Introduction
Esophagectomy (EE) with immediate esophagoplasty is a high-risk intervention associated with high complication rates (up to 74 % [1])) and hospital mortality (up to 4.5 % in highly specialized centers [2]). Anastomosis insufficiency and conduit necrosis are the most dangerous postoperative complications. They determine duration of hospitalization, mortality, price of treatment, and patient’s quality of life after surgery. Traditionally, after EE surgeons strictly prohibit peroral feeding for 5–7 postoperative days. The body’s energy requirements during this period are met through parenteral or enteral nutrition via nasojejunal tube or jejunostomy. Jejunostomy is good not only for enteral nutrition in the early postoperative period, but also for prevention of weight loss during complications, adjuvant chemotherapy, or other situations when oral food intake does not meet the body’s needs [3, 4]. However, despite all the advantages of the technique, jejunostomy often leads to complications such as malabsorption (10–39 %), leakage (1.4–25 %), peristomal skin irritation around the tube (0.4–16 %), which worsen patient quality of life [5]. Additionally, in 7 % of patients, tube dislocation causes intestinal obstruction [6]. Furthermore, enteral tube feeding reduces saliva production and promotes oral cavity colonization by pathogenic microorganisms, increasing the risk of infectious complications [7].
Recently, researchers have proved the safety of early oral feeding (EOF) in upper gastrointestinal (GI) surgery [8]. It has been officially included in most enhanced recovery protocols. However, in esophageal surgery, the possibility of EOF is doubted by many specialists [9–12]. The Enhanced Recovery After Surgery (ERAS) society recommends early initiation of enteral nutrition after EE and concludes that additional researches are required to find the optimal nutrient delivery route [13]. But Sindler D.L. in the meta-analysis, conversely, summarizes that EOF after EE does not increase postoperative complications rate and promotes bowel function activation [14]. Therefore, researches in this area are in progress.
Objective
To evaluate the effect of EOF after EE with immediate esophagoplasty on postoperative complications and laboratory blood test dynamics.
Materials and methods
Study Design
From 2012–2024, 500 elective EE with immediate esophagoplasty were performed at the A.V. Vishnevsky National Medical Research Center of Surgery (NMRCS) for benign (54 %) and malignant (46 %) esophageal diseases. All patients were managed under the rationally accelerated perioperative rehabilitation (RAPOR) program based on interdisciplinary interaction and personalized patient treatment tactics.
We conducted a retrospective study for evaluation patients’ treatment outcomes. The study was approved by the local ethics committee of the NMRCS, protocol No. 007-2024 dated 25.10.2024.
Within the RAPOR program framework, all patients were examined by an interdisciplinary team at the pre-hospital stage to determine complication risk and prescribe preoperative preparation plan. Patients were evaluated for body mass index (BMI), Charlson comorbidity index, anesthesiological risk according to American Society of Anesthesiologists (ASA) standards, cardiac complication risk according to Lee scale, nutritional deficiency risk according to Nutrition Risk Screening 2002 (NRS-2002), as well as clinical-laboratory and instrumental data. Preoperative surgical correction of cardiac and vascular pathology was performed in 35 (7 %) patients. Nutritional status correction was needed in 109 (21.8 %) patients. It included gastrostomy installation (5.2 %), nasogastric tube placement (3.8 %), stenting (0.8 %) or esophageal bougienage (3.4 %), sipping prescription (6.2 %) and/or dietary changes for patients with benign esophageal diseases and when neoadjuvant treatment was indicated in oncological patients. Preparation was performed outpatient for weeks or even months. When ambulatory nutritional status correction was impossible (malignant esophageal neoplasm) or ineffective, 6.2 % of patients received combined nutrition (parenteral and enteral) for 7–14 days in hospital before surgery. The objective criterion for completing prerehabilitation was increased/normalized serum prealbumin level.
