Preoperative management of patients with oesophageal cancer: Expert recommendations

Abstract

The aim of this work is to establish recommendations for the preoperative evaluation and selection of patients with malignant oesophageal neoplasms, who are candidates for surgical resection with curative intent, based on the consensus established by a group of experts.

Using the Delphi methodology and after 2 rounds of evaluation, responses were obtained from 37 experts to 47 questions about the preoperative management of oesophageal cancer, considering consensus if there was a mean score greater than 8 (range between 0–10).

Of the respondents, 54% were women, with a mean age of 50.2 years, with more than 15 years of average experience in oesophageal surgery. In the preoperative evaluation, agreement was obtained on most of the recommendations, with the indication of a staging laparoscopy being the only one where it was not reached. In the preoperative optimisation, agreement was reached on nutritional, anaemia, physical status, respiratory and comorbidities evaluation, but no agreement was reached on recommending immunonutrition or echocardiography routinely. In the inoperability criteria were included ECOG greater than 1, impaired lung function, and/or Child B or C liver cirrhosis. Agreement was reached on considering unresectable tumours with invasion of the tracheobronchial tract, large vessels, and spinal column, multivisceral metastases, and/or peritoneal carcinomatosis.

Therefore, the recommendations established in this manuscript may be useful to support decision-making in daily clinical practice, with a high degree of consensus that decisions regarding the management of these patients should be made on an individual basis and within a multidisciplinary committee of experts.

Resumen

El objetivo de este trabajo es establecer recomendaciones en la evaluación y selección preoperatoria de pacientes con neoplasias esofágicas malignas, candidatos a una resección quirúrgica con intención curativa, en base al consenso establecido por un grupo de expertos.

Mediante la utilización del método Delphi y tras 2 rondas de evaluación se obtuvieron respuestas de 37 expertos a 47 preguntas acerca del manejo preoperatorio del cáncer esofágico, considerándose consenso si existía una puntuación media mayor de 8 (rango de 0 a 10).

De los encuestados, el 54 % eran mujeres, siendo la edad media del grupo de 50,2 años, con más de 15 años de experiencia media en cirugía esofágica. En la evaluación preoperatoria se obtuvo acuerdo en la mayor parte de recomendaciones, siendo la indicación de una laparoscopia de estadificación la única donde no se alcanzó. En la optimización preoperatoria se estableció acuerdo en la evaluación nutricional, anemia, actividad física y respiratoria y comorbilidades, sin conseguirlo para recomendar la inmunonutrición, una prueba de esfuerzo o ecocardiografía de forma rutinaria. En los criterios de inoperabilidad se recomienda que los pacientes no deban tener un ECOG mayor de 1, una función pulmonar deteriorada y/o una cirrosis hepática Child B o C. Se consigue acuerdo en considerar irresecables tumores con invasión del traqueobronquial, de grandes vasos y columna vertebral, metástasis multiviscerales y/o carcinomatosis peritoneal, individualizando los casos de oligometástasis.

Por tanto, las recomendaciones establecidas en este manuscrito pueden ser de utilidad para el apoyo a la toma de decisiones en la práctica clínica diaria, existiendo un elevado grado de consenso en que las decisiones en torno al manejo de estos pacientes deben realizarse de forma individualizada y en el seno de un comité multidisciplinar de expertos.

Introduction

Surgery is the mainstay of curative treatment for oesophageal cancer, although due to its complexity it is associated with considerable postoperative complication and mortality rates.¹

Many diverse strategies have been used to try to improve these results, with less aggressive surgical approaches and techniques, optimisation of the nutritional and physical status of patients preoperatively, more advanced perioperative care, etc. In order to reduce the number of complications, a thorough preoperative evaluation with optimisation of the areas where deficits are found and the meticulous selection of candidates who are going to undergo this procedure are basic measures, with various guidelines and consensus documents having been published that attempt to guide and standardise the diagnostic and perioperative care strategies for these patients.^2–4

The objective of this study was to establish a series of recommendations to consider in the preoperative evaluation and selection of patients with malignant oesophageal neoplasms who were candidates for curative-intent surgical resection, based on the consensus established by a group of experts in this area.

