Abstract
Objectives:
To perform a systematic review of otolaryngologic presentation rates to emergency department settings before and after lockdown due to the COVID-19 pandemic.
Sources:
PubMed, Scopus, and CINAHL.
Methods:
A systematic search was conducted following PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-analyses) for studies describing otolaryngologic presentations to emergency department and rapid access clinic settings both in the before-lockdown and after-lockdown periods. The start of after-lockdown period varied based on initiation of lockdown, ranging from March 1st to June 1st of 2020 across general emergency department studies.
Results:
A total of 14 studies were included in this review. About 10 were general emergency departments, 3 were specifically pediatric emergency departments, and 1 study focused on the geriatric population (>65 years). A total of 13 790 patients were included, with 9446 in the before-lockdown period (68.5%) and 4344 in the after-lockdown period (31.5%). Meta-analysis of proportions for otolaryngologic presentations across general emergency departments was performed. Comparison of weighted proportions found significant differences between before-lockdown and after-lockdown presentation rates for infectious etiologies, tonsillitis specifically, foreign bodies, non-infectious airway issues, and epistaxis among these studies.
Conclusions:
The increased proportions of various non-infectious presentations (eg, epistaxis, foreign bodies, and airway issues) following lockdown might be associated with proportional decreases in infectious pathologies, given decreased social contact to prevent SARS-CoV-2 transmission. Overall, it is important for otolaryngologists to recognize what presentations might more commonly be seen and require evaluation and potential intervention in light of a global pandemic.
Introduction
In the first few months of 2020, the emergence of the COVID-19 pandemic dramatically shifted the landscape of medical care throughout the world. As the disease rapidly transmitted from country to country, it placed an unprecedented burden on healthcare systems which required the cessation of many non-emergent services. The United States Centers for Medicare and Medicaid Services (CMS), along with many other national health agencies throughout the world, issued recommendations urging healthcare facilities and physicians to delay “nonessential surgeries and other procedures.” 1 Most countries also issued stay-at-home, or lockdown, orders for various periods of time throughout the pandemic, and urged patients not to present to the emergency department (ED) if avoidable. Surgical specialties with high rates of elective or non-urgent cases, such as otolaryngology, experienced dramatic declines in their patient census. 2 Despite the decrease in otolaryngology patient populations, there were still patients throughout the pandemic who presented for evaluation by an otolaryngologist. In this systematic review and meta-analysis, we sought to evaluate the impact of the COVID-19 pandemic on otolaryngologic volume in ED settings during the after-lockdown period (ALP) as compared with the before-lockdown period (BLP).
Methods
Search Criteria
The study was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 3 This study was not prospectively registered. No review protocol was established before study initiation. A comprehensive search was performed in the following 3 databases: PubMed (U.S. National Library of Medicine, National Institutes of Health), Scopus (Elsevier), and CINAHL (EBSCO). The search strategy used both subject headings (eg, MeSH in PubMed) and keywords for the following concepts and/or keywords: otolaryngology, ear nose and throat, volume, and COVID-19. The search strategy was modified for the other databases, utilizing similar keywords, and substituting MeSH terms with appropriate subject headings as available. The databases were searched from January 1st, 2020 to June 1st, 2022 to exclude non-SARS-CoV-2-related studies. The reference lists of relevant articles and citing articles were manually searched to both identify additional articles and confirm the search strategy. All references were uploaded to the review management software Covidence (Veritas Health Innovation Ltd, Melbourne, Australia) and screened for relevance.
Selection Criteria
Studies that described otolaryngology volume in EDs and rapid access clinics (RACs) before and during the pandemic were included in the analysis. All study designs were considered for inclusion. Specifically, studies must have included the number of cases of otolaryngology-related presentations both in the BLP and ALP. Studies must have defined the lockdown period and reported the periods in which patients were included. These periods varied based on the start of lockdown in given countries and specific hospital systems, with ALP start dates ranging across studies from March 1st to June 1st of 2020 across general ED studies. Patients of all ages were included in this study to capture a full spectrum of otolaryngologic patients seen during the pandemic; however, studies were stratified by patient population for meta-analysis. Studies must have described multiple otolaryngologic conditions that presented for evaluation. Exclusion criteria included non-English studies, nonhuman studies, and non-SARS-CoV-2 related studies. Studies without a pre-pandemic comparison or an overall otolaryngology patient population as comparator were excluded. Studies without otolaryngology-specific data (eg, aggregated with other surgical subspecialties) were excluded. Studies reporting data from outpatient otolaryngology clinic visits or elective surgical cases were excluded.
