ORIGINAL ARTICLE Laparoscopic Transhiatal Esophagectomy for Invasive Esophageal Adenocarcinoma Kelly R. Haisley 1,2 & Walaa F. Abdelmoaty 2,3 & Christy M. Dunst 2,3,4 Received: 8 October 2019 /Accepted: 16 December 2019 # 2020 The Society for Surgery of the Alimentary Tract Abstract Background Esophagectomy is a fundamental step to achieve long-term disease-free survival in esophageal cancer. While various approaches have been described, there is no consensus on the single best technique to optimize operative and oncologic outcomes. We aim to report the modern experience with laparoscopic transhiatal esophagectomy (LTHE) for invasive adenocarcinoma. Methods We reviewed all patients who underwent LTHE with extended lymph node dissection for distal esophageal adenocar- cinoma (EAC) at our institution between 2007 and 2016. Pre-operative characteristics, operative details, postoperative compli- cations, and long-term outcomes were tracked by review of the electronic medical record and patient surveys. Survival rates were calculated with Kaplan-Meier curves. Results Eighty-two EAC patients underwent LTHE during the study period (84% male, mean age 65, mean BMI 27.8, large). Most patients were clinical stage III (42.7%) and 68.3% had received neoadjuvant chemoradiation (nCRT). Laparoscopy was successful in 93.9%, with five cases requiring conversion to open (6.1%). The median lymph node harvest was 19. Overall complication rate (major and minor) was 45.5% and ninety-day mortality was 4%. Overall 5-year survival was 52% (77% for stage 1, 57% for stage 2, 37% for stage 3). Conclusions Laparoscopic transhiatal esophagectomy has an important role in current esophageal cancer treatment and can be performed with curative intent in patients with distal esophageal tumors. In addition to the well-known advantages of laparos- copy, the increased mediastinal visibility and a modern focus on oncologic principles seem to have a positive impact on cancer survival compared to the open transhiatal approach. Keywords Esophageal adenocarcinoma . Transhiatal esophagectomy . Minimally invasive esophagectomy Introduction Esophageal cancer is increasingly prevalent and highly mor- bid. While significant advances in multimodality treatment have substantially changed the therapeutic protocols for this disease, surgical excision of the esophagus remains a funda- mental step to achieve long-term disease-free survival. 1,2 Although various approaches for esophagectomy have been described, there has never been a consensus on the single best technique to optimize both operative and oncologic outcomes. 3 In general, modern surgical technique has tended to favor a transthoracic esophageal dissection (either with a cervical (Three Field) or intra-thoracic (Ivor Lewis) anastomo- sis) in cases of esophageal cancer, ostensibly to maximize nodal harvest and oncologic outcomes. Transhiatal esophagectomy (THE), in which the resection is accomplished through abdominal and cervical fields alone without a dedicated transthoracic component, has been uti- lized for many years in an attempt to avoid the morbidity of a thoracotomy and of intrathoracic esophageal anastomotic leaks. 4,5 Several published series have shown THE, * Christy M. Dunst [email protected] 1 The Ohio State University, Wexner Medical Center, 410 W 10th Ave, Columbus, OH 43210, USA 2 Providence Portland Medical Center, 4805 NE Glisan St, Portland, OR 97213, USA 3 The Foundation for Surgical Innovation and Education, 4805 NE Glisan St #6N60, Portland, OR 97213, USA 4 The Oregon Clinic, Division of Gastrointestinal & Minimally Invasive Surgery, 4805 NE Glisan St Suite 6N60, Portland, OR 97213, USA Journal of Gastrointestinal Surgery https://doi.org/10.1007/s11605-019-04506-4
Laparoscopic Transhiatal Esophagectomy for Invasive Esophageal Adenocarcinoma
Jan 30, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Laparoscopic Transhiatal Esophagectomy for Invasive Esophageal AdenocarcinomaKelly R. Haisley1,2 & Walaa F. Abdelmoaty2,3 & Christy M. Dunst2,3,4
Received: 8 October 2019 /Accepted: 16 December 2019 # 2020 The Society for Surgery of the Alimentary Tract
Abstract Background Esophagectomy is a fundamental step to achieve long-term disease-free survival in esophageal cancer. While various approaches have been described, there is no consensus on the single best technique to optimize operative and oncologic outcomes. We aim to report the modern experience with laparoscopic transhiatal esophagectomy (LTHE) for invasive adenocarcinoma. Methods We reviewed all patients who underwent LTHE with extended lymph node dissection for distal esophageal adenocar- cinoma (EAC) at our institution between 2007 and 2016. Pre-operative characteristics, operative details, postoperative compli- cations, and long-term outcomes were tracked by review of the electronic medical record and patient surveys. Survival rates were calculated with Kaplan-Meier curves. Results Eighty-two EAC patients underwent LTHE during the study period (84% male, mean age 65, mean BMI 27.8, large). Most patients were clinical stage III (42.7%) and 68.3% had received neoadjuvant chemoradiation (nCRT). Laparoscopy was successful in 93.9%, with five cases requiring conversion to open (6.1%). The median lymph node harvest was 19. Overall complication rate (major and minor) was 45.5% and ninety-day mortality was 4%. Overall 5-year survival was 52% (77% for stage 1, 57% for stage 2, 37% for stage 3). Conclusions Laparoscopic transhiatal esophagectomy has an important role in current esophageal cancer treatment and can be performed with curative intent in patients with distal esophageal tumors. In addition to the well-known advantages of laparos- copy, the increased mediastinal visibility and a modern focus on oncologic principles seem to have a positive impact on cancer survival compared to the open transhiatal approach.
