Figure 1. Visual analog scale.
Figure 2. Stepwise management of chronic rhinosinusitis with nasal polyps.
Figure 3. Evaluation of intervention effect on newly diagnosed patients.
Figure 4. Evaluation of intervention impact on patients with various degree of control.
Rizk HG, Ferguson BJ. Categorizing Nasal Polyps by Severity and Controller Therapy. Arch Otolaryngol Head Neck Surg. 2012;138(9):846–853. doi:10.1001/archoto.2012.1736
Author Affiliations: Departments of Otolaryngology–Head and Neck Surgery, Hôtel-Dieu de France Hospital, Saint Joseph University, Beirut, Lebanon (Dr Rizk), and University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania (Dr Ferguson).
Objective To analyze a new categorization of chronic rhinosinusitis with nasal polyposis (CRSwNP) according to severity and controller therapy, modified from the 2007 stepwise treatment for asthma.
Design Prospective categorization of 50 new or established patients over a 3-month period.
Setting Tertiary center of Hôtel-Dieu de France Hospital, Beirut, Lebanon.
Patients Fifty consecutive adult patients with nasal polyposis referred by primary care physicians for sinonasal complaints or by pulmonologists for worsening lower airway control including asthma.
Interventions All patients were categorized according to CRSwNP severity scale and treated with stepwise therapy based on the study schematic.
Main Outcome Measures The Modified Sinonasal Questionnaire and the visual analog scale were used to assess patients' response to the treatment regimen and to modify the severity scale and the step treatment accordingly.
Results Ten new medication-naïve patients with CRSwNP were categorized by severity, which directed the subsequent treatment plan. All patients showed improvement in severity with the stepwise treatment plan. Thirty-seven of 40 patients with established CRSwNP seen in follow-up were easily categorized by degree of control, and subsequent treatment directed by therapy guidelines resulted in improvement.
Conclusions Initial categorization of medication-naïve patients with CRSwNP and patients with established CRSwNP facilitated delivery of successful directed stepwise therapy that resulted in improvement of classification of severity in most patients. This proposed classification may provide a useful template for future studies comparing patients with CRSwNP.
Chronic rhinosinusitis with nasal polyposis (CRSwNP) has a wide variability of presentation and impact on quality of life in affected patients. Several symptom surveys, computed tomographic (CT) grading, endoscopic scores, and quality of life (QOL) surveys are used in evaluating CRS. Within the subcategory of CRSwNP, no standardized clinical categorization relative to severity or cause is currently widely used1,2. This is a handicap in comparing objective presence and size of nasal polyps both in patients undergoing no medical therapy and in patients undergoing medical therapy. In the extreme example, oral prednisone will usually cause nasal polyps to significantly regress, but the nasal polyps will return when the systemic steroid therapy is stopped. To categorize the degree of nasal polyps between patients and over time, it is important to know what medication and how much is required to control the patients' nasal polyps.
A similar problem presents itself with asthma. A patient with asthma being treated with steroids may have no symptoms of asthma compared with a patient undergoing biweekly bronchodilator therapy, but the excellent control of the patient's asthma with steroid treatment comes with a significant cost in terms of adverse effects of long-term steroid use. Thus, to compare the degree of asthma between 2 patients or to assess asthma in a single patient over time, it is important to know the patients' pulmonary function, what medications are being used, what doses are required to control the asthma. This categorization of asthma relative to objective severity (including symptom scores) combined with medication requirements has been performed in asthma according to the stepwise approach based on the principles published in 2007 by National Asthma Education and Prevention Program, Expert Panel Report 3 (EPR-3)3:
Assessment of severity to initiate therapy and to control symptoms, with adjustment made in stepwise fashion
Patient education and environmental control
Management of comorbidities
Selection of medication
The stepwise approach to treatment of asthma was created to facilitate classification of patients and provide them with the best treatment for their condition. Indeed, it was based on assessment of impairment (ie, symptoms and spirometry results) as well as risk of subsequent asthma attacks. It is only with EPR-3 guidelines that this approach became evidence based.3
The present report describes a pilot study meant to analyze the validity of a newly devised categorization of CRSwNP according to severity and controller therapy and a modification of the stepwise approach treatment for asthma published in 2007 by EPR-3. We hope it will serve as the foundation for future work that allows categorization of CRSwNP by severity score as well as controlling therapy. The controlling therapy is administered in a stepwise fashion based on evidence of efficacy similar to the step therapy for asthma in EPR-3. Medications with weak evidence of efficacy for CRSwNP were placed in the last step, and the clinician is advised to use his or her best clinical judgment. Finally, we found cost constraints to alter some of the stepwise recommendations, especially for medications in the final step such as omalizumab or antifungal agents that were cost prohibitive.
