Medical therapy versus sinus surgery by using balloon

sinus dilation technology: A prospective multicenter study

Spencer C. Payne, M.D.,1 Pablo Stolovitzky, M.D.,2 Neelesh Mehendale, M.D.,3 Keith Matheny, M.D.,4
William Brown, M.D.,5 Anthony Rieder, M.D.,6 Douglas Liepert, M.D.,7 Ewen Tseng, M.D.,4
Andrew Gould, M.D.,8 Scott Powell, M.D.,9 Daniel Van Himbergen, M.D.,10 Boris Karanfilov, M.D.,11
Daniel Harfe, M.S.E., M.B.A.,12 Laura England, Ph.D.,12 and Christopher Melroy, M.D.13

Y
P

ABSTRACT

Background: Although previous studies of sinus surgery that used balloon catheter dilation technology for the paranasal sinuses (balloon sinus dilation
[BSD]) demonstrated safety and efficacy, data that compare BSD with continued medical management (MM) are lacking.
Objective: To evaluate the outcomes of sinus surgery when using BSD instruments versus MM for patients with chronic rhinosinusitis for whom MM failed.
Methods: Adult patients with chronic rhinosinusitis for whom a minimum of 3 weeks of oral antibiotics, 4 weeks of daily saline solution therapy, and 4
weeks of daily nasal corticosteroids failed were included. Qualifying participants were allowed to self-select sinus surgery with BSD (either an office or
operating room setting) or continued MM. The primary end point was the comparison of change in the Chronic Sinusitis Survey score from baseline to 24
weeks. Secondary end points included comparisons of change for the Rhinosinusitis Disability Index (RSDI) and the Sino-Nasal Outcome Test (SNOT-20).
Results: A total of 198 patients were enrolled (146 surgery and 52 MM). Of the patients who chose BSD, 72% (105/146) had their procedures completed
in an office setting. Overall, BSD instruments were successful in dilating 97.6% of targeted sinuses (561/575). Patients who chose BSD showed a significantly
greater improvement in the Chronic Sinusitis Survey score versus MM (42.0 versus 27.0, p 0.001). Results from the RSDI and SNOT-20 surveys showed
similar improvements for surgery versus MM (RSDI, 36.0 versus 18.1, p 0.001; SNOT-20, 1.7 versus 1.0, p 0.002).
Conclusion: Patients who selected sinus surgery in which BSD instruments were used on the peripheral sinuses demonstrated significantly greater
improvements in quality of life compared with those who elected ongoing MM. These results were achieved through office-based procedures with the patient
under local anesthesia in the majority of patients.
(Am J Rhinol Allergy 30, 279 286, 2016; doi: 10.2500/ajra.2016.30.4346)

hronic rhinosinusitis (CRS) is a debilitating disease, with substantial quality of life (QOL) impact. Treatment initially consists of medical management (MM) for which the concept has evolved
over time. Although, historically, this option has included antibiotics,
best evidence indicates that topical corticosteroids supplemented by
nasal saline solution rinses with oral corticosteroids and/or antibiotics as adjunctive measures when appropriate are preferred.1 Although MM is successful in managing the disease for a majority of
patients, a large number of patients continue to experience symptoms.2,3 On failure of medical treatment, patients are faced with a
difficult choice: continue with MM or undergo sinus surgery.
Balloon sinus dilation (BSD) technology is available as an additional surgical tool in the otolaryngologic armamentarium to widen

O
D

O
N

From the 1Department of Otolaryngology - Head & Neck Surgery, University of

Virginia, Charlottesville, Virginia, 2ENT of Georgia, Atlanta, Georgia, 3ENT Associates of Texas, McKinney, Texas, 4Collin County ENT, Frisco, Texas, 5South Florida
ENT Associates, Miami, Florida, 6Alliance ENT and Hearing Center, Wauwatosa,
Wisconsin, 7Allied ENT Specialty Center South Bend, Indiana, 8Advanced ENT and
Allergy, Louisville, Kentucky, 9Florida E.N.T. & Allergy, Tampa, Florida, 10Virginia Ear
Nose and Throat, Henrico, Virginia, 11Ohio Sinus Institute, Dublin, Ohio, 12Acclarent,
Inc., Menlo Park, California, and 13Georgia Nasal and Sinus Institute, Savannah, Georgia
This study was sponsored by Acclarent, Inc. Financial and logistical support for the
study was provided by Acclarent, Inc
S.C. Payne and B. Karanfilov consultant for Acclarent. P. Stolovitzky, C. Melroy and
A. Gould consultant for Acclarent and speakers for Acclarent. K. Matheny and E.
Tseng are speakers for Acclarent. W. Brown and S. Powell are former consultants for
Acclarent. D. Harfe and L. England were employees Acclarent during the time of the
study. The remaining authors have no conflicts of interest pertaining to this article
Interim results from this study were presented at the 60th Annual Meeting of the
American Rhinologic Society, September 20, 2014, Orlando, Florida
Address correspondence to Spencer C. Payne, M.D., Department of Otolaryngology
Head and Neck Surgery, University of Virginia Health System; P.O. Box 800713,
Charlottesville, VA 22908-0713
E-mail address: SCP9X@virginia.edu
Published online June 17, 2016
Copyright 2016, OceanSide Publications, Inc., U.S.A.