Keywords Esophageal adenocarcinoma . Transhiatal esophagectomy .Minimally invasive esophagectomy
Introduction
Esophageal cancer is increasingly prevalent and highly mor- bid. While significant advances in multimodality treatment
have substantially changed the therapeutic protocols for this disease, surgical excision of the esophagus remains a funda- mental step to achieve long-term disease-free survival.1,2
Although various approaches for esophagectomy have been described, there has never been a consensus on the single best technique to optimize both operative and oncologic outcomes.3 In general, modern surgical technique has tended to favor a transthoracic esophageal dissection (either with a cervical (Three Field) or intra-thoracic (Ivor Lewis) anastomo- sis) in cases of esophageal cancer, ostensibly to maximize nodal harvest and oncologic outcomes.
Transhiatal esophagectomy (THE), in which the resection is accomplished through abdominal and cervical fields alone without a dedicated transthoracic component, has been uti- lized for many years in an attempt to avoid the morbidity of a thoracotomy and of intrathoracic esophageal anastomotic leaks.4,5 Several published series have shown THE,
* Christy M. Dunst [email protected]
1 The Ohio State University, WexnerMedical Center, 410W 10th Ave, Columbus, OH 43210, USA
2 Providence Portland Medical Center, 4805 NE Glisan St, Portland, OR 97213, USA
3 The Foundation for Surgical Innovation and Education, 4805 NE Glisan St #6N60, Portland, OR 97213, USA
4 The Oregon Clinic, Division of Gastrointestinal & Minimally Invasive Surgery, 4805 NE Glisan St Suite 6N60, Portland, OR 97213, USA
Journal of Gastrointestinal Surgery https://doi.org/10.1007/s11605-019-04506-4
Materials and Methods
Patient Enrollment
Under local IRB approval, we performed a retrospective re- view of all patients at a single institution who completed a LTHE over a 10-year period between April of 2007 and July of 2016. A single cohort was analyzed of all patients with invasive esophageal adenocarcinoma of the distal esophageal or GE junction tumors (Seiwert I–III). All patients with such tumors were taken for laparoscopic transhiatal esophagecto- my, which was the standard operative approach at our institu- tion during this time. Patients with more proximal tumors (> 5 cm above the GE junction) were not eligible for THE in our practice and undwerwent transthoracic approaches to focus on perihilar lymph node dissection. Any patient whose surgery was converted to open for any reason was excluded from the outcome analysis to focus on laparoscopic results.
Data Collection
The electronic medical record for each patient was analyzed to obtain pre-operative demographics of interest including age, gender, and BMI, as well as important tumor characteristics, particularly clinical stage (American Joint Committee on Cancer (AJCC) 7th edition15) and location (Seiwert tumor classification16) as well as and any neoadjuvant chemoradio- therapy (nCRT) received. Details of the operative report and postoperative course were reviewed and all complications were stratified according to the Clavien-Dindo classification system.17 Anastomotic leaks were further classified according to their management into minor leaks (treated conservatively with NPO, antibiotics, etc.) or major leaks (required interven- tional procedures such as stenting and reoperation).
Patients were contacted to complete the validated Gastrointestinal Quality of Life Index (GIQLI) form to assess long-term outcomes.18 Survival was determined through que- ry of the Social Security Death Index and survival rates were calculated using Kaplan-Meier survival curves.
Operative Approach
Under general anesthesia, patients are positioned supine with arms tucked at the sides with the neck slightly extended and the head tilted to the right. A diagnostic laparoscopy is per- formed and assuming no metastatic disease, the lesser sac is opened to mobilize the greater curvature of the stomach, tak- ing it off the transverse colon while avoiding injury to the gastroepiploic arcade. The short gastric arteries are divided and a gentle Kocher maneuver is performed by freeing the attachments of the lateral duodenum just enough to allow the pylorus to be brought up to the level of the right crus.
The gastro-hepatic ligament is opened and an extended celiac lymph node dissection is performed, starting at the left hepatic artery and coming down across the common hepatic to the base of the left gastric artery, taking all the fatty nodal tissue between the IVC and the right crus and dividing the left gastric artery at its base with a linear stapler.
A complete circumferential hiatal dissection is performed, keeping as much of the fatty and loose tissue as possible with the specimen and taking a slip of hiatal muscle tissue circumferentially to provide a wide radial margin around the GE junction. The dissection is carried proximally, including mediastinal pleura on both sides of the resection to provide a radial margin along the thoracic esophagus. Anteriorly, the dissection is performed along the posterior boarder of the pericardium. Finally, the posterior dissection is completed by skeletonizing the aorta and taking the esophageal mesentery to allow for an en bloc resection. Care must be take in the area between the aorta and the azygous to prevent injury to the thoracic duct, which can be accomplished by avoiding the deep groove between the two. Mobilization is continued as high into the chest as possible under direct laparoscopic visu- alization, typically to the level of the carina. This dissection allows for removal of all the periesophageal lymph node tissue (stations 8 and 9) but not reliably the subcarinal packet (station 7). An approximately 3-cm wide gastric conduit can then be created using a linear stapler.
