Quality of Life and Olfaction Changes Observed with Short-Term Medical Management of Chronic Rhinosinusitis

Abstract

BACKGROUND:

Patients with chronic rhinosinusitis (CRS) commonly experience both reduced quality-of-life (QOL) and olfactory dysfunction (OD). Literature on the impacts of appropriate medical therapy (AMT) for CRS on QOL and OD is limited, and the focused design of these studies may limit their applicability to usual clinical practice.

METHODS:

Adults with symptomatic CRS were prospectively enrolled (November 2016–October 2018) to an observational, multi-institutional study. Individualized AMT was initiated per standard practice according to evidence-based guidelines. Endoscopy examination (Lund-Kennedy), olfactory function (Sniffin’ Sticks) testing, and QOL survey responses (22-item SinoNasal Outcome Test [SNOT-22]; Questionnaire of Olfactory Disorders-negative statements [QOD-NS]) were obtained at enrollment and follow-up.

RESULTS:

Baseline measures demonstrated heterogeneity of QOL and OD. After an average of 7.8 weeks, within-subject median SNOT-22 total improved 39.5% (n=39, p<0.001) relative to baseline, including 50% (p=0.014) improvement for item #21 “sense of smell/taste.” QOD-NS improvement was also statistically significant (p=0.044). Sniffin’ Sticks score relative improvement of 10.9% (n=33, p=0.109) was not statistically significant and lacked correlation with SNOT-22 total scores (R=−0.247, p=0.165) or QOD-NS total scores (R=−0.016, p=0.930), but correlated moderately with endoscopy score (R=−0.436, p=0.018).

CONCLUSIONS:

Participants with varied impacts of CRS, treated with individualized short-term AMT, demonstrate significant improvements in CRS- and olfactory-specific QOL measures, without corresponding improvement in clinically measured olfactory function. Olfactory function changes moderately correlated with endoscopy score changes, but lacked association with QOL measurements.

INTRODUCTION

Patients with chronic rhinosinusitis (CRS) commonly experience reduced quality-of-life (QOL) as well as olfactory dysfunction (OD).^1–4 CRS-related reduction in QOL is also highly prevalent, though variable in magnitude and multifactorial in etiology. QOL reduction in CRS may result directly from diminished overall and psychological health, as well as indirectly from impaired function and productivity.^5,6 The prevalence of clinically measured (Sniffin’ Sticks olfactory testing) OD in patients with CRS is up to 80%,^3,7 and the symptomatic manifestations of OD as dysosmia are pervasive in CRS definitions, ‘cardinal symptoms,’ and outcome assessments.^1,4,8,9 Dysosmia may contribute to reduced QOL through negative impacts on functions of daily living (e.g., detection of odors and food enjoyment), and olfaction-specific QOL improves after endoscopic sinus surgery (ESS) for patients who have failed medical management of CRS.³

The impact of a short course of appropriate medical therapy (AMT; e.g., topical and systemic corticosteroids, oral antibiotics, and nasal saline irrigations) on QOL and olfaction-specific QOL in patients with CRS is poorly understood.^4,10,11 Previous studies on the QOL and OD impacts of AMT for CRS focused on patients that already failed AMT, but whom elected ongoing medical management instead of ESS.^12,13 Follow-up measures in these studies are often obtained after 6 months of continued medical therapy, in contrast to assessment of the patient’s response after an initial short trial of AMT.^12,13 The narrow focus of this existing data on medically-refractory patients treated with a prolonged course of AMT limits its applicability to this scenario which is uncommon in usual clinical practice. A more common scenario in the practices of this multi-institutional study is to initiate a short trial (1–2 months) of AMT, then evaluate the patient’s response to therapy prior to considering additional medical and surgical treatments. Thus there is currently a lack of information available to prognosticate QOL and olfaction-specific QOL changes for CRS patients in common “real world” clinical scenarios.

