A Novel Endoscopic Finding of Achalasia: “Endoscopic Vertebrae Sign”

Abstract

Introduction

Achalasia is an esophageal motility disorder that significantly impairs quality of life. Recently, peroral endoscopic myotomy has yielded satisfactory treatment outcomes. Although the current gold standard is esophageal high-resolution manometry (HRM), early endoscopic detection is essential for accurate diagnosis. Dilatation of the esophageal lumen during endoscopy is subjective and has not been fully evaluated. We focused on the extramural compression of the vertebrae in the esophagus, which may reflect dilatation of the esophageal lumen, named the endoscopic vertebrae sign (EVS), and examined the possibility of the EVS as a novel endoscopic finding of achalasia.

Methods

Forty-three patients were diagnosed with achalasia using HRM between July 2013 and November 2022. Five who underwent surgical treatment and one for whom esophagogram was unavailable were excluded, resulting in 37 patients in the achalasia group. Among those who underwent comprehensive medical checkups and esophagogastroduodenoscopy screening at our hospital during the same period, all age- and sex-matched individuals were randomly extracted and 74 were set as controls. The rates of EVS, endoscopic esophageal dilatation at the endoscopist’s discretion, and number of visible vertebrae in a single endoscopic view were retrospectively analyzed.

Results

In the achalasia group, the proportion with EVS was 86.5%, with endoscopic esophageal dilatation at the endoscopist’s discretion 76%, and the number of visible vertebrae in patients with EVS was 3.2 ± 0.9. These findings differed significantly in controls: 9.5%, 0%, and 1.4 ± 0.4, respectively (p < 0.001).

Conclusion

EVS may be one of the endoscopic findings associated with esophageal dilatation and suggestive of achalasia.

Introduction

Achalasia is defined as an esophageal motility disorder (EMD) of unknown etiology, characterized by failure of relaxation of the lower esophageal sphincter (LES) and impaired peristaltic movement of the lower esophageal body [1]. The global incidence and prevalence range from 0.03 to 1.63 per 100,000 persons per year and 1.8 to 12.6 per 100,000 persons per year, respectively [2, 3]. Although achalasia is a rare benign disease, symptoms include dysphagia, regurgitation of ingested food, chest pain, and weight loss, which significantly impair quality of life. Therefore, it is important to accurately diagnose achalasia because effective treatments have become available in recent years. A definitive diagnosis of achalasia is made by esophageal high-resolution manometry (HRM), which is the current gold standard test assessing LES relaxation and esophageal body motility [4]. Esophagography is also an important diagnostic modality. However, these modalities are not always performed in every institute. Given that esophagogastroduodenoscopy (EGD) is generally performed as an initial examination for symptomatic patients, endoscopic detection of achalasia is essential for accurate diagnosis of achalasia. According to Descriptive Rules for Achalasia of the Esophagus, 4th Edition [1], endoscopic findings of achalasia include: (1) dilatation of the esophageal lumen, (2) abnormal retention of food and/or liquid remnants in the esophagus, (3) whitish change and thickening of the esophageal mucosal surface, (4) functional stenosis of the esophagogastric junction, and (5) abnormal contraction of the esophagus. Dilatation of the esophageal lumen is one of the typical endoscopic findings suggesting esophageal achalasia; however, its evaluation can be influenced by each endoscopist subjectively. We focused on the extramural compression of the vertebrae in the esophagus which reflects dilatation of the esophageal lumen. This endoscopic finding had the potential to be an objective evaluation method for esophageal dilatation. We named this endoscopic finding the endoscopic vertebrae sign (EVS), a clear hemispherical extramural compression image visible on the posterior esophageal wall. The aims of this study were to evaluate the potential of EVS to be an objective endoscopic evaluation method for esophageal dilatation and to verify the possibility of EVS as a novel endoscopic finding of achalasia.

Methods

Study Design

This was a single-center retrospective case-control study.

