Factors Affecting Nasal Discharge in Chronic Rhinosinusitis Patients

ABSTRACT

Background

Nasal discharge is a cardinal symptom of chronic rhinosinusitis (CRS); however, patients often use variable descriptions for this symptom. The 22‐item Sinonasal Outcomes Test (SNOT‐22) presents four different questions regarding this symptom. The objective of this cross‐sectional analysis was to determine which patient‐specific factors influence the severity of nasal drainage.

Methodology

Demographic information, medical history, and SNOT‐22 scores were recorded at baseline as part of a multi‐institutional, prospective cohort study of patients with CRS refractory to medical therapy. Severe symptoms were defined as scores ≥ 3 on the SNOT‐22 item response. Bivariate odds ratio (OR) calculations and multivariate regression analysis were completed.

Results

A total of 1020 patients were included, equally split based on gender and polyp status. On the SNOT‐22, postnasal discharge (PND) was the most severe discharge symptom reported, with an average score of 2.98 [SD ± 1.5] and 67.3% of patients reporting severe disease. Female sex (OR = 2.0, [95% CI: 1.5–2.5]), diagnosis of CRS without nasal polyps (OR = 1.5, [95% CI: 1.1–2.0]), prior endoscopic sinus surgery (ESS) (OR = 1.5, [95% CI: 1.1–2.0]), and allergies (OR = 1.4, [95% CI: 1.1–1.8]) were significantly associated with severe PND, while comorbid acid reflux was not (OR = 1.1, [95% CI: 0.8–1.6]; p = 0.53).

Conclusion

Of the various forms of nasal discharge, patients with CRS report worse symptoms related to PND. This is the first study to evaluate the effects of demographic and comorbid conditions on the presentation and severity of PND. Female sex, CRSsNP, prior ESS, and comorbid allergies are associated with higher odds of having severe PND.

Level of Evidence

3

Trial Registration: www.clinicaltrials.gov: NCT02720653 and NCT01332136

This study aimed to identify patient‐specific factors influencing the severity of nasal drainage in chronic rhinosinusitis (CRS) patients. It found that postnasal discharge (PND) was the most severe symptom, with female gender, CRS without nasal polyps, prior endoscopic sinus surgery, and allergies significantly associated with severe PND. Comorbid acid reflux was not significantly associated with severe PND.

1. Introduction

Chronic rhinosinusitis (CRS) is a significant health condition in the United States, with CRS patients reporting severe symptoms at baseline [1, 2, 3]. Nasal discharge is a cardinal symptom of CRS and can be described in various ways. In fact, the Sino‐Nasal Outcomes Test (SNOT‐22) survey tool includes four different descriptions of nasal discharge including “need to blow nose,” “runny nose,” “post‐nasal discharge” (PND), and “thick nasal discharge” [4]. In a 2019 study of SNOT‐22 questionnaire data, 3 of the 5 most important symptoms of CRS were thick nasal discharge, need to blow the nose, and PND [5].

There exists considerable variability in CRS symptom burden between demographics such as age, race, and sex [6, 7]. For example, females with CRS have been observed to have worse quality‐of‐life (QOL) at baseline compared to males with CRS [8]. Asthma, allergies, and acid reflux can affect the upper airway and modulate symptom presentation and severity, and CRS subtype and nasal polyp status also impact the disease course [9]. In addition to severity at presentation, these patient‐specific factors can impact treatment outcomes. A recent study observed that 45% of patients had the persistence of nasal discharge even after endoscopic sinus surgery (ESS) [10]. This phenomenon leaves clinicians unclear on if this discharge is due to persistent inflammation from CRS or other comorbidities affecting the upper airway.

The goal of this study was to provide further insight into how different groups of patients experience nasal discharge symptoms in the setting of CRS and explore the impact of comorbid conditions. We hypothesize that of the four nasal discharge questions, there will be differences in symptom severity at baseline, that different demographic groups will experience these symptoms differently and worsen with comorbid conditions of the upper and lower airway [1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 14, 15, 16]. Results of this study will inform clinicians on counseling patients with CRS regarding their nasal discharge.

