The clinical relevance of postoperative routine contrast swallow following esophagectomy

Abstract

Background

Limited evidence exists regarding the role of the routine contrast swallow in determining the safety of initiating oral intake following esophagectomy. Therefore, this study evaluated its clinical relevance in decision-making regarding oral intake.

Methods

A descriptive, observational, single-center study included esophageal cancer patients who underwent esophagectomy with gastric conduit reconstruction between March 2019 and March 2021. Patients were kept nil by mouth until postoperative day (POD) 3 and began oral intake from POD 4 if the contrast swallow was normal. The primary outcome was routine use (POD 2–6) of the contrast swallow and its impact on oral intake decisions. Predictors of an abnormal contrast swallow were explored using logistic regression.

Results

A contrast swallow was routinely performed in 80 of 110 patients (73%), of which 48 (44%) had a normal contrast swallow, while 32 (29%) had an abnormal contrast swallow. Ultimately, in 28 patients (25%), findings of the contrast swallow led to a decision to delay oral intake. All abnormal contrast swallows (n=32) involved aspiration (n=30, 94%) or laryngeal penetration (n=2, 6%). Aspiration occurred alongside delayed gastric conduit emptying in 2 cases, and anastomotic leakage was suspected in 2 others but not confirmed on computed tomography (CT) scan. No statistically independent predictors of an abnormal contrast swallow were identified in multivariable analysis.

Conclusions

Routine contrast swallow following esophagectomy led to withholding oral intake in 25% of cases. Future research should explore a more selective approach in specific patient categories.

Highlight box.

Key findings

• Routine contrast swallow following esophagectomy resulted in withholding oral intake in 25% of patients.

• Decisions were primarily based on observed swallowing dysfunction, particularly aspiration.

What is known and what is new?

• Delayed gastric emptying and swallowing impairment are among the most common functional complications after esophagectomy.

• This study shows that routine use of contrast swallow led to protocol deviations in 25% of cases, as the results of the contrast swallow influenced the decision not to initiate oral intake mostly because of aspiration, potentially preventing the adverse effects of aspiration such as pneumonia.

What is the implication, and what should change now?

• A more selective approach, guided by clinical swallowing assessment by a speech-language pathologist and focused on identifying high-risk patients, may better inform the use of contrast swallows.

• Future research should explore a more selective approach in specific patient categories, as well as the potential benefits of earlier timing following esophagectomy.

Introduction

Background

Esophagectomy for esophageal cancer is associated with significant postoperative morbidity. Among the most common functional complications are delayed gastric conduit emptying and swallowing impairment (1). Delayed gastric conduit emptying typically results from impaired gastric motility and pyloric dysfunction, often linked to vagal nerve transection during surgery, which limits the conduit’s ability to empty effectively into the intestine (1). Swallowing dysfunction, commonly referred to as dysphagia, is attributed to several factors, including vocal cord immobility, reconstruction route, and prolonged duration of surgery (2,3).

A common method to assess gastric conduit passage and swallowing function is the contrast swallow, which was historically used to detect anastomotic leakage. However, evidence indicates that routine contrast swallow for leakage detection offers no additional benefit when there is no clinical suspicion of leakage, with poor sensitivity and high false negative error rates (4-7). A retrospective study, however, showed that performing a routine contrast swallow on postoperative day (POD) 3 or 4 to evaluate passage after esophagectomy enabled earlier nasogastric tube removal and earlier resumption of oral intake compared to the routine practice of capping the nasogastric tube at POD 5 (8). In this light, contrast swallow exams may be a valuable part of the enhanced recovery after surgery (ERAS) protocol after esophagectomy.

Rationale and knowledge gap

To date, limited evidence exists regarding the role of a routine contrast swallow in determining the timing of oral intake initiation following esophagectomy.

Objective

Therefore, the aim of the current study was to assess the clinical relevance of the routine contrast swallow in decision-making regarding oral intake after esophagectomy for esophageal cancer in a Western high-volume center. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2037/rc) (9).

