Efficacy and adverse reactions of endoscopic submucosal dissection and endoscopic submucosal tunnel dissection in early esophageal cancer patients

Abstract

BACKGROUND

Esophageal cancer remains a notable global health burden. Endoscopic submucosal dissection (ESD) is a standard treatment for early esophageal cancer, but it has limitations, including risk of incomplete resection and adverse events. Endoscopic submucosal tunnel dissection (ESTD) is a new technique that may offer advantages.

AIM

To compare the efficacy and safety of ESD and ESTD in patients with early esophageal cancer.

METHODS

In this retrospective study, 257 patients were divided into ESD (n = 156) and ESTD (n = 101) groups. The primary endpoints were efficacy (en bloc resection, R0 resection, and recurrence rates) and adverse reactions. The secondary endpoints included operative indicators, inflammatory markers (interleukin-8, tumor necrosis factor-alpha, and high-sensitivity C-reactive protein) measured preoperatively and on postoperative days (3 and 7), and quality of life (General Quality of Life Inventory-74 assessed preoperatively and at 6 months). Multivariate logistic and linear regression models were used to adjust for potential confounders, including demographics and tumor characteristics.

RESULTS

The R0 resection rate in the ESTD group (96.04%) was substantially higher than that in the ESD group (89.10%). The three-year recurrence rate was lower in the ESTD group (4.95%) compared with that in the ESD group (12.82%). The overall incidence of adverse reactions in the ESTD group was lower than that in the ESD group (7.92% vs 17.31%). Compared with ESD, ESTD was associated with a larger dissection area (1548.29 vs 1395.76), higher dissection speed (18.85 vs 16.34), and less intraoperative bleeding (19.42 vs 26.83). The postoperative levels of interleukin-8, tumor necrosis factor-alpha, and high-sensitivity C-reactive protein were consistently low in the ESTD group. At six months, the patients with ESTD exhibited much higher scores across all quality of life domains (all P < 0.05) compared with those in the ESD group. Multivariate analysis confirmed that ESTD independently improved R0 resection [adjusted odds ratios (aOR) = 3.21], reduced three-year recurrence (aOR = 0.31), lowered adverse events (aOR = 0.40), improved the operative indicators, lowered inflammation, and increased the patients’ quality of life (all P < 0.05).

CONCLUSION

ESTD is superior to ESD in treating early esophageal cancer, and it has higher rates of curative resection and lower long-term recurrence and adverse events. It improves operative efficiency, attenuates postoperative inflammation, and enhances quality of life.

Key Words: Endoscopic submucosal dissection; Endoscopic submucosal tunnel dissection; Early esophageal cancer; Treatment outcome; Adverse reactions

Core Tip: This retrospective study provides comparative evidence on the use of endoscopic submucosal tunnel dissection (ESTD) vs conventional endoscopic submucosal dissection in treating early esophageal cancer. We found that ESTD is associated with a high R0 resection rate, low recurrence, and few adverse reactions. Moreover, ESTD demonstrates superior surgical efficiency (short procedure time and reduced bleeding) and attenuates postoperative inflammatory responses, which correlate with improved long-term quality of life. These results position ESTD as a potentially more effective and less traumatic minimally invasive option compared with other approaches, warranting further prospective validation.

INTRODUCTION

Esophageal cancer remains a critical global health burden, with early detection and effective treatment being crucial for improving patient outcomes. Early esophageal cancer, typically confined to the mucosal layer, offers an opportunity for curative endoscopic interventions[1,2]. Endoscopic submucosal dissection (ESD) has become a standard approach for removing early-stage lesions, providing high rates of en bloc resection and enabling precise histopathological assessment. However, ESD has limitations, including the risk of incomplete resection and adverse events such as perforation and bleeding[3-5]. These challenges have spurred interest in alternative techniques that can enhance efficacy while minimizing complications.

Endoscopic submucosal tunnel dissection (ESTD) represents a novel advancement in endoscopic therapy for early esophageal cancer. This technique involves creating a submucosal tunnel from a small entry incision, allowing for controlled and extensive dissection along the lesion. The tunneling approach may offer several advantages over traditional ESD, including enhanced visualization of the tumor margins, reduced trauma to surrounding tissues, and potentially low rates of adverse reactions. By maintaining the structural integrity of the esophageal wall, ESTD could minimize the risk of complications associated with multiple entry points required in ESD[6-8]. Understanding the potential benefits and limitations of ESTD compared with ESD is essential for optimizing clinical practice.

