Key Points
Question
Does stellate ganglion block (SGB) improve COVID-19−induced smell distortion (parosmia) symptoms and disease-specific quality of life compared to a placebo?
Findings
This randomized, double-blinded, placebo-controlled clinical trial including 57 participants, mepivacaine vs saline injection was administered at the stellate ganglion in patients with self-reported parosmia at least 6 months after COVID-19 infection. At 1 month and 3 months postinjection, there was no clinically meaningful difference between groups in parosmia-specific patient-reported outcome measures.
Meaning
SGB did not provide benefit beyond placebo effect and should not be recommended as treatment for COVID-19−induced parosmia.
Abstract
Importance
Smell distortion (parosmia) is a symptom of post–COVID-19 condition that persists and has immeasurable detriments on quality of life. To date, a standard criterion for treatment of this bothersome condition has not been identified.
Objective
To determine whether stellate ganglion block (SGB) is effective and safe in improving olfactory dysfunction in patients with persistent COVID-19−induced parosmia.
Design, Setting, and Participants
This was a randomized, double-blinded, placebo-controlled clinical trial conducted from October 2023 to September 2024 at a single center study (the Washington University in St Louis/Barnes Jewish Hospital). A volunteer sample of 192 individuals were screened; 57 were enrolled after meeting eligibility criteria (age 18-70 years with self-reported parosmia of ≥6 months since COVID-19 infection, and a screening score of ≥40 on the Parosmia Olfactory Dysfunction Outcomes Rating [DisODOR] scale). Of 135 excluded, the most common reasons were prior SGB (n = 42) and parosmia resolution or non−COVID-19−induced parosmia (n = 28). Data were analyzed from September to October 2024.
Intervention
Ultrasonography-guided injection of 6 to 8 mL of active mepivacaine, 1%, or saline, 0.9%, at the right or left (randomized 1:1) stellate ganglion was performed by a board-certified anesthesiologist and pain medicine specialist.
Main Outcome and Measure
Between-subject difference of 25% in proportion of responders (defined by a 15-point decrease in DisODOR score) from baseline.
Results
The study evaluated 48 participants, 32 randomized to SGB (median [range] age, 45 [19-64] y; 25 [81%] female), and 16 to placebo (median [SD] age, 45 [26-64] y; 13 [81%] female). Time since COVID-19 infection was similar between groups (SGB, 35.3 vs placebo, 30.6 months; MD = −3.1 months; 95% CI, −10.9 to 3.7). Three-month response rate was 43% (n = 13) for SGB and 38% (n = 6) for placebo (difference, −5%; 95% CI, −32% to 33%). There was no between-group difference in clinical global impression of improvement.
Conclusions and Relevance
This randomized clinical trial found that SGB is not superior to placebo in treating COVID-19−induced parosmia, and thus, should not be recommended as treatment.
Trial Registration
ClinicalTrials.gov Identifier: NCT06253806
This randomized clinical trial investigates whether stellate ganglion block is effective and safe in improving olfactory dysfunction in patients with persistent COVID-19−induced parosmia.
Introduction
Olfactory dysfunction (OD), which includes hyposmia (smell loss) and parosmia (smell distortion), is common after COVID-19 infection.1 More specifically, persistent parosmia may affect nearly 50% of patients with symptoms of post–COVID-19 condition.2,3 A recent study4 suggests that 55% of patients who experience parosmia during a COVID-19 infection still have parosmia 1 year after the infection. Parosmia has been shown to have substantial psychosocial and health impacts, with reduced enjoyment of food, weight changes, nutritional insufficiency, social and professional disturbances, and decreased emotional well-being.5,6 Safety hazards have also been reported due to inability to detect burning smells or other environmental dangers.7
Despite the burden of COVID-19−induced parosmia, no standard of care treatment has been established. Routinely recommended therapies for postinfectious OD include olfactory training and topical steroids. However, evidence mainly supports these interventions for hyposmia rather than parosmia, with some studies noting parosmia as an adverse event in response to olfactory training.8 Other therapies have been evaluated for postinfectious OD via randomized clinical trial with limited success, including vitamin A, theophylline, and sodium citrate, but were also designed for the treatment of hyposmia rather than parosmia.9 Given the psychosocial impact of parosmia, there is a critical need for the development and validation of a treatment option specifically for this population.
The stellate ganglion block (SGB) is proposed to inhibit the sympathetic neural connections within the head, neck, and upper extremity, improve regional blood flow, reduce adrenal hormone concentration, and even, reestablish circadian rhythms through modulation of melatonin.10,11,12,13 SGB has been used to treat other diseases involving neuroimmunologic feedback loops, such as chronic regional pain condition, posttraumatic stress disorder, and migraines.14 It is hypothesized that symptoms of post–COVID-19 condition, including parosmia, may be due to autonomic nervous system dysfunction and immune-driven inflammatory processes creating a positive feedback loop that results in sympathetic hyperactivation.14 Thus, SGB has been proposed to improve olfactory disorders through temporary blockage of the sympathetic chain, which allows the inflammatory and autonomic nervous systems to reset and resolve the dysautonomia believed to be contributing to smell distortion.14,15,16,17
Our previous work15 suggests that SBG may benefit COVID-19−induced OD, including parosmia. Other retrospective studies also suggest that SGB may improve subjective COVID-19−induced parosmia.16,17,18 However, no randomized clinical trial has yet established the efficacy of SGB for this condition. Thus, we sought to assess the efficacy of SGB in the treatment of COVID-19−induced parosmia. We also report further on the safety of SGB and whether psychosocial and mental health metrics are associated with response to SGB in patients with parosmia.