O
C

sinus outflow tracts. When compared with traditional excisional

endoscopic sinus surgery (ESS), sinus surgery by using BSD technology (hereafter referred to as the BSD or surgery group) is relatively less invasive because sinus outflow tracts can be widened
without tissue removal. Although not suitable for all types of sinus
disease or anatomy, several prospective multicenter studies have
shown BSD to be a safe and effective option for surgically addressing
the symptoms of CRS.46 More recently, randomized studies that
compared BSD with traditional ESS were published and demonstrate
promising results for BSD.710
It is notable, however, that studies that compare BSD to ongoing
medical therapy for patients with CRS are not available in the literature. As a result, controversy regarding the strength of the evidence
in support of BSD technology persists.11 Without such data, it is
difficult for surgeons to communicate predicted relative outcomes
when discussing surgical versus continued medical options with their
patients. In many instances, BSD can now be routinely and safely
performed in a physician office setting,1215 which potentially shifts
the benefit-to-risk ratio as patients consider their treatment options.
The objective of our study was to prospectively follow up the outcomes of two groups of patients with CRS; patients in both groups
had a recent, well-defined course of MM that failed. One group chose
ongoing MM alone, whereas the other group elected sinus surgery by
using BSD instruments, followed by MM as needed.

METHODS
Study Overview
This multicenter study (NCT01685229) enrolled patients between
September 2012 and August 2014 at 24 centers in the United States,
after institutional review board approval at all the sites. Adult patients with CRS whose symptoms were unresolved after a defined
course of MM were eligible to enroll. In consultation with their
surgeon, enrolled patients elected either ongoing MM or sinus surgery by using BSD instruments in an office or operating room (OR)

American Journal of Rhinology & Allergy

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm

279

setting. Patients were followed up at 2 (BSD only), 12, and 24 weeks

after enrollment, with the primary end point being the relative difference in improvement in the Chronic Sinusitis Survey (CSS)16 score
at 24 weeks.

Table 1 Rationale provided by patients for electing continued

medical management over sinus surgery
Reason

Patients, no. (%)

(N 52)

Inclusion and Exclusion Criteria

Doesnt feel sick enough

Concerned about surgical complications
Concerned about recovery time
Other
Limited time for surgery

19 (36.5)
8 (15.4)
7 (13.5)
6 (11.5)
2 (3.8)

Adult patients ages 19 years with a diagnosis of CRS as per the

2007 American Academy of Otolaryngology - Head & Neck Surgery
(AAO-HNS) Rhinosinusitis Task Force17 were eligible to enroll. All
patients must have medical therapy that failed, including a minimum
of a 3-week course of broad-spectrum or culture-directed antibiotics,
4 weeks of a topical intranasal steroid, and 4 weeks of daily intranasal
saline solution therapy. Additional medication beyond this defined
minimum was at the discretion of the investigator and customized to
the patients disease process. Medical failure had to occur within 90
days before the study to ensure enrollment of recently treated patients. In addition to the strictly defined criteria for medical failure,
radiographic evidence of paranasal sinus inflammation from a computed tomography (CT) image that was taken after the conclusion of
the medical regimen was required for all the patients. A baseline
mean Sino-Nasal Outcome Test (SNOT-20)18 index score of 0.6
(range, 05) was also required for enrollment.
Patients with peripheral sinus (i.e., maxillary, frontal, or sphenoid)
disease that was not amenable to treatment with BSD instruments
were excluded, as were pregnant or lactating women. For example,
subjects who required a large antrostomy to excise intrasinus polyps
were not enrolled. In this regard, patients were eligible when the
treating physicians believed that relieving outflow obstruction was
anticipated to resolve the patients underlying disease. Ethmoidectomy, polypectomy, septoplasty, and turbinate reduction were not
exclusionary procedures.

Treatment Arms

O
D

Baseline Assessments

Baseline characteristics collected from each patient included medical history, Lund-Mackay CT score (score range, 024),19 and LundKennedy endoscopy score (score range, 020).20 The actual failed
medical therapy regimen was collected at baseline, as was an assessment of missed days from work or school for the 90-day period before
enrollment.

Primary and Secondary Outcomes

The CSS is a six-question disease-specific survey that measures
both symptom burden and medication usage in the preceding 8
weeks, with a normalized score range of 0 to 100.16 A patient with a
low CSS score is experiencing more symptoms and/or greater medication usage than a patient with a higher CSS score. The difference
between the baseline and 24-week CSS scores was calculated for the

280

O
N

All enrolled patients were eligible for ESS with BSD instruments.
Patients were then allowed to self-select a plan of continued MM (the
MM arm) or proceed with surgery (the BSD arm). A nonvalidated survey was used to assess the rationale for patient selection of
continued MM, despite all the patients being surgery eligible.
For the patients in MM, the ongoing medical regimen was dictated
by the patients specific disease process and customized by the treating physician investigator. A single universal MM was not prescribed
by the protocol. For the surgical patients, surgery was conducted by
using transnasal BSD instruments (Acclarent, Inc., Menlo Park, CA) in
an office or OR setting. The choice of surgery location was driven by
surgeon recommendation and patient preference. To mirror realworld practice, adjunctive procedures, such as polypectomy, septoplasty, and ethmoidectomy, were permitted. After surgery, ongoing
MM for the BSD arm was prescribed by the operating surgeon, again
customized to the disease process of each individual patient.

Y
P

MM and BSD groups. The primary outcome was the relative betweengroup difference in baseline to 24-week improvement in the CSS
scores.
Secondary end points included the relative difference between the
MM and BSD groups in baseline to 24-week change scores of
the SNOT-20 and Rhinosinusitis Disability Index (RSDI)21 surveys.
The SNOT-20 measures disease-specific QOL by using a 20-question
survey,18 whereas the RSDI (total score range, 0120) measures the
effect of CRS on physical, emotional, and functional domains by using
30 questions. For both the SNOT-20 and RSDI, a higher score indicated an increased impact of sinus disease on the patient. Additional
secondary end points included an evaluation of missed days of work
and/or school, oral steroid usage, topical steroid usage, and antibiotic
usage. These secondary end points were based on a patient questionnaire with a 90-day recall period. Safety was tracked by collection of
all medication-, procedure-, or device-related adverse events. Revision procedures on sinuses successfully dilated during the index
procedure were also tracked, as were the number of patients in MM
who elected to have BSD after enrollment (known as crossover
patients).

