Nature, severity, and impact of chronic oropharyngeal dysphagia following curative resection for esophageal cancer: a cross-sectional study

Summary

Chronic oropharyngeal dysphagia (COD) and aspiration after esophageal cancer surgery may have clinical significance; however, it is a rarely studied topic. In a prospective cross-sectional observational study we comprehensively evaluated the nature, severity, and impact of COD, its predictors, and the impact of the surgical approach and site of anastomosis. Forty participants were recruited via purposive sampling from the (Irish) National Center between November 2021 and August 2022. Swallow evaluations included videofluoroscopy [Dynamic Imaging Grade of Swallowing Toxicity v2 (DIGESTv2), MBS Impairment Profile, Penetration-Aspiration Scale)]. Functional Oral Intake Scale (FOIS) identified oral intake status. The patient reported outcome measures of swallowing, and Quality of Life (QL) included EAT-10 and MD Anderson Dysphagia Inventory (MDADI). Fourteen (35%) participants presented with COD on DIGESTv2 and 10% had uncleared penetration/aspiration. Avoidance or modification of diet on FOIS was observed in 17 (42.5%). FOIS was associated with pharyngeal dysphagia (OR = 4.05, P = 0.046). Median (range) EAT-10 and MDADI Composite results were 3(0–30) and 77.9(60–92.6), respectively. Aspiration rates significantly differed across surgical groups (P = 0.029); only patients undergoing transhiatal surgery aspirated. Survivors of esophageal cancer surgery may have COD that is undiagnosed, potentially impacting swallow-related QL. Given the small number of aspirators, further research is required to determine whether aspiration risk is associated with surgical approach. A FOIS score below 7 may be a clinically useful prompt for the MDT to refer for evaluation of COD following curative intent surgery. These data present findings that may guide preventive and rehabilitative strategies toward optimizing survivorship.

Graphical Abstract

INTRODUCTION

Esophageal cancer (EC) is the seventh most common cancer worldwide and is responsible for approximately 1 in every 20 cancer deaths. ^1,2 Five-year survival rates are low at approximately 20–25% overall, rising to approximately 50% for patients with locally advanced disease who can be treated with curative intent, and up to 90% for patients diagnosed with Stage I cancer.^2–4 Many factors have improved survival trends, including the centralization of EC surgeries, the use of trimodality therapy and pre- or perioperative chemotherapy, the development of enhanced recovery after surgery protocols, and increased early diagnosis through surveillance of Barrett’s esophagus cohorts.^5–7

With improved prospects of survivorship, there is an increased clinical and research focus on health-related quality of life (HR-QL), in particular in patients who have undergone surgery.^8–10 The nutritional consequences, and the impact on muscle mass and conditioning, have dominated reported research; however, the impact of surgery on swallowing function has been rarely studied.¹¹ It is known notwithstanding that symptoms of dysphagia can be present years following surgery, and that dysphagia may be an independent predictor of HR-QL in survivorship.^12–15 In this context, however, few studies have explored the nature and severity of oropharyngeal dysphagia throughout survivorship, and studied it systematically. Accordingly, further investigation is necessary to gain a more comprehensive understanding of the nature and severity of chronic oropharyngeal dysphagia (COD) through survivorship, as well as its psycho-social impact on patients. In addition, the prediction of COD could potentially assist with surgical decision-making and appropriate referrals for swallow assessments and support.

In this cross-sectional prospective observational study, we aimed to define the nature and severity of COD at a minimum of 12 months post EC resection, and its impact on aspiration risk and QL, and to determine risk factors relating to operative approaches.