Early esophageal cancer presents unique challenges because of its anatomical location and the need for precise resection to achieve optimal outcomes. The goal of endoscopic therapy is not only to remove the visible tumor but also to ensure clear margins to prevent recurrence. Although ESD has demonstrated effectiveness in achieving these goals, the increasing complexity of lesions and the rising incidence of adverse events necessitate the exploration of alternative approaches. ESTD’s ability to create a continuous submucosal tunnel provides a platform for thorough, meticulous dissection, which can improve the completeness of resection and reduce the likelihood of residual diseases. Moreover, the minimally invasive nature of ESTD may accelerate recovery and improve patient comfort[9-11].

Comparative studies that evaluated the efficacy and safety of different endoscopic techniques for early esophageal cancer are limited, particularly regarding ESTD. Most existing data are from adult populations treated with ESD, leaving gaps in knowledge about the relative merits of ESTD[12,13]. A comprehensive evaluation of ESTD vs ESD in patients with early esophageal cancer is necessary to guide clinical decision-making. Such a study would provide valuable insights into the comparative outcomes of the two approaches, helping clinicians choose the ideal treatment for their patients. Addressing these knowledge gaps is crucial for optimizing treatment strategies and improving patient care in early esophageal cancer management. This study aims to fill this gap by comparing the efficacy and adverse reactions of ESD and ESTD in patients with early esophageal cancer, thereby informing optimal therapeutic strategies for this patient population.

MATERIALS AND METHODS

Inclusion and exclusion criteria

A retrospective analysis was conducted on 257 patients with early esophageal cancer who underwent endoscopic treatment at our hospital from April 2018 to March 2022. Depending on the treatment method, the 257 patients were divided into ESD (n = 156) and ESTD (n = 101) groups.

The inclusion criteria were as follows: (1) Diagnosis of early esophageal cancer confirmed by endoscopic biopsy and histopathological examination in accordance with the relevant diagnostic criteria of the Japanese Esophageal Society Guidelines for Diagnosis and Treatment of Esophageal Cancer[14]; (2) Imaging studies and endoscopy confirmed that the lesion was confined to the mucosal layer, with an intact submucosal layer and without regional lymph node or distant metastasis; (3) Assessed as Tis or T1a stage in accordance with the American Joint Committee on Cancer’s Tumor, Node, and Metastasis Staging System[15]; (4) Met the criteria for ESD established by Japanese Esophageal Society and the European Society of Gastrointestinal Endoscopy[14,16]; and (5) Complete clinical data, surgical records, and postoperative follow-up data.

The exclusion criteria were as follows: (1) Presence of severe cardiopulmonary insufficiency, hepatic or renal dysfunction, or other underlying diseases; (2) History of malignancy at other sites or concurrent malignancies; (3) Allergy to the dye used in the procedure, anesthetic drugs, or submucosal injection agents; (4) Previous history of endoscopic procedures or surgical operations on the esophagus; (5) Presence of moderate to severe esophageal varices; (6) Presence of active systemic infection or recent (within two weeks) use of anti-inflammatory medications (excluding low-dose aspirin for cardiovascular prophylaxis); (7) Recent use of anticoagulants or antiplatelet drugs; and (8) Lesions with a circumferential extent greater than 75% or lesions centered within 2 cm of the dentate line at the esophagogastric junction.

Ethical statement

This study strictly adhered to the Declaration of Helsinki and related ethical guidelines for medical research. It received formal approval from the Medical Ethics Committee of Qingdao Central Hospital. In accordance with the Ethical Review Measures for Biomedical Research Involving Human Subjects regarding the ethical requirements for retrospective studies and considering the characteristics of this study (no direct intervention, low risk, and adequate protection of subject privacy), the Medical Ethics Committee unanimously agreed to waive the informed consent procedure after review.