To create a stepwise approach for the treatment of CRSwNP, we defined criteria for severity and adequacy of control, summarized respectively in Table 1 and Table 2. We used the Sinonasal Questionnaire (SNQ) (Table 3), validated by Dixon et al,4 for screening of CRS in patients with lower airway disease, to which we added an assessment of olfaction to categorize patients. We also used the visual analog scale (VAS) for this purpose (Figure 1), already validated in the European Position Paper on Rhinosinusitis and Nasal Polyps 2007 (EPOS).2 When there was discordance between the 2 scores, we used the score placing the patient in the more severe or lesser degree of control category. The Modified SNQ was scored from 0 to 3, as an average of 6 symptoms. Scores of 0 to less than 0.5 were considered asymptomatic; 0.5 to less than 1.5, mild; 1.5 to less than 2.5, moderate; 2.5 to 3, severe. For any discordance between Modified SNQ score, VAS, and any other items such as the need for surgery, number of purulent exacerbations, number of exacerbations requiring oral corticosteroids, and degree of control of comorbidities, the finding placing the patient in the most severe category was used. We standardized the treatment plan, based on the categorization of degree of control and severity, into a 5-step stepwise management plan (Figure 2). Medication choices were based on evidence presented for CRSwNP in EPOS.2
To analyze the applicability of this management scheme, a prospective study was conducted in the tertiary center of Hôtel-Dieu de France Hospital over the course of 3 months, with a minimum follow-up of 3 months for the patients recruited at the end of the study period. Fifty patients older than 12 years were included who were referred either by a primary care physician for sinonasal complaints or by a pulmonologist for difficulty controlling the lower airway comorbidity, usually asthma. These patients were diagnosed as having CRSwNP using clinical history, physical examination, CT scan of the sinuses, and endoscopic examination of the nasal cavities.
All statistical analysis used a χ2 Yates test with a statistical significance set at P < .05.
Ten medication-naïve patients with CRSwNP were assessed for severity and for treatment initiation based on the criteria summarized in Table 1. The sex ratio was 1:1. The mean age was 36.6 years (age range, 12-65 years). Seven patients had a comorbidity: 5 asthma, 1 poorly controlled asthma, and 1 Kartagener syndrome. Of the 10 new patients, 2 were classified as having mild disease and were treated with a step 2 treatment regimen, with improvement noted on follow-up. Five were classified as having moderate disease; 3 were treated with a step 3 treatment regimen, with improvement in SNQ and VAS scores noted on follow-up, while the other 2 patients underwent surgery because of the presence of extensive disease found on CT scan (Lund-Mackay score, 24). Indeed, the extent of disease precluded efficient administration of topical corticosteroids but was not severe enough to warrant treatment with oral corticosteroids. Postoperatively, one of these patient was categorized as having mild disease and maintained treatment with the step 2 treatment regimen, and the other was categorized as having moderately severe disease and maintained treatment with the step 3 regimen.
Two new patients were classified as having severe disease. The condition in 1 of these patients failed to improve significantly despite treatment with a short course of oral corticosteroids (VAS >8 after step 5 treatment). The other patient refused oral corticosteroids. Computed tomographic imaging showed extensive sinonasal disease (Lund-Mackay score >20), and both patients underwent endoscopic sinus surgery. Postoperatively, both were downstaged to a step 3 therapy and were doing well at last follow-up (2 months postoperatively).
