Nursing outcome comparison of laparoscopy-assisted, pneumomediastinoscopy-assisted esophagectomy vs conventional thoracoscopic-assisted, three-incision esophagectomy in early esophageal cancer

Abstract

BACKGROUND

Surgical resection is an important component of curative therapy for early and locally advanced esophageal cancers. Despite the advantages of conventional open esophagectomy, it results in considerable surgical trauma, cardiopulmonary burden, and a high rate of postoperative respiratory and cardiac complications. To address these limitations, mediastinoscopy-assisted and pneumomediastinum-based esophagectomy techniques have been introduced and refined. Nursing teams play a central role in coordinating mobilization schedules, respiratory exercises, drain care, and nutritional advancement as well as in monitoring early signs of clinical deterioration. As such, detailed reports addressing relevant nursing outcomes of new mediastinoscopy-assisted strategies are of particular interest.

AIM

To compare perioperative and early recovery outcomes between laparoscopy-assisted pneumomediastinoscopy-assisted esophagectomy (LAPA-PMAE) and conventional thoracoscopic-assisted, three-incision esophagectomy for early esophageal cancer.

METHODS

This retrospective cohort study screened 136 consecutive patients with resectable early thoracic esophageal cancer who underwent minimally invasive esophagectomy at the First Affiliated Hospital of Fujian Medical University (Fuzhou, Fujian Province, China). After excluding patients who underwent emergency or non-elective surgery, combined major procedures on other organs, had incomplete perioperative or follow-up data, died during the index operation, or fulfilled other exclusion criteria, 110 patients were divided into 2 groups: Conventional thoracoscopic-assisted, three-incision esophagectomy (control, n = 60); and LAPA-PMAE (observation, n = 50).

RESULTS

Chronic obstructive or other chronic lung diseases were more prevalent in the observation group, whereas other baseline variables were similar. Operative duration and blood loss were lower in the observation group. Patients in this group achieved earlier ambulation and oral intake, had a shorter drain duration, shorter postoperative hospital stay, and a more favorable recovery speed index. Overall complication rates were comparable, whereas postoperative arrhythmia occurred only in the control group. After adjusting for baseline factors, the LAPA-PMAE approach remained associated with a shorter hospital stay, suggesting that the observed difference in hospital length of stay was not fully accounted for by the measured risk variables. Lymph node yield was comparable between groups, whereas patient-controlled analgesia use and rescue analgesic use were lower in the observation group.

CONCLUSION

LAPA-PMAE in a single supine position through cervical and upper abdominal access is associated with a lighter perioperative profile, faster early recovery, and shorter hospitalization.

Key Words: Early esophageal cancer; Laparoscopy-assisted pneumomediastinoscopy; Single-position esophagectomy; Minimally invasive esophagectomy; Perioperative nursing; Enhanced recovery after surgery; Postoperative complications; Impaired pulmonary function

Core Tip: This study compares perioperative and early recovery outcomes between laparoscopy-assisted pneumomediastinoscopy-assisted esophagectomy and conventional thoracoscopic-assisted esophagectomy in patients with early esophageal cancer. The retrospective analysis of 110 patients demonstrates that the laparoscopy-assisted pneumomediastinoscopy-assisted esophagectomy approach yields shorter operative duration, reduced blood loss, earlier ambulation and oral intake, decreased drain duration, and shorter postoperative hospital stay with comparable complication rates, suggesting a favorable perioperative profile and accelerated recovery.

INTRODUCTION

Esophageal cancer remains a significant global health burden, with > 600000 new cases and > 500000 deaths reported worldwide in 2020; a substantial proportion of this burden is concentrated in Eastern Asia[1,2]. China accounts for a large proportion of these cases and deaths, and esophageal squamous cell carcinoma remains the predominant pathological subtype in many high-risk areas[3,4]. Although screening and early detection programs have expanded, a considerable number of patients are still diagnosed at stages in which multimodal treatment and major surgery are required, and overall survival remains unsatisfactory in many populations[5].

Surgical resection is an important component of the curative therapy for early and locally advanced esophageal cancers. Conventional open esophagectomy has the advantage of good exposure; however, it results in considerable surgical trauma, cardiopulmonary burden, and a high rate of postoperative respiratory and cardiac complications. Minimally invasive esophagectomy (MIE) was developed to reduce access trauma, while maintaining oncological principles. Randomized trials and large cohort studies have reported that hybrid or fully minimally invasive approaches can reduce major complications and shorten the length of hospital stay(s) compared with open techniques, while providing at least comparable long-term survival and oncological adequacy[6-10]. Despite these advantages, conventional thoracoscopic and thoracolaparoscopic procedures often require single-lung ventilation and prolonged thoracic manipulation, which can be challenging in patients with limited cardiopulmonary reserves.