Patients

Figure 1 shows the flowchart of this study. From 378 cases in which EGD and esophageal HRM were performed at our hospital between July 2013 and November 2022, 43 patients diagnosed with achalasia types I to III using HRM were extracted. Five cases after surgical treatment and one case for which esophagogram data were unavailable were excluded, resulting in 37 cases (achalasia group). Among those who underwent comprehensive medical checkups and EGD screening at our hospital during the same period, all age- and sex-matched individuals per patient with achalasia were extracted. Initially, one researcher randomly numbered the individuals. Another researcher then randomly selected 2 individuals per patient with achalasia (e.g., 2 age- and sex-matched individuals per patient with achalasia). Thus, 74 patients were randomly selected by age- and sex-matching (control group). Moreover, 44 and 5 patients with esophagogastric junction outflow obstruction (EGJOO) and absent contractility, respectively, diagnosed using HRM during the same period with available endoscopic images, were also extracted for comparison with other EMDs.

Fig. 1.

Flowchart of this study.

High-Resolution Manometry

The catheter was inserted nasally, positioned to extend over both the upper and lower esophageal sphincters, and fixed to the nose using tape. After 5 min of rest, 10 mouthfuls of water (5 mL/mouthful) were placed with the patient in the supine position. At our institute, before October 2021, HRM was performed based on the Chicago classification (CC) version 3.0 [5] protocol, while after that date it was performed based on the CC version 4.0 [6] protocol. In CC version 4.0, additional swallows, such as five mouthfuls of water in the upright position, were performed. HRM was performed using a Starlet^® HRM system (Star Medical, Tokyo, Japan). This system has a catheter with a 36-channel solid-state sensor spaced at 1-cm intervals (Unisensor, Attikon, Switzerland). Esophageal motility was described according to CC version 3.0 or version 4.0. The upper limit of normal for the integrated relaxation pressure was set at 26 mmHg, and the normal range for distal contractile integral was set between 1,000 and 10,000 mmHg · s · cm, which also accounted for the differences in integrated relaxation pressure and distal contractile integral based on the type of HRM catheter used [7].

Esophagogram

An esophagogram was obtained based on the Descriptive Rules for Achalasia of the Esophagus, 4th Edition [1]. Morphologic type on esophagogram was classified into straight (St) type and sigmoid (Sg) type. Dilatation was classified into grades I to III according to the maximum transverse diameter (d) of the esophagus, drawn perpendicularly to the longitudinal axis. Grade I is d <3.5 cm, grade II is 3.5≤ d <6.0 cm, and grade III is d ≥6.0 cm.

Definition of EVS

EVS was defined as one or more clearly recognized hemispherical extramural compressions of the posterior wall of the esophagus. Extramural compression that was not hemispherical was excluded from the EVS. Representative images of patients with and without EVS are shown in Figure 2a, b and Figure 2c, d, respectively.

Fig. 2.

Representative images of endoscopic vertebrae sign (EVS) (a, b) and non-EVS (c, d).

Endoscopic Evaluation

EGD was performed via endoscopy using GIF-Q260, GIF-H260, GIF-H260Z, GIF-XP290N, GIF-H290, or GIF-H290Z (Olympus Co., Tokyo, Japan), with or without sedation. Endoscopic images captured before and closest to HRM were used in this study. Three board-certified fellows from the Japan Gastroenterological Endoscopy Society retrospectively reviewed all endoscopic images of the esophagus without detailed patient information. Representative images of EVS and non-EVS were shown to the three endoscopists before image evaluation. The EVS, esophageal dilatation at the endoscopist’s discretion, the number of visible vertebrae in a single endoscopic view and previously described endoscopic findings of the esophagogastric junction (EGJ) – specifically, the esophageal rosette sign [8] and the gingko leaf sign [9] – were evaluated. Additionally, EVS in EGJOO and absent contractility were also assessed. If the diagnoses of the three endoscopists varied, the final decision was made based on the majority.

Symptom Evaluation

Chief complaints were extracted from medical records. In the achalasia group, Eckardt score [10] was assessed at the time of HRM.