Treatment method

Preoperative preparation: All patients underwent preoperative tests, including complete blood count, coagulation profile, liver and kidney function tests, and serum electrolyte level assessments. Endoscopic ultrasound (GF-UCT180, Olympus, Japan) was performed to determine the extent and depth of lesion invasion. Patients routinely fasted for 6-8 hours before the procedure. Prophylactic antibiotics were administered intravenously 30 minutes prior to the operation, followed by endotracheal intubation and general anesthesia.

Surgical procedure: All procedures in this study were performed by a single senior endoscopist with over 10 years of clinical experience and who had completed more than 200 ESD and 50 ESTD procedures prior to the study period, thereby minimizing the influence of the learning curve. The ESD procedure was as follows: A gastroscope (GIF-Q290, Olympus, Japan) was inserted orally into the esophagus. The lesion location was identified using standard white-light endoscopy and narrow-band imaging. The morphology of the intraepithelial papillary capillary loop was observed via magnifying endoscopy with narrow-band imaging, and iodine staining was used to delineate the lesion margins. Electrocautery markings were made 4-5 mm outside the lesion margin by using a needle-type knife (KD-1V, Olympus, Japan). As required, 0.4% sodium hyaluronate (Approval No. HJ20170211, Meiji Seika Pharma Co., Ltd., Japan) was injected submucosally to elevate the lesion. The mucosa was incised lateral to the anal-side marking line by using a needle-type knife, and the surrounding mucosa was cut with an IT knife (KD-611U, Olympus, Japan). The lesion tissue was gradually dissected. During the procedure, a hook knife (KD-620QR, Olympus, Japan) and a dual knife (KD-655U, Olympus, Japan) were employed as needed to assist in dissection. After lesion resection, the wound was carefully examined for any injury to the muscularis propria, active bleeding, or exposed vessel stumps. When necessary, thorough hemostasis was achieved using an IT knife, argon plasma coagulation (APC 2, ERBE, Germany), or hemostatic forceps (FD-410 LR, Olympus, Japan).

For the ESTD procedure, the steps of scope insertion, lesion localization, boundary marking, submucosal injection, and postoperative wound management were similar to those for ESD. The difference was in the additional step of establishing a submucosal tunnel in ESTD. After submucosal injection, an entry point was created lateral to the anal-side marking line. An IT knife was used to incise the mucosa around the lesion, establishing a submucosal tunnel between the oral-side mucosa and the muscularis propria. The lesion tissue was gradually dissected along the tunnel. When the tunnel met the opening at the anal side, the entire lesion was completely resected.

Postoperative management and follow-up: Postoperatively, patients were asked to routinely fast for 12-48 hours. A nasogastric tube was used for gastrointestinal decompression, and vital signs were closely monitored. The patients were observed for symptoms, such as fever, cough, sputum production, chest pain, dyspnea, hematemesis, and melena. Physical examinations were conducted to check for signs of peritoneal irritation and subcutaneous emphysema, and the volume and characteristics of the fluid in the nasogastric tube were monitored. Routine postoperative treatments included acid suppression, fluid replacement, nutritional support, and anti-infective therapy. On the third day after surgery, the patients were started on a liquid diet.

A follow-up plan was established, with endoscopic examinations scheduled at 1, 3, 6, and 12 months postoperatively and with biopsy if necessary. For patients without residual lesions or recurrence, annual follow-up examinations were scheduled thereafter. For those with residual lesions or recurrence, further treatment decisions, including additional endoscopic therapy or surgical intervention, were made on the basis of the specific condition assessed by the medical team.

