Prevalence and predictors of depression and anxiety in patients with chronic rhinosinusitis: a systematic review and meta-analysis

Abstract

Background

Chronic rhinosinusitis (CRS) is a persistent inflammation of the sinuses. As a result of long-term discomfort, patients may experience symptoms of common mental disorders such as anxiety and depression. This may affect the quality of life and disease progression. However, there is still uncertainty about the extent of the problem.

Objective

This meta-analysis aimed to determine the prevalence of depression and anxiety symptoms in patients with CRS.

Search strategy

We searched PubMed, Embase, Web of Science, Cochrane Library, and CBM databases for relevant studies published before 15 July 2022 in patients with CRS with concomitant depression and anxiety symptoms.

Data collection and analysis

Two authors independently performed screening and quality assessment using validated tools. Extraction of data using predefined standardised data collection spreadsheets. Heterogeneity and inconsistency were checked using the I² statistic.

Results

The meta-analysis included 32 articles involving 56 933 patients. The prevalence of depression and anxiety symptoms was estimated at 24.7% (95% CI, 21.3% to 28. 1%) and 29.7% (95% CI, 19.3% to 40.2%). Subgroup analysis revealed the following: (1) CRS without nasal polyps (CRSsNP): 26.2% (95% CI, 21.9% to 30.5%), CRS with nasal polyps(CRSwNP): 20% (95% CI, 15.9% to 24%); (2) Female patients: 36. 1% (95% CI, 25.3% to 46.9%), male patients: 24.3% (95% CI, 12. 1% to 36.6%); and (3) The average age≤50 years patients: 29.8% (95% CI, 21.3% to 38.2%), the average age>50 years patients: 22. 1% (95% CI, 17.1% to 27%).

Conclusion

A significant proportion of people with CRS have symptoms of depression and anxiety, and early screening for depression and anxiety in people with CRS is critical. And, more attention needs to be given to females and patients with CRSsNP during screening.

PROSPERO registration number

CRD42022345959).

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The research samples were drawn from four continents, demonstrating racial and ethnic diversity.

• We explored sources of heterogeneity through subgroup analyses, sensitivity, and random-effects meta-regression.

• There was a high degree of heterogeneity in the findings, which could be attributed to the diversity of screening tools and research methods.

• The potential impact of comorbid symptoms on the prevalence of depressive symptoms in patients with chronic rhinosinusitis cannot yet be analysed due to insufficient data.

Introduction

Chronic rhinosinusitis (CRS) is one of the most common chronic diseases characterised by sinuses and nasal mucosal inflammation for at least 12 consecutive weeks. CRS is increasingly common in primary care and Ear, Nose and Throat (ENT) practice. The estimated prevalence of CRS varies from 5.5% in Brazil to 28% in Iran, 8% in China, 11% in South Korea, 4.8% to 12% in the USA, and 16% in the Netherlands, based on the symptom-only epidemiology criteria in the European Position Paper on Rhionsinusitis and Nasal Polyps (EPOS) guidelines.¹ As we all know, CRS is a considerable burden on society and healthcare systems in terms of direct or indirect healthcare costs. A previous study showed that direct medical expenditures for CRS treatment were increasing, ranging from $5560 per person to $5955 each year, resulting in total expenditure in 2011 from $60.6 billion to $64.9 billion.² Indirect costs include the reduced productivity of patients and their caregivers due to illness or treatment.

Depression and anxiety affect patients’ perceptions of the disease, leading to magnified symptoms and increased distress.³ It may explain why objective results on nasal endoscopy and sinus CT scores often correlate poorly with actual patient-reported sinus symptoms.⁴ For CRS patients, depression can affect the quality of life and the treatment effect. One plausible explanation for this could be that individuals with CRS may have poor treatment adherence due to mental health issues, and this may nourish a vicious cycle of mental health problems and physical health problems. Besides the rising treatment costs, the poor treatment effects often lead to the discontent of patients and frustration of medical staff. Considering the negative impact of depression on CRS patients, it is essential to screen for unbroken depression. Furthermore, it is important to comprehend the intricate relationships that exist between comorbidities and depression linked to CRS.