The relationship between changes in clinically measured olfaction and patient-reported QOL, specifically for patients managed with short-term AMT, is also unknown. Given the time and cost burden of performing Sniffin’ Sticks clinical olfaction testing, it is not practical to perform on all patients, which limits clinical use outside of research. Consistency in QOL and olfactory function measures is plausible based on underlying inflammatory mechanisms, but skepticism is supported by a well published history of incongruence between clinical testing and patient-based measures in CRS.^14–18 Specific investigation of the relationship between Sniffin’ Sticks testing and more readily available patient reported measures (i.e., proxy measures) is particularly valuable in context of the common clinical scenario of short-term AMT.

The current investigation aims to accomplish two main objectives in a relatable cohort of CRS patients and common clinical scenario. The first objective is to describe changes in clinically-measured olfactory function (Sniffin’ Sticks), and patient reported QOL outcomes (22-item SinoNasal Outcome Test [SNOT-22]; Questionnaire of Olfactory Disorders-negative statements [QOD-NS]) after a short course of AMT. The second objective is to evaluate the relationship between changes in these patient-reported outcomes (e.g., SNOT-22 and QOD-NS) and clinically measured outcomes (e.g., Sniffin’ Sticks and endoscopy scores).

MATERIALS and METHODS

Study Design and Sample Population

Enrollment was conducted as part of an investigator-monitored, observational research study of human subjects funded by the National Institute on Deafness and Other Communication Disorders (Bethesda, MD.; Federal grant #R01 DC005805). Study participants were prospectively recruited from a patient population presenting to academic, rhinology centers located in the United States at either Oregon Health and Science University (OHSU, Portland, OR.), the Medical University of South Carolina (Charleston, SC.), the University of Utah (Salt Lake City, UT.), the University of Colorado (Aurora, CO.), and the University of Virginia (Charlottesville, VA.). The Institutional Review Board affiliated with each enrollment location approved study protocols while providing continuing review for patient and data safety per good clinical practice guidelines. Patients provided written, informed consent after initial enrollment meetings to ensure voluntary participation involving minimal risk and no deviations from the standard of care (SOC) for the treatment of CRS.

Inclusion Criteria and Clinical Follow-Up

Symptomatic, adult study participants received a confirmed diagnosis of CRS, with or without nasal polyposis, from a fellowship trained rhinologist following criteria established by current Clinical Practice Guidelines of the American Academy of Otolaryngology-Head and Neck Surgery.¹ Additionally patients confirmed the ability to tolerate medical therapy regimens of systemic corticosteroids, broad-spectrum or culture directed antibiotics, and topical corticosteroid sprays/irrigations. Participants needed to complete all research study questionnaires in the English language. Patients were individually counseled for follow-up appointments after completion of adequate trials of appropriate medical therapy prescribed for symptom management or resolution including, but not limited to: nasal saline irrigations, topical corticosteroid sprays/irrigations, systemic corticosteroids, antibiotics, macrolide therapy, decongestants, antihistamines, or leukotriene modifiers. This investigation attempted to characterize subsequent differences in primary outcome measures associated with appropriate medical therapy.

Exclusion Criteria

Study participants were not considered for study participation if they had previous physician diagnoses of: Granulomatosis with polyangiitis (GPA), sarcoidosis, dementia or Alzheimer’s disease, Parkinson’s disease, immunodeficiency, and/or a history of major head trauma/traumatic brain injury resulting in olfactory loss. Additional exclusions were considered if study participants did not complete all study related questionnaires or provide complete medical history during baseline enrollment meetings.

Clinical Measure of Disease Severity

Study participants completed a comprehensive review of medical and social history during baseline enrollment meetings. Clinical measures of CRS disease severity were also collected for diagnostic purposes per the SOC and sourced as research data when available. Bilateral endoscopy examinations were conducted in clinic both prior to study enrollment and during follow-up appointments using rigid endoscopes with 0–30° visualization (Karl Storz, Tuttlingen, Germany). Pathology of the paranasal sinuses, including the presence of nasal polyposis, was quantified by each enrolling physician using Lund-Kennedy staging (score range: 0–20).¹⁹

