Abstract
Background
Complementary and alternative medicine (CAM) is frequently used in the treatment of chronic rhinosinusitis (CRS) in developed countries. With a plethora of CAM therapies available, their effectiveness and safety are poorly understood in the context of CRS.
Objectives
This article aims to critically appraise the evidence for CAM use in CRS through a systematic review of current literature that investigate the effects of CAM on symptoms and clinical status of adults with CRS.
Study Design
Systematic review and qualitative analysis.
Review Methods
A comprehensive systematic review of the literature was conducted by the authors using 5 databases from inception to July 2017: CINAHL, Cochrane, Embase, PubMed, and SCOPUS. Inclusive medical subject headings and keywords consisted of, but were not limited to, sinusitis and complementary therapies, naturopathy, or traditional Chinese medicine. PRISMA guideline was followed. Using templates by Cochrane Public Health Group and Newcastle-Ottawa Scale, each author extracted data, assessed bias, and computed minimal clinically important difference. Any conflicts were resolved through discussion.
Results
In total, 7 of 7141 articles from 1995 to 2016 were included. Three randomized controlled trials and 4 observational studies were organized into 4 categories of CAM: naturopathy, Chinese medicine, homeopathy, and others. Limited evidence supported the use of Pimpinella anisum and crenotherapy for CRS. Data available on Chinese medicine, homeopathy, and liposomal therapy in CRS were inconclusive due to inherent flaws in the studies.
Conclusion
Overall, there is very limited evidence to support the use of CAM in the treatment of CRS. No significant adverse effects have been found. Given its widespread use, more rigorous data from high-quality research are needed before it can be routinely recommended.
Keywords
Introduction
Chronic rhinosinusitis (CRS) refers to inflammation of the nasal cavity and paranasal sinuses. 1 It is estimated that 5% to 15% of the population worldwide suffers with CRS. 2 The annual cost of CRS to the Canadian society is assessed at $1.3 billion. 3 The health status and quality of life (QoL) among individuals with CRS are comparable to other chronic diseases, such as congestive heart failure and cancer.4,5
The conventional therapy for CRS includes saline irrigation, intranasal corticosteroids, and endoscopic sinus surgery (ESS), but the results are highly variable.1,6 Although the vast majority of CRS cases are managed effectively, 7 many continue to suffer and seek other treatments because the relapse rate can be as high as 47.5%. 6 In a cohort study, CRS patients were twice as likely than the general population to inquire about complementary and alternative medicine (CAM). 8 Up to 79% of individuals in developed countries9,10 and 40% of CRS patients in North America11–13 have reported using at least one CAM therapy.
CAM, defined as “any non-mainstream practice used together with, or in place of, conventional medicine approved by the governing body,” 14 has many different types. Evolved from traditional practices in 19th century Europe, naturopathy includes colloidal silver, xylitol, and Pimpinella anisum, a herb that has been widely used as analgesic, expectorant, and carminative. 15 Homeopathy was developed in 18th-century Germany with a theory that a disease can be cured by minimal dose of a substance that produces similar symptoms. 16 Traditional Chinese medicine includes plants, Tai Chi, and acupuncture, a reflex therapy that affects the nervous system and autonomic regulations.17,18 The use of CAM may lead to adverse events, increasing the risk of drug interactions and morbidity. 19 Given its popularity, potential side effects, and the lack of consensus on its effectiveness as a treatment for CRS, this article aims to critically appraise the evidence for CAM use in CRS through a systematic review.
