Translation, cultural adaptation, and validation of the NOSE-Perf scale to Brazilian Portuguese

Highlights • NOSE-Perf scale quantifies symptoms resulting from nasal septal perforation.• The Portuguese version of the NOSE-perf scale showed internal consistency.• The NOSE-Perf scale translation showed good reliability parameters.• The proposed instrument is valid for measuring nasal septal perforation symptoms.


Introduction
Nasal septal perforation (NSP) is a pathological condition that establishes communication between the two nasal cavities due to a septal defect.It emerges from a range of pathologies, including traumatic, iatrogenic, drug use, and various inflammatory, rheumatic, infectious, and neoplastic diseases. 1,2ts clinical repercussions are very diverse, ranging from asymptomatic patients to symptoms that significantly disrupt the Quality of Life (QoL), such as nasal obstruction, whistling, crusting, epistaxis, pain, dryness, and nasal deformity. 3---6Thus, its treatment is something challenging for rhinologists and can range from conservative measures, such as nasal irrigation and emollients, to the use of septal buttons or surgical interventions, with a variety of techniques described in the literature. 2,7,8 fundamental part of the management of any condition is to understand the impact of each symptom on the patient's QoL.QoL assessment instruments are important not only to better understand the disease but also to standardize the way of reporting treatment outcomes.To this end, the NOSE-Perf scale was recently described as a validated disease-specific symptom score for NSP. 9 However, it is currently only available in English, and there are no similar instruments validated for the Brazilian Portuguese-speaking population.
The aim of this study was to perform the Translation, Cultural Adaptation, and Validation of the NOSE-Perf Scale to Brazilian Portuguese.

Methods
This study was developed in two stages.The first consisted of the translation and cultural adaptation of the original questionnaire into Brazilian Portuguese and the second of its prospective validation.

Ethical considerations
This study was authorized by the Ethics Committee of the University, under registration number 59360722.0.0000.5404.Informed consent was obtained from all participants.

Translation and cultural adaptation
Translation and cultural adaptation followed the recommendations of the ISPOR Task Force 10 and was divided into 7 steps, as follows.

Participants
We included 16 consecutive patients from the rhinology outpatient clinic with the diagnosis of septal perforation were, and 16 participants without septal perforation and sinonasal comorbidities for the control group.For the group with septal perforations, the inclusion criteria were patients older than 18 years, literate, with a diagnosis of septal perforation, without other sinonasal comorbidities, and who did not undergo any intervention between the first and second answering of the questionnaire.Exclusion criteria: active nasosinusal disease, previous skull base surgery, history of sinonasal malignancy, previous head and neck radiation therapy, craniofacial syndrome, current use of intranasal drugs, cleft palate, or previous cleft palate repair.

Application of the final version of the Questionnaire
Participants were instructed to answer the final version of the NOSE-Perf in Portuguese by themselves.A physician was always present, but without intervening.In addition, to assess clarity, the patients were asked to classify the question as ''easy to understand'', ''difficult to understand'', and ''did not understand'' for each question.
The control group answered the questionnaire only once to perform validation.The perforation group answered the questionnaire again after one month to assess the stability of the responses using test-retest reliability.

Statistical analyses
The sample size was calculated according to Bonett's sample size requirements 11 using the Arifin sample size calculator (https://wnarifin.github.io/sscweb.html) and based on the estimate of the questionnaire being filled out by the same individual on two different occasions, hoping to find an ICC of 0.9 a 0.1 precision, and an alpha value of 0.05.This resulted in a sample size of 15 patients, taking into account that, although unexpected, we could have a dropout, the final sample size was calculated to be 16, similar to Camina et al. 12 The same number of patients was chosen to form the control group.
IBM ® SPSS ® software version 29.0 was used for analysis.Qualitative variables are presented in absolute and relative frequencies.The quantitative variables' means, standard deviations, and minimum and maximum amplitudes were determined.Sociodemographic and clinical characteristics were analyzed using the Chi-Squared, Fisher Exact, and independent sample t-tests.To assess internal consistency, Cronbach's alpha coefficient was calculated, with values between 0.7 and 0.95 considered acceptable.Test---retest reliability was analyzed using Spearman's correlation coefficient.The Intraclass Correlation Coefficient (ICC) with absolute-agreement, 2-way mixed-effects model was determined by correlating the measurements obtained in the first and the second visit.
For discriminant validity, the total scores of the quiz of the two groups were compared.Data were analyzed using a nonparametric Mann---Whitney U test.A p-value of < 0.05 was considered significant.

