Sympathicotomy for isolated facial blushing: long-term follow-up of a randomized trial

Abstract

Open in new tabDownload slide

OBJECTIVES

Thoracoscopic sympathicotomy may be an effective treatment for disabling facial blushing in selected patients. Short- and mid-term results are good but very long-term results are scarce in the medical literature and there is no knowledge which extent of sympathicotomy is better long-term for isolated facial blushing.

METHODS

We previously randomized 100 patients between a rib-oriented R2 or R2–R3 sympathicotomy for isolated facial blushing, and reported local effects, side effects and quality of life after 12 months. In the present study, we sent identical questionnaires to all patients after a median of 16 years (interquartile range 15–17 years).

RESULTS

The response rate was 66%. Overall, 82% reported excellent or satisfactory results on facial blushing, with significant better local effect after R2 sympathicotomy compared with R2–R3 sympathicotomy. Patients who underwent R2 sympathicotomy were also significantly more satisfied with the operation. We found no significant difference between R2 and R2–R3 sympathicotomy in quality of life or rates of compensatory sweating (77%) and recurrence of blushing (41%) which was milder than preoperatively in most patients.

CONCLUSIONS

R2 sympathicotomy should be the preferred approach for isolated facial blushing because of better local effect and higher satisfaction rates. Although this was a very long-term follow-up of the only randomized trial of its kind the response rate was limited leaving a risk of undetected bias.

INTRODUCTION

Emotional blushing is a phenomenon characterized by reddening of the facial skin in response to emotional or social stimuli, first described by Charles Darwin [1], and a normal physiological response mediated by increased signalling in the sympathetic nervous system that leads to cutaneous vasodilation [2]. In some individuals, intense emotional blushing may be disabling and have severe negative impacts on social functioning [3]. Several non-surgical treatments including psychotherapeutic modalities [4] and local- or systemic pharmacotherapy [5, 6] may be utilized, but these methods often provide insufficient relief and patients may seek surgical intervention by thoracoscopic sympathicotomy, which was first used for facial blushing in 1985 [7]. Over the years, different levels of the sympathetic chain have been targeted by various surgical techniques. The most used procedure for isolated facial blushing is a rib-oriented R2 or R2–R3 sympathicotomy where the sympathetic trunk is simply transected rather than a sympathectomy, where a segment of the sympathetic chain is resected, but the term sympathectomy is often used synonymously for both procedures. A systematic review of the literature reported good short- and mid-term results with relief of blushing in 78% of patients and overall satisfaction in 84% [8]. Compensatory sweating was the most common adverse effect (74%), and gustatory sweating was seen in up to 24% of patients. There is only one publication in the medical literature on long-term effects after sympathicotomy for isolated facial blushing [9]: In a large Swedish study, Smidfelt and Drott reported 73.5% satisfaction and 14% primary failure rate in 536 patients. Their study was a retrospective follow-up study which prompted us to investigate long-term results in a prospective cohort that was randomized between two extents of sympathicotomy for isolated facial blushing [10].

PATIENTS AND METHODS

We previously randomized 100 patients to rib-oriented R2 or R2–R3 sympathicotomy for isolated facial blushing during a 6-year period from 2005 to 2011 (10) where the technical aspects have been described in detail. A standardized quality of life (QoL) Short Form 36 questionnaire and a self-made questionnaire regarding local effects and side effects were obtained from all patients in the out-patient clinic on the day before surgery and in responding patients 1 year after the operation. In the present follow-up study, which was approved by the Regional Ethics Committee of Southern Denmark (Journal nr. 21/58252), we sent electronic identical questionnaires linked to the national social security numbers of the 93 patients who responded in the initial study. The questionnaires assessed the effect of the operation on facial blushing as well as possible side effects. In short, we asked patients to report the degree of effect achieved from the procedure on blushing as excellent, satisfactory, some effect or no effect. We also asked patients to comment on occurrence and location of any compensatory sweating, defined as abnormal excessive sweating after the operation. To determine the severity of compensatory sweating, patients were asked if they had to change clothes during the day due to this side effect. Additionally, the questionnaire assessed gustatory sweating, defined as facial sweating while eating certain foods, and dry hands with subsequent need for hand lotion. Finally, patients were asked to report their level of satisfaction with the operation, whether they regretted it, and whether they would recommend it to others with severe facial blushing. In addition, we used a Short Form 36 questionnaire to assess QoL [11]. The primary end points of this study were local effects on facial blushing and side effects after 16 years. QoL was a secondary end point All information regarding QoL, local effects and side effects in patients before and 1 year after the operation was available from the original manuscript [10].

