Abstract
Objective:
To critically review the literature on nasal nitric oxide (nNO) and its current clinical and research applicability in the diagnosis and treatment of different sinonasal inflammatory diseases, including acute bacterial rhinosinusitis (ABRS), allergic rhinitis (AR), and chronic rhinosinusitis (CRS).
Methods:
A search of the PubMed database was conducted to include articles on nNO and sinonasal diseases from January 2003 to January 2020. All article titles and abstracts were reviewed to assess their relevance to nNO and ABRS, AR, or CRS. After selection of the manuscripts, full-text reviews were performed to synthesize current understandings of nNO and its applications to the various sinonasal inflammatory diseases.
Results:
A total of 79 relevant studies from an initial 559 articles were identified using our focused search and review criteria. nNO has been consistently shown to be decreased in ABRS and CRS, especially in cases with nasal polyps. While AR is associated with elevations in nNO, nNO levels have also been found to be lower in AR cases with higher symptom severity. The obstruction of the paranasal sinuses is speculated to be an important variable in the relationship between nNO and the sinonasal diseases. Treatment of these diseases appears to affect nNO through the reduction of inflammatory disease burden and also mitigation of sinus obstruction.
Conclusion:
nNO has been of increasing interest to researchers and clinicians over the last decade. The most compelling data for nNO as a clinical tool involve CRS. nNO can be used as a marker of ostiomeatal complex patency. Variations in measurement techniques and technology continue to impede standardized interpretation and implementation of nNO as a biomarker for sinonasal inflammatory diseases.
Keywords
Introduction
There has been an increasing interest in nitric oxide (NO) as a biomarker for various inflammatory pathologies of the upper and lower respiratory tracts. NO plays a role in many physiologic pathways within the body, including blood circulation, immune function, and inflammatory processes.1-3 NO is produced by several different nitric oxide synthases (NOS) within the respiratory tract, endothelium, and immune cells.4,5 Within the lower respiratory tract, major sources of NO are from inducible nitric oxide synthase 2 (iNOS2), which is triggered by Type 2 inflammation and eosinophilia. 6 Similarly, iNOS2 generates NO in the upper respiratory airway, especially within the paranasal sinuses. 7
In the early 1990s, NO was found to be elevated in the exhaled breath of asthmatic patients.8-10 NO in the lower airway, commonly measured as fractionated exhaled nitric oxide (FeNO), serves as a clinically relevant biomarker for asthma. 10 The diagnostic accuracy of FeNO for asthma has shown moderate sensitivity and specificity overall depending on cutoff values between 20 and 40 parts per billion (ppb). 11 FeNO levels are significantly decreased in patients undergoing asthma therapy with inhaled corticosteroids, leukotriene receptor antagonists, and biologic therapies, potentially also serving as a tool for monitoring disease treatment.6,12-15
Given the utility of NO as a biomarker for lower airway disease, additional research has been performed on NO in the upper aerodigestive tract, including the sinonasal cavities. Nasal nitric oxide (nNO), as measured by aspiration of the nasal cavities with velum closure, is hypothesized to be the most effective measurement for various sinonasal inflammatory diseases due to the high concentration of NO in the paranasal sinuses. 7 Currently, nNO has shown the most promise as a diagnostic tool in primary ciliary dyskinesia (PCD), which is characterized by very low nNO levels.16-18 Most recently, Shapiro et al 17 established the diagnostic accuracy of nNO for PCD with a sensitivity and specificity of 97.5% and 96%, respectively, when compared to the gold standard of electron microscopy. Initially thought to be only useful as a screening test, nNO has now been recommended by the European Respiratory Society (ERS) task force and American Thoracic Society (ATS) for diagnostic use of PCD.19-21
The use of nNO as a diagnostic test for PCD has further increased the interest in the utility of nNO in other sinonasal pathologies. The objective of this review is to highlight the relationship between nNO and various sinonasal pathologies, including acute bacterial rhinosinusitis (ABRS), allergic rhinitis (AR), and chronic rhinosinusitis (CRS). Although systematic reviews have been performed on each of these pathologies singly, a general overview of nNO in sinonasal pathologies has yet to be published.22-25 A synthesis of the literature will help direct further research needs on the role of nNO as a biomarker for upper airway diseases.