Observation indicators: The observation indicators in this study were divided into primary and secondary endpoints. The primary endpoints were efficacy and adverse reactions, and the secondary endpoints included operative indicators, inflammatory markers, and quality of life: (1) Efficacy: This indicator includes the en bloc resection, R0 resection, and recurrence rates. En bloc resection refers to the removal of the specimen as a single, intact piece of tissue. R0 resection indicates that the lateral and basal margins of the resected specimen are free of cancer cells upon pathological examination. In this study, tumor recurrence rates were recorded at 1, 2, and 3 years postoperatively. Recurrence was confirmed by follow-up endoscopy and pathological biopsy showing low-/high-grade intraepithelial neoplasia, squamous cell carcinoma, or adenocarcinoma[14,16]; (2) Adverse reactions: According to the American Society for Gastrointestinal Endoscopy lexicon for endoscopic adverse events[17], adverse reactions primarily include perforation, delayed bleeding (hematemesis or melena occurring more than 24 hours after surgery and requiring endoscopic hemostasis or blood transfusion), esophageal stricture (postoperative dysphagia confirmed by endoscopy, with an esophageal lumen diameter of ≤ 3 mm), and infection (postoperative fever with a temperature of ≥ 38.5 °C, accompanied with elevated white blood cell count in routine blood tests or radiologically confirmed pulmonary or mediastinal infections). Severity was classified as mild (conservative management only), moderate (endoscopic intervention required), or severe (surgery or intensive care unit admission needed); (3) Operative indicators: These indicators included operation time (t), dissection area (S), dissection speed (v), and intraoperative bleeding volume. Operation time refers to the total time from when the endoscope enters the esophagus to locate the lesion until the lesion is completely resected, the wound is treated, and the endoscope is withdrawn. Dissection area S = [(a + b)/2] × [(c + d)/2] (where a and b are the maximum and minimum lengths of the lesion, respectively, and c and d are the maximum and minimum widths of the lesion, respectively). Dissection speed is calculated as v = S/t; (4) Inflammatory markers: The serum levels of interleukin-8 (IL-8), tumor necrosis factor-alpha (TNF-α), and high-sensitivity C-reactive protein (hs-CRP) were measured using a fully automated chemiluminescence immunoassay analyzer (Cobas e 801, Roche Diagnostics, Switzerland) preoperatively, on postoperative day 3, and on postoperative day 7; and (5) Quality of life: Quality of life was assessed using the General Quality of Life Inventory (GQOLI-74) questionnaire preoperatively and 6 months postoperatively. The questionnaire consists of four dimensions, namely, psychological function, physical function, social function, and material living status, comprising a total of 74 items. A five-point Likert scale (1 to 5 points) was used, where each dimension score is calculated as the sum of all item scores in that dimension divided by the number of items (i.e., the average score) and then multiplied by 20 to convert it to a 0-100 scale. High scores indicate high quality of life in that dimension. The Cronbach’s α coefficient for this scale is 0.86[18].

Statistical analysis

This study used SPSS 29.0 statistical software (IBM Corp., Armonk, NY, United States) for data processing and analysis. All statistical tests were two-tailed, and P < 0.05 was considered statistically significant. The Shapiro-Wilk test was used to assess the normal distribution characteristics of continuous variables. According to this test, all continuous variables in this study followed a normal distribution and are reported as mean ± SD. Independent sample t-tests were used for between-group comparisons. Categorical variables are expressed as n (%) and were compared between groups via the χ² test.

For the primary and secondary outcomes, we performed multivariate analyses to adjust for potential confounders, including age, sex, body mass index, diabetes, hypertension, tumor histology (esophageal squamous cell carcinoma vs Barrett’s esophagus-associated adenocarcinoma), depth of invasion (M1/M2/M3), maximum tumor diameter, tumor location, and circumferential extent. Multivariate logistic regression was used for binary outcomes (R0 resection, recurrence, adverse events), and the results are reported as adjusted odds ratios (aORs) and 95% confidence intervals (CIs). Multivariate linear regression was employed for continuous outcomes (inflammatory markers, quality of life scores, and operative indicators), with the results reported as adjusted coefficients (β values) and 95%CI.

The primary endpoints (efficacy and adverse reactions) had no missing data. For the secondary endpoints, the proportion of missing values was < 3% (due to missed blood draws or incomplete quality of life questionnaires). Multiple imputation that used a fully conditional specification with five imputations was performed to manage the missing data. Sensitivity analyses using complete-case analysis yielded consistent results.

RESULTS

Baseline characteristics

The baseline demographic characteristics, including age, gender, body mass index, smoking and drinking habits, diabetes, and hypertension, of the ESD and ESTD groups were comparable (all P > 0.05, Table 1). No significant differences were observed between the two groups regarding disease characteristics, such as histological type, tumor, node, and metastasis staging, depth of invasion, maximum tumor diameter, tumor location, macroscopic type, and circumferential extent (all P > 0.05, Table 2).