The neck is then opened along the medial boarder of the sternocleidomastoid and the proximal esophagus is identified and freed circumferentially. Using careful blunt dissection and avoiding injury to the recurrent laryngeal nerve, the proximal portion of the esophagus is circumferentially mobilized as far distally into the chest as is possible, ideally meeting up with the transhiatal dissection from below. The final central attach- ments of the esophagus that could not be reached from either the cervical or abdominal fields are then divided by passing
J Gastrointest Surg
vein stripper into the lumen of the esophagus and out through the abdominal portion of the specimen. Traction is applied to the vein stripper causing the esophagus to be inverted from the upper mediastinum and removed. The gastric conduit can then be brought up through the empty posterior mediastinum and a cervical anastomosis is performed with a leak test following completion.
A jejunostomy feeding tube is placed laparoscopically and a closed suction drain is left in the neck. An NG tube is inserted into the conduit and its position confirmed. Two drains are also placed into the mediastinum, one into the right chest and one into the left chest, both being brought out the hiatus and through abdominal trocar sites. The hiatal opening can be loosely reapproximated and secured to the conduit to prevent twisting or paraconduit herniation, while taking care not to pinch or compromise the blood supply to the right gastroepiploic arcade feeding the conduit.
Postoperative Protocol
All patients have an epidural placed for perioperative pain control as has been the standard of care at our institution. Following completion of their operation, patients are taken to the ICU for overnight observation and unless there are issues or concerns, they are transferred to the floor on the first post-operative day. All attempts are made to remove NG tubes, A-lines, and Foleys early in the hospitalization, prefer- ably post-op day one. Supplemental tube feeds are started in the first few days postoperatively and uptitrated to goal. Surgical drains are removed as output decl ines . Investigations for anastomotic problems (contrast study or endoscopy) are performed selectively when clinically indicat- ed. Our typical hospitalization lasts at least 6 days. In our recovery protocol, patients are kept NPO for a full 21 days following their procedure and maintain their nutrition via J- tube feeds, after which time they are allowed to slowly ad- vance their diets at home.
Following their discharge, esophageal cancer patients are evaluated every 3 months for the first 2 years, then yearly for 5 years with routine laboratory analysis and standard surveil- lance CT imaging to watch for signs of recurrence.
Results
Demographics
Of the 91 patients in our database who underwent LTHE dur- ing the study period, 82 were performed for an indication of invasive adenocarcinoma at the GE junction (39% Siewert I, 57% Siewert II, 4% Siewert III). The remaining patients were excluded as their indication for esophagectomy was due to an alternate diagnosis such as squamous cell cancer (n = 1), high-
grade dysplasia (n = 1), an end-stage motility disorder (n = 4), GIST (n = 1), or recalcitrant stricture (n = 1).
Demographic characteristics for the final population are listed in Table 1. In line with the natural history of esophageal cancer in the USA, the population was largely male (84.1%) with a mean age of 65 (range 44–88) and a mean BMI of 27.8 (range 16.0–47.1). Twenty-one patients (25.6%) were clinical- ly stage I, with 26 patients (31.7%) clinically stage II, and 35 patients (42.6%) clinically stage III, with the proportion of patients in each stage remaining relatively stable over time. Of the stage 1 patients, tumor depth was T1a 9%, T1b 62%, and T2 29%. More than two-thirds of patients (68.3%) had received neoadjuvant chemoradiation (nCRT) prior to their resection, namely those with locally advanced tumors and with an increasing percentage of stage II patients receiving nCRT in the later years of the study as national treatment guidelines changed to include this recommendation. Five of these patients had their procedures converted to open and as a result did not receive the laparoscopic mediastinal nodal dis- section, and were excluded from final further outcomes anal- ysis. This resulted in a final study cohort of 77 patients.
Operative Outcomes
The laparoscopic THE was technically successful in 93.9% of cases, with only 5 cases requiring conversion to open in order to complete the procedure. Mean intraoperative blood loss was 128 ml and the median lymph node harvest was 19.
Table 1 Patient demographics (n = 82)
% or range
Seiwert classification
nCRT, neoadjuvant chemoradiotherapy; BMI, body mass index
J Gastrointest Surg
Median postoperative length of stay was 12 days. There were no intra-operative mortalities associated with LTHE.
Complications
The overall complication rate was 45% (50 complications in 35 patients: 6 grade I, 22 grade II, 12 grade IIIb, 8 grade Iva), with specific complications listed in Table 2. The most com- mon complication was anastomotic leak, which occurred at a rate of 22%. Fortunately, all leaks in this series were treated conservatively by either simply prolonging NPO status (n = 2 patients), administering antibiotics (n = 4), or placing an en- doscopic stent (n = 11). No leak required returning to the op- erating room for a takedown or revision of the anastomosis. There were no in-hospital mortalities following LTHE.
