Keywords
turbinates - rhinoplasty - rhinitis - allergic - randomized controlled trial - steroids
Introduction
Inferior turbinate surgery, which is a common nasal procedure,[1]
[2]
[3] is indicated in patients with nasal obstruction associated with turbinate hypertrophy
when nonsurgical strategies fail. It is commonly performed in association with procedures
to correct other causes of nasal obstruction, deviated nasal septum, and internal
valve stenosis.[4]
[5]
[6] Rhinoseptoplasty is also often performed concomitantly with inferior turbinate surgery
with the aim of preserving the total breathing area which is surgically reduced, even
though there is no evidence of efficacy regarding this aspect.[7]
[8]
[9]
[10]
[11]
[12]
Allergic rhinitis (AR) is one of the main causes of turbinate enlargement.[13] It is characterized by hypersensitivity induced by immune-mediated inflammation
resulting from the exposure of the nasal mucosa to specific allergens.[4]
[14]
[15] Symptoms of AR include rhinorrhea, nasal obstruction, sneezing, and postnasal drip
that resolve spontaneously or following treatment.[15] The inferior turbinate mucosa plays a central role in the pathophysiology of AR.[16]
[17] The continuous allergic process causes engorgement of the turbinate tissue, producing
obstructive symptoms. In addition, the inflammatory process affecting the turbinate
epithelium and submucosa has also been linked to non-obstructive symptoms such as
nasal secretion, discharge, and sneezing.[16]
Considering the major role played by the inferior turbinates in AR, it would be fair
to hypothesize that surgical reduction of the tissue affected by inflammation might
curtail the allergic process, in addition to increasing nasal patency.[15]
[16] However, the literature still lacks controlled studies investigating this topic.
The high prevalence of AR, which may affect up to 40% of the world population, with
an economic impact close to US$ 4.5 billion reported in the United States,[14] warrants the search for treatment alternatives/strategies. Therefore, the present
study aims to evaluate the impact of inferior turbinate surgery on non-obstructive
allergic symptoms (NOAS) in patients with AR.
Materials and Methods
We performed a secondary analysis of unpublished data from two double-blind randomized
controlled trials (RCTs) previously performed by our research group to evaluate the
clinical impact of inferior turbinate reduction on the quality of life (QoL) of patients
undergoing primary rhinoseptoplasty.[9]
[10] These RCTs were designed to measure the relative change in QoL scores obtained with
the validated Brazilian Portuguese version of the Nasal Obstruction Symptom Evaluation
(NOSE-p) scale.[18]
The present secondary analysis was granted approval by the Research Ethics Committee
at our institution (number 20180109), and investigators signed a nondisclosure agreement
for the use of secondary data.
RCT Sample and Protocol
Details of the RCTs have been described elsewhere.[9]
[10] Briefly, patients with an existing diagnosis of AR were recruited at the ear, nose,
and throat (ENT) outpatient clinic (Facial Plastic Surgery Division, Otology and Head
& Neck Surgery) of a tertiary university hospital (Hospital de Clínicas de Porto Alegre)
in southern Brazil. Individuals aged ≥ 16 years, with indication for primary rhinoseptoplasty
for functional and/or esthetic reasons were eligible. The exclusion criteria were
previous nasal surgery, nasal obstruction resulting strictly from inferior turbinate
enlargement, and/or concomitant surgical procedures, such as functional endoscopic
sinus surgery, adenoidectomy, blepharoplasty, or otoplasty.
In the study by Lavinsky-Wolff et al.,[9] 50 patients were randomized to receive submucosal electrocauterization of the inferior
turbinates (25 patients) associated with rhinoseptoplasty, or rhinoseptoplasty only
(25 patients). In the study by De Moura et al.,[10] 50 patients were randomized to receive endoscopic partial inferior turbinectomy
associated with rhinoseptoplasty (25 patients), or rhinoseptoplasty only (25 patients).
Both studies employed a randomization sequence generated using a web-based tool (http://randomization.com/) by an independent investigator. A 1:1 allocation ratio was used to randomize patients
with or without inferior turbinate surgery with blocks including 4 to 6 participants.
Investigators involved in the recruitment and evaluation of the patients were blinded
to allocation sequence.
In both RCTs, participants were evaluated pre and postoperatively using a standard
protocol administered by trained investigators who were blinded to the study intervention.
