2.1

Patients

The present investigation was approved in 2009 by our Section’s in-house ethical committee. The study was carried out in accordance with the principles of the Helsinki Declaration.

Adult patients with evidence of polyposis involving at least two recesses without extending to the nasal cavity (multiple polyps occupying the middle meatus, grade 2), or polyps extending beyond the middle meatus (grade 3) , were first treated medically for 3 months with local mometasone furoate 200 μg daily (100 μg/nostril), the dosage recommended by Stjarne et al. , or with fluticasone furoate 110 μg daily (55 μg/nostril), and oral therapy with methylprednisolone (32 mg daily on days 1–5, then 16 mg daily on days 6–10, then 8 mg daily on days 11–20), as suggested by Van Zele et al. . If this medical therapy failed, patients underwent FESS (ethmoidectomy, middle antrostomy, sphenoidectomy and/or frontal sinusotomy, depending on the location of the polyps), with or without septoplasty/turbinoplasty, under general anesthesia. Patients with polyps confined to only one recess (grade 1) were not treated surgically.

Between January 2011 and December 2014, 210 patients underwent FESS for CRSwNP, performed by the same surgeon (G.B.). Exclusion criteria were a postoperative follow-up shorter than 6 months and a diagnosis of eosinophilic granulomatosis with polyangitis (EGPA) (formerly Churg-Strauss syndrome), which left 179 consecutive patients eligible, who were enrolled in the present study.

For all patients, the preoperative diagnostic work-up included nasal endoscopy (with rigid 0° and 30° [Ø 4 mm] optical instruments), nasal cytology, serum eosinophil and basophil counts, assays of total and specific serum IgE for common airborne allergens, serum eosinophil cationic protein (ECP) levels, and paranasal high-resolution computerized tomography (CT). Asthmatic patients underwent spirometry with methacholine challenge.

After surgery, patients performed nasal irrigations with isotonic saline solution twice a day (using 20 ml per irrigation) and local nasal therapy with mometasone furoate 200 μg daily (100 μg/nostril), using the dosage recommended by Stjarne et al. , or fluticasone furoate 110 μg daily (55 μg/nostril). All laboratory tests were performed before surgery, at least 3 months after withdrawing oral steroids, and at least 1 month after stopping nasal steroid treatment.

During periods of pollination, patients with a known allergy to pollen were treated with antihistamines. Patients with asthma received appropriate therapy.

2.2

Clinical history and features

We identified patients with a history of aspirin intolerance. The American Rheumatism Association criteria, revised in 2012 by the International Chapel Hill Consensus Conference Nomenclature of Vasculitides , were considered for the diagnosis of EGPA. Cases with a clinical history of diabetes or endoscopic evidence of pharyngo-laryngeal inflammation due to gastroesophageal reflux disease (GERD) were also recorded.

2.3

Laboratory features: serum eosinophil and basophil counts, serum total and specific IgE assay, serum ECP

All patients had 3 blood samples taken 12 months and 6 months before surgery, and on the day of FESS to obtain their eosinophil and basophil counts. Absolute values and percentages were recorded, and the means and standard deviations are reported. We also measured serum total IgE and specific IgE for dermatophagoides pteronyssinus and farinae, birch pollen, pellitory, grass mix, cat and dog dander, alternaria alternata, aspergillus fumigatus, and common ragweed.

The normal reference range for serum ECP is between 2.3 and 16 μg/L (according to the Laborationslista, Uppsala University Hospital, Art. Nr. 40284 Sj74a, April 22, 2008 ). For the purposes of the present study, a serum ECP concentration > 16 μg/L was considered significant.

All assays were performed at the same laboratory (the EIA Unit, Laboratory Medicine Service, Padova General Hospital).

2.4

Histopathological features: eosinophil component in polyps

Three high-power fields (400 × magnification) from each surgical specimen were examined to quantify the eosinophil component, and two patterns of polyposis were recognized: (i) polyposis rich in eosinophils (≥ 10 eosinophils per field); and (ii) polyposis poor in eosinophils (< 10 eosinophils per field).