A recent meta-analysis has investigated the prevalence of symptoms of depression in CRS patients. In this study,⁵ the impacts of many factors, such as CRS subtype, the severity of illness, and comorbidities on the prevalence of depression symptoms are still unclear. Therefore, we plan to make a supplement for the above-mentioned study, investigating the prevalence of depression and anxiety symptoms in CRS. And to explore the factors that influence the prevalence of depressive symptoms in CRS by examining CRS subtypes, the objective severity of CRS, gender, age, and comorbidities.

Methods

This study was conducted according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement.

Eligibility criteria

(1) The types of studies were cross-sectional, case–control, and cohort studies (only baseline data were extracted from the study). (2) For patients with CRS who meet standard diagnostic criteria, there are no restrictions on the cause and type of CRS, race, age, or gender. (3) The prevalence of depression and anxiety symptoms was reported. Assessments for depression and anxiety symptoms include a previous doctor’s diagnosis or screening through a validation questionnaire.

Studies were excluded. (1) Studies of mixed groups with various sinus diseases, unless data from patients with CRS can be separated. (2) Duplicate published studies. However, we allow studies that analyse different aspects of the same sample (eg, different subtypes of the disease and gender), and we do not impose any language restrictions on eligible studies.

Literature search

Two independent authors searched PubMed, Embase, Web of Science, Cochrane Library, and CBM databases for relevant studies on depression and anxiety symptoms in patients with CRS from the establishment of these databases to 15 July 2022. (Detailed English search strategies for PubMed and Web of Science databases can be found in online supplemental appendix A. Search strategies for other databases are available on request.) We also screened references in relevant articles or reviews for eligible reports that were not collected in the above databases.

Quality assessment and risk of bias assessment

Two evaluators independently assessed the quality of the selected studies using The Joanna Briggs Institute Prevalence Critical Appraisal Tool. The 10 questions in the assessment tool include the following: (1) Was the sample representative of the target population? (2) Were participants recruited in an appropriate way? (3) Was the sample size adequate? (4) Were the study subjects and the setting described in detail? (5) Was the analysis conducted with sufficient coverage of the sample? (6) Were objective, standard criteria used for the measurement? (7) Was the condition measured reliably? (8) Was there appropriate statistical analysis? (9) Are all important confounding factors/subgroups identified and accounted for? (10) Were subpopulations identified using objective criteria? These questions can be answered with yes, no, unclear, or not applicable.⁶ We use a rating system of 0–5 for low quality and 6–10 for high quality. If sample size calculations were not reported, the sample size was assessed based on whether the study was able to detect a 25% prevalence of CRS depression within a 5% margin of error and 95% confidence level. Therefore, the minimum sample size was 288, and any differences were identified through discussion between the two evaluators.

Data extraction

Data were extracted by two investigators using a standardised, predefined data collection spreadsheet. The extracted contents include the following: (1) study characteristics, including first author name and year of publication; (2) sample characteristics, such as sample size, average age, and proportion of female participants; (3) classification of CRS, severity of disease, and comorbidities; (4) assessment methods for depression and anxiety symptoms; and (5) prevalence of depression and anxiety symptoms. Any disagreements that occurred were resolved through discussion or negotiation with the third investigator.

Statistical analysis

Meta-analysis was performed using Stata SE V.17.0 (Stata Corporation). Heterogeneity is defined by I², a statistic that describes the percentage of variation between studies due to heterogeneity rather than chance and does not inherently depend on the number of studies considered; 50% was used as a criterion to verify heterogeneity between studies. When there was no statistical heterogeneity between results (I²<50%, p>0.1), a fixed-effects model was used for meta-analysis. When there was statistical heterogeneity (I²>50%, p<0.1), a random-effects model was used for the meta-analysis. To explore potential causes of heterogeneity, we performed regression analyses by country, method of evaluating depression and anxiety symptoms, and sample size to determine whether covariates could account for heterogeneity between studies. We also performed subgroup analyses to examine the effects of age, gender, CRS severity, CRS type, and comorbidities on the prevalence of depression and anxiety symptoms in CRS.

Patient and public involvement

None.