Methods
A search of the PubMed database was conducted to include relevant articles on nNO measurement and sinonasal inflammatory diseases from January 2003 to January 2020. Articles were included if they measured nNO in the adults or children with any of the following sinonasal pathologies: ABRS, AR, and/or CRS. Retrospective, prospective studies, and any systematic reviews and meta-analyses of nNO and ABRS, AR, and CRS were included. Articles were excluded if they were not written in English, did not report nNO levels, or did not include nNO in the context of sinonasal disease. All article titles and abstracts were reviewed (JB, ML) to assess their relevance and included according to the inclusion criteria. Any disagreement on inclusion of articles was settled by the senior authors (JH, KL). After selection of relevant manuscripts, full-text reviews were performed to synthesize current understanding of nNO, and its application to sinonasal diseases.
Results
A total of 559 articles were identified using our focused search criteria after removing duplicates. After reviewing article titles and abstracts, 79 relevant manuscripts were selected for full review. The following sections will discuss nNO in the context of ABRS, AR, and CRS, respectively.
Discussion
Techniques for Measuring Nasal Nitric Oxide
Accurate measurements of nNO must rely on the sampling of the sinonasal cavities without including lower airway sources and thus require velum closure and nasal air aspiration.26-29 The 2 most common techniques to achieve accurate sampling involves velum closure via oral exhalation against resistance or with breath holding. Aspiration of the nasal cavity is then performed via an occlusive nasal olive which is attached to the measurement device.30,31 Both methods have shown adequate velum closure and repeatability. 32 Oftentimes, researchers have taken measurements from both nares and average the 2 values; studies have shown very little variability in laterality of nNO in healthy controls. 33 Early studies have also suggested nNO sampling during humming may reflect ostiomeatal complex (OMC) obstruction, but more recent data show the humming technique does not predict radiographic OMC obstruction more effectively than the velum closure technique.34-37 The ATS and ERS published a standardized protocol for nNO measurement in 2005, but since that time, advancements in technology and new data suggest that a new protocol is needed. 29
In regard to the equipment used for nNO measurement, early studies utilized chemiluminescence NO analyzers, which have become the gold standard for measuring nNO. 38 Chemiluminescence analyzers remain very expensive, and their availability in clinical practice is limited. 39 Most research regarding standardization of nNO measurement utilizes a chemiluminescence analyzer due to its use as a validated diagnostic tool for PCD.17,21 A nNO value of >77 nl/min (calculated by measured ppb multiplied by flow rate) makes the diagnosis of PCD unlikely. 31 Many variables, however, have been shown to effect these values, including aspiration flow rate, nasal secretions and edema, atopic status, and intranasal steroid use.22,30,40,41
Newer devices that utilize electrochemical technology to measure NO levels have been more widely available and are less expensive.41,42 The NIOX MINO and VERO devices (Circassia Limited, Oxford, UK) have been approved by the US Food and Drug Administration (FDA) for FeNO measurements for asthma, but have also been utilized in nNO research. Numerous studies have utilized this device and have shown standardized reference values, repeatability, and good patient tolerance. Harris et al 43 compared values from a portable electrochemical sensor device to a standard chemiluminescence device for the diagnosis of PCD, and concluded that the handheld electrochemical NO device was an effective and reliable diagnostic tool. For this handheld device, however, nNO measurements were obtained during normal breathing without velum closure, which may have been a confounding factor. A similar study by Montella et al 44 comparing the 2 types of devices using ATS guidelines for measurement, found significant correlation of NO values in both PCD and healthy control populations.