Methods

Study design and population

This descriptive, observational cohort study included all consecutive patients who were treated for esophageal cancer with an esophagectomy at the University Medical Center Utrecht from March 2019 to March 2021. Procedures included transthoracic or transhiatal esophagectomy with gastric conduit reconstruction using an intrathoracic or cervical handsewn end-to-side anastomosis. There were no exclusion criteria. Data were extracted from a prospectively maintained institutional database. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional ethical board of the University Medical Center Utrecht (No. 13-061). Informed consent was waived as this study utilized data from routine clinical care in an existing prospective database, which was reviewed and exempted from the Medical Research Involving Human Subjects Act (WMO) by our institutional ethics board.

Surgical procedure

The preferred surgical approach was a transthoracic esophagectomy with gastric conduit reconstruction and two-field lymphadenectomy, performed using a total robot-assisted technique. Transhiatal procedures were reserved for patients with significant comorbidities, as determined during multidisciplinary team meetings, and were carried out using open, laparoscopic, and robot-assisted techniques. Depending on the surgical approach, either a cervical or intrathoracic hand-sewn end-to-side anastomosis was constructed.

ERAS protocol

Since 2019 a late oral feeding protocol has been implemented at the University Medical Center Utrecht as part of the ERAS protocol, in which patients remain nil by mouth until POD 3. On POD 4, a contrast swallow is performed, and if the results and clinical situation are satisfactory, oral intake is gradually initiated with sips of water followed by a liquid diet. Additionally, the intraoperatively placed nasogastric tube is removed if the contrast swallow does not show dilation of the gastric conduit.

Patients routinely receive an intraoperative placed jejunostomy, and enteral tube feeding begins on POD 1, increasing gradually if bowel movements have resumed and there are no signs of ileus. Upon discharge, patients continue enteral tube feeding via jejunostomy alongside a liquid diet.

Contrast swallow

Per protocol, patients were scheduled to undergo a contrast swallow on POD 4; however, due to logistical constraints, a timeframe of POD 2–6 was considered acceptable as ‘per protocol’. A contrast swallow was performed using water-soluble contrast (Visipaque; GE Healthcare, Oslo, Norway). While the patient swallowed the contrast, real-time images were taken under fluoroscopy from two angles: anterior-posterior and lateral. The contrast swallow was postponed in case of a complicated course or prolonged intensive care unit (ICU) stay, After the contrast swallow, an expert radiologist assessed and reported on three main aspects: (I) swallowing function (laryngeal penetration—defined as passage of contrast into the larynx without passing the vocal cords; and/or aspiration—defined as passage of contrast beyond the vocal cords and into the airway/lungs); (II) dilation and passage of the gastric conduit; and (III) signs of leakage at the esophagogastric anastomosis. Delayed gastric conduit emptying was defined on the contrast swallow based on the expert radiologist’s judgement of delayed contrast passage (10).

If no abnormalities were reported on these aspects, the contrast swallow was considered normal. If the radiological report was inconclusive, the contrast swallow was retrospectively reexamined by the study team (C.D.K., K.H.E.S. and R.v.H.).

A speech-language pathologist was involved during the contrast swallow based on clinical indication, such as in cases of weakness or suspicion of recurrent laryngeal nerve (RLN) injury, to assist the examination by providing swallowing instructions. As their involvement was indication-based, the presence of a speech-language pathologist differed between patients.

Outcomes

The primary outcome was the routine use of the contrast swallow and the decision to delay oral intake based on its findings. Secondary outcomes included: reasons for protocol non-adherence, involvement of a speech-language pathologist during and following the contrast swallow, the use of a computed tomography (CT) scan due contrast swallow findings, median day of oral intake initiation, median day of nasogastric tube removal, and diet at discharge. Additionally, postoperative outcomes, such as complications, ICU stay, and length of hospital stay, were assessed.