Methods
The study protocol was approved by Washington University Institutional Review Board (HRPO# 202304090; available in Supplement 1). All potential participants provided verbal informed consent during recruitment; written informed consent was obtained at the initial study visit before commencement of research activities. We followed the Washington University reporting guidelines.
Trial Design
This was a single-center, double-blinded, placebo-controlled randomized clinical trial conducted at Washington University in St Louis. Participants in the active group received unilateral SGB and those in the placebo group received unilateral saline injection. Unilateral injections were given because our previous pilot study did not demonstrate any substantial changes in olfaction after participants received a second contralateral injection.15
Participants
Participants were included if they met the following criteria: age 18 to 70 years; diagnosed with COVID-19 at 6 months or more before enrollment and had self-reported parosmia; able to read, write, and understand English; and a screening score of 40 or greater on the Parosmia/Distortion Olfactory Dysfunction Outcomes Rating (DisODOR) scale,19 and 25 or greater at the initial injection visit. Demographic information collected, including race and ethnicity, were self-reported.
Participants were excluded if they had a history of smell loss or change before COVID-19 infection; a history of medical conditions known to affect olfactory or nerve function (eg, chronic rhinosinusitis, prior skull base or sinus surgery, neurodegenerative disease); current use of olfactory dysfunction therapy, including topical steroids and olfactory retraining; an inability to tolerate injection to the neck; a contraindication to SGB (eg, unilateral vocal cord paralysis, severe pulmonary or cardiac disease, glaucoma); current use of blood thinners or antiplatelet agents; allergy to local anesthetic; inability to extend the neck for any reason; and/or had a history of prior SGB. From October 2023 to May 2024, we assessed 192 individuals for eligibility (Figure 1). The most common reasons for exclusion were prior SGB (n = 42) and resolved parosmia or non−COVID-19− related OD (n = 28).
Figure 1. CONSORT Flow Diagram.
DisODOR indicates parosmia/distortion olfactory dysfunction outcomes rating; OD, olfactory dysfunction; and SGB, stellate ganglion block.
Interventions
Unilateral injections at the stellate ganglion were performed by a board-certified anesthesiologist (L.W.C.) at Washington University’s Clinical Translational Research Unit. The laterality of injection was randomized 1:1.
Participants were asked to refrain from food and drink for 8 hours prior to injection. The randomized side of the neck was prepared in a sterile manner. An ice pack and eye patch were placed on the participant’s ipsilateral eye at least 10 minutes prior to injection. After visualizing the stellate ganglion under ultrasonography guidance and confirming location of neurovascular structures, lidocaine, 1.0%, was injected superficially in the region to provide local anesthesia. Next, under sonography guidance, a 21-gauge ultrasonography needle was advanced in-plane in the prevertebral fascia above the longus coli muscle. Then, injection of 6 to 8 mL of mepivacaine, 1.0% (active group), or saline, 0.9% (placebo group) was administered around the stellate ganglion.
Because Horner syndrome is expected after SGB, participants were asked to keep the ice pack over their eye for 10 to 20 minutes postinjection and to keep the eye patch on for at least 2 hours postinjection to ensure masking to the intervention. Participants were assessed prior to, during, and after injection for adverse events, which, if present, were documented for severity, drug-relatedness, and expectedness. Vital signs were also monitored prior to, during, and after injection, until patients were deemed safe for discharge. Follow-up was conducted via emailed surveys and/or telephone calls to assess long-term adverse events at 1 month and 3 months postinjection.
Outcomes
Patient-reported outcome measures (PROMs) included Parosmia/Distortion Olfactory Dysfunction Outcomes Rating (DisODOR),19 Clinical Global Impression–Severity and Improvement scales (CGI-I, CGI-S)20 for OD, an adapted Long-COVID Questionnaire (LCQ),21 a derived Olfaction Catastrophizing Scale (OCS),22 Hospital Anxiety and Depression Scales (HADS),23 and the University of Pennsylvania Smell Identification Test (UPSIT).24 CGI-S and UPSIT were collected at baseline, and CGI-I at 1- and 3-month follow-up visits. All other measures were collected 3 times: at baseline and at the 1- and 3-month follow-up visits. DisODOR was assessed at screening and all visits.
The primary outcome was a 25% between-group proportion difference in responders. Responders were defined by 15-point decrease in DisODOR score at 1 or 3 months from baseline.19 ΔDisODOR was defined as baseline score minus follow-up visit score, and ΔΔDisODOR represented the change at each time, between treatment groups.