Finally, a 23-year-old patient with Kartagener syndrome presented with acute exacerbation of CRSwNP with purulence. After a course of culture-directed antibiotics, his condition improved, and he was classified as having mild CRSwNP and treated with step 2 treatment with successful control of symptoms. Results of this group of patients are summarized in Table 4.
Forty patients with CRSwNP already undergoing medical therapy were seen in follow-up and categorized usingthe criteria outlined in Table 2.Three patients were excluded because they declined to participate in the evaluation or because they were unreliable. For the final analysis of control, 37 patients were evaluated, of whom 22 were male patients (59%). Mean age was 43 years (age range, 12-74 years). Twenty-three patients had lung comorbidities (1 cystic fibrosis, 2 aspirin-exacerbated respiratory disease, 1 chronic obstructive pulmonary disease, and 19 asthma). Ten had concomitant poor asthma control and poor control of sinonasal symptoms.
Fourteen patients had well-controlled sinonasal symptoms and were maintained on the same therapy. Thirteen patients without lower airway control problems had poorly controlled sinonasal symptoms, and therapy was adjusted according to step therapy protocol.
Under step therapy, all patients with poorly controlled conditions showed marked improvement at follow-up. Finally, 10 patients had symptoms that were very poorly controlled. These patients improved following a short course of oral corticosteroid therapy or after surgery, and their conditions were downstaged after reevaluation. Results from this group of patients are summarized in Table 5.
Of treated patients seen in follow-up, women were less likely to have well-controlled sinonasal symptoms than men: 10 of 14 well-controlled patients were men compared with only 4 women in the well-controlled group. Overall, 10 of 22 men were well controlled and 4 of 15 women were well controlled. These results were statistically significant (P <.05). Of patients presenting with CRSwNP without prior treatment, there was no sex difference in the spectrum of disease severity.
Two of 10 very poorly controlled patients and 2 of 13 not-well-controlled, step-4 patients refused treatment with oral corticosteroids and preferred surgery. In these cases, no oral steroids were given. One not-well-controlled, step-2 patient had extensive disease noted on CT scan precluding efficient administration of intranasal corticosteroids and this patient underwent surgery.
A comparison of the evaluation of the newly diagnosed patients before and after intervention showed a downstaging of all patients in the moderate and severe category. Also, the evaluation of patients seen at follow-up after intervention showed an increase in the number of well-controlled patients with no very poorly controlled patients (Figure 3 and Figure 4). This change in categorization after the intervention was also statistically significant (P <.05).
The medical management of nasal polyps is directed toward minimizing inflammation, whereas surgical control focuses on removing the mechanical obstruction within the sinus and nasal cavities.5 The physiopathologic characteristics and origins of CRSwNP are still a subject of much debate. Some authors consider it as a subgroup of CRS,6 whereas others consider CRS with and without NP as 2 separate entities.7,8
The diagnosis of objective CRS cannot be reliably made based on symptoms alone. In 3 studies with sample sizes ranging from 46 to over 700 patients with presumed sinusitis based on symptoms, only 35% to 54% had CT confirmation of CRS.9- 11 If one restricts analysis to patients who have endoscopic evidence of nasal polyps, then the number of patients with objective evidence of sinus disease by CT becomes substantially higher.12 We required both endoscopic and sinus CT validation of CRSwNP.
Symptom scores and radiologic and endoscopic scores such as the Lund-Mackay staging systems have been widely studied, with only a poor to absent correlation with objective disease severity and symptoms. In a large multicenter prospective study conducted on 1840 patients undergoing endoscopic sinus surgery, Hopkins et al13 found that 21% of these patients had Lund-Mackay scores close to those in asymptomatic patients. Moreover, other studies have shown a lack of correlation between preoperative Lund-Mackay staging system and postoperative outcome. Although it is clear that symptom scores are important in treatment decision, stratification of patients using validated outcome questionnaires is often dismissed on the basis that these assessment tools are subjective.