O
C

Sample Size Calculation and Statistical Analysis

The study was powered to detect a difference of 12.5 points in the

CSS total change score between the MM and BSD groups when
assuming a common standard deviation of 25, no more than a 1:3
ratio between the MM and BSD arms, 80% power, and a two-sided
of 0.05. Smith et al.22 showed a 15.7-point difference between the MM
and ESS arms in a study of similar design, with a MM:ESS ratio of
1:2.2. Given the availability of office-based BSD, it was anticipated
that a higher ratio of patients would choose BSD relative to MM when
compared with a choice of ESS versus MM, and the study was
powered accordingly. It was calculated that a sample size of 129
patients for BSD and 43 patients for MM would be required to
demonstrate a 12.5-point CSS superiority for the BSD arm over the
MM arm. Given the follow-up challenges demonstrated in earlier
studies of ESS,22,23 we anticipated considerable loss to follow up and
allowed for up to 350 patients to enroll.
A Bayesian interim analysis framework was predefined and implemented in the study to adaptively stop the enrollment early should
certain criteria be met. Interim analyses on the primary end point
were prespecified when 125 and 250 patients completed 24 weeks of
follow-up. Enrollment could be stopped early for superiority if the
Bayesian predictive probability of BSD superiority was at least 95%. A
Bayesian predictive probability of 95% means that there is a 95%
probability that the results observed in the interim analysis (i.e., after
125 patients) would not change if the study were allowed to be fully
enrolled (i.e., to 350 patients). Conversely, if the Bayesian predictive
probability of superiority at the interim analysis dropped below 10%
(i.e., there is only a negligible chance that surgery with BSD tools
would prove superior should the study be allowed to be fully enrolled), the enrollment could be stopped early for futility. During the
course of the study, it was anticipated that some of the patients who
initially chose MM would later cross over and choose surgery, and
this was permitted by the study protocol after the 12-week follow-up

JulyAugust 2016, Vol. 30, No. 4

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm

Table 2 Baseline demographics and characteristics of the MM and BSD groups

Characteristic

MM (n 52)

BSD (n 146)

Age, mean SD (range), y

Men, no. (%)
Medical history, no. (%)
Previous sinus surgery
Nasal polyposis
Septal deviation
Asthma
ASA intolerance
Allergic rhinitis
Depression
Oral steroid dependency
Current smoker
Lund-Mackay CT score, mean SD*
Lund-Kennedy endoscopy score, mean SD#

46.4 13.3 (2372)

21 (40.4)

50.4 14.8 (2192)

71 (48.6)

12 (23.1)
6 (11.5)
36 (70.6)
12 (23.1)
2 (3.8)
47 (90.4)
5 (9.6)
4 (7.7)
4 (7.7)
8.7 5.2
5.5 2.9

42 (28.8)
24 (16.4)
94 (64.8)
40 (27.4)
4 (2.8)
100 (68.5)
20 (13.7)
3 (2.1)
10 (6.9)
8.4 5.1
5.1 3.3

p
0.086
0.335
0.473
0.502
0.495
0.587
0.999
0.002
0.483
0.079
0.999
0.715
0.381

Y
P

O
C

MM Medical management; BSD balloon sinus dilation; SD standard deviation; ASA acetylsalicylic acid; CT computed tomography.
*Lund-Mackay CT score range, 0 24.
#Lund-Kennedy endoscopy score range, 0 20.
Table 3 Baseline quality of life and medication usage
MM, mean SD
(no.) (n 52)
CSS total score
RSDI total score
SNOT-20 score
Missed work and/or school*
Oral antibiotics*
Oral steroids*
Nasal spray*

34.3 18.4 (52)

39.5 20.0 (52)
2.4 0.9 (52)
1.5 2.4 (49)
22.8 6.1 (52)
8.4 10.0 (52)
47.2 28.1 (52)

O
N

BSD, mean SD
(no.) (n 146)

28.5 17.9 (145)

47.9 25.3 (146)
2.5 1.0 (146)
3.0 9.2 (125)
28.1 15.4 (146)
7.9 10.4 (146)
53.0 29.6 (146)

0.048
0.032
0.622
0.257
0.018
0.788
0.220

MM Medical management; BSD balloon sinus dilation; SD standard deviation; CSS Chronic Sinusitis Survey (score range, 0 100); RSDI
Rhinosinusitis Disability Index (score range, 0 120); SNOT-20 20-item Sino-Nasal Outcome Test (mean score range, 0 5).
*The average number of days (of the past 90 days).

visit. Data before the crossover were included in the MM arm,

whereas post-crossover data were censored from the MM arm so as to
not blend surgical and nonsurgical data.
Categorical data were summarized by using frequency tables,
which presented the patient counts and the percentage of patients in
that group that were in that category. Pearson 2 and t-tests were used
for analyzing differences in baseline comorbidities. The t-tests were
used to assess for improvement between baseline and follow-up time
points, and to assess between-group changes. In addition, for distributions that were suspected to violate normality assumptions, nonparametric Wilcoxon signed rank tests were calculated. Multivariate
linear regression was performed to adjust for potential confounding
baseline variables when assessing the relative CSS change between
the MM and surgery groups. Final models were constructed by using
clinical judgment and a stepwise selection with a significance value of
0.05 for entry into the model and 0.10 for exit from the model. All
statistical analyses were performed by using SAS 9.4 (SAS Institute,
Cary, NC).

O
D

RESULTS
Early Stopping for Superiority
Bayesian analysis was conducted on the primary end point (the
relative difference in baseline to 24-week CSS improvement for MM
versus BSD) after the first 125 patients had completed 24 weeks of
follow-up. Patients who selected BSD demonstrated a significantly
greater improvement in CSS compared with those patients who

elected continued MM (p 0.001). The Bayesian predictive probability of superiority exceeded 95%, which permitted early enrollment
stopping of the study for superiority. At the time of the early stopping
analysis, 198 patients had been enrolled, which thus represented the
full study cohort.