Methods
Search Strategy and Selection Criteria
A comprehensive review of the literature was conducted using 5 databases: CINAHL, Cochrane, Embase, PubMed, and SCOPUS. Search terms were identified based on the expertise of an otolaryngologist (C. J. C.) and CAM literature. The review protocol was predetermined, created, and disseminated among the authors and 2 medical librarians. Selection process and flow diagram were developed based on PRISMA guideline. 20
Medical subject headings (MeSH) and search strings used in this study were sinusitis in conjunction with complementary therapies, naturopathy, homeopathy, traditional Chinese medicine, or chiropractic. Each descriptor consisted of up to more than 100 terms. For example, a MeSH term of complementary therapies included, but were not limited to, acupuncture, traditional medicines, mesotherapy, and organotherapy. To find all relevant articles, the search incorporated additional keywords and truncation symbol, including rhinosinusitis and homeopath*. The search was conducted from the time of database inception to July 2017. Studies in any language were screened (but only titles with a full English translation were included). Search of gray literature, consisting of research data that are unpublished or exist outside of traditional publications, was conducted through Canadian Electronic Library, Grey Literature Report, and OpenGrey. All articles were reviewed using Rayyan application. 21
The search results were screened by 2 authors (C. J. C. and A. J. J.) independently using the titles and abstracts. Full-text articles were reviewed to determine eligibility based on inclusion and exclusion criteria (Table 1). The authors of respective studies were contacted to obtain and clarify relevant information. The primary outcomes of interest were CRS symptoms and clinical status assessed through validated tests, including sinonasal outcome test 20/22 (SNOT 20/22), Chronic Sinusitis Survey (CSS), short-form healthy survey (SF-12/36), visual analogue scale (VAS), nasal endoscopy (NE) score, and radiologic (Lund–Mackay) score. Secondary outcomes included tolerability, complications, and mucosa ciliary beat frequency.
Inclusion and Exclusion Criteria.
Abbreviations: CAM, complementary and alternative medicine; CRS, chronic rhinosinusitis; CSS, Chronic Sinusitis Survey; RCT, randomized controlled trial; SNOT-20/22, sinonasal outcome test 20/22; VAS, visual analogue scale.
Critical Appraisal and Data Extraction
Included studies were analyzed by 2 reviewers (C. J. C. and A. J. J.). Randomized controlled trials (RCTs) were subjected to critical appraisal and bias risk assessment using the “Data Extraction and Assessment Template,” published by the Cochrane Public Health Group. 22 Non-RCTs, including case control and cohort studies were analyzed using Newcastle-Ottawa Scale (NOS). 23 The level of evidence for each study was determined based on the Oxford Centre for Evidence-Based Medicine guideline. 24
Data from all studies, including study characteristics, methods, participants, interventions, and outcomes, were extracted by the 2 reviewers individually using the template created by the Cochrane Public Health Group. 22 Extracted data were compared to identify discrepancies, which were subsequently resolved via discussions.
Data Synthesis
For quantitative data on CRS symptoms and QoL, minimal clinically important difference (MCID) was computed using a distribution-based method by multiplying standard deviation of sample mean at baseline by 0.5. Corresponding to Cohen’s estimate for a medium effect size, 27 this MCID value was found to be reliable for various research instruments, including SNOT-22, VAS, and CSS in CRS research.27–29,32–34 Observed score changes were compared to the MCID to assess clinical relevance. Wherever possible, numerical values of the outcomes with 95% confidence interval (95% CI) were provided for each treatment. In addition, qualitative analysis was performed by pooling the data of each category of CAM for meta-synthesis.
Results
A total of 7141 studies were identified from the 5 databases and 7134 were excluded (Figure 1). Numerous CAM therapies for CRS were identified. Based on the established criteria, 7 articles from 1995 to 2016, involving a total of 583 subjects, were included and categorized into 4 categories. Both authors independently used the Cochrane “Data Extraction and Assessment Template” and NOS to critically appraise each article for bias, extract data, and compute MCID. The results and levels of evidence are summarized in Table 2.

Flow diagram of literature search and study selection.
Summary of Studies Investigating the Use CAM in CRS.
Abbreviations: CAM, complementary and alternative medicine; CRS, chronic rhinosinusitis; NOS, Newcastle-Ottawa Scale; RCT, randomized controlled trial; SNOT-20, sinonasal outcome test 20.
