The Optimal Head Position for Distributing Topical Nasal Medication Using the Mucosal Atomization Device

1. PURPOSE The purpose of this study is to determine if the Lying Head Back or Head Down and Forward head position for administering topical atomized nasal medication results in greater distribution and efficacy of budesonide administered via the mucosal atomization device in chronic rhinosinusitis patients who have had previous sinus surgery.
2. HYPOTHESIS The investigators hypothesize that endoscopic sinonasal mucosal scores will be lower (improved) in patients oriented in the LHB than the HDF position while applying medication after 12 weeks of treatment.
3. JUSTIFICATION: Topical medications administered by nebulizers are frequently utilized for patients suffering from refractory CRS. Nebulizers are useful as they atomize medication to increase contact with surrounding mucosa. This technique is most useful for widespread inflammation of the olfactory cleft, ethmoid, sphenoid and frontal sinuses where increased inflammation may obstruct drainage pathways. Obstructed sinus ostia may prevent mucosal clearance, impair ciliary function and subsequently contribute to infection. Topical medication, such as steroid therapy is a safe and effective method to locally reduce inflammation. The success of atomized nasal medications is dependent on multiple factors such as gravity, obstructive features of sinonasal anatomy and patient interpretation of the preferred delivery method. Several delivery methods for topical nasal therapies have previously been evaluated to determine the optimal head position for administration. However, a review of the literature suggests that the optimal method of administering atomized nasal steroids to the paranasal sinuses in human subjects has yet to be determined. It has been previously suggested that successful distribution of intranasal steroid sprays requires topical agents to contact and remain on desired mucosa to prevent disease reoccurrence. In a prospective single-cohort study, healthy subjects were recruited to evaluate the distribution and clearance rate of fluorescein-labeled nasal spray distributed using a squeeze bottle, nasal gel or nasal spray device. The investigators found there was no significant difference in clearance rate between the delivery methods. This trial provides the preliminary evidence to support further investigation on the rate of distribution from intranasal steroid administration in CRS patients. The findings cannot be generalized to a wider CRS population, as patients enrolled were healthy with no history of CRS or sinus surgery. Similarly, the investigators did not report the head position utilized to administer the treatment. Therefore, there remains no formal consensus for the optimal head position for CRS patients suffering from recurrent symptoms for the distribution and clearance rate of intranasal steroid medication. Clinician experience at the St. Paul's Sinus Centre (SPSC) has recommended that patients should orient themselves in the Lying Head Back (LHB) position to effectively treat edematous mucosa of the ethmoid roof and frontal sinus recess. These instructions are based on evidence collected in a previous cadaveric trial completed at our centre (in press). Our team evaluated the effect of head position on the distribution of fluorescent nasal spray within the paranasal sinuses of cadaver specimens. The investigators utilized two commonly instructed head positions, the Lying Head Back (LHB) or Mygind position and the Head Down and Forward (HDF) or Moffat's position. In this trial the investigators oriented cadaver specimens in either study position and evaluated the presence of fluorescent nasal spray across eleven clinically relevant anatomical areas. the investigators concluded that the LHB position had significantly greater frequency of distribution (OR = 1.85, 95% CI: 3.0, 6.9, p< 0.001) to all evaluated areas than the HDF position. This trial was limited by a number of factors, which include a small sample size (n=20). Similarly, cadaveric specimens cannot be generalized to humans, as moist and ciliated mucosa, and nasal polyps are not represented. Therefore, it remains controversial the optimal head position for the deposition of intranasal medication.
4. OBJECTIVES: Primary Objective The primary objective of this clinical trial is to compare endoscopic sinonasal mucosal inflammation between patients oriented in the LHB and HDF position after 12 weeks of treatment. Subject sinus cavities will be assessed by a validated endoscopic mucosal staging system first at baseline, then at 6 and 12 weeks of treatment. Total scores will be compared between the treatment groups. The Lund-Kennedy (LKES) and Philpott-Javer (PJES) endoscopic staging systems will be utilized to quantify the severity of sinonasal disease. The Principal Investigator will complete all endoscopic evaluations and will be blind to treatment allocation. Secondary Objectives: The secondary objectives of this trial will be to compare LKES and PJES scores to compare sensitivity to change during the treatment period and degree of correlation between total scores and change in scores between the scoring systems. Inter-rater and intra-rater reliability will be calculated and compared for each scoring system. The purpose of this secondary outcome is to validate the PJES with the "gold standard" and widely used LKES, in terms of overall score and sensitivity to treatment.