Statistical analysis

All analyses were performed using SPSS version 30.0.0 (IBM SPSS Statistics). Continuous variables were expressed as means with standard deviations (SDs) or medians with interquartile ranges (IQR), depending on data distribution. Categorical variables were presented as numbers with percentages. To explore potential predictors of an abnormal contrast swallow, both univariable and multivariable logistic regression analyses were performed. Independent variables considered included age, body mass index (BMI), gender, tumor location, surgical approach, location of the anastomosis, exposure to preoperative radiation, suspicion of RLN injury, and presence of a speech-language pathologist during the contrast swallow. Each factor with P<0.100 in the univariable analysis was included in the multivariable analysis. Results are expressed as odds ratios (ORs) with 95% confidence intervals (CI). P<0.05 was considered statistically significant.

Results

Patient characteristics

The characteristics of the 110 included patients that underwent an esophagectomy from March 2019 to March 2021 and followed the late oral feeding protocol are shown in Table 1. The majority were male (n=82, 75%), with a mean age of 67 years. Most patients had a tumor of the distal esophagus (n=72, 66%) or gastroesophageal junction (n=24, 22%). All patients were treated with a minimally invasive esophagectomy (MIE) or robot-assisted minimally invasive esophagectomy (RAMIE), and 69% (n=76) underwent transthoracic esophagectomy (cervical anastomosis: n=37, 34%; intrathoracic anastomosis: n=39, 35%), while 31% (n=34) underwent transhiatal esophagectomy.

Table 1. Characteristics of all included patients (n=110).

Characteristics	Values
Gender
Male	82 [75]
Female	28 [26]
Age at time of operation, years	66.7±8.6
BMI, kg/m2	25.1±4.2
ASA-score
I	3 [3]
II	53 [48]
III	50 [46]
IV	4 [4]
Tumor location
Proximal	3 [3]
Mid	11 [10]
Distal	72 [66]
GEJ	24 [22]
Clinical T-stage
cT1	3 [3]
cT2	16 [15]
cT3	80 [73]
cT4	11 [10]
Clinical N-stage
cN0	50 [46]
cN+	59 [54]
cNx	1 [1]
Tumor histology
Adenocarcinoma	88 [80]
Squamous cell carcinoma	22 [20]
Neoadjuvant treatment
No neoadjuvant treatment	17 [16]
Chemotherapy	13 [12]
Chemoradiotherapy	80 [73]
Approach
TH (cervical anastomosis)	34 [31]
TT (cervical anastomosis)	37 [34]
TT (intrathoracic anastomosis)	39 [35]
MIE versus RAMIE
MIE	27 [24]
RAMIE	83 [74]

Data are presented as n [%] or mean ± SD. ASA, American Society of Anesthesiologists; BMI, body mass index; GEJ, gastroesophageal junction; MIE, minimally invasive esophagectomy; N, node; RAMIE, robot-assistant minimally invasive esophagectomy; SD, standard deviation; T, tumor; TH, transhiatal; TT, transthoracic.

Use of contrast swallow

Figure 1 presents a flowchart depicting the use of the contrast swallow in our cohort (n=110). Of the 110 patients, 8 (7%) did not undergo a routine contrast swallow, primarily due to a complicated postoperative course followed by death (n=3, 3%) or an unsafe swallow function caused by clinical aspiration (n=3, 3%). In 22 cases (20%), the contrast swallow was not performed between POD 2–6, due to respiratory insufficiency (n=7, 6%), (suspected) anastomotic leakage (n=5, 5%), both (n=3, 3%), or other reasons (n=7, 6%). These patients were excluded from further analyses.

Figure 1.

Flowchart depicting the (per protocol) use of the contrast swallow in our cohort (n=110) and the subsequent decision to expand oral intake. POD, postoperative day.