Results

We identified a total of 3887 studies from five databases. After removing duplicates (n=1210), the titles and abstracts of 2677 studies were examined. After removing 2536 ineligible studies, 141 full-text articles were reviewed. Finally, 32 studies were included in our review. The literature screening process and results are shown in figure 1.

Figure 1.

Literature screening procedure and results: PRISMA diagram. PRISMA, Preferred Reporting Items for Systematic reviews and Meta-Analyses.

Study and participant characteristics

The characteristics of the literature are shown in table 1. Among the 32 studies (n=56 933), 32 reported the prevalence of depression symptoms and 12 reported the prevalence of anxiety symptoms. The 32 samples were from the USA (n=17),^{3 7–22} the UK (n=3),^23–25 Canada (n=3),^26–28 Korea (n=4),^29–32 China (n=3),^33–35 Egypt (n=1),³⁶ and Sweden (n=1).³⁷ Seven studies reported the prevalence of depression symptoms in men (n=551) and women (n=644); 11 studies reported the prevalence of depression symptoms in patients with CRSwNP (n=6135) and CRSsNP (n=11 697). In the studies involving CRS subtypes, 34. 4% of the patients had nasal polyps; eight studies (n=1183) reported the prevalence of depression symptoms in patients with comorbidities; nine studies (n=1434) reported Lund-Kennedy Endoscopy scores. To assess depression and anxiety symptoms, 22 studies used measures such as Patient Health Questionnaire-9 (PHQ-9), and seven studies used standardised diagnostic codes (International Classification of Diseases and Korean Classification of Diseases).

Table 1.

The characteristics of the literature (n=32)

Study	Country	Age (mean)	Depression evaluation	Anxiety evaluation	Gender (female %)	N	Depression symptoms prevalence, n (%)	Anxiety symptoms prevalence, n (%)
Davis et al 20057	USA	–	PHQ-9	BSI	45.3	95	25 (26.3)	17 (17. 9)
Smith et al 2005	USA	47.1	ICD-10	–	62.2	119	16 (13.4)	–
Brandsted and Sindwani 2007	USA	Depression, 45.92; not depression, 44.39	SF-8	–	59.7	106	27 (25.5)	–
Wasan et al 20079	USA	43.4	HADS	HADS	70	143	34 (23.8)	–
Mace et al 200810	USA	46.7	ICD-10	–	64.7	102	23 (22.5)	–
Smith et al 201012	USA	48.6	ICD-10	–	48.7	302	48 (15.9)	–
Soler and Smith 2010	USA	49.5	–	–	52	127	17 (13.4)	–
Litvack et al 2011	USA	Depression, 46.7; not depression, 50.5	PHQ-9	–	44.7	76	19 (25)	–
Sahlstrand-Johnson et al 2011	Sweden	–	HADS	HADS	41	171	23 (13. 5)	42 (24. 6)
Nanayakkara et al 2013	UK	43.5	HADS	HADS	1	57	6 (10. 5)	18 (31. 6)
Alt et al 2013	USA	PSQI ‘Poor’ sleep quality, 48.7; PSQI ‘Good’ sleep quality, 50.7	–	–	54.9	268	49 (18.3)	–
Jung et al 2014	Korea	45.56	BDI	–	72	25	10 (0. 4)	–
Tomoum et al 2015	Egypt	53.69	HADS	HADS	57.3	124	24 (19.4)	29 (23. 4)
Steele et al 201515	USA	Normal weight 47.0; overweight, 57.0; obese, 52.0	ICD-9	–	56	241	38 (15.8)	–
Schlosser et al 2016	Canada	57.64	BDI	–	51.5	42	13 (31)	–
Schlosser et al 2016	USA	50.5	PHQ-2	–	51.6	685	67 (9. 8)	–
Cox et al 2016	USA	48.59	PHQ-2	–	42.9	70	47 (67. 1)	–
Cox et al 2017	USA	49.12	PHQ-2	–	47	68	47 (69. 1)	–
Hanna et al 201727	Canada	–	PHQ-9	–	49.8	261	50 (19.2)	–
Mattos et al 2017	USA	52.70	–	–	56	109	20 (18.3)	–
Phillips et al 2017	USA	51.1	PHQ-2	–	48.9	94	28 (29.8)	–
Kim et al 2019	Korea	–	KCD	KCD	58.8	16 224	3772 (23.2)	6067 (37.4)
Ospina et al 2019	Canada	–	PHQ-9	–	43.7	142	22 (15. 5)	–
Choi et al 2020	Korea	–	ICD-10	ICD-10	1	21 707	2125 (9.8)	2134 (9.8)
Kim et al 2020	Korea	–	KCD		58.2	14 762	3444 (23.3)	5523 (37.4)
Ranford et al 2020	UK	45.38	PHQ-9	GAD-7	42	100	27 (27)	14 (14)
Vandelaar et al 2020	USA	52.4	PHQ-9	–	55.5	216	46 (21.3)	–
Vogt et al 2021	UK	CRSwNP, 48.82; CRSsNP, 47.37	HADS	–	56.5	95	47 (49.4)	37 (38.9)
Gill et al 2022	USA	57.2	FCI	FCI	65.3	69	31 (44.9)	39 (56.5)
Feng et al 2019	China	45.2	SDS	SAS	57.7	160	58 (36.3)	57 (35.6)
Feng et al 2020	China	47.3	SDS	SAS	36.9	78	28 (35.9)	25 (32)
Chen and Zhang 2021	China	41.88	SDS	SAS	47.4	95	20 (21)	–

