Chronic Rhinosinusitis (CRS) is an inflammatory disorder of the nose and paranasal sinuses lasting for at least 12 weeks and affects 10.9% of the European population [1, 2]. While 1-4% of the general population report having Nasal Polyps (NP), the occurrence of CRSwNP has been shown to be more prevalent in patients reporting a positive family history of NP and in patients with respiratory phenotypes such as asthma and Allergic Rhinitis (AR) [3-6]. The prevalence of AR is 10-30% and asthma 1-18% in the general population their risk is strongly associated to a family history of asthma or allergy [7-9]. Thus, early awareness and treatment of CRSwNP might prevent recalcitrant disease forms and asthma [10, 11]. The aim of this prospective controlled study was to evaluate association between endoscopic, radiologic, self-reported CRSwNP, and family history in detecting CRSwNP. The hypothesis was that radiologic, self-reported and heredity of NPs would all have high predictability of CRSwNP phenotype.

The subject characteristics used in this study are presented in Table 1. The proportion of subjects reporting AR was significantly higher in the control group than in the CRS group (p = 0.03). The proportion of subjects reporting intranasal corticosteroid use, and having undergone previous sinonasal operations were significantly higher in the CRS group compared to the control group (p<0.001, p = 0.001, respectively, Table 1). The median age of the CRS group was significantly higher compared to the control group (p<0.001, Table 1). When comparing to the CRS subgroups, the proportion of patient reporting asthma or AERD was significantly higher in the CRSwNP group compared to the CRSsNP group (p<0.001, p = 0.048, respectively, Table 1). The median age of the CRSwNP group was significantly higher compared to the CRSsNP group (p < 0.001, Table 1). The median total LM score of sinus CT scans of the CRSwNP group was significantly higher compared to the CRSsNP group (p<.001, Table 1). Higher proportion of CRS patients reported obstruction, postnasal drip and/or facial pain/pressure as one of their worst symptom compared to controls (p<0.003, Table 1).

Fig. (1). Evaluation of radiologic signs of nasal polyposis from sinus Computed Tomography (CT) scans. Sinus CT scans that were routinely performed due to clinical purposes. The radiologic NP score was formed based on collected data of the CT scans to evaluation form that was filled by an experienced Radiologist (AM) blinded to patient´s history (12). Sinus mucosal findings were scored as 0=no change, 1=mucosal thickening, 2= polypous mucosal thickening ±discharge. Detectability of turbinate structures was scored as 0=detected; 1= not detectable. The reasons for “not detectable” responses were poor visualization of middle turbinate due to polypoid change or operative modification of turbinate, both considered to be signs of NP. The radiologic NP score 8-16 was suggestive for clinical CRSwNP phenotype. (A) No CT signs of polypous sinus mucosa. Turbinate anatomy is detectable. Radiologic NP score =0. (B) Polypous maxillary sinus mucosa on both sides. No mucosal swelling of anterior ethmoidal cells. Turbinate anatomy is detectable. Radiologic score =4. (C) Polypous maxillary sinus mucosa. Mucosal swelling of anterior ethmoidal cells. Turbinate anatomy is detectable. Radiologic score =6. (D) No polypous sinus mucosa. Right inferior turbinate fine structure is undetectable. Radiologic score =1. (E) Mucosal swelling of right maxillary sinus as well as anterior ethmoidal cells (not shown) and posterior ethmoidal cells. Polypous mucosa of left maxillary sinus. The anatomical fine structure of Inferior and middle turbinates is undetectable. Radiologic score 11. (F) Polypous mucosa of maxillary sinuses and anterior ethmoidal cells (not shown) and posterior ethmoidal cells. The anatomical fine structure of middle turbinates is undetectable. The fine structure of inferior turbinates is detectable. Radiologic score 14.

Total radiologic NP score gave an AUROC of 0.95 (95% CI 0.91-1.00), p<0.001 (Fig. 2). With the threshold value total radiologic NP score ≥ 8/16, the sensitivity for detecting CRSwNP from CRSwNP group was 94% while specificity was 82% (Fig. 2). When using the cut-off value of eight in Radiologic NP score, 6% of CRSwNP patients did not have radiologic signs of NPs while 18% of CRSsNP patients had radiologic signs of NPs (p<0.001, Table 1). Total LM score gave an AUROC of 0.84 (95% CI 0.75-0.94), p<0.001 (Fig. 2). With the threshold LM score ≥ 9/24 the sensitivity and specificity for detecting CRSwNP from CRSwNP group was 78% and 69% respectively.

A total of 33% of CRSwNP patients reported not having NPs and 18% of CRSsNP patients reported having NPs (p<.001, Table 1). While 27% of patients with radiologic signs of NPs reported not having NPs, 22% of patients with radiologic signs of NPs reported having NPs (p<0.001, Table 1).

Binary logistic regression was conducted to evaluate the positive and negative family history of NPs and its association with the control and CRS groups. When evaluating the family history, 14% of controls and 21% of CRS group had positive family history of NP (p=.45, Table 3). The corresponding percentages of positive family history of asthma were also not significantly different between controls (40%) and CRS (42%) (p=0.084). Interestingly, we found a significant association with positive family history of AR, 56% in controls to 32% in the group CRS (p=0.015, Table 3).

