Carcinoma esophagus is an aggressive malignancy with an increasing incidence. Its virulence in terms of symptoms and mortality justifies a continued search for an optimal therapy. Relative 5 year survival rates for patients with esophageal cancer have improved over. The 3 most successive decades (1). Although the reason for this trend is surely multifactorial, the improvement in survival corresponds with the wide – spread acceptance and use of multimodality treatment approach. This has led some investigators to examine the value of surgery in combined modality treatment regimen and others to suggest a diminished role for radical resection in the modern management of esophageal cancer. The former promise is based on the hypothesis driven scientific testing of surgery in the context of clinical trials and is appropriate. However, the later subjective conclusion is at best premature and at worst irrational and has the potential to compromise delivery of a treatment that offers the best opportunity for long term survival.


Surgical resection is still the standard treatment in esophageal cancer management and remains the bench mark to which other modalities are compared. Advances in anaesthesia, refinement of surgical techniques, improved perioperative care and pain control, proper selection and staging of a patient and R0 resection (macroscopic and microscopic clearance of tumor) have resulted in improved loco regional control and long term survival with reduced operative morbidity and mortality, especially when performed by expert team in experienced centres with high volume practice (2,3,4,5).


There is no issue that engenders more debate among experts than the optimal extent of surgery in patients with esophageal cancer (6). Depending upon performance status, tumor location and extent, the operative approach has traditionally varied from conventional transthoracic esophagectomy or limited esophagectomy without a thoracotomy to more radical or enbloc esophagectomy with 2 field or 3 field lymphadenectomy. During the last 15 years, several minimal invasive techniques have also been introduced.


The aim of this article is to outline surgical options and summarise available data about their effectiveness. Issues which remain controversial are – optimal surgical approach – transthoracic (TTE) Vs Transhiatal esophagectomy (THE), extent of lymphadenectomy (2 fields or 3 fields), value of using minimal invasive techniques and role of salvage esophagectomy.



The two most common surgical approaches for curative resection of esophageal cancer are transthoracic (TTE) and transhiatal esophagectomy (THE). Proponents of TTE claim its oncological superiority in terms of complete tumor clearance and thorough lymphadenectomy thereby minimising the potential for occult residual disease and tumor spillage. The direct transthoracic approach provides adequate exposure and reduces risk of injury to adjacent structures including azygous vein, thoracic duct, aorta and bronchus. However, supporters of THE claim its oncological equivalence to TTE in terms of outcome, minimal respiratory complications, reduced lethal complication of mediastinitis due to anastomotic leakage, shortened duration of surgery and thus decreased morbidity and mortality.


Two large meta – analysis and four phase III randomised controlled trials have compared TTE with THE. Rindanietal (7) reviewed 44 studies conducted from 1986 to 1996 which included 5483 patients. Hulscheretal (8) did meta – analysis of 50 studies conducted from 1990 – 1999 which included 7527 patients. Both series concluded that overall perioperative complications and survival at 3 or 5 years did not differ significantly between the two procedures. There was a significant higher perioperative mortality rates for patients who underwent TTE whereas those patients treated with THE had a higher incidence of anastomotic leaks, strictures and recurrent laryngeal nerve injury.


Four phase III randomised controlled trials conducted by Hulscher (9), Chuetal (10), Jacobi (11) and Goldmincetal (12) comparing TTE with THE have been published. Three of these phase III trials (10, 11,and 12) had very small sample size, so on definitive conclusions can be drawn from their data. In all three trials, post – operative mortality and morbidity rates were similar for both groups and overall survival was unaffected by choice of surgical approach. A fourth trial from Netherland (9) involved 220 patients with adenocarcinoma of the mid or distal esophagus who were randomly assigned to undergo either THE or TTE. Significantly more lymph nodes were dissected from patients in TTE group (31 Vs 16, P < 0.001). A significantly higher incidence of pulmonary complications and increased time in both intensive care unit and hospital were recorded in patients who underwent TTE, but there was no significant difference in post – operative mortality between the two procedures. The investigators noted a trend towards benefit in disease – free and overall survival at 5 years with TTE, the survival curves began to separate at 3 yrs.