2. Methods

2.1. Study Population and Inclusion Criteria

This study is a secondary data analysis of a prospective, longitudinal, multi‐institutional study conducted at Oregon Health and Science University (Portland, OR), the Medical University of South Carolina (Charleston, SC), Stanford University (Palo Alto, CA), University of Utah (Salt Lake City, UT), and University of Calgary (Calgary, Alberta, Canada). Inclusion criteria were adult patients (age ≥ 18 years) with medically refractory CRS who were enrolled from 2011 to 2020. Diagnosis of CRS was confirmed using current guidelines on Adult Sinusitis from the American Academy of Otolaryngology [17, 18]. Given that these individuals were evaluated at tertiary rhinology centers, they had typically failed standard medical therapy and frequently demonstrated more severe disease phenotypes, as patients with adequate response to initial medical management are less likely to be referred to or seek care at such specialized centers. Patients who were under 18 years of age or whose CRS symptoms resolved with the first trial of medical therapy were excluded. Refractory to medical treatment was defined as patients who were unable to achieve adequate symptom control following a minimum of a 3‐week course of broad‐spectrum or culture‐directed antibiotics, along with a 3‐week trial of topical steroid therapy. By narrowing inclusion criteria to medically refractory patients, the study aimed to minimize variability introduced by patients with milder or well‐controlled CRS. Informed consent was obtained from all participants, in English, at the enrollment visit. The institutional review board at each participating institution provided approval and regulatory oversight. While patients with cystic fibrosis were eligible for enrollment in the longitudinal study, they were excluded from this secondary data analysis due to the high heterogeneity of symptom presentation.

2.2. Data Collection

Cross‐sectional analyses focused on data collected at the baseline enrollment visit. Demographic information included age, sex, race, ethnicity, household income range, education, and insurance. Medical history was also collected and included a history of sinus surgery, a diagnosis of nasal polyps, allergic fungal sinusitis (AFRS), mucocele, septal deviation, recurrent acute rhinosinusitis (RARS), turbinate hypertrophy, asthma, allergies, aspirin‐exacerbated respiratory disease (AERD), acid reflux, and obstructive sleep apnea (OSA). Clinical examinations such as physician exam, nasal endoscopy, and sinonasal computed tomography (CT) scans were also performed per the standard of care.

2.3. Disease‐Specific Outcome Measurements

Subjects completed the SNOT‐22, a CRS‐specific quality of life instrument (2006, Washington University, St. Louis, MO). This survey is a highly validated instrument consisting of 22 questions (scale: 0 = “No problem,” 1 = “Very mild problem,” 2 = “Mild or slight problem,” 3 = “Moderate problem,” 4 = “Severe problem,” and 5 = “Problem as bad as it can be”) with lower scores associated with less severe symptom burden [4]. The SNOT‐22 questions of interest in this study pertain to nasal discharge and include “need to blow nose,” “runny nose,” “post‐nasal discharge,” and “thick nasal discharge.” In addition to measuring the average and proportional severity of each SNOT‐22 question, this study evaluated the proportion of subjects reporting a severe PND score. Mild scores were recategorized as PND item scores between 0 and 2, and severe scores were recategorized as PND item scores between 3 and 5 [19].

