Retrograde Cricopharyngeal Dysfunction: Patient Characteristics and Outcomes in Australia

ABSTRACT

Objective

Retrograde cricopharyngeal dysfunction (RCPD) is characterised by an inability to burp. Typical first‐line management involves injection of Botulinum A toxin (BonT‐A) into the cricopharyngeus. We present our series of 109 RCPD patients.

Methods

A retrospective chart review was conducted between January 2018 and November 2024. Demographics and clinical data were extracted. From late 2021 onwards, we also started collecting scores from a 7‐point Likert questionnaire based on the six cardinal symptoms of RCPD (RCPD‐Q in short).

Results

A109 patients were identified. Sixty‐two (56.9%) were female. Mean age was 30.0 (range: 14–72). The majority (91.7%) were of Caucasian ethnicity. Eighty patients completed the RCPD‐Q at their initial consultation, with a mean score of 27.6/36 (range: 13–36). Seventy‐four patients underwent initial treatment. At first follow‐up within 4 weeks, 53 patients (71.6%) had complete resolution, 11 (14.9%) had partial resolution and seven (9.5%) had no change in symptoms. In the 30 patients who completed follow‐up at > 3 months after treatment, 17 patients (56.7%) reported complete resolution, three patients (10%) reported partial resolution and 10 patients (33.3%) reported no resolution or reversion to baseline symptoms.

Conclusion

Younger age, treatment under general anaesthesia, and higher doses of BonT‐A were associated with higher rates of success.

Level of Evidence

Level 3.

1. Introduction

Retrograde cricopharyngeal dysfunction (RCPD) is a relatively new diagnosis characterised by an inability to burp [1]. The pathophysiology of this condition is thought to be a failure of the Cricopharyngeus muscle (CP) relaxation when there is distension of the upper oesophagus; the CP otherwise relaxes normally during swallowing [2]. The typical first‐line management of this condition is injection of Botulinum A toxin (BonT‐A) into the cricopharyngeus (CP) muscle [3]. BonT‐A injection can be performed under general anaesthetic via direct esophagoscopy, or under in‐office with EMG guidance. In‐office injection can be performed via an anterior trans‐airway approach first described by Wajsberg et al. [4], or a lateral trans‐cervical approach first described in Doruk et al. [5]. There is, thus far, no well‐established guideline as to the optimum method and dose of injection.

As there are also no formal diagnostic criteria for RCPD, treatment success is not well‐defined, and in most published series, it is only reported as a binary yes or no outcome. Some have reported that patients who have undergone BonT‐A injection for RCPD have lasting success in symptom resolution beyond the anticipated 3‐month duration of action [3]. However, it is unclear whether patients' symptoms remain completely in remission or whether there is some degree of recurrence.

In this series, we characterise the first 109 patients presenting with symptoms suggestive of RCPD to a tertiary Laryngology practice. We report our use of a patient‐reported questionnaire to assess the six cardinal symptoms of RCPD, with the aim of quantifying patient responses to treatment, and also present our treatment outcomes associated with various demographic factors and treatment modalities.

2. Methods

A retrospective chart review was conducted of all RCPD patients seen at a single institution between January 2018 and November 2024. Demographics and clinical data, including treatment information, were extracted from the medical records. From late 2021 onwards, we also started collecting questionnaire data consisting of six questions on the cardinal symptoms of RCPD. Scores from this RCPD questionnaire (RCPD‐Q) were used to characterise the type and severity of baseline symptoms, as well as to quantify response to treatment over time. Approval was obtained from the University of Sydney Human Research Ethics Committee (2023/HE000647).

Treatment was overseen by two fellowship‐trained Laryngologists (DN, TS), assisted by a clinical fellow. Treatment under general anaesthesia (GA) was performed in the operating theatre and involved direct esophagoscopy with either a Holinger‐Benjamin Slimline Diverticuloscope or Dedo Laryngoscope and injection of either Onabotulinumtoxin‐A (Botox Allergan, Irvine, USA) or IncobotulinumtoxinA (Xeomin Merz Therapeutics, Frankfurt am Main, Germany) into four equal and equidistant aliquots, along the posterior half of the CP muscle. Treatment under local anaesthesia (LA) was performed in‐office and involved electromyography (EMG) guided injection into one side of the CP muscle, either via a lateral transcervical [5] or anterior trans‐airway approach [4] as previously described. Patients undergoing treatment under LA were counselled on the possible need for multiple procedures before an evident clinical effect. Consistent with the recent series published by Kutler et al., contralateral injections performed within 8 weeks of the initial in‐office injection were considered part of the same treatment round [6].