Maintenance of normothermia and normovolemia, protective lung ventilation, prevention of postoperative nausea and vomiting, thromboembolic and infectious complications were conducted during the intraoperative period. All patients underwent sub/total EE, with 2.6 % in combination with gastrectomy or subtotal proximal gastric resection. All operations were performed by one surgeon using open transthoracic (44.6 %) or transhiatal (55.4 %) approaches with manual cervical anastomosis. Gastric tube (92.6 %), colon (6.2 %), or combined gastric-intestinal conduit (1.2 %) were used for esophagoplasty. The median intervention duration was 390 minutes. Jejunostomy was performed intraoperatively in 12 (2.4 %) patients due to high risk of complications in postoperative period. In our opinion, preventive jejunostomy after EE with esophagoplasty is inexpedient. Decompressive nasogastric tube and cervical drainage installation were discontinued in 2016.
Starting from 2017, 97.4 % of patients were extubated in the operating room and transferred to ICU for further observation. The median ICU stay was 0.67 days. On the morning of postoperative day (POD) 1, patients were transferred to the specialized department, mobilized. Urinary catheter and pleural drainage were removed provided aerostasis and less than 400 ml drainage output.
Standardly, all patients were prohibited oral water intake until POD 5 (first radiological or CT control day), prescribed infusion therapy of 30 ml/kg, including total parenteral nutrition 25–30 kcal/kg (standard postoperative nutrition group (SPN)). With satisfactory cervical anastomosis quality (positive intraoperative assessment of gastric tube and esophageal stump viability by surgeon), absence of aspiration and anastomotic insufficiency on radiological control, patients were allowed water intake from POD 1 and sipping from POD 2 (early oral feeding group, EOF). In patients with initial malnutrition, EOF was combined with parenteral nutrition. Daily fluid volume consumed both orally and parenterally was 30 ml/kg. Pureed food intake was possible from POD 4 after satisfactory second CT control.
Postoperative complications were evaluated according to the unified Esophagectomy Complications Consensus Group registry [15]. Isolated pleural effusion related to early drainage removal on POD 1 was not considered as a complication. Additionally, laboratory parameters (total lymphocyte count, albumin, prealbumin, transferrin levels) were evaluated preoperatively and on POD 1, 3–5, and 8–10.
Statistical analysis
Statistical data analysis was performed using R program, version 4.3.1, 2023. Descriptive statistics for categorical variables are presented as absolute and relative frequencies, for quantitative variables as mean with standard deviation (M ± σ) and median with interquartile range (Me [Q1; Q3]). P-values for categorical factors correspond to chi-square test, for quantitative variables — t-test.
Logistic regression including covariates (gender, age, BMI, comorbidity index, operation time) was used to analyze the association of EOF with complication development. For regression coefficients, p-values and 95 % confidence intervals calculated by Wald method were provided, using robust standard error estimates (sandwich type HC4).
Mixed linear models including covariates (gender, age, BMI, comorbidity index, operation time); analyzed parameter values before surgery (adjustment for baseline parameter values); categorical variable corresponding to the time point at which the parameter was measured (before surgery, POD 1, 3–5, 8–10) in interaction with nutrition type (EOF or SPN), as well as random intercept at patient level were used for laboratory parameter dynamics analysis. Models were fitted using maximum likelihood method with lmer function in lme4 package.
Overall significance of nutrition group association with laboratory parameter dynamics was evaluated using model-submodel tests with Satterthwaite correction (in lmerTest package). Mean difference evaluation in laboratory parameter values between groups was performed using Wald method for linear combinations of mixed model. Normal approximation was used for p-value and confidence interval calculation.
Regression model correctness analysis was performed visually using QQ-plots for residuals, residuals vs. fitted plots, scale-location diagrams, residuals vs. leverage plots, and partial residuals plots. Model quality was satisfactory in all cases.
Multiple comparison corrections were not applied. Data gap completion was not performed. Available case analysis was used. Statistical tests were performed at α=0.05 level.