Naturopathy
Two studies exploring the effects of naturopathy on CRS were identified. In a single-blinded RCT of 48 patients with CRS without nasal polyps (CRSsNPs), 33 intranasal administration of P. anisum led to a greater improvement (P = .012) in SNOT-22 score (mean decrease of 20.8 ± 16.2 from baseline of 44.9, 95% CI: 14.2–27.4, P < .001) than intranasal fluticasone spray alone (10.36 ± 10.30 from 44.7, 95% CI: 5.79–14.9, P < .001) at 4 weeks posttreatment. A similar but insignificant (P = .067) relationship was observed in Lund–Mackay scoring of CT between P. anisum (mean decrease of 2.22 ± 2.94, 95% CI: 0.92–3.53, P = .002) and control (0.76 ± 1.39, 95% CI: 0.05 to 1.48, P = .038). P. anisum was associated with nasal discharge (n = 7), nausea (n = 3), and headache (n = 1), but no serious side effects. The MCID of SNOT-22 for the P. anisum group was −10.4. However, the role of placebo in both groups cannot be ruled out. The lack of blinding increases the risk of performance bias, and the exclusion of CRS with nasal polyps (CRSwNP) limits the generalizability of this study.
A longitudinal case study 34 focused on crenotherapy in 137 elderly patients (age ≥ 65 years) with moderate CRSsNP. They found that a 12-day course of inhalation of hyperthermal water rich in mineral salts improved olfactory function based on VAS questionnaires (4.07 ± 1.83 to 1.37 ± 1.33, P < .05) and NE score (5.2 ± 1.4 to 1.4 ± 1.3, P < .05) at 1 month. These benefits were sustained at 6 months, and no adverse events were reported. NaCl group did not influence VAS (4.0 ± 3.13 to 4.0 ± 3.3, P > .05) or NE (5.3 ± 2.9 to 5.2 ± 3.0, P > .05). The MCID of VAS for crenotherapy was −0.92. Crenotherapy was not compared to any of the conventional treatments for CRS. The age limitation of this study restricts applicability to the general population. The study was not blinded and did not outline the method for NE score. The NOS analysis showed a moderate risk of bias in selection and outcome, as the participants were from a single health center, and the recruitment process was not described.
Summary
Very limited evidence supports the use of naturopathy to treat CRS. There is some evidence to suggest that the use of Pimpinella anisum may improve symptomatology in CRSsNP. More research is warranted before a recommendation can be made.
Chinese Medicine
We found 3 studies examining the use of Traditional Chinese Medicine in CRS patients who met our inclusion criteria. A double-blind RCT of Chinese herbal medicine (CHM) containing Tsang-Erh-San with Houttuynia was conducted among 53 Taiwanese (TW) elderly patients (age ≥ 65 years) with CRSsNP. 35 At 8 weeks, there were no differences in effectiveness between the CHM and erythromycin control group in the Taiwanese version of the SNOT-20 (TWSNOT-20) score (47.8 ± 2.8 vs 48.2 ± 2.9, P = .838), NE (3.8 ± 0.3 vs 3.9 ± 0.2, P = .45), and bacterial culture rate (44.2% vs 48.1%, P = .834). CHM or erythromycin alone did not influence endoscopic score (mean decrease of 0.4, P = .165; decrease of 0.5, P = .077) and bacterial growth (decrease of 4%, P = .844; increase of 15%, P = .17). TWSNOT-20 scores, however, improved with CHM (mean decrease of 6.9, P = .029) or control (5.7, P = .041), without any significant adverse effects. The MCID of TWSNOT-20 for CHM was −1.75. The validity of these improvements is unclear due to the lack of placebo control.
A single-blind RCT 36 investigated the efficacy of acupuncture in 66 adult patients with CRSsNP. At 12 weeks, SF-36 improved with conventional medication consisting of xylometazoline and oral corticosteroid (mean increase of 15.25 ± 26), but not Chinese acupuncture (0.12 ± 23.8) or sham acupuncture at nonacupoints (0.69 ± 21.8); however, the P values were not listed. An improvement in CSS score was observed with conventional medication (32 ± 41), acupuncture (mean increase of 11 ± 23), and sham acupuncture at nonacupoints (8 ± 24). No statistical differences were found, except for SF-36 mental health (95% CI: 1–25, P = .03) between acupuncture (−3 ± 10) and conventional medication (10 ± 20). Due to missing data, the MCID of CSS and SF-36 were not determined. This study was not blinded, and the conventional medications used as a control had considerable intragroup variability. No baseline statistical analysis was performed. An analysis with the Cochrane template revealed a high risk of selection and attrition bias. There was no allocation concealment, and the outcome data were incomplete due to the withdrawal of 18 subjects.