METHODS

The STROBE Statement guidelines¹⁶ were adhered to (see Appendix). Forty participants (10 females, 30 males) were recruited from the National Center for Esophageal Cancer in Ireland via purposive sampling. Ethical approval was granted by the Tallaght University Hospital Research Ethics Committee. Eligible participants had undergone a transthoracic (two- or three-stage en bloc) resection or transhiatal resection at least 12 months prior. A three-stage approach included an abdominal and neck exploration for mid and upper-ECs (mostly squamous cell carcinomas), with a two-stage approach including an abdominal exploration for most lower third and junctional tumors (typically adenocarcinoma).¹⁷ One- and two-field lymphadenectomies were completed. Exclusion criteria included a history of oropharyngeal dysphagia, neurological impairment, head and neck cancer, recurrent esophageal disease, severe COPD, and cognitive impairment. The first author accessed a dataset established in the data collection site that contained details of all patients who had undergone esophageal resection at the National Center of Esophageal Cancer. Using this information along with data collected from the hospital information system, the first author selected eligible participants, to whom a participant information leaflet and cover letter were posted, followed by a phone call 7 days later. Eligible participants from anywhere in Ireland were selected in chronological order from the database. The intention was to recruit roughly the same number of participants who underwent cervical and thoracic anastomoses. It was not known whether any of these participants had ongoing swallowing difficulties. Those who wished to participate in this prospective study, and who met the inclusion criteria outlined above, were provided with an appointment time for the next available assessment. Informed verbal and written consent were obtained. The same protocol was used for all participants.

A Patient and Public Involvement group, including two EC survivors (>5 years post-surgery), assisted with the selection of appropriate outcomes and assessment tools. Data collection took place from November 2021 to August 2022. Each participant underwent a standardized videofluoroscopy (VFSS) exam and completed swallowing related questionnaires. See the outcomes measured and tools utilized in Table 1.

Table 1

Outcomes and tools

	Outcomes measured	Source
Swallow safety, efficiency, pathophysiology	Presence of pharyngeal dysphagia	Dynamic Imaging Grade of Swallowing Toxicity v2 (DIGEST v2)24
	Presence and depth of laryngeal penetration/aspiration, response to airway invasion	Penetration-Aspiration Scale (PAS)23
	Nature of oropharyngeal swallow impairment	Modified Barium Swallow Impairment Profile (MBSImP)25
Oral intake status	Functional Oral Intake	Functional Oral Intake Scale (FOIS)31
Patient reported dysphagia	Self-reported functional impact of swallowing	EAT-1026
Reported QL impact	Self-reported impact of oropharyngeal dysphagia on QL	MD Anderson Dysphagia Inventory (MDADI)32
Clinical information	Age, sex, type of cancer, location of cancer, stage of cancer	Medical files
Surgical information	Surgical approach, site of anastomosis, date of surgery, type of neoadjuvant treatment, length of stay in ICU, months since surgery	Surgical files

Videofluoroscopies were conducted by the first named author (AG; an experienced speech and language therapist) and a radiographer using continuous fluoroscopy, and recorded on TIMS DICOM (Foresight Imaging LLC) whilst captured at 25 frames per second, a commonly used frame rate in Europe¹⁸ due to the 50-Hz power standard. Consistencies were prepared according to the International Dysphagia Diet Standardisation Initiative (IDDSI) framework¹⁹ and using standardized recipes including 20% weight to volume (w/v) concentration of barium contrast (Maxibar™ 98.45% w/v)^20,21. Participants were initially seated upright with head and neck in a neutral position to gain a lateral view of the lips and anterior oral cavity, the soft palate superiorly, the cervical spine posteriorly, and the proximal cervical esophagus inferiorly. Barium trials consisted of IDDSI Level 0 water presented by cup (2 × 5 mL, 2 × 10 mL, 2 × uncontrolled volume sequential sips), IDDSI Level 4 water by spoon (2 × 5 mL bolus), and half a Level 7 cracker (Carr’s Table Water Crackers) with 3 mL of level 4 barium on top. Participants were repositioned to an anteroposterior viewing plane, which showed the inferior pharynx superiorly and cervical esophagus (if still intact) inferiorly. During trials, the radiographer tracked the bolus inferiorly towards the gastric cardia/body (if still intact). Barium trials included Level 0 water by cup (2 × 20 mL), Level 4 water by spoon (2 × 5 mL), and 1 × 13 mm barium tablet. Further information on the protocol is included in Supplemental Table 1. Inter-rater agreement, completed by two experienced MBSImP certified clinicians who were blinded to participant information and results, was completed on 25% of the VFSS ratings and measured using a two-way random model assessment on SPSS.²² Airway protection was determined using PAS (Penetration-Aspiration Scale), an eight-point ordinal scale. The worst score for each consistency was recorded.²³ Results were grouped into safe (1), penetrating (2–5), and aspirating (6–8) categories. Presence and severity of pharyngeal dysphagia were rated using the Dynamic Imaging Grade of Swallowing Toxicity v2 (DIGESTv2) protocol; an ordinal scale (0 = no pharyngeal dysphagia, 4 = life threatening pharyngeal dysphagia²⁴). For the purpose of this study, pharyngeal dysphagia refers to a diagnosis of dysphagia on the DIGESTv2 only (i.e. an abnormal score). The DIGESTv2 captures only two, albeit important, markers; swallow safety and swallow efficiency. Additional swallow pathophysiological findings were rated using the MBS Impairment profile (MBSImP). The MBSImP includes physiologic measures of oral, pharyngeal and esophageal function during swallowing, rated on ordinal scales from 0 to 2, 0 to 3, or 0 to 4 (where 0 is normal), with the highest score representing the Overall Impression Score.²⁵ For this study, the term ‘chronic’ refers to ongoing dysphagia a minimum of 12 months post surgery.