Results
Study group characteristics
Comparison of EOF and SPN group patients is presented in Table 1.
| Parameter | SPN (n = 404) |
EOF (n = 96) |
Total (n = 500) |
p-value |
|---|---|---|---|---|
| Age (years) | ||||
| М ± σ | 57.6 ± 13.3 | 50.6 ± 13.8 | 56.3 ± 13.7 | < 0.001 |
| Me (Q1; Q3) | 59.0 (50.0; 66.3) | 52.5 (40.8; 62.3) | 59.0 (48.0; 66.0) | |
| Gender | ||||
| Male | 240 (59.4 %) | 59 (61.5 %) | 299 (59.8 %) | 0.79 |
| Female | 164 (40.6 %) | 37 (38.5 %) | 201 (40.2 %) | |
| Weight (kg) | ||||
| М ± σ | 69.5 ± 16.1 | 70.9 ± 15.5 | 69.8 ± 16.0 | 0.43 |
| Me (Q1; Q3) | 68.0 (56.0; 80.0) | 69.0 (58.0; 82.6) | 69.0 (57.0; 80.0) | |
| BMI (kg/m2) | ||||
| М ± σ | 24.1 ± 5.19 | 23.9 ± 4.61 | 24.0 ± 5.08 | 0.78 |
| Me (Q1; Q3) | 23.4 (20.3; 27.6) | 23.7 (20.7; 26.7) | 23.5 (20.4; 27.4) | |
| BMI category | ||||
| < 18.5 kg/m2 | 60 (14.9 %) | 14 (14.6 %) | 74 (14.8 %) | 0.94 |
| 18.5–24.9 kg/m2 | 191 (47.2 %) | 44 (45.8 %) | 235 (47 %) | |
| > 25 kg/m2 | 153 (37.8 %) | 38 (39.6 %) | 191 (38.2 %) | |
| Smoking | ||||
| No | 300 (74.3 %) | 66 (68.8 %) | 366 (73.2 %) | 0.34 |
| Yes | 104 (25.7 %) | 30 (31.3 %) | 134 (26.8 %) | |
| Diagnosis | ||||
| Achalasia | 77 (19.1 %) | 37 (38.5 %) | 114 (22.8 %) | < 0.001 |
| Esophageal stricture | 100 (24.8 %) | 28 (29.2 %) | 128 (25.7 %) | |
| Cancer | 204 (50.4 %) | 26 (27.1 %) | 230 (45.9 %) | |
| Esophagealfistula | 6 (1.5 %) | 0 (0 %) | 6 (1.2 %) | |
| Other | 17 (4.2 %) | 5 (5.2 %) | 22 (4.4 %) | |
| NRS-2002 | ||||
| М ± σ | 2.78 ± 0.871 | 2.54 ± 0.724 | 2.73 ± 0.849 | 0.0071 |
| Me (Q1; Q3) | 3.00 (2.00; 3.00) | 2.00 (2.00; 3.00) | 3.00 (2.00; 3.00) | |
| Comorbidity index | ||||
| М ± σ | 3.68 ± 2.49 | 2.41 ± 2.24 | 3.44 ± 2.49 | < 0.001 |
| Me (Q1; Q3) | 3.00 (2.00; 6.00) | 2.00 (0; 4.00) | 3.00 (1.00; 5.00) | |
| Lee Scale | ||||
| М ± σ | 1.25 ± 0.546 | 1.18 ± 0.459 | 1.24 ± 0.530 | 0.18 |
| Me (Q1; Q3) | 1.00 (1.00; 1.00) | 1.00 (1.00; 1.00) | 1.00 (1.00; 1.00) | |
| ASA | ||||
| М ± σ | 2.69 ± 0.658 | 2.43 ± 0.677 | 2.64 ± 0.669 | < 0.001 |
| Me (Q1; Q3) | 3.00 (2.00; 3.00) | 2.00 (2.00; 3.00) | 3.00 (2.00; 3.00) | |
| Weight loss > 10 % in 6 months | ||||
| No | 283 (70.0 %) | 75 (78.1 %) | 358 (71.5 %) | 0.14 |
| Yes | 121 (30.0 %) | 21 (21.9 %) | 142 (28.5 %) | |
| Total intraoperative infusion therapy volume, minus diuresis and blood loss compensation (ml/kg/h) | ||||
| М ± σ | 4.42 ± 1.71 | 4.45 ± 1.53 | 4.43 ± 1.68 | 0.88 |
| Me (Q1; Q3) | 4.29 (3.29; 5.22) | 4.32 (3.19; 5.54) | 4.29 (3.26; 5.26) | |
| Duration ofoperation (min) | ||||
| М ± σ | 417 ± 106 | 349 ± 79.8 | 404 ± 105 | < 0.001 |
| Me (Q1; Q3) | 415 (330; 481) | 335 (295; 385) | 390 (320; 470) | |
| Duration of anesthesia (min) | ||||
| М ± σ | 512 ± 115 | 430 ± 90.0 | 496 ± 115 | < 0.001 |
| Me (Q1; Q3) | 500 (420; 590) | 413 (360; 466) | 480 (400; 570) | |