Material and method

The data presented in this study were obtained by using the Delphi method, with the collaboration of physicians who are experts in the management of patients with oesophageal cancer. To this end, a questionnaire was prepared and sent to 40 experts selected for their extensive experience and from different geographical areas within Spain.

After completing the first round of the questionnaire, the experts were invited to complete the full survey. In this case they were given the average scores obtained for each recommendation after the first response.

After two rounds of responses to the survey, the data obtained were compiled and analysed to establish recommendations by agreement of the experts on the preoperative management of patients with oesophageal cancer.

Questionnaire preparation

Based on current knowledge of oesophageal cancer surgery, its outcomes, and factors related to increased risk of morbidity and mortality, a total of 47 questions were developed, divided into the following thematic blocks:

• 1

  Diagnostic management and preoperative treatment (11 questions)

• 2

  Preoperative evaluation and optimisation (15 questions)

• 3

  Inoperability criteria (9 questions)

• 4

  Unresectability criteria (12 questions)

All of the questions from each block can be viewed in Tables 1–4.

Table 1.

Results relating to diagnostic management and preoperative treatment.

Recommendation	Mean score (SD)
	1st round	2nd round
Endoscopy with multiple biopsy is recommended to establish the diagnosis of malignant oesophageal tumours	9.53 (±1.07)	9.81 (±.51)
A cervical-thoracic-abdominal-pelvic computed tomography (CT) is recommended to assess tumour extension and establish correct preoperative staging	9.75 (±1.41)	9.89 (±.51)
An upper gastrointestinal endoscopic ultrasound is recommended to assess tumour extension and establish correct preoperative staging	9.16 (±1.50)	9.32 (±.94)
A total body positron emission tomography (PET) is recommended to assess tumour extension and establish correct preoperative staging	9.03 (±1.71)	9.35 (±1.20)
Fibrobronchoscopy is recommended to rule out tracheobronchial involvement in patients with neoplasias of the middle and proximal third of the oesophagus	8.81 (±1.53)	8.70 (±1.35)
Evaluation and diagnostic and therapeutic decisions are recommended in all cases in the context of a multidisciplinary committee of experts	10 (±0)	10 (±0)
A staging laparoscopy with peritoneal lavage is recommended to rule out disease Metastatic disease in advanced tumours of the EGJ (>T3 or N+)	7.67 (±2.83)	7.91 (±1.76)
It is recommended to evaluate the clinical response to neoadjuvant therapy by CT and/or PET before performing resective surgery and once an appropriate time has passed after completing neoadjuvant therapy	9.87 (±.55)	9.89 (±.31)
It is recommended to follow the guidelines set by the European Society for Medical Oncology (ESMO) in relation to the therapeutic management of these patients according to their clinical stage	9.40 (±1.18)	9.67 (±.47)
It is recommended to perform resective surgery with curative intent between 3−6 weeks after completing chemotherapy treatment	9 (±1.58)	9.40 (±1.01)
It is recommended to perform resective surgery with curative intent between 6−10 weeks after completing chemotherapy treatment	8.78 (±2.13)	9.37 (±.92)

Table 2.

Results relating to preoperative evaluation and optimisation.