Oral intake initiation after routine contrast swallow

A contrast swallow was routinely performed in 80 of 110 patients (73%), of which 48 (44%) had a normal contrast swallow, while 32 (29%) had an abnormal contrast swallow, as depicted in Figure 1. Ultimately, the contrast swallow findings led to withholding oral intake initiation in 25% of patients (n=28), deviating from the standard oral feeding protocol.

Normal contrast swallow

Of the 48 patients who had a normal contrast swallow, oral intake was initiated in 43 (90%). Five patients did not start oral intake due to increasing inflammatory parameters/increasing oxygen requirements (n=2), anastomotic leakage on CT (n=1), unsafe swallowing function by speech-language pathologist assessment (n=1), and high nasogastric tube output (n=1). An overview of the outcomes is presented in Table 2.

Table 2. Findings from the contrast swallow, decisions regarding oral intake initiation, and postoperative outcomes, stratified by contrast swallow results and oral intake initiation status.

Variables	Normal contrast swallow				Abnormal contrast swallow
	All (n=48)	Oral intake initiated (n=43)	Oral intake not initiated (n=5)		All (n=32)	Oral intake initiated (n=4)	Oral intake not initiated (n=28)
Contrast swallow
Day contrast swallow	4 [4–4]	4 [4–4]	4 [4–5]		4 [4–4]	4 [4–4]	4 [4–4]
Abnormal findings contrast swallow†	0	0	0		32 [100]	4 [100]	28 [100]
Laryngeal penetration	–	–	–		2 [6]	1 [25]	1 [4]
Aspiration	–	–	–		30 [94]	3 [75]	27 [96]
Signs of delayed gastric emptying	–	–	–		2 [6]	0	2 [7]
Suspicion of anastomotic leakage	–	–	–		2 [6]	0	1 [4]
Involvement speech-language pathologist
During contrast swallow	21 [44]	20 [47]	1 [20]		8 [25]	2 [50]	6 [21]
Following contrast swallow	7 [15]	4 [9]	3 [60]		30 [94]	4 [100]	26 [93]
CT because of findings contrast swallow	0	0	0		3 [9]	0	3 [11]
Oral intake
Day oral intake initiation	4 [4–5]	4 [4–4]	6 [5–12]		8 [5–15]	4 [4–4]	9 [6–19]
Reason oral intake not initiated
Findings contrast swallow	0	–	0		26 [82]	–	26 [93]
Other	5 [10]	–	5 [100]		2 [6]	–	2 [7]
Day removal of nasogastric tube	4 [4–5]	4 [4–5]	4 [4–6]		5 [4–7]	4 [3–5]	5 [4–7]
Postoperative outcomes
Complications	35 [73]	31 [72]	4 [80]		27 [84]	3 [75]	24 [86]
Anastomotic leakage	9 [19]	8 [19]	1 [20]		5 [16]	0	5 [18]
Pneumonia	15 [31]	13 [30]	2 [40]		11 [34]	0	11 [39]
Recurrent laryngeal nerve injury	3 [4]	2 [5]	1 [20]		10 [32]	1 [25]	9 [32]
Length of hospital stay (days)	11 [8–13]	11 [8–14]	9 [8–12]		10 [8–13]	10 [8–14]	10 [8–14]
Length of ICU-stay (days)	1 [1–2]	1 [1–2]	1 [1–2]		1 [1–1]	1 [1–1]	1 [1–1]
Oral intake at discharge‡
Solid	1 [2]	1 [2]	0		0	0	0
Liquid	40 [83]	36 [84]	4 [80]		23 [72]	4 [100]	19 [68]
No	6 [13]	5 [12]	1 [20]		9 [28]	0	9 [32]

Data are presented as n [%] or median [IQR]. ^†, contrast swallow could show more than one abnormal finding; ^‡, 1 patient died during hospitalization. CT, computed tomography; ICU, intensive care unit; IQR, interquartile range.