Outcomes measured include RCPD‐Q scores and patient self‐reported resolution of symptoms (complete resolution, partial resolution, or no change) in the short term (< 4 weeks) and long term (> 3 months). Descriptive statistical analysis was performed with Microsoft Excel, and statistical analysis was performed with Orange Data Analytics v3.34.0. Univariate analysis was performed for patients who completed the relevant follow‐up visits using the chi‐squared test for categorical variables and one‐way ANOVA for continuous variables.

3. Results

A 109 patients (62 female, 47 male) were identified and met the inclusion criteria. The mean age was 30.0 (range, 14–72; SD, 10.1). The majority (91.7%) were of Caucasian ethnicity, with six individuals of East Asian descent, two of Middle Eastern descent, and one of South Asian descent (Table 1).

TABLE 1.

Patient demographics.

	Total (n = 109)
Gender
Male	47 (43.1%)
Female	62 (56.9%)
Age (mean)	30.0 (14–72, SD: 10.1)
Ethnicity
Caucasian	100 (91.7%)
East Asian	6 (5.5%)
Middle Eastern	2 (1.8%)
South Asian	1 (0.9%)

Eighty patients completed the RCPD‐Q at their initial consultation. Each question in the RCPD‐Q was rated on a 7‐point Likert scale (0–6) for a maximum score of 36. The mean initial RCPD‐Q score was 27.6 (range: 13–36, standard deviation 5.02). The questionnaire and sub‐scores of each individual domain are presented in Table 2. The most prominent symptom was the inability to burp, followed by abdominal bloating, then socially awkward noises in the upper abdomen, chest and throat.

TABLE 2.

RCPD questionnaire and breakdown at presentation.

	Mean	Range	Standard deviation
Inability to burp	5.6	3–6	0.7
Abdominal bloating/discomfort/nausea	5.5	4–6	0.69
Socially awkward noises	5.0	0–6	1.4
Excessive flatulence	4.6	0–6	1.6
Difficulty vomiting	4.0	0–6	1.8
Social Inhibition due to the above symptoms	2.9	0–6	2.3

Seventy‐four patients underwent initial treatment at our centre. Overall, 53 out of 74 (71.6%) patients reported complete resolution after initial treatment, 11 (14.9%) reported partial resolution and seven (9.5%) reported no change. At follow‐up beyond 3 months, the proportion of patients with complete resolution had dropped to 56.7%.

3.1. Treatment Under GA

Sixty‐eight patients (91.9%) underwent initial treatment under GA. Access to the CP muscle was inadequate in two patients, and the procedure was abandoned. Sixty‐six patients (89.2%) completed initial treatment under GA. Mean dose of BonT‐A given under GA was 81.4u (range: 50–100u). At short term follow‐up (< 4 weeks), 48 patients (72.7%) reported complete resolution of symptoms with a mean RCPD‐Q score of 5.0 (SD: 3.0), 11 patients (16.7%) reported partial resolution with a mean post‐op RCPD‐Q score of 19.2 (SD: 3.8), four patients (6.0%) reported no improvement with a mean post‐op RCPD‐Q score of 24.3 (SD: 4.5). Three patients were pending or lost to follow‐up. Thirty‐three patients (50.0%) reported mild complications, the majority of which were self‐limiting dysphagia, which resolved in 1–2 weeks.

Thirty patients completed a follow‐up visit more than 3 months after the initial GA procedure, with 17 patients (56.7%) reporting complete resolution, three patients (10.0%) reporting partial resolution and 10 patients (33.3%) reporting no resolution or reversion to baseline symptoms.

Nine patients who had either no or partial resolution of symptoms underwent further procedures. Three out of nine had a repeat procedure under GA, and six of nine had a repeat procedure under LA. Seven out of nine of these patients reported complete resolution of symptoms after the repeat procedure. No patients underwent oesophageal dilation at any point. One patient failed a total of three BonT‐A procedures. This patient was counselled that RCPD may not be the correct diagnosis. However, as the patient had exhausted other treatment options with Gastroenterology as well and was still very bothered by his symptoms, we eventually proceeded with an open cricopharyngeal myotomy, which unfortunately also did not alleviate his symptoms.