2.4. Statistical Analysis

Study participant data was coded by each enrollment site using unique study identification numbers to protect patient confidentiality and compliance with the Health Insurance Portability and Accountability Act (HIPAA) of 1996 and guidelines for good clinical practice [20]. Variables selected for analysis were closely reviewed by authors to ensure that analysis was scientifically driven by a priori hypotheses. Statistical analyses were completed using SPSS statistical software (ver.29.0; IBM Corporation, Armonk, NY). Descriptive statistics were completed while assumptions of normality were confirmed for all continuous/scaled study data using Shapiro–Wilk testing and graphical distributions. Differences in bivariate proportions of interest were evaluated using both the likelihood ratio chi‐square (Χ ²) test and determinations of odds ratio (OR) values to determine the magnitude of effect estimates. Binary logistic regression models were created to evaluate the multivariate associations between independent covariates and the differential likelihood of reporting severe PND at the time of initial presentation. Candidate covariates were screened for univariate significance (p < 0.05) against recategorized severe PND scores as the primary outcome of interest and a referent cohort. Final modeling was completed using manual forward selection and backward elimination (p < 0.05) procedures in a stepwise process. Goodness‐of‐fit for final logistic modeling was assessed using Hosmer‐Lemeshow chi‐square (Χ ²) statistics. Adjusted OR values and 95% confidence intervals (CI) are reported along with type‐I error probabilities (p values) for all statistical associations of interest.

3. Results

3.1. Baseline Results

A total of 1020 patients with CRS were included in this analysis. The average age was 50.6 years [Range: 18–92] and just over half were female (51.6%). The race of most patients was White (85.0%) and non‐Hispanic/Latino ethnicity (94.9%). Over half of the patients had previously undergone ESS (53.8%) and the most common comorbid diagnosis was allergies (47.3%). A full summary of demographic and comorbidity information can be found in Table 1.

TABLE 1.

Baseline characteristics of study population.

Variable	N	(%)/Mean [±SD]
Age (years at enrollment)	1020	50.6 [±15.3]
Sex—Male	494	(48.4%)
Race—White	867	(85.0%)
Race—African American/Black	79	(7.7%)
Race—Asian	35	(3.4%)
Race—American Indian/Alaska Native	7	(0.7%)
Race—Native Hawaiian/Pacific Islander	1	(0.1%)
Race—Other/unknown/biracial	31	(3.0%)
Ethnicity—Hispanic/Latino	52	(5.1%)
Ethnicity—Non‐Hispanic/Latino	968	(94.9%)
Prior ESS	539	(52.8%)
CRSwNP	444	(43.5%)
Allergies (mRAST/skin prick confirmed)	482	(47.3%)
Asthma	414	(40.6%)
Acid reflux	162	(15.9%)
COPD	49	(4.8%)
Diabetes mellitus (type I/II)	80	(7.8%)
Depression (self‐reported)	198	(19.4%)
AERD/ASA sensitivity	110	(10.8%)
AFRS	38	(3.7%)
OSA	108	(10.6%)
Smoking Status (any, current, former)	138	(13.5%)

Abbreviations: AERD: aspirin exacerbated respiratory disease, AFRS: allergic fungal rhinosinusitis, ASA: aspirin, COPD: chronic obstructive pulmonary disease, CRSwNP: chronic rhinosinusitis with nasal polyps, ESS: endoscopic sinus surgery, mRAST: microradioallergosorbent test, OSA: obstructive sleep apnea.

Baseline measurement of nasal discharge symptoms on the SNOT‐22 revealed PND as the most severe among all patients (Table 2). PND also had the highest proportion of patients reporting severe disease (67.3%), followed by thick nasal discharge (65.0%). The difference between the proportion of patients reporting severe PND and severe thick nasal discharge was not significantly different (2.3%, [95% CI: −1.8%, 6.3%], p = 0.27). Notably, runny nose was least likely to be reported as a severe feature of nasal drainage (48.5%) compared to other symptoms.

TABLE 2.

Baseline nasal discharge score; SD, standard deviation.

SNOT‐22 question	Mean	SD	Proportion of severe disease (3–5)
Need to blow nose	2.70	1.44	61.6%
Runny nose	2.34	1.46	48.5%
Post‐nasal discharge	2.98	1.50	67.3%
Thick nasal discharge	2.87	1.57	65.0%

3.2. PND Specific Analysis

Given that PND was the most severe symptom among all patients, additional analysis was performed. The mean score of PND in the cohort was 2.98. Figure 1 depicts the distribution of PND scores among the total cohort of study participants, with four reported as the most frequent score. Bivariate comparisons were performed to assess how various demographic and comorbidity variables affect the proportion of patients experiencing severe PND. The OR values associated with severe PND were calculated between each independent cofactor against a referent group (Table 3). Patients with female sex, a history of prior ESS, CRS without nasal polyps (CRSsNP), and comorbid allergies had increased odds of having severe PND compared to their referent groups. A history of smoking, comorbid asthma, and reflux did not significantly associate with differential odds of severe PND.