EAT-10 is a patient-reported self-administered tool that is validated with a wide array of swallowing disorders.^26–29 Scores ≥3 indicate difficulty swallowing^26,30 (range: 0–40, where 40 = severe impairment). The Functional Oral Intake Scale (FOIS) indicates a feeding tube requirement (1–3), modified/avoided food/fluid consistencies (4–6), or no food/fluid restrictions (7).³¹ The MDADI is a self-administered questionnaire containing 19 items with adequate validity and reliability in identifying the impact of dysphagia in head and neck cancer patients.³² (Normalized score of 20 = extremely low functioning, 100 = high functioning/QL³³).

Statistical analysis

SPSS Version 24 (SPSS, Inc, Chicago, IL, USA) was used for data analysis. Subgroup analyses examined differences in dysphagia, aspiration and QL results (DIGESTv2, PAS, MBSImP, EAT-10, FOIS, MDADI) for various subgroups (surgical approach, site of anastomosis, age, cancer location, cancer type, radiotherapy). A dichotomous variable (aspirators/non-aspirators) was used given the small number of aspirators (n = 2). A mean rank Kruskal–Wallis H-test was utilized as the non-parametric equivalent of one-way ANOVA when distributions of each group did not have the same shape, and as an alternative to the Mann–Whitney U when independent variables contained more than 2 categories. Fisher Helton Exact Test was used as the sample size was small and non-parametric.³⁴ For aim 3, regression analysis assessed for predictors [age, months since surgery, cancer type, cancer stage, surgical approach, site of anastomosis, radiotherapy, days in ICU, self-reported dysphagia (EAT-10), oral intake status (FOIS) of COD (DIGESTv2)]. Univariate binary logistic regression analyses were completed because the independent variables were dichotomous.

One participant did not complete the questionnaires and these missing data were excluded.

RESULTS

40 patients (30 M/10F) were recruited, with a mean (range) age of 66 (43–84) years, with 23 (57.5%) adenocarcinoma, 12 (30%) squamous cell cancer, 3 (7.5%) high grade dysplasia, and 2 (5%) intramucosal carcinoma. 24 (60%) tumors were at the esophagogastric junction, 10 (25%) in the lower esophagus, and 6 (15%) in the mid esophagus. 21 participants (52.5%) had undergone a radical two-stage en bloc esophagectomy, 10 (25%) a radical three-stage en bloc resection, and 9 (22.5%) a transhiatal resection, all with gastric conduit reconstruction. The mean (range) time since surgery was 52 months (13–125). Inter-rater agreement on videofluoroscopy ratings was good-to-excellent: 82.2% (i.e. ICC 2,2 = 0.822) on the MBSImP (98.8% of responses were within one point difference or less), and 93.75% and 94.9% on PAS and DIGESTv2 scores, respectively.