Improved tumor and nodal clearance, slight benefit in survival and mounting evidence to support improved outcome in high volume centres with regard to esophageal resections have tilted balance towards transthoracic esophagectomy.


Na – not available, NS – Not significant, SCC – Squamous cell cancer, AC – adenocarcinoma.



The aim of primary surgery in treatment of carcinoma esophagus is definite cure. To obtain this goal, R0 resection is of paramount importance, which includes a thorough lymphadenectomy. However, one of the most controversial questions remains the extent of lymph node dissection (the so called third field).


There is a dichotomy in views regarding the role of lymphadenectomy. Opponents argue that the presence and survival remains unchanged despite removal of these nodes (13). For proponents, presence of lymph node involvement even at a distance from a primary tumor justifies an aggressive approach of radical esophagectomy combined with thorough lymphadenectomy (14).


Extended 3 field lymphadenectomy (3FLND) became widely practised in Japan as evidence by a nationwide study reporting the results of the three field lymphadenectomy performed at 35 institutions (15). From that report it appeared that almost 1 in 3 patient (30%) had unforeseen metastasis in cervical lymph nodes. The authors also claimed an improved disease free and overall 5 year survival as compared with esophagectomy with 2 FLND. On the contrary, surgeons in the West, in part influenced by the minimalistic attitude, have been sceptical and reluctant to adopt the procedure because of the fear of increasing morbidity and mortality by adding bilateral cervical lymphadenectomy. Moreover in North America and Europe, most cancers occur in distal esophagus and gastro esophageal junction with adenocarcinoma being predominant histologic type and more tiet towards option of chemo radiation in West (16). The data from Western world are, therefore, scare and hence the controversy remains.


The arguments that favour more extended lymphadenectomy are – optimal staging, prolonged tumor control and improved cure rates without increasing hospital morbidity and mortality. Available data indicate that unforeseen lymph node involvement in the neck is encountered in approximately 30% of the patients after 3 FLND. Even in tumor of GE junction, upto 20% of patients in the T3N+ setting have unforeseen positive nodes in the neck. The high prevalence of unforeseen cervical LN involvement and the resulting potential stage migration clearly has therapeutic consequences. First, the inability to assess adequately LN involvement in T1b tumor still makes patient selection for non – surgical treatment modalities like endoscopic mucosal resection and photo dynamic therapy very unreliable. In this group, surgery therefore remains the standard. Secondly, when using radio therapy in the induction mode, extension of the irradiated field to cervical region should be seriously considered in tumors of the tubular esophagus, irrespective of the histologic type and location. Third and most importantly, as long as accurate clinical staging can only be obtained through extended 3 FLND, patient selection for multi – modality treatment, i.e induction therapy and assessment of outcome remains very unreliable. Finally and equally important, 3 FLND, by detecting otherwise unknown positive nodes in the third field, refines the true No population. Thus allowing to assess real prognosis (17). Radical esophagectomy with 3 FLND is decreasing loco regional recurrence substantially below 10% in several published reports. Moreover extended lymphadenectomy seems to defer onset of loco regional recurrence and generalised metastasis for upto 3 years or more. Many studies have also indicated a distinctive survival benefit. There are no randomised trials to the effect of 3 FLND in local control disease free and overall survival, but there is evidence from various studies indicative of benefit.


Lerutetal (18) published a study in 1999 reflecting single institutional experience with extensive lymphadenectomy. Unforeseen involvement of cervical LN was seen in 30% cases. In T3N+ tumors of GEJn, 16.6% positive neck nodes were detected. There was a tendency towards better survival in favour of 3 FLND especially in stage III, IV diseases (21% Vs 12% only with simple esophagectomy without 3 FLND).