| Approach | ||||
| Transthoracic | 194 (48.1 %) | 28 (29.2 %) | 222 (44.5 %) | 0.0012 |
| Transhiatal | 209 (51.9 %) | 68 (70.8 %) | 277 (55.5 %) | |
| Type of transplant | ||||
| Gastric tube | 366 (90.8 %) | 96 (100 %) | 462 (92.6 %) | 0.0086 |
| Colon | 31 (7.7 %) | 0 (0 %) | 31 (6.2 %) | |
| Combined | 6 (1.5 %) | 0 (0 %) | 6 (1.2 %) | |
According Table 1, EOF group more often included younger and less comorbid patients with benign esophageal diseases after faster transhiatal EE with immediate gastric tube reconstruction. These patients more often met the EOF group inclusion criteria.
Effect of EOF on postoperative complications
When comparing EOF and SPN groups, statistically significant reduction in overall complication frequency was noted with oral feeding initiation from POD 1 (Table 2). The frequency of anastomotic insufficiency and conduit necrosis in groups was comparable. Low complication frequency and early oral feeding initiation led to significant reduction of hospitalization duration to 8 days. No fatal outcomes were noted in the EOF group.
| Parameter | SPN (n = 404) |
EOF (n = 96) |
Total (n = 500) |
p-value |
|---|---|---|---|---|
| Complications | ||||
| No | 272 (67.3 %) | 82 (85.4 %) | 354 (70.8 %) | < 0.001 |
| Yes | 132 (32.7 %) | 14 (14.6 %) | 146 (29.2 %) | |
| Anastomotic insufficiency/ conduit necrosis | ||||
| Yes | 29 (7.2 %) | 4 (4.2 %) | 33 (6.6 %) | 0.4 |
| No | 375 (92.8 %) | 92 (95.8 %) | 467 (93.4 %) | |
| Hospitalization duration (days) | ||||
| М ± σ | 11.8 ± 8.57 | 9.57 ± 6.91 | 11.3 ± 8.32 | 0.0086 |
| Me (Q1; Q3) | 10.0 (8.00; 11.3) | 8.00 (7.00; 9.00) | 9.00 (8.00; 11.0) | |
| Mortality | ||||
| No | 397 (98.3 %) | 96 (100 %) | 493 (98.6 %) | 0.41 |
| Yes | 7 (1.7 %) | 0 (0 %) | 7 (1.4 %) | |
In the first research stage, a basic regression model was created to evaluate postoperative complication risk (Table 3). It included age and gender as “universal confounders”, BMI (nutritional status evaluation criterion), comorbidity index (concomitant pathology severity criterion), and operation duration (operation complexity criterion).
| Adjusted odds ratios (OR) | ||||
|---|---|---|---|---|
| Parameter | OR | p- value | 95 % CI | |
| L | U | |||
| (Constant) | 0.0794 | 0.003 | 0.0151 | 0.417 |
| Age (years) | 1.01 | 0.489 | 0.984 | 1.03 |
| Female | 0.809 | 0.333 | 0.526 | 1.24 |
| BMI | 0.975 | 0.209 | 0.937 | 1.01 |
| Comorbidity index | 1.13 | 0.062 | 0.994 | 1.28 |
| Duration of operation (min) | 1.00 | 0.001 | 1.00 | 1.01 |
Based on this model, the association between complication development and nutrition type was tested. Considering associations with age, gender, BMI, comorbidity index, and operation duration, we established that EOF acted as a protective factor and significantly reduced complication probability by 50 % (5–75 %) (Table 4).