Patient characteristics
Table 1 summarizes the patient' characteristics.There was no statistical difference between the groups regarding age, gender, and education.
In our sample, there were seven different etiologies for septal perforations.In six patients, the NSP was a late sequela of granulomatosis with polyangiitis, all with stable disease and no other sinonasal repercussions, 5 idiopathic cases, 2 patients with iatrogenic perforation after septoplasty, 1 case with late sequelae of Leishmaniasis, without active disease or other sinonasal repercussions, 1 case after facial trauma and 1 patient with sequelae of acute invasive fungal rhinosinusitis, with the cured disease.

Questionnaire Translation and cultural adaptation
The final version of the NOSE-Perf questionnaire translated into Brazilian Portuguese is shown in Fig. 1.
Fig. 2 compares the original NOSE-Perf sentences in English with their respective translations into Portuguese.

Clarity
Fig. 3 illustrates how the 32 participants rated the clarity of the questions.For questions 1, 2, 4, and 11, only one patient rated them as ''difficult to understand''.For the other questions, all 32 patients rated them as ''easy to understand''.None of the participants rated a question as ''I did not understand''.

Discriminant validity and questionnaire scores
Table 2 shows the total score of the scale for the two groups evaluated, with their mean and minimum amplitudes.The NSP group obtained a mean total score of 13.8 ± 12.6.The control group obtained a mean total score of 2.3 ± 1.8.There was a statistical difference between the groups, with p < 0.001, demonstrating good discriminant validity.
Table 3 details the mean scores for each of the questions and the two responses from the NSP group.For these patients, crusting was the main problem, with a score of 1.87 and 1.69 (first and second visit, respec-    tively), while whistling had the lowest score (0.81 and 0.69).The other symptoms assessed had similar scores.
As for the control group, the most reported symptom was facial pain, with a score of 0.69, while whistling and epistaxis were not scored by any participant.Table 4 demonstrates the means scores, SD value, and comparison between the NSP group and control group.It also describes the item-total correlation for each question, which showed a good overall correlation, with all questions above 0.5.

