Septoplasty: is it possible to identify potential “predictors” of surgical success?
Septoplasty is one of the most frequent surgical procedures performed by otolaryngologists. Despite successful surgical correction, many
patients are not satisfied with their outcomes. So far, in clinical practice there is no consensus of opinion about the reliability of objective
measurements of nasal patency and the correlation between objective measurements and subjective nasal patency symptoms. This study
aims to assess the reasons for patient dissatisfaction after septoplasty and optimise pre-operative diagnostic management to predict surgical
outcomes. We analysed 494 patients undergoing septoplasties with turbinoplasty by subjective Nasal Obstruction Symptom Evaluation
questionnaire (NOSE) and objective active anterior rhinomanometric measurements before surgery and after 6 months. In our series, 17%
had postoperative septal re-displacement; all patients had an anterior deviations at baseline. We found that the type of septal deviation,
anterior vs posterior, was a significant predictor of postoperative functional improvement, whereas demographic characteristics as age,
gender and smoke habit were not. Our data suggest that the anterior segment of the nasal septum was the most critical area for nasal airway
resistance and more difficult to manage because it is likely to re-displace vs the posterior one and for this reason it represents a negative
predictor of postoperative satisfaction.
Nasal obstruction is one the most common symptoms observed in rhinological practice 1 and septoplasty is one of the most frequent surgical procedures performed by otolaryngologists 2. Despite successful surgical correction, many patients are not satisfied with their postoperative outcomes, which may lead to medicolegal problems 2.
So far, to avoid postoperative dissatisfaction, surgeons have developed a variety of techniques, but sometimes, regardless of the surgical technique, nasal obstruction can persist 3.
On the other hand, inadequate surgical procedures lead patients to re-consult an otolaryngologist. This happens because there are often many underestimated causes for nasal obstruction, such as allergy, nasal valve collapse or lateral wall insufficiency, which may alter the physiological dynamics of nasal airflow 1.
In addition to persistent septal deviation, septal perforation, synechiae, loss of nasal dorsum and tip support are the main reasons for persistent postoperative nasal obstruction and represent the most frequent complications after septoplasty 2 4.
In clinical practice there is still no consensus of opinion about the reliability of objective measurements of nasal patency and the correlation between objective measurements and subjective symptoms reported by patients 5. Indeed, there is no diagnostic tool to predict postoperative outcomes, objectively evaluate the geometry of the nose and measure nasal resistances and degree of obstruction. For instance, several studies have shown no correlation between objective measurements and patient satisfaction, whereas other studies did not use validated questionnaires to assess postoperative outcomes after septoplasty 5.
To the best of our knowledge, this is the first prospective study examining, in a large sample size, the medium-term outcomes of septoplasty in terms of both objective and subjective assessment, with the use of active anterior rhinomanometry (AAR) and a validated questionnaire specifically designed for nasal obstruction, the Nasal Obstruction Symptoms Evaluation (NOSE).
Materials and methods
This is a prospective cohort study including 537 consecutive patients who underwent septoplasties with turbinoplasty performed at the ear nose and throat (ENT) Departments of the “Federico II” University and Cardarelli Hospital in Naples. Patients were enrolled between 2010 and 2017. We analysed 494 septoplasties that completed the study with at least 6 months follow-up; there were 284 females and 210 males, 18 to 50 years old (Table I). Enrolled subjects gave informed consent to the study, which was approved by the local Board of Medical Ethics. The study was carried out according to the Declaration of Helsinki.
Inclusion criteria were: patients complaining of nasal obstruction due to septal deviation and inferior turbinate hypertrophy who underwent septoplasty and turbinoplasty; nasal endoscopy confirming septal deviation and negative for chronic rhinosinusitis; previous maxillofacial computed tomography scan (CT) confirming septal nasal deviation and negative for chronic rhinosinusitis.