Endpoints

The primary endpoint was set as the rate of presence of EVS. The secondary endpoints were the rate of endoscopic esophageal dilatation at the endoscopist’s discretion and the number of visible vertebrae in a single endoscopic view. These outcomes were compared between the two groups. The interobserver agreement of esophageal dilatation at the endoscopist’s discretion and EVS in the achalasia group was assessed using Fleiss’ kappa. Additionally, the relationship between the number of visible vertebrae in a single endoscopic view and the esophageal diameter on the esophagogram was analyzed as a supplementary result. Furthermore, EVS in EGJOO and absent contractility was also compared with that in the case of achalasia. Moreover, factors associated with EVS were explored by comparing EVS-positive and EVS-negative patients in both the achalasia and control groups. Finally, the proportion of EVS among patients with the esophageal rosette sign and the gingko leaf sign, which could be assessed retrospectively, was evaluated.

Statistical Analysis

χ² and Fisher’s exact tests were used for between-group comparisons of qualitative variables, and the Mann-Whitney U test was used for the comparison of quantitative variables. All statistical analyses were performed using SPSS software for Windows (version 25.0; IBM Corp., Armonk, NY, USA). Statistical significance was set at p < 0.05. Additionally, Fleiss’ kappa was used to assess interobserver agreement.

Ethical Approval

This study was conducted according to the ethical principles of the Declaration of Helsinki for medical research involving human participants. The study protocol was approved by the Ethics Committee of Toranomon Hospital, Japan. The requirement for informed consent was waived because the data were anonymized. The information disclosure document for this study has been published on our hospital’s website.

Results

Patient Characteristics

Patient characteristics are shown in Table 1. The mean age of the 37 patients with achalasia was 58 ± 17 years, with a male-to-female ratio of 16:21, whereas the controls were of the same age, with a male-to-female ratio of 32:42. Body mass index (BMI), chief complaints, Eckardt scores, type of achalasia on HRM, morphologic type on esophagogram, maximum diameter on esophagogram, and dilatation grading on esophagogram are presented in Table 1.

Table 1.

Patient characteristics

​	Achalasia group (N = 37)	Control group (N = 74)
Age, mean, years	58±17	58±17
Male/female ratio, n	16:21	32:42
Body mass index, mean, kg/m2*	22.2±3.8	23.1±3.8
Chief complaint, n	Dysphagia: 34	–
	Chest pain: 2	​
	Regurgitation: 1	​
Eckardt score, mean (points)	3.9±1.8	–
Type of achalasia on HRM, n (Ⅰ:Ⅱ:Ⅲ)	14:18:5	–
Morphologic type on esophagogram, n (St:Sg)	29:8	–
Maximum diameter on esophagogram, mean, mm	39±12	–
Dilatation grading on esophagogram, n (Ⅰ:Ⅱ:Ⅲ)	14:20:3	–

HRM, high-resolution manometry; St, straight type; Sg, sigmoid type.

*p = 0.418.

Endpoints

Table 2 presents the primary and secondary endpoints of this study. In the achalasia group, the proportion of patients with EVS was 86.5%, that of patients with endoscopic esophageal dilatation at the endoscopist’s discretion was 76%, and the number of visible vertebrae in EVS-positive patients was 3.2 ± 0.9. In the control group, the proportion with subjective endoscopic esophageal dilatation was 0%, the proportion with EVS was 9.5%, and the number of visible vertebrae in EVS-positive patients was 1.4 ± 0.4 – all significant differences with the achalasia group (p < 0.001). The maximum number of visible vertebrae in a single endoscopic view was five and two in the achalasia and control groups, respectively. The interobserver agreement of the endoscopic findings in the achalasia group is shown in Table 3. The Fleiss’ kappa value of EVS was 0.63, which was higher than 0.47 of endoscopic esophageal dilatation at the endoscopist’s discretion.

Table 2.

Endpoints

​	Achalasia group (N = 37)	Control group (N = 74)	p value
EVS	86.5% (32/37)	9.5% (7/74)	<0.001
Endoscopic esophageal dilatation at the endoscopist’s discretion	75.6% (28/37)	0%	<0.001
Number of visible vertebrae in a single endoscopic view in EVS-positive patients, mean	3.2±0.9	1.4±0.4	<0.001
Maximum number of visible vertebrae in a single endoscopic view	5	2	–

EVS, endoscopic vertebrae sign.