Nature, severity, and impact of chronic oropharyngeal dysphagia

14 (35%) participants presented with dysphagia on the DIGESTv2 (score > 0), 11 (27%) had mild dysphagia, (score = 1), 1 (3%) had moderate dysphagia (score = 2) and 2 (5%) had severe dysphagia (score = 3). All participants had at least one impaired component on MBSImP; 40 (100%) had impaired swallow initiation, 36 (90%) had pharyngeal residue, 32 (80%) had reduced anterior hyoid excursion and impaired esophageal clearance. The frequency of impairment across components is shown in Figure 1. Bars are highlighted when frequency is ≥70% to visually convey the six most frequently impaired components (oral residue, swallow initiation, laryngeal elevation, anterior hyoid excursion, pharyngeal residue, and esophageal clearance). The components whose median (interquartile range (IQR)) results were impaired (>0) include oral residue 2(0,2), swallow initiation 2(2,3), laryngeal elevation 1(0,1), anterior hyoid excursion 1(1,1), pharyngeal stripping 1(1,1), upper esophageal segment opening 1(1,1), tongue base retraction 1(1,2), pharyngeal residue 2(1,2), and esophageal clearance 1.5(1,2). The median scores for each MBSImP Component are presented in Figure 2.

Fig. 1

Percentage of participants with impairments on 17 MBSImP components.

Fig. 2

The median scores for MBSImP components.

The median (IQR) scores for dysphagia, aspiration and swallow physiology are: DIGESTv2: 0 (0,1), PAS: 1(1,1), MBSImp Total Oral Score: 4(3,6) and Total Pharyngeal Score: 7(3.25,9). Twenty (54%) participants scored >3 on the EAT10 which indicates abnormal swallowing.²⁶ The median (IQR) results for patient-reported outcomes are: FOIS: 7(6,7), EAT-10: 3(0,14), MDADI Global: 5(4,5) and MDADI Composite 77.9 (60.0,92.6). 17 (43%) participants scored <7 on the FOIS indicating either avoidance (30%) or modification of certain/all food consistencies (13%). Table 2 provides descriptive results (mean, SD, range, median, and IQR) for the entire cohort regarding dysphagia and aspiration outcomes, swallow pathophysiology, and patient-reported outcomes (relating to oral intake status, dysphagia, and QL). Individual participant level breakdown of each of these scores is shown in Supplemental Table 2.

Table 2

Dysphagia, aspiration, and QL results across the entire cohort

	Mean +/− Std Dev scores	Median/IQR	Range
DIGESTv2 (n = 40)	.49+/− 0.79	.00/0–1	0–3
PAS (n = 40)	1.54 +/− 1.502	1/1–1	1–8
MBSImP Total Oral (n = 40)	4.82 +/− 2.293	4/3.0–6.0	1–10
MBSImP Total Pharyngeal (n = 40)	6.36 +/− 3.004	7/3.25–9.0	1–11
FOIS (n = 40)	6.49 +/−.683	7/6–7	5–7
EAT-10 (n = 39)	7.79 +/− 9.163	3/0–14	0–30
MDADI Global (n = 39)	4.21 +/− 1.105	5/4–5	1–5
MDADI Composite (n = 39)	76.302 +/− 18.9564	77.895/60.0–92.6	32.6–100

With respect to aspiration, 2 (5%) participants had uncleared penetration on the PAS, 2 (5%) had uncleared laryngeal aspiration, 4 (10%) effectively cleared laryngeal penetration, and 32 (80%) did not have laryngeal penetration or aspiration. A visual of a participant aspirating a sip of thin fluid is displayed in Figure 3.

Fig. 3

Aspiration observed on videofluoroscopy.

The background information across thoracic and cervical (two- and three-stage) groups is outlined in Table 3.