Sample Size and Randomization
Sample size was estimated using prior pilot study results.15 A 25% between-group difference in responder proportion was considered clinically meaningful. Using G*Power, 3.1.2 (Heinrich-Heine-Universität Düsseldorf),25 and Fisher exact test, a sample size of 114 was estimated.26 Estimating 20% dropout, up to 140 participants were approved. Planned interim analyses included safety analysis after 20 participants enrolled, and futility analysis after 40 participants completed 1-month surveys. Permuted blocks of varying sizes were used to achieve a randomization sequence resulting in a 2:1 allocation ratio for the active SGB group and the placebo group.
Blinding
Double-blinding was preserved by Barnes Jewish Hospital investigational drug pharmacy via distribution of identical-appearing treatment vials labeled “preservative-free mepivacaine HCL 1% or 0.9% NaCl 8 mL.” Additionally, eye patch and cold pack placement described under intervention were used to disguise potential but expected Horner syndrome after SGB. After assessing for initial adverse effects, assessment of the blind was conducted by asking participants what treatment they believed they received.
Statistical Analysis
Between-group differences in proportion of responders was assessed using proportion difference (PD) with 95% CIs, following per-protocol analysis principle. Interim analysis was conducted by calculating conditional power based on 40 participants to detect the clinically significant between-group difference of 25% at the end of the study (114 patients). Conditional power less than 20% would suggest futility of continuing the study.27 Data analysis was completed using SPSS, version 29.0.2.0 (IBM Corp), from September to October 2024.
Results
Baseline Characteristics
Of 57 enrolled participants, 48 were randomized and received injections through June 12, 2024 (Figure 1). Of these, 32 were randomized to SGB and 16 to placebo. Baseline demographic characteristics were similar between groups (Table 1). Median (range) age at the time of enrollment was 45.0 (19-64) years and 54.5 (26-64) years for SGB and placebo, respectively (median difference [MD], 5.0; 95% CI, −2.0 to 13.0). Both groups were 81% female and 19% male (SGB, 25 and 7, and placebo, 13 and 3, respectively; PD, 0%; 95% CI, −21% to 26%).
Table 1. Baseline Characteristics of Participants Randomized to Stellate Ganglion Block (SGB) or Placebo Group.
| Characteristic | Study group, No. (%) | Difference (95% CI)a | |
|---|---|---|---|
| SGB (n = 32) | Placebo (n = 16) | ||
| Demographic | |||
| Age, median (range), y | 45.0 (19 to 64) | 54.5 (26 to 64) | 5.0 (−2.0 to 13.0) |
| Sex | |||
| Female | 25 (78) | 13 (81) | 3 (−23 to 24) |
| Male | 6 (22) | 3 (19) | |
| Race | |||
| Black | 1 (3) | 2 (12) | −9 (−33 to 6) |
| White | 31 (97) | 14 (88) | |
| Ethnicity | |||
| Hispanic or Latino | 3 (9) | 1 (6) | 3 (−20 to 19) |
| Not Hispanic or Latino | 29 (91) | 15 (94) | |
| Time since COVID-19 diagnosis, median (range), mo | 35.3 (8.4 to 49.6) | 30.6 (3.0 to 50.5) | −3.1 (−10.9 to 3.7) |
| Measures of disease severity, median (range) | |||
| DisODOR total score | 63 (33 to 97) | 66 (26 to 95) | −1.0 (−13.0 to 11.0) |
| CGI-S smell loss | |||
| No loss | 1 (3) | 1 (6) | 3 (−11 to 25) |
| Very mild to mild loss | 6 (19) | 5 (31) | 13 (−11 to 39) |
| Moderate to severe loss | 25 (78) | 10 (63) | −16 (−42 to 10) |
| CGI-S smell distortion | |||
| No distortion | 3 (9) | 2 (12) | 3 (−14 to 27) |
| Very mild to mild distortion | 1 (3) | 1 (6) | 3 (−11 to 25) |
| Moderate to severe distortion | 28 (88) | 13 (82) | −6 (−32 to 14) |
| LCQ total score | 12.0 (0 to 37) | 14.5 (0 to 42) | 2.0 (−5.0 to 9.0) |
| OCS | 35.0 (10 to 51) | 33.5 (18 to 52) | −2.0 (−8.0 to 6.0) |
| HADS-anxiety total score | 7.0 (0 to 21) | 9.0 (2 to 18) | 1.0 (−2.0 to 4.0) |
| HADS-depression total score | 7.0 (0 to 17) | 6.0 (2 to 13) | −1.0 (−3.0 to 2.0) |
| UPSIT total score | 23.5 (8 to 37) | 25.0 (20 to 38) | 3.0 (−2.0 to 8.0) |
| Expectation (How confident are you that this treatment will improve your smell loss?) | |||
| Not at all | 0 (0) | 0 (0) | 0 (−11 to 19) |
| Slightly confident | 9 (28) | 4 (25) | −3 (−26 to 24) |
| Somewhat confident | 12 (38) | 9 (57) | 19 (−10 to 44) |
| Very confident | 7 (22) | 2 (12) | −9 (−29 to 17) |
| Extremely confident | 4 (12) | 1 (6) | −6 (−23 to 17) |
| Blind assessment | |||
| Active medication | 18 (56) | 12 (75) | 19 (−10 to 41) |
| Placebo | 14 (44) | 4 (25) | |
Abbreviations: CGI-S, Clinical Global Impression–Severity; DisODOR, Parosmia/Distortion Olfactory Dysfunction Outcomes Rating; HADS, Hospital Anxiety and Depression Scales; LCQ, adapted Long-COVID Questionnaire; OCS, Olfaction Catastrophizing Scale; UPSIT, University of Pennsylvania Smell Identification Test.
aDifference between groups uses proportion difference with Fisher exact confidence interval estimate for categorical variables, and median difference with Hodges-Lehman confidence interval for continuous variables.