Survival/Oncologic Outcomes
Two patients did have early post-operative mortalities (within 90 days of the index operation), one due to an acute MI and a second due to colon herniation with colonic necrosis and sep- sis that could not be rescued despite early recognition and surgery. Complete Kaplan-Meier survival estimates are de- tailed in Table 3 with overall 5-year survival for the entire
cohort of 52% (Fig. 1). With respect to stage-specific out- comes, 5-year survival was 77% for stage 1, 57% for stage 2, and 37% for stage 3 (Fig. 2). The 5-year survival for pa- tients treated with surgery alone (n = 23) was 81%, though this population consisted largely of clinically early stage tumors (16 stage 1 (69.6%)) with a minority of more advanced tumors who declined nCRT for various reasons with 6 stage 2 (26.1%), and a single stage 3 (4.3%) patient. For patients who had received nCRT (n = 54), those with a resulting path- ologic complete response (n = 17, 31.5%) had a 5-year surviv- al of 59% compared with a 5-year survival of only 30% for those with residual disease after nCRT (n = 37) (Fig. 3). The median survival for patients with residual node positive dis- ease after CRT was 21 months.
Quality of Life
Fifteen patients completed the long-term quality of life survey at a median follow-up time of 52 months (range 36 to 88 months). With higher scores representing an improved overall quality of life, the mean GIQLI score for the study population was 112 (range 75–133, maximum possible score of 144), slightly below the population norm of 122.6. Condition-specific scores were relatively similar between the study patients and population norms in terms of GI symptoms (62 vs 62.9), emotional symptoms (20 vs 22.3), and social condition (14 vs 14.8), though physical scores were a bit lower (15 vs 23.5) (Fig. 4).18
Discussion
The transhiatal esophagectomy (THE) was first proposed in 1913.4,5 It would take almost 20 years for the first successful THE to be performed, and even then, little enthusiasm devel- oped for the relatively blind and blunt approach. In the mid- 1970s, however, the THE was resurrected by Orringer, and further developed in the 1990s by DePaula and Swanstrom through the application of minimally invasive techniques, which allowed for significantly improved mediastinal visual- ization and dissection.6,7 Nevertheless, its use has remained somewhat limited in cases of esophageal cancer due to theo- retical concerns over the inability to perform an adequate me- diastinal lymphadenectomy. In fact, a recent study of trends in surgical treatment for esophageal cancer showed that while there is a worldwide shift towards increasing use of minimally invasive or hybrid techniques, only 4% of surgeons utilize the LTHE approach.19
This is perhaps surprising given that many previous studies of LTHE from institutions other than our own have been quite promising with findings of decreased pulmonary complica- tions, shorter hospitalizations, lower 30-day hospital mortali- ty, compared to the transthoracic approach, in the setting of
Table 2 Post-operative complications stratified according to Clavien- Dindo classification
Complication grade n % Description (n)
Grade I 6 12% 1. Anastomotic leak (2) 2. RLN palsy (1) 3. Ileus (1) 4. Lymph leak (1) 5. Wound infection (1)
Grade II 22 44% 1. Atrial fibrillation (9) 2. Anastomotic leak (4) 3. Pneumonia (4) 4. Pneumatosis intestinalis (1) 5. UTI (1) 6. Wound infection (3)
Grade IIIa 0 0% –
Grade IIIb 12 24% 1. Acute postoperative hemorrhage (1) 2. Anastomotic leak (11)
Grade IVa 8 16% 1. Atrial fibrillation (1) 2. Mediastinal empyema (1) 3. Pulmonary embolism (2) 4. Respiratory failure (1) 5. Renal failure (1) 6. TEF (1) 7. Tracheal injury (1)
Grade IVb 0 0% –
Grade V 2 4% 1. Colon herniation (1) 2. Myocardial infarction (1)
RLN, recurrent laryngeal nerve; TEF, tracheo-esophageal fistula; UTI, urinary tract infection
J Gastrointest Surg
both benign disease and cancer.7,10–13,20 Furthermore, new studies have started to suggest nearly equivalent oncologic outcomes between LTHE and LTTE, including a randomized trial out of the Netherlands that reported 5-year survivals of 34% and 36%, respectively (P = 0.71, per protocol analysis) with no statistical difference in survival between the two.21
While the overall morbidity rate in our own study was 45%, this reflects a very aggressive data collection to capture both major and minor complications and without a direct compar- ison in our own data, it is difficult to equate this to other morbidity studies that have collected data or defined morbid- ity differently.
Contrary to the theoretical concerns of oncologic inferiority of the LTHE approach for esophageal cancer, the survival rates in our population treated with LTHE compare favorably to the survival outcomes reported for patients treated with the more traditional transthoracic approach in other publications. The American Cancer Society, based on population data, re- ports that the average 5-year survival node negative esopha- geal cancer is 45%, which drops to 24% for patients whose disease has spread to surrounding tissues or regional lymph nodes, and just 5% if metastatic disease is present.22 In our study population, the 5-year survival rate of 52% for the
cohort is actually superior to the overall survival rates typical- ly quoted for esophageal cancer. In addition, our 5-year stage and treatment-specific survival rates compare favorably with previously published reports including the landmark CROSS trial which quotes similar numbers in terms of survival.23,24 In the series of 100 en bloc transthoracic esophagectomies pub- lished by the DeMeester group, an identical overall 5-year survival was found of 52% as we had in this series of transhiatal esopahgectomies.25 While we cannot make any direct statistical comparisons between these individual study populations as they certainly vary in terms of stage, histology, and demographics, and we found no compelling evidence in our population that the LTHE obviously leads to worse long- term outcomes.