Olfactory Dysfunction Measurement

For the primary outcome of interest for this investigation, comprehensive bilateral olfactory function was measured using Sniffin’ Sticks (Burghart Messtechnik, Wedel, Germany) which evaluate three separate domains of olfactory function: odorant threshold (score range: 1–16; higher score indicating greater olfactory function), discrimination (score range: 0–16), and identification (score range: 0–16). Odorant threshold (n-butanol target) was evaluated in a ‘staircase procedure’ using pen triplets in which odorant thresholds are detected on a continuum of dilution steps until the odorant can be accurately distinguished from 2 blank pens offered in random sequence. Odorant discrimination was conducted using a presented sequence of pen triplets in which 2 pens have the same odorant. Study participants are directed to identify the single pen with a different odorant from the presented sequence. Odorant identification was evaluated using 16 pens containing common odorants presented individually. Respondents are directed to select the correct odorant from 4 multiple choice options. Correctly identified threshold (T), discrimination (D), and identification (I) item scores, are summarized as a composite TDI total score (score range: 1–48). Olfactory function was evaluated both at baseline enrollment and post-treatment follow-up appointments.

Patient Reported Outcome Measures

Study participants were asked to provide complete responses to two patient-reported outcome measures (PROMs) during both baseline enrollment meetings and post-treatment follow-up appointments with minimal time burden. The 22-question SinoNasal Outcomes Test (SNOT-22) is a widely used, validated, survey instrument designed to quantify the severity of symptoms associated with sinonasal disorders using Likert scale (item score range: 0–5) response options (Washington University, St. Louis, MO.).⁸ Validated factor analysis of SNOT-22 item scores in a patient population with CRS identified five symptom domains which can be categorized and summarized into: rhinologic symptoms (score range: 0–30), extranasal rhinologic symptoms (score range: 0–15), ear and/or facial symptoms (score range: 0–25), psychological dysfunction symptoms (score range: 0–35), and sleep dysfunction symptoms (range: 0–25).²⁰ One specific survey question of the SNOT-22 (Q#21) queries respondents to rank their ‘Sense of smell/taste.’ Higher SNOT-22 total score (range: 0–110) and domain scores reflect and overall worse sinonasal symptom severity. A minimal clinically important difference (MCID) of at least 8.9 points in SNOT-22 total scores post-treatment has been historically described for patients with CRS using distribution-based methodology.^8,21,22

Additionally, participants also provided responses to the 17-item Questionnaire of Olfactory Disorders-negative statements (QOD-NS) survey. The QOD-NS is a validated, olfaction-specific questionnaire designed to capture patient perception and impact of olfactory dysfunction using Likert scale response options (item score range: 0–3) whereas higher summarized response scores (range: 0–51) represent greater impacts on normal, daily activity, patient preferences, and emotional status due to olfactory impairment.²³ An average MCID value for the QOD-NS has been described as at least 5.2 points in patients electing surgical intervention for CRS using distribution-based determinations.²⁴

Prescribed Medication History

During both baseline enrollment meetings and post-treatment follow-up clinical appointments, study participants were asked to recall how many days they had used particular sinonasal medications out of the preceding 90 days including: topical nasal corticosteroid sprays, topical nasal corticosteroid rinses/irrigations/drops, oral antibiotics for CRS, oral corticosteroids for CRS, saline irrigations/rinses, decongestants, antihistamines for CRS, and/or leukotriene modifiers.

Database Management and Statistical Methods

Investigational data was secured through the assignment of unique study identification numbers for study participants and removal of all protected health information prior to transfer into a centralized database in a closed environment at OHSU (Access; Microsoft Corporation; Redmond, WA.) in compliance with the Health Insurance Portability and Accountability Act of 1996. All descriptive and statistical comparisons were completed using SPSS software (version 24.0; IBM Corporation, Armonk, NY.). Statistical analyses were guided after an evaluation of all scaled measures for assumptions of normality and linearity using both graphical analysis and Shapiro-Wilk testing as appropriate. Means, median values, and interquartile range (IQR) were provided for nonparametric test comparisons including Wilcoxon signed rank testing for matched pairs over time. Two-tailed Spearman’s rank correlations were also completed to evaluate significant associations between all scaled variables of interest while statistical comparisons were determined significant using a conventional 0.050 type-I error probability (p-value).