FIGURE 1.

Distribution of postnasal discharge severity categories for all study participants at the time of enrollment. SNOT‐22 severity ranges between 0 (“No problem”) and 5 (“Problem as bad as it can be”). In this study, severe disease was categorized as a score of ≥ 3. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]

TABLE 3.

Post‐nasal discharge bivariate analysis.

Independent cofactors	% of mild PND (n = 330)	% of severe PND (n = 690)	p value	OR (95% CI)
Sex—Female	131 (39.7%)	395 (57.2%)	< 0.001	2.03 (1.56, 2.66)
Race—White	275 (31.7%)	592 (68.3%)	0.31	1.21 (0.84, 1.73)
Race—Non‐White	55 (35.9%)	98 (64.1%)
Ethnicity—Hispanic/Latino	20 (38.5%)	32 (61.5%)	0.34	0.75 (0.42, 1.34)
Prior ESS	153 (28.4%)	386 (71.6%)	0.004	1.47 (1.13, 1.91)
CRSsNP	168 (50.9%)	408 (59.1%)	0.013	1.40 (1.07, 1.82)
Allergies (mRAST/skin prick confirmed)	134 (27.8%)	348 (72.2%)	0.003	1.49 (1.14, 1.94)
Asthma	123 (29.7%)	291 (70.3%)	0.14	1.23 (0.94, 1.61)
Acid reflux	49 (30.2%)	113 (69.8%)	0.53	1.12 (0.78, 1.62)
COPD	12 (24.5%)	37 (75.5%)	0.22	1.50 (0.77, 2.92)
Diabetes mellitus (type I/II)	31 (38.8%)	49 (61.3%)	0.21	0.74 (0.46, 1.18)
Depression (self‐reported)	66 (33.3%)	132 (66.7%)	0.74	0.95 (0.68, 1.32)
AERD/ASA sensitivity	28 (25.5%)	82 (74.5%)	0.10	1.46 (0.93, 2.28)
AFRS	11 (28.9%)	27 (71.1%)	0.65	1.18 (0.58, 2.41)
OSA history	42 (38.9%)	66 (61.1%)	0.13	0.73 (0.48, 1.09)
Smoking history (any, current, former)	50 (36.2%)	88 (63.8%)	0.30	0.82 (0.56, 1.19)

Abbreviations: AERD: aspirin exacerbated respiratory disease, AFRS: allergic fungal rhinosinusitis, ASA: aspirin, COPD: chronic obstructive pulmonary disease, CRSsNP: chronic rhinosinusitis without nasal polyps, ESS: endoscopic sinus surgery, mRAST: microradioallergosorbent test, OSA: obstructive sleep apnea.

Preliminary multivariate logistic regression modeling considered those candidate variables screened for significant bivariate associations with severe PND, including female sex, history of prior ESS, CRSsNP, and comorbid allergies. The final logistic model found each covariate to be retained and associated with increased odds of reporting severe PND compared to each referent group (Table 4). Females had the highest odds of reporting severe PND, followed by CRSsNP, a history of prior ESS, and comorbid allergies. The Hosmer‐Lemeshow Χ ² statistic indicated strong modeling accuracy (p > 0.10) for the final binary logistic model.

TABLE 4.

Composite multivariate analysis.

Final multivariate cofactors	df	OR	(95% CI)	p value
Sex—Female	1	1.96	(1.46, 2.52)	< 0.001
Sex—Male (referent)	—
Prior ESS	1	1.51	(1.14, 2.00)	0.004
No prior ESS (referent)	—
CRSsNP	1	1.51	(1.14, 1.99)	0.004
CRSwNP (referent)	—
Allergies (mRAST/skin prick confirmed)	1	1.39	(1.05, 1.83)	0.021
No allergies (referent)	—
Constant term	1	2.43	—	< 0.001

Abbreviations: CRSsNP: chronic rhinosinusitis without nasal polyps, CRSwNP: chronic rhinosinusitis with nasal polyps, ESS: endoscopic sinus surgery, mRAST: microradioallergosorbent test.