Table 3

Comparison of background information, across surgical groups

	Thoracic group	Cervical group		χ2(2), P
	Two-stage n = 21 (52.5%)	Three-stage n = 10 (25%)	Transhiatal n = 9 (22.5%)
Age: mean rank (mean)	19.88 (65.1)	16.85 (62.8)	26 (71.7)	3.037 .219
Sex	3 females 18 males	4 females 6 males	3 females 6 males	2.749 .253
Cancer Type	Adenocarcinoma =18 scc =2 Intramucosal =1	Adenocarcinoma =2 scc =8	Adenocarcinoma =3 scc =2 HighGrade Dysplasia =3 Intramucosal =1	13.73 .001
Recurrent nerve palsy	0	1	0	ns
Location of cancer in esophagus	Middle:0 Lower:3 EGJ:18	Middle:5 Lower:2 EGJ:3	Middle:1 Lower:5 EGJ:3	13.053 .001
Months since surgery   Mean rank (Mean)	12 (44)	28.9 (79)	31 (76)	23.552 <0.001
Extent of Lymphadenectomy	1 field:0 2 field:12 Unspecified:9	1 field:0 2 field:6 Unspecified:4	1 field:2 2 field:2 Unspecified:5	.001 .999
Neo-adjuvant treatment	Chemotherapy:14 Radiotherapy:0 Chemoradiotherapy:9	Chemotherapy:2 Radiotherapy:0 Chemoradiotherapy:7	Chemotherapy:2 Radiotherapy:0 Chemoradiotherapy:0	14.389 <0.001

Key: scc = squamous cell carcinoma, EGJ = esophagogastric junction

Oropharyngeal dysphagia, aspiration, and dysphagia-related QL results across cervical and thoracic subgroups

Comparing cervical and thoracic anastomotic cohorts (Table 4), the median (IQR) results for dysphagia (DIGESTv2), penetration/aspiration (PAS), oral and pharyngeal pathophysiological impairments (MBSImP Total Oral and MBSImP Total Pharyngeal Scores), oral intake status (FOIS), self-reported dysphagia (EAT-10), and QL (MDADI) are the following: DIGESTv2: 0.00 (.0,1) vs 0.29 (.0,1) (P = 0.246), PAS: 1(1,1) vs 1(1,1) (P = 0.816), MBSImP Total Oral Scores: 4(3,5) vs 5(3.5,7) (P = 0.098), MBSImP Total Pharyngeal Scores: 7(5,8) vs 7(3,10) (P = 0.935), FOIS: 6(6,7) vs 7(6,7) (P = 0.251), EAT-10: 5(.0, 17) vs 2.5(.1, 12) (P = 0.648), MDADI Global: 5(4,5) vs 5 (4,5) (P = 0.853) and MDADI Composite: 73.7(55.8, 93.7) vs 81.6(67.1, 91.6) (P = 0.811).

Table 4

Median (IQR) and mean rank dysphagia, aspiration and QL results across the cervical and thoracic subgroups

Median (IQR)    Mean rank, P-value	Site of Anastomosis
	Cervical (n = 19)	Thoracic (n = 21)
DIGESTv2	.00 (.0,1)	.29 (.0,1)
	χ2(1) = 1.349, P = 0.246
PAS	1 (1,1)	1 (1,1)
	χ2(1) = 0.054, P = 0.816
MBSImP Total Oral	4 (3,5)	5 (3.5,7)
	χ2(1) = 2.731, P = 0.098
MBSImp Total Pharyngeal	7 (5,8)	7 (3,10)
	χ2(1) = 0.007, P = 0.935
FOIS	6 (6,7)	7 (6,7)
	χ2(1) = 1.318, p = 0.251
	Cervical (n = 19)	Thoracic (n = 20)
EAT-10	5 (.0,17)	2.5 (.0, 12)
	χ2(1) = 0.208, P = 0.648
MDADI Global	5 (4,5)	5 (4,5)
	χ2(1) = 0.034, P = 0.853
MDADI Composite	73.737 (55.789,93.684)	81.579 (67.105,91.579)
	χ2(1) = 0.057, P = 0.811

Note: One participant did not provide responses to the EAT-10 and MDADI.