Table 3.

Interobserver agreement in the achalasia group

​	Fleiss’ kappa value
Endoscopic esophageal dilatation on endoscopist’s subjective discretion	0.47
EVS	0.63

EVS, endoscopic vertebrae sign.

Figure 3 shows the relationship between the number of vertebrae on endoscopy and the dilatation grading on the esophagogram. Dilatation grading on esophagogram was classified as grade I in 14 patients, grade II in 20 patients, and grade III in 3 patients, while the mean number of visible vertebrae was 2.3 ± 1.5 in grade I, 2.9 ± 1.3 in grade II, and 4.4 ± 0.5 in grade III, with significant differences among the three groups (p = 0.015). A scatter plot of the number of visible vertebrae in a single endoscopic view, and the esophageal diameter on the esophagogram is shown in Figure 4. The correlation coefficient between the two is 0.469, indicating a positive correlation.

Fig. 3.

Relationship between number of vertebrae in endoscopy and dilatation grading on esophagogram. Dilatation was classified into grades I to III according to the maximum transverse diameter (d) of the esophagus.

Fig. 4.

Scatter plot of number of visible vertebrae in a single endoscopic view and esophageal diameter on esophagogram.

Table 4 presents the comparison with EGJOO and absent contractility. The proportion of EVS in the achalasia group was significantly higher than that in both the EGJOO and absent contractility group (p < 0.001).

Table 4.

EVS in EGJOO and absent contractility

​	EVS
EGJOO	6.8% (3/44)*
Absent contractility	0% (0/5)*

EVS, endoscopic vertebrae sign; EGJOO, esophagogastric junction outflow obstruction.

*p < 0.001 compared with achalasia.

Tables 5, 6 show the comparison between EVS-positive and EVS-negative patients in both the achalasia and control groups. In the achalasia group, the BMI of EVS-positive patients was significantly lower than that of EVS-negative patients (p = 0.048). Regarding achalasia subtype on HRM, the proportion of type III achalasia was 9.5% (3/32) in EVS-positive patients and 40% (2/5) in EVS-negative patients; however, there was no significant difference among the three subtypes (p = 0.167). Additionally, the mean maximum esophageal diameter was greater in EVS-positive patients than in EVS-negative, although this difference was not statistically significant (p = 0.153). No significant differences in age, sex, or BMI were observed between EVS-positive and EVS-negative individuals in the control group.

Table 5.

Comparison of EVS-positive and EVS-negative patients with achalasia

​	EVS-positive achalasia (N = 32)	EVS-negative achalasia (N = 5)	p value
Age, mean, years	58±17	62±20	0.589
Male/female ratio, n	13:19	3:2	0.634
Body mass index, mean, kg/m2	21.7±3.7	25.2±3.3	0.048
Chief complaint, n	Dysphagia: 30	Dysphagia: 4	0.283
	Chest pain: 1	Chest pain: 1	​
	Regurgitation: 1	Regurgitation: 0	​
Eckardt score, mean (points)	3.9±1.8	3.8±2.1	0.880
Type of achalasia on HRM, n (Ⅰ:Ⅱ:Ⅲ)	13:16:3	1:2:2	0.167
Morphologic type on esophagogram, n (St:Sg)	25:7	4:1	1.000
Maximum diameter on esophagogram, mean, mm	40±12	31±14	0.153
Dilatation grading on esophagogram, n (Ⅰ:Ⅱ:Ⅲ)	11:18:3	3:2:0	0.489

EVS, endoscopic vertebrae sign; HRM, high-resolution manometry; St, straight type; Sg, sigmoid type.

Table 6.

Comparison of EVS-positive and EVS-negative controls

​	EVS-positive controls (N = 7)	EVS-negative controls (N = 67)	p value
Age, mean, years	61±17	58±17	0.506
Male/female ratio, n	4:3	30:37	0.696
Body mass index, mean, kg/m2	22.8±3.8	23.2±3.9	0.561

EVS, endoscopic vertebrae sign.