Exclusion criteria were: age < 18 years, previous nasal surgery, patients requiring rhinoseptoplasty, malformative and neuropsychiatric diseases, previous diagnosis of chronic rhinosinusitis (cystic fibrosis, immotile cilia syndrome etc.), uncontrolled rhinitis and asthma (to prevent interference of medical therapy with surgical and functional outcomes). Furthermore, we excluded patients with serious pathologies such as previous head or neck radiotherapy, immunologic disorders, diabetes, heart disease, neoplasms etc.
Before surgery (T0) enrolled subjects underwent:
We distinguished anterior and posterior septal deviation in relation to the nasal valve (Table I). In particular, we considered anterior the deviations corresponding to the dorso-caudal septal cartilage as previously described 10 11.
Enrolled subjects underwent septoplasty and turbinoplasty as described by Cottle and later by Sulsenti 12 13 under general anaesthesia by the same three experienced consultants (MI, EC, FR). The maxilla-premaxilla approach combining mobilisation and/or removal of any deranged portion of the bony and/or cartilaginous septum, followed by reconstruction of the septum support and the submucous inferior turbinoplasty (without bony resection) with lateral out-fracture of inferior turbinates were performed 13. The septum was fixed through trans-septal sutures and the septal caudal incision was closed. Nasal packing was used and removed 48 hours after surgery.
After removal of nasal packing, all patients subjectively recovered normal nasal flow; they were treated with intranasal saline irrigation alone, twice a day for 20 days.
At 6 months’ post-operative evaluation (T6), patients repeated endoscopy to assess septal displacement and rhinomanometric evaluation, and were asked to fill out the Nose Scale questionnaire. As reported in the literature, we considered surgical success as a significant decrease in the postoperative NOSE score 8.
Statistical analyses were performed using the SPSS-PC program (version 16; SPSS Inc., Chicago, IL). Three dependent AAR variables were considered: nasal resistance in the wider nasal cavity (WNR), nasal resistance in the narrower nasal cavity (NNR), and total nasal resistance (TNR), each recorded at sample pressures of 150 Pa 14. A Student’s t test was used to evaluate differences in nasal obstruction metrics at T0 and T6. To evaluate the relationship between nasal obstruction metrics and patient characteristics using linear regression we created a multivariable model. Pearson’s correlation was used to determine the correlation between two numerical variables (Wessa P. Free Statistics Software, Office for Research Development and Education, version 1.1.23-r7, 2017L; www.wessa.net).
For data analysis, we used the Δ, i.e. the difference in values detected at baseline and at T6 (Δ = T0-T6).
A p-value was considered statistically significant if < 0.05.
At 6 months’ post-operative evaluation, we observed septal re-displacement (R) in 84 of (17%) 494 enrolled patients. According to the type of septal deviation, we observed septal re-displacement in 84 out of 312 anterior septal deviation patients; whereas we observed no displacement in patients with posterior septal deviation at baseline. Furthermore, we observed the following complications: 3 (0.6%) infections, 6 (1.2%) haemorrhage, 12 (2.5%) valvular and 82 (16.5%) intra-nasal synechias, 6 (1.2%) perforations and 13 (2.6%) haematomas.
The analysis demonstrated that the type of deviation (D) anterior (a) vs posterior (p) and the occurrence of a re-displacement (R) at T6 represented significant (p < 0.01) predictors of the change (= Δ, i.e. differences between parameters at T0 and T6) in subjective, ΔNOSE (T0NOSE-T6NOSE) and objective, ΔTNR (T0TNR-T6TNR), ΔWNR (T0WNR-T6WNR), ΔNNR (T0NNR-T6NNR) measures. Only for ΔNNR did the presence of allergy represent an additional significant predictor (Table II).
By applying Pearson’s test, we found no correlation between ΔNOSE, ΔTNR, ΔWNR, ΔNNR and age, (respectively r = 0.021; r = – 0.026; r = – 0.013; r = 0.041. p > 0.05).