It must be recognized that there have been many advances in modern medicine and surgery that directly impact the po- tential role for LTHE in esophageal cancer. First, the majority of cancers in the Western world are now reflux related and located in the distal esophagus making access to the tumor location amenable to a transhiatal approach.24 Next, improve- ments in laparoscopy, lighting, magnification, and instrumen- tation have made the visibility into the mediastinum through the hiatus substantially better, allowing for hemostasis and
Table 3 Kaplan-Meier estimated survival Number
died
Number
alive
5-year
survival
Mean
month
Median
month
By stage
Stage 2 8 16 57% 36.8 28
Stage 3 18 16 37% 22.1 26 20
By treatment
+nCRT +pCR 6 11 59% 30.1 18 0.0035
+nCRT -pCR 20 17 30% 30.1 26 21 45
Median overall survival is undefined because fewer than 50% of all patients died
Fig. 1 Overall survival Fig. 2 Stage-specific survival
J Gastrointest Surg
direct dissection. Additionally, nCRT has become standard of care for locally advanced tumors, which some speculate makes the nodal harvest less important, which has always been the main criticism of the LTHE approach.
While it is well documented that lymph node status is the single best prognostic factor in esophageal cancer, the ques- tion of the importance of nodal harvest in esophageal cancer remains somewhat controversial. It is important to understand that nodal drainage of tumors in the lower esophagus is some- what unpredictable and even distal GE junction tumors may have isolated mediastinal node involvement,26 arguing for the…
Received: 8 October 2019 /Accepted: 16 December 2019 # 2020 The Society for Surgery of the Alimentary Tract
Abstract Background Esophagectomy is a fundamental step to achieve long-term disease-free survival in esophageal cancer. While various approaches have been described, there is no consensus on the single best technique to optimize operative and oncologic outcomes. We aim to report the modern experience with laparoscopic transhiatal esophagectomy (LTHE) for invasive adenocarcinoma. Methods We reviewed all patients who underwent LTHE with extended lymph node dissection for distal esophageal adenocar- cinoma (EAC) at our institution between 2007 and 2016. Pre-operative characteristics, operative details, postoperative compli- cations, and long-term outcomes were tracked by review of the electronic medical record and patient surveys. Survival rates were calculated with Kaplan-Meier curves. Results Eighty-two EAC patients underwent LTHE during the study period (84% male, mean age 65, mean BMI 27.8, large). Most patients were clinical stage III (42.7%) and 68.3% had received neoadjuvant chemoradiation (nCRT). Laparoscopy was successful in 93.9%, with five cases requiring conversion to open (6.1%). The median lymph node harvest was 19. Overall complication rate (major and minor) was 45.5% and ninety-day mortality was 4%. Overall 5-year survival was 52% (77% for stage 1, 57% for stage 2, 37% for stage 3). Conclusions Laparoscopic transhiatal esophagectomy has an important role in current esophageal cancer treatment and can be performed with curative intent in patients with distal esophageal tumors. In addition to the well-known advantages of laparos- copy, the increased mediastinal visibility and a modern focus on oncologic principles seem to have a positive impact on cancer survival compared to the open transhiatal approach.
Keywords Esophageal adenocarcinoma . Transhiatal esophagectomy .Minimally invasive esophagectomy
Introduction
Esophageal cancer is increasingly prevalent and highly mor- bid. While significant advances in multimodality treatment
have substantially changed the therapeutic protocols for this disease, surgical excision of the esophagus remains a funda- mental step to achieve long-term disease-free survival.1,2
Although various approaches for esophagectomy have been described, there has never been a consensus on the single best technique to optimize both operative and oncologic outcomes.3 In general, modern surgical technique has tended to favor a transthoracic esophageal dissection (either with a cervical (Three Field) or intra-thoracic (Ivor Lewis) anastomo- sis) in cases of esophageal cancer, ostensibly to maximize nodal harvest and oncologic outcomes.