Kato etal (19) did a study comparing 3 FLND with 2 FLND operative mortality was 2.6% and 12.3% in 3 FLND and 2 FLND group respectively. 5 year survival was 48.7% and 33.7% with 3 FLND and 2 FLND respectively.


Nishihiraetal (20) in a prospective randomised trial compared 3 FLND with 2 FLND. The 5 year survival was 64.8% and 48% after 3 FLND and 2 FLND respectively.


Akiyama etal (21) reported a 5 year survival of 54% after 3 FLND in node negative patients as compared with 34% after 2 FLND (P = 0.004). In this study, patients with positive nodal disease also had an overall superior 5 year survival of 43% as compared with 28% after 2 FLND (P = 0.0008)


Nishimakietal (22) reported an overall 5 year survival rates of upto 68% with 3 FLND.


Altorkietal (23) published the results of prospective observational study of esophagectomy with 3 FLND performed in 80 patients, Unforeseen metastasis to cervical LN occurred in 36% patients irrespective of the cell type or location of tumor. The 5 year SR for node negative patients was 88%. The hospital mortality rate was 4% only. Van de Ven C etal (24) published a study to provide support to the concept of 3 FLND. They concluded that 3 FLND improves staging and decrease loco regional recurrences. Cervical LN metastases in absence of thoracic LN involvement were seen in 20% of GE junction tumors and 35.3% of distal esophageal tumors supporting the fact that there is an important skipping phenomenon which has therapeutic implications.


All results regarding the extended esophagectomy with 3 FLND are reported form experienced institutions; this technique is not widely adopted in the world. These results show low mortality rates with acceptable morbidity rates and low loco regional recurrence following 3 FLND. To our knowledge, the therapeutic value of 3 FLND is unproven in randomised trials; however, the better long term survival rates reported by experienced centres after 3 FLND, with acceptable morbidity and mortality are encouraging. The combination of chemo therapy and radio therapy has an established role in cancer treatment such as anal cancers but has not replaced surgery which still remains gold standard in esophageal cancers. Thus surgeons must do their best by removing potentially tumor by making extended esophagectomy and 3 FLND as standard in selected patients. Nishimakietal (25) reported survival to be extremely poor even after 3 FLND in patients with 5 or more positive nodes, with simultaneous metastasis to cervical, mediastinal and abdominal LN, with cervical mets from a lower esophageal cancer and with intramural metastasis. Roderetal (26) reported ratio of involved to removed LN as important prognostic indicator. Ratio of less than 0.2, i.e. less than 20% of removed LN involved suggest a good prognosis and benefit from 3 FLND in terms of loco regional recurrence and survival.






Large sized incisions are usually needed on the thorax and abdomen to safely perform a curative esophagectomy; however, such surgical modalities result in a heavy operative burden for the patient. Particularly, a radical esophagectomy with 3 FLND places a maximum operative burden on patients with esophageal cancer.


However, owing to recent remarkable advancements in optical technology, endoscopic instruments are now able to access the thoracic and abdominal cavities through a small incisional wound. The mobilisation of stomach and esophagus can be done using such endoscopic instruments. This type of operation is called minimal invasive surgery. If such minimal invasive surgery is feasible for performing a curative esophagectomy for esophageal cancer, and reducing the operative burden thus resulting in a decrease in post – operative complications, then a good patient QOL can be expected.


MIE surgery offers several potential benefits. Two of most frequent complications following esophagectomy are pneumonia and pulmonary failure. Indeed patients who develop pneumonia following esophagectomy face upto a 20% risk of death (27). The avoidance of laparotomy and thoracotomy incisions may very well impact on the incidence of these complications. Techniques have now been developed for both a complete transhiatal laparoscopic and a combined thoracoscopic/ laparoscopic esophagectomy. The former approach simplifies patient approach and doesn’t require single lung ventilation.