In the SGB group, 18 participants (56%) received left-sided injections and 14 (44%), right-sided injections; in the placebo, 7 (44%) received left-sided injections and 9 (56%), right-sided injections. Forty-seven participants (98%) completed 1-month surveys, and 46 participants (96%) completed 3-month surveys.
Median time since COVID-19 diagnosis between groups was also similar (SGB, 35.3 vs placebo, 30.6 months; MD, −3.1 months; 95% CI, −10.9 to 3.7). Similarly, baseline parosmia severity according to DisODOR score was similar between groups (SGB, 63 vs placebo, 66; MD, −1.0; 95% CI, −13.0 to 11.0). UPSIT was only collected at baseline prior to receiving the study treatment and was similar between groups (SGB, 23.5 vs placebo, 25; MD, 3.0; 95% CI, −2.0 to 8.0). Although the range of UPSIT scores indicated a high burden of hyposmia in some individuals, there was no correlation found between UPSIT and DisODOR scores (Spearman ρ, −0.11; 95% CI, −0.39 to 0.19), suggesting that smell identification did not correlate with patient-reported impact on quality of life due to parosmia.
Baseline CGI-S for smell loss and smell distortion were also similar between groups (Table 1). Notably, both smell loss and smell distortion were prevalent in both groups, with 25 participants (78%) in the SGB and 10 (63%) in the placebo groups noting moderate to severe smell loss, and 28 (88%) in SGB and 13 (82%) in the placebo noting moderate to severe smell distortion.
Interim analysis for safety was performed after enrollment of 20 participants, and no serious adverse events were noted. A second interim analysis was conducted for futility after 40 participants completed the 1-month follow-up. There were clinically meaningful responses in 11 participants (42%) in the SGB and 6 (43%) in the placebo group. Conditional power was 2.6%, therefore, enrollment was stopped, and remaining follow-up occurred through September 16, 2024, for the 48 patients randomized that had already received an injection.
Olfaction Outcomes
At 1-month follow-up, DisODOR score decreased by a median (range) of 9.0 (−19.0 to 54.0) points for SGB and 9.0 (−13.0 to 82.0) points for placebo (MD, 0; 95% CI, −10.0 to 15.5). Based on the minimal clinically important difference, 12 participants (39%) in the SGB and 7 (44%) in the placebo group were responders (PD, 5%; 95% CI, −22% to 33%).
At 3-month follow-up, DisODOR score decreased by a median (range) of 14.0 (−21.0 to 59.0) points for SGB and 10.5 (−16.0 to 60.0) points for placebo (MD, −1.0; 95% CI, –14.0 to 17.0), corresponding to 43% (n = 13) SGB and 38% (n = 6) placebo responders (PD, −6%; 95% CI, −32% to 33%). Figure 2 demonstrates the change in DisODOR scores at the 1- and 3-month follow-ups.
Figure 2. Within-Subject Change in DisODOR (ΔDisODOR) from Baseline to 1 Month and Baseline to 3 Months.
Dotted line at 15 points represents the threshold for the magnitude of change in DisODOR that defines a responder. Between-group ΔΔDisODOR (95% CI) is listed above each time point. The reference line at ΔDisODOR = 0 suggests no change in DisODOR from baseline. DisODOR indicates parosmia/distortion olfactory dysfunction outcomes rating, and SGB, stellate ganglion block.
Changes in CGI-I scores at 1- and 3-month follow-ups are shown in Table 2. No clinically relevant improvement in smell loss was noted at either time point. CGI-I for smell distortion did not reflect clinically important differences at 1- or 3-month follow-up in either SGB or placebo participants. When evaluating participant perception of improvement, 8 participants (26%) in the SGB group and 4 (25%) in the placebo group rated smell distortion slightly or much better at 1 month (PD, −1%; 95% CI, −24% to 27%). At 3 months, 12 participants (40%) in the SGB group and 5 (31%) in the placebo group noted slightly or much better smell distortion (PD, −9%; 95% CI, −33% to 20%).