Titles and abstracts were first independently screened by 2 reviewers (C.C.M. and S.R.S.) to select for records that met inclusion criteria. Disagreements were resolved in a discussion with a third reviewer (S.A.N.). Next, the full texts were independently assessed by both C.C.M. and S.R.S. to identify which reports satisfied all the inclusion and exclusion criteria to be included in the final analysis. S.A.N. resolved any conflicts. The level of evidence for each selected study was evaluated with the 2011 criteria of the Oxford Center for Evidence-Based Medicine. 4 The JBI Critical Appraisal Checklist for Prevalence Studies was utilized to assess the quality of included studies as outlined in methodological guidance for systematic reviews on observational epidemiological studies. 5 Two authors (C.C.M. and S.R.S.) reviewed all included studies independently and rated each JBI checklist item as “yes,” “no,” “unclear,” or “not applicable.” All disagreements were resolved by discussion with a third author (S.A.N.). A score was determined for each study by adding each checklist item marked “yes,” with total scores ranging from 0 if no criteria were met to 9 if all criteria were met. Studies were rated as low risk or good quality when scoring 5 and above and were included in the analysis.6,7
Data Extraction
Two reviewers (C.C.M. and S.R.S.) independently extracted the data and compared for accuracy. Author, year of publication, demographics, including age and gender, were recorded. Number of otolaryngology presentations by diagnosis were extracted.
Statistical Analysis
Meta-analysis of proportions was performed using MedCalc 20.106 (MedCalc Software, Ostend, Belgium). Proportions (otolaryngologic presentations) were expressed as percentages with their 95% confidence intervals. If there was high heterogeneity (I 2 > 50%), the random-effects model was used; if low heterogeneity, then a fixed-effects model has been considered allowable. In addition, a comparison of weighted proportions was done to compare otolaryngologic disease presentations between BLP and ALP. Finally, the Sterne and Egger tests were performed for further assessment of risk of publication bias.8,9 Potential publication bias was evaluated by visual inspection of the funnel plot and Egger’s regression test, which statistically examines the asymmetry of the funnel plot. A P value of <.05 was considered to indicate a significant difference for all statistical tests.
Results
Search Results and Study Characteristics
The literature search yielded 1073 unique articles. Title and abstract screening excluded 822 records. A full text review of the remaining studies excluded 237 studies, leaving 14 studies for inclusion in the final data extraction and analysis.10-23 All 14 studies were observational studies reporting prevalence in both the BLP and ALP. Ten studies were inclusive of all patient populations in general EDs. Three studies included data from pediatric EDs specifically. One study from a general ED reported only data from geriatric patients (defined by age >65 years) and was therefore not included in the overall meta-analysis but proportions from this study are discussed separately. Figure 1 outlines the full selection process. Descriptive features of the included observational studies are summarized in Tables 1 and 2. Articles selected for inclusion were all level 4 studies based on the Oxford Level of Evidence and were published between 2020 and 2022 from 9 different countries. Critical appraisal of studies indicated an acceptably low risk or good quality with all 14 studies scoring between 6 and 8 on JBI assessment scores (Table 3). A funnel plot with Egger et al’s 8 test (0.4665, 95% CI −11.39-12.32, P = .931) demonstrated relatively low asymmetry and all studies were within the funnel, suggesting low publication bias (Figure 2).

PRISMA flow chart of study selection.
Characteristics of the 10 Studies Reporting Data From General EDs.
Abbreviations: ALP, after-lockdown period; BLP, before-lockdown period; NR, not reported; OLE, Oxford level of evidence.
Characteristics of the 3 Studies Reporting Pediatric ED Data and 1 Study Reporting Data From Geriatric Patients.
Abbreviations: ALP, after-lockdown period; NR, not reported; OLE, Oxford level of evidence.
Table Demonstrating Risk of Bias Assessments Over the Measures Included in the JBI Prevalence Checklist.
Abbreviations: JBI, Joanna Briggs Institute; N, no; U, unclear; Y, yes.
Was the sample frame appropriate to address the target population?
Were study participants sampled in an appropriate way?
Was the sample size adequate?
Were the study subjects and the setting described in detail?
Was the data analysis conducted with sufficient coverage of the identified sample?
Were valid methods used for the identification of the condition?