Handheld electrochemical NO analyzers have thus been shown to be as effective for measuring nNO compared to the gold standard of chemiluminescence. Due to their relatively low cost and wider availability, they appear to be a more practical tool for nNO measurement in clinical practice. Further studies are required to obtain FDA approval for nNO measurement using these devices.
nNO in Acute Bacterial Rhinosinusitis
ABRS is a clinical diagnosis based on the duration of symptoms and the cardinal symptoms of facial pain/pressure, nasal obstruction, and purulent nasal drainage. 45 The addition of an effective biomarker for ABRS would help improve disease monitoring and expedite appropriate treatment. 46 nNO has been hypothesized to be an effective biomarker due to its involvement in inflammatory reactions within the paranasal sinuses. 2 Table 1 summarizes the nNO data collected in patients with ABRS.
Summary of Acute Bacterial Rhinosinusitis Studies.
Abbreviations: ABRS, acute bacterial rhinosinusitis; CT, computed tomography; OMC, ostiomeatal complex; PAR, perennial allergic rhinitis.
In an early study, a cohort of pediatric patients with maxillary sinusitis was noted to exhibit decreased nNO at the time of diagnosis and an initial increase in nNO after treatment with antibiotics. 47 This study, however, was not performed with velum closure, thus limiting the interpretation of the data. Lanz et al 48 similarly measured nNO in ABRS patients while humming and found similar results of low nNO with increased values after treatment. The findings of decreased nNO values in the acute disease state and subsequent increases in nNO after treatment are seemingly paradoxical to the use of nNO as an inflammatory marker. The paradoxical relationship between nNO and the inflammatory burden in ABRS has been speculated to be caused by sinus obstruction. In ABRS, sinus obstruction is hypothesized to inhibit the ability to sample nNO produced by the sinuses during acute infections. In a prospective study in which 50 patients were followed during an ABRS episode, a decrease in nNO values correlated with the evidence of sinus inflammation and OMC obstruction on CT imaging, and elevated levels of C-reactive protein. 49 These findings have highlighted the importance of OMC obstruction in ABRS as the likely cause of decreased nNO in acutely infected sinuses.
nNO has been shown to correlate with ABRS, but given the overlapping values with other disease processes, it may be limited as a clinical tool for diagnosis. Wen et al 50 evaluated patients with perennial allergic rhinitis (PAR) and concomitant unilateral sinusitis, and attempted to identify sensitivity and specificity of nNO as a diagnostic tool for ABRS. The group identified a nNO cut-off value of <286 ppb for patients with comorbid PAR and ABRS with a sensitivity of 95.5% and specificity of 86.1%. Interpretation of this study is limited because there were no objective measures to diagnose unilateral sinusitis in these patients and the presence of an atopic disease.
Despite the heterogenous methods of nNO sampling, the general consensus is that nNO is decreased in active ABRS. Although there is a lack of data for a cut off value for nNO in ABRS, it may be used as a tool for monitoring the response to treatment.
nNO in Allergic Rhinitis
Given the ability of NO to measure eosinophilic inflammation in the lower airway, there has been interest in nNO for the diagnosis and management of AR. Early studies identified elevated FeNO in AR patients compared to normal healthy controls, but due to the high prevalence of comorbid asthma in this population, it was difficult to determine the individual contributions of asthma and AR to FeNO values.48,51 nNO, if performed correctly, isolates the nasal airway, and thereby reduces the sampling of NO from the lower airway, and contributions of comorbid asthma.26,52-55 Nonetheless, based on a small cohort study of AR, Nesic et al 56 has suggested a nNO value of 564 ppb as a diagnostic cutoff specifically for AR. The sensitivity and specificity of such a value were 83% and 80%, respectively. Still, other studies have suggested values as low as 38.5 ppb and as high as 1614 ppb.26,57,58 The wide variability in these measurements highlight the importance of standardization of measurement techniques. The low values likely represent sampling of the lower airway due to nasal exhalation rather than aspiration sampling during velum closure.58-60 Such wide variability was confirmed by a high heterogeneity amongst studies included in a systematic review and meta-analysis on nNO in AR. 24
In AR, nNO and disease-specific symptom severity via Sinonasal Outcome Test (SNOT)−22 questionnaires have actually demonstrated an inverse relationship.47,61 Takeno et al 62 evaluated nNO at the inferior turbinate and middle meatus in AR patients whose disease severity were stratified as mild or moderate/severe. nNO was found at higher levels in AR patients because of the eosinophilic inflammation associated with the disease pathophysiology, but as disease severity, including nasal obstruction, worsened, nNO values decreased accordingly. The lower nNO values in AR patients with significant nasal obstructive symptoms were thought to be secondary to increased sinus obstruction from disease burden, which in turn limits the sampling of NO levels in the sinus cavities. 40
As expected, treatment of AR is associated with a decrease in nNO levels.63,64 This relationship is hypothesized to be in part due to reduction of ongoing eosinophilic inflammation of the sinonasal cavities. 52 Antosova et al analyzed nNO values for AR after treatment with either oral antihistamines alone or in combination with nasal corticosteroids. The patients treated with combined therapy showed a statistically significant decrease in nNO, while the patients taking antihistamines alone did not. 65 This study concluded that combination therapy is more effective in reducing overall production of nNO. There are additional studies that do not show a decrease in nNO with AR treatment; however, these conflicting results may reflect confounding variables, including disease severity, treatment methods and duration of treatment.35,36 Table 2 provides a summary of studies evaluating nNO and treatment of AR.