Table 2. Impact of Treatment Assignment to Stellate Ganglion Block (SGB) or Placebo Group for 1- and 3-Month Outcome Measures.
| Response | No. (%) | |||||
|---|---|---|---|---|---|---|
| 1-mo outcome | 3-mo outcome | |||||
| SGB (n = 31) | Placebo (n = 16) | Difference (95% CI)a | SGB (n = 30) | Placebo (n = 16) | Difference (95% CI)a | |
| ΔDisODOR, median (range) | 9.0 (−19 to 54) | 9.0 (−13 to 82) | 0.0 (−10 to 15) | 14.0 (−21 to 59) | 10.5 (−16 to 60) | −1.0 (−14 to 17) |
| Responderb | ||||||
| Yes | 12 (39) | 7 (44) | 5 (−22 to 33) | 13 (43) | 6 (38) | −6 (−32 to 23) |
| No | 19 (68) | 9 (32) | 17 (57) | 10 (62) | ||
| CGI-I smell loss | ||||||
| Slightly to much better | 8 (26) | 4 (25) | −1 (−24 to 27) | 13 (43) | 6 (37) | −6 (−32 to 23) |
| About the same | 22 (71) | 11 (69) | −2 (−24 to 33) | 13 (43) | 10 (63) | 19 (−10 to 44) |
| Slightly to much worse | 1 (3) | 1 (6) | 3 (−11 to 25) | 4 (14) | 0 (0) | −13 (−30 to 8) |
| LCQ total, median (range) | 44 (21 to 68) | 45 (11 to 49) | 1.0 (−2 to 1) | 44 (16 to 74) | 44 (11 to 53) | 0.0 (−5 to 1) |
| ΔOCS score, median (range) | 8.0 (−6 to 32) | 5.5 (−5 to 38) | 0.5 (−6 to 6) | 6.5 (−13 to 31) | 4.5 (−8 to 33) | 1.0 (−7 to 8) |
| ΔHADS anxiety, median (range) | 1.0 (−6 to 10) | 1.0 (−5 to 10) | 1.0 (−2 to 4) | 0.5 (−8 to 18) | 1.0 (−8 to 8) | 0.0 (−2 to 4) |
| ΔHADS depression, median (range) | 1.0 (−7 to 10) | 1.0 (−4 to 8) | 0.0 (−2 to 3) | 8.0 (−6 to 32) | 2.5 (−7 to 7) | 1.0 (−2 to 4) |
| Are you satisfied with your treatment? | ||||||
| Somewhat to completely satisfied | 8 (26) | 5 (31) | 5 (−19 to 33) | 11 (37) | 5 (31) | −5 (−30 to 23) |
| Neither satisfied nor dissatisfied | 10 (32) | 4 (25) | −7 (−30 to 21) | 1 (3) | 5 (31) | 28 (6 to 52) |
| Somewhat to completely dissatisfied | 13 (42) | 7 (44) | 2 (−25 to 30) | 18 (60) | 7 (44) | 16 (−14 to 46) |
| Would you recommend this treatment? | ||||||
| Yes | 21 (68) | 11 (69) | 1 (−27 to 26) | 14 (47) | 11 (69) | 22 (−8 to 46) |
| No | 10 (32) | 5 (31) | 16 (53) | 5 (31) | ||
Abbreviations: CGI-S, Clinical Global Impression–Severity; DisODOR, Parosmia/Distortion Olfactory Dysfunction Outcomes Rating; HADS, Hospital Anxiety and Depression Scales; LCQ, adapted Long-COVID Questionnaire; OCS, Olfaction Catastrophizing Scale.
aProportion difference with Fisher exact confidence interval estimate for categorical variables, median difference with Hodges-Lehman confidence interval for continuous variables.
bA responder is defined as a participant with a 15-point decrease in DisODOR score.
Mental Health and Outcomes of Post–COVID-19 Condition
The impact on olfaction catastrophizing, anxiety, depression, and symptoms of post–COVID-19 condition were also assessed at 1- and 3-month postinjection follow-ups (Table 2). At 1 month, the median OCS score decreased by 8.0 points in the SGB and 5.5 points in the placebo group (MD, 0.5; 95% CI, −6.0 to 6.0). At 3 months, there was a 6.5-point decrease for SGB and 4.5-point decrease for placebo (MD, 1.0; 95% CI, −7.0 to 8.0). There were no clinically or statistically relevant between-group differences in anxiety or depression symptoms at either time point. Among the 11 symptoms of post–COVID-19 condition that were tracked using the LCQ questionnaire, there were no clinically significant between-group differences in improvement of tiredness and/or fatigue, shortness of breath, brain fogginess, cough, depression, low-grade fevers, palpitations, dizziness, muscle pains, or joint pains at 1- and 3-month follow-ups. However, 1-month nonresponders in both groups had higher scores on OCS (MD, 11.0; 95% CI, 4.0 to 18.0), HADS-A (MD, 4.0; 95% CI, 1.0 to 6.0), and HADS-D (MD, 3.0; 95% CI, 1.0 to 5.0) compared to responders (total of 19 participants). At 3 months, nonresponders had higher scores on OCS (MD, 19.0; 95% CI, 10.0 to 26.0) and HADS-D (MD, 3.0; 95% CI, 1.0 to 6.0) compared to responders (19 participants).
Preintervention Expectation and Treatment Satisfaction
Most patients in both groups (21 [66%] in SGB and 13 [81%] in placebo) reported slightly confident to somewhat confident that treatment would improve their smell (PD, 16; 95% CI, −10 to 41). Among the 19 responders at 1 month, 16 (85%) were only slightly or somewhat confident in the treatment, whereas only 3 (16%) were very or extremely confident in the treatment preintervention. Similarly, at 3 months, 15 responders (79%) were only slightly or somewhat confident in the treatment, and 4 (21%) were very or extremely confident in the treatment preintervention.