4. Discussion

Nasal discharge is a cardinal symptom of CRS and is described in various ways by patients. PND is the most common and most severe descriptor of nasal discharge, with 70% of patients considering it an important aspect of their disease [5]. This suggests posterior symptoms are more severe than anterior symptoms, which have been observed in the literature [15, 21]. PND has previously been implicated in numerous related disease processes of the upper airway, such as acid reflux, asthma, chronic cough, and allergies. The entanglement of PND with these potentially confounding diseases makes it difficult to counsel patients on how their unique medical history may affect symptom severity. This study found that female sex, prior ESS, CRSsNP, and diagnosis of allergies were associated with higher odds of having severe PND, while diagnosis of acid reflux and asthma did not display a significant association. The information in this analysis will aid otolaryngologists in understanding how patient‐specific factors may affect the presentation of this important CRS symptom.

Previous literature has found females report more severe PND in both CRS with and without nasal polyps, which was consistent with this multivariate analysis [15]. Anatomic size, tobacco susceptibility, and hormonal factors have previously been cited as reasons for why these differences exist between sexes [22]. Additionally, it has been thought that the increased responsiveness of the female immune system may be a factor in severe PND. Females with CRS with nasal polyps (CRSwNP) and asthma were observed to have high levels of autoantigen‐specific IgG in polyp tissue compared to males [16]. This suggests there exist global inflammatory mechanisms to explain the difference in PND severity between males and females with CRS.

Nasal polyposis in CRS has been previously linked to inflammatory processes, as evidenced by eosinophilic mediators in polyp tissue of CRS patients [23, 24, 25]. The presence of polyps has also been observed to affect disease presentation. Patients with CRSwNP typically report more anterior symptoms, such as the need to blow their nose, nasal obstruction, or congestion, while patients without polyps report more posterior symptoms, such as PND [11, 19]. Our results support these findings, with CRSsNP patients reporting more severe PND than CRSwNP patients. One hypothesis for this difference could be due to mechanical obstruction caused by polyps diverting more of the patient's attention to anterior symptoms.

Another group with increased odds of experiencing severe PND is patients with a prior history of ESS. This observation aligns with other research that reports 20 to 30% of patients do not experience a clinically significant improvement in overall symptoms after surgery [5]. This also corroborates research showing 45% of patients experience persistent nasal discharge after ESS [10]. Surgical intervention is typically reserved for patients with disease refractory to medical therapy. Thus, the cohort of patients undergoing surgery may have more severe PND to begin with, and it may be more likely for this to persist post‐operatively.

Sinusitis‐related PND has been implicated both as a cause of chronic cough and as a factor in asthma exacerbation. It is thought there is a common type 2 (T2) inflammatory mediator between these diseases defined by increases in interleukins 4, 5, and 13 cytokines [26]. This phenomenon is summarized by the unified airway theory, which proposes that diseases of the upper and lower airway share a common inflammatory process and require an integrated diagnostic and therapeutic approach [27]. It is interesting that we did not observe significant bivariate associations between asthma diagnosis and severe PND. This is in contrast to prior literature that showed a correlation between poorly controlled asthma and PND [28]. Additionally, a retrospective study by Nakajima et al. found that asthma patients with PND had a significantly longer time until cough disappearance compared to those without PND symptoms [29]. This may indicate that PND is linked to specific phenotypes of asthma, such as cough‐variant or cough‐predominant [29].