The proportion of EVS among patients with previously described endoscopic findings of the EGJ is shown in Table 7. The esophageal rosette sign was positive in 22 patients, and gingko leaf sign was positive in 11 patients, while 4 patients were negative for both findings. EVS was positive in 90.9% of esophageal rosette sign-positive patients, 81.8% of gingko leaf sign-positive patients, and 75.0% of patients who were negative for both.

Table 7.

Proportion of EVS-positive patients according to other endoscopic findings of the EGJ associated with achalasia

​	Esophageal rosette sign-positive (N = 22)	Gingko leaf sign-positive (N = 11)	Both negative (N = 4)
EVS-positive patients, n (%)	20 (90.9)	9 (81.8)	3 (75.0)

EVS, endoscopic vertebrae sign; EGJ, esophagogastric junction.

Discussion

The present study demonstrated that the number of visible vertebrae in a single endoscopic view and esophageal dilatation on esophagogram were proportional, and EVS was seen in 86.5% of patients with achalasia, a significant difference (p < 0.001) from the control group. In the interobserver agreement, the Fleiss’ kappa value of the EVS was 0.63, which was higher than that of endoscopic esophageal dilatation at the endoscopist’s discretion (0.47).

EMDs, including achalasia, can cause dysphagia or chest pain and are diagnosed using esophageal HRM or esophagography. Although EGD is performed to rule out organic diseases such as esophageal cancer or gastroesophageal reflux disease, typical endoscopic findings that could indicate the presence of EMDs are sometimes observed during EGD [11]. Moreover, EGD plays a pivotal role in the exclusion of pseudoachalasia and other mechanical obstructions that may result in symptoms similar to achalasia [4].

Endoscopic detection of achalasia is sometimes difficult because the esophageal mucosa may appear normal in early or type III achalasia, which can lead to a delayed diagnosis [12]. Previously, it had been reported that characteristic endoscopic findings of achalasia were present in only approximately half of the patients [13]. Recently, several endoscopic features have been identified in patients with achalasia. Iwakiri et al. [8] reported the endoscopic features of non-visibility of the esophageal palisade vessels and the appearance of rosette-like esophageal folds, “esophageal rosette” in the lower esophagus during deep inspiration as characteristic findings of primary achalasia. Minami et al. [14] reported the presence of the longitudinal superficial wrinkles of esophageal mucosa, called the “pinstripe pattern.” Gomi et al. [15] reported the “champagne glass sign,” defined when the distal end of the LES relaxation failure is proximal to the squamocolumnar junction and the squamocolumnar junction is dilated in the retroflex view. Hoshikawa et al. [9] reported that some patients with achalasia without “esophageal rosette” had the morphology that appeared to be gingko leaf-shaped without being able to see the full extent of esophageal palisade vessel or the rosette-like folds, called “gingko leaf sign.” These findings may serve as useful indicators of achalasia in some patients. It is important to fully understand the characteristic endoscopic findings and observe them carefully when achalasia is suspected.

Few studies have reported the dilatation of the esophageal lumen, which is a diagnostic feature of EGD in patients with achalasia, in detail. According to the Descriptive Rules for Achalasia of the Esophagus, 4th Edition [1], the presence or absence of dilatation of the esophageal lumen was referred to; however, details such as the degree or evaluation method were not provided. In the present study, the number of visible vertebrae in a single endoscopic view was related to the esophageal diameter on the esophagogram. In the interobserver agreement of EVS and endoscopic esophageal dilatation at the endoscopist’s discretion, the Fleiss’ kappa value was 0.63 with substantial agreement and 0.47 with moderate agreement, based on previous literature [16]. This result showed that the EVS is a more objective evaluation method for esophageal dilatation in EGD. Additionally, since EVS was seen significantly more in the achalasia than the control group, this finding has the possibility to be one of the findings to suspect achalasia. However, 9.5% of the control group also had EVS. This might be due to the patient’s bone structure and body shape, although no significant difference in BMI was observed between EVS-positive and -negative controls in this study. The average and maximum number of visible vertebrae in EVS-positive patients were 3.2 ± 0.9 and 5 in the achalasia group, 1.4 ± 0.4 and 2 in the control group, respectively. We observed a significant difference in the number of visible vertebrae between the two groups, which might reflect esophageal dilatation in the achalasia group. In the comparison with EGJOO and absent contractility, EVS in the achalasia group was significantly higher. However, the number of absent contractility was too small and inadequate to confirm the difference between two groups.

