Stellate Ganglion Block for the Treatment of COVID-19-Induced Olfactory Dysfunction: A Prospective Pilot Study

Abstract

Objective:

The objective of this study was to explore the safety and feasibility of stellate ganglion blocks (SGBs) to treat persistent COVID-19-induced olfactory dysfunction (OD). Secondarily, the goal was to determine effect sizes to plan a future randomized clinical trial.

Study Design:

Prospective case series

Setting:

Quaternary care academic medical center

Methods:

In this single-arm pilot trial, adult participants with a COVID-19 diagnosis ≥ 12 months prior to enrollment with OD underwent bilateral SGBs. Subjects were followed for 1 month after completion of SGB. The primary outcome measure was the change in Clinical Global Impression - Improvement scale for smell loss. Secondary outcome measures included change in University of Pennsylvania Smell Identification Test (UPSIT) and Olfactory Dysfunction Outcomes Rating (ODOR).

Results:

Twenty participants were enrolled with a mean (SD) age of 46 (11) years and a mean (SD) duration of OD of 21 (5) months. At 1-month, 10 (50%) participants experienced at least slight subjective improvement in their OD, 11 (55%) attained a clinically meaningful improvement in smell identification using the UPSIT, and 7 (35%) achieved a clinically meaningful improvement in olfactory-specific QOL measured by the ODOR. The median difference between UPSIT scores at baseline and 1-month was 6 (95% CI, 3 to 11), exceeding the MCID of 4. There were no serious adverse events.

Conclusion:

Sequential SGBs for COVID-19-associated OD were safe and associated with modest improvements in subjective olfaction, odor identification, and olfactory-specific QOL. A placebo-controlled trial is warranted to determine the efficacy of SGBs for COVID-19-associated OD.

INTRODUCTION

Millions of people have developed persistent olfactory dysfunction (OD) following COVID-19 infection, and few treatments for COVID-19-induced OD have proven effective. The stellate ganglion block (SGB) inhibits sympathetic neuronal connections and reduces circulating adrenal hormone levels.¹ The SGB has been used successfully in a multitude of disorders, including post-traumatic stress disorder², migraine³, and complex regional pain syndrome.⁴ A meta-analysis of 12 clinical trials found that SGB was superior to placebo in reducing pain scores among patients with various sympathetic hyperactivity-associated disorders.⁵ Many “long COVID” symptoms, those that persist after recovery from acute COVID-19 infection, are hypothesized to be, at least in part, a result of sympathetic hyperactivity resulting in positive feedback loops.⁶ Anecdotal evidence and case reports describe success with the use of stellate ganglion blocks (SGBs) for both COVID-19-induced OD and other post-viral causes. ^7–9 This study aims primarily to explore the safety and feasibility of administering SGBs for the treatment of persistent COVID-19-induced OD and secondarily to generate pilot data to guide effect size estimates for a future appropriately powered, randomized clinical trial.

METHODS

This study was a single-arm, prospective case series with recruitment and follow-up occurring from September 8, 2022, to December 12, 2022. The study protocol was reviewed and approved by the Washington University School of Medicine Institutional Review Board (IRB# 202207170) and the study was registered on clinicaltrials.gov (NCT05445921). Adults with a COVID-19 diagnosis at least 12 months prior to recruitment with subjective and objective OD were eligible. Those with a history of other known causes of OD or contraindications to the SGB were excluded. Enrolled participants underwent bilateral SGBs with mepivacaine performed 1 week apart by a pain management physician (L.W.C.). Further study procedure details are available in Supplement 1. Participants were followed for 1 month after completion of both SGBs.

The primary outcome was the 7-point Likert Clinical Global Impression – Improvement (CGI-I) scale for smell loss at 1 month.¹⁰ Secondary outcome measures included CGI-I for parosmia, participant satisfaction with the SGB, and changes in UPSIT and Olfactory Dysfunction Outcomes Rating (ODOR) scores.^11–13 The minimal clinically important difference (MCID) for UPSIT and ODOR is 4 and 15, respectively. Outcomes were assessed prior to the first SGB, 1 week after the first SGB, and 1 month following the second SGB. Sample size calculation was done based on a 71% self-reported response rate observed in a previous study done in patients with non-COVID post-viral OD.¹⁴ Data were analyzed following intention-to-treat principles.