RESULTS

Final Study Population

A total of 48 patients with CRS met eligibility criteria, provided informed consent for study participation, and complete baseline enrollment measures between November, 2016 and October, 2018. A total of 39 (81%) study participants provided post-treatment clinical follow-up and PROM surveys an average of 7.8 [±2.8] weeks after enrollment for evaluations of medical therapy effectiveness on symptom reduction or resolution. A total of 33 (69%) study participants completed both follow-up endoscopy exams and Sniffin’ Sticks testing to assess changes in olfactory function. Baseline descriptors of study participant demographics per NIH criteria, comorbid conditions, endoscopy scores, olfactory dysfunction, and PROM scores are provided in Table 1. Baseline Sniffin’ Sticks testing identified a total of 13 participants (27%) with anosmia (TDI 1–15), 19 participants (40%) with hyposmia (TDI 16–30), and 16 participants (33%) with normosmia (TDI 31–48), categorized by TDI score according to Sniffin’ Sticks testing instructions.²⁵ Prescribed medication history from the preceding 90 days is described in Table 1.

Table 1:

Descriptors of final study population at study enrollment (n=48).

Descriptor:	N (%)	Mean [±SD]	Median [IQR]	Range (LL, UL)
Age (years)	----	49.5 [±18.6]	55.5 [35.0]	(18, 88)
Male	25 (52%)	----	----	----
Female	23 (48%)	----	----	----
White / caucasian	40 (83%)	----	----	----
African American	5 (10%)	----	----	----
Asian	1 (2%)	----	----	----
Hispanic/Latino ethnicity	1 (2%)	----	----	----
Nasal polyposis	29 (60%)	----	----	----
Septal deviation	17 (35%)	----	----	----
Turbinate hypertrophy	8 (17%)	----	----	----
Previous endoscopic sinus surgery	17 (35%)	----	----	----
Asthma	16 (33%)	----	----	----
Migraine	10 (21%)	----	----	----
AERD / ASA intolerance	4 (8%)	----	----	----
GERD	9 (19%)	----	----	----
Diabetes mellitus (Type I/II)	4 (8%)	----	----	----
Depression (physician diagnosed)	13 (27%)	----	----	----
Anxiety (physician diagnosed)	11 (23%)	----	----	----
Oral corticosteroid dependency	1 (2%)	----	----	----
Smoking / tobacco use (current)	3 (6%)	----	----	----
Alcohol use (current)	26 (54%)	----	----	----
Allergic rhinitis	25 (52%)	----	----	----
Autoimmune disorders (unspecified)	0 (0%)	----	----	----
Lund-Kennedy endoscopy score	----	6.7 [±3.6]	5.0 [5.0]	(0, 13)
Sniffin’ Sticks / TDI total score	----	24.2 [±9.3]	26.0 [16.9]	(9, 45)
Threshold score	----	4.3 [±3.3]	3.6 [5.8]	(1, 16)
Discrimination score	----	10.0 [±3.2]	10.0 [6.0]	(4, 16)
Identification score	----	10.0 [±4.0]	11.0 [7.0]	(1, 15)
SNOT-22 total score	----	42.8 [±19.4]	46.0 [32.0]	(10, 93)
Rhinologic symptoms	----	16.0 [±7.1]	16.5 [12.0]	(2, 28)
Extranasal rhinologic symptoms	----	7.3 [±3.8]	7.5 [6.0]	(0, 14)
Ear / facial symptoms	----	7.4 [±5.1]	6.0 [7.0]	(1, 25)
Psychological dysfunction symptoms	----	10.3 [±7.2]	9.0 [12.0]	(0, 27)
Sleep dysfunction symptoms	----	10.6 [±6.4]	11.0 [11.0]	(0, 25)
Q21: “Sense of smell/taste”	----	2.8 [±1.9]	3.0 [4.0]	(0, 5)
QOD-NS total score	----	10.2 [±8.4]	8.0 [14.0]	(0, 32)
Prescribed Medication History (out of past 90 days):
Topical nasal corticosteroid sprays	29 (60%)	24.5 [±32.0]	6.0 [60.0]	(0, 90)
Topical nasal corticosteroid rinses/irrigations	17 (35%)	17.2 [±31.1]	0.0 [28.0]	(0, 90)
Oral antibiotics for CRS	21 (44%)	6.9 [±9.9]	0.0 [10.0]	(0, 40)
Oral corticosteroids for CRS	14 (29%)	5.6 [±11.2]	0.0 [9.0]	(0, 60)
Saline irrigations/rinses	34 (71%)	33.0 [±35.5]	20.5 [68.0]	(0, 60)
Decongestants for CRS	22 (46%)	13.9 [±26.2]	0.0 [18.0]	(0, 90)
Antihistamines for CRS	21 (44%)	19.6 [±31.7]	0.0 [38.0]	(0, 90)
Leukotriene modifiers	9 (19%)	11.1 [±27.8]	0.0 [0.0]	(0, 90)