To address these limitations, mediastinoscopy-assisted and pneumomediastinum-based esophagectomy techniques have been introduced and refined. Early descriptions of mediastinoscope-assisted transhiatal esophagectomy showed that a transcervical and transhiatal route could be used as a minimally invasive option for thoracic esophageal cancer, potentially decreasing pulmonary complications by avoiding formal thoracotomy[11]. More recently, mediastinoscopic esophagectomy under pneumomediastinum was developed, which combines mediastinal insufflation with transcervical access to improve visualization and enable systematic lymph node dissection. Inflatable video-assisted mediastinoscopic transhiatal esophagectomy has emerged as a prominent technique in this field. Clinical series and reports addressing the learning curve have shown that the procedure is practical in carefully selected patients, even those with early stage disease and serious cardiopulmonary comorbidities, and demonstrates perioperative outcomes comparable with those of thoracoscopy-assisted procedures[12,13].

There is increasing awareness that outcomes after esophagectomy should not be evaluated solely based on operative mortality, anastomotic insufficiency, or long-term survival. Postoperative milestones, functional recovery, and quality of life are increasingly recognized to be important endpoints, particularly in older patients and those with poor baseline pulmonary function. Quality of life following inflatable mediastinoscopic transhiatal esophagectomy was assessed and found to be acceptable in the short term for selected patients, with early physical recovery possible after this procedure[14-16]. Nevertheless, most of the literature addressing mediastinoscopy-assisted techniques focused primarily on technical aspects, lymph node dissection, overall complication rates, and prognosis, while providing limited details about structured, nursing-specific recovery parameters, such as time to ambulation, oral intake, drain management, and readiness for discharge.

Perioperative nursing practice has evolved alongside advances in surgical techniques, largely influenced by enhanced recovery-after-surgery (ERAS) protocols. Enhanced recovery pathways for esophagectomy and, more broadly, major gastrointestinal surgery emphasize preoperative optimization, standardized anesthesia and analgesia, careful fluid management, early mobilization, and early nutritional support. Several studies have demonstrated that a higher adherence to these components is associated with shorter intensive care unit (ICU) and hospital stays, reduced pulmonary complications and arrhythmias, and improved early postoperative recovery[17,18]. Nursing teams are integral to these pathways and play a central role in coordinating mobilization schedules, respiratory exercises, drain care, and nutritional advancement as well as in monitoring early signs of clinical deterioration. Against this background, detailed reports addressing relevant nursing outcomes of new mediastinoscopy-assisted strategies are of particular interest.

Although minimally invasive and transcervical approaches have rapidly progressed, comparative evidence remains limited. Direct comparisons between laparoscopy-assisted pneumomediastinoscopy-assisted esophagectomy (LAPA-PMAE) and standard thoracoscopic-assisted three-incision esophagectomy remain relatively rare, especially in patients with early esophageal cancer and compromised pulmonary function, who represent a key target group for less-invasive mediastinal access. Existing comparative accounts often report combined complication rates and other broad perioperative endpoints; however, specific recovery milestones are not always quantified, and adjusted analyses that account for baseline cardiopulmonary risk are uncommon[13,15]. Furthermore, many studies are from high-volume centers, and technical details, nursing protocols, and discharge criteria are not always reported in a way that can be translated directly into everyday perioperative practice.

The present study, therefore, was designed to contribute to this area by examining perioperative and early postoperative outcomes in a cohort of patients with early esophageal cancer who underwent either LAPA-PMAE or conventional thoracoscopy-assisted three-incision esophagectomy at a single center. The analysis focused on operative variables, early recovery milestones, postoperative complications, and nursing-relevant endpoints and used multivariable models to explore the association between surgical approach, postoperative hospital length of stay (LoS), and composite complication outcomes.

At our center, LAPA-PMAE is performed with the patient in a single supine position through a cervical incision and upper abdominal access without a thoracic incision, using single-port mediastinoscopic and laparoscopic techniques. This approach has been mainly offered to patients with early esophageal cancer and impaired pulmonary function, who are considered at higher risk for conventional three-incision thoracoscopic esophagectomy. However, the nursing-related recovery outcomes for this specific single-position strategy have not yet been characterized in detail.

MATERIALS AND METHODS

Study design and setting

This retrospective cohort study was conducted at the First Affiliated Hospital of Fujian Medical University (Fuzhou, Fujian Province, China). Consecutive patients with early esophageal cancer who underwent MIE were screened for eligibility during the study period. Patients who underwent LAPA-PMAE in the single supine position comprised the observation group. Patients who underwent conventional thoracoscopy-assisted, three-incision esophagectomy were included in the control group.