Transhiatal esophagectomy (THE), in which the resection is accomplished through abdominal and cervical fields alone without a dedicated transthoracic component, has been uti- lized for many years in an attempt to avoid the morbidity of a thoracotomy and of intrathoracic esophageal anastomotic leaks.4,5 Several published series have shown THE,
* Christy M. Dunst [email protected]
1 The Ohio State University, WexnerMedical Center, 410W 10th Ave, Columbus, OH 43210, USA
2 Providence Portland Medical Center, 4805 NE Glisan St, Portland, OR 97213, USA
3 The Foundation for Surgical Innovation and Education, 4805 NE Glisan St #6N60, Portland, OR 97213, USA
4 The Oregon Clinic, Division of Gastrointestinal & Minimally Invasive Surgery, 4805 NE Glisan St Suite 6N60, Portland, OR 97213, USA
Journal of Gastrointestinal Surgery https://doi.org/10.1007/s11605-019-04506-4
Materials and Methods
Patient Enrollment
Under local IRB approval, we performed a retrospective re- view of all patients at a single institution who completed a LTHE over a 10-year period between April of 2007 and July of 2016. A single cohort was analyzed of all patients with invasive esophageal adenocarcinoma of the distal esophageal or GE junction tumors (Seiwert I–III). All patients with such tumors were taken for laparoscopic transhiatal esophagecto- my, which was the standard operative approach at our institu- tion during this time. Patients with more proximal tumors (> 5 cm above the GE junction) were not eligible for THE in our practice and undwerwent transthoracic approaches to focus on perihilar lymph node dissection. Any patient whose surgery was converted to open for any reason was excluded from the outcome analysis to focus on laparoscopic results.
Data Collection
The electronic medical record for each patient was analyzed to obtain pre-operative demographics of interest including age, gender, and BMI, as well as important tumor characteristics, particularly clinical stage (American Joint Committee on Cancer (AJCC) 7th edition15) and location (Seiwert tumor classification16) as well as and any neoadjuvant chemoradio- therapy (nCRT) received. Details of the operative report and postoperative course were reviewed and all complications were stratified according to the Clavien-Dindo classification system.17 Anastomotic leaks were further classified according to their management into minor leaks (treated conservatively with NPO, antibiotics, etc.) or major leaks (required interven- tional procedures such as stenting and reoperation).
Patients were contacted to complete the validated Gastrointestinal Quality of Life Index (GIQLI) form to assess long-term outcomes.18 Survival was determined through que- ry of the Social Security Death Index and survival rates were calculated using Kaplan-Meier survival curves.
Operative Approach
Under general anesthesia, patients are positioned supine with arms tucked at the sides with the neck slightly extended and the head tilted to the right. A diagnostic laparoscopy is per- formed and assuming no metastatic disease, the lesser sac is opened to mobilize the greater curvature of the stomach, tak- ing it off the transverse colon while avoiding injury to the gastroepiploic arcade. The short gastric arteries are divided and a gentle Kocher maneuver is performed by freeing the attachments of the lateral duodenum just enough to allow the pylorus to be brought up to the level of the right crus.
The gastro-hepatic ligament is opened and an extended celiac lymph node dissection is performed, starting at the left hepatic artery and coming down across the common hepatic to the base of the left gastric artery, taking all the fatty nodal tissue between the IVC and the right crus and dividing the left gastric artery at its base with a linear stapler.
A complete circumferential hiatal dissection is performed, keeping as much of the fatty and loose tissue as possible with the specimen and taking a slip of hiatal muscle tissue circumferentially to provide a wide radial margin around the GE junction. The dissection is carried proximally, including mediastinal pleura on both sides of the resection to provide a radial margin along the thoracic esophagus. Anteriorly, the dissection is performed along the posterior boarder of the pericardium. Finally, the posterior dissection is completed by skeletonizing the aorta and taking the esophageal mesentery to allow for an en bloc resection. Care must be take in the area between the aorta and the azygous to prevent injury to the thoracic duct, which can be accomplished by avoiding the deep groove between the two. Mobilization is continued as high into the chest as possible under direct laparoscopic visu- alization, typically to the level of the carina. This dissection allows for removal of all the periesophageal lymph node tissue (stations 8 and 9) but not reliably the subcarinal packet (station 7). An approximately 3-cm wide gastric conduit can then be created using a linear stapler.
The neck is then opened along the medial boarder of the sternocleidomastoid and the proximal esophagus is identified and freed circumferentially. Using careful blunt dissection and avoiding injury to the recurrent laryngeal nerve, the proximal portion of the esophagus is circumferentially mobilized as far distally into the chest as is possible, ideally meeting up with the transhiatal dissection from below. The final central attach- ments of the esophagus that could not be reached from either the cervical or abdominal fields are then divided by passing
J Gastrointest Surg
vein stripper into the lumen of the esophagus and out through the abdominal portion of the specimen. Traction is applied to the vein stripper causing the esophagus to be inverted from the upper mediastinum and removed. The gastric conduit can then be brought up through the empty posterior mediastinum and a cervical anastomosis is performed with a leak test following completion.
A jejunostomy feeding tube is placed laparoscopically and a closed suction drain is left in the neck. An NG tube is inserted into the conduit and its position confirmed. Two drains are also placed into the mediastinum, one into the right chest and one into the left chest, both being brought out the hiatus and through abdominal trocar sites. The hiatal opening can be loosely reapproximated and secured to the conduit to prevent twisting or paraconduit herniation, while taking care not to pinch or compromise the blood supply to the right gastroepiploic arcade feeding the conduit.