N, sample size; SD, standard deviation; LL, lower limit; UL, upper limit; AERD, aspirin exacerbated respiratory disease; ASA, acetylsalicylic acid; GERD, gastroesophageal reflux disease; CT, computed tomography; TDI, odorant threshold, discrimination, and identification; SNOT-22, 22-item Sinonasal Outcomes Test; QOD-NS, Questionnaire of Olfactory Disorders – negative statements. CRS, chronic rhinosinusitis. Due to the observational study design, baseline CT scores were not available for all study participants; IQR, interquartile range.

Follow-up Differences in Olfactory Dysfunction and PROM Scores

Study participants were asked to repeat all study related measures during subsequent clinical follow-up appointments to evaluate the effectiveness of short-term, prescribed medical therapy on symptom reduction or resolution. Lund-Kennedy endoscopy scores were not found to significantly improve (p=0.237) in patients with follow-up from 6.9 [±3.5] to 6.2 [±3.3] on average. Similarly, olfactory cleft endoscopy scores did not significantly improve (p=0.834) on average from 4.4 [±3.1] to 4.5 [±3.8] in that group. Median post-treatment changes in olfactory function, endoscopy and PROM scores for study participants with follow-up are compared in Table 2.

Table 2:

Comparisons between baseline and post-treatment changes in olfactory function and PROM scores for study participants with follow-up.

Outcomes:	Baseline Median [IQR]	Follow-up Median [IQR]	Difference (Δ) Median [IQR]	Matched pairing test statistic	p-value
Lund-Kennedy endoscopy score (n=33)	7.0 [6.0]	6.0 [5.0]	0.0 [5.5]	WSR=−1.08	0.283
Sniffin’ Sticks / TDI total score (n=33)	24.8 [36.0]	27.5 [12.6]	1.5 [4.1]	WSR= 1.60	0.109
Threshold score	3.0 [5.5]	4.0 [5.5]	0.0 [1.8]	WSR= 0.85	0.396
Discrimination score	10.0 [6.0]	11.0 [5.8]	1.0 [5.5]	WSR= 0.77	0.444
Identification score	12.0 [8.0]	12.0 [5.0]	0.0 [2.5]	WSR= 1.09	0.277
SNOT-22 total score (n=39)	43.0 [30.0]	26.0 [22.0]	−12.0 [32.0]	WSR= −3.66	<0.001
Rhinologic symptoms	15.0 [12.0]	9.0 [6.0]	−4.0 [12.0]	WSR= −3.67	<0.001
Extranasal rhinologic symptoms	7.0 [7.0]	5.0 [6.0]	−2.0 [5.0]	WSR= −3.01	0.003
Ear / facial symptoms	6.0 [7.0]	4.0 [6.0]	−1.0 [6.0]	WSR= −3.15	0.002
Psychological dysfunction symptoms	9.0 [12.0]	6.5 [9.0]	−3.0 [8.0]	WSR= −2.96	0.003
Sleep dysfunction symptoms	9.0 [10.0]	5.0 [10.0]	−3.0 [9.0]	WSR= −3.29	0.001
Q21: “Sense of smell/taste”	3.0 [4.0]	1.5 [3.0]	0.0 [2.0]	WSR= −2.45	0.014
QOD-NS total score (n=39)	8.0 [13.0]	8.0 [10.0]	−2.0 [5.3]	WSR= −2.01	0.044