–, not available; BDI, Beck Depression Inventory II; BSI, Brief Symptom Inventory; CRSsNP, chronic rhinosinusitis without nasal polyps; CRSwNP, chronic rhinosinusitis with nasal polyps; FCI, Functional Comorbidity Index; GAD-7, General Anxiety Disorder-7; HADS, Hospital Anxiety and Depression Scale; ICD-9, International Classification of Diseases, Ninth Revision; ICD-10, International Classification of Diseases, Tenth Revision; KCD, Korean Classification of Diseases; N, sample size; PHQ-2, Patient Health Questionnaire-2; PHQ-9, Patient Health Questionnaire-9; PSQI, Pittsburgh Sleep Quality Index; SAS, Self-rating Anxiety Scale; SDS, Self-rating Depression Scale; SF-8, Short Form-8.

Study quality

The quality assessment of the included studies is summarised in table (detailed results can be found in online supplemental appendix B). We did not find any low-quality studies. The main factors affecting the quality of these studies were inadequate sample sizes, samples that were not representative of the target population, and failure to identify and account for all important subgroup differences.

Prevalence of depression symptoms in CRS

In this meta-analysis, 32 studies reported the prevalence of depression symptoms in patients with CRS. The random-effect prevalence estimates are shown in figure 2. A crude estimate of the prevalence of depression symptoms in CRS was 24.7% (95% CI 21.3% to 28.1%).

Figure 2.

Prevalence of depression symptoms in CRS (n=32). Weights and between-subgroup heterogeneity test are from random-effects model. CRS, chronic rhinosinusitis.

Prevalence of anxiety symptoms in CRS

In this meta-analysis, 12 studies reported the prevalence of anxiety symptoms in patients with CRS. The random-effect prevalence estimates are shown in figure 3. The prevalence of anxiety symptoms in CRS was roughly 29.7% (95% CI 19.3% to 40.2%).

Figure 3.

Prevalence of anxiety symptoms in CRS (n=12). Weights are from random-effects model. CRS, chronic rhinosinusitis.

Subgroup analysis

As shown in table 2, subgroup analysis showed that the prevalence of depression symptoms was 26.2% (95% CI, 0.219 to 0.305) in CRSsNP patients and 20% (95% CI, 15.9% to 24%) in CRSwNP patients. Female patients: 36.1% (95% CI, 0.253 to 0.469), male patients: 24.3% (95% CI, 12.1% to 36.6%); the average age≤50 years patients: 29.8% (95% CI, 21.3% to 38.2%), the average age>50 years patients: 22. 1% (95% CI, 17.1% to 27%); patients with a Lund-Kennedy endoscopy score≤6: 46. 1% (95% CI, 28.5 to 63.8%), patients with a Lund-Kennedy endoscopy score>6: 16.5% (95% CI, 10.5% to 22.5%); CRS patients with concomitant asthma: 32% (95% CI, 15% to 49. 1%); CRS patients with allergies: 35.4% (95% CI, 17% to 53.8%); CRS patients with a history of previous sinus surgery: 48. 1% (95% CI, 20.7% to 75.6%). (Detailed results can be found in online supplemental appendix C, Fig. C.1–5.)