Patients and eligibility criteria

In this study, 136 consecutive patients with resectable early thoracic esophageal cancer who underwent MIE at the authors’ hospital were screened for eligibility. Patients were excluded if any of the following criteria applied: Underwent emergency or non-elective surgery (n = 7); required combined major procedures on other organs (n = 3); incomplete key perioperative or follow-up data (n = 6); died during the index operation (n = 6); or fulfilled other exclusion criteria (n = 4). After exclusion, 110 patients were included in the final analytical cohort, with 50 allocated to the LAPA-PMAE group and 60 to the conventional thoracoscopic-assisted three-incision esophagectomy group (Figure 1).

Figure 1 Flow diagram of patient selection and group allocation.

In routine clinical practice, the surgical approach was selected according to clinical indications rather than by a prespecified allocation protocol. LAPA-PMAE was preferentially considered for patients with early-stage disease and impaired pulmonary function who were considered to have limited tolerance for prolonged one-lung ventilation during conventional thoracoscopic-assisted three-incision esophagectomy. Final selection of the surgical approach was made after multidisciplinary discussion, considering tumor stage, pulmonary reserve, anesthetic risk, and technical feasibility. Therefore, the higher proportion of chronic obstructive pulmonary disease (COPD) or other chronic lung disease in the observation group reflected the clinical indication profile of this procedure and should be considered when interpreting between-group comparisons.

The baseline variables included age, sex, body mass index, smoking history, hypertension, diabetes, chronic obstructive or other chronic lung diseases, American Society of Anesthesiologists (ASA) class, forced expiratory volume in 1 second (FEV₁), FEV₁/forced vital capacity, serum albumin, and hemoglobin levels. Tumor-related variables included tumor location, clinical T stage, clinical N stage, and neoadjuvant therapy.

Surgical procedures

LAPA-PMAE: Patients in the observation group underwent a single-position, minimally invasive procedure while supine. Under general anesthesia with endotracheal intubation, patients were positioned supine with slight neck extension and no intraoperative repositioning was required. A small cervical incision was made along the left neck border to expose the cervical esophagus. After the creation of the pneumomediastinum, a single-port video mediastinoscope was introduced through the cervical access. Dissection of the thoracic esophagus and periesophageal lymphatic tissue was performed along the mediastinum under endoscopic guidance without a thoracic incision.

The abdominal phase was performed laparoscopically through an upper abdominal working incision and a trocar placement pattern. The stomach was mobilized, the lymph nodes in the abdomen were dissected, and a gastric conduit was formed. Specimens were removed using a cervical incision. Cervical esophagogastric anastomosis was performed according to the surgeon’s preference. Mediastinal and upper abdominal drains were placed as required before the patient left the operating room. In summary, the procedure was completed with the patient in a single supine position through a cervical incision and upper abdominal access using a single-port endoscopic technique.

Conventional thoracoscopic-assisted three-incision esophagectomy: Patients in the control group underwent conventional thoracoscopy-assisted three-incision esophagectomy. The lateral decubitus position was used during the thoracic phase, with one-lung ventilation. Thoracoscopic ports were inserted, the thoracic esophagus was mobilized, and mediastinal lymph node dissection was performed under visualization using a thoracoscope. Stomach mobilization and abdominal lymphadenectomy were performed with the formation of a gastric conduit, either laparoscopically or by open surgery, according to surgeon and institutional norms. The cervical phase consisted of the exposure of the cervical esophagus, specimen removal, and cervical esophagogastric anastomosis. All procedures were performed or supervised by surgeons with extensive experience with MIE and the transcervical mediastinal approach at the First Affiliated Hospital of Fujian Medical University.

Perioperative management and nursing protocol

All patients were managed according to a standardized institutional perioperative pathway that incorporated the key elements of ERAS. Before the operation, patients were educated about the procedure, respiratory exercises, mobilization goals, and postoperative pain management. Nutritional status and comorbidities were evaluated and optimized.

The key nursing process was structured around preoperative respiratory preparation, intraoperative coordination, postoperative respiratory care, pain assessment, early mobilization, nutritional advancement, and drain management. Before surgery, nurses instructed patients to perform deep-breathing and effective coughing exercises at least three times daily and documented patient compliance when available. For patients with impaired pulmonary function, sputum expectoration training and individualized respiratory guidance were reinforced. During surgery, the nursing team coordinated the supine position, protected the cervical and upper abdominal access sites, secured anesthesia and drainage tubes, and monitored hemodynamic and oxygenation changes during pneumomediastinum establishment. After surgery, pain was assessed regularly using the institutional pain assessment form, and analgesic requirements were communicated to the surgical and anesthesia teams. Patients who were hemodynamically stable, had no active bleeding, and had adequate oxygenation were encouraged to sit at the bedside and stand within the first postoperative day, followed by progressive ward ambulation. Respiratory training, assisted coughing, and chest physiotherapy were continued several times daily according to tolerance. Oral intake and drain removal were advanced according to predefined clinical criteria, including anastomotic safety, return of intestinal function, drainage volume, and absence of suspected bleeding, chyle leakage, or anastomotic failure. The same institutional ERAS-based nursing pathway was applied to both groups, while procedure-specific intraoperative coordination differed according to the access route and operative position.