Summary of Key Allergic Rhinitis Studies.
Abbreviation: AR, allergic rhinitis.
nNO as a biomarker for AR is also affected by concomitant inflammation of the sinuses, thereby affecting the utility of nNO as a diagnostic tool for AR alone. Patients with AR and concurrent CRS with or without nasal polyps have significantly lower nNO compared to both healthy controls and patients with AR alone.35,66 nNO can additionally be used to lateralize the sinus disease in AR patients when comparing measurements from either nares: a lower nNO when compared to that from the contralateral nasal cavity may suggest concomitant sinus disease. 50 Current data suggest that although nNO is elevated in active AR, it does not appear to be the optimal method for diagnosis or monitoring treatment response. 24 Further research on the effects of concomitant inflammatory diseases on nNO in patients with AR is required24,25
nNO in Chronic Rhinosinusitis
The most promising area for nNO as a biomarker has been identified in CRS patients both with polyps (CRSwNP) and without polyps.22,23 Initial theory has suggested that the inherent inflammation in CRS would result in an overall increase in nNO, as evident in the relationship between AR and nNO levels. In an early study by Lundberg et al35, however, nNO was noted to be paradoxically low in patients with CRSwNP; this relationship was attributed to sinus cavity obstruction that decreases the total volume of nasal airflow available for nNO sampling.34,67 Additional key studies, as summarized in Table 3, have confirmed that low nNO is seen in both CRSwNP and CRSsNP patients.34,62,67-79
Summary of Key Chronic Rhinosinusitis Studies.
Abbreviations: CRSsNP, chronic rhinosinusitis without nasal polyposis; CRSwNP, chronic rhinosinusitis with nasal polyposis.
Several studies have attempted to establish a diagnostic cutoff value for CRSwNP.72,75 Bommarito et al, for example, suggested a diagnostic cutoff of value for nNO < 442 ppb with a specificity of 91% and sensitivity of 87%. Jeong et al suggested a much lower value of <163 ppb, demonstrating a specificity and sensitivity of 93% and 81%, respectively. The difference between these values is likely related to the phenotypic and endotypic heterogeneity of the CRSwNP population. These studies furthermore included patients with concomitant AR which may explain the large variation. There is currently no accepted standardized nNO value for the diagnosis of either CRSwNP or CRSsNP.