Baseline Characteristics and Patient Disposition

One hundred ninety-eight patients were enrolled at 24 centers in
the United States: 146 patients (74%) chose BSD and 52 (26%) chose
MM. The reasons that patients chose continued MM despite being
eligible for surgery, with the most common reason being Doesnt feel
sick enough, are summarized in Table 1. Follow-up rates for the
primary end point (CSS at 24 weeks) were 87.0% for the BSD group
and 71.1% for the MM group. Baseline characteristics are shown in
Table 2. The patients who chose MM had a greater prevalence of
allergic rhinitis (p 0.002) but otherwise had similar comorbidities as
patients who chose surgery. Notably, the baseline CT (p 0.72,
Lund-MacKay) and endoscopy (p 0.38, Lund-Kennedy) were similar between the groups.
The baseline medical failure regimen (previous 90 days) and baseline QOL results for the MM and surgery groups are shown in Table
3. The groups had similar oral and nasal steroid usage in the 90 days
before enrollment; the surgery group showed greater antibiotic usage
(p 0.012). Baseline SNOT-20 scores were not statistically different
between the groups, whereas the RSDI (p 0.03) and CSS (p 0.048)
both showed a statistically greater impact on QOL for the surgery
group compared with the MM group.

American Journal of Rhinology & Allergy

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm

281

282

0.001
0.001
0.001
69.2 21.3 (76)
11.9 19.3 (77)
0.9 0.8 (77)

CSS Chronic Sinusitis Survey; RSDI Rhinosinusitis Disability Index; SNOT-20 20-item Sino-nasal Outcome Test; MM medical management; BSD balloon sinus dilation; SD standard
deviation.

Baseline characteristics (Tables 2 and 3) demonstrated a difference

in the presence of allergic rhinitis and previous antibiotic use between

27.7 17.1 (86)

46.3 24.0 (86)
2.4 1.0 (86)
0.001
0.001
0.001

28.5 17.9 (145)

47.9 25.3 (146)
2.5 1.0 (146)

70.4 21.1 (128)

11.6 19.0 (127)
0.8 0.8 (128)

0.001
0.001
0.001

Baseline,
mean SD
(no.)
p

At 24 Weeks,
mean SD
(no.)
Baseline,
mean SD
(no.)

BSD with or without Adjunctive Procedures

61.3 18.9 (32)

22.6 23.2 (32)
1.4 1.0 (32)
34.3 18.4 (52)
39.5 20.0 (52)
2.4 0.9 (52)

p
MM

At 24 Weeks,
mean SD
(no.)

CSS
RSDI
SNOT-20

Regression Modeling

Baseline,
mean SD
(no.)

O
D

O
N

Y
P

O
C
Table 4 Change from baseline for CSS, RSDI, and SNOT-20 for MM, BSD, and BSD alone groups

Primary and Secondary Outcomes

Changes from baseline to 24 weeks for CSS, RSDI, and SNOT-20
are provided in Table 4. Relative comparisons of the baseline to
24-week improvement are provided in Table 5. Both the MM and
surgery groups demonstrated significant improvement from baseline to 24 weeks in CSS, RSDI, and SNOT-20 scores. Patients who
elected BSD showed significantly greater improvement in CSS (p
0.001), RSDI (p 0.001), and SNOT-20 (p 0.002) survey scores at
24 weeks compared with the MM group. The relative difference in
improvement for the CSS, RSDI, and SNOT-20 end points when the
groups were compared are shown in Fig. 1. Worthy of note is that,
although the study was not originally powered to isolate BSD as a
standalone procedure, a subgroup analysis did confirm that, when
the patients who underwent adjunctive procedures (i.e., a mixed/
hybrid surgery) were excluded, BSD remained superior to MM in
improving QOL as measured by the CSS (p 0.0027), SNOT-20
(p 0.0049), and RSDI (p 0.0006) (Table 5). In addition, improvement in CSS did not differ between isolated BSD and BSD with
ethmoidectomy (p 0.1071), and between BSD and BSD with any
adjunctive procedure (p 0.8741).
Changes in antibiotic, nasal steroid, and oral steroid usage are
presented in Table 6, as is a comparison of the number of missed
work and/or school days for the two groups. At 24 weeks, both
groups demonstrated large reductions in antibiotics and oral steroid usage as well as fewer missed days of work and/or school.
The MM group showed an increase in nasal steroid usage, whereas
the BSD group demonstrated a decrease (BSD versus MM, p
0.005). None of the patients in the BSD group required a revision
procedure during the 24-week follow-up. Seven of the 52 patients
(13.5%) who initially chose MM subsequently crossed over and
requested and received sinus surgery with BSD technology before
the 24-week end point.

p
BSD Alone

There were 146 patients who elected sinus surgery by using BSD
tools alone or in a mixed/hybrid setting: 105 (71.9%) of these
procedures occurred in an office setting with the patient under
local anesthesia and 41 procedures (28.0%) occurred in the OR. A
total of 575 sinuses were targeted for dilation, with overall technical success of 97.6% (561 dilated/575 attempted). In-office technical success was 96.8%, and OR technical success was 100%.
Adjunctive procedures were performed more frequently in the OR
setting compared with the office setting: ethmoidectomy (68.3%
OR versus 1.0% office), polypectomy (26.8% OR versus 2.9% office), turbinate reduction (48.8% OR versus 14.3% office), and
septoplasty (51.2% OR versus 1.0% office). The median time to
return to normal activities (surgical recovery) was 2.0 days for the
office patients and 5.0 days for the OR patients (p 0.001). Of the
patients who had BSD in the office, 45% returned to normal activity within 1 day after surgery.
Two procedure-related adverse events were reported. A 92-yearold patient had a transient ischemic attack 3 days after the procedure, secondary to discontinuation of clopidogrel-aspirin due to a
nosebleed. This event was fully resolved within 1 day. In the other
event, in one subject a dental crown was broken during extubation.
One sinus infection that occurred 2 weeks after surgery was identified by the investigational site as possibly procedure related, in
that it was not possible to rule out surgical manipulation as a
contributor to the inflammation that led to the infection. There
were no serious device-related adverse events reported during the
study.