3.2. Treatment Under LA

Eight patients (10.8%) completed initial treatment under LA, including the two patients who previously failed access under GA. Mean dose of BonT‐A given under LA was 45.6u (range: 25–85u). At short‐term follow‐up, five patients (62.5%) reported complete resolution of symptoms with a mean RCPD‐Q score of 4.8. Three patients (37.5%) reported no improvement in symptoms with a mean RCPD‐Q score of 28.6. Three patients had mild complications including vocal fold paresis (n = 2) and self‐limiting dysphagia (n = 1).

On univariate analysis (Table 3), factors associated with initial treatment success included younger age at presentation (Figure 1), initial treatment under GA, as well as a higher dose of BonT‐A (Figure 2). No factors were found to be associated with lasting treatment success (more than 3 months) in our series.

TABLE 3.

Univariate analysis of factors predicting initial and lasting success.

Variable	Initial success (< 4 weeks)	No success (< 4 weeks)	Lasting success (> 3 months)	No lasting success (> 3 months)
Gender	Male = 29	Male = 5	Male = 7	Male = 5
	Female = 35	Female = 2	Female = 16	Female = 7
	χ 2 3.42 (p = 0.181)		χ 2 0.97 (p = 0.615)
Age	ANOVA F = 3.831 (p = 0.027)		ANOVA 3.260 (p = 0.051)
Treatment type	GA = 59	GA = 4	GA = 20	GA = 10
GA vs. LA	LA = 5	LA = 3	LA = 3	LA = 2
	χ 2 8.42 (p = 0.015)		χ 2 0.56 (p = 0.754)
BonT‐A dose	ANOVA F = 4.360 (p = 0.017)		ANOVA 1.334 (p = 0.278)
BonT‐A type	OnabotulinumtoxinA = 33	OnabotulinumtoxinA = 5	OnabotulinumtoxinA = 11	OnabotulinumtoxinA = 9
	IncobotulinumtoxinA = 32	IncobotulinumtoxinA = 2	IncobotulinumtoxinA = 12	IncobotulinumtoxinA = 3
	χ 2 1.91 (p = 0.753)		χ 2 3.79 (p = 0.436)

Note: Statistically significant associations highlighted in boldface.

Abbreviations: BonT‐A, Botulinum neurotoxin Type A; GA, general anaesthetic; LA, local anaesthetic.

FIGURE 1.

Correlation of age at presentation with short term success in treatment of retrograde cricopharyngeal dysfunction.

FIGURE 2.

Correlation of initial dose of Botulinum neurotoxin Type‐A with short‐term success in treatment of Retrograde cricopharyngeal dysfunction including local and general anaesthesia modalities.

4. Discussion

In this series, we describe our Australian cohort of RCPD patients and report our results with BonT‐A injection to the CP muscle. Our cohort is similar in terms of age, gender and ethnic distribution to previously reported series from North America and Europe. Since Bastian and Smithson first drew attention to RCPD in 2019 [1], early series primarily reported qualitative, descriptive accounts of symptomatology. Although this was invaluable in helping to characterise the disease entity, a quantitative assessment is required to standardise diagnosis and systematically assess and document treatment outcomes.

Doruk et al. [7] had earlier presented their use of an 11‐question questionnaire to characterise RCPD patients at presentation and to assess and document response to treatment. In our series, we present a shorter six‐question questionnaire, which places a greater weight on the physical symptoms of RCPD (accounting for five out of six questions). With the reduced cognitive load of a shorter questionnaire, we have opted for a seven‐point Likert scale, which offers advantages in terms of greater precision and improved discriminatory power for statistical analysis.

Our patients had a mean score of 27.6 out of a possible 36 prior to treatment. The five physical domain questions consistently had high mean sub‐scores, with the lowest being 4.0, highlighting the substantial physical burden of RCPD. In contrast, the only psychosocial question (social inhibition due to symptoms) had a relatively lower mean sub‐score of 2.9. Although this might suggest that RCPD patients are not socially affected by their symptoms, qualitative research on the quality of life impact by Miller et al. suggests otherwise [8]. It is more likely that the phrasing of our question did not capture the true extent of the psychosocial burden experienced by our patients. Future efforts should focus on formally developing and validating a patient‐reported questionnaire for RCPD that can be used across institutions to standardise symptom assessment and reporting.