In a prospective pilot study of 55 African American patients with CRS, 37 Chinese herbal tea containing Biyan pian led to considerable improvements in CSS total scores (mean increase of 11.1 from baseline of 79.4 ± 12.3, P = .02), but not SF-36 (a mean increase of 4.9 from 70.3 ± 21.4, P > .05) at 6 weeks after treatments. The MCID of CSS for Biyan pian herbal tea was 6.15. When assessed with NOS, this study was associated with a moderate risk of bias in selection as well as high risks in comparability and outcome. Only 27 participants complied with the regimen, and data were collected from self-reports.
Summary
The risk of biases in these studies is substantial, providing inconclusive and insufficient evidence to support the use of Chinese traditional treatment in CRS.
Homeopathy
One prospective cohort study 8 followed 134 adult CRS patients for 8 years to assess the efficacy of classical, individualized homeopathy delivered by 61 homeopathic physicians in different health centers. The authors focused on the use of homeopathic medicine in general rather than the specific treatments provided. Therapies consisting of various remedies and potencies were not standardized, but the most common homeopathic agents were Sepia, Pulsatilla, Lycopodium, Phosphorus, Carcinosium, Nux vomica, Sulphur, Natrium muriaticum, Staphisagria, and Silicea; these agents made up half of all the prescriptions. The study reported long-term improvements in physical component of SF-36 with greatest benefit at 2 years (mean increase of 5.53, 95% CI: 2.84–8.21, P < .001). The MCID of SF-36 for classical homeopathy was 6.35. The study population was heterogeneous with various comorbidities. During the study period, 43.3% of participants used other CRS treatments, including conventional medications, surgery, and acupuncture. This study was not controlled, and neither was it blinded nor randomized. In addition to a moderate risk of selection bias, high risks of bias in comparability and outcome were revealed in the NOS. The outcomes were measured solely based on self-reports, and follow-up was not satisfactorily reported.
Summary
Based on the limited data, there is poor evidence to support the use of Homeopathy in CRS.
Others—Liposome
One controlled pilot study 38 investigated the efficacy of liposomal nasal spray in 90 adult CRS patients by comparing it to the intranasal steroid spray. At 3 months, both liposome and beclomethasone spray independently achieved significant benefits (P < .05) in sinusitis symptoms (mean decrease of 2.73 ± 2.85 and 1.74 ± 3.15, respectively), rhinoscopy score (1.93 ± 1.64 and 1.96 ± 1.15), and SNOT-20 score (14.1 ± 12.7 and 13.9 ± 19.2). The MCID of SNOT-20 for liposomal nasal spray was −5.39. Observed differences between the groups did not reach statistical significance. Without any serious side effects, 16.7% in the liposome group and 43.3% in the control group reported headache, rhinitis, and bronchitis. The study was not blinded nor randomized, and no baseline statistical analysis was performed. Using the NOS, this study was found to have moderate risk of bias in selection, comparability, and outcome. The study did not control for various confounding factors nor provide any explanations for 27 participants who withdrew from the study.
Summary
Due to the high risk of bias with this study, there is poor evidence to support the use of liposomal nasal spray in treating CRS.
Discussion
Many different CAM therapies are available for CRS. In this study, numerous remedies and interventions were identified and classified into 4 groups—naturopathy, Chinese medicine, homeopathy, and “other.” Three noninferiority RCTs concluded that CAM may be as effective as the conventional treatments, while 4 observational studies reported benefits of CAM in managing CRS. However, our critical review suggests that there is very limited evidence to recommend these therapies in clinical settings due to inherent flaws and significant biases in the studies. Only 2 studies34,35 have used an established diagnostic criteria of CRS [American Academy of Otolaryngology (AAO) or European Position Paper on Rhinosinusitis and Nasal Polyps (EPOS)], illustrating another major flaw in the current literature. The level of evidence among the included studies ranged from IIb to IV.