There were no significant differences across subgroups (type of cancer, location of cancer, age, site of anastomosis or neo-adjuvant treatment) for any dysphagia, aspiration or QL outcomes (Supplemental Table 3). However, as aspiration was only evident in patients who had undergone a transhiatal resection, a subgroup analysis was conducted across the three surgical groups to further explore whether the surgical approach might have been relevant. Results revealed a statistically significant difference (χ²(2) = 7.07, P < 0.029). None from the transhiatal group penetrated but 4 (19%) patients who had a two-stage resection, and 2 (20%) who had a three-stage resection, showed penetration (χ²(2) = 0.039, P = 0.981). 4 (44.4%) patients who had undergone a transhiatal resection had dysphagia on the DIGEST compared to 6 (28.5%) two- and 4 (40%) three-stage resection patients (χ2(2) = 0.823, P = 0.663). 7 (77.7%) in the transhiatal cohort scored below 7 on the FOIS, compared with 3 (33.3%) two and 3 (30%) three-stage (χ2(2) = 5.796, P = 0.055). For EAT-10, 6 (66.7%) in the transhiatal cohort had abnormal swallowing compared with 10 (47.6%) two-stage and 5 (50%) three-stage (χ2(2) = 0.754, P = 0.686).

Clinical and surgical variables associated with pharyngeal dysphagia

FOIS scores were associated with pharyngeal dysphagia in this population (P = 0.047) based on DIGESTv2 results. The odds ratio of dysphagia (on DIGESTv2) for participants who scored <7 on FOIS (i.e., abnormal) was 4.05 (95% CI 1.03–16.00). The remaining variables (age, type, and stage of cancer, months since surgery, radiotherapy, surgical approach, site of anastomosis, days in ICU, and EAT-10) were not found to be predictors of pharyngeal dysphagia as illustrated in Table 5.

Table 5

Univariate regression model representing factors associated with pharyngeal dysphagia (DIGESTv2)

Predictor of pharyngeal dysphagia	P-value	Odds ratio	95% CI
Age > 65	.666	-	-
Type of cancer	.966	-	-
Stage of cancer	.880	-	-
Months since surgery	.336	-	-
Radiotherapy	.446	-	-
Surgical approach	.659	-	-
Site of anastomosis	.372	-	-
Days in ICU	.738	-	-
EAT-10	.064	-	-
Abnormal FOIS (<7)	.046	4.05	1.025–16.007

DISCUSSION

Long-term oropharyngeal dysphagia (COD) affects EC survivors, with 35% of participants in our study presenting with COD and 10% experiencing uncleared laryngeal penetration/aspiration a minimum of 12 months (mean 51 months) following curative-intent EC surgery. With respect to swallow physiology, 73% of participants presented with impaired oral components, 100% had impaired pharyngeal components, and 80% had impaired esophageal clearance. In addition, 43% of participants continued to modify or avoid certain food consistencies because of dysphagia and other aero-digestive issues, with a consequent adverse impact on HR-QL. These data may be important in providing reference data for these cohorts, and also to guide support considerations to EC survivors as current international clinical practice guidelines do not recommend referrals to swallow specialists for assessment.^35,36

The factors underpinning prolonged oropharyngeal dysphagia following EC surgery are unclear. There were no significant differences in dysphagia, aspiration or QL outcomes depending on the site of anastomosis (cervical -v- thoracic). Of the surgical approaches, a transhiatal approach was the only operation associated with aspiration, and although non-significant, the transhiatal group was also poorer than two- and three-stage cohorts for dysphagia observed on VFSS, dysphagia reported by participants, and functional oral intake. Nevertheless, given that only two participants aspirated, a larger study would be required to determine if the surgical approach is relevant. In addition, the cervical anastomosis in a transhiatal resection is no different to that of a three-stage resection, so it is possible that the age profile of these patients, with the transhiatal group significantly older, may be relevant, notwithstanding, age per se (> vs < 65 years) did not significantly impact on this finding. Months since surgery was longest for the transhiatal participants so this is unlikely to be a cause. Patients who are deemed to be at a higher operative risk based on respiratory factors are typically selected for a transhiatal resection so it is therefore possible that their respiratory status contributed to these obervations.³⁷ There were also statistically significant differences across surgical groups for the cancer type, location of cancer, and neo-adjuvant treatments received so any of these factors may likewise be relevant.