Table 2.

Subgroup analysis of the prevalence of depression symptom

	Meta-analysis		Heterogeneity
Subgroup	No. of studies	Prevalence (95% CI)	I², %	P value
CRS subtype
CRSwNP	11	0.2 (0.159 to 0.24)	64.3	0.002
CRSsNP	11	0.262 (0.219 to 0.305)	73.5	<0.001
Age
≤50	13	0.298 (0.213 to 0.382)	93.7	<0.001
>50 gender	8	0.221 (0.171 to 0.270)	79.9	<0.001
Female	7	0.361 (0.253 to 0.469)	86	<0.001
Male L-K score	7	0.243 (0.121 to 0.366)	93.7	<0.001
>6	4	0.165 (0.105 to 0.225)	60.3	0.056
≤6 comorbidity	5	0.461 (0.285 to 0.6380)	96.2	<0.001
Asthma	4	0.320 (0.150 to 0.491)	90.6	<0.001
Allergy	4	0.354 (0.170 to 0.538)	87.4	<0.001
Prior sinus surgery	3	0.481 (0.207 to 0.756)	93.9	<0.001

≤6, Lund-Kennedy endoscopy scores ≤6 points; >6, Lund-Kennedy endoscopy scores＞6 points; ≤50, the average age of CRS patients ≤50 years; >50, the average age of CRS patients >50 years; Allergy, CRS patients with allergy; Asthma, CRS patients with asthma; CRS, chronic rhinosinusitis; CRSsNP, chronic rhinosinusitis without nasal polyps; CRSwNP, chronic rhinosinusitis with nasal polyps; CRSwNP, chronic rhinosinusitis with nasal polyps; L-K score, Lund-Kennedy endoscopic score; Prior sinus surgery, CRS patients with a history of previous sinus surgery.

Meta-regression analyses

Meta-regression analysis showed that sample size (coefficient = −4. 81E-06, t =−1.02, r=0.318), country (coefficient=−0.0071089, t=−0.61, r=0.544), and age (coefficient=−0.0007087, t = −0.09, r=0.932) had no significant effect on heterogeneity. However, the method of evaluating depression symptoms (coefficient=0.0222019, t=2.52, r=0.018) was a source of heterogeneity.

Publication bias and sensitivity analysis

The funnel plot and Egger test did not show significant publication bias in depression (Egger test: t=2. 03, r=0.051) or anxiety (t=0. 88, r=0.398). Sensitivity analyses did not find that outliers significantly altered the pooled prevalence of depression and anxiety symptoms in CRS (Online supplemental appendix D).

Discussion

We investigated the prevalence of depression and anxiety symptoms in CRS patients. The results showed that the prevalence of depression symptoms was about 24.7%, and anxiety symptoms was about 29.7%, broadly consistent with the previous research results.⁵ Our study revealed that the prevalence of depression symptoms is higher in patients with CRSsNP than in those with CRSwNP. CRSwNP patients may experience more severe symptoms of nasal congestion and anosmia, while CRSsNP patients may experience more severe pain.^{26 38} It is not clear whether certain major symptoms play a more critical role in the development of depression symptoms. According to the imaging and endoscopic grading systems, nasal polyps undoubtedly cause more severe objective symptoms. However, for many diseases, including CRS, objective symptoms do not reflect the full burden of disease experienced by the individual patient.³⁹ Rehab⁴⁰ discovered that individuals with CRSsNP experience a greater burden of disease and have a worse general Health-Related Quality of Life (HRQoL) compared with those with CRSwNP. This may explain the higher prevalence of depression symptoms in CRSsNP patients than in CRSwNP patients.