At 24 Weeks, mean SD
(no.)

Surgical Data and Safety

JulyAugust 2016, Vol. 30, No. 4

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm

Table 5 Change comparisons for CSS, RSDI, and SNOT-20 for MM, BSD, and BSD Alone groups
Intrapatient Change*

CSS
RSDI
SNOT-20

MM, mean SD
(no.)

BSD with or without Adjunct

Procedures, mean SD (no.)

p#

BSD Alone,
mean SD (no.)

p#

27.0 19.5 (32)

18.1 16.8 (32)
1.0 1.0 (32)

42.0 23.7 (127)

36.0 26.4 (127)
1.7 1.1 (128)

0.001
0.001
0.002

42.3 25.1 (76)

34.6 25.4 (74)
1.6 1.1 (75)

0.005
0.001
0.005

Y
P

CSS Chronic Sinusitis Survey; RSDI Rhinosinusitis Disability Index; SNOT-20 20-item Sino-Nasal Outcome Test; MM medical management;
BSD balloon sinus dilation; SD standard deviation.
*Only includes patients with both baseline and 24-week data.
#The p values reflect the difference between BSD groups and MM.

the two treatment arms. Stepwise regression modeling was performed to determine if the treatment arm or other patient characteristics may remain as independent predictors of improvement. The
following variables, in addition to treatment arm, remained in the
model after analysis: the presence of polyposis, septal deviation,
depression, baseline Lund-MacKay CT score, and baseline CSS score.
Regression results demonstrated that patients who elected BSD
showed a significantly greater improvement in CSS scores compared
with patients who elected continued MM (p 0.001), after adjustment
for potential confounding variables. The results of the linear regression model, which also demonstrated that patients with polyposis,
depression, and worse CT scores did worse in general, regardless of
treatment arm (Table 7). The results were unchanged when limiting
the analysis to patients in MM and patients undergoing BSD without
adjunctive procedures.

DISCUSSION

O
D

O
N

In this prospective multicenter study, patients in whom a welldefined medical regimen had failed and who were considered
candidates for ESS with balloon dilation instruments were given
the option to select continued MM or surgery with balloon dilation
technology. As anticipated, a relatively small proportion of the
patients (26%) chose continued medical therapy. The study required failure of an appropriate medical regimen, and this failure
was required to have occurred within a 90-day period before
enrollment. These criteria, coupled with the recent availability of
office-based balloon dilation, likely led to the large majority of
patients choosing the surgical option rather than to continue on a
medical therapy regimen. Interestingly, the primary reason that
subjects gave for choosing MM over BSD (Doesnt feel sick
enough) was based on their self-assessment of symptom severity
(Table 1). The next most common reasons for declining surgery
were fear of surgical complications and concerns about recovery
time.
Similar to a comparative study of ESS versus MM by Smith et
al.,22,24 patients in our study who chose MM over surgery had
better CSS and RSDI scores at baseline, although their SNOT-20
scores were the same. This is despite a higher prevalence of allergic
rhinitis for the MM group and radiographically and endoscopically
assessed disease severity that was the same as the group of subjects
who chose surgery. This further confirmed our knowledge that
there is often an imperfect correlation between perceived disease
severity and objective findings. Presumably, the impact of certain
symptoms on overall QOL was not adequately accounted for by
these symptom scales and/or differed based on the underlying
disease phenotype.
This study demonstrated that patients who elected sinus surgery
with BSD technology in a standalone or mixed/hybrid surgery demonstrated a significantly greater improvement in QOL compared with
patients who elected continued MM, and this superiority was consistent across the three validated instruments used. Patients in MM were

permitted to crossover and select surgery if they thought that continued MM was unsatisfactorily addressing their symptoms. It was
anticipated that the lower QOL (worse feeling) of the patients in MM
would crossover and exit the MM group, thereby enriching the mean
score of those who remained in the MM group. That we demonstrated
strong superiority despite this bias in favor of MM is encouraging.
That being said, our study did not measure overall satisfaction with
treatment choice, so it remains possible that, despite the surgical arm
having a greater improvement in QOL, those who opted for a continued medical approach may have been more happy and/or satisfied
with their outcome.
To our knowledge, this is the first study of BSD that used the CSS
instrument to evaluate outcomes. Unlike the RSDI and SNOT-20
instruments, the CSS includes an assessment of the intensity of medical therapy and, therefore, can provide additional information beyond the RSDI and SNOT-20.25 The CSS outcomes for the BSD arm of
this study, as well as the CSS outcomes for previous ESS22,2634 studies
that used the CSS as an outcome measure, are shown in Fig. 2. Given
the differences in patient population, surgical techniques, and follow-up length, it is not possible to make direct comparisons among
studies. It is notable, however, that the baseline score in this study
was lower than in all others (i.e., greater disease impact), potentially
because the protocol mandated a recent MM failure, and medication
usage was captured within the CSS.
This was also the first study, to our knowledge, that compared
ongoing MM to BSD surgery in patients with CRS. There are limited
reports in the traditional ESS literature that compare MM with surgery, most demonstrate that ESS conferred no additional benefit
when compared with ongoing MM.23,33,35 However, the initial MM
before enrollment was either not well characterized or was minimal
compared with generally accepted criteria for maximal medical failure, which made it difficult to interpret the results. More recently,
Smith et al.22 published an ESS versus MM study by using welldefined medical failure criteria, which provides a good model for
future investigations.
In their study, Smith et al.22 used a design feature that we adopted,
which allowed patients to self-select into MM or surgical arms. The
study demonstrated that patients who had ESS had a significantly
greater improvement in CSS and RSDI scores (SNOT-20 was not
used) compared with MM. As in our study, the patients were diagnosed with CRS according to the 2007 AAO-HNS Rhinosinusitis Task
Force criteria.17 The required medical failure regimen was similar to
our study in that the investigators required a minimum of 3 weeks of
antibiotics and 3 weeks of topical steroids (compared with the 3
weeks of antibiotics, 4 weeks of topical steroids, and 4 weeks of nasal
irrigation therapy required in our study).
Similar to the article by Smith et al.22, our study did not require oral
steroids as a mandatory component of medical therapy for inclusion.
Despite this, of the 30 patients enrolled in the study with nasal polyps,
the majority (21/30) received a course of oral steroids. An important