| Adjusted odds ratios (OR) | ||||
|---|---|---|---|---|
| Parameter | OR | p- value | 95 % CI | |
| L | U | |||
| (Constant) | 0.114 | 0.013 | 0.0207 | 0.628 |
| Age (years) | 1.01 | 0.635 | 0.981 | 1.03 |
| Female | 0.781 | 0.262 | 0.507 | 1.20 |
| BMI | 0.979 | 0.286 | 0.941 | 1.02 |
| Comorbidity index | 1.12 | 0.079 | 0.987 | 1.28 |
| Duration of operation (min) | 1.00 | 0.004 | 1.00 | 1.01 |
| Early oral feeding | 0.498 | 0.035 | 0.260 | 0.952 |
After constructing a single-factor model, similar results were obtained. EOF reduced complication rate by 35–80 % (Table 5).
| Unadjusted odds ratios (OR) | ||||
|---|---|---|---|---|
| Parameter | OR | p- value | 95 % CI | |
| L | U | |||
| (Constant) | 0.487 | 0.000 | 0.396 | 0.600 |
| Early oral feeding | 0.351 | 0.001 | 0.190 | 0.646 |
Evaluation of EOF effect on laboratory parameters
Descriptive statistics for laboratory parameters are presented in Table 6. Since 12-year patient treatment results were evaluated retrospectively, 50–95 % of parameters are missing.
| Parameter | Before surgery | POD 1 | POD 3–5 | POD 8–10 |
|---|---|---|---|---|
| Absolute lymphocyte count (×109 /l) | ||||
| М ± σ | 1.77 ± 0.560 | 1.08 ± 0.449 | 1.12 ± 0.493 | 1.42 ± 0.576 |
| Me (Q1; Q3) | 1.76 (1.40; 2.19) | 1.04 (0.740; 1.44) | 1.05 (0.710; 1.39) | 1.43 (0.978; 1.73) |
| Data | 94 (18.8 %) | 76 (15.2 %) | 93 (18.6 %) | 68 (13.6 %) |
| Albumin (g/l) | ||||
| М ± σ | 40.3 ± 5.15 | 29.9 ± 4.11 | 31.6 ± 3.85 | 33.3 ± 4.37 |
| Me (Q1; Q3) | 40.5 (37.9; 43.0) | 30.0 (28.0; 32.1) | 32.0 (29.0; 34.2) | 33.0 (30.6; 36.0) |
| Data | 357 (71.4 %) | 305 (61.0 %) | 308 (61.6 %) | 246 (49.2 %) |
| Transferrin (mg/dl) | ||||
| М ± σ | 237 ± 63.1 | 173 ± 52.8 | 168 ± 48.9 | 168 ± 50.5 |
| Me (Q1; Q3) | 231 (195; 269) | 154 (140; 200) | 165 (131; 193) | 166 (136; 197) |
| Data | 152 (31.6 %) | 43 (8.6 %) | 133 (26.6 %) | 128 (25.6 %) |
| Prealbumin (g/l) | ||||
| М ± σ | 0.300 ± 0.174 | 0.380 ± 0.215 | 0.196 ± 0.150 | 0.209 ± 0.143 |
| Me (Q1; Q3) | 0.250 (0.220; 0.300) | 0.340 (0.210; 0.560) | 0.145 (0.120; 0.200) | 0.160 (0.130; 0.220) |
| Data | 172 (34.4 %) | 23 (4.6 %) | 148 (29.6 %) | 105 (21.0 %) |
For each parameter, an association test of its dynamics with EOF was conducted, considering adjustment for parameter value before surgery. Since laboratory parameter values were not available for all patients, this association was then studied in detail using a reference patient example (60-year-old male with comorbidity index of 3 and preoperative albumin 40 g/l; absolute lymphocyte count 1.8 × 109/l; prealbumin 0.3 g/l; transferrin 237 mg/dl). Model-based calculated values and differences between them for the reference patient with adjustment for gender, age, comorbidity index, operation duration, and preoperative parameter level in SPN and EOF groups are reflected in Figures 1 and 2.