To evaluate whether the patients’ starting point impacted postoperative functional outcomes, we used a t-test for paired samples comparing T0 and T6 values of NOSE, TNR, WNR, NNR. We found that parameters (mean ± SD) improved (p < 0.01) in all subjects (T6NOSE 24.06 ± 19 vs T0NOSE 49.60 ± 20; T6TNR 0.25 ± 0.15 vs T0TNR 0.50 ± 0.23; T6WNR 0.39 ± 0.14 vs T0WNR 0.57 ± 0.14; T6NNR 0.44 ± 0.16 vs T0NNR 0.76 ± 0.22).
To evaluate whether the type of deviation (D), anterior (a) vs posterior (p), the occurrence of displacement (R), allergy status (A) and smoking (s) habit (independent variables) influenced the improvement (Δ) of each outcome measure (dependent variable), we used a t test.
All obstruction metrics improved (p < 0.01) after surgery in both anterior and posterior deviations, although the change (Δ) was more relevant (p < 0.01) in the anterior-a- (aΔNOSE: 28.16 ± 24.7; aΔTNR: 0.35 ± 0.28; aΔWNR: 0.27 ± 0.19; aΔNNR: 0.39 ± 0.28) than in the posterior-p- ones (pΔNOSE: 21.07 ± 12.5; pΔTNR: 0.08 ± 0.16; pΔWNR: 0.02 ± 0.1; pΔNNR: 0.17 ± 0.20).
Our data demonstrated that allergic (A) subjects presented a more statistically relevant change (Δ) (p < 0.05) only for AΔWNR (0.16 ± 0.19) and AΔNNR (0.28 ± 0.26) than non-allergic (NA) ones (NAΔNNR: 0.36 ± 0.28; NAΔWNR: 0.20 ± 0.22).
In addition, we found no differences (p > 0.05) between smokers – s – (sΔNOSE: 27.48 ± 19.8; sΔTNR: 0.27 ± 0.28; sΔWNR: 0.19 ± 0.20; sΔNNR: 0.30 ± 0.27) and non-smokers – ns – (nsΔNOSE: 24.77 ± 21.91; nsΔTNR: 0.24 ± 0.27; nsΔWNR: 0.17 ± 0.20; nsΔNNR: 0.31 ± 0.27).
We found a positive significant (p < 0.05) correlation between NOSE and AAR measures. For instance, in all sample populations we found a positive significant (p < 0.05) correlation between T0NOSE and T0TNR (r: 0,58), T0WNR (r: 0,51), and T0NNR (r: 0,57), and between T6NOSE and T6TNR (r: 0,43), T6WNR (r: 0,27), and T6NNR (r: 0,40) (Figs. 1, 2). In the posterior deviations we also found a positive significant (p < 0.05) correlation between T0NOSE and T0TNR (r: 0,59) (Fig. 3), T0WNR (r: 0,27), and T0NNR (r: 0,53), and between T6NOSE and T6TNR (r: 0,18). Conversely, we found no correlation between T6NOSE and T6WNR (r: 0,06), and T6NNR (r: – 0,03).
In anterior deviations, we found a significant positive correlation (p < 0.05) between T0NOSE and T0TNR (r:,38), T0WNR (r: 0,44) and T0NNR (r: 0,45), and between T6NOSE and T6TNR (r: 0,55), T6WNR (r: 0,61), T6NNR (r: 0,53) (Fig. 4). In the re-displaced septoplasties, we found no correlation between T6NOSE and T6TNR (r: 0,012), T6WNR (r: 0,09), and T6NNR (r: 0,12) (Fig. 5).
Comfortable nasal breathing is a condition that is directly related to QoL. Currently, there is evidence that the patient’s perspective on treatment outcomes is a crucial element for improving high-quality care and patient-rated therapeutic outcomes in terms of symptoms, and that it can provide a much more realistic picture of the effectiveness of a treatment than those of objective outcomes 15.