Our findings indicate the presence of allergies is significantly associated with an increased odds of having severe PND. One hypothesis for this correlation is due to the impairment of the epithelial barrier in the upper airway, which may increase exposure to allergens and nasal inflammation [30]. Another mediator of this relationship between CRS and allergies may be due to shared inflammatory pathways driving each disease. This link has been supported by several studies that have described associations of atopic disease and sinusitis [31, 32]. However, this relationship is contentious in the literature, with some studies refuting that allergies, allergic rhinitis, or atopy influence CRS severity or diagnosis rates [33, 34, 35]. Despite this discordance, allergy testing and treatment are still considered a reasonable option for CRS patients as the benefits exceed the associated risks [5]. This analysis further supports the role that allergies have in the CRS disease course, specifically their negative effect on PND severity.

Acid reflux is a common condition, affecting as much as 20% of the adult population [36]. Our analysis did not demonstrate an increased odds of severe PND in patients with reflux, which was contrary to our original hypothesis. This result was unexpected as there is literature demonstrating a relationship between CRS and reflux [37, 38]. CRS patients have been shown to exhibit higher esophageal and oropharyngeal pH values and a higher concentration of pepsin compared to those who did not have CRS [37]. Some research has also suggested that PND and excessive throat phlegm are a bodily response to reflux [11]. As a result, it has been historically difficult to differentiate PND as resultant of CRS or acid reflux. The discrepancy between our analysis and the broader literature may have been caused by a difference in the incidence of acid reflux within this cohort (15.9%) compared to the broader US population (20%) [11, 14]. Another confounding factor was whether patients were receiving reflux treatment at the time of their baseline SNOT22 survey, as some data have shown an improvement in CRS symptoms after treatment with proton‐pump inhibitors [39].

Our study provides valuable insights about how different groups of patients experience symptoms related to CRS, specifically PND. There are, however, some limitations to highlight while interpreting these findings. First, the prevalence of acid reflux in this cohort is lower than in the US population [11, 14]. Since, the non‐significant correlation between acid reflux and PND contrasts with prior literature, it is worth considering that this study sample may not reflect the general population. Secondly, this study did not account for potential interactive effects between the severity and treatment status of comorbidities included in the multivariate analysis. A future area of research could consider stratifying each comorbidity group into treated and untreated cohorts, then correlating each group with severe PND. Third, since this cohort is comprised of patients who have previously failed medical therapy, the burden of disease may be more severe than that of those responsive to medication. Furthermore, this study is a secondary analysis of a previously conducted study. Since the data were not collected specifically to address the research question, there is a risk that certain variables relevant to the study's focus may be prone to bias. This design also limited the patient population that could be studied, as the prior analysis only comprised patients with medically refractory CRS; patients who respond to standard medical therapy are typically excluded, as resolution of symptoms obviates the need for further investigation into underlying contributory factors. To mitigate this bias, the authors carefully reviewed the available data to select variables most closely aligned with the research question and applied appropriate statistical methods to account for potential confounders. Lastly, the focus of the present study was a cross‐sectional analysis of CRS symptoms. This study design may introduce recall and selection bias, in which patients may report symptom severity at only one point in time, rather than an aggregate experience, potentially reflecting their single most bothersome symptom at the time of their office visit. Although measures such as clear survey wording and detailed instructions to patients were implemented to mitigate these risks, they remain inherent to the study design. Future studies should consider including both a cross‐sectional and longitudinal approach to mitigate these risks.

5. Conclusion

Nasal discharge symptoms are prevalent and severe in CRS patients but present in a variety of ways. PND has been shown to be particularly burdensome to patients and thus was investigated with a multivariate analysis. Females, CRSsNP, those with allergies, and a history of ESS were found to have higher odds of experiencing severe PND. Asthma and acid reflux were not found to be associated with significant changes. As this was a baseline cross‐sectional study, future research should focus on how these patient‐specific factors impact treatment response and clinical outcomes.

Conflicts of Interest

The authors declare no conflicts of interest.

Patel P. A., Kallenberger E. M., Nguyen S. A., et al., “Factors Affecting Nasal Discharge in Chronic Rhinosinusitis Patients,” The Laryngoscope 135, no. 11 (2025): 4086–4092, 10.1002/lary.32288.