Recently, minimal invasive techniques of esophageal resection have been used. These procedures incorporate various combinations of laparoscopy. Thoracoscopy and hand assisted manoeuvers. Minimal invasive surgery offers many theoretical advantages including minimising post – operative pain, reducing length of stay in intensive care units and hospitals and decreasing overall cost. These factors improve quality of care and outcome cost, enabling patients to regain a normal life style quickly. Hand assisted laparoscopic techniques can reliably mimic open esophagectomy and may be the ideal bridge between open and totally laparoscopic or thoracoscopic procedures. The hand assisted approach preserves the advantage of open procedures by offering tactile sensation or feedback and improved hand – eye coordination, while maintaining the benefit of minimal invasive procedure. The surgeon’s ability to feel the tumor during the procedure helps ensure wide field dissection and adequate surgical margins. Proponents of minimally invasive approaches argue that mediastinal exposure is improved and lymph node dissection is made easier by use of laparoscopic or thoracoscopic techniques. LN retrieval seems adequate and comparable with open procedure. However multiple body compartments are still opened despite incisions being small which is major contributing factor to the morbidity and mortality associated with esophagectomy. Indeed, this may negate the theoretical advantage of MIE.


Data comparing MIE with open surgery are accumulating. Most of these data are in the form of single institution studies using historical controls for comparison. Nguyen etal (28) found that MIE was associated with less blood loss, shorter length of surgery and shorter length of stay in the intensive care unit. The operative mortality was 4.3% and mean number of LN removed 10 and 3 year survival was 57%. Luketichetal (29) evaluated outcome of MIE in 222 patients with esophageal cancer. The operative mortality was 1.4% and median hospital stay was 7 days. Survival rates at 40 months were 70%, 20%, 25% for stage I, II, III disease respectively. A large meta – analysis review (30) was performed demonstrating the widening application of minimally invasive approaches to esophagectomy. Twenty three articles concerning the topic met the stringent requirement for review and included 1398 patients. Overall mortality (2.3%), morbidity (46%) and leak rates (7.7%) were similar to open procedures. Lymph node yield for an R0 resection was similar to published results of open series. Avitaletal (31) published results of laparoscopic THE showing operative mortality of 4.5% and average LN yield of 14 nodes.


Long term follow up is needed to confirm that these values translate into satisfactory oncological outcome. Adequate randomised data is not currently available to satisfactorily confirm its oncological equivalence to open techniques in terms of lymphadenectomy and R0 resection. Based on limited knowledge available today, short team outcomes after MIE are atleast comparable with, if not better than, outcomes associated with open procedures. Whether MIE is superior to the time honoured open techniques as an oncologic surgical procedure remains to be defined. Experience with abdominal surgeries has shown that minimally invasive techniques eventually prevail over open techniques. This should hold true for MIE as long as the procedures do not Jeopardize. The basic principle of oncologic surgery – complete resection of the tumor and its lymph node bearing territories.




Some recent studies have suggested that definitive chemo radio therapy has an equivalent impact on the long term survival to a radical esophagectomy in patients with resectable esophageal cancer. These results suggest definitive chemo radiation to be useful as a standard therapeutic option for esophageal cancer, even though significant toxicity develops during or even after the treatment (32). However, the persistence or recurrence of local disease is not uncommon after definite chemo radio therapy, thus suggesting the potential need of salvage treatment in patients receiving this treatment.


For superficial tumors which remain persistent or recurrent at the local site, an endoscopic mucosal resection is an excellent salvage treatment. However, for a persistent or recurrent tumor involving deeper layer of the esophageal wall or recurrent tumors occurring in adjacent lymph nodes, an esophageal resection is necessary as a salvage treatment. At present, whether such a salvage esophagectomy is feasible and provides a survival benefit remains uncertain because of apaucity in the volume of such salvage surgery to provide solid evidence. Recent studies have shown that salvage esophagectomy results in long term survival in subset of patients with an early pathologic stage, a prolonged time to relapse and R0 surgical resection, although patients undergoing a salvage esophagectomy for tumor recurrence after definitive chemo radio therapy do tend to have an increased mortality, morbidity and length of hospitalisation in comparison to patients undergoing a planned esophagectomy after preoperative chemo radiation.