For instance, symptoms score questionnaires provide valuable information about how patients perceive the severity of their nasal obstruction and, consequently, about the degree of postoperative satisfaction 15.
Sometimes, after surgery nasal obstruction can persist with consequent dissatisfaction and poor QoL 2; other times persistent septal deviations are associated with good patient satisfaction, whereas completely corrected deviations are not.
According to the literature, the risk of relapse or re-displacement after septoplasty is a common condition. Very often, an excellent postoperative result immediately after surgery evolves in a partial or total failure a few months later on the reappearance of deviation to some degree. This mostly depends on septal cartilaginous structures, which tend to retain the “memory” of the deviation due to their elasticity and so, over time, tend to curve again 16.
This phenomenon is the result of two, extrinsic and intrinsic, forces acting on the cartilaginous nasal septum and causing, if they are not released during surgery, relapse. The extrinsic forces are exerted on by deviated nasal bony structures, whereas the intrinsic forces are due to the impaired growth of septal cartilage or to trauma deforming tissue ultrastructure 16. Another common cause of septal displacement is the difficult repositioning of the caudal margin of the quadrangular cartilage along its bony rail 13.
Thus, despite structural differences, all relapsed patients will predictably share certain features: surgically distorted native anatomy, disrupted tissue planes, altered microcirculation and potentially postsurgical mucosal dysfunction 17.
At baseline anterior deformities not only prevailed (Table I) in our surgical population, but also showed higher NOSE and ARR scores suggesting the prevalence of anterior deviations in candidates for septoplasty and a higher impact on symptom severity and nasal resistances than posterior ones.
Furthermore, our data demonstrated that the type of deviation, anterior vs posterior, represented significant predictors of the change (Δ) both in subjective and objective measures. According to the literature, these findings suggest a role of the type of deviation as a potential predictor for surgical success in terms of symptoms score improvement 18. The study of Shiryaeva et al. using only subjective scores demonstrated that the type of septal deviation, unilateral vs bilateral, and preoperative obstruction scores, may aid in predicting outcomes of nasal surgery. However, they did not evaluate the anterior vs posterior septal deviation and did not use objective measurements 18.
In our population, the presence of allergy represented an additional significant predictor; our findings demonstrated that allergic subjects presented a more relevant change (Δ) in ARR (ΔWNR and ΔNNR) scores than non-allergic (NA) ones, probably for worse starting scores.
In allergic subjects, the hypertrophy of turbinates plays a primary role in the genesis of nasal obstruction, but while septoplasty corrects the septal deviation, literature data has shown that it is unclear whether turbinoplasty is beneficial in the long term for treatment of nasal obstruction. Recent studies reported improvement in symptoms and subjective evaluation at early post treatment evaluation and decreased benefits at the long term observation 19. As allergy is a chronic inflammatory disease, in the long-term it tends to cause turbinate hypertrophy even after surgery 20. Our data showed worse baseline AAR subscores in allergic individuals, although they did not clarify the impact of the allergy on septoplasty outcomes.
In any case, these data pointed out the utility of studying preoperative allergic status as an additional putative predictive factor for surgical outcomes 19.
Based on our results, smoking status, as well as age and gender, were not crucial as predictors of postoperative results. In particular, our findings confirmed recent research demonstrating that both smokers and non-smokers benefit from septoplasty 21.
At T6 we found that all parameters improved (p = 0.01) in the overall population, in agreement with the literature (satisfactory outcomes in 65% to 80% of patients) 17.
So far, the correlation between subjective nasal patency symptoms and objective nasal measurements is still debated in the literature. For instance, some authors reported the presence of correlation, whereas others did not 22 23.