When conducting association test, p < 0.05 was obtained for albumin (p = 0.0021) and prealbumin (p = 0.025). No association between transferrin dynamics (p = 0.16), absolute lymphocyte count (p = 0.53), and postoperative nutrition type was found.
Association using reference patient example showed:
- In the EOF group, mean albumin level was significantly higher on POD 3–5 (Figure 2).
- In the EOF group, mean prealbumin level was significantly higher on POD 1. However, POD 1 prealbumin level is known only for 4.6 % of patients, while on POD 3–5 and 8–10 for 29.6 % and 21 %. In our view, the absence of substantial POD 1 data affected the obtained results.
- In the EOF group, mean transferrin level was significantly higher on POD 8–10. However, no association between transferrin level and nutrition type was found in the general test, so additional studies are required to confirm results.
Fig. 1. Albumin, prealbumin, transferrin and the absolute number of lymphocytes dynamics for the reference patient in the standard and early oral feeding groups
Fig. 2. The difference in the expected albumin, prealbumin, transferrin and the absolute number of lymphocytes dynamics value between the standard and early oral feeding groups
Discussion
At the beginning of the 21st century frequency of anastomotic insufficiency reached 35 %. For its reduction surgeons suggested to increase oral water and nutrient intake prohibition to 28 days [16]. This allowed to reduce anastomotic insufficiency rate from 12.7 % to 2.7 % [10]. In similar work, Bolton J.S. et al. delayed oral feeding until POD 12 and achieved anastomotic insufficiency reduction from 23 % to 3 % [11]. In both studies, jejunostomy was routinely performed and enteral nutrition prescribed for all patients.
However, over time, the ERAS paradigm began gaining popularity, and physicians started investigating EOF effects on postoperative complications. Chinese specialists were among the first to prove EOF possibility and safety. They first tested the protocol on 68 patients [17]. Then conducted a randomized study including 280 patients (140 in EOF group) [18]. They performed thoracolaparoscopic EE with immediate esophagoplasty and manual three-layer cervical anastomosis. Oral water intake was possible from surgery day, sipping from POD 1, soft food intake (noodles, rice, bread) from POD 2. Overall complications and anastomotic insufficiency were comparable in both groups (34.3 % and 3.6 % in EOF group, 38.6 % and 4.3 % in SPN group). Additionally, EOF promoted earlier flatulence, defecation and better quality of life 2 weeks after surgery. However, it should be noted that only patients without severe somatic pathology (Charlson index 0–2) were included. It also explains low complication rate, absence of hospital mortality and conduit necrosis.
Meanwhile in Europe, EOF safety was first studied in 50 oncological patients after Lewis operation with stapler intrapleural anastomosis. Their treatment results were compared with retrospective data. Anastomotic insufficiency frequency was 14 % in EOF group and 24 % in SPN group, but difference was statistically insignificant. Researchers concluded that early oral feeding does not increase complication risk [19]. They also evaluated long-term EOF effects. It appeared that in the first month, mean BMI reduction in EOF group was greater due to absence of nutritional jejunostomy. However, no significant differences between groups were subsequently found [20]. The same scientists launched a randomized study including 132 patients (65 in EOF group). Anastomotic insufficiency frequency in EOF and SPN groups did not differ statistically significantly (18.5 % and 16.4 % respectively) [21]. They also evaluated long-term results: in EOF group, overall, 3-year and disease-free 5-year survival were statistically higher [22].
In Japan, EOF after upper GI operations was also investigated, concluding that it is not only safe but also promotes bowel function recovery and preoperative patient quality of life [23]. However, in this study, among 54 patients in EOF group, only 33 underwent EE with immediate esophagoplasty.
