the
abscesses ranged from 11 mm to 28 mm with a mean of
18 mm. The size differences between these two groups,
which was statistically significant, can be viewed in
Table III. Of the eight false positives, three patients had
drainage procedures without procurement of pus, and
five were medically managed successfully. The diameter
of the abscess cavity on US in these five patients ranged
from 11 mm to 21 mm, with a mean of 15 mm. Of the
three patients undergoing negative procedures, two had
bedside needle aspiration without evidence of purulence,
and one underwent a negative incision and drainage in
the operating room based on clinical exam. Three
patients diagnosed with PTA clinically but managed
medically also had US findings consistent with abscess.
Either these patients responded promptly to initiation of
medical management, or the parent opted to forego elec-
tive surgical intervention as initial therapy.
US did not reveal a PTA in 26 (60.5%) of the chil-
dren enrolled. The breakdown of the US diagnoses for
these patients can be seen in Table IV. Two of these chil-
dren underwent drainage procedures following clinical
diagnosis of PTA but with no purulence identified in
TABLE I.
Patient Demographics.
No. of Patients
Mean Age (SD), yr
Median Age (IQR), yr
Age Range, yr
Gender
Race
Ultrasound cohort
43
12.0 (5.3)
13 (7–17)
2–20
M: 23, F: 20 W: 7, A: 35, H: 1
A
5
African American; H
5
Hispanic; IQR
5
interquartile range; SD
5
standard deviation; W
5
white.
TABLE II.
Ultrasound Results and Surgical Findings.
Surgical Intervention
Ultrasound Results
Positive
Negative
Total
None or negative I&D
8
26
34
Positive I&D
9
0
9
Total
17
26
43
I&D
5
Incision and drainage.
TABLE III.
Comparison of True Positive and False Positive Abscess Cavity
Dimensions.
True Positive
False Positive
No.
9
8
Range, mm
7–32
11–28
Mean, mm
25
18
P
value
<
.05
P
value calculated using two-tailed Mann-Whitney test.
Laryngoscope 125: December 2015
Fordham et al.: Transcervical US in Pediatric PTA
63