Statistical analysis

Local effects and side effects on facial blushing between the two extents of sympathicotomy were calculated by cross-tabulation (chi-squared test with linear-by-linear association). McNemar’s test and Stuart-Maxwell’s test for marginal homogeneity were used to calculate paired local effects and side effects within the two extents of sympathicotomy over time. We used a mixed effect linear regression model to calculate QoL, as presented by Short Form 36 scores [11]. The mixed model was used to analyse the QoL outcome between the two extents of sympathicotomy before the operation, 1 year after the operation and at 16 years after the operation. Additionally, QoL alterations within each extent of sympathicotomy were analysed. QoL outcomes were adjusted for interaction between levels of sympathicotomy and time of follow-up, as well as gender. P values <0.05 were considered statistically significant. All data were analysed in the STATA 17.0 software.

RESULTS

The response rate was 66% (61 out of 93 patients) after a median of 16 years [interquartile range (IQR) 15–17 years] from surgery. Two patients were lost to follow-up because their social security numbers were no longer active, suggesting that they had emigrated, and the remaining 30 patients did not respond to the questionnaire despite 3 reminders. The sympathetic chain was transected at R2 in 30 patients and at R2–R3 in 31 patients. The overall male/female ratio was 17/44: 2/28 in the R2 group and 15/16 in the R2–R3 group. The overall median age at follow-up was 49 years (IQR 42–57 years) with a median age of 49 years (IQR 42–59 years) in the R2 group and 44 years (IQR 41–56 years) in the R2–R3 group.

Primary outcome

Effect of surgery

Overall, 82% of the patients reported an excellent or satisfactory result after surgery which is shown in detail in Table 1. With respect to local effect on facial blushing we found a significant difference in favour of R2 sympathicotomy (P = 0.009). We also found significant and opposite changes over time within each type of surgery: Within the R2 group there was a significant increase in local effect over time (P = 0.011), but in the R2–R3 group there was a significant decrease in local effect over time (P = 0.046).

Recurrence of facial blushing

Table 1 shows that 25 patients (41%) experienced recurrence of facial blushing during the follow-up period with no significant difference between R2 and R2–R3 (P = 0.50). In most patients (64%) the severity was lower than preoperatively. Seven patients in the R2–R3 group and 2 patients in the R2 group reported recurrence of blushing symptoms that were identical to the preoperative level.

Compensatory sweating

Table 1 demonstrates that compensatory sweating occurred in 47 (77%) of the responding patients without significant difference between R2 and R2–R3 (P = 0.944). Table 1 also list the anatomical locations of compensatory sweating. On the abdomen, the back, and the thighs there was no difference but ‘sweating in other anatomic locations’ occurred significant more frequently after R2 sympathicotomy (n = 14 vs n = 4; P = 0.002). Severe compensatory sweating (defined as episodes with a need to change clothes due to compensatory sweating) occurred in 13% of patients who had compensatory sweating with no significant difference between the two groups. Compensatory sweating decreased significantly in almost all anatomical domains over time for both extents of sympathicotomy except compensatory sweating in ‘other anatomical locations’ within the R2 group and thighs within the R2–R3 group, where the amount of compensatory sweating was similar after 1-year and 16-year follow-up.

