38 / 412
reinnervation of opposing muscle groups by the same nerve,
leading the muscles to contract simultaneously. In canine
models, reinnervating axons begin reaching the vocal fold
muscles within 3 months of injury. Therefore, treatments to
prevent unfavorable synkinesis would need to be adminis-
tered within this time frame.
22,23
The objective of this study was to determine how often
patients with UVFP present to the Washington University
School of Medicine (WUSM) within 3 to 4 months of
symptom onset. These patients might be eligible for a
clinical trial of early intervention for the prevention of
synkinesis. A review of literature revealed no previous
studies focusing on time of presentation of patients with
UVFP. Etiology and laterality data were also collected and
reported.
Methods
Approval for the study was obtained from the WUSM
Institutional Review Board. Adult patients (
.
18 years of
age) seen between January 1, 2002, and January 1, 2012,
with a diagnosis of unilateral vocal fold paresis or paralysis
(based on
International Classification of Diseases, Ninth
Revision
and
Current Procedural Terminology
codes) were
identified via a query of electronic medical records. In most
cases, the diagnosis of UVFP was made by the attending
physician based only on examination of the larynx; electro-
myography or cricoarytenoid joint palpation was performed
in only a small percentage of patients. In all cases, the diag-
nosis was confirmed by flexible fiberoptic examination or
videostroboscopy.
Each patient’s chart was reviewed and data were col-
lected and stored in an electronic database. Data included
age at initial visit, sex, date of visit to a WUSM otolaryn-
gologist, date of presentation to a non-WUSM otolaryngolo-
gist (in any), primary symptom, date of symptom onset, side
of paralysis, etiology of paralysis, initial and subsequent
treatment received, date of initial treatment, voice improve-
ment, and fiberoptic examination vocal fold movement out-
comes. From this information, the time intervals from
symptom onset to initial presentation to an otolaryngologist
and the time interval from initial presentation to referral to
WUSM (if made), were calculated. The specific reason for
referral to the university was not usually recorded in the
chart, but very few had been previously treated. Diagnosis
of idiopathic UVFP was confirmed by negative imaging
along the course of the vagus and recurrent laryngeal
nerves. Patients with incomplete records were excluded.
Data were analyzed by first determining percentages of
men and women in the study population, laterality of
UVFP, and etiology. To evaluate laterality, a calculation for
left-right skew was devised as follows:
Skew
5
#
cases on right
=
total
#
cases
ð
Þ
0
:
5
:
With this formula, it can be seen that if there is a perfect 50-50
split, the skew is zero; as the proportion of left-sided cases
increases, the value becomes more negative (moves to the left),
and so on, with a maximum/minimum value of
6
0.5. This
value was determined and plotted for each etiology. Statistical
significance was determined based on a null hypothesis that the
frequency on each side was the same, using a
x
2
test.
Median presentation time to an outside otolaryngologist
was compared with median referral presentation time to a
WUSM otolaryngologist. For this analysis, patients who ini-
tially presented to the WUSM were excluded. Due to wide
variance in presentation times and nonnormally distributed
data, the mean presentation times were not presented. To com-
pare median presentation times between the 2 groups, a paired
Wilcoxon rank sum test was used. Histograms were also gen-
erated incorporating all data, displaying number of patients
and cumulative percentage of patients by time of presentation.
Percentages of patients per etiology who presented at 2, 3, and
4 months after symptom onset were also calculated.
Results
Of the charts reviewed, 938 patients met inclusion criteria
for this study; 497 (53%) were women and 441 (47%) were
men. The average age of patients was 56.9 years (range, 18-
93 years). Overall, 621 (66.2%) patients had left-sided
UVFP, while 317 (33.8%) patients had right-sided UVFP.
Table 1
displays the etiologies of unilateral vocal fold
paralysis. In total, 522 (55.6%) patients had UVFP due to
iatrogenic effects related to surgery. The most frequently
observed surgery related to UVFP was thyroid/parathyroid
surgery, noted in 158 (16.8%) patients. Lung surgery (n =
73 [7.8%]), cardiac surgery (n = 58 [6.2%]), and cervical
spine surgery (n = 48 [5.1%]) were the next most common
surgical causes of UVFP.
In total, 358 patients (38.2%) had UVFP due to causes
not directly related to surgical intervention. Malignancy was
the cause of UVFP in 167 (17.8%) of patients. Lung malig-
nancy (n = 73 [7.8%]), metastatic malignancy (n = 24
[2.6%]), skull base malignancy (n = 18 [1.9%]), and direct
invasion by thyroid malignancy (n = 14 [1.5%]) were most
common. Idiopathic UVFP was noted in 124 patients
(13.2%). Other less common causes of UVFP included intu-
bation (n = 58 [6.2%]), trauma (n = 30 [3.2%]), cerebral
vascular accident (CVA; n = 18 [1.9%]), and neck radiation
(n = 8 [0.9%]).
Table 2
shows the laterality of UVFP based on etiology.
In total, 622 (66.2%) patients had left-sided UVFP. This
table shows the difference between right- and left-sided
UVFP. Left-skewed etiologies of UVFP, represented by
negative values, and right-skewed etiologies of UVFP, rep-
resented by positive values, are plotted in
Figure 1
. In
addition to the expected left-sided predominance of
intrathoracic etiologies (lung surgery, cardiac surgery, eso-
phageal surgery, and lung malignancy), other significantly
left-sided causes included idiopathic, intubation, carotid sur-
gery, and skull base malignancy. There were no etiologies
that were significantly skewed to the right.
Table 3
shows the median time of presentation for the
92 patients (9.8% of study population) who initially
Spataro et al
18