Recommendation	Mean score (SD)
	1st round	2nd round
Nutritional status assessment and optimisation, if necessary, is recommended in all patients diagnosed with oesophageal cancer	9.96 (±.17)	9.97 (±.16)
Preoperative assessment of anaemia and iron deficiency and optimisation, if necessary, is recommended in all patients diagnosed with oesophageal cancer	9.96 (±.17)	10 (0)
It is recommended to try to obtain a preoperative haemoglobin level greater than 13 g/dl in all patients undergoing resective surgery	8.65 (±1.85)	8.70 (±.84)
Psychological evaluation and therapy, if necessary, is recommended for all patients diagnosed with oesophageal cancer	8.21 (±2.37)	8.51 (±1.32)
It is recommended that all patients be assessed by a professional specialised in physical rehabilitation / physiotherapy to achieve preoperative optimisation of their performance status	9.09 (±1.74)	9.35 (±.75)
It is recommended to stop toxic habits (alcohol and tobacco) at least 4 weeks before the intervention	9.71 (±.92)	9.94 (±.22)
Individualised evaluation of the patient and optimisation of accompanying comorbidities is recommended in all patients who are going to undergo resective surgery for oesophageal cancer	9.93 (±.24)	10 (0)
Assessment of respiratory function (such as spirometry) is recommended for correct evaluation and selection of operable patients	9.34 (±1.45)	9.43 (±.86)
Stress tests as preoperative assessment tests for patients with malignant oesophageal tumour should be performed in all cases	5.84 (±2.27)	5.45 (±2.14)
Cardiological evaluation by echocardiography is recommended in all candidates for surgery, except in those with a known high cardiovascular risk	5.78 (±2.90)	5.70 (±2.09)
Preoperative evaluation by a cardiologist is recommended in all patients with a history of cardiac ischemia, heart failure, cerebrovascular events, insulin-dependent diabetes, or renal failure (creatinine > 2.0 mg/dl)	8.43 (±2.15)	7.91 (±1.42)
The use of immunonutrition is recommended in the 7−10 preoperative days (mean responses in the first round: 7)	6.96 (±2.70)	7.27 (±1.50)
It is recommended to limit preoperative fasting to 2 h for liquids and 6 h for solids in patients without dysphagia	8.87 (±2.09)	8.57 (±1.61)
The use of parenteral nutrition is recommended in those patients in whom for any reason an adequate nutritional status has not been achieved, either orally or enterally, before the intervention	8.06 (±2.53)	8.80 (±1.39)
The application of enhanced prehabilitation and recovery protocols (ERAS) is recommended in patients who are going to undergo resective surgery for oesophageal cancer	9.53 (±1.01)	9.67 (±.52)

Table 3.

Results relating to inoperability criteria.

Recommendation	Mean score (SD)
	1st round	2nd round
The rating of operable/non-operable in a patient diagnosed with oesophageal cancer will be made on an individual basis in the context of a multidisciplinary evaluation and decision by experts	9.96 (±.17)	10 (±0)
Patients with impaired physical status, assessed by the ECOG scale, are considered inoperable, with ECOG 0 or 1 being considered requirements for surgery	8.56 (±1.84)	8.86 (±.97)
Patients with impaired lung function (FEV less than 1.25−1.5 l and/or FEV1 less than 40%) are considered inoperable	8.58 (±1.89)	8.40 (±1.21)
An anaerobic threshold of less than 11 ml/min/kg and/or significant myocardial ischaemia in a stress test is not considered a criterion for inoperability, but it is a criterion for a higher risk of postoperative complications, and therefore should be evaluated in conjunction with the patient's performance status	8.58 (±1.52)	8.80 (±.70)
Age alone is not considered a criterion for inoperability, but should be evaluated jointly with the patient's performance status	9.43 (±1.26)	9.08 (±1.27)
Patients with Child B and C cirrhosis of the liver are considered inoperable. Oesophagectomy may be considered In highly selected Child A patients who have not presented complications secondary to cirrhosis	8.71 (±1.95)	8.81 (±1.30)
Chronic renal failure in patients with oesophageal cancer is not considered a criterion for inoperability. The patient's baseline status and chronic kidney disease should be assessed by a multidisciplinary team that includes a nephrologist to make an individualised decision	9.43 (±.84)	9.37 (±.63)
The presence of generalised arteriosclerosis increases morbidity and mortality in patients who are going to undergo oesophageal surgery. However, it should not be used as the sole criterion for considering inoperability	9.15 (±1.24)	9.16 (±.64)
Palliative treatment is recommended for patients with unresectable, locally advanced, recurrent or metastatic tumours, with a score <60% on the Karnofsky functional assessment scale (KPS) or ≥3 on the ECOG scale	9.59 (±.91)	9.70 (±.46)

Table 4.

Results relating to unresectability criteria.