Table 1 Comparison of demographic characteristics between two groups, n (%).

Parameter	ESD group (n = 156)	ESTD group (n = 101)	t/χ2	P value
Age (years), mean ± SD	65.84 ± 8.85	64.68 ± 9.42	1.001	0.318
Gender			0.059	0.808
Male	120 (76.92)	79 (78.22)
Female	36 (23.08)	22 (21.78)
BMI (kg/m2), mean ± SD	23.51 ± 3.22	24.16 ± 3.54	1.508	0.133
Smoking habit			0.116	0.734
Frequent	94 (60.26)	63 (62.38)
Occasional/non-smoker	62 (39.74)	38 (37.62)
Drinking habit			0.295	0.587
Frequent	78 (50.00)	54 (53.47)
Occasional/non-drinker	78 (50.00)	47 (46.53)
Diabetes			0.857	0.355
Yes	31 (19.87)	25 (24.75)
No	125 (80.13)	76 (75.25)
Hypertension			0.084	0.772
Yes	49 (31.41)	30 (29.70)
No	107 (68.59)	71 (70.30)

ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection; BMI: Body mass index.

Table 2 Comparison of disease characteristics between two groups, n (%)/mean ± SD.

Parameter	ESD group (n = 156)	ESTD group (n = 101)	t/χ2	P value
Histological type			0.345	0.557
ESCC	109 (69.87)	74 (73.27)
BE-EAC	47 (30.13)	27 (26.73)
TNM staging			0.086	0.769
Tis	62 (39.74)	42 (41.58)
T1a	94 (60.26)	59 (58.42)
Tumor invasion depth			0.441	0.802
M1	53 (33.97)	37 (36.63)
M2	62 (39.74)	36 (35.64)
M3	41 (26.28)	28 (27.72)
Maximum tumor diameter (mm)	33.86 ± 10.98	31.92 ± 9.75	1.440	0.151
Tumor location			1.756	0.416
Upper esophagus	16 (10.26)	6 (5.94)
Middle esophagus	58 (37.18)	36 (35.64)
Lower esophagus	82 (52.56)	59 (58.42)
Macroscopic type			0.637	0.888
Elevated type	40 (25.64)	29 (28.71)
Flat type	88 (56.41)	52 (51.49)
Depressed type	23 (14.74)	16 (15.84)
Mixed types	5 (3.21)	4 (3.96)
Circumferential extent			1.139	0.286
< 50%	98 (62.82)	70 (69.31)
50%-75%	58 (37.18)	31 (30.69)

ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection; ESCC: Esophageal squamous cell carcinoma; BE-EAC: Barrett’s esophagus-associated adenocarcinoma; TNM: Tumor, node, and metastasis.

Efficacy

The en bloc resection rate was high in both groups (P > 0.05). However, the R0 resection rate in the ESTD group (96.04%) was significantly higher than that in the ESD group (89.10%, P = 0.047, Table 3). The recurrence rates in the first and second years after the operation did not differ significantly between the groups (both P > 0.05). However, at the three-year follow-up, the recurrence rate in the ESTD group (4.95%) was significantly lower than that in the ESD group (12.82%, P = 0.038, Table 4).

Table 3 Comparison of the resection effect between two groups, n (%).

Parameter	ESD group (n = 156)	ESTD group (n = 101)	χ2	P value
En bloc resection rate	151 (96.79)	99 (98.02)	0.039	0.844
R0 resection rate	139 (89.10)	97 (96.04)	3.932	0.047

ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection.

Table 4 Comparison of the recurrence rates between two groups, n (%).

Parameter	ESD group (n = 156)	ESTD group (n = 101)	χ2	P value
Postoperative 1 year	5 (3.21)	2 (1.98)	0.039	0.844
Postoperative 2 years	12 (7.69)	4 (3.96)	1.463	0.227
Postoperative 3 years	20 (12.82)	5 (4.95)	4.324	0.038

ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection.