Postoperative pain management consisted of a regular nursing team-assessed multimodal schedule. Respiratory therapy consisted of incentive spirometry, chest physical therapy, coughing, and oxygen therapy if needed, and early mobilization was recommended. Sitting at the bedside and standing were planned within the first postoperative day for clinically stable patients, followed by progressive walking in the ward under nursing supervision.

Oral intake was advanced in a stepwise manner. Initially, patients were administered parenteral or enteral tube feeding. After confirmation of anastomotic integrity and the return of intestinal function, clear fluids were started, and the intake quantity gradually increased toward a soft diet. Nurses observed the patients for tolerance to feeding, as indicated by signs such as nausea, vomiting, and cramping in the abdomen, and documented the observations in the medical records.

Drain management was performed according to standardized criteria. The drainage volume and characteristics were monitored, and the drain sites were inspected regularly. Removal of the mediastinal and abdominal drains was considered when the drainage volume decreased and there were no signs of active bleeding, chyle leakage, or anastomotic failure.

Data collection and variable definitions

Clinical information was obtained from electronic medical and perioperative nursing records. The duration of surgery was defined as the interval from skin incision to wound closure, and was documented in minutes. Estimated blood loss was recorded from anesthesiology records. Lymph node-related variables included the total number of harvested lymph nodes and the number of positive lymph nodes. Analgesic-related variables included patient-controlled analgesia use, rescue analgesic use, and the number of rescue analgesic administrations during the early postoperative period.

Recovery milestones were calculated from the conclusion of surgery onward. Time to first ambulation was defined as the time elapsed from the end of surgery to the first time standing or walking out of bed under nursing supervision. Time to first flatus was defined as the time from the end of surgery to the time of the first documented passage of gas per rectum. The time to first oral intake was defined as the time from the end of surgery to the first permitted oral fluid intake. Drain duration was defined as the time from the end of surgery to the removal of the last mediastinal or abdominal drain. Postoperative LoS was defined as the time from surgery to discharge in days, and 30-day readmission was ascertained from hospital records.

A composite recovery speed index was constructed from the standardized values of the main recovery milestones to summarize the overall recovery performance, with lower values indicating faster recovery and higher values indicating slower recovery, according to predefined coding rules.

Postoperative complications that occurred during the index admission were detected by reviewing clinical notes, radiology reports, operation notes, and discharge summaries. Pulmonary complications included pneumonia, atelectasis requiring bronchoscopy or escalation of respiratory support, and clinically significant pleural effusion requiring intervention. Arrhythmia was defined as a new supraventricular tachyarrhythmia or other clinically relevant arrhythmia requiring pharmacological or interventional treatment after surgery. Anastomotic leakage was diagnosed based on contrast studies, endoscopy, or enteric content drainage. Deep vein thrombosis was diagnosed using duplex ultrasonography or other imaging modalities. Reoperation was defined as any unplanned return to the operating room during the index admission. The composite complication end point was defined as the occurrence of any recorded postoperative complications.

Outcomes

The primary outcome was postoperative LoS. Other important endpoints included times to first ambulation, first flatus, and first oral intake, and drain duration, composite recovery speed index, and 30-day readmission. The safety results included single postoperative complications and combined complication endpoints. Additional perioperative endpoints included lymph node yield and postoperative analgesic-use variables.

Statistical analysis

Statistical analyses were performed using SPSS version 29.0.1.0.171 (IBM Corp., Armonk, NY, United States). Continuous variables were examined for distributional characteristics. Variables with approximate normal distribution are expressed as mean ± SD, and the t test was used for inter-group comparison. Skewed variables were reported as median and interquartile range and compared using the Mann-Whitney U test. Categorical variables are expressed as n (%), and were compared using the χ² test or Fisher’s exact test as needed. All statistical tests were two-tailed, and differences with P < 0.05 were considered to be significant.

Multivariable models were used to describe the associations between the surgical approach and the selected outcomes. An ordinary least squares regression model was constructed for the postoperative LoS with the surgical group as the main exposure. Covariates in this model included age, sex, ASA class, FEV₁, presence of chronic obstructive or other chronic lung disease and receipt of neoadjuvant therapy. A logistic regression model was fitted to the composite complication end points using the same set of covariates.

Ethics approval

The study protocol was approved by the Ethics Committee of First Affiliated Hospital of Fujian Medical University, and was conducted in accordance with the Declaration of Helsinki and the applicable local regulations, approval No. FMU[2015]084-3.