A speech-language pathologist was involved in 21 patients (44%) during contrast swallow for routine swallowing assessments (n=16, 33%) or hoarseness/dysphonia (n=5, 10%). Median day of nasogastric tube removal and oral intake initiation was POD 4 (IQR, 4–5). Among patients with a normal contrast swallow, 35 (73%) had postoperative complications, including 15 (31%) with pneumonia. RLN injury occurred in 3 patients (4%), all on the left side (grade 1, Clavien Dindo 1), 2 of whom were suspected and confirmed prior to the contrast swallow. The median ICU-stay was 1 day (IQR, 1–2 days). Median hospital stay was 11 days (IQR, 8–13 days). One patient died during hospitalization, due to postoperative complications.

At discharge, 40 patients (83%) were on a liquid diet according to protocol, while 6 patients (13%) had no oral intake due to anastomotic leakage (n=4), suspicion of chyle leakage (n=1) or a dilated gastric conduit (n=1). One patient was discharged on a solid diet after jejunostomy luxation.

Abnormal contrast swallow

Among the 32 patients with an abnormal contrast swallow, aspiration was the primary finding in 30 (94%), often with a late or absent cough reflex (n=9). Two also showed delayed gastric emptying. In two other patients, contrast swallow findings raised suspicion of anastomotic leakage, which was not confirmed on CT. In the remaining 2 patients (6%), the contrast swallow showed laryngeal penetration.

Oral intake was withheld in 28 patients (88%) (Figure 1), based solely on contrast swallow findings in 26 patients, and in 2 patients due to respiratory insufficiency shortly after the contrast swallow. Four patients (13%) started oral intake despite abnormal contrast swallow results, including aspiration (n=3) and laryngeal penetration (n=1), as a speech-language pathologist assessed their swallowing function as safe. A speech-language pathologist was involved in 8 patients (25%) during the contrast swallow, for routine swallowing assessment (n=6; 19%) or hoarseness/dysphonia (n=2, 6%). In 30 patients (94%), a speech-language pathologist was involved following the contrast swallow. CT scans were performed following contrast swallow results in 3 patients (9%): 2 for suspected anastomotic leakage, which was not confirmed on CT in either case, and 1 for suspected conduit torsion, which was confirmed on CT.

Median day of nasogastric tube removal was POD 5 (IQR, 4–7). Median day of oral intake initiation was POD 8 (IQR, 5–15). Among patients with an abnormal contrast swallow, 27 (84%) had postoperative complications, including 11 (34%) with pneumonia. RLN injury occurred in 10 (7 left, 2 bilateral, 1 right; all grade 1, Clavien-Dindo 1), 3 of whom were suspected and confirmed prior to the contrast swallow. The median ICU-stay was 1 day (IQR, 1–1 days). The median hospital stay was 10 days (IQR, 8–13 days).

At discharge, 23 patients (72%) were on a liquid diet, and 9 (28%) had no oral intake due to an unsafe swallowing function/aspiration (n=5), anastomotic leakage (n=3), and dilated gastric conduit (n=1).

Potential predictors

Table 3 provides an overview of the potential predictors considered in the logistic regression analysis. Univariable analyses revealed no significant predictors of an abnormal contrast swallow, and multivariable analysis similarly identified no independent predictors.

Table 3. Logistic regression analysis of potential predictors for an abnormal contrast swallow.

Variables	Univariable analysis				Multivariable analysis
	OR	95% CI	P value*		OR	95% CI	P value*
Age	1.049	0.994–1.107	0.08		1.046	0.991–1.104	0.10
BMI	0.946	0.838–1.068	0.37
Gender
Male	1.000	–	–
Female	0.450	0.144–1.405	0.17
Tumor location
Cervical/proximal	1.000	–	–
Mid	0.600	0.027–13.582	0.75
Distal/GEJ	0.667	0.040–11.104	0.78
Surgical approach
Transhiatal	1.000	–	–
Transthoracic	0.709	0.269–1.867	0.49
Location anastomosis
Cervical	1.000	–	–
Intrathoracic	0.612	0.238–1.575	0.31
Preoperative radiation
No, CT/no therapy	1.000	–	–
Yes, CRT	0.733	0.272–1.979	0.54
Suspicion of RLN injury prior to contrast swallow
No	1.000	–	–
Yes	2.379	0.375–15.110	0.36
Speech-language pathologist involved†
No	1.000	–	–		1.000	–	–
Yes	0.429	0.160–1,145	0.09		0.448	0.165–1.215	0.12