To evaluate treatment success, we utilised both RCPD‐Q scores and patient‐reported success on a three‐point scale: complete resolution, partial resolution and no change. After initial treatment, 71.6% of patients reported complete resolution. However, at follow‐up beyond 3 months, this proportion declined to 56.7%. These findings contrast some initial series that reported initial and sustained success exceeding 95% [1, 3, 9]. Notably, those studies presented success using a binary (yes/no) approach, which we hypothesise reduced the ability to discern nuances in outcome. Our success rates align with more recent series that adopt a more detailed outcome scale [10]. In our cohort, we observed a drastic improvement in mean RCPD‐Q scores (from 27.6 to 5.0) in patients who reported complete resolution at initial follow‐up within a month. In contrast, patients reporting partial resolution only saw a modest improvement (from 27.6 to 19.2). This differentiation highlights the importance of employing more nuanced and quantitative measures to assess treatment effectiveness.

Younger age at presentation, higher initial BonT‐A dosage, and undergoing initial treatment under GA were significantly associated with higher rates of initial success. The prevailing hypothesis for sustained symptom relief beyond the typical 3–4 month duration of action of BonT‐A is that patients may ‘re‐train’ their CP muscle, possibly developing new neural pathways that support a lasting burp reflex even after the effects of BonT‐A subside. Although statistical significance was established for the association between age and initial success, a trend towards significance was observed when examining age and lasting success (F = 2.155, p = 0.086). We hypothesise that with a larger sample size, this trend may become statistically significant. Younger patients may have better neural plasticity, which could enhance their ability to successfully develop and maintain new neural pathways for a burp reflex. These findings lend support to the CP retraining/neural pathway development hypothesis and highlight the value of encouraging patients with RCPD to seek treatment earlier.

The association of higher BonT‐A dosage with success is not surprising. The need for higher doses (in the range of 75–100u) to optimise success has also been anecdotally reported by other authors [3, 9] and formally reported in a recent meta‐analysis [11]. Initially, we administered 50u of BonT‐A under GA and offered repeat procedures for non‐responding patients. However, over time, we increased the initial dose over the years to increase the likelihood of success with the first procedure. Although we found an association between the route of injection (direct esophagoscopy under GA vs. EMG‐guided injection under LA) and the rate of initial treatment success, this may be confounded by differences in BonT‐A dosage. In our series, injections under GA typically involved doses about twice as high as those under LA and were administered bilaterally. In contrast, LA injections were administered unilaterally to minimise the risk of bilateral vocal fold palsy resulting from inadvertent diffusion into the posterior cricoarytenoid muscle. Although multivariate analysis could help clarify the true association, it was not feasible in our study due to the limited sample size.

The primary limitation of our study is its retrospective nature. Over the years, as new literature became available, we adapted and refined our practice. This explains the variability in treatment protocols and outcome measures across our cohort. Although our study represents one of the larger RCPD series in the literature, the sample size remains relatively small, limiting the robustness of statistical analysis.

5. Conclusion

We present our series of 109 patients with suspected RCPD, detailing patient characteristics, treatment modalities and response to treatment. Higher treatment success was associated with younger age at presentation, higher initial doses of BonT‐A, and treatment performed under GA. This study contributes to the growing body of evidence on RCPD and its management. The RCPD‐Q provided a quantitative outcome measure. To enable meaningful comparisons of different treatment protocols and their outcomes, further refined quantitative outcome measures should be developed and validated.

Funding

The authors have nothing to report.

Conflicts of Interest

The senior author (A/Prof. D. Novakovic) and second author Dr. Sarah Emmett are specialty editors of ANZJS, but had no role in the review process for this article.

Leong Z. H., Emmett S., Sandham K., Stewart T., and Novakovic D., “Retrograde Cricopharyngeal Dysfunction: Patient Characteristics and Outcomes in Australia,” ANZ Journal of Surgery 96, no. 3 (2026): 602–606, 10.1111/ans.70518.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.