Overall, there were mixed results on satisfaction with treatment, with 8 participants (26%) in the SGB and 5 (31%) in the placebo reporting somewhat to completely satisfied at 1-month postintervention. At 3 months, 11 (37%) in the SGB and 5 (31%) in the placebo group were somewhat to completely satisfied. Responders were more likely to report being somewhat to completely satisfied with treatment: 47% (n = 9) at 1 month, and 68% (n = 13) at 3 months, whereas nonresponders reported any level satisfaction of 21% (n = 4) at 1 month, and 16% (n = 3) at 3 months.
Assessment of Blinding
In the placebo group, 12 of 16 participants (75%) thought they had received the active intervention, whereas 18 of 32 participants (56%) in the intervention group thought that they had received the active intervention (PD, 19%; 95% CI, −10 to 41). These findings suggest successful blinding.
Adverse Events
No serious or unexpected adverse events were reported to the study team by participants. Nonserious and expected adverse events (AEs) immediately postinjection are summarized in Table 3. The most common AEs were transient voice hoarseness (9 participants in SGB), nasal congestion (6 in SGB), headache or migraine (5 in SGB; 1 in placebo), and light-headedness or dizziness (5 in SGB; 1 on placebo).
Table 3. Adverse Events at Baseline Visit After Treatment Injection, Stellate Ganglion Block (SGB) vs Placebo.
| Adverse eventa | No. (%) | |
|---|---|---|
| SGB | Placebo | |
| Voice hoarseness | 9 (28) | 0 |
| Nasal congestion | 6 (19) | 0 |
| Headache/migraine | 5 (16) | 1 (5) |
| Light-headedness/dizziness | 5 (16) | 1 (5) |
| Hypertension | 5 (16) | 0 |
| Globus sensation | 4 (13) | 0 |
| Eye throbbing, warmth | 2 (6) | 0 |
| Chest tightness/pressure | 1 (3) | 1 (5) |
| Vasovagal syncope | 1 (3) | 0 |
| Shortness of breath | 1 (3) | 0 |
| Sensation of warmth in lips | 1 (3) | 0 |
| Postnasal drip | 1 (3) | 0 |
| Throat pressure | 1 (3) | 0 |
| Ear fullness | 1 (3) | 0 |
| Nausea | 0 | 1 (5) |
aCategories are not mutually exclusive.
Discussion
Parosmia is commonly noted after COVID-19 infection; however, to our knowledge, there is no validated treatment for this condition. In this randomized clinical trial of SGB for the treatment of COVID-19−induced parosmia, we found no clinically meaningful improvement over that of placebo, with a 3-month response rate of 43% and 38% for SGB and placebo, respectively. Therefore, despite anecdotal and retrospective evidence supporting SGB, and its enthusiastic reception in the community, we cannot recommend this treatment for COVID-19−induced parosmia based on the results of this study.
The high response rate among participants receiving placebo highlights the substantial impact of psychological factors on treatment outcomes. Nonetheless, a substantial placebo response of at least 38% in the overall cohort after 3 months warrants further attention. In the present study, somewhat or slight confidence in improvement rather than very or extreme confidence were more likely to respond to either treatment. According to the desire-expectation model of emotions, placebo responses are driven by decreased negative or increased positive emotional states for avoidance and approach goals, respectively.28 Receiving an injection expected to alleviate parosmia aligns with an avoidance goal, for which we expect a higher expectation of improvement to estimate higher satisfaction, and therefore, response to treatment. Surprisingly, our results demonstrate that participants who enrolled with lower expectations were more likely to be responders in both groups.
Notably, there was no association of baseline catastrophizing, anxiety, or depression ratings with treatment response. This suggests that parosmia improvement was not associated with baseline psychosocial measures. However, at follow-up, patients with higher depression and catastrophizing scores consistently reported fewer clinically meaningful improvements in DisODOR at that time point, regardless of treatment group. This aligns better with the desire-expectation placebo-effect model, suggesting negative emotional states were associated with worse outcomes. While the directionality of the relationship between these factors cannot be established in the present study, these findings imply that patient emotional states at time of assessment are associated with treatment response. The observation that patients with lower expectations were more likely to respond may suggest that those with stronger reliance on an external treatment (reliance on external locus of control) were less likely to report benefit in parosmia PROMs or improved mental health PROMs.29 Alternatively, patients with higher expectations may have been disappointed if they did not feel improvement, leading to higher ratings of catastrophizing, anxiety, or depression. Future studies should consider additional routes of gauging expectations, attitudes around treatment, satisfaction, and directionality of associations between psychosocial measures and parosmia PROMs. A comprehensive approach to treatment should consider targeting both the physiological sources of parosmia and patients’ mental and emotional well-being, with the goal of providing subjective relief from this sequela of post–COVID-19 condition.