RESULTS

A total of 20 participants were enrolled, and all participants completed the study. The mean (SD) age was 46 (11) years, 16 (80%) participants were women, 20 (100%) were white, and the mean (SD) duration of OD was 21 (5) months.

At 1 week after the first SGB, 8 (40%) and 9 (45%) participants reported at least slight improvement in smell loss and parosmia, respectively. The median difference in UPSIT scores from baseline to 1 week was 4.5 (95% CI: 2 to 6.5), and 12 (60%) participants attained a clinically meaningful improvement in smell identification. The median difference in ODOR scores from baseline to 1 week was −5.5 (95% CI: −11 to −1), and 5 (25%) participants achieved a clinically meaningful improvement in olfaction-related QOL as measured by the ODOR. Eight (40%) participants reported being at least somewhat satisfied with the SGB and 17 (85%) reported they would recommend it to a close friend or family member with the same condition.

At 1-month, 10 (50%) and 9 (45%) participants experienced at least slight improvement in smell loss and parosmia, respectively. The median difference in UPSIT scores from baseline to 1 month was 6 (95% CI: 3 to 11), and 11 (55%) participants attained a clinically meaningful improvement in smell identification. The median difference in ODOR scores from baseline to 1 month was −11 (95% CI: −19 to 5.5), and 7 (35%) participants achieved a clinically meaningful improvement in olfactory-specific QOL. Eleven (55%) participants reported being at least somewhat satisfied with the SGB and 17 (85%) reported they would recommend it to a close friend or family member with the same condition.

CGI-I responses at each follow-up, and UPSIT and ODOR scores at baseline and each follow-up are summarized in Table 1. UPSIT scores at baseline and each follow-up are illustrated in Figure 1. Reported adverse effects are shown in Table 2. There were no serious adverse events, and all adverse events with the exception of fatigue and shoulder tightness were expected. All adverse events were transient; most resolved by the end of the 10-minute post-procedure observation period, and all resolved within a few hours. Horner’s syndrome, which is expected with a stellate ganglion block, resolved in about 50% of participants within an hour and within 2–3 hours for the remainder of participants. Vision was not affected.

Table 1.

CGI-I, UPSIT, and ODOR Outcomes at Baseline, One-Week, and One-Month

Outcome Measure	Baseline (n = 20)	One Week (n = 20)			One Month (n = 20)
		n (%)			n (%)
CGI-I Smell Lossa
About the same	NA	12 (60)			10 (50)
Slightly better	NA	8 (40)			6 (30)
Moderately better	NA	0 (0)			3 (15)
Much better	NA	0 (0)			1 (5)
CGI-I Parosmiaa
About the same	NA	11 (55)			11 (55)
Slightly better	NA	9 (45)			6 (30)
Moderately better	NA	0 (0)			2 (10)
Much better	NA	0 (0)			1 (5)
	Median (Range)	Median (Range)	Median difference (95% CI)	Participants Reaching MCID, n (%)	Median (Range)	Median difference (95% CI)	Participants Reaching MCID, n (%)
UPSITb	16.5 (4–32)	21 (10–31)	4.5 (2 to 6.5)	12 (60)	22.5 (10–35)	6 (3 to 11)	11 (55)
ODORc	62 (16–85)	57 (13–82)	−5.5 (−11 to −1)	5 (25)	50.5 (2–75)	−11 (−19 to −5.5)	7 (35)

Abbreviations: 95% CI, 95% confidence interval; CGI-I, Clinical Global Impression of Improvement; UPSIT, University of Pennsylvania Smell Identification Test; ODOR, Olfactory Dysfunction Outcome Rating; NA, not applicable

^aThe CGI-I a modified 7-point Likert scale used to assess the perceived change of smell loss and parosmia following the intervention.

^bThe UPSIT has a minimum score of 0 and maximum score of 40 with lower scores indicating a greater degree of impairment. An UPSIT score of >33 for men and >34 for women is considered normosmic, and the minimal clinically important difference is a change of 4 points.