Adverse reactions

The overall incidence of procedure-related adverse reactions in the ESTD group (7.92%) was significantly lower than that in the ESD group (17.31%, P = 0.032). The occurrences of specific adverse events, including perforation, delayed bleeding, esophageal stricture, and infection, are shown in detail in Table 5. However, when the adverse events were graded by severity, no significant difference was found in the incidence of moderate-to-severe events between the ESD and ESTD groups (7.05% vs 1.98%, P = 0.128).

Table 5 Comparison of the incidence of procedure-related adverse reactions between two groups, n (%).

Parameter	ESD group (n = 156)	ESTD group (n = 101)	χ2	P value
Adverse reaction incidence rate	27 (17.31)	8 (7.92)	4.592	0.032
Delayed bleeding	8 (5.13)	3 (2.97)
Esophageal stricture	16 (10.26)	8 (7.92)
Infection	2 (1.28)	1 (0.99)
Moderate-to-severe events	11 (7.05)	2 (1.98)	2.312	0.128

ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection.

Operative indicators

While the operation time was comparable between groups (P > 0.05), the ESTD group demonstrated a larger dissection area (1548.29 ± 398.74 vs 1395.76 ± 423.58, P = 0.004), a higher dissection speed (18.85 ± 3.45 vs 16.34 ± 3.12, P < 0.001), and a significantly reduced intraoperative bleeding volume (19.42 ± 6.37 vs 26.83 ± 7.24, P < 0.001) compared with the ESD group (Figure 1).

Figure 1 Comparison of the operative indicators between two groups. A: Operation time; B: Dissection area; C: Dissection speed; D: Intraoperative bleeding volume. ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection.

Inflammatory markers

The preoperative levels of the inflammatory markers (IL-8, TNF-α, and hs-CRP) were similar between the groups (all P > 0.05). On postoperative days 3 and 7, all three inflammatory markers were consistently and significantly lower in the ESTD group compared with the ESD group (all P < 0.01, Table 6).

Table 6 Comparison of the inflammatory markers between two groups, mean ± SD.

Parameter	ESD group (n = 156)	ESTD group (n = 101)	t	P value
IL-8 (pg/mL)
Preoperative	15.32 ± 3.25	15.28 ± 3.43	0.090	0.929
Postoperative 3 days	20.58 ± 4.76	18.94 ± 4.12	2.844	0.005
Postoperative 7 days	15.89 ± 3.82	14.76 ± 3.05	2.629	0.009
TNF-α (pg/mL)
Preoperative	8.45 ± 1.58	8.41 ± 1.62	0.205	0.838
Postoperative 3 days	11.02 ± 2.34	10.15 ± 2.01	3.081	0.002
Postoperative 7 days	8.76 ± 1.95	8.12 ± 1.74	2.691	0.008
hs-CRP (mg/L)
Preoperative	3.44 ± 1.02	3.39 ± 1.08	0.372	0.710
Postoperative 3 days	10.38 ± 3.16	9.31 ± 2.85	2.744	0.006
Postoperative 7 days	6.05 ± 1.83	5.42 ± 1.76	2.726	0.007

ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection; IL-8: Interleukin-8; TNF-α: Tumor necrosis factor-alpha; hs-CRP: High-sensitivity C-reactive protein.

Quality of life

The preoperative quality of life scores across all four dimensions of GQOLI-74 showed no significant intergroup differences (all P > 0.05). During the six-month postoperative assessment, the patients in the ESTD group reported significantly higher scores in psychological functioning (91.24 ± 3.12 vs 89.96 ± 3.58, P = 0.004), physical functioning (90.67 ± 3.88 vs 89.15 ± 4.15, P = 0.004), social functioning (89.43 ± 4.05 vs 87.92 ± 3.74, P = 0.002), and material living status (91.85 ± 3.96 vs 90.54 ± 3.81, P = 0.009) compared with those in the ESD group (Figure 2).

Figure 2 Comparison of General Quality of Life Inventory-74 scores between two groups (points). A: Psychological functioning; B: Physical functioning; C: Social functioning; D: Material living status. ESD: Endoscopic submucosal dissection; ESTD: Endoscopic submucosal tunnel dissection.