RESULTS

Study population and baseline characteristics

Of 136 patients who were screened, 110 fulfilled the eligibility criteria and were included in the final cohort (Figure 1). Among the included patients, 50 underwent LAPA-PMAE in a single supine position (observation) and 60 underwent conventional thoracoscopy-assisted three-incision esophagectomy (control).

Baseline demographics, nutritional indices, pulmonary function, clinical stage, and neoadjuvant treatment are summarized in Table 1. Overall, the 2 groups were well-matched in the majority of baseline characteristics, such as age, body mass index, FEV₁, FEV₁/forced vital capacity, albumin, hemoglobin, sex distribution, smoking history, hypertension, diabetes, ASA classification, tumor location, clinical T stage, clinical N stage, and neoadjuvant treatment.

Table 1 Baseline characteristics and preoperative risk profile, n (%)/mean ± SD.

Variable	Control group	Observation group	Test	P value
Age (years)	61.17 ± 9.40	62.68 ± 8.14	Welch t	0.368
BMI (kg/m2)	22.77 ± 2.77	23.35 ± 2.76	Welch t	0.273
FEV1 (L)	2 ± 0.44	1.93 ± 0.40	Welch t	0.432
FEV1/FVC (%)	67.03 ± 7.68	68.92 ± 9.18	Welch t	0.250
Albumin (g/L)	38.44 ± 3.41	39.11 ± 3.37	Welch t	0.305
Hemoglobin (g/L)	120.82 ± 13.39	124.58 ± 12.40	Welch t	0.129
Gender			χ2 test	0.533
Female	15 (25)	10 (20)
Male	45 (75)	40 (80)
Smoking history			χ2 test	0.330
No	28 (46.70)	28 (56)
Yes	32 (53.30)	22 (44)
COPD or other chronic lung disease			χ2 test	0.036
No	49 (81.70)	32 (64)
Yes	11 (18.30)	18 (36)
Hypertension			χ2 test	0.385
No	36 (60)	34 (68)
Yes	24 (40)	16 (32)
Diabetes			χ2 test	0.541
No	49 (81.70)	43 (86)
Yes	11 (18.30)	7 (14)
ASA class			χ2 test	0.997
1	11 (18.30)	9 (18)
2	38 (63.30)	32 (64)
3	11 (18.30)	9 (18)
4	0 (0)	0 (0)
Tumor location			χ2 test	0.663
Upper	15 (25)	9 (18)
Middle	24 (40)	21 (42)
Lower	21 (35)	20 (40)
cT stage			χ2 test	0.637
1	29 (48.30)	20 (40)
2	23 (38.30)	21 (42)
3	8 (13.30)	9 (18)
4	0 (0)	0 (0)
cN stage			χ2 test	0.905
0	45 (75)	37 (74)
1	15 (25)	13 (26)
2	0 (0)	0 (0)
3	0 (0)	0 (0)
Neoadjuvant therapy			χ2 test	0.610
No	40 (66.70)	31 (62)
Yes	20 (33.30)	19 (38)

BMI: Body mass index; FEV: Forced expiratory volume; FVC: Forced vital capacity; COPD: Chronic obstructive pulmonary disease; ASA: American Society of Anesthesiologists; cT: Clinical T stage; cN: Clinical N stage.

In addition, there was a difference in the rates of COPD and chronic lung disease, which were more common in the observation group than in the control group. This imbalance was consistent with the clinical indication for the LAPA-PMAE approach, which was preferentially used in patients with limited pulmonary reserve.

Intra- and perioperative parameters

Perioperative parameters are summarized in Table 2. Operative duration was shorter and the estimated blood loss was lower in the observation group than in the control group, which is consistent with the lighter intraoperative burden in this cohort. The total number of harvested lymph nodes and the number of positive lymph nodes did not differ significantly between the two groups. Patient-controlled analgesia use, rescue analgesic use, and the number of rescue analgesic administrations were lower in the observation group, indicating lower postoperative analgesic requirements after the LAPA-PMAE approach.

Table 2 Intraoperative, oncological, and analgesic parameters, n (%)/mean ± SD/median (interquartile range).