2.5

Follow-up

Patients underwent postoperative nasal endoscopy under topical anesthesia (lidocaine 20 mg/ml plus xylometazoline 1 mg/ml; 1:1 ratio). Endoscopic follow-ups were performed using rigid 0° and 30° (Ø 4 mm) instruments and scheduled 3, 6, 12, 24, and 36 months after surgery. We considered patients with recurrences as those with endoscopic evidence of at least grade I polyposis.

2.6

Statistical analysis

Fisher’s exact test was used to investigate the association between sinonasal polyp recurrences and clinical, laboratory and pathological variables in a univariate statistical setting. A p-value < 0.05 was considered significant, while values in the range of 0.10 > p ≥ 0.05 were assumed to indicate a statistical trend.

A multivariate logistic model was developed, adding the clinical, laboratory and pathological parameters for which Fisher’s exact test disclosed a p value < 0.20. The results were expressed as odds ratios (ORs); p values and 95% confidence intervals (CIs) were also calculated. During the analysis, the model was checked for multicollinearity with a variance inflation factor test.

In accordance with Steyerberg et al. , the quality of the model was assessed by calibration. Applying the Hosmer and Lemeshow test, a non-significant result (p > 0.05) meant that the model fitted the data well. The effectiveness of the discrimination was assessed on the scale proposed by Hosmer and Lemeshow , where: an area under the receiver operating characteristic (ROC) curve (AUC [ROC]) 0.5 = no discrimination; AUC (ROC) between 0.7 and 0.8 = acceptable discrimination; AUC (ROC) between 0.8 and 0.9 = excellent discrimination; AUC (ROC) beyond 0.90 = outstanding discrimination. Additional statistics derived from the model (sensitivity, specificity, positive predictive accuracy, negative predictive accuracy, and accuracy) were also calculated.

The STATA™ statistical package 8.1 (Stata Corp, College Station, TX, USA) was used for all analyses.

2.1

Patients

The present investigation was approved in 2009 by our Section’s in-house ethical committee. The study was carried out in accordance with the principles of the Helsinki Declaration.

Adult patients with evidence of polyposis involving at least two recesses without extending to the nasal cavity (multiple polyps occupying the middle meatus, grade 2), or polyps extending beyond the middle meatus (grade 3) , were first treated medically for 3 months with local mometasone furoate 200 μg daily (100 μg/nostril), the dosage recommended by Stjarne et al. , or with fluticasone furoate 110 μg daily (55 μg/nostril), and oral therapy with methylprednisolone (32 mg daily on days 1–5, then 16 mg daily on days 6–10, then 8 mg daily on days 11–20), as suggested by Van Zele et al. . If this medical therapy failed, patients underwent FESS (ethmoidectomy, middle antrostomy, sphenoidectomy and/or frontal sinusotomy, depending on the location of the polyps), with or without septoplasty/turbinoplasty, under general anesthesia. Patients with polyps confined to only one recess (grade 1) were not treated surgically.

Between January 2011 and December 2014, 210 patients underwent FESS for CRSwNP, performed by the same surgeon (G.B.). Exclusion criteria were a postoperative follow-up shorter than 6 months and a diagnosis of eosinophilic granulomatosis with polyangitis (EGPA) (formerly Churg-Strauss syndrome), which left 179 consecutive patients eligible, who were enrolled in the present study.

For all patients, the preoperative diagnostic work-up included nasal endoscopy (with rigid 0° and 30° [Ø 4 mm] optical instruments), nasal cytology, serum eosinophil and basophil counts, assays of total and specific serum IgE for common airborne allergens, serum eosinophil cationic protein (ECP) levels, and paranasal high-resolution computerized tomography (CT). Asthmatic patients underwent spirometry with methacholine challenge.