Multivariate regression analyses

After all covariates listed in Tables 1 and 2 were adjusted, ESTD remained significantly associated with a higher likelihood of R0 resection (aOR = 3.21, 95%CI: 1.18-8.73, P = 0.022) and a lower risk of three-year recurrence (aOR = 0.31, 95%CI: 0.11-0.87, P = 0.026) compared with ESD. The overall incidence of adverse reactions was also significantly lower in the ESTD group after adjustment compared with the ESD group (aOR = 0.40, 95%CI: 0.17-0.94, P = 0.035). The linear regression models indicated that ESTD independently predicted a larger dissection area (adjusted β = 148.52 mm², 95%CI: 48.63-248.41, P = 0.004), had a higher dissection speed (adjusted β = 2.51 mm²/minute, 95%CI: 1.21-3.81, P < 0.001), and reduced the intraoperative bleeding volume (adjusted β = -7.39 mL, 95%CI: -9.92 to -4.86, P < 0.001) compared with ESD. Furthermore, ESTD was associated with low IL8 levels on postoperative day 7 (adjusted β = -1.15 pg/mL, 95%CI: -1.98 to -0.32, P = 0.007) and high physical functioning scores on GQOLI-74 at six months (adjusted β = 1.56 points, 95%CI: 0.49-2.63, P = 0.004). These results confirm that the observed benefits of ESTD are independent of potential confounders.

DISCUSSION

The comparison of ESD and ESTD in the treatment of early esophageal cancer produced several important findings that could guide clinical practice. Our study highlighted differences in efficacy, adverse reactions, operative indicators, inflammatory markers, and quality of life between the two procedures. These differences suggest potential mechanisms underlying the observed outcomes and are critical for understanding the advantages and limitations of each technique.

The R0 resection rate in the ESTD group was higher than that in the ESD group. The R0 resection rate is a crucial indicator of complete tumor removal with clear margins; complete tumor removal is essential for preventing recurrence. The improved R0 resection rate in the ESTD group may be attributed to the unique tunneling approach used in this technique. ESTD involves creating a submucosal tunnel from a small entry incision, allowing enhanced visualization and highly precise dissection along the tumor margins. This method potentially reduces the risk of incomplete resection by providing a clear field for the endoscopist to work in. The tunneling approach may help maintain the structural integrity of the esophageal wall during the procedure, leading to enhanced margin control and reduced chances of residual tumor cells[19,20]. By contrast, traditional ESD requires multiple entry points, which can complicate the procedure and increase the risk of incomplete resection[21].

Another notable finding is the low recurrence rate three years after operation in the ESTD group. Although no statistically significant differences in recurrence rates were found after one and two years, the long-term benefit observed in the ESTD group suggests a sustained advantage of this technique. The reduced recurrence rate could be linked to the high R0 resection rate because achieving clear margins substantially reduces the likelihood of residual cancer cells leading to recurrence. This tunneling technique may facilitate a thorough examination and removal of surrounding tissue, ensuring that all affected areas are addressed. This comprehensive approach could contribute to enhanced long-term outcomes and reduce the need for additional treatments or surgeries[22-24].

The overall incidence of procedure-related adverse reactions in the ESTD group was lower than that in the ESD group. Adverse events, such as perforation, delayed bleeding, esophageal stricture, and infection, can have serious implications for patient recovery and quality of life. The reduction in adverse reactions in ESTD could be due to the minimally invasive nature of this tunneling approach. By limiting the number of entry points and maintaining the integrity of the esophageal wall, ESTD may minimize trauma to surrounding tissues, reducing the risk of complications. This tunneling technique allows for controlled and gradual dissection, which can help prevent sudden pressure changes that might lead to perforations or other injuries[25,26]. This careful approach likely contributed to the low incidence of adverse reactions observed in our study.