Most of the patients in the present study continued to experience the burden of COVID-19−induced parosmia. Given the broader quality-of-life challenge posed by this manifestation of post–COVID-19 condition and the failure of SGB as an effective treatment, researchers should continue to search for a suitable therapy. Another recent trial30 explored gabapentin as treatment for COVID-19−induced parosmia, but did not show any benefit. A separate study31 suggested that a combination of palmitoylethanolamide and luteolin with alpha-lipoic acid and olfactory therapy may resolve COVID-19−induced parosmia. However, this study was not placebo-controlled and lacked appropriate measures for assessing parosmia improvement. Considering the media-driven push for various therapies without evidence to support their use, more well-designed clinical trials with validated measures of disease-specific improvement should be conducted to prevent harm and to identify efficacious treatments for individuals with COVID-19−induced parosmia.
Strengths and Limitations
The strengths of the study were its well-preserved blinding, placebo-control, and SGB administration by an experienced specialist at a high-volume center. Limitations include lack of objective assessment for parosmia, which remains poorly defined in the literature.
Conclusions
The findings of this randomized clinical trial indicate the nonsuperiority of SGB with mepivacaine compared to saline injection placebo for the treatment of COVID-19−induced parosmia. Interestingly, there was close to 40% response rate in both the active intervention and placebo groups, indicating a very strong placebo response in this population. Given the results of this study, SGB should not be recommended for treatment of COVID-19−induced parosmia. Additional studies should be performed to evaluate the high placebo response rate in this patient population. Alternatives should also be evaluated for the treatment of COVID-19−induced parosmia.
Trial Protocol
Data Sharing Statement
References
- 1.Menni C, Valdes AM, Freidin MB, et al. Real-time tracking of self-reported symptoms to predict potential COVID-19. Nat Med. 2020;26(7):1037-1040. doi: 10.1038/s41591-020-0916-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lerner DK, Garvey KL, Arrighi-Allisan AE, et al. Clinical features of parosmia associated with COVID-19 infection. Laryngoscope. 2022;132(3):633-639. doi: 10.1002/lary.29982 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hopkins C, Surda P, Vaira LA, et al. Six month follow-up of self-reported loss of smell during the COVID-19 pandemic. Rhinology. 2021;59(1):26-31. [DOI] [PubMed] [Google Scholar]
- 4.Callejón-Leblic MA, Martín-Jiménez DI, Moreno-Luna R, et al. Analysis of prevalence and predictive factors of long-lasting olfactory and gustatory dysfunction in COVID-19 patients. Life (Basel). 2022;12(8):1256. doi: 10.3390/life12081256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Burges Watson DL, Campbell M, Hopkins C, Smith B, Kelly C, Deary V. Altered smell and taste: anosmia, parosmia and the impact of long Covid-19. PLoS One. 2021;16(9):e0256998. doi: 10.1371/journal.pone.0256998 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stoller E, Hentati F, Kallogjeri D, Piccirillo JF, Lenze EJ, Farrell NF. Depression, anxiety, and suicidal ideation in adults with COVID-induced parosmia. Int Forum Allergy Rhinol. 2023;13(12):2240-2243. doi: 10.1002/alr.23211 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Santos DV, Reiter ER, DiNardo LJ, Costanzo RM. Hazardous events associated with impaired olfactory function. Arch Otolaryngol Head Neck Surg. 2004;130(3):317-319. doi: 10.1001/archotol.130.3.317 [DOI] [PubMed] [Google Scholar]
- 8.Damm M, Pikart LK, Reimann H, et al. Olfactory training is helpful in postinfectious olfactory loss: a randomized, controlled, multicenter study. Laryngoscope. 2014;124(4):826-831. doi: 10.1002/lary.24340 [DOI] [PubMed] [Google Scholar]
- 9.Addison AB, Wong B, Ahmed T, et al. Clinical Olfactory Working Group consensus statement on the treatment of postinfectious olfactory dysfunction. J Allergy Clin Immunol. 2021;147(5):1704-1719. doi: 10.1016/j.jaci.2020.12.641 [DOI] [PubMed] [Google Scholar]
- 10.Kang CK, Oh ST, Chung RK, et al. Effect of stellate ganglion block on the cerebrovascular system: magnetic resonance angiography study. Anesthesiology. 2010;113(4):936-944. doi: 10.1097/ALN.0b013e3181ec63f5 [DOI] [PubMed] [Google Scholar]
- 11.Treggiari MM, Romand JA, Martin JB, Reverdin A, Rüfenacht DA, de Tribolet N. Cervical sympathetic block to reverse delayed ischemic neurological deficits after aneurysmal subarachnoid hemorrhage. Stroke. 2003;34(4):961-967. doi: 10.1161/01.STR.0000060893.72098.80 [DOI] [PubMed] [Google Scholar]