Finally, we wanted to evaluate the proportion of subjects not knowing their family history of NP, asthma or AR. Control group and CRS group knew similarly their family history of asthma (p=0.077, Table 4), and AR (p=0.082, Table 4). In contrast, the knowledge of family history of NPs in the CRS group was significantly poorer compared to the control group (adjusted OR=6.02, 95% CI 1.98-18.30, p=0.002, Table 4).

This study was carried out to study the relationship between endoscopic-, radiologic-, self-reported CRSwNP, and family history in detecting CRSwNP. We hypothesised that radiologic, self-reported and heredity of NPs could be able to predict CRSwNP phenotype. Studies support that early awareness and early treatment of CRSwNP might prevent progressive and severe disease forms as well as development of asthma [10, 11]. Thus studying predictive potential of radiologic and reported NP signs could be useful in situations where nasal endoscopy is not available, such as in primary care or during visits at another specialist other than an Otorhinolaryngologist.

Fig. (2). The Receiver operating characteristic (ROC) discrimination curve of prediction models for CRSsNP group (n=55) over CRSwNP group (n=18). Total radiologic NP score gave an AUROC of 0.95 (95% CI 0.91-1.00), p<0.001. With the threshold value total radiologic NP score ≥ 8/16, the sensitivity was 94% and specificity 82% for detecting CRSwNP from CRSsNP group. Total LM score gave an AUROC of 0.84 (95% CI 0.75-0.94), p<0.001. With the threshold value LM score ≥ 9/24 the sensitivity was 78% and specificity 69% for detecting CRSwNP from CRSsNP group.

We were able to develop a radiologic NP score with good specificity and sensitivity to detect CRSwNP from CRSsNP. The score was based on findings in nasal turbinates and maxillary, anterior/posterior ethmoid cells showing that polypoid sinus mucosa and/or non-detectable middle/inferior turbinate(s) were most indicative for CRSwNP phenotype. Total LM showed a moderate specificity and sensitivity to detect CRSwNP from CRSsNP. Larger well-characterized prospective cohort studies are needed to further evaluate predictive algorithms based on CT scan findings.

In addition to radiologic NP score, self-reported NPs associated with CRSwNP phenotype. This might indicate that in situations of lacking the possibility to perform nasal endoscopy, early recognition of CRSwNP might benefit form information obtained from both sinus CT scans and patients, in addition to clinical examination and proper patient interview. In this study, patients were asked the presence of NPs when referred to a CRS-surgical consultation at the University Hospital, which might have increased self-awareness of NPs. Hence true positive and false negative rates of self-reported NPs might be lower in general population which might lead to poorer predictability of self-reported NPs for CRSwNP phenotype. A study of school students showed that of 131 diagnosed asthma cases, 118 (90%) were aware about their diagnosis indicating the need for health education of airway diseases [15].

Strikingly, contrary to expectation, family history of NPs did not associate with CRSwNP or CRS. This could be associated with the observation that CRS patients were uncertain concerning family history when reporting NPs compared to controls. Additional evidence for this line of thinking comes from other studies showing poor accuracy in reporting family history of diseases such as colorectal cancer or prostate cancer [16-18]. It could thus be speculated that, younger age of controls, as in our study, could in part explain them to be more astute when seeking information and hence be more aware of family history and health related issues. On the other hand, our models were adjusted by age, gender, presence of AR and asthma.

Family history of allergy and/or asthma was given with higher certainty both in CRS and control groups compared to family history of NP, which was given with uncertainty. Poor knowledge of NPs heredity could indicate an unmet need to improve patient education in order to detect patients at risk to develop recalcitrant CRSwNP early and to improve their respiratory health. Our study also highlights an important need to improve questionnaires to better detect positive family history of CRSwNP. The question “Do you have a family history of NP” will putatively need additional questions such as “Has a family member complained symptoms of CRS such as loss of nasal obstruction, discharge, facial pain or loss of sense of smell?” or “Has a family member undergone a paranasal sinus operation and/or polypectomy?”. Future studies are therefore needed to validate whether additional questions would improve possibilities to gather knowledge of inherited CRSwNP.

This study has several limitations. The sample size was small and not population based. CRS population presented a random sample of CRS patients visiting CRS-surgical consultation and thus the limited number of CRSwNP patients. There might have been limitations in the variables collected from patient records. We acknowledge that lack of SNOT22, endoscopic NP score, NP eosinophilia, and socioeconomic status limits the interpretation of the findings.

The study demonstrated that radiologic NP score and self-reported NPs associated with endoscopic NPs. This might be clinically useful in situations where nasal endoscopy is unavailable, in order to hit early progressive CRSwNP. Family history of CRSwNP had less predictive value. Validation studies with larger population are still needed.

This study was funded in part by the grants of the Finnish cultural foundation, the Hospital Districts of Tampere and of Helsinki and Uusimaathe Tampere Tuberculosis association, the Finnish Association of Otorhinolaryngology and Head and Neck Surgery, Finnish Medical Foundation, the Finnish Society of Allergology and Clinical Immunology, Jane and Aatos Erkko Foundation, Paulo Foundation, Yrjö Jahnsson Foundation and Väinö and Laina Kivi Foundation. The authors report no conflicts of interest.