Variable	Control group	Observation group	Test	P value
Operative time (minutes)	266.78 ± 36.44	234.52 ± 41.53	Welch t	< 0.001
Estimated blood loss (mL)	284.12 ± 75.72	229.36 ± 94.80	Welch t	0.001
ICU stay (hours)	0 (0-39.50)	0 (0-0)	Mann-Whitney U	0.055
ICU admission			χ2 test	0.075
No	36 (60)	38 (76)
Yes	24 (40)	12 (24)
Total lymph nodes harvested	19.98 ± 4.28	19.14 ± 3.89	Welch t	0.282
Positive lymph nodes	0.00 (0.00-1.00)	0.00 (0.00-1.00)	Mann-Whitney U	0.662
PCA use			χ2 test	0.007
No	13 (21.7)	23 (46.0)
Yes	47 (78.3)	27 (54.0)
Rescue analgesic use		χ2 test	0.009
No	31 (51.7)	38 (76.0)
Yes	29 (48.3)	12 (24.0)
Rescue analgesic times	0.00 (0.00-1.00)	0.00 (0.00-0.00)	Mann-Whitney U	0.011

ICU: Intensive care unit; PCA: Patient-controlled analgesia.

ICU admission and ICU stay were numerically lower in the observation group than in the control group, but the between-group differences did not reach statistical significance. Therefore, no definitive conclusion regarding ICU resource use can be drawn from this dataset.

Primary outcome and postoperative recovery milestones

The primary and secondary recovery results are reported in Table 3 and illustrated in Figure 2A. The observation group achieved several key rehabilitation benchmarks more rapidly than the control group. In particular, the time to first ambulation was significantly reduced, reflecting earlier postoperative mobilization. The time to first oral intake was also significantly shorter in the observation group, indicating an earlier reintroduction to oral feeding. The drain duration was shorter in the observation group, which was consistent with a faster postoperative recovery trajectory.

Figure 2 Data for each experimental group. A: Postoperative recovery milestones and length of stay by group; B: Composite recovery speed index by group (excluding hospital length of stay). R² = 0.14. LoS: Length of stay.

Table 3 Primary and key secondary recovery outcomes, n (%)/mean ± SD/median (interquartile range).

Variable	Control group	Observation group	Test	P value
Time to first ambulation (hours)	31 (24.80-36)	25 (20.20-32.50)	Mann-Whitney U	0.002
Time to first flatus (hours)	60.50 (49-72)	58 (50-67.80)	Mann-Whitney U	0.336
Time to first oral intake (hours)	93.50 (84.80-101.20)	74.50 (66.20-87.20)	Mann-Whitney U	4.726e-06
Drain duration (hours)	157 (135-185.20)	132 (104.80-156.80)	Mann-Whitney U	< 0.001
Postoperative LoS (days)	12.60 (10.90-14.80)	11.80 (9.20-13.80)	Mann-Whitney U	0.005
Recovery speed index (z-mean)	0.25 ± 0.52	-0.30 ± 0.39	Welch t	4.534e-09
30-day readmission			χ2 test	0.523
No	53 (88.30)	46 (92)
Yes	7 (11.70)	4 (8)

LoS: Length of stay.

There was no apparent difference in the time to first flatus between the groups, suggesting that the recovery of gastrointestinal function may have been comparable between the approaches in this series. The primary endpoint of the postoperative LoS was reduced in the observation group. A composite measure of recovery speed based on various recovery milestones also favored the observation group, providing an overview of recovery velocity consistent with the results of single milestones. The distribution of the recovery speed index by group is presented in Figure 2B. Thirty-day readmission occurred infrequently in both groups, and no clear between-group differences were observed.

Postoperative complications and safety outcomes

Postoperative complications are reported in Table 4. Pulmonary complications occurred at similar rates in both groups. Importantly, this parity was noted in the presence of a greater baseline prevalence of COPD or other chronic lung disease in the observation group (Table 1), suggesting that postoperative pulmonary risk may not have been markedly elevated in the observation arm of this particular dataset.

Table 4 Postoperative complications (safety outcomes), n (%).

Variables	Control group	Observation group	Test	P value
Pulmonary complications			Fisher	0.753
No	54 (90)	46 (92)
Yes	6 (10)	4 (8)
Anastomotic leak			Fisher	0.500
No	58 (96.70)	50 (100)
Yes	2 (3.30)	0 (0)
Arrhythmia			Fisher	0.008
No	52 (86.70)	50 (100)
Yes	8 (13.30)	0 (0)
Deep vein thrombosis			Fisher	1.000
No	58 (96.70)	49 (98)
Yes	2 (3.30)	1 (2)
Reoperation			Fisher	0.500
No	58 (96.70)	50 (100)
Yes	2 (3.30)	0 (0)

Anastomotic leakage, deep vein thrombosis, and reoperation were rare in both groups, with no evident differences. Arrhythmia was observed in the control group, but not in the observation group, which was consistent with a more favorable pattern for the observation approach with regard to postoperative rhythm disturbance in this cohort.

Adjusted analyses for postoperative LoS and composite complications

Results of the adjusted analyses are reported in Tables 5 and 6. In the multivariable ordinary least squares regression model for postoperative LoS, surgical approach remained associated with a shorter hospital stay after adjustment for age, sex, ASA class, FEV₁, COPD or other chronic lung disease, and neoadjuvant therapy. The adjusted group coefficient indicated fewer postoperative hospital days in the observation group than in the control group, suggesting that the observed difference in the LoS was not fully explained by the measured baseline risk factors.