Internal consistency and test-retest reliability
Cronbach's alpha score for the Nose-Perf score was 0.996, indicating high internal consistency.
Spearman's correlation coefficient was 0.994 (p < 0.001), indicating the high reliability of repeated measures.The intraclass correlation coefficient was 0.992 with a 95% Confidence Interval (95% CI) of 0.978---0.997,indicating high reliability.Fig. 4 represents the Bland-Altman plot of the difference between the questionnaire scores in the first and second visits, indicating that their mean values are within two standard deviations (1.96 × SD: Standard Deviation), demonstrating agreement between both visits' answers.NSP can lead to significant impairment in patients' quality of life.Using the SNOT-22, Leong and Webb 13 assessed the impact of NSP symptoms on QoL, comparing scores with those of patients with CRS.Both scores were high, indicating great impairment in QoL and, despite being higher in the CRS group, there was no significant difference between them.Subsequently, Khong and Leong 14 reported that SNOT-22 scores of patients with NSP were comparable to other diseases with a major impact on QoL, such as recalcitrant chronic rhinosinusitis, empty nose syndrome, and neurogenic facial pain syndromes.In both studies, the authors highlighted the limitation of using a non-specific instrument to assess NSP symptoms and the importance of creating one that addresses specific symptoms of the disease, such as nasal crusting, epistaxis, and whistling noise.
In this context, the NOSE-Perf scale was developed, a disease-specific instrument to assess and quantify symptoms resulting from septal perforation. 9After its validation and publication, the same group used this instrument to evaluate NSP baseline symptoms in 202 patients. 6The most prevalent symptom was crusting, followed by congestion, breathing difficulty, obstruction, rhinorrhea, epistaxis, and whistling.In our study, despite crusting was also the most common, the prevalence of other symptoms was variable, differing from those reported by Taylor et al. 6 Our NSP patients were also less symptomatic, with a mean total NOSE-Perf score of 13.8 compared to 23.7 in the aforementioned study. 6Although NSP is a condition that potentially affects QoL, our sample reported a minor impact, with individual average for each of the 12 items not reaching 2 out of a possible 4 points (Table 3).Compared to previous studies that reported symptoms related to NSP, there is also a significant variability in symptom prevalence, 15---21 thus it is expected their respective impacts on QoL would also vary.
NSP clinical manifestations are related to the size and location of the NSP. 3,14,22Anterior NSPs are the most frequent and usually more symptomatic, while posterior or superior NSPs are often asymptomatic and more commonly caused by systemic diseases. 3,22,23Therefore, these differences between published studies can be explained in part by different samples regarding the etiology and location of NSP, in addition to ethnically and regionally different populations.
Evaluating the etiologies of NSP, we found a high prevalence of sequelae of rheumatological diseases and idiopathic cases, and a relatively small percentage of cases secondary to surgical procedure, a frequent cause of NSP. 2,4This can be explained by the study being carried out in a tertiary university reference center, with a low volume of septoplasties and a high prevalence of rare diseases and complex cases.However, discussion regarding etiology is limited by the small number of patients participating in this study.
This study highlights the clinical heterogeneity of the disease and the importance of having an instrument that explicitly evaluates each symptom so that it is possible to understand each case better, making the treatment more accurate and personalized.
In practical terms, the questionnaire proved easy and quick to complete and may also give patients greater insight into the severity of their symptoms and the outcomes of medical interventions.
The Brazilian Portuguese version of the NOSE-Perf scale showed excellent clarity and understanding by the participants, good reliability, and proved to be valid for use in the Brazilian population.This instrument may contribute to standardizing the documentation of NSP symptoms among Brazilian physicians and provide a deeper understanding of this disease's impact on the Brazilian population's QoL.It may also be used to report treatment outcomes and enable comparison with other international studies using the NOSE-Perf scale.
This study had the limitation of producing an instrument to be self-applied, limiting its use in the illiterate population.Another challenge of any cultural adaptation of an instrument for the Brazilian population is the fact that it is a large country with culturally heterogeneous regions.

Conclusion
The Brazilian version of the NOSE-Perf scale is a reliable and valid instrument for measuring symptoms in patients with nasal septal perforations.

Figure 1
Figure 1 Final version of NOSE-Perf scale for Brazilian Portuguese.

Figure 2
Figure 2 Comparison between the original NOSE-Perf scale and the version translated into Portuguese.The original sentences are in italics and their respective translations are in bold.

Figure 3
Figure 3 Clarity of the Nose-Perf questions translated to Portuguese.Participants were asked if the question was easy to understand, hard to understand, or if they did not understand.

Figure 4
Figure 4 Bland and Altman graphic representation of reproducibility.DIF, The difference between the total scores from the first and second visits.MEAN, Mean values of the questionnaire scores.The darker lines represent Standard deviation values.Portuguese, following the Principles of Good Practice recommended by the ISPOR taskforce, 10 ensuring equivalence between the original version and the one adapted to the Brazilian culture.
a t-test.b Chi-Squared test.c Fisher exact test.

Table 2
NOSE-Perf-scale total score for both groups, showing mean, minimum and maximum score and standard deviation.

Table 3
Comparison of mean scores for each question for the nasal septal perforation (NSP) Group in their first and second visits.

Table 4
Comparison between each question for the nasal septal perforation group (NSP) and Control Group (CG).Item-total correlation for each question for the NSP group.
a Pearson correlation.b Mann---Whitney U test.