Recommendation	Mean score (SD)
	1st round	2nd round
The assessment of tumour resectability in a patient diagnosed with oesophageal cancer will be carried out in the context of a multidisciplinary evaluation and decision by experts	9.90 (±.29)	9.94 (±.22)
In general, tumour invasion of the tracheobronchial tree is considered a criterion for unresectability in malignant tumours of the oesophagus	9.81 (±.64)	9.97 (±.16)
In general, tumour invasion of the large vessels is considered a criterion for unresectability in malignant tumours of the oesophagus	9.84 (±.44)	9.97 (±.16)
In general, tumour invasion of the spine is considered a criterion for unresectability in malignant tumours of the oesophagus	9.84 (±.44)	9.94 (±.22)
Tumour invasion of structures such as the pleura, pericardium, lung, and diaphragm is not considered a criterion for unresectability on its own	9.32 (±1.42)	9.45 (±.42)
Tumours of the cervical oesophagus (less than 5 cm from the cricopharyngeal muscle) are considered unresectable and are eligible for treatment with chemotherapy and radiotherapy	9.43 (±1.34)	9.81 (±.39)
Multivisceral metastatic tumour involvement is considered a criterion for unresectability in malignant oesophageal tumours	9.90 (±.39)	9.89 (±.31)
Oligometastatic involvement (1 organ with 3 metastases or 1 extra-regional lymph node station with metastasis) in selected patients (e.g. young and with good functional capacity) should be evaluated individually to establish a criterion for unresectability	9.25 (±1.52)	9.32 (±.62)
The presence of non-regional affected lymph nodes is considered metastatic disease and is therefore considered a criterion for unresectability in malignant oesophageal tumours	8.68 (±1.34)	8.75 (±1.09)
The presence of affected cervical lymph nodes (except station 1 and levels VI and VII) is considered extra-regional metastatic disease, regardless of tumour location or histology according to the 8th edition of the AJCC TNM and is therefore considered a criterion for unresectability in malignant oesophageal tumours	8.65 (±1.57)	8.67 (±.88)
In patients with malignant oesophageal tumours, the presence of positive supraclavicular lymphadenopathy is considered a criterion for unresectability	7.96 (±2.86)	8.32 (±1.22)
The presence of carcinomatosis and/or positive peritoneal cytology is considered a criterion for unresectability in malignant oesophageal tumours	9.12 (±2.13)	9.40 (±.79)

The invitation to participate was sent by email and participants agreed to fill out the form created using Google Forms.

To answer each question, the experts had to choose a score from the ordinal scale between 0 and 10, with 0 being completely disagree and 10 being completely agree.

Analysis of results

Descriptive analysis of the data was carried out using the SPSS v22.0 programme, showing the results of the qualitative variables as frequencies and percentages and the values assigned to each statement according to the mean and standard deviation (SD), considering that consensus or recommendation was established for each statement if there was a mean score greater than 8 (in a range of 0–10) in the results of the second round of questions.

The data provided by the participants were handled anonymously and were only used for statistical purposes.

Results

Of the total of 40 specialists who were invited to participate in the study, 37 (92.5%) correctly completed the 2 rounds of responses to the questionnaire.

54% of the participants were women, with the mean age of the respondents being 50.2 years (SD: + 3.4). Most of them were specialist physicians (45.9%) and section heads (45.9%). Most of the respondents worked in centres with more than 700 beds and the average number of years of experience in oesophageal surgery in the group was 16.1 years (SD: +4.2) (Table 5).

Table 5.

Respondent characteristics.

Concept	Mean (SD)/n (%)
Age	50.2 years (SD ± 3.4)
Sex
Man	17 (45.9%)
Woman	20 (64.1%)
Position
Specialist physician	17 (45.9%)
Head of section	17 (45.9%)
Head of department	3 (8.1%)
Hospital
Under 350 beds	1 (2.7%)
Between 350 and 700 beds	7 (18.9%)
Over 700 beds	29 (78.4%)
Years of experience in oesophageal surgery	16.1 years (SD ± 4.2)

In relation to the responses obtained in the questionnaire, the results are presented below divided into the 4 thematic blocks:

Block 1: Diagnostic management and preoperative treatment (Table 1). In this block, average scores are obtained that allow recommendations to be made in most of the statements raised. The highest scores of agreement (average score equal to or greater than 9) are achieved in points such as performing an endoscopy, cervico-thoraco-abdomino-pelvic CT and PET in the preoperative study, decision-making in the context of a multidisciplinary committee, following the guidelines set by ESMO,⁵ evaluating the clinical response after neoadjuvant treatment with CT and/or PET and performing surgery after 3−6 weeks after chemotherapy. With lower scores, but with a sufficient degree of agreement to be recommended (average score between 8 and 8.99), it is recommended to perform a fiberoptic bronchoscopy in neoplasms of the middle and proximal third and schedule surgery after 6−10 weeks after completion of neoadjuvant treatment with chemotherapy and radiotherapy.