Was the condition measured in a standard, reliable way for all participants?
Was there appropriate statistical analysis?
Was the response rate adequate, and if not, was the low response rate managed appropriately?

Funnel plot evaluating for publication bias.
Overall, the included studies had 13 790 patients total, with 9446 in the BLP (68.5%) and 4344 in the ALP (31.5%). Table 4 shows the results of meta-analyses of proportions of various diagnoses in the BLP, ALP, and a comparison of these weighted proportions. Infections in the BLP were seen in 29% of patients in the included studies compared with 23.2% of patients in the ALP. Comparison between BLP and ALP demonstrated a statistically significant (P < .0001) difference of 5.9% between the proportion of infections in the 2 populations. Additionally, tonsillitis showed a difference in proportions of 4.9% (P = .0009) less patients in the ALP. While the overall proportion of patients presenting with infectious pathology, along with tonsillitis, demonstrated significant difference, there were less remarkable differences for other specific infectious etiologies. Uncomplicated upper respiratory infections (URIs) accounted for 17% of presentations in the BLP versus 15.8% in the ALP. For various upper airway abscesses (peritonsillar, parapharyngeal, retropharyngeal), there were proportionally fewer in the ALP compared with BLP. The ALP decrease in these abscesses, however, was only 1.52% with a P-value of .045, demonstrating marginal statistical significance. The proportion of BLP presentations of epistaxis was 12.51%, while the proportion of ALP presentations of epistaxis was 16.93%. The difference in presentations of epistaxis between BLP and ALP was significant (Table 4). Other parameters investigated with meta-analysis are also summarized in Table 4, with all bleeding, otologic pathology, foreign body presentations, and non-infectious airway presentations all demonstrating significant difference in proportions between ALP and BLP. The change in proportions of presentations of these diagnoses between BLP and ALP is demonstrated in Figure 3.
Results of Meta-Analysis of Proportions for Various Otolaryngology Diagnoses and Categories of Disease Across 10 Included Studies.
Abbreviations: ALP, after-lockdown period; BLP, before-lockdown period; CI, confidence interval; LL, lower limit; UL, upper limit; URI, upper respiratory infection.

Forest plot demonstrating the change in proportion for various otolaryngologic presentations between the BLP and ALP.
Discussion
This systematic review and meta-analysis analyzed the proportion of patients presenting with various otolaryngologic complaints both in the BLP and ALP out of the total number of otolaryngology presentations. Specific otolaryngologic diagnoses as well as categories of disease were analyzed for changes in rates of presentation in the BLP compared with ALP. Overall rates for ED visits decreased significantly as lockdown was implemented in countries. In the United States, the CDC reported a 42% decline in ED volume after a state of national emergency was initially declared. 24 These numbers fluctuated with transmission rates of the virus, which likely contributes to variability seen between studies. Given the various measures employed during the ALP of the COVID-19 pandemic (social isolation, quarantine, stay-at-home orders) and dramatically decreased interpersonal contact overall, the authors suspected that rates of infectious conditions and infections seen during the pandemic might be decreased in the ALP across studies. 25
As hypothesized, all infectious etiologies meta-analyzed were found to have decreased ALP presentation rates with variable overall decreases and statistical significance. Decreases in the proportion of total infections and tonsillitis, 5.45% and 4.90% respectively, were statistically significant. Given stay-at-home orders and other social distancing measures implemented to prevent disease transmission, it is not surprising that the proportion of otolaryngologic infectious presentations were significantly decreased in the ALP. Many of the prevention methods implemented to mitigate the spread of COVID-19 are also effective in preventing transmission of common otolaryngologic infections. It is important to consider that the proportional decreases in infectious pathology, while statistically significant for some diagnoses, could also be explained by other factors. Recommendations from public health officials and hospital policy changes might have led to more stringent triaging of patients for mild to moderate presentations of otolaryngologic disease, both self-enforced among patients and for admission into EDs and hospitals. Although this study cannot directly assess the impact lockdown and social distancing measures had on overall rates of otolaryngologic infections, the results of this study demonstrate both absolute and proportional decreases in infectious presentations to ED settings during the ALP.