Hong et al. pointed out that only the preoperative NOSE scale, reflects post-septoplasty outcomes. Although the study by Hong provided useful information, it included a non-homogeneous (43 males and 6 females), small sample of patients undergoing septoplasty with or without turbinoplasty treated with different surgical techniques and different follow-up times (1, 3 or 6 months). Furthermore, the authors used acoustic rhinometry as an objective measurement that produces detailed evaluation of the geometry of nasal cavities, but it does not evaluate the ﬂow ﬁeld or provide any information on the physiology of nasal pressure 1.
In our study population, we found a significant correlation between subjective and objective measures (Figs. 1-5), whereas among re-displaced anterior deviations, we did not find any correlation at T6 (Fig. 5).
This lack of correlation, again in accordance with the literature, pointed out that the AAR provides only a global assessment and does not consider the local details of the nasal air flow (i.e. the anterior regions of inferior meatus or nasal valve) that are often extremely significant from clinical and QoL standpoints 22 25.
From a merely speculative point of view, we can hypothesise that re-displacement occurs almost exclusively in the nasal valve region, where the ARR presents some limitations in the evaluation of nasal air flow 22 23; in such cases a dilation test could be useful 26.
On the basis of these observations, in our opinion surgeons should always perform both objective and subjective nasal measurements before and after septoplasty, even if the predictive value of these methods is still controversial 22 23.
In our clinical setting we always administer the subjective NOSE questionnaire and the objective ARR. This assessment helps us to evaluate the nasal airways and to select patients who would benefit most from septoplasty.
Literature data have also demonstrated that patients with significant nasal obstruction may have a small septal deviation, whereas other patients with severe anatomical deformity may have mild symptoms 27.
Therefore, the goal of septoplasty should not be to obtain a perfectly midline aligned septum, but to improve breathing, making the nasal cavities pervious 13.
Consequently, in clinical practice the presence at nasal endoscopy and/or imaging as CT of non-perfectly aligned portions of the septum or the presence of septal spurs in different nasal subsites that do not impact nasal airflow (posterior regions of nasal cavities, high regions of the septal cartilage) should not be considered a surgical failure, especially in association with improvement in objective and subjective measures. Although preoperative CT scan can be used for further examination of nasal anatomy and can be helpful in identifying ancillary sinonasal pathologies, there is little correlation between septal deviation findings on CT scan and symptoms of nasal obstruction as demonstrated by previous studies that did not support a role for CT as a clinically meaningful or necessary test to investigate residual postoperative nasal obstruction 28 29.
Since patients generally tend to rate the results of their septal surgery less positively as the postoperative period gets longer, it should be mandatory to evaluate surgical outcomes in the long term; although the most literature data on septoplasty outcomes reports postoperative results after 6 months 27 30 31.
Another limitation of this study is the lack of nasal decongestion, but given the large sample size reported in the study the decongestion test was not available for all patients. However, in our opinion this limitation does not affect the validity of the study. Indeed, some authors reported a mean nasal airway resistance reduction after decongestion only of 33% 32 and many others do not perform decongestion in nasal assessment 33 34.
In conclusion, our data suggest that anterior septal deviations significantly impact the QoL, they are more difficult to manage and disposed to re-dislocate than posterior ones. In particular, the nasal valve is the hardest site to evaluate with objective measures like AAR 2.
Surgeons should always perform both objective and subjective nasal measurements before and after septoplasty, although their predictive value is still controversial. Furthermore, as previously demonstrated, CT does not correlate with septal deviation findings, and is not recommended as a postoperative diagnostic tool.
Figures and tables
|Population (N = 494)|
|Sex||284 F; 210 M|
|Smoking habit||142 (28%)|
|Septal deviation: anterior deviations posterior deviations||312 (63%) 182 (37%)|
|Obstruction metrics||R 2||F (ANOVA)||Independent variables||P|
|ΔNOSE||0.298||103.9||D and R *||0.001|
|ΔTNR||0.441||139.4||D and R *||0.001|
|ΔWNR||0.601||369.7||D and R *||0.001|
|ΔNNR||0.355||89.7||D, R * and A||0.001|