Although there are a number of validated and commonly used symptom scores in the literature, we chose the SNQ, which is a 5-item questionnaire, because of its excellent sensitivity and specificity for identifying chronic sinonasal disease.4
Based on the recommendations of the EPOS 2007,2 we also used the VAS for a number of CRS symptoms. However, both of these surveys fail to address olfaction, which has been shown to be disproportionately reduced in CRSwNP.12 And so Subjective Evaluation of Olfaction integrated into the SNQ was thus used to help categorize the disease severity and degree of control in our recruited patients. The need for surgery and oral corticosteroids in these patients as well as the number of acute exacerbations requiring antibiotics in the past year and the presence and degree of control of comorbidities—notably asthma—were also used in our new classification to categorize patients.
We built the stepwise management rationale based on the suggestions for treatment in EPOS 2007.2 Few of the suggested pharmaceutical and nasal wash interventions are approved by the US Food and Drug Administration for use in CRSwNP. Saline irrigations correspond to the lowest level of management and, along with mucolytics, are considered step 1 of treatment and are reserved for the least severe disease. Use of saline irrigations improved symptom scores in several randomized controlled trials in patients with CRS.2 These outcome measures were reached in isotonic as well as hypertonic saline irrigation.2 Theories about the mechanism of action include improving mucociliary flow, flushing away irritative substances, allergens and nitric oxide, and hydrating nasal mucosa.2,14,15 No trials, however, have been specifically conducted for evaluation in the setting of CRSwNP.2 Mucolytics such as guaifenesin are particularly useful in patients with altered quality of mucus, such as patients with human immunodeficiency virus or cystic fibrosis.2 In patients with purulent nasal polyposis, culture-directed antibiotics are considered the first step of treatment, and recategorization of the patient will be necessary after clearing of the acute exacerbation.
Intranasal corticosteroids have a level Ia of evidence both for improving symptoms and reducing nasal polyps and are considered as a step-2 treatment when used once daily. Step-3 treatment increases intranasal corticosteroids use to twice daily and/or adds a leukotriene receptor antagonist such as montelukast or zafirlukast to the treatment regimen. Recent investigations are showing promising results in interventions with leukotriene modulators, especially in patients with associated asthma and lower respiratory problems,4,14- 16 although there is only level III evidence for their clinical utility.2 However, their role could not be clearly evaluated in our series because most of the patients requiring a step 3 of treatment were already taking them for associated asthma.
Also included in this step are antihistamine sprays. In step 4, we suggest the use of 5-lipoxygenase inhibitors such as zileuton. Although we did not use it in our population because of expense, we still think it can be a viable option for refractory disease during step 4. Furthermore, in step 4 of treatment, we can add oral antihistamines. Although they cannot reduce polyp size significantly, they can provide symptomatic relief, as supported by level Ib of evidence in patients with allergic diathesis.2,5 Finally, step 5 of the treatment regimen corresponds to the addition of a short course of oral corticosteroids. Polyps usually respond well to intranasal corticosteroids even if they are not eosinophilic, such as in cystic fibrosis. Because of significant morbidity in prolonged use of systemic corticosteroids, we place them in step 5.
The role of antifungal therapy is poorly understood in nasal polyps owing to a lack of controlled studies. Fungi are variably cultured from sinuses of healthy people, and positive culture findings are not necessarily correlated with disease. However, at this step of treatment, we should consider possible fungal causes in refractory symptoms, and a trial of antifungal therapy can be conducted with amphotericin B irrigation or systemic antifungal agents appropriate to cultures.2,5,14 We did not assess antifungal therapy in patients in the current pilot study because it was prohibitively expensive for our study population. Finally, omalizumab, an anti-IgE monoclonal antibody, has been shown to be effective in moderate to severe allergic asthma and allergic rhinitis.5 Only 1 of our patients was given omalizumab, patient 11 (Table 5), who had severe lower and upper airway disease. The patient opted for surgery after 8 weeks (4 injections) because he could no longer afford the treatment, although he noted mild benefit with omalizumab.