PROM, patient reported outcome measures; TDI, odorant threshold, discrimination, identification; SNOT-22, 22-item Sinonasal Outcome Test; QOD-NS, Questionnaire of Olfactory Disorders – negative statements. WSR, Wilcoxon Signed Rank test statistic for nonparametric data (standardized); IQR, interquartile range.

Significant reduction in average and median symptom severity scores were reported across all five symptom domains of the SNOT-22 survey, while 23/39 (59%) reported improvements in SNOT-22 total scores of one MCID value or greater. Additionally, responses to the anchor-based survey item of the SNOT-22 which evaluates a patients’ “Sense of smell/taste” significantly improved (p=0.014) on average from 2.6 [±2.0] to 1.7 [±1.6]. Study participants however were not found to experience significant differences in olfactory function using Sniffin’ Sticks / TDI testing (p≥0.109) or endoscopy findings (p=0.283). Statistically significant improvement was reported overall using QOD-NS survey responses following prescribed medical therapy, but with only 8/39 (21%) of patients reporting improvement at least one MCID value in magnitude.

Follow-up Differences in Prescribed Medication Therapy

Differences in prescribed medication use (days out of the previous 90) between baseline enrollment and follow-up were evaluated (Table 3). Study participants reported significantly greater median duration of saline irrigations and prescribed topical nasal steroid rinses/irrigations (e.g., budesonide) during the observational period. Similarly, the prevalence of patients reporting nasal daily steroid rinses/irrigations also significantly increased from 39% to 69% (p=0.008).

Table 3:

Comparisons between baseline and post-treatment changes in median prescribed medication use days (out of the previous 90) for study participants with follow-up (n=39)

Medication use variables:	Baseline Median [IQR]	Follow-up Median [IQR]	Difference (Δ) Median [IQR]	Matched pairing test statistic	p-value
Topical nasal corticosteroid sprays	6.5 [60.0]	5.0 [45.0]	0.0 [9.3]	WSR= −0.28	0.782
Topical nasal corticosteroid rinses/irrigations	0.0 [30.0]	40.0 [90.0]	0.0 [50.0]	WSR= 3.39	0.001
Oral antibiotics for CRS	5.0 [10.0]	0.0 [10.0]	0.0 [11.0]	WSR= −1.75	0.081
Oral corticosteroids for CRS	0.0 [10.0]	5.0 [12.0]	0.0 [10.0]	WSR= 0.67	0.503
Saline irrigations/rinses	25.0 [79.0]	50.0 [90.0]	5.0 [37.0]	WSR= 3.18	0.001
Decongestants for CRS	0.0 [10.0]	0.0 [5.0]	0.0 [5.0]	WSR= −1.82	0.069
Antihistamines for CRS	0.0 [45.0]	0.0 [10.0]	0.0 [7.0]	WSR= −1.88	0.061
Leukotriene modifiers	0.0 [0.0]	0.0 [0.0]	0.0 [0.0]	WSR= 1.66	0.098

IQR, interquartile range; N, sample size; CRS, chronic rhinosinusitis; WSR, Wilcoxon Signed Rank test statistic for nonparametric data (standardized).