From the 4 categories of CAM, the strongest evidence was found for naturopathy. Crenotherapy was shown to have some benefit in managing CRS in the elderly population. VAS score decreased by 2.7, which was above the MCID threshold of 0.92. Crenotherapy has been shown to boost the immune system via anti-inflammatory and mucolytic functions. 39 We also found that P. anisum may improve CRS symptoms. The observed improvement in SNOT-22 score of −20.8 was significantly higher than the reported MCID of −9.25,26 In vitro studies have demonstrated its anti-microbial, spasmolytic, and secretolytic activities.40,41
The use of traditional Chinese medicine in the treatment of CRS cannot be supported at this time, and Chinese acupuncture seems to be inferior to conventional medical therapy. 36 Acupuncture is theorized to work by influencing autonomic regulation, blood flow, and inflammatory mediators in the nasal mucosa; its effects may depend on the acupoints.42,43
Likewise, homeopathy cannot be recommended for CRS. Common remedies including Sepia, Pulsatilla, Lycopodium, and Carcinosinum 44 showed immunomodulatory and anti-microbial properties in other studies.45–47 The evidence in the published literature however does not demonstrate clear health benefits.
This review found inconclusive data to support the use of liposomal nasal spray in CRS. Previous studies have suggested that there is a benefit in allergic rhinitis.48,49 Liposomes are thought to help stabilize the nasal mucosa to protect against pathogens 50 and to reduce inflammation. 51
Some promising CAM therapies for CRS that have been excluded in this study are xylitol, manuka honey, and dead sea salt (DSS). In previous RCTs,52,53 sinonasal irrigation with 5% xylitol led to greater improvement in SNOT-20/22 compared to saline control. Manuka honey spray also led to an improved QoL score in a subset of adults with allergic fungal rhinosinusitis. 54 DSS spray reduced CRS symptoms as effectively as steroid–saline combination. 55 However, all of these studies included patients who underwent ESS less than 6 months from the onset of CAM therapy. This is a crucial confounding factor given that biological changes and clinical improvements associated with the surgery may not plateau until at least 6 months.56,57
This systematic review emphasizes the lack of quality studies that provide conclusive evidence for CAM in CRS. It also has some limitations. Due to the strict inclusion and exclusion criteria that attempted to minimize bias, the number of included studies was very low, and meta-analysis was not possible due to the heterogeneity of included studies. Non-English literatures were screened but were excluded if no English translation was available. It was felt that using validated outcome measures would limit the risk of bias in the included studies, but even still, the majority of the studies had a significant risk of bias. In addition to the previously mentioned flaws, of the included studies, only 2 of the 7 studies used standardized diagnostic criteria for CRS, which is a significant potential source of bias.
In summary, the evidence for CAM in CRS is extremely limited. Future CAM studies should focus on specific agents and use validated outcome measurements. To reduce selection, attrition, and performance bias, they may utilize placebo controls, randomization, and blinding. Without high-quality scientific evidence, it is difficult to draw definitive conclusions about the effectiveness of CAM in CRS. Further research is warranted, as CAM practices continue to increase and evolve among CRS patients.
Conclusion
There is very limited evidence to support the use of CAM in the treatment of CRS. Given its widespread use, more rigorous data are needed before it can be routinely recommended.
Footnotes
Acknowledgments
The authors would like to thank Ms. Jackie Phinney, Ms. Courtney Boudreau, and Mr. Donaldo Canales for providing guidance on incorporating appropriate literature search and analysis strategy in this review.
Author Contributions
C. J. C. and A. J. J. were responsible for study conception, design, data analysis, data collection, and drafting of the article.
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) received no financial support for the research, authorship, and/or publication of this article.