Gustatory sweating

Gustatory sweating was reported by 38% (Table 1). The main triggers were spicy food and fruits. There were no statistically significant differences between the two extents of sympathicotomy or within the R2–R3 group over time but it increased significantly within the R2 group from 1-year to late follow-up (P = 0.025).

Dry hands

Forty patients (66%) reported dry hands at late follow-up after sympathicotomy. There were no significant differences between the two extents of sympathicotomy, nor any notable changes over time within each group.

Overall satisfaction

At late follow-up, 46 patients (75%) reported that they were satisfied with the overall result, which was significantly higher (P = 0.009) in the R2 group (n = 27; 90%) compared with the R2–R3 group (n = 19; 61%). Fifty patients (82%) would recommend the surgery to other patients with severe facial blushing and significantly more frequent (P = 0.023) in the R2 group (n = 28; 93%) compared with the R2–R3 group (n = 22; 71%).

Regretting sympathicotomy

At late follow-up 7 patients (11%) reported that they regretted surgery which was more common after R2–R3 (n = 6 vs n = 1; P = 0.050). The patient in the R2 group regretted surgery due to a unilateral Horner’s syndrome, although it diminished over the years and had almost disappeared at late follow-up. The 6 patients in the R2–R3 group regretted surgery due to compensatory sweating (n = 2) or inadequate local effect on facial blushing with compensatory sweating (n = 4).

Secondary outcome

Total SF36

There were no significant differences in Total SF36 score between the two extents of sympathicotomy at the three time points (before, at 1 year and at late follow-up) as seen in Fig. 1. Within the R2–R3 group Total SF36 score increased significantly after 1 year (P = 0.001) and at late follow-up (P = 0.002). There were no significant changes in Total SF36 score over time within the R2 group.

Physical health

We did not find any significant changes in Physical Health SF36 scores between the two extents of sympathicotomy before the operation, at 1-year follow-up or at late follow-up. At late follow-up, however, the mean Physical Health SF36 score declined significantly over time in the R2 group (P = 0.016) as seen in Fig. 2.

Mental health

There were no significant differences in Mental Health SF36 score between the two extents of sympathicotomy at the 3 time points (Fig. 3). Mental Health increased significantly after 1 year (P < 0.001) and at late follow-up (P < 0.001) within the R2–R3 group. There were no significant changes in mental health over time in the R2 group.

DISCUSSION

Our study demonstrates that most patients who underwent sympathicotomy for isolated facial blushing achieved an excellent or satisfactory long-lasting effect suggesting that sympathetic surgery has a clear role in selected patients with disabling symptoms when non-surgical treatments fail. Our study also shows that among the two most used surgical procedures to treat isolated facial blushing the R2 sympathicotomy resulted in significantly better long-term outcomes compared with the R2–R3 sympathicotomy. Although QoL was similar between the two groups, patients who had an R2 sympathicotomy reported significantly better local effects and fewer recurrences at late follow-up. We do not have a good explanation for these findings since the R2 level was also transected in patients who underwent an R2–R3 procedure. Likewise, we have no good explanation why the intensity of recurrence was higher in the R2–R3 group where it was more often described as being identical to the blushing experienced before surgery. Our results therefore document that an R2 sympathicotomy should be the preferred approach for patients with isolated facial blushing.

Compensatory sweating is a well-known side effect after sympathicotomy for facial blushing as well as primary hyperhidrosis. The reported incidence varies in the literature between 35% and 95% and it may lose intensity over time or remain the same for years [12]. Our results suggest that compensatory sweating decreases substantially from 1 to 16 years after surgery regardless of the extent of sympathicotomy. This finding contradicts another long-term follow-up by Smidfelt and Drott [9] who did a large retrospective investigation of 648 patients with facial blushing operated between 1989 and 1998 and a mean follow-up of 14.6 years. They found that compensatory sweating did not decrease over time. In fact, most of their patients described an increase in compensatory sweating at late follow-up. Smidfelt and Drott did not specify outcomes for different levels of targeting the sympathetic chain but reported a 73.5% overall satisfaction rate: they also reported a primary failure rate of 14%, which they defined as the proportion of patients who were not satisfied with the effect on blushing. In any case, we believe our study adds valuable information because we present results from a prospectively randomized cohort.