5. RESEARCH METHOD: Study Design: Parallel, single-blind, randomized controlled trial of subjects previously randomized to LHB or HDF treatment arms (from Part 1) in order compare endoscopic sinonasal mucosal scores after 12 weeks of treatment. Patients who meet the inclusion and exclusion criteria will be recruited for the study at the clinic. After study patients voluntarily consent to participate in the study, data collection will begin. Randomization Treatment Arm #1: Administering topical fluorescent-labeled nasal spray in the LHB position. Treatment Arm #2: Administering topical fluorescent-labeled nasal spray in the HDF position. Adult patients suffering from refractory CRS symptoms post-FESS will be randomized to either Treatment Arm #1 or #2 in equal ratio. The sequence will be computer generated by the trial statistician. The sequence will be ordered random-permuted blocks to ensure allocation remains concealed. The sequence will be maintained in a computer located in the St. Paul's Sinus Centre (SPSC) with strict passwords necessary for access. Non-transparent sealed envelopes with treatment assignments will be kept in a secure environment and accessible only to study personnel at the SPSC. When study personnel are notified of a potential candidate, inclusion and exclusion criteria will be evaluated. When consent has been provided the next consecutive envelope will be taken and treatment group assigned. After treatment is assigned, the patient will preserve the envelope to ensure the Principal Investigator remains blinded to their treatment allocation. Blinding: Part 1: As patients will be required to orient themselves in the appropriate study positions, blinding will not be feasible for subjects. However, the Principal Investigator and two additional senior Rhinologists evaluating endoscopic images for mucosal disease severity will be blinded to each subject's allocated head position. Patients will be randomized to a head position group, either the Lying Head Back (LHB) or Head Down and Forward (HDF) position. Patients will remain in the allocated group and administer their medication in their respective head position for 12 weeks. As above, patients will not be blind to which position they have been randomized. However, the Principal Investigator and two additional senior Rhinologists will be blind to which position the subject had been oriented throughout the treatment period (12 weeks). Images of each sinus cavity will be captured at baseline and after 12 weeks and evaluated using the LKES and PJES scoring systems. Total scores and change in scores will be compared. The randomization sequence will not be broken until the final analysis is complete. Conduct of Study: Baseline and Intervention: Adult CRS patients having previously received FESS, who meet the inclusion/exclusion criteria and provide consent will be randomized to either the LHB or HDF positions to administer a budesonide topical nasal corticosteroid spray to their sinus cavities. Baseline characteristics will be collected from patient charts as outlined above. Each subject in both study groups will receive a sterile 3ml syringe fixed with a Mucosal Atomization Device (MAD) tip (Wolf-Tory Medical, Salt Lake City, UT). This device is advantageous as its produces a fine mist, increasing the surface area for absorption within the paranasal sinuses.4 Budesonide is a well-tolerated, glucocorticoid steroid that is primarily used for treatment in asthma and nasal polyposis.20 At SPSC, it is standard practice to prescribe budesonide to treat postoperative sinonasal inflammation utilizing the MAD. From discussion with senior Rhinologists at SPSC, patients are typically instructed to administer treatment twice a day for 6-12 weeks. However, dosage and frequency may be modified to suit individual circumstances. For the purpose of this trial, patients will be instructed to load each syringe with 2cc of the budesonide (1mg in 2cc saline) solution to be distributed to the right and left sides of the nose. A designated research assistant will collect baseline data and describe and demonstrate the allocated study position in a separate room from the Principal Investigator. To ensure that positioning has been correctly communicated, subjects will then orient themselves in the appropriate position with guidance from research staff if required. Subjects randomized to the LHB position will be instructed to lay supine on the clinical table, with their head hanging over the edge of the bed as far as possible without discomfort (Figure 1). Subjects randomized to the HDF position will be instructed to kneel down, placing the top of their head on the ground and forehead close to the knees with the nostrils facing upwards (Figure 2). Once positioned, subjects will insert the MAD-syringe at a 45-degree angle into the nasal valve, direct the tip to the lateral epicanthus of the ipsilateral orbit and depress the syringe plunger completely. Subjects will remain positioned for 60 seconds. At this time a designated research associate will be measure the angle of the subjects head in relation to the horizontal plane, using an analog protractor. As all subjects may not be able to orient themselves in a single standardized angle, the investigators will note the angle most comfortable to determine a range of possible orientations. Subjects will be instructed to administer budesonide 1mg/2cc nebules daily for 6 weeks using the MAD syringe, in the head position allocated to which they were randomized at the baseline visit. After this time, subjects will return to SPSC for endoscopic examination of their sinus cavities. Bilateral nasal endoscopy will be performed using a 2.3mm Karl Storz rigid nasal endoscope. The Principal Investigator capturing images of the sinus cavities will be blind to which head position group the subject was randomized. Each sinus cavity will be assessed independently with images captured and stored on a password-protected, centrally located computer. Once endoscopic images are captured for patients returning after 12 weeks of treatment, three independent senior Rhinologists will score each sinus cavity according to the LKES and PJES criteria.18,19 This will be completed twice, with a 1-week break between assessments. Responses will be tabulated on separate, password-protected spreadsheets with each Rhinologist receiving a study code to prevent identification. A blind statistician will summarize the data and compare the total and change in endoscopic scores between the treatment groups. Secondary analysis will be completed to determine the extent of correlation and sensitivity between the LKES and PJES responses. Inter and intra-rater reliability will also be computed to determine the extent of agreement between the scoring systems. Data collection: The following data will be collected at the initial clinic visit:

• Baseline characteristics will be collected including age (years), sex (male or female), number of cigarettes per day (count), height (centimeters), weight (kilograms), handedness (right or left), number of prior sinus surgeries (count), preoperative disease severity (Lund-Mackay CT Score). The following data will be collected at the 6 and 12 week follow-up visits:
• PJ Endoscopic scores
• LK Endoscopic scores
• SNOT-22 scores

1. STATISTICAL ANALYSIS:

Sample size determination:

From our recent trial evaluating the LHB and HDF positions in human cadavers, the incidence of total spray distribution was 76% and 41% in the LHB and HDF positions. Incorporating these findings, the investigatorsperformed a sample size calculation using a Type I error (alpha) of 5%, Type II error (beta) of 20% and an effect size of 73%. The effect size was calculated using the incidence proportions from our centre's previous trial in cadaver specimens, which was identified as clinically relevant and practically achievable. A two-proportion calculation was used to compare the average distribution of all anatomical sites between the head position groups. The calculation revealed a necessary sample size of 30 patients per group. Based on clinical experience of the Principal Investigator it is estimated that 60% of patients requiring sinus surgery (n=400/year) will be eligible for enrollment (n=280/year). Accounting for 70% enrollment of eligible subjects (n=196/year), the investigators estimate that 16 patients can be recruited every month. Adjusting for a dropout rate of 20%, the investigators will increase our total sample size (n=60) by 6 patients. the investigators estimated that our desired sample size adjusting for loss-to-follow-up (n = 66) will be achieved in 4 months (Figure 3). Patients will be recruited in equal ratio to the experimental and control groups (33 in each arm).

Baseline characteristics analysis:

Baseline characteristics of the two groups will be reported. Descriptive statistics will be reported included mean, median, standard deviation and range. These will be reported for age, height, weight, number of prior surgeries and preoperative disease severity. The proportion of males and females, right and left handed subjects and frequency of blood clots and adhesions present prior to the intervention will also be described.

End of Study Analysis:

Subject sinus cavities will be assessed using the LKES and PJES scales. Measurements will be made at baseline and 5 weeks after treatment. The LKES assesses each sinus (ethmoid, maxillary, sphenoid, frontal) for the presence of polyps (0=absent, 1=present in middle meatus only, 2=beyond middle meatus); presence of edema, scarring or crusting (each evaluated as 0=absent, 1=mild, 2=severe) and presence of discharge (0=absent, 1=clear and thin, 2=thick or purulent). A total, bilateral score is calculated for all sinuses evaluated. The PJES is a similar scoring system that evaluates each sinus cavity individually to yield a total bilateral score out of a total of 80. Additional points are awarded for the presence allergic mucin or purulent discharge. Two endoscopic scores will be recorded for each subject, using both scales for baseline and 5 weeks post treatment. The difference in baseline and 5 week results will be calculated for each subject and summed together to yield an average change for the LHB and HDF groups. Outcome data will be summarized by mean, median, standard deviation and interquartile range. The Students t-test will be used to test if the changes in endoscopic scores are significantly different between the treatment groups. Pearson's (R) correlation will be calculated to determine the extent of linear association between the LKES and PJES results. From a previous study evaluating the association between endoscopic scoring systems, an a priori decision has been made to categorize R < 0.49 as a weak correlation, 0.50 < R < 0.69 as a moderate correlation and R > 0.70 as a strong correlation. Linear correlation will be calculated in terms of absolute score and change in score between the study time points.

To determine the extent of clinical agreement between the senior Rhinologists evaluating each subject, intra and inter-rater kappa scores will be calculated. It has been determined a priori that scores < 0 will be assigned no agreement, 0.01 - 0.20 poor agreement, 0.21-0.40 slight agreement, 0.41-0.60 fair agreement, 0.61-0.80 good agreement, 0.81-0.92 very good agreement and 0.93-1.00 excellent agreement.