Bivariate Associations between Measures of Post-treatment Change

Unadjusted, two-tailed, bivariate correlations were also evaluated between Sniffin’ Sticks / TDI scoring, Lund-Kennedy endoscopy scores, and all patient reported outcome measures for both baseline associations and for associations in measures of post-treatment change (Table 4). Significant bivariate correlation was found between Sniffin’ Sticks total scores and Lund-Kennedy endoscopy scores (p=0.002) indicating a weak association between better olfactory function and less overall sinonasal disease severity. Similar weak associations existed between higher baseline Sniffin’ Sticks total scores and significantly lower individual item scores (Q#21) of the SNOT-22 survey which operationalize less impairment in a patients sense of smell/taste (p=0.001). Following short-term medical therapy, bivariate association between change in endoscopy scores and Sniffin’ Sticks total scores was identified (p=0.018) but reflected a lack of significant improvement in both scores.

Table 4:

Two-tailed, bivariate Spearman’s rank correlation coefficients between Sniffin’ Sticks TDI score, endoscopy scores, and patient reported outcome measures. Correlation (R) and statistical significance (p-value) reported for both baseline values (n=48) and the change from baseline to follow-up (n=33).

Sniffin’ Sticks / TDI total score	Baseline Score		Change (Δ): Baseline to Follow-up
	R	p-value	R	p-value
Lund-Kennedy endoscopy score	−0.431	0.002	−0.436	0.018
SNOT-22 total score	0.124	0.402	−0.247	0.165
Rhinologic symptoms	−0.167	0.256	−0.163	0.364
Extranasal rhinologic symptoms	0.163	0.269	−0.004	0.981
Ear / facial symptoms	0.153	0.301	−0.275	0.122
Psychological dysfunction symptoms	0.226	0.123	−0.237	0.184
Sleep dysfunction symptoms	0.202	0.169	−0.218	0.222
Q21: “Sense of smell/taste”	−0.454	0.001	−0.080	0.659
QOD-NS total score	−0.234	0.110	−0.016	0.930

DISCUSSION

The relationship between CRS and both reduced olfactory function and reduced QOL is well established, particularly for patients with severe disease refractory to medical therapy. In these patients, improvements in both olfaction and QOL are observed after ESS, and to a lesser degree with long-term continued medical therapy.^{12,13,26–28} However, short-term changes in clinically-measured olfaction and patient-reported QOL are not understood for the average patient with CRS (not only those patients who are medically refractory surgical candidates) who presents to rhinology clinic and is treated with AMT.^4,10,11 This multi-institutional observational cohort study aimed to identify these short-term changes associated with patient-specific AMT assigned by the treating provider according to evidence-based guidelines. This observational methodology revealed that a majority of patients had already completed multiple medical therapies for CRS at presentation to the enrolling rhinology center, and thus the “comprehensive” medical therapy intervention we anticipated studying was in reality “appropriate” additions and modifications to existing therapy. This short-course of individualized AMT was associated with significant within-subject improvements in patient-reported outcomes for the total cohort; improvements exceeding the MCID of the SNOT-22 and QOD-NS were observed in 59% (n=23) and 21% (n=8) of subjects, respectively. In contrast to these improvements in patient-reported outcomes, no significant improvements were identified for clinically-measured olfaction (Sniffin’ Sticks TDI total score and all subscores [T, D, and I]) or endoscopy score (Lund-Kennedy score).

Observation of QOL and OD outcome changes in response to individualized AMT, without a specific predefined regimen of therapy, supported our goal of understanding these relationships in real-world clinical practice. Patient-reporting of therapies has inherent drawbacks such as recall bias, but this method was felt to more accurately relate changes in outcomes to actual therapies used, rather than what may have been prescribed. In order to identify which specific therapies the observed outcome changes were attributable to, differences in patient-reported medical therapy use from baseline to follow-up were retrospectively determined. Significant increases in use were only reported for saline irrigations and prescribed topical nasal steroid rinses/irrigations (i.e., budesonide irrigations - the only steroid irrigation utilized by participating institutions). Thus changes associated with AMT may be primarily attributable to effects of saline and topical budesonide irrigations. This is particularly interesting and unanticipated considering the significant improvements seen in short-term patient-reported outcome measures, as it suggests that these interventions which are normally viewed as longer term maintenance may have significant effects on QOL even in the short term (average 7.8 weeks). The specific types of therapy reported is also important to consider in relating these findings to other patient populations with different regimens of medical therapy.