Meanwhile, against the background of increasing EOF popularity, a retrospective study proving jejunostomy benefit and oral water intake prohibition until POD 15 was published in USA in 2018. This allowed to reduce anastomotic insufficiency from 14.5 % to 4.2 % [9]. Danish scientists reached similar conclusions based on retrospective analysis [12].
Meta-analyses that included the studies above proved EOF safety. In X. Li et al. meta-analysis, delayed feeding initiation advantage in open esophageal surgery and EOF advantage in minimally invasive surgery are mentioned [24]. In more recent D.L. Sindler et al. meta-analysis, EOF does not increase complication rate and promotes GI recovery [14]. M. Shi et al. meta-analysis concludes that EOF improves patient quality of life. However, due to small sample size, additional research in this area is required [25]. Interestingly, all 3 meta-analyses evaluate the same studies of patients operated from 2004–2015.
At NMRCS 500 EE with immediate esophagoplasty procedures were performed within RAPOR program framework. But only 45.9 % of patients were operated for malignant neoplasm. Foreign studies included only oncological patients. All patients underwent open transthoracic or transhiatal EE with manual cervical anastomosis. 96 (19.2 %) patients followed EOF protocol. EOF was permitted exclusively with good gastric conduit and esophageal stump blood supply and quality of esophageal anastomosis. This group included younger and less comorbid patients after shorter interventions. Therefore, overall complication frequency and specifically anastomotic insufficiency/conduit necrosis in EOF group were lower than in SPN group (14.6 % vs. 32.7 % [p < 0.001] and 4.2 % vs. 7.2 % [p = 0.04]). Also, due to patient selection for EOF initiation, this group had no fatal outcomes.
Additionally, at NMRCS, dynamics of laboratory parameter in EOF and SPN groups were evaluated. Nutritional status changes were assessed based on plasma protein concentrations: albumin, transferrin, and prealbumin. They have different half-life periods — 20 days, 7 days, and 2 days [26]. Absolute leukocyte count was measured to evaluate immunosuppression. Despite absence of several measurements, calculations allowed detecting only significant albumin level increase on POD 3–5 in EOF group. This was probably related to less fluid retention in the body due to larger volumes of water and nutrients consumed orally. On POD 8–10, albumin level difference was eliminated. It can be concluded that postoperative nutrition type is not reflected in laboratory parameter dynamics with short half-life periods. However, literature data for obtained result comparison could not be found.
Perioperative EE management with immediate esophagoplasty within RAPOR program framework allowed to reduce overall anastomotic insufficiency and conduit necrosis frequency to 6.6 %. Additionally, it allowed some patients to initiate oral water intake from POD 1 and pureed food intake from POD 4 safely. This promoted not only successful early patient mobilization but also reduced postoperative hospital stay duration to POD 8. Previously, when conducting prospective randomized study, we noted that with EOF initiation, patients had significantly earlier flatulence and defecation [27]. However, at retrospective study its evaluation was impossible.
Conclusion
Within RAPOR program framework, early oral feeding after esophagectomy with immediate esophagoplasty is effective and safe. Its routine practice application is possible only in high-volume centers with close interdisciplinary interaction and personalized patient treatment tactics. Additional multicenter studies are required for confirmation. Furthermore, additional research is needed to evaluate long-term early oral feeding effects on patient nutritional status.
Disclosure. The authors declare no competing interests.
Author contribution. All authors according to the ICMJE criteria participated in the development of the concept of the article, obtaining and analyzing factual data, writing and editing the text of the article, checking and approving the text of the article.
Ethics approval. This study was approved by the local Ethical Committee of Federal State Budget Institution “A.V. Vishnevsky National Medical Research Center of Surgery” of the Ministry of Health of the Russian Federation (reference number: 7-25.10.2024).
Funding source. This study was not supported by any external sources of funding.
Data Availability Statement. The data that support the findings of this study are available from the corresponding author upon reasonable request.