While the overall volume for all diagnoses decreased in the ALP, the proportion of specific otolaryngologic presentations was increased relative to overall otolaryngologic volume during this time. While the proportional rates of many infectious diagnoses decreased in the ALP, there were several other diagnoses with proportionally increased presentation rates during this time. The increased proportion of various non-infectious pathology could very well be explained by proportional decreases across infectious categories, however, there are also plausible explanations for many of these proportional increases. Epistaxis is an incredibly common otolaryngologic presentation to EDs, accounting for approximately 1 in 200 ED visits in typical years within the United States. 26 The significant increase in the proportion of patients with epistaxis may also be explained by the primary treatment modality for patients suffering with moderate or severe cases of COVID-19: oxygen therapy. One study looking at epistaxis rates among patients hospitalized for COVID-19 and receiving oxygen therapy noted increased frequency for epistaxis, and most commonly among patients who already had non-allergic rhinitis or relatively dry nasal epithelium. 27 Finally, there is also the potential that COVID-19 could be directly associated with epistaxis, given a high predilection for nasal mucosal cells and the inflammation associated with the virus. 28 This prospective study of 40 patients discussed that the highest prevalence for COVID-19 was found in nasal mucosal cells and olfactory epithelium and correlated these findings with increased rates of epistaxis among patients hospitalized for COVID-19.
Foreign body presentations also had a significant increase in proportion for ALP presentations. While there is similar potential for a relative increase given the drop in infectious presentations, there are also several etiology-specific possibilities. A retrospective study at a single institution which looked specifically for foreign body ingestion during the pandemic found a more than 2.5-fold increase in these referrals, particularly so for button batteries and magnet ingestion. 29 The authors of this study theorized that disruption to home environments (eg, increase in work-from-home opportunities, school closures) may have altered parental surveillance and led to an increased risk for pediatric ingestion. This was further suggested in other studies documenting significant increases in foreign-body ingestions, which mentioned that most ingestions in children occur in the home environment and increased time spent at home by children was seen during the pandemic.30,31 Although some included studies reported foreign bodies by anatomical location, there was not enough data reported to analyze presentation rates by location.
There were several limitations to this study that may impact the generalizability of the data. First, there was significant heterogeneity among studies likely indicative of differences in study design and methodology. There was not complete uniformity in diagnoses selected for reporting, time frames in which patients were included, nor the clinical setting in which patients were encountered (ED, RAC, inpatient settings). Studies with overall otolaryngology complaint comparator populations were utilized to try to minimize some of this variability, yet there is still inconsistency with the specific data reported. The risk of bias was low overall; however, it was generally unclear whether these observational studies had adequate sample sizes to reflect the population under investigation. While there was significant heterogeneity among the included studies, there was low concern for publication bias given the funnel plot and Egger’s test findings. Future studies with larger patient populations across multiple institutions that assess the relationship between timing of lockdown and rates of SARS-CoV-2 infection with presentation rates of otolaryngologic diagnoses would help reduce some of the potential bias and variability seen among included studies.
While the COVID-19 pandemic caused significant decreases across all fields within medicine, surgical specialties with high rates of elective or non-urgent disease, such as otolaryngology, experienced significant declines in patient volume. Among studies reporting otolaryngologic presentations to EDs and RACs in both the BLP and ALP, there were significant decreases in the proportions of overall otolaryngology-related infections and significant increases in the proportions of various pathologies. It is important for otolaryngologists to recognize the impact quarantine, social isolation, and decreased overall interpersonal contact has had and may continue to have on otolaryngologic presentations in ED settings. Implementation of telehealth services, pre-appointment screening, more stringent triaging of patients, and outright cessation of elective visits and procedures were all utilized following initial outbreak and lockdown during the COVID-19 pandemic to help mitigate transmission of the virus and overburdening of healthcare systems.32-35 Such measures were particularly important in otolaryngology, given the prevalence of aerosol-generating procedures in the specialty. 36 Despite these measures leading to dramatic reductions in patient censuses, presentations of various otolaryngologic manifestations were still seen in ED settings. The recognition of presentations that may still be expected amid social isolation and stringent triaging can help otolaryngologists to be better prepared in effectively diagnosing and managing patients they may encounter as transmission rates of SARS-CoV-2 and subsequent variants continue to change.
Footnotes
Author Contributions
Christopher C. Munhall: Study design, data collection and analysis, drafting and revision of manuscript. Sunny Shah: Data collection and analysis, drafting and revision of manuscript. Shaun A. Nguyen: Data analysis, drafting and revision of manuscript. Ted A. Meyer: Drafting and revision of manuscript. Rodney J. Schlosser: Manuscript revision. David R. White: Drafting and revision of manuscript.
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.