Some of the medications suggested in steps 4 and 5, such as zileuton and antifungal therapy, were prohibitively expensive for us to assess. When given a choice, the patient would consistently opt for surgery rather than these medical options. We still left them as options based on the EPOS findings.2
We assessed responses to each newly added therapy after 1 to 2 weeks.5 The minimum length of follow-up was 3 months. Although the aim was to assess the efficacy of each new intervention, longer follow-up is needed in such a chronic disease to assess adaptive tolerance to medication and to look for sustained efficiency of each intervention. Indications for sinus surgery following medical therapy are poorly defined.5,17 In the pilot categorization table, we considered sinus surgery in poorly or not-well-controlled disease under step 4 or step 5 of treatment. Four patients unwilling to try oral corticosteroids (2 very poorly controlled step-4 patients and 2 not-well-controlled step-4 patients) and 3 patients with extensive disease found on CT scan that would preclude the use of topical corticosteroids (2 new patients with moderate disease, 1 not-well-controlled step-2 patient) underwent surgery. This illustrates the necessity to be fluid with recommendations. On the other hand, most patients with CRSwNP in this pilot study were easily categorized and either maintained on current therapy or improved on step therapy.
The timing of surgery in symptomatic patients with extensive disease on CT scan requires further study or deference to patient preference and response. The role of surgery in patients with CRSwNP differs from that CRS without NP. Without NP, limited surgery addressing an obstruction can restore function to the sinuses. However, in CRSwNP, the goal of surgery is not only to remove the obstructions but to marsupialize the nasal cavity so that the anatomic obstructions that might preclude application of topical washes and topical steroids are minimized. This allows more effective application of topical anti-inflammatory medications and possibly antimicrobial agents in the future to retard the return of the hyperplastic mucosal disease.2,5,18
The stepwise approach to treatment is a guideline that still needs further testing. However, our pilot study showed that it was easily implemented in 80% of newly diagnosed patients and in 92% of established patients already undergoing treatment.
In conclusion, this prospective study demonstrates the implementation and response to change of categorization for severity by stepwise treatment for CRSwNP. This standardized schema resulted in improved QOL for more than 90% of patients treated. Longer follow-up is needed in future studies to assess sustained efficiency of each intervention. For most patients the classification allows a standardized stepwise approach to CRSwNP that usually improves severity and QOL. Nevertheless, exceptions and patient preferences should still be taken into consideration, and the use of a standardized classification and treatment algorithm is never a substitute for sound clinical judgment. This schema allows for patients being treated with high-dose medical therapy required for control of CRSwNP to be categorized based on medication required for control in addition to presence of symptoms and NP. Further studies are needed to assess the validity of this new categorization approach in the clinical setting.
Correspondence: Habib G. Rizk, MD, Department of Otolaryngology–Head and Neck Surgery, Hôtel-Dieu de France Hospital, PO Box 16-6830, Beirut, Lebanon (firstname.lastname@example.org).
Submitted for Publication: March 11, 2012; final revision received July 8, 2012; accepted July 12, 2012.
Author Contributions: Dr Ferguson had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Rizk and Ferguson. Acquisition of data: Rizk. Analysis and interpretation of data: Rizk and Ferguson. Drafting of the manuscript: Rizk and Ferguson. Critical revision of the manuscript for important intellectual content: Ferguson. Statistical analysis: Rizk. Study supervision: Ferguson.
Financial Disclosure: Dr Ferguson is on the Advisory Boards of Teva (makers of a nasal steroid spray), Meda (makers of a nasal antihistamine spray), and Sunovion (makers of a nasal steroid spray).
Previous Presentation: This article was presented as a poster at the American Academy of Otolaryngology–Head and Neck Surgery Annual Meeting; September 26 to 29, 2010; Boston, Massachusetts.
Additional Contributions: We thank Amine Haddad, MD, FRCS, Bassam Tabchy, MD, FRCS, Simon Rassi, MD, Walid Abou Hamad, MD, Georges Dabar, MD, and Zeina Aoun Bacha, MD, for their help and guidance in constructing the clinical database.