Table 5 Adjusted analysis for postoperative length of stay by ordinary least squares regression.

Predictors	Beta	95%CI	P value
const	16.10	11.09-21.12	< 0.001
Group	-1.49	-2.59 to -0.38	0.009
Age	-0.04	-0.10 to 0.02	0.176
Gender	-0.86	-2.20 to 0.47	0.202
ASA	0.46	-0.45 to 1.37	0.314
FEV1	-0.57	-1.88 to 0.74	0.389
COPD	-0.21	-1.48 to 1.07	0.750
Neoadjuvant	0.70	-0.43 to 1.82	0.222

ASA: American Society of Anesthesiologists; FEV: Forced expiratory volume; COPD: Chronic obstructive pulmonary disease; CI: Confidence interval.

Table 6 Adjusted model for composite postoperative complications.

Predictors	OR	95%CI	P value
Surgical approach	0.24	0.07-0.79	0.019
Age (per 1 year)	1.00	0.94-1.06	0.997
Gender (male vs female)	0.64	0.21-1.91	0.425
ASA (per 1 class)	0.52	0.19-1.39	0.191
FEV1 (per 1 L)	0.38	0.10-1.39	0.143
COPD (yes vs no)	1.76	0.50-6.18	0.377
Neoadjuvant therapy (yes vs no)	0.30	0.09-1.01	0.053

Composite postoperative complications refers to the occurrence of any of the following: Pulmonary complications, anastomotic leak, arrhythmia, deep vein thrombosis, and reoperation. Model: Binomial logistic regression adjusted for age, gender, American Society of Anesthesiologists class, forced expiratory volume in 1 second, chronic obstructive pulmonary disease, and neoadjuvant therapy. ASA: American Society of Anesthesiologists; FEV: Forced expiratory volume; COPD: Chronic obstructive pulmonary disease; OR: Odds ratio; CI: Confidence interval.

In the adjusted logistic regression model using a composite endpoint of postoperative complications, the surgical approach was associated with lower odds for the complications included in the composite outcome. Neoadjuvant therapy exhibited a borderline pattern toward lower odds in this model; however, the estimate was imprecise and should be interpreted with caution.

DISCUSSION

This retrospective cohort comparison examined the perioperative and recovery outcomes of LAPA-PMAE and conventional thoracoscopy-assisted three-incision esophagectomy for early esophageal cancer. In this dataset, the observation approach showed a consistent pattern of faster postoperative recovery, reflected by the earlier achievement of core postoperative milestones and a shorter postoperative LoS. These endpoints are commonly used to describe short-term functional recovery after esophagectomy and are central to contemporary enhanced recovery pathways. The ERAS Society’s recommendations for esophagectomy emphasize standardized perioperative care elements that support early mobilization, early nutritional progression, and streamlined postoperative management to improve recovery metrics and reduce LoS[19].

The observed differences in operative duration and blood loss are clinically relevant because they describe a perioperative profile with lower operative exposure and measured intraoperative burden in the observation group. Similar perioperative patterns have been described in the recent literature addressing inflatable video-assisted mediastinoscopic transhiatal and related transmediastinal approaches, in which investigators reported feasibility and evolving technical refinements while continuing to evaluate comparative perioperative outcomes[12,20,21]. These publications also highlighted that outcomes may be influenced by patient selection, institutional pathways, and the experience curve of technically demanding procedures, which are highly relevant when interpreting observational comparisons[13,20]. Meanwhile, studies of transcervical inflatable mediastinoscopic esophagectomy and related techniques have reported acceptable short-term outcomes and lymph node yields when compared with thoracoscopic-assisted MIE but also underline the importance of careful patient selection and standardized technical protocols[22,23].

The recovery endpoints used in this study align with how recovery programs operationalize progress during the early postoperative period. The ERAS guidelines for esophagectomy and subsequent implementation studies describe standardized perioperative pathways in which early mobilization, careful fluid and analgesia management, and early nutritional progression are treated as structured targets. Higher pathway adherence has been associated with shorter hospitalization and improved complication profiles[17,19,24]. Enhanced recovery frameworks in broader surgical populations have also been linked to shorter hospital stays and more efficient use of critical care resources when compliance with the pathway elements is high[24-26]. In this context, the recovery pattern observed here can be described as coherent across multiple domains rather than driven by a single administrative metric. A composite recovery measure can complement the LoS by summarizing the milestone performance across domains, which may be useful when local discharge processes contribute to LoS variations.