The observed discrepancy between patient-reported improvement and lack of statistical significance for clinically-tested outcomes has multiple possible interpretations. Median TDI scores for Sniffin’ Sticks testing did improve by 1.5 (IQR 4.1) from 24.8 (IQR 36.0) at baseline to 27.5 (IQR 12.6) at follow-up, though this change did not reach significance (p=0.109). This lack of significant change in clinical tests may simply reflect small sample size (TDI and Lund-Kennedy scores, n=33; SNOT-22 and QOD-NS surveys, n=39) and cohort heterogeneity (increased variance) precluding statistical power to resolve a real difference. The sample size was slightly greater for patient-reported (n=39) compared to clinically-tested measures (n=33), but the median change of −12.0 (IQR 32.0) in SNOT-22 score from 43.0 (IQR 30.0) at baseline to 26.0 (IQR 22.0) at follow-up was also of greater magnitude compared to TDI scores change. The specific SNOT-22 survey question (Q#21) on ‘sense of smell/taste’ was explored for further insight into this difference in patient-reported and clinically-tested findings, and to evaluate its utility as potential proxy measure for clinically measured olfaction.

The ‘olfactory impairment’ (Q#21) and ‘nasal obstruction’ (Q#22) items added to the SNOT-20 questionnaire resulted in the now validated and most widely used SNOT-22 survey.^8,29 Although not validated independent from the SNOT-22, we considered scores for Q#21 as a simple subjective query of dysosmia. Scores for Q#21 demonstrated uniquely significant correlation to baseline TDI score and Lund-Kennedy endoscopy scores, contrasting the lack of correlation between all SNOT-22 subdomain scores and these clinically measured outcomes. Consistent with these findings, dysosmia has been described as the only symptom significantly associated with radiographic findings of CRS, differentiating patients from those with radiographically normal sinuses.³⁰ Despite this relationship of dysosmia to objective measures, we found SNOT-22 Q#21 scores improved along with patient-reported measures (SNOT-22 total, SNOT-22 subdomains, and QOD-NS scores), incongruent to the lack of significant change in objective measures (TDI and Lund-Kennedy score). Therefore, change in individual symptom severity (e.g., dysosmia) after short-term AMT may have less relationship to empirically measurable alternations in a physiologic process (e.g., olfactory function), than to more global change in perception of disease severity and QOL. These results should again be interpreted in context of low statistical power, as a potential relationship between symptom severity and measured olfactory function may have been missed due to the limited sample size and the heterogeneity of baseline olfactory function in the study cohort.

Despite an anticipated cost of reduced statistical power, this study was specifically designed as an observational cohort of all enrolled patients in order to maximize real-world clinical applicability. In general, patients with more severe baseline impairment have greater opportunity for improvement with treatment, while normal baseline function may preclude additional improvement (ceiling effect). In the current study 33% (n=16) of the cohort was characterized as normosmic at baseline by Sniffin’ Sticks testing, limiting their potential for further improvement with intervention. Inclusion of all patients regardless of baseline impairment creates a sample cohort more closely resembling the population, but dilutes power to detect statistically significant differences. For this reason, studies with an objective of identifying significant changes attributable to an intervention may select a cohort of participants with high baseline impairment. Publication bias may then contribute to promoting such studies in the literature, and although valid and valuable for select patients, their clinical applicability may be limited. Although the unrefined observational cohort utilized in this study was chosen to maximize applicability of findings to clinical practice, other sources of selection bias such as performing this study at academic rhinology centers should still be considered.

CONCLUSION

Adult patients enrolled in this observational multi-institutional cohort study demonstrated varied baseline impairment in olfactory function and QOL, and were treated with appropriate medical therapy consisting primarily of nasal saline and topical steroid irrigations. At short-term follow-up (average 7.8 weeks), significant improvements were observed for patient-reported measures of QOL (SNOT-22 and QOD-NS), but not for clinically measured olfactory function (Sniffin’ Sticks TDI total score). Olfactory function in this cohort correlated with endoscopy score, but lacked correlation to change in QOL.