Scores were not high enough to establish a recommendation regarding performing a staging laparoscopy with peritoneal lavage to rule out metastatic disease in advanced tumours of the EGJ (> T3 or N+), where the average score obtained was 7.67 and 7.91 in the first and second rounds respectively.

Block 2: Preoperative assessment and optimisation (Table 2). In this section, the recommendations with the highest level of agreement (mean score equal to or greater than 9) were nutritional assessment and optimisation; anaemia; physical and respiratory functional activity, and associated comorbidities, in the context of the application of multimodal rehabilitation protocols. The recommendations for psychological assessment and therapy, cardiological assessment, maintaining a preoperative fast of 2 h for liquids and 6 h for solids, and the use of parenteral nutrition in those patients in whom, for whatever reason, an adequate nutritional status has not been achieved by oral or enteral route before the intervention, also achieved a sufficient level of agreement (mean score between 8 and 8.99.

Consensus was not strong enough to recommend routinely performing a stress test or echocardiography or to recommend the use of immunonutrition in the preoperative period.

Block 3: Inoperability criteria (Table 3). The recommendations that obtained a high degree of agreement from the experts (mean score greater than 9) were: to individually and multidisciplinarily evaluate the operability of a patient and not to consider age, chronic kidney disease, or generalised atherosclerosis as exclusive criteria in themselves for establishing inoperability. To a lesser degree of agreement, but with sufficient power to maintain its recommendation (mean score between 8 and 8.99), it was established that patients who are going to undergo an oesophagectomy should not have an ECOG greater than 1, impaired lung function and/or Child B or C liver cirrhosis.

Block 4: Unresectability criteria (Table 4). In this section, the experts agree on all the statements presented. Thus, it is recommended (with average scores greater than 9) that the resectability of this type of tumour be evaluated in a multidisciplinary area of experts, considering as unresectable those tumours with invasion of the tracheobronchial tree, the large vessels and the vertebral column, multivisceral metastases and/or peritoneal carcinomatosis. In the case of oligometastatic involvement, it is recommended that the therapeutic strategy be evaluated individually.

A lower degree of consensus (between 8 and 8.99 average score) was achieved in relation to the recommendations regarding non-regional, cervical and supraclavicular adenopathic involvement (Table 4).

Discussion

Surgery for oesophageal cancer is associated with high morbidity and mortality rates (postoperative complication rates close to 60%),⁶ and a comprehensive, multidisciplinary⁷ preoperative assessment is recommended to select candidates for surgery and optimise their clinical and performance status before the intervention. As a result, structured, homogeneous and systematised guidelines and recommendations are required, which combine scientific evidence and the experience of teams of professionals specialised in the management of these patients, in order to offer quality care.

This manuscript establishes recommendations for the preoperative evaluation and optimisation of patients with oesophageal neoplasia based on the agreement and expert opinion of 37 professionals who responded to a survey online.

Multimodal rehabilitation protocols began to be applied in oesophageal cancer surgery with the intention of providing guidelines to improve the condition of patients and facilitate their recovery after surgical aggression. Many of the measures proposed in these protocols⁸ have been included in this survey, thus reinforcing the degree of recommendation established in previous publications. In this context, measures such as optimising nutritional, physical and anaemia status are basic pillars of these protocols and are recommended by experts with a high level of consensus.^9,10

Currently, there is controversy over the use of immunonutrition, a point upon which the experts in this survey did not reach a consensus for its indication, with publications showing disparate results when evaluating the benefit of its use compared to other enteral nutrition formulas.^11–13