The operative indicators also showed favorable results for ESTD. Although the operation time was similar in the two groups, the ESTD group demonstrated a larger dissection area, higher dissection speed, and smaller intraoperative bleeding volume compared with the ESD group. This improved operative efficiency is consistent with a recent meta-analysis conducted by Liu et al[19], who concluded that ESTD offers faster resection for large superficial esophageal lesions compared with ESD. The increased dissection area and speed could be attributed to the streamlined access provided by the tunneling technique. This method allows for continuous, uninterrupted dissection along the length of the lesion, potentially improving efficiency and reducing procedural time. The reduction in intraoperative bleeding may be due to the enhanced visualization and control offered by the tunneling approach; reduced intraoperative bleeding facilitates targeted hemostasis and minimizes blood loss. These improvements in operative indicators suggest that ESTD could offer practical benefits in terms of surgical precision and safety[27,28].

The inflammatory markers, including IL-8, TNF-α, and hs-CRP, were consistently lower in the ESTD group compared with the ESD group postoperatively. The reduction in these markers indicates a less pronounced inflammatory response after ESTD. The tunneling technique’s ability to minimize tissue damage and maintain the structural integrity of the esophagus likely plays a role in this reduced inflammation. By avoiding excessive manipulation of surrounding tissues, ESTD may limit the release of pro-inflammatory mediators, resulting in a mild systemic inflammatory response. This response could translate into rapid recovery and reduced postoperative complications for patients undergoing ESTD[22,29].

The quality of life assessments revealed higher scores in psychological functioning, physical functioning, social functioning, and material living status in the ESTD group six months after surgery compared with the ESD group. These improved quality of life measures could be linked to the low incidence of adverse reactions and reduced inflammatory response associated with ESTD. Patients who experience few complications and recover quickly are likely to report exceptional overall well-being. The precision and thoroughness of ESTD may provide patients with high confidence in their treatment outcomes, positively affecting their mental health and daily functioning[30,31]. These findings underscore the importance of considering quality of life when evaluating treatment options for early esophageal cancer.

Despite the promising results, our study has several limitations. First, the retrospective, single-center, non-randomized design is subject to selection and observer bias. Although we performed multivariate adjustment for numerous confounders, unmeasured confounders (e.g., operator experience within the learning curve, patient frailty, and socioeconomic status) may still influence the outcomes. Second, the generalizability of our findings is limited because all procedures were performed by a single expert endoscopist at a high-volume center; the reproducibility of ESTD in low-volume settings or by less experienced operators remains unknown. Third, GQOLI-74 has not been specifically validated in post-endoscopic esophageal cancer populations, and we did not collect baseline psychosocial factors (e.g., anxiety and depression) that could influence quality of life independently of the procedure. Fourth, the measurement of the inflammatory markers at only three time points may have missed dynamic changes, and perioperative medications (antibiotics and proton pump inhibitors) could have modulated these markers, although the standardized protocol mitigates between-group differences. Fifth, follow-up beyond three years is ongoing, and long-term recurrence and survival data are needed. Sixth, we did not perform a formal cost-effectiveness analysis, which is essential for informing healthcare policies. Seventh, although the handling of missing data via multiple imputation is robust, it assumes missingness at random, which may not hold. Finally, the lack of blinding of the outcome assessors for some endpoints (e.g., operative indicators) may have introduced detection bias.

Future prospective, multicenter, randomized controlled trials with blinded outcome assessment are urgently needed to confirm our findings. These trials should also include prespecified subgroup analyses (e.g., by tumor location, circumferential extent, and histology) to identify patients who derive remarkable benefits from ESTD. Moreover, long-term follow-up (≥ 5 years) and cost-effectiveness analyses are warranted. The learning curve for ESTD should be formally characterized using cumulative sum analysis to guide training and credentialing. The effect of ESTD on patient-reported outcomes beyond six months and on health resource utilization should also be examined.

CONCLUSION

This study suggests that ESTD may offer potential advantages over traditional ESD in the treatment of early esophageal cancer. ESTD achieves high R0 resection and low recurrence rates potentially because of its ability to provide enhanced visualization and highly precise dissection along tumor margins. Furthermore, ESTD may result in only a few adverse reactions possibly because of its minimally invasive nature and reduced tissue trauma. The procedure also shows promise in improving operative efficiency and reducing intraoperative bleeding, which could enhance patient outcomes. Moreover, patients undergoing ESTD may experience a low inflammatory response and improved quality of life postoperatively. These findings indicate that ESTD could be a viable alternative to ESD, offering enhanced therapeutic outcomes without compromising patient safety.