O
C

American Journal of Rhinology & Allergy

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm

283

O
D

284

Figure 1. Improvement in quality of life from baseline to 24 weeks in

Chronic Sinusitis Survey (CSS), Rhinosinusitis Disability Index (RSDI),
and Sino-Nasal Outcome Test (SNOT-20) scores for the medical management (MM) and balloon sinus dilation (BSD) groups

difference between these two studies, however, may be the timing of

the medical failure. We required patients to have had a medical
regimen that failed within the 90 days immediately before enrollment

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm
0.7 2.8 (31)
65.3 35.8 (31)

47.2 28.1 (52)

2.8 6.7 (31)

8.4 10.0 (52)

22.8 6.1 (52)

0.0 0.0 (32)

1.5 2.4 (52)

0.012

0.001

0.001

0.001

53.0 29.6 (146)

7.9 10.4 (146)

28.1 15.4 (146)

3.0 9.2 (145)

Baseline, mean
SD (n)

BSD

44.6 42.7 (128)

0.8 3.7 (130)

2.4 6.6 (130)

0.2 0.8 (121)

At 24 Weeks,
mean SD
(no.)

MM medical management; BSD balloon sinus dilation; SD standard deviation.

#
The average number of days (of the past 90 days); only includes patients with both baseline and 24-week data.

Days of missed
work and/or
school#
Days of oral
antibiotics use#
Days of oral
steroids use#
Days of nasal
spray use#

At 24 Weeks,
mean SD
(no.)

Baseline in
days, mean
SD (no.)

MM

Table 6 Changes in missed work and/or school days, antibiotic, oral steroid, and nasal steroid usage

O
N
p

0.001

0.001

0.001

7.9 9.3 (31)

20.5 10.1 (31)

1.3 2.2 (30)

0.382

0.070
0.461

2.0 4.4 (109)

26.0 15.8 (130)
6.5 9.9 (130)
0.057

17.6 46.0 (31)

9.1 46.6 (128)

0.005

BSD, mean SD
(no.)

Intrapatient Change*
MM, mean
SD (no.)

O
C

Y
P

JulyAugust 2016, Vol. 30, No. 4

Table 7 Stepwise linear regression analysis of potential confounding baseline variables on relative CSS change between MM and BSD
groups
Variable

Univariate Models
Estimate

Intercept
BSD (1 BSD, 0 MM)
Baseline CSS score (range, 0100)
Sex (1 female, 0 male)
Age, y
Previous sinus surgery (1 yes, 0 no)
Current polyposis (1 yes, 0 no)
Septal deviation (1 yes, 0 no)
Asthma (1 yes, 0 no)
ASA sensitivity (1 yes, 0 no)
Allergic rhinitis (1 yes, 0 no)
History of depression (1 yes, 0 no)
Oral steroid dependency (1 yes, 0 no)
Current smoker (1 yes, 0 no)
Lund-Mackay CT score (024)
Lund-Kennedy endoscopy score

15.04163
0.69852
2.00612
0.08788
5.25122
17.3644
6.51511
8.78525
4.50758
8.61401
8.28138
6.99786
1.03604
0.12403
1.29782

Stepwise Model
p

Estimate

0.0012
.0001
0.5951
0.4924
0.2059
0.0006
0.0977
0.0373
0.6767
0.0412
0.1213
0.6137
0.8943
0.7497
0.0269

40.34603
12.73975
0.68911

.0001
0.0019
.0001

Y
P

21.19213
7.29158

11.40103

O
C
1.05683

.0001
0.0253

0.0069

0.0023

CSS Chronic Sinusitis Survey; MM medical management; BSD balloon sinus dilation; CT computed tomography.

which they could be assigned to a MM treatment group that had

already proved ineffective. As mentioned previously, we adopted
design features used in the study sponsored by the American Rhinologic Society that modeled the real-world decision-making that
occurs in everyday practice.22 Risks to such a design include bias that
can be introduced due to differences in measured and unmeasured
baseline characteristics between the groups, which potentially affect
the magnitude of change in outcome measures.
A single standardized postenrollment medication regimen was not
implemented in this multicenter study. This approach was similar to
other recent comparative effectiveness studies in sinus surgery in which
medication requirements were customized to the individual disease
process, as determined by the clinician who provides direct patient
care.22,24,38,39 Given the heterogeneity of sinus disease pathogenesis and
the individually customized nature of medical therapy, we did not
believe that it was possible or appropriate to strictly control or limit
medication options. The need for medications was assessed as an outcome measure and was not considered a fixed design element.
The patient population studied may only represent a select subset
of patients with rhinosinusitis, so the external validity of the study
may not extend to all patients with rhinosinusitis. Although the QOL
metrics at baseline were comparable with the study by Smith et al.,22
the average CT score (Lund-Mackay) as noted in Table 2 was in the
mild-to-moderate range (8.7 for MM and 8.4 for BSD arms). Further,
the appropriateness for balloon dilation of the peripheral sinuses was
left to the discretion of the surgeon, without rigidly defined criteria.
Also, the incorporation of adjunctive procedures (although in only a
minority of the patients studied) prevented a complete advocating of
BSD technology as a standalone procedure when compared with
medical therapy. That being said, when these mixed/hybrid cases
were excluded, improvement in the CSS was superior in the BSD arm
compared with MM (p 0.0027).
The BSD arm included surgery performed in the office and OR
settings, with a variety of adjunctive procedures. We chose not to
limit the surgical options of the investigator in an attempt to mirror
real-world practice as closely as possible. It is acknowledged, however, that, given this mixed/hybrid surgical model, we could not
definitively parse out the relative contributions of the balloon dilation
versus the adjunctive procedures with this study design. Although
the study was not powered to detect differences in various subgroup
outcomes, exploratory analyses demonstrated that the change in CSS