^†, speech-language pathologist present/involved during the contrast swallow. *, P<0.05. BMI, body mass index; CI, confidence interval; CRT, chemoradiotherapy; CT, chemotherapy; GEJ, gastroesophageal junction; OR, odds ratio; RLN, recurrent laryngeal nerve.

Discussion

Key findings

In this observational cohort, a contrast swallow was routinely performed in 80 of 110 patients (73%), of which 48 (44%) had a normal contrast swallow, while 32 (29%) had an abnormal contrast swallow. The contrast swallow findings led to withholding oral intake initiation in 25% of patients (n=28), deviating from the standard oral feeding protocol. All abnormal contrast swallows (n=32) involved either aspiration (n=30, 94%) or laryngeal penetration (n=2, 6%). In two patients, aspiration occurred alongside signs of delayed gastric conduit emptying and in two other patients, suspected anastomotic leakage was observed but not confirmed on CT.

Strengths and limitations

This study has considerable strengths. The use of a prospective database and a clinical protocol has resulted in a highly comprehensive dataset, and the included population serves as a strong representation of the Western esophageal cancer population. Additionally, multiple aspects were considered, beyond just the findings of the contrast swallow and the initiation of oral intake, such as involvement of a speech-language pathologist, removal of the nasogastric tube, and oral intake at discharge, all of which contribute to providing a thorough overview.

However, this study also has several limitations that should be acknowledged. Firstly, the retrospective nature of the study introduces variability in the surgical procedures performed, including transhiatal and transthoracic approaches, which may affect outcomes and limit the comparability of the data. Additionally, while a descriptive study design was chosen with the primary aim of exploring the clinical relevance of the routine contrast swallow within our current ERAS protocol, this design does not allow for the establishment of causal relationships. Although data were obtained from a prospectively maintained database, the findings related to the contrast swallow were based on radiologists’ assessments/reports. As a result, there was no standardized assessment, and it was assumed that any aspects not explicitly mentioned in the report were considered normal. Lastly, since multiple radiologists contributed to the reports, there may be an increased risk of interobserver variability, which could affect the consistency of the findings.

Comparison with similar research

A contrast swallow is one of the most commonly used examinations after esophageal surgery. As the contrast swallow is a routine practice following esophagectomy in many centers, this study aimed to evaluate whether its use in daily clinical practice actually achieves its intended purpose, namely to guide safe initiation of oral intake after esophagectomy. It has several benefits, including low costs and its status as a relatively safe, non-invasive first-line investigation (7). Despite these advantages, the contrast swallow as routine practice also presents logistical challenges, as its coordination requires careful alignment among all involved parties (i.e., speech-language pathologists, surgeons, nursing staff, radiology technicians) and is also dependent on the patient’s physical status. Most research on its clinical value focuses on its sensitivity in detecting anastomotic leakage. However, previous research has shown low diagnostic sensitivity for anastomotic leakage, and therefore, it is not recommended to use this examination routinely for detecting this complication (4,7,11). Limited research has been conducted on additional aspects of the contrast swallow in relation to oral intake. One such aspect is aspiration. Not only are respiratory complications the most common cause of postoperative mortality after esophagectomy, but the likelihood of developing pneumonia is also directly related to the degree of swallowing dysfunction (12). In the current study, aspiration and laryngeal penetration were found in all abnormal contrast swallows. Pneumonia rates were comparable between the normal (31%) and abnormal (34%) contrast swallow groups. Due to the observational nature of the study and multifactorial etiology of pneumonia, it is difficult to assess whether withholding oral intake influenced the occurrence of pneumonia. Consequently, no causal inferences can be made.