The examinations recommended for the diagnosis and preoperative evaluation of these patients and tumour staging are included within a fairly standardised clinical practice, with most of the recommended complementary tests well established in clinical practice guidelines and little variation among experts. However, there are points of controversy such as the indication for staging laparoscopy, where no consensus was reached. The literature has shown that the performance of this laparoscopy can change the therapeutic strategy in 10%–20% of patients according to some series¹⁴ and, therefore, some societies recommend its performance in locally advanced tumours (T3/T4) of the cardia.^15,16 Although there is no homogeneous attitude in the indication of this procedure, in many centres it is performed systematically due to its potential benefits and greater sensitivity and specificity to detect metastatic and/or adenopathic lesions.¹⁷

In this sense, in addition to an individualised evaluation of each case, the creation of predictive models, which support decision-making, could help to better establish the most correct management guidelines and the cost-effectiveness ratio of these strategies.¹⁸

The criteria for inoperability in patients undergoing surgery for oesophageal cancer have been modified due to increasingly exhaustive pre-surgical evaluation and preparation and the improvement and less aggressiveness of anaesthetic and surgical techniques that have achieved a reduction in postoperative morbidity and mortality. The increased risk of complications in patients with impaired functional status, renal failure, cirrhosis or severe pulmonary and/or cardiovascular pathology is known.^19,20 In fact, due to their clinical importance, many of these comorbidities are included in different predictive models of morbidity and mortality risk after oesophagectomy.²¹

Different studies have found advanced age (over 70 or 75 years)^22,23 to be an independent factor of poor prognosis in patients who are candidates for oesophagectomy due to oesophageal cancer, as it increases the probability of dying not only in the short but also in the long term. However, age alone is not considered a criterion for inoperability by experts, as shown by some studies,²⁴ in which the degree and number of complications appears greater in older patients, but a significant increase in mortality is not demonstrated.^25,26

The complementary tests to carry out a correct preoperative evaluation of the performance status of patients are often dependent on the protocols established in each centre and, although there is a basic series of recommended and necessary explorations during the preoperative study agreed upon and recommended in many guidelines and protocols,⁸ there may be controversy on some points. Thus, although some published studies²⁷ suggest that routine preoperative evaluation of cardiac function may be useful to predict potential complications, carrying out stress tests or advanced test to assess cardiopulmonary function are not unanimously or systematically recommended in the guidelines or by the experts.

The criteria for unresectability in oesophageal cancer changed in recent years and patients who were considered unresectable years ago due to metastatic involvement may now be considered candidates for surgery with curative intent in certain contexts. In general, based on the agreement of experts, tumours with invasion of structures such as the large vessels, the tracheobronchial tree or the spine are considered unresectable, although there are already publications of case series that show the feasibility of multivisceral resective surgery in certain specific circumstances, with acceptable morbidity and valid oncological outcomes.^28–31

At this point, a topic of controversy at present is lymph node involvement. In fact, there is a lack of unification for the definition of the different lymph node stations,³² the most correct extension of lymphadenectomy³³ and the locations considered unresectable, these points being currently a matter of debate among the groups with the greatest experience in this field and where the lowest degree of agreement among the experts was found.

Currently, with the advance of new chemotherapy drugs, more selective and effective radiotherapy treatments and the appearance of immunological therapies, the expected survival in patients with stage IV has been increasing. Therefore, it is increasingly common to propose an aggressive, resective approach with curative intent in patients with limited metastatic disease and preserved performance status. Today, palliative treatment remains the therapeutic option indicated in most patients with metastatic³⁴ disease. However, based on growing evidence, surgical resection could be considered indicated in certain contexts. In this sense, the OMEC (Oligometastatic Esophagogastric Cancer)³⁵ group defines the concepts of oligometastatic disease (generally including organ involvement with less than 3 lesions or lymph node involvement in some non-regional station, although with clarifications depending on the affected organ) and attempts to establish the bases and indications for the most appropriate therapies in this type of patient, where resective surgery and systemic treatment have shown improvements in survival rates.)^36–38

This paper therefore presents a series of recommendations, supported by the experience of 37 experts to preoperatively evaluate, select and optimise patients who are candidates for resective surgery with curative intent for oesophageal cancer, which may be useful to support decision-making in the context of daily clinical practice. Despite some controversial points, there is a high degree of consensus that decisions regarding the management of these patients should be made on an individual basis and within a multidisciplinary committee of experts.

This paper has not been sent for publication in any other journal.