O
D

O
N

Figure 2. Chronic Sinusitis Survey (CSS) results from the balloon sinus
dilation (BSD) arm of this study as well as CSS results from historic
endoscopic sinus surgery (ESS) studies.

because we wanted to ensure that all the patients had an aggressive

and recent medical treatment plan. It is not apparent whether the
study by Smith et al.22 required failure of medical therapy to have
occurred within the 90 days before enrollment.
Another strong point of our study was the requirement that enrollment CTs be completed after the failure of a well-defined regimen of
medical therapy. Although this may be the case in other studies as
well, this stipulation has not been previously described. Although it
has been repeatedly shown that CTs do not correlate with QOL
measures,36,37 this specific timing requirement ensured that patients
had demonstrable disease before enrollment.

Limitations
Although the study was comparative, the MM and BSD groups
were not randomized. Given the strictly defined inclusion requirement for intensive medical failure, we believed it would be difficult to
enroll patients who were surgery eligible into a randomized study in

American Journal of Rhinology & Allergy

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm

285

for the patients who had office-based BSD was the same as that for
subjects with BSD performed in the OR (BSD office versus BSD OR,
p 0.33), and the change in CSS for BSD without adjunctive procedures (i.e., pure BSD) remained superior to MM (BSD without
adjunctive procedures versus MM, p 0.0027).
Compliance (follow-up) rates for the primary end point were
87.0% for the BSD group and 71.1% for the MM group. These rates
were similar to those reported in Smith22 (75.8% ESS, 67.9% MM)
and reflective of the general challenges of follow-up in QOL studies. Although there were no statistical differences in baseline characteristics between patients who successfully reported 24-week
CSS values and those who did not, it is not possible to fully
determine whether the patients who were missing would have
reported similar results to those who returned.

14.

15.

16.
17.

CONCLUSION

18.

19.

This was the first prospective multicenter study, to our knowledge,

to compare ongoing MM with sinus surgery by using balloon dilation
instruments for patients with CRS for whom recently strictly defined
intensive medical regimen failed. Patients with mild-to-moderate disease who chose sinus surgery with BSD instruments in a standalone
or mixed/hybrid approach had significantly greater enhancements in
QOL compared with those who elected ongoing medical therapy.

20.
21.
22.

23.

ACKNOWLEDGMENTS
We thank Ryan Shields and Suzanne Wilson of Acclarent, Inc., for
providing programming and data analysis support, and John Ellison
for providing editorial assistance.

24.

T
25.

REFERENCES
1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

286

Rudmik L, Hoy M, Schlosser RJ, et al. Topical therapies in the

management of chronic rhinosinusitis: An evidence-based review
with recommendations. Int Forum Allergy Rhinol 3:281298, 2013.
Subramanian HN, Schechtman KB, and Hamilos DL. A retrospective
analysis of treatment outcomes and time to relapse after intensive
medical treatment for chronic sinusitis. Am J Rhinol 16:303312, 2002.
Lal D, Scianna JM, and Stankiewicz JA. Efficacy of targeted medical
therapy in chronic rhinosinusitis, and predictors of failure. Am J
Rhinol Allergy 23:396400, 2009.
Weiss RL, Church CA, Kuhn FA, et al. Long-term outcome analysis of
balloon catheter sinusotomy: Two-year follow-up. Otolaryngol Head
Neck Surg 139(suppl. 3):S38S46, 2008.
Levine HL, Sertich AP II, Hoisington DR, et al. Multicenter registry of
balloon catheter sinusotomy outcomes for 1,036 patients. Ann Otol
Rhinol Laryngol 117:263270, 2008.
Stankiewicz J, Truitt T, and Atkins J Jr. One-year results: Transantral
balloon dilation of the ethmoid infundibulum. Ear Nose Throat 89:72
77, 2010.
Plaza G, Eisenberg G, Montojo J, et al. Balloon dilation of the frontal
recess: A randomized clinical trial. Ann Otol Rhinol Laryngol 120:
511518, 2011.
Achar P, Duvvi S, and Kumar BN. Endoscopic dilatation sinus surgery (FEDS) versus functional endoscopic sinus surgery (FESS) for
treatment of chronic rhinosinusitis: A pilot study. Acta Otorhinolaryngol Ital 32:314319, 2012.
Cutler J, Bikhazi N, Light J, et al. Standalone balloon dilation versus
sinus surgery for chronic rhinosinusitis: A prospective, multicenter,
randomized, controlled trial. Am J Rhinol Allergy 27:416422, 2013.
Bikhazi N, Light J, Truitt T, et al. Standalone balloon dilation versus
sinus surgery for chronic rhinosinusitis: A prospective, multicenter,
randomized, controlled trial with 1-year follow-up. Am J Rhinol
Allergy 28:323329, 2014.
Halderman AA, Stokken J, Momin SR, et al. Attitudes on and usage
of balloon catheter technology in rhinology: A survey of the American Rhinologic Society. Am J Rhinol Allergy 29:389393, 2015.
Albritton FD IV, Casiano RR, and Sillers MJ. Feasibility of in-office
endoscopic sinus surgery with balloon sinus dilation. Am J Rhinol
Allergy 26:243248, 2012.
Levine SB, Truitt T, Schwartz M, and Atkins J. In-office stand-alone
balloon dilation of maxillary sinus ostia and ethmoid infundibula in