Postoperative Protocol
All patients have an epidural placed for perioperative pain control as has been the standard of care at our institution. Following completion of their operation, patients are taken to the ICU for overnight observation and unless there are issues or concerns, they are transferred to the floor on the first post-operative day. All attempts are made to remove NG tubes, A-lines, and Foleys early in the hospitalization, prefer- ably post-op day one. Supplemental tube feeds are started in the first few days postoperatively and uptitrated to goal. Surgical drains are removed as output decl ines . Investigations for anastomotic problems (contrast study or endoscopy) are performed selectively when clinically indicat- ed. Our typical hospitalization lasts at least 6 days. In our recovery protocol, patients are kept NPO for a full 21 days following their procedure and maintain their nutrition via J- tube feeds, after which time they are allowed to slowly ad- vance their diets at home.
Following their discharge, esophageal cancer patients are evaluated every 3 months for the first 2 years, then yearly for 5 years with routine laboratory analysis and standard surveil- lance CT imaging to watch for signs of recurrence.
Results
Demographics
Of the 91 patients in our database who underwent LTHE dur- ing the study period, 82 were performed for an indication of invasive adenocarcinoma at the GE junction (39% Siewert I, 57% Siewert II, 4% Siewert III). The remaining patients were excluded as their indication for esophagectomy was due to an alternate diagnosis such as squamous cell cancer (n = 1), high-
grade dysplasia (n = 1), an end-stage motility disorder (n = 4), GIST (n = 1), or recalcitrant stricture (n = 1).
Demographic characteristics for the final population are listed in Table 1. In line with the natural history of esophageal cancer in the USA, the population was largely male (84.1%) with a mean age of 65 (range 44–88) and a mean BMI of 27.8 (range 16.0–47.1). Twenty-one patients (25.6%) were clinical- ly stage I, with 26 patients (31.7%) clinically stage II, and 35 patients (42.6%) clinically stage III, with the proportion of patients in each stage remaining relatively stable over time. Of the stage 1 patients, tumor depth was T1a 9%, T1b 62%, and T2 29%. More than two-thirds of patients (68.3%) had received neoadjuvant chemoradiation (nCRT) prior to their resection, namely those with locally advanced tumors and with an increasing percentage of stage II patients receiving nCRT in the later years of the study as national treatment guidelines changed to include this recommendation. Five of these patients had their procedures converted to open and as a result did not receive the laparoscopic mediastinal nodal dis- section, and were excluded from final further outcomes anal- ysis. This resulted in a final study cohort of 77 patients.
Operative Outcomes
The laparoscopic THE was technically successful in 93.9% of cases, with only 5 cases requiring conversion to open in order to complete the procedure. Mean intraoperative blood loss was 128 ml and the median lymph node harvest was 19.
Table 1 Patient demographics (n = 82)
% or range
Seiwert classification
nCRT, neoadjuvant chemoradiotherapy; BMI, body mass index
J Gastrointest Surg
Median postoperative length of stay was 12 days. There were no intra-operative mortalities associated with LTHE.
Complications
The overall complication rate was 45% (50 complications in 35 patients: 6 grade I, 22 grade II, 12 grade IIIb, 8 grade Iva), with specific complications listed in Table 2. The most com- mon complication was anastomotic leak, which occurred at a rate of 22%. Fortunately, all leaks in this series were treated conservatively by either simply prolonging NPO status (n = 2 patients), administering antibiotics (n = 4), or placing an en- doscopic stent (n = 11). No leak required returning to the op- erating room for a takedown or revision of the anastomosis. There were no in-hospital mortalities following LTHE.
Survival/Oncologic Outcomes
Two patients did have early post-operative mortalities (within 90 days of the index operation), one due to an acute MI and a second due to colon herniation with colonic necrosis and sep- sis that could not be rescued despite early recognition and surgery. Complete Kaplan-Meier survival estimates are de- tailed in Table 3 with overall 5-year survival for the entire
cohort of 52% (Fig. 1). With respect to stage-specific out- comes, 5-year survival was 77% for stage 1, 57% for stage 2, and 37% for stage 3 (Fig. 2). The 5-year survival for pa- tients treated with surgery alone (n = 23) was 81%, though this population consisted largely of clinically early stage tumors (16 stage 1 (69.6%)) with a minority of more advanced tumors who declined nCRT for various reasons with 6 stage 2 (26.1%), and a single stage 3 (4.3%) patient. For patients who had received nCRT (n = 54), those with a resulting path- ologic complete response (n = 17, 31.5%) had a 5-year surviv- al of 59% compared with a 5-year survival of only 30% for those with residual disease after nCRT (n = 37) (Fig. 3). The median survival for patients with residual node positive dis- ease after CRT was 21 months.