Explanations of findings

In our center, the contrast swallow was primarily introduced to detect early signs of delayed gastric emptying. This condition arises from a combination of anatomical and physiological changes in the gastric conduit following esophagectomy and can complicate postoperative recovery, especially when oral intake is resumed too fast. The extension of the gastric conduit can cause aspiration or anastomotic leakage due to tension of the filled conduit on the anastomosis. Moreover, several patient and procedure-related factors have been shown to influence or be associated with delayed gastric conduit emptying, such as female gender, pre-existing pulmonary comorbidity, and longer operative time (13). In our cohort, delayed gastric conduit emptying was suspected based on the contrast swallow findings in only two patients (3%), both in combination with aspiration, leading to withholding of oral intake according to the protocol. However, previous studies have reported delayed gastric conduit emptying rates as high as 39% (13,14). This discrepancy raises the question of whether our findings reflect an actual lower incidence, potentially influenced by factors such as late oral intake initiation, or whether the contrast swallow may have limited sensitivity in detecting delayed gastric conduit emptying.

Involvement of a speech-language pathologist after esophagectomy has already been shown to offer significant clinical benefits. A study comparing postoperative outcomes between patients who received a swallowing intervention led by a speech-language pathologist after esophagectomy and a control group found that the intervention group had a significantly earlier initiation of oral intake (8 vs. 11 days, P=0.009), and a significantly shorter hospital stay in patients without complications (POD 14 vs. 18, P=0.001) and with RLN injury (POD 21.5 vs. 30, P=0.003) (15). In our cohort, a speech-language pathologist was present during the contrast swallow in 44% patients with a normal contrast swallow, compared to only 25% of patients with an abnormal contrast swallow. Speech-language pathologists were involved during the contrast swallow based on clinical indication. Multivariable analysis showed a trend toward decreased odds of an abnormal contrast swallow when a speech-language pathologist was present during the examination (OR =0.448, 95% CI: 0.165–1.215, P=0.12). As speech-language pathologist involvement was indication-based, their presence differed between patients and may represent a potential confounding factor, since tailored swallowing instructions during the examination could influence aspiration and penetration findings. Early integration of such care into the ERAS protocol could potentially shorten hospital stay and improve overall recovery, while also preventing unnecessarily prolonged periods without oral intake.

Implications and actions needed

Overall, this study demonstrates that routine implementation of the contrast swallow led to deviations from the oral intake protocol in 25% of patients, as the results of the contrast swallow influenced the decision not to initiate oral intake, mostly because of aspiration. Despite the relatively small sample size, this study also explored potential predictors of an abnormal contrast swallow. Although not statistically significant, older age appeared suggestive of increased risk, consistent with previous literature (3). Low BMI has also been identified as a risk factor in earlier studies, though this was not confirmed in our data (3). Taken together, these factors, along with a clinical swallowing assessment by a speech-language pathologist, may help guide the selective use of contrast swallows after esophagectomy.

Conclusions

In conclusion, 73% of patients in our cohort underwent a routine contrast swallow, and in 25% of cases, the findings led to withholding oral intake initiation. Decisions to delay the initiation of oral intake were primarily based on observed swallowing dysfunction (aspiration and laryngeal penetration). Ideally, a more targeted approach, guided by clinical swallowing assessment and focused on identifying patients at higher risk of abnormal findings, may better inform the use of contrast swallows. Future research should explore a more selective approach in specific patient categories as well as a possible earlier timing following esophagectomy.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional ethical board of the University Medical Center Utrecht (No. 13-061). Informed consent was waived as this study utilized data from routine clinical care in an existing prospective database, which was reviewed and exempted from the Medical Research Involving Human Subjects Act (WMO) by our institutional ethics board.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2037/dss

jtd-18-02-66-dss.pdf ^{(39.9KB, pdf)}

DOI: 10.21037/jtd-2025-aw-2037