^cThe ODOR has a minimum score of 0 and a maximum score of 112 with higher scores indicating a greater degree of impairment and limitation. The minimal clinically important difference is a change of 15 points.

Figure 1.

UPSIT Scores at Baseline, One-Week, and One-Month. This figure represents a box and whisker plot of UPSIT scores at baseline, one-week, and one month, which correspond to visits 1, 2, and 3, respectively.

Table 2.

Adverse Effects Reported by Study Participants

Adverse Effecta	No. (%) All completed procedures (n = 40)
Horner syndrome	39 (97.5)
Hoarseness	17 (42.5)
Globus	14 (35)
Brief lightheadedness	9 (22.5)
Numbness of the face	8 (20)
Mild shortness of breath	7 (17.5)
Nasal congestion	6 (15)
Pain at the injection site	6 (15)
Fatigueb,c	5 (12.5)
Cough	2 (10)
Weakness of the arm	2 (10)
Weakness of the neck/back	2 (10)
Tightness of the neck	2 (10)
Arm heaviness	1 (5)
Bruising of the injection site	1 (5)
Chest heaviness	1 (5)
Headache	1 (5)
Palpitations	1 (5)
Tightness of the shoulderb	1 (5)

^aDisplayed adverse effects were all found to be at least possibly related to the study, three additional adverse effects, bruising of the eyelid, persistent numbness of the neck, and photosensitivity, were also reported but their relatedness to the study was determined to be unlikely

^bAdverse effect was determined to be unexpected and was reported to the institutional review board

^cOne instance of persistent fatigue was reported and was determined to be unexpected; all other instances of fatigue were brief and were determined to be expected

DISCUSSION

This study found that among 20 participants with COVID-19-associated OD receiving sequential SGBs, 10 (50%) experienced at least slight subjective improvement in smell loss, 11 (55%) attained a clinically meaningful improvement in smell identification, and 7 (35%) achieved a clinically meaningful improvement in olfactory-specific quality of life at 1 month.

The SGBs were well-tolerated with no serious adverse events. Two common transient adverse events, hoarseness and globus, can be explained by the relationship of the recurrent laryngeal nerve with the stellate ganglion, where spill-over of the local anesthetic may mimic symptoms of a transient recurrent laryngeal nerve palsy. Therefore, the block can even rarely cause stridor, which is why bilateral stellate ganglion blocks are not recommended to be performed together. The transient arm weakness experienced by a small proportion of our cohort is likely due to the relationship of the stellate ganglion to the brachial plexus. Similarly, the symptoms of chest heaviness and shortness of breath may be explained by local anesthetic spillover to the phrenic nerve. Other rare but serious adverse events reported in the literature but not observed in our case series are pneumothorax, severe hypertension, seizure, total spinal anesthesia, transient locked-in syndrome, paratracheal hematoma, and soft tissue infection/osteomyelitis.^15,16 A randomized clinical trial investigating SGBs for PTSD reported zero of these serious adverse events following a total of 148 right-sided SGBs with active medication.² Most, if not all of these serious adverse events, can likely be avoided by having an experienced clinician administering the SGB.

The mechanistic explanation for the SGB’s efficacy is uncertain and its reported effectiveness has been previously limited to anecdotal evidence and case reports. Although SGBs have not yet been deemed an efficacious therapy for COVID-19-induced OD, they are currently being offered to patients at high out-of-pocket costs. Given this and the need to identify effective treatments for this condition, rigorous studies assessing the efficacy of the SGB are needed.

This pilot study is limited by its lack of a control group and small sample size. However, the goal of this study was to generate estimates of effect sizes and safety data rather than prove efficacy of the SGB in patients with COVID-19-associated OD. Additionally, neither participants nor study team members were blinded to the intervention. Further investigation of the SGB in a double-blinded, randomized clinical trial is needed to establish it as an efficacious therapy for COVID-19-induced OD.

CONCLUSION

This study found that sequential SGBs were safe and associated with modest improvements in subjective olfaction, odor identification, and olfactory-specific QOL. With the establishment of safety and effect sizes for the outcome measures, an appropriately powered, placebo-controlled randomized clinical trial can be planned to determine the efficacy of the SGB.