Limitations

Our study has important limitations because of potential reporting bias and because of the low number of patients. The response rate was just 66% and despite reminders almost one-third of the patients were lost at follow-up. We have no solid explanation for the low response rate, but it appears to be a general problem in surgical research. A recent review demonstrated that the average response rate in surgical research is just 65% when done by postal surveys and as low as 46% after web-based online surveys [13]. In that context, one may even speculate that a response rate of 66% after 16 years was good, but it certainly leaves room for undetected bias. In addition, we also recognize that our original study may have been underpowered to detect relevant clinical differences in both primary and secondary outcomes, but we still believe that the data presented here are relevant because they represent very long-term outcomes from the only randomized trial of its kind. The gender distribution in our study was clearly asymmetric with more women in the R2 group. In theory, this could influence differences in reported outcomes that were unrelated to the operation per se if changes in QoL over a 16-year period is different between genders. It is therefore noteworthy that patients in the R2 group reported lower numerical QoL scores than the R2–R3 group (Figs 1–3) even though this was not significant in our multivariate statistical analysis. In contrast, Smidfelt and Drott [9] reported that women were significantly more satisfied than men despite there was no difference in local effect and suggested that the explanation was a higher prevalence of compensatory sweating in males (85.4% vs 76.2%).

We previously reported on the nature of facial blushing and its surgical management [14]. In short, patients with disabling blushing who seek medical help should be seen by a dermatologist or an internist to exclude important underlying pathologic conditions. Non-surgical treatments should be emphasized as first-line management but options for isolated facial blushing are limited [14]. In addition to medical treatment, psychological therapy such as cognitive-behavioral therapy has successfully improved comping mechanisms for fear of blushing [15, 16]. Provided non-surgical methods fail and there is still an indication for treatment our results demonstrate that thoracoscopic sympathicotomy may effectively cure facial blushing. We emphasize that selection of patients for the operation is important. The key to success is recognizing that the type of blushing most likely to benefit from sympathicotomy is mediated by the sympathetic nerves and is the uncontrollable, rapidly developing blush that is typically elicited by receiving attention from other people [14]. It is also key to success that blushing must be of major concern to the patient, enough to tolerate a substantial amount of compensatory sweating after the operation.

Presented at the EACTS Annual Meeting 2023, Vienna, Austria, 5-7 October 2023.

ACKNOWLEDGEMENTS

The authors thank OPEN It & Data Management, OPEN, Open Patient Data Explorative Network, Odense University Hospital, Region of Southern Denmark. The authors want to acknowledge statistician Søren Möller at OPEN It & Data Management for his valuable input with the statistical analysis in the present study.

CONSENT TO PUBLICATION

This follow-up study was approved by the Regional Council of the Region of Southern Denmark (file number: 21/55832).

FUNDING

No remuneration was paid to the patients, and none of the doctors or researchers involved had any commercial interest in the project. The project team received no financial support.

Conflict of interest: No conflicts of interest for this manuscript but Peter Licht has otherwise received honoraroum for scientific lectures by Medtronic and Ethicon Johnson&Johnson.

DATA AVAILABILITY

The data underlying this article will be shared on reasonable request to the corresponding author.

Author contributions

Finn Amundsen Dittberner: Formal analysis; Investigation; Methodology; Project administration; Writing—original draft; Writing—review & editing. Ole Dan Jørgensen: Methodology; Supervision. Hans Kristian Pilegaard: Supervision. Lars Ladegaard: Supervision. Peter Bjørn Licht: Conceptualization; Project administration; Resources; Supervision; Writing—original draft; Writing—review & editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Paula Moreno, Yuji Shiraishi and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• IQR

  Interquartile range

 
• QoL

  Quality of life