Evaluations were performed preoperatively and at 7, 30, 60, and 90 days postoperatively.
The diagnosis of AR was based on the presence of two or more of the following symptoms
(as stated in the 2008 Allergic Rhinitis and its Impact on Asthma [ARIA] guidelines):
watery rhinorrhea, sneezing, nasal obstruction, nasal pruritus, and allergic conjunctivitis
for more than 1 hour on most days.[15] Patients were asked about symptoms of AR at all follow-up visits, and were classified
according to symptom intensity and frequency into an intermittent or persistent category
and a mild or moderate/severe category according to the ARIA guidelines.[15]
Starting on postoperative day 30, a topical nasal steroid (budesonide 100 micrograms
twice a day) was prescribed for 30 days to all patients with mild and persistent symptoms
and to those with moderate/severe symptoms.[15] From postoperative day 60 to 90, budesonide 100 micrograms once a day was prescribed
if symptoms improved, or budesonide 200 micrograms twice a day if symptoms persisted.
All patients were advised to use h1-antihistamines as required.
At baseline, the NOSE-p scale[18] was administered to all patients to assess outcomes in nasal obstruction in the
trials. A score of 0 indicates no problems with nasal obstruction, and a score of
100 indicates the worst possible problems with nasal obstruction.
At all follow-up visits, the patients answered a standardized questionnaire about
medications used for AR. The questionnaire included questions about whether the patient
was using topical nasal corticosteroids and whether there was a need for oral antihistamine
medication in the past 30 days.
Assessment of NOAS
In the present study, NOAS were evaluated in RCT participants. As a first step, data
from each trial were analyzed separately. In the turbinate surgery group (which included
50 patients), 25 patients underwent submucosal electrocauterization and 25 patients
underwent partial inferior turbinectomy. These data were combined for analysis, with
patients who had undergone inferior turbinate surgery forming an intervention group
and patients receiving only rhinoseptoplasty analyzed as a control group ([Fig. 1]). Non-obstructive allergic symptoms at 90 days were compared between the groups.
Also, NOAS recorded before surgery were compared with NOAS at 90 days in both groups.
Fig. 1 Study protocol.
Primary Outcome Measure
Patients were asked about clinical symptoms of AR, as stated in the ARIA guidelines,[15] at all follow-up visits using a standard protocol administered by trained investigators
who were blinded to the study intervention. Non-obstructive allergic symptoms were
defined as discharge, sneezing, itching, or allergic conjunctivitis lasting at least
1 hour and occurring on most days.
The main outcome measure in the present study was a composite variable defined as
the presence of one or more NOAS at baseline and 90 days postoperatively.
Secondary Outcome Measure
The frequency of topical nasal steroid and of oral antihistamine use 90 days postoperatively
was analyzed as a secondary outcome measure.
Statistical Analysis
Statistical analysis was performed with IBM SPSS Statistics for Windows, Version 20.0
(IBM Corp., Armond, NY, USA). Data were presented as means, standard deviation, number
or percentage. Statistical significance was established at p = 0.05. Continuous variables were compared using the Student t-test. The chi-squared test was used for comparison of non-parametric variables. The
frequency of allergic symptoms at 90 days postoperatively was analyzed using a Poisson
regression model for robust variance adjusted for the use of topical nasal steroid
and oral antihistamine. A generalized estimating equation (GEE) was used to compare
the frequency of each allergic symptom preoperatively and at 90 days postoperatively.
Results
One hundred patients were analyzed, 50 in the intervention group and 50 in the control
group ([Fig. 1]). In each group, 3 patients were lost to follow-up, adding to a total of 6 patients
(6%). Preoperatively, most patients reported moderate/severe allergic symptoms and
nasal obstruction. The mean age was 34 ± 13 years in the intervention group and 34 ± 16
years in the control group. There were no differences in the mean NOSE score, age,
sex, years of schooling, nasal obstruction, or intensity of ARIA symptoms between
the groups at baseline ([Table 1]).