Comparisons between EVS-positive and EVS-negative patients in the achalasia group revealed that EVS-positive patients had significantly lower BMI. This difference may be the result of a higher proportion of type III achalasia in the EVS-negative group. Type III achalasia has been reported to be associated with less weight loss compared with other subtypes [17], and esophageal dilatation on esophagogram is typically milder than in other subtypes [18]. In contrast, no significant differences in patient background, including BMI, were observed between EVS-positive and EVS-negative individuals in the control group. Based on these findings, EVS might be particularly effective in patients with type I and II achalasia, who typically have lower BMI and more prominent esophageal dilatation.

Regarding the association with previously described endoscopic findings of the EGJ, EVS was positive in 90.9% of esophageal rosette sign-positive patients and 81.8% of gingko leaf sign-positive patients. In contrast, EVS was positive in 75.0% of the patients who were negative for both findings in this study. We consider that EVS may serve as an endoscopic finding suggestive of achalasia even in some patients without specific endoscopic findings of the EGJ associated with achalasia.

Although achalasia significantly impairs quality of life, therapeutic intervention has enabled good treatment outcomes because of recent advances in endoscopic treatment when the condition is diagnosed accurately. An endoscopic treatment method known as peroral endoscopic myotomy (POEM) was introduced in 2010 [19]. Since then, POEM has become one of the most important, minimally invasive treatment options for achalasia. Many studies have reported favorable outcomes with a high safety profile and efficacy for POEM, and this is now endorsed by major endoscopic societies [20]. A large multicenter retrospective study in Japan reported that response rates (Eckardt scores ≤3) were 95.1% at 3 months postoperatively and 94.7% at 1 year postoperatively [21].

The present study had some limitations. First, this was a retrospective study with a small sample size conducted at a single institute. Second, endoscopic images, including EVS, were retrospectively reviewed. EVS can be influenced by certain factors, such as the amount of insufflation, and its reproducibility has not been verified in this study. Third, esophagogram data in the control group were lacking. Finally, symptoms in the control group were not evaluated. However, to the best of our knowledge, this is the first report of endoscopic findings reflecting esophageal dilatation. As little information about the endoscopic findings related to esophageal dilatation has been reported, such information may be helpful in daily practice. Additionally, more information regarding endoscopic findings suggestive of achalasia is desirable because POEM can be performed accurately when achalasia is diagnosed. Further accumulation of similar cases and a prospective study are required to verify the usefulness of the EVS as an endoscopic finding for achalasia. Furthermore, additional effect of EVS in diagnosing achalasia and usefulness of diagnosis especially in early or type III achalasia requires further investigation in the future.

In conclusion, a novel endoscopic finding “EVS” was significantly more frequent in the patients with achalasia than the control group. EVS has the potential to be an endoscopic finding associated with esophageal dilatation and a suggestive finding of achalasia. The EVS may be a useful evaluation method in daily practice.

Statement of Ethics

The study protocol was reviewed and approved by the Ethics Committee of Toranomon Hospital (Approval No. 1625). The requirement for informed consent was waived because the data were anonymized, which was approved by the Ethics Committee of Toranomon Hospital.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funding.

Author Contributions

Concept and design and drafting of the manuscript: Y.O. Acquisition, analysis, and interpretation of data and review and approval of the manuscript: Y.O., Y.S., K.N., H.O., D.K., and S.H. Endoscopic evaluation: Y.O., Y.S., and K.N. Supervision: D.K. and S.H.

Funding Statement

This study was not supported by any sponsor or funding.

Data Availability Statement

The data used to support the findings of this study are not publicly available due to privacy reasons but are available from the corresponding author upon reasonable request.