- 12.Mulvaney SW, McLean B, de Leeuw J. The use of stellate ganglion block in the treatment of panic/anxiety symptoms with combat-related post-traumatic stress disorder; preliminary results of long-term follow-up: a case series. Pain Pract. 2010;10(4):359-365. doi: 10.1111/j.1533-2500.2010.00373.x [DOI] [PubMed] [Google Scholar]
- 13.Lipov EG, Joshi JR, Sanders S, Slavin KV. A unifying theory linking the prolonged efficacy of the stellate ganglion block for the treatment of chronic regional pain syndrome (CRPS), hot flashes, and posttraumatic stress disorder (PTSD). Med Hypotheses. 2009;72(6):657-661. doi: 10.1016/j.mehy.2009.01.009 [DOI] [PubMed] [Google Scholar]
- 14.Fischer L, Barop H, Ludin SM, Schaible HG. Regulation of acute reflectory hyperinflammation in viral and other diseases by means of stellate ganglion block. A conceptual view with a focus on COVID-19. Auton Neurosci. 2022;237:102903. doi: 10.1016/j.autneu.2021.102903 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Peterson AM, Miller BJ, Kallogjeri D, et al. Stellate ganglion block for the treatment of COVID-19-induced olfactory dysfunction: a prospective pilot study. Otolaryngol Head Neck Surg. 2024;170(1):272-276. doi: 10.1002/ohn.530 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Galvin V, Cheek DJ, Zhang Y, Collins G, Gaskin D. Short communication: stellate ganglion blockade for persistent olfactory and gustatory symptoms post-COVID-19. Local Reg Anesth. 2023;16:25-30. doi: 10.2147/LRA.S402197 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Naimi BR, Garvey E, Chandna M, et al. Stellate ganglion block for treating post-COVID-19 parosmia. Int Forum Allergy Rhinol. 2024;14(6):1088-1096. doi: 10.1002/alr.23314 [DOI] [PubMed] [Google Scholar]
- 18.Gunder N, Hummel T. Parosmia in patients with post-infectious olfactory dysfunction in the era of COVID-19-associated olfactory impairment. HNO. 2024;72(9):649-656. doi: 10.1007/s00106-024-01470-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ioerger P, Kallogjeri D, Roland L, Schneider JS, Piccirillo JF, Farrell NF. Development and validation of the parosmia olfactory dysfunction outcomes rating (DisODOR). Otolaryngol Head Neck Surg. 2023;169(6):1654-1661. doi: 10.1002/ohn.477 [DOI] [PubMed] [Google Scholar]
- 20.Dunlop BW, Gray J, Rapaport MH. Transdiagnostic clinical global impression scoring for routine clinical settings. Behav Sci (Basel). 2017;7(3):40. doi: 10.3390/bs7030040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hughes SE, Haroon S, Subramanian A, et al. Development and validation of the symptom burden questionnaire for long COVID (SBQ-LC): Rasch analysis. BMJ. 2022;377:e070230. doi: 10.1136/bmj-2022-070230 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sullivan MJ, Bishop SR, Pivik J. The pain catastrophizing scale: development and validation. Psychol Assess. 1995;7(4):524. doi: 10.1037/1040-3590.7.4.524 [DOI] [Google Scholar]
- 23.Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361-370. doi: 10.1111/j.1600-0447.1983.tb09716.x [DOI] [PubMed] [Google Scholar]
- 24.Doty RL, Shaman P, Dann M. Development of the University of Pennsylvania Smell Identification Test: a standardized microencapsulated test of olfactory function. Physiol Behav. 1984;32(3):489-502. doi: 10.1016/0031-9384(84)90269-5 [DOI] [PubMed] [Google Scholar]
- 25.Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods. 2009;41(4):1149-1160. doi: 10.3758/BRM.41.4.1149 [DOI] [PubMed] [Google Scholar]
- 26.Schulz KF, Altman DG, Moher D; CONSORT Group . CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. J Clin Epidemiol. 2010;63(8):834-840. doi: 10.1016/j.jclinepi.2010.02.005 [DOI] [PubMed] [Google Scholar]
- 27.Mehta CR, Pocock SJ. Adaptive increase in sample size when interim results are promising: a practical guide with examples. Stat Med. 2011;30(28):3267-3284. doi: 10.1002/sim.4102 [DOI] [PubMed] [Google Scholar]
- 28.Price DD, Fields HL. The contribution of desire and expectation to placebo analgesia: implications for new research strategies. In: Harrington A, ed. The Placebo Effect: An Interdisciplinary Exploration. Harvard University Press; 1997:117-137. [Google Scholar]
- 29.Anderson S, Stebbins GT. Determinants of placebo effects. Int Rev Neurobiol. 2020;153:27-47. doi: 10.1016/bs.irn.2020.03.029 [DOI] [PubMed] [Google Scholar]
- 30.Mahadev A, Hentati F, Miller B, et al. Efficacy of gabapentin for post-COVID-19 olfactory dysfunction: the GRACE randomized clinical trial. JAMA Otolaryngol Head Neck Surg. 2023;149(12):1111-1119. doi: 10.1001/jamaoto.2023.2958 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Cantone E, D’Ascanio L, De Luca P, et al. Persistent COVID-19 parosmia and olfactory loss post olfactory training: randomized clinical trial comparing central and peripheral-acting therapeutics. Eur Arch Otorhinolaryngol. 2024;281(7):3671-3678. doi: 10.1007/s00405-024-08548-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol
Data Sharing Statement