In terms of implications on eating patterns, FOIS scores below 7 (i.e., modifying/avoiding certain food textures) were associated with pharyngeal dysphagia (based on DIGESTv2), and had a 4.05 times higher chance of having dysphagia compared to those who scored 7 (did not restrict/avoid food consistencies) (OR 4.05, 95% CI 1.03–16.00). This is important for the multi-disciplinary team (MDT) to be aware of because even if a patient is not reporting symptoms of dysphagia, their FOIS score may suggest that they would be appropriate for a swallow assessment referral. The FOIS is not a dysphagia screening tool and was initially developed to evaluate oral intake/tube-feeding status, therefore, further research could explore the validity and reliability of existing MDT dysphagia screening tools to prevent missed opportunities for swallow assessment referrals.

If these data have clinical importance and relevance, how is it predicted, prevented, and treated? We recently conducted a systematic review which highlighted a paucity of evidence on dysphagia rehabilitation exercises both prior to and following EC surgery.³⁸ Future studies could evaluate the efficacy of exercises specifically targeting the most frequently impaired pathophysiological impairments and symptoms of dysphagia detected in this current study including oral and pharyngeal residue, reduced laryngeal elevation, reduced hyoid movement, and delayed swallow initiation. Oral and pharyngeal residue may potentially be caused by reduced lingual and pharyngeal strength³⁹ secondary to sarcopenia, a known cause of dysphagia acutely post-esophagectomy,⁴⁰ and therefore exercises targeting sarcopenic dysphagia may be beneficial to treat chronic symptoms. Research investigating the intricate inter-relationship between the neo-esophagus and pharyngeal swallow in this population is also warranted as some researchers have suggested that the esophagus has the ability to modulate the motor pattern of the oropharyngeal swallow,⁴¹ which could explain atypical behaviors seen in this study such as delayed swallow initiation. Reduced hyo-laryngeal movement could potentially be a consequence of esophageal shortening and stomach pull-up. Previous research identified how chronic hypertensive peristalsis exists in the upper esophagus following EC surgery⁴² and how upper esophageal sphincter relaxation is impacted by esophagogastric issues such as gastro-esophageal reflux and achalasia,^43,44 in what is hypothesized to be a protective mechanism against aspiration. It is therefore plausible that the upper esophageal sphincter may be hypertensive following esophageal resection to compensate for the absent lower esophageal sphincter, thereby tethering hyo-laryngeal movement and altering swallow physiology throughout the pharyngeal phase. It may be advantageous to consider further interventions such as a free flap to improve hyolaryngeal function or behavioral strategies that relax the upper esophageal sphincter.

This study’s limitations include the use of assessment tools that have not yet been validated with EC patients. This was unavoidable as this research is in its infancy. The DIGESTv2 was the only tool use used to diagnose dysphagia in this study. We recognize that this tool only evaluates airway protection and swallow efficiency (residue) as part of this diagnosis. The MBSImP allowed for greater exploration of pathophysiological function across all phases of swallowing. PAS was used to rate airway protection, however this measurement has been considered twice given that it is also a component of DIGESTv2. The FOIS was used to determine the participant’s status of oral intake and although it was subsequently found to have an association with dysphagia, which may suggest that it could be used to prompt the MDT to refer a patient for a dysphagia assessment, it would be beneficial for future researchers to explore the reliability and validity of existing MDT screens for oropharyngeal dysphagia with this population. This study moreover included a relatively small sample size, particularly in the subgroup analyses. Furthermore, those who chose to participate may have had heightened concerns about their swallowing so results may not reflect the wider EC survivor population. Limitations may be remediated in future research through consecutive admission sampling and multi-site recruitment.