Quality of Life
Fifteen patients completed the long-term quality of life survey at a median follow-up time of 52 months (range 36 to 88 months). With higher scores representing an improved overall quality of life, the mean GIQLI score for the study population was 112 (range 75–133, maximum possible score of 144), slightly below the population norm of 122.6. Condition-specific scores were relatively similar between the study patients and population norms in terms of GI symptoms (62 vs 62.9), emotional symptoms (20 vs 22.3), and social condition (14 vs 14.8), though physical scores were a bit lower (15 vs 23.5) (Fig. 4).18
Discussion
The transhiatal esophagectomy (THE) was first proposed in 1913.4,5 It would take almost 20 years for the first successful THE to be performed, and even then, little enthusiasm devel- oped for the relatively blind and blunt approach. In the mid- 1970s, however, the THE was resurrected by Orringer, and further developed in the 1990s by DePaula and Swanstrom through the application of minimally invasive techniques, which allowed for significantly improved mediastinal visual- ization and dissection.6,7 Nevertheless, its use has remained somewhat limited in cases of esophageal cancer due to theo- retical concerns over the inability to perform an adequate me- diastinal lymphadenectomy. In fact, a recent study of trends in surgical treatment for esophageal cancer showed that while there is a worldwide shift towards increasing use of minimally invasive or hybrid techniques, only 4% of surgeons utilize the LTHE approach.19
This is perhaps surprising given that many previous studies of LTHE from institutions other than our own have been quite promising with findings of decreased pulmonary complica- tions, shorter hospitalizations, lower 30-day hospital mortali- ty, compared to the transthoracic approach, in the setting of
Table 2 Post-operative complications stratified according to Clavien- Dindo classification
Complication grade n % Description (n)
Grade I 6 12% 1. Anastomotic leak (2) 2. RLN palsy (1) 3. Ileus (1) 4. Lymph leak (1) 5. Wound infection (1)
Grade II 22 44% 1. Atrial fibrillation (9) 2. Anastomotic leak (4) 3. Pneumonia (4) 4. Pneumatosis intestinalis (1) 5. UTI (1) 6. Wound infection (3)
Grade IIIa 0 0% –
Grade IIIb 12 24% 1. Acute postoperative hemorrhage (1) 2. Anastomotic leak (11)
Grade IVa 8 16% 1. Atrial fibrillation (1) 2. Mediastinal empyema (1) 3. Pulmonary embolism (2) 4. Respiratory failure (1) 5. Renal failure (1) 6. TEF (1) 7. Tracheal injury (1)
Grade IVb 0 0% –
Grade V 2 4% 1. Colon herniation (1) 2. Myocardial infarction (1)
RLN, recurrent laryngeal nerve; TEF, tracheo-esophageal fistula; UTI, urinary tract infection
J Gastrointest Surg
both benign disease and cancer.7,10–13,20 Furthermore, new studies have started to suggest nearly equivalent oncologic outcomes between LTHE and LTTE, including a randomized trial out of the Netherlands that reported 5-year survivals of 34% and 36%, respectively (P = 0.71, per protocol analysis) with no statistical difference in survival between the two.21
While the overall morbidity rate in our own study was 45%, this reflects a very aggressive data collection to capture both major and minor complications and without a direct compar- ison in our own data, it is difficult to equate this to other morbidity studies that have collected data or defined morbid- ity differently.
Contrary to the theoretical concerns of oncologic inferiority of the LTHE approach for esophageal cancer, the survival rates in our population treated with LTHE compare favorably to the survival outcomes reported for patients treated with the more traditional transthoracic approach in other publications. The American Cancer Society, based on population data, re- ports that the average 5-year survival node negative esopha- geal cancer is 45%, which drops to 24% for patients whose disease has spread to surrounding tissues or regional lymph nodes, and just 5% if metastatic disease is present.22 In our study population, the 5-year survival rate of 52% for the
cohort is actually superior to the overall survival rates typical- ly quoted for esophageal cancer. In addition, our 5-year stage and treatment-specific survival rates compare favorably with previously published reports including the landmark CROSS trial which quotes similar numbers in terms of survival.23,24 In the series of 100 en bloc transthoracic esophagectomies pub- lished by the DeMeester group, an identical overall 5-year survival was found of 52% as we had in this series of transhiatal esopahgectomies.25 While we cannot make any direct statistical comparisons between these individual study populations as they certainly vary in terms of stage, histology, and demographics, and we found no compelling evidence in our population that the LTHE obviously leads to worse long- term outcomes.
It must be recognized that there have been many advances in modern medicine and surgery that directly impact the po- tential role for LTHE in esophageal cancer. First, the majority of cancers in the Western world are now reflux related and located in the distal esophagus making access to the tumor location amenable to a transhiatal approach.24 Next, improve- ments in laparoscopy, lighting, magnification, and instrumen- tation have made the visibility into the mediastinum through the hiatus substantially better, allowing for hemostasis and
Table 3 Kaplan-Meier estimated survival Number
died
Number
alive
5-year
survival
Mean
month
Median
month
By stage
Stage 2 8 16 57% 36.8 28
Stage 3 18 16 37% 22.1 26 20
By treatment
+nCRT +pCR 6 11 59% 30.1 18 0.0035
+nCRT -pCR 20 17 30% 30.1 26 21 45
Median overall survival is undefined because fewer than 50% of all patients died
Fig. 1 Overall survival Fig. 2 Stage-specific survival
J Gastrointest Surg
direct dissection. Additionally, nCRT has become standard of care for locally advanced tumors, which some speculate makes the nodal harvest less important, which has always been the main criticism of the LTHE approach.
While it is well documented that lymph node status is the single best prognostic factor in esophageal cancer, the ques- tion of the importance of nodal harvest in esophageal cancer remains somewhat controversial. It is important to understand that nodal drainage of tumors in the lower esophagus is some- what unpredictable and even distal GE junction tumors may have isolated mediastinal node involvement,26 arguing for the…
Related Documents