Table 1
Characteristics of the intervention and control groups
|
Characteristic
|
Intervention group (n = 50)
|
Control group (n = 50)
|
|
Mean age (SD), years
|
34 (13)
|
34 (16)
|
|
Female gender, no. (%)
|
27 (54)
|
26 (52)
|
|
Years of schooling, no. (%)
|
|
≤ 8
|
13 (26)
|
18 (36)
|
|
9–11
|
24 (48)
|
22 (44)
|
|
≥ 12
|
13 (26)
|
10 (20)
|
|
NOAS
|
|
Sneezing
|
23 (46)
|
23 (46)
|
|
Discharge
|
17 (34)
|
25 (50)
|
|
Itching
|
20 (40)
|
24 (48)
|
|
Allergic conjunctivitis
|
9 (18)
|
12 (24)
|
|
Nasal obstruction
|
47 (94)
|
48 (96)
|
|
Nose score, mean (SD)
|
70.2 (24.4)
|
78.7 (16.4)
|
|
ARIA symptom intensity, no. (%)
|
|
Mild
|
5 (10.2)
|
4 (8.2)
|
|
Moderate-severe
|
45 (89.8)
|
46 (91.8)
|
|
Seasonal symptoms, no. (%)
|
36 (72)
|
34 (68)
|
|
Worst in, no. (%)
|
|
Winter
|
21 (42)
|
22 (44)
|
|
Spring
|
7 (14)
|
7 (14)
|
|
Other
|
8 (16)
|
5 (10)
|
|
Use of topical nasal steroid at baseline, no. (%)
|
22 (44)
|
24 (48)
|
|
Deviated septum, no (%)
Amount nasal obstruction caused by deviated septum, no. (%)
|
48 (96)
|
45 (90)
|
|
< 25%
|
1 (2)
|
0
|
|
25–50%
|
4 (8)
|
7 (14)
|
|
50–75%
|
14 (28)
|
14 (28)
|
|
> 75%
|
29 (58)
|
24 (48)
|
Abbreviations: ARIA, Allergic Rhinitis and its impact on Asthma; NOAS, non-obstructive
allergic symptoms; NOSE, Nasal Obstruction Symptom Evaluation; SD, standard deviation.
Septal deviation was diagnosed in 93 participants (93%) – 48 in the intervention group
and 45 in the control group ([Table 1]). The groups were similar regarding the severity of septal deviation. Patients were
submitted to rhinoseptoplasty regardless of the presence of septal deviation to harvest
cartilage grafts for the procedure.
When each allergic symptom was analyzed independently, a reduction in the intragroup
frequency was noted 90 days postoperatively versus preoperatively (p < 0.01) for all symptoms, except allergic conjunctivitis (intervention group, p = 0.112; control group, p = 0.253). There was no difference in the frequency of each symptom between the groups
at 90 days ([Table 2]). Regarding patients with seasonal complaints, there was no difference between the
groups in the number of patients with exacerbations either at baseline (p = 0.389) or 90 days postoperatively (p = 0.881).
Table 2
Frequency of non-obstructive allergic symptoms in the intervention and control groups
prior to surgery and 90 days postoperatively
|
Symptom, n (%)
|
Intervention group (n = 50)
|
Control group (n = 50)
|
|
|
Preoperative
|
90 days
|
p time
|
Preoperative
|
90 days
|
p time
|
p 90 days
|
|
Sneezing
|
23/50 (46)
|
7/47 (15)
|
< 0.01
|
23/50 (46)
|
11/47 (23)
|
< 0.01
|
0.216
|
|
Itching
|
20/50 (40)
|
9/47 (19)
|
< 0.01
|
24/50 (48)
|
10/47 (20)
|
< 0.01
|
0.783
|
|
Discharge
|
17/50 (34)
|
11/47 (23)
|
< 0.01
|
25/50 (50)
|
8/47 (17)
|
< 0.01
|
0.843
|
|
Allergic conjunctivitis
|
12/50 (24)
|
7/47 (15)
|
0.137
|
9/50 (18)
|
5/47 (10)
|
0.253
|
0.128
|
Generalized estimating equation (GEE) for comparisons between the groups. (p time, intragroup difference in pre and postoperative frequency); (p 90 days, between-group
difference at 90 days postoperatively).
Both the use of topical steroids and the presence of NOAS were similar in the two
groups prior to the surgery. At 90 days, the groups remained similar regarding nasal
obstruction (p = 0.769) as well as regarding NOSE scores (p = 0,769). The same was true for NOAS at 90 days (p = 0.835), even after adjustment for topical nasal steroid and oral antihistamine
use (p = 0.899) ([Table 3]).