The prevalence of depression symptoms is higher in CRS patients with a Lund-Kennedy endoscopic score of ≤6 compared with those with a score of >6. This situation seems to be explained by the lack of correlation between patients’ subjective symptoms and objective measures of CRS severity.^{36 41} The findings suggest that patient-reported outcome indicators, particularly in patients with comorbid depression and anxiety, such as SNOT-22 CRS symptom scores, correlate less well with objective indicators such as CT imaging or nasal endoscopy scores.²⁴ Psychological disorders and environmental factors can influence symptom perception in CRS. Patients with symptoms of anxiety or depression may tend to amplify their symptoms. Therefore, clinicians should proceed with caution when assessing whether patients with CRS will develop depression. Additionally, further exploration of the relationship between subjective symptoms and the prevalence of depression symptoms is necessary.

Among the enrolled patients, younger participants had a higher prevalence of depression symptoms. This is consistent with Hanna’s²⁷ findings, which showed that younger people were more likely to have worse SNOT-22 scores and higher rates of depression and pain, and higher SNOT-22 scores were highly and independently associated with depression and pain scores. This may be attributed to that young people have to deal with more complex social relationships and bear more financial pressure. It should be noted that this finding is not consistent throughout the literature. In some of the literature, older age is a high-risk factor for depression in patients with CRS.⁴²

The prevalence of depressive symptoms is higher in females than in males. A large body of literature suggests that inflammation acts through well-established mechanistic pathways to depression in some individuals.^{43 44} In a study, females and males had similar IL-6 and TNF-α responses following low-dose endotoxin administration. However, females reported a greater increase in depression than males. Compared with males, females may be more susceptible to emotional and behavioural changes caused by inflammation and may be more prone to emotional symptoms in response to the inflammation.⁴⁵ These findings suggest that female has a susceptibility to inflammation, and emotional effects might lead to gender differences in depression.

An investigation of the prevalence of depression and anxiety symptoms among paediatric CRS patients was not addressed in our study. It is also known that CRS has an impact on children’s health. Compared with adult CRS, paediatric CRS presents unique characteristics. Therefore, special attention should be given to the physical and psychological care of children with CRS.^{46 47} CRS has a significant impact on HRQoL. Studies have shown that children with chronic rhinitis and sinusitis have a significantly lower HRQoL compared with normal children, affecting their emotional state and mental health in addition to their general physical condition.⁴⁸ In addition, current research examining the association between depression and inflammation has focused on adults, but a study by Colasanto et al ⁴⁹ found that increased levels of inflammatory markers were associated with future depression in children and adolescents. Therefore, it is necessary to focus on depression and anxiety states in children with CRS in future studies.

Although we sought to investigate the prevalence of anxiety and depression symptoms in CRS patients, there exists relatively little data on anxiety symptoms in CRS patients. Anxiety and depression often coexist in the general population, with up to 67% of people with major depression meeting the criteria for anxiety.⁵⁰ Future research needs to pay more attention to anxiety disorders. Most studies have only reported the number of depression cases in CRS patients. However, they have not provided sufficient data for relative risk analysis of accompanying illnesses such as nasal polyps, asthma, and allergies. Accordingly, it cannot confirm the potential impact of these factors on the prevalence of depression symptoms in CRS. Most studies used cross-sectional data and baseline data from cohort studies, limiting the analysis of covariates that may be associated with the prevalence of depression and anxiety symptoms in CRS patients. The included study samples also included patients with CRS who had failed drug therapy or were awaiting surgery. Many complex conditions may contribute to overestimating the prevalence.

The data in this study are from seven countries, most of which are from high-income countries. So, these findings may not apply to low-income and middle-income countries. This study is highly heterogeneous due to the diverse screening tools, different research methods, and insufficient sample size in most studies. Random-effects meta-analysis could be used to explain the heterogeneity among studies.⁵¹ However, this model gives higher weight to smaller studies, limiting the validity of the obtained results.

Conclusions

Our findings indicate that approximately 25% of CRS patients suffer from depression symptoms, and close to 30% of CRS patients experience anxiety symptoms. Given this study’s marked heterogeneity, the conclusion might still be tentative. We also need more research data to further investigate the relative risk factors of depression and anxiety symptoms in CRS patients. In essence, it could be very meaningful to identify CRS patients with high-risk depression and anxiety early in clinical practice and provide the precautionary measures to deal with CRS-related depression and anxiety symptoms.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Not applicable.