O
D

adults with chronic or recurrent acute rhinosinusitis: A prospective,

multi-institutional study with-1-year follow-up. Ann Otol Rhinol
Laryngol 122:665671, 2013.
Karanfilov B, Silvers S, Pasha R, et al. Office-based balloon sinus
dilation: A prospective, multicenter study of 203 patients. Int Forum
Allergy Rhinol 3:404411, 2013.
Sikand A, Silvers SL, Pasha R, et al. Office-based balloon sinus
dilation: 1-year follow-up of a prospective, multicenter study. Ann
Otol Rhinol Laryngol 124:630637, 2015.
Gliklich RE, and Metson R. Techniques for outcomes research in
chronic sinusitis. Laryngoscope 105(pt. 1):387390, 1995.
Rosenfeld RM, Andes D, Bhattacharyya N, et al. Clinical practice
guideline: Adult sinusitis. Otolaryngol Head Neck Surg 137(suppl.):
S1S31, 2007.
Piccirillo JF, Merritt MG Jr, and Richards ML. Psychometric and
clinimetric validity of the 20-Item Sino-Nasal Outcome Test (SNOT20). Otolaryngol Head Neck Surg 126:4147, 2002.
Lund VJ, and Mackay IS. Staging in rhinosinusitis. Rhinology 31:183
184, 1993.
Lund VJ, and Kennedy DW. Staging for rhinosinusitis. Otolaryngol
Head Neck Surg 117(pt. 2):S35S40, 1997.
Benninger MS, and Senior BA. The development of the Rhinosinusitis
Disability Index. Arch Otolaryngol Head Neck Surg 123:11751179, 1997.
Smith TL, Kern RC, Palmer JN, et al. Medical therapy vs surgery for
chronic rhinosinusitis: A prospective, multi-institutional study. Int
Forum Allergy Rhinol 1:235241, 2011.
Ragab SM, Lund VJ, and Scadding G. Evaluation of the medical and
surgical treatment of chronic rhinosinusitis: A prospective, randomised, controlled trial. Laryngoscope 114:923930, 2004.
Smith TL, Kern R, Palmer JN, et al. Medical therapy vs surgery for
chronic rhinosinusitis: A prospective, multi-institutional study with
1-year follow-up. Int Forum Allergy Rhinol 3:49, 2013.
Quintanilla-Dieck L, Litvack JR, Mace JC, and Smith TL. Comparison
of disease-specific quality-of-life instruments in the assessment of
chronic rhinosinusitis. Int Forum Allergy Rhinol 2:437443, 2012.
Soler ZM, and Smith TL. Quality-of-life outcomes after endoscopic
sinus surgery: How long is long enough? Otolaryngol Head Neck
Surg 143:621625, 2010.
Al Badaai Y, and Samaha M. Outcome of endoscopic sinus surgery
for chronic rhinosinusitis patients: A Canadian experience. J Laryngol Otol 124:10951099, 2010.
Smith TL, Litvack JR, Hwang PH, et al. Determinants of outcomes of
sinus surgery: A multi-institutional prospective cohort study. Otolaryngol Head Neck Surg 142:5563, 2010.
Litvack JR, Griest S, James KE, and Smith TL. Endoscopic and quality-of-life outcomes after revision endoscopic sinus surgery. Laryngoscope 117:22332238, 2007.
Smith TL, Mendolia-Loffredo S, Loehrl TA, et al. Predictive factors
and outcomes in endoscopic sinus surgery for chronic rhinosinusitis.
Laryngoscope 115:21992205, 2005.
Catalano P, and Roffman E. Outcome in patients with chronic sinusitis after
the minimally invasive sinus technique. Am J Rhinol 17:1722, 2003.
Metson R, and Gliklich RE. Endoscopic treatment of sphenoid sinusitis. Otolaryngol Head Neck Surg 114:736744, 1996.
Gliklich RE, and Metson R. Effect of sinus surgery on quality of life.
Otolaryngol Head Neck Surg 117:1217, 1997.
Metson R, and Gliklich RE. Clinical outcome of endoscopic surgery for
frontal sinusitis. Arch Otolaryngol Head Neck Surg 124:10901096, 1998.
Ragab SM, Lund VJ, Scadding G, et al. Impact of chronic rhinosinusitis therapy on quality of life: A prospective randomized controlled
trial. Rhinology 48:305311, 2010.
Wabnitz DA, Nair S, and Wormald PJ. Correlation between preoperative symptom scores, quality-of-life questionnaires, and staging
with computed tomography in patients with chronic rhinosinusitis.
Am J Rhinol 19:9196, 2005.
Hopkins C, Browne JP, Slack R, et al. The Lund-Mackay staging
system for chronic rhinosinusitis: How is it used and what does it
predict? Otolaryngol Head Neck Surg 137:555561, 2007.
DeConde AS, Mace JC, Bodner T, et al. SNOT-22 quality of life domains
differentially predict treatment modality selection in chronic rhinosinusitis. Int Forum Allergy Rhinol 4:972979, 2014.
DeConde AS, Mace JC, Alt JA, et al. Investigation of change in cardinal
symptoms of chronic rhinosinusitis after surgical or ongoing medical
management. Int Forum Allergy Rhinol 5:3645, 2015.
e

O
N

26.

27.

28.

29.

30.

31.
32.
33.
34.
35.

36.

37.

38.

39.

Y
P

O
C

JulyAugust 2016, Vol. 30, No. 4

Delivered by Ingenta to: Guest User IP: 178.62.71.132 On: Tue, 08 Nov 2016 15:23:20
Copyright (c) Oceanside Publications, Inc. All rights reserved.
For permission to copy go to https://www.oceansidepubl.com/permission.htm