Table 3
Frequency of participants with one or more non-obstructive allergic symptoms and medication
use at 90 days postoperatively
|
Variable, n (%)
|
Intervention group (n = 47)
|
Control group (n = 47)
|
p
|
|
One or more NOAS
|
20/47 (42)
|
21/47 (44)
|
0.835[a]/0.899[b]
|
|
Topical nasal steroid (yes)
|
16/47 (34)
|
28/47 (59)
|
0.014[a]/0.037[b]
|
|
Oral antihistamine (yes)
|
2/47 (4)
|
10/47 (20)
|
0.016/ [a]0.030[b]
|
Abbreviation: NOAS, non-obstructive allergic symptoms.
p: Poisson regression for robust variables.
a unadjusted.
b adjusted for steroid and oral antihistamine use. p < 0.05, Pearson χ2.
At 3 months postoperatively, nasal spray use was less frequent in the intervention
group versus controls (34% versus 59% respectively; p = 0.014). A protective effect (relative risk [RR]: 0.35; 95%CI: 0.151–0.810) was
observed for steroid nasal spray in the intervention group. The frequency of oral
antihistamine use was also lower in the intervention versus the control group (2%
versus 23%; p = 0.016). As a result, a protective effect for medication use was detected in the
intervention group (RR: 0.145; 95%CI: 0.030–0.699) ([Table 3]).
Discussion
The present study is the third by our group evaluating the impact of inferior turbinate
surgery in patients submitted to rhinoseptoplasty. Lavinsky-Wolff et al.,[9] in 2013, and De Moura et al.,[10] in 2018, focused on QoL related to nasal obstruction measured by the NOSE scale.
In the present study, a secondary analysis was performed by aggregating data from
these two previous RCTs, focusing on NOAS. In addition to improvement in nasal obstruction,
patients undergoing inferior turbinate surgery also often report improvement in all
AR symptoms. However, analysis of this relationship is complex and confounded by topical
steroid use and by the difficulty in assessing these symptoms objectively.
To the best of our knowledge, the present study is the first to evaluate in a controlled
manner, albeit based on secondary data, the impact of nasal turbinate surgery on NOAS
in patients undergoing rhinoseptoplasty. The data analyzed in this study were produced
with randomization and blinding of participants and investigators and having a control
group of patients undergoing rhinoseptoplasty without inferior turbinate reduction.
Previous RCTs[19]
[20] have compared several interventions for turbinate reduction with the aim of evaluating
the impact on NOAS and nasal obstruction, and noted improvement in NOAS as well as
obstruction symptoms.
An important finding of the present study was a lower frequency of topical nasal steroid
use in the intervention group observed 90 days after the procedure. The use of steroids
is a potential confounding factor. Nevertheless, for ethical reasons, the medication
was provided to patients with persistent or moderate-severe intensity AR symptoms.
To control for this potential bias, a single prescription protocol was used for all
participants in both RCTs, and contemplated in our regression model.
Also noteworthy was the lower frequency of antihistamine use at 90 days postoperatively
in the intervention group. The use of this medication is also a potential confounding
factor; however, antihistamines were provided to participants, as well as nasal steroids.
Patients were guided to use the medication in the presence of symptoms, with the frequency
of use informed and recorded during the follow-up consultations.
Measuring the efficacy of an intervention in multifactorial diseases is challenging
and further complicated by the use of one or several medications that directly impact
symptoms, producing changes in the intensity of clinical complaints.[9] For that reason, the RCTs on which the present study was based were designed as
pragmatic trials, which measured effectiveness and included real-world patients recruited
from an outpatient clinic, influenced by aspects of their usual routine.
A possible explanation for the impact of inferior turbinate surgery on NOAS could
be found in the pathophysiology of AR. Turbinate epithelium is a key site for the
genesis of allergic reactions.[13] Consequently, tissue reduction might lead to fewer/less intense symptoms.[16]
[17] Apparently, resection of autonomic nervous in the submucosa, which are responsible
for the control of vessel engorgement and mucus and immunoglobulin production, leads
to improvement in sneezing and rhinorrhea.[16] The replacement with scar tissue contributes to the improvement in discharge.[16] In a study comparing the histopathological characteristics of two groups with AR
submitted to turbinectomy, Lukka et al.[13] found restoration of a histological pattern that was similar to that of patients
without the disorder.