Safety outcomes in this dataset were broadly similar between the groups, with most individual complications occurring infrequently. Multiple contemporary series have identified new-onset atrial fibrillation after esophagectomy as a frequent postoperative complication, with reported incidences commonly in the range of one-fifth to one-third of patients undergoing transthoracic procedures[27,28]. A key finding was the between-group difference in postoperative arrhythmia. Postoperative atrial fibrillation and related atrial arrhythmias after esophagectomy are well-recognized clinically significant events, as they have been associated with prolonged hospital stay and increased postoperative morbidity in contemporary series[29,30]. Therefore, the arrhythmia differences observed in this study are of interest for postoperative and nursing surveillance. Moreover, the number of events was small and the estimates should be interpreted cautiously within the context of this cohort. Interpretation of pulmonary outcomes was also influenced by baseline risk differences. This baseline difference should be considered when interpreting the recovery outcomes. Although multivariable models adjusted for FEV1 and COPD or other chronic lung disease, residual confounding related to pulmonary reserve, anesthetic decision-making, and surgeon selection cannot be fully excluded in this retrospective cohort. An important feature of this cohort was that the observation group contained more patients with chronic obstructive or other chronic lung diseases at baseline, whereas the postoperative pulmonary complication rates were not higher in this group. The procedure was performed with patients in a single supine position through a cervical incision and upper abdominal access without a thoracic incision, which avoided intraoperative repositioning and may reduce the need for prolonged one-lung ventilation. From a nursing perspective, these features are important because they influence the profile of postoperative pain, respiratory care, and the potential for early mobilization in patients with poor pulmonary reserves. The current literature addressing mediastinoscopic and transcervical approaches highlights the potential benefit of selecting patients who are at an increased risk for undergoing traditional transthoracic approaches. However, comparative data consist of heterogeneous reports and are commonly restricted by selection and technical variations[12,20,21].

Adjusted analyses were performed to report associations after adjusting for baseline characteristics. In the present study, the surgical approach remained associated with LoS in the adjusted model and with the composite complication endpoint. The use of a composite complication endpoint can be methodologically helpful when individual complications are uncommon; however, it also combines clinically distinct events and should be read in parallel with the individual complication profile. Similar analytical strategies and interpretive caution were common in recent comparative studies and pooled analyses of mediastinoscopy-assisted and transmediastinal esophagectomy variants[12,20,21].

Technical learning curve effects are another important consideration when interpreting our findings. MIE in general, and transcervical or mediastinoscopy-assisted techniques in particular, are technically demanding. Pooled analyses of MIE have suggested that > 100 cases may be required before key outcomes such as operative time, blood loss, and complication rates stabilize[31,32]. Learning curve reports from programs implementing minimally invasive transcervical esophagectomy describe staged improvements in efficiency and safety over time, and stress the need for structured training and team experience[32-34]. The current study did not formally analyze the learning curve, and it is possible that some of the observed differences in perioperative metrics reflected changing experiences during the study period.

Despite the strengths of this study, several limitations must be acknowledged. The retrospective design means that residual confounding factors cannot be excluded[13,20]. Individual complications occurred infrequently, which reduced the statistical power for single endpoints and widened the uncertainty around between-group estimates. Learning curve effects for mediastinoscopic and transcervical techniques may also have influenced the observed perioperative metrics; however, these were not formally analyzed in this study. Finally, this analysis focused on short-term perioperative and recovery outcomes, and long-term oncological endpoints were not assessed[21,35,36]. Although lymph node yield and analgesic-use variables were added to the revised analysis, detailed analgesic dosing, patient-controlled analgesia settings, and serial postoperative pain scores were not completely captured in the retrospective nursing records. Therefore, pain management should still be interpreted as a potential contributor to early mobilization and feeding tolerance.

Overall, the present findings provide clinically interpretable, nursing-relevant outcomes for an emerging esophagectomy strategy that combines pneumomediastinoscopic and laparoscopic components. These results are consistent with the short-term recovery profile characterized by earlier milestone attainment and shorter hospitalization in the observation group, along with generally similar short-term safety outcomes, with a notable difference in arrhythmia occurrence.

CONCLUSION

In this retrospective cohort of patients with early esophageal cancer, including those with impaired pulmonary function, LAPA-PMAE performed in a single supine position through a cervical incision and upper abdominal access without a thoracic incision was associated with shorter operative duration, lower blood loss, and a faster early recovery profile than conventional thoracoscopic-assisted, three-incision esophagectomy. The observation group achieved key postoperative milestones, such as earlier ambulation, oral intake, and drain removal, and a shorter postoperative hospital stay, while overall complication rates were similar and postoperative arrhythmias occurred less frequently in these patients. The results provide nursing-focused data regarding a single-position pneumomediastinal approach in a selected group of patients who may be less suitable candidates for traditional three-incision esophagectomy.