nNO has also been studied as a tool for disease monitoring and treatment response. 80 Specifically, the use of nNO to help delineate patients with CRSwNP from those with CRSsNP has significant clinical relevance. Ambrosino et al 22 evaluated the largest pool of CRS patients to date and evaluated the difference in nNO values among CRSsNP, CRSwNP, and healthy patients. This meta-analysis consisted of 23 prospective studies which compared CRSsNP to healthy patients (7 studies), CRSwNP to healthy patients (2 studies), CRSwNP to CRSsNP patients (6 studies), and a mix of both CRSsNP and CRSwNP to healthy patients (8 studies). Overall, the largest cohort of patients had CRSwNP. The standardized mean difference (SMD) between CRSwNP and healthy controls was statistically significant at −1.49, suggesting that CRSwNP patients had decreased nNO levels when compared to healthy controls. Additionally, when CRSwNP were compared to CRSsNP, the SMD was still significant at −1.44. Interestingly, when meta-analysis regression was performed on nNO aspiration flow rate, a significant impact on nNO measurement was observed; this may be an important variable to standardize in future studies. The difference in nNO between CRSsNP and healthy controls was no longer statistically significant when corticosteroid use was considered. Overall, this study showed that nNO is indeed lower in CRSwNP than CRSsNP, although no specific nNO values or cutoffs are established.
The degree of nasal polyposis has been determined to be inversely correlated with nNO, allowing nNO to serve as a helpful biomarker to monitor disease treatment.67-69 Ragab et al performed a prospective trial on CRSwNP and CRSsNP patients, in which nNO was longitudinally followed with either surgical or medical treatments. After an initial medical trial of nasal steroid sprays, patients with persistent symptoms were randomized to a medical arm of steroid sinus rinses and oral erythromycin. In addition to these medications, CRSwNP patients received an additional dose of oral steroids and topical steroid drops. The surgical arm underwent endoscopic sinus surgery (ESS) with post-operative nasal steroid rinses and topical sprays. Interestingly, there was no statistically significant difference in nNO between the surgical and medical arm at 6 and 12 months of therapy, but both arms showed increases in nNO which was statistically significant. 81
Other studies have shown that CRS patients who undergo ESS exhibit significant increases in their nNO levels post-operatively.81,82 Fu et al followed CRSwNP and CRSsNP patients in the post-operative period and measured nNO at 3, 6, and 12 months following surgery. Patients with CRSwNP had a significant increase in nNO postoperatively that plateaued at 3 months, while CRSsNP patients achieved a plateau of nNO at 6 months postoperatively. The difference in the nNO plateau may indicate different inflammatory pathways between the 2 disease phenotypes. In CRSwNP, quality-of-life measures, as measured by SNOT-22, were noted to have more significant improvements in patients with lower nNO in the preoperative period. 73 An additional study showed the post-operative increase in nNO for CRSwNP is proportional to the degree of polyp size reduction.64,67 CRS patients with olfactory dysfunction have low nNO as expected, and nNO has a positive correlation with olfactory function after treatment. 83
The current understanding is that re-establishing sinus cavity patency provides the largest effect on nNO. For CRS, nNO has also been correlated with CT scores, showing decreased nNO levels in patients with elevated Lund-Mackay score.40,72,81 Symptom severity scores also inversely correlate with nNO measurements, as patients with increased SNOT-22 scores has been noted to have lower nNO, though this is not confirmed in other studies.77,81,82 Although there are no specific values that indicate adequate treatment or resolution of symptoms, further studies may help delineate the prognostic value of nNO in the post-operative period.
Although very few studies have been performed on allergic fungal rhinosinusitis (AFRS) and nNO, there is some data that suggests it may be useful in the diagnosis and management of this CRS variant. In 1 study, patients with unilateral AFRS were found to have very low nNO in the affected side with a cutoff of 239 ppb. The suggested cutoff value allowed for a diagnostic sensitivity and specificity of 79% and 87.2%, respectively. 84
Conclusion
nNO has been of increasing interest over the last decade. The most compelling data for nNO as a clinical tool is in ABRS and CRS patients, in whom nNO levels have found to be lower than in healthy individuals. AR generally has been associated with elevated nNO levels, but nNO levels have also been found to be lower in AR cases with more severe nasal obstructive symptoms. These relationships between nNO levels and various sinonasal pathologies highlight the important role of the OMC as a major site of pathologic obstruction. nNO may thus be expected to predict and monitor OMC patency in patients with sinonasal pathologies. Variations in measurement techniques and technology, however, continue to impede standardized interpretation and implementation of nNO as a widely used clinical tool.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported with a Medarva Foundation Research Grant.