After surgery, patients performed nasal irrigations with isotonic saline solution twice a day (using 20 ml per irrigation) and local nasal therapy with mometasone furoate 200 μg daily (100 μg/nostril), using the dosage recommended by Stjarne et al. , or fluticasone furoate 110 μg daily (55 μg/nostril). All laboratory tests were performed before surgery, at least 3 months after withdrawing oral steroids, and at least 1 month after stopping nasal steroid treatment.

During periods of pollination, patients with a known allergy to pollen were treated with antihistamines. Patients with asthma received appropriate therapy.

2.2

Clinical history and features

We identified patients with a history of aspirin intolerance. The American Rheumatism Association criteria, revised in 2012 by the International Chapel Hill Consensus Conference Nomenclature of Vasculitides , were considered for the diagnosis of EGPA. Cases with a clinical history of diabetes or endoscopic evidence of pharyngo-laryngeal inflammation due to gastroesophageal reflux disease (GERD) were also recorded.

2.3

Laboratory features: serum eosinophil and basophil counts, serum total and specific IgE assay, serum ECP

All patients had 3 blood samples taken 12 months and 6 months before surgery, and on the day of FESS to obtain their eosinophil and basophil counts. Absolute values and percentages were recorded, and the means and standard deviations are reported. We also measured serum total IgE and specific IgE for dermatophagoides pteronyssinus and farinae, birch pollen, pellitory, grass mix, cat and dog dander, alternaria alternata, aspergillus fumigatus, and common ragweed.

The normal reference range for serum ECP is between 2.3 and 16 μg/L (according to the Laborationslista, Uppsala University Hospital, Art. Nr. 40284 Sj74a, April 22, 2008 ). For the purposes of the present study, a serum ECP concentration > 16 μg/L was considered significant.

All assays were performed at the same laboratory (the EIA Unit, Laboratory Medicine Service, Padova General Hospital).

2.4

Histopathological features: eosinophil component in polyps

Three high-power fields (400 × magnification) from each surgical specimen were examined to quantify the eosinophil component, and two patterns of polyposis were recognized: (i) polyposis rich in eosinophils (≥ 10 eosinophils per field); and (ii) polyposis poor in eosinophils (< 10 eosinophils per field).

2.5

Follow-up

Patients underwent postoperative nasal endoscopy under topical anesthesia (lidocaine 20 mg/ml plus xylometazoline 1 mg/ml; 1:1 ratio). Endoscopic follow-ups were performed using rigid 0° and 30° (Ø 4 mm) instruments and scheduled 3, 6, 12, 24, and 36 months after surgery. We considered patients with recurrences as those with endoscopic evidence of at least grade I polyposis.

2.6

Statistical analysis

Fisher’s exact test was used to investigate the association between sinonasal polyp recurrences and clinical, laboratory and pathological variables in a univariate statistical setting. A p-value < 0.05 was considered significant, while values in the range of 0.10 > p ≥ 0.05 were assumed to indicate a statistical trend.

A multivariate logistic model was developed, adding the clinical, laboratory and pathological parameters for which Fisher’s exact test disclosed a p value < 0.20. The results were expressed as odds ratios (ORs); p values and 95% confidence intervals (CIs) were also calculated. During the analysis, the model was checked for multicollinearity with a variance inflation factor test.

In accordance with Steyerberg et al. , the quality of the model was assessed by calibration. Applying the Hosmer and Lemeshow test, a non-significant result (p > 0.05) meant that the model fitted the data well. The effectiveness of the discrimination was assessed on the scale proposed by Hosmer and Lemeshow , where: an area under the receiver operating characteristic (ROC) curve (AUC [ROC]) 0.5 = no discrimination; AUC (ROC) between 0.7 and 0.8 = acceptable discrimination; AUC (ROC) between 0.8 and 0.9 = excellent discrimination; AUC (ROC) beyond 0.90 = outstanding discrimination. Additional statistics derived from the model (sensitivity, specificity, positive predictive accuracy, negative predictive accuracy, and accuracy) were also calculated.

The STATA™ statistical package 8.1 (Stata Corp, College Station, TX, USA) was used for all analyses.