In conclusion, in unselected survivors we report a high incidence of COD after EC surgery and the most frequently impaired physiological components, providing novel data to the EC survivorship literature. For most, airway protection appears to be relatively intact, however the surgical approach used, may be associated with chronic aspiration. The FOIS is associated with pharyngeal dysphagia and could be a valuable prompt for the wider MDT to improve referrals for oropharyngeal swallow assessment. Further research is warranted to explore peri-rehabilitative dysphagia interventions and improve quality of life. Finally, the study suggests an expanded role and value of the speech and language therapist within the ideal modern multidisciplinary EC team.

ACKNOWLEDGMENTS/FUNDING

Funding for the main author’s PhD was provided through Trinity College Dublin Provost Award.

The authors declare that they have no known conflicts of interest that could have influenced the production of the manuscript. No funding has been sourced for this research.

CREDIT AUTHOR STATEMENT

Anna Gillman (Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Resources, Visualization, Writing—original draft, Writing—review & editing), Ciarán Kenny (Formal analysis, Writing—review & editing), Michelle Hayes (Methodology, Writing—review & editing), Margaret Walshe (Supervision, Writing—review & editing), John Reynolds (Methodology, Supervision, Writing—review & editing), Julie Regan (Conceptualization, Methodology, Supervision, Writing—review & editing).

Specific author contributions: Study conception/design: Anna Gillman, Prof Regan, Prof Reynolds. Material preparation, data collection: Anna Gillman. Statistical analysis: Prof Kenny & Anna Gillman. First draft: Anna Gillman. Review, edit, and approve final manuscript: all authors.

Appendix: STROBE statement

	Item no	Recommendation
Title and abstract	1	(a) Indicate the study’s design with a commonly used term in the title or the abstract	Yes
		(b) Provide in the abstract an informative and balanced summary of what was done and what was found	Yes
Introduction
Background/rationale	2	Explain the scientific background and rationale for the investigation being reported	Yes
Objectives	3	State specific objectives, including any pre-specified hypotheses	Yes
Methods
Study design	4	Present key elements of study design early in the paper	Yes
Setting	5	Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data collection	Yes
Participants	6	(a) Give the eligibility criteria, and the sources and methods of selection of participants	Yes
Variables	7	Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if applicable	Yes
Data sources/measurement	8*	For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe comparability of assessment methods if there is more than one group	Yes
Bias	9	Describe any efforts to address potential sources of bias	Yes
Study size	10	Explain how the study size was arrived at	Yes
Quantitative variables	11	Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and why	Yes
Statistical methods	12	(a) Describe all statistical methods, including those used to control for confounding	Yes
		(b) Describe any methods used to examine subgroups and interactions	Yes
		(c) Explain how missing data were addressed	Yes
		(d) If applicable, describe analytical methods taking account of sampling strategy	Yes
		(e) Describe any sensitivity analyses	NA
Results
Participants	13*	(a) Report numbers of individuals at each stage of study—e.g. numbers potentially eligible, examined for eligibility, confirmed eligible, included in the study, completing follow-up, and analyzed	Yes
		(b) Give reasons for non-participation at each stage	NA
		(c) Consider use of a flow diagram	NA
Descriptive data	14*	(a) Give characteristics of study participants (e.g. demographic, clinical, social) and information on exposures and potential confounders	Yes
		(b) Indicate number of participants with missing data for each variable of interest	Yes
Outcome data	15*	Report numbers of outcome events or summary measures	Yes
Main results	16	(a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (e.g. 95% confidence interval). Make clear which confounders were adjusted for and why they were included	Yes
		(b) Report category boundaries when continuous variables were categorized	Yes
		(c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period
Other analyses	17	Report other analyses done—e.g. analyses of subgroups and interactions, and sensitivity analyses	Yes
Discussion
Key results	18	Summarize key results with reference to study objectives	Yes
Limitations	19	Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction and magnitude of any potential bias	Yes
Interpretation	20	Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from similar studies, and other relevant evidence	Yes
Generalizability	21	Discuss the generalizability (external validity) of the study results	Yes
Other information
Funding	22	Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on which the present article is based	Yes