There is no consensus regarding the ideal technique for inferior turbinate surgery.[2] The most widely used techniques rely on “cold” volume reduction, such as partial
turbinectomy, and on thermal methods, such as electrocauterization.[21] The underlying rationale is usually to avoid aggressive reductions in tissue volume
to prevent empty nose syndrome.[2]
[22]
[23] In the present study, no differences were detected in the frequency of NOAS at 90
days regardless of the type of technique employed for turbinate reduction.
The improvement in NOAS detected in our population supports previous findings from
the literature. Mori et al.[16]
[24] showed evidence of improvement not only in nasal obstruction following submucosal
turbinectomy, but also in sneezing, discharge, and itching. Chen et al.[19] and Caffier et al.[25] reported similar results. More recently, Hamerschmidt et al.[26] studied a cohort including two groups with inferior turbinate hypertrophy undergoing
turbinoplasty, one with NOAS and the other with nasal obstruction. The group with
allergy symptoms had significant reduction of NOAS after turbinoplasty, as compared
with the preoperative period. It should be noted, however, that none of these studies
involved randomization of participants and blinding of participants and investigators.
There was no difference in the frequency of NOAS between our groups at 90 postoperative
days. This might be linked to the increased frequency of clinical treatment of AR
in the control group. To control for this potential confounding factor, we performed
an analysis adjusted for topical steroid and oral antihistamine use. Another possible
explanation for the improvement in NOAS is the potential influence of septal deviation
treatment on allergic symptoms. According to Kim et al.,[27] septoplasty may have a positive impact on NOAS in patients with a diagnosis of AR.
That study reported an improved postoperative score obtained with Rhinasthma,[28] a validated instrument for assessment of AR and asthma, in a group submitted to
septoplasty associated with submucosal turbinoplasty in comparison to a group submitted
only to submucosal turbinoplasty. In theory, normalization of air flow could reduce
inflammation in the nasal cavities, impacting NOAS.
Interestingly, in our current analysis, we found that patients undergoing turbinate
surgery used fewer topical corticosteroids and fewer antihistamines 90 days after
the procedure. The latter was not evident in the data from the individual trials that
make up our study[9]
[10] ([Fig. 2]). It is possible that these patients were actually using these medications for nasal
obstruction, and not for NOAS, which may have shown more improvement in the inferior
turbinate group. We did not include measurements of nasal patency, so it is also possible
that discrepancies in acoustic rhinomanometry could have some influence in our findings.
It is also possible that turbinate surgery could have accelerated the clinical improvement
of AR symptoms; however, this contrasts with the lack of between-group difference
in NOAS observed in our study.
Fig. 2 Forest plot showing risk of symptoms and medication use 90 days after nasal turbinate
reduction.
Despite the strict methodological criteria employed in the present study, some limitations
must be addressed. The RCTs whose data we analyzed were designed for assessment of
nasal obstruction-related QoL. Information regarding NOAS was collected using the
ARIA guidelines[15] questionnaire designed for clinical diagnosis of AR, but visual/verbal analogue
scales were not used to gauge the intensity of symptoms . Also, we did not use any
laboratory test for AR diagnosis. Our patient follow-up was limited to 90 days after
the procedure. However, it is generally assumed that at 3 months there is maximal
scar tissue contraction; therefore, if no difference is NOAS was detected at 90 days,
it seems unlikely that this difference would emerge in the long term.[10] Another issue may be the sample size, which was originally calculated to detect
differences in the NOSE scale, but not in the AR outcomes. Given the frequencies of
NOAS that were observed, our study is powered to detect a 38% reduction in the incidence
of NOAS (assuming a baseline incidence of 21/47, one-tailed chi-square test). Because
the incidences in the control and intervention groups were virtually the same, we
believe that our observation is correct. However, a larger study would be required
to corroborate our findings.
Conclusion
Turbinate reduction performed in association with rhinoseptoplasty did not reduce
the frequency of non-obstructive allergic symptoms 90 days after the procedure more
than rhinoplasty alone. However, the observed decrease in nasal steroid and oral antihistamine
use suggests an impact of turbinate reduction on medication use in patients with AR
undergoing rhinoseptoplasty.