The management of oesophageal cancer will undoubtedly continue to evolve as improvements in technology, combined with a greater understanding of genomics and biology of tumors, better define effective therapeutic interventions into strategies for clinical management. The role of surgery is likely to change over time, but will continue as a primary, or secondary, treatment modality for a substantial number of patients with esophageal cancer. In patients with confined disease (stage I, IIa), resection can be curative and the acceptable outcome achieved by surgeons who are experienced at esophagectomy is unlikely to be challenged by other forms therapy. Since the vast majority of patients with locally advanced disease who undergo chemo radio therapy have residual disease in situ at completion of treatment, common sense dictates that surgery should be beneficial in achieving a long term disease free state. Conversely, it is difficult to contemplate how surgery would benefit patients who achieve a complete pathological response with combined modality therapy, and therefore, they should not be exposed to the risks associated with oesophageal resection. Unfortunately, at present we are unable to pinpoint those patients who require some kind of therapeutic intervention and those for whom particular treatment should be avoided. However, it is likely that in the not too distant future, we will be able to establish a genetic finger print for each individual patient that will dictate the most efficacious therapeutic strategy in each setting. It is important for us to separate fact from fiction when assessing the effectiveness of therapeutic intervention and avoid the pitfall inherent in personal bias. For example, although MIE has many theoretical benefits, its worth has not been vigoursly examined and, therefore, the actual advantages over open esophagectomy are unclear and unproven. Likewise, the widespread use of preoperative chemo radio therapy on the basis of its presumed therapeutic benefit should be weighed against the insufficient evidence for an actual advantage over resection alone. The challenge for us is not to accept the status quo and to investigate new better options, but to do so in a scientifically sound way that leaves little doubt to the accuracy of therapeutic claims. In a disease where as many deaths occur as new cases are reported each year, no therapeutic option, including surgery, should be diagnostically accepted and defended. We should all continue to search for more effective applications of current treatments, and vigorously explore innovative alternatives which must be validated through the clinical trial process.




    • 1) Jemal A, Siegel R, Ward E etal Cancer statistics 2007. CA cancer J Clin 2007; 57: 43 – 66.


    • 2) Whooley BP, Law S, Murthy SC, Wongs Analysis of reduced death and complication rates after esophageal resection. Ann Surg 2001; 233 : 338 – 44


    • 3) Karl RC, Schreiber R, Boul Ware D, Bakers Coppola D factors affecting morbidity, mortality and survival in patients undergoing Ivor Lewis esophagectomy Ann Surg 2000; 231 – 635 – 43


    • 4) Orringer MB, Marshall B, Iannettoni MD THE: Clinical experience and refinements Ann Surg 1999; 230: 392 – 400


    • 5) Chang Ac, Birkmeyer JD The volume performance relationship in esophagectomy thoracSurgClin 2006; 16 : 87 – 94


    • 6) Pennathur A, Luketich JD Resection of esophageal cancer and strategies for optimal management Ann ThoracSurg 2008; 85: 5751 – 6


    • 7) Rindanib, Martin CJ, Cox MR THVS Ivor Lewis esophagectomy:Istherea difference? Aust NZ J Surg 1999 : 69 : 187 – 94


    • 8) Hulscher JB, Tijssen JG, obertoph, Van Lanschot JJ TTVS TH resection for cancer esophagectomy: a meta – analysis. Ann O thoracsurg 2001 ; 72 : 306 – 13


    • 9) Hulscher JB, Van Sandick JW, de Boer AG etal Extended TT resection compared with limited TH resection compared with limited TH resection for adenocarcinoma of the esophagus N Engl J Med 2002; 347 : 1662 – 9


    • 10) Chu KM, Lwast, Fok M, Wong J. A prospective randomised comparison of TH and TT resection for lower one third esophageal cancer Am J Surg 1997; 174 : 320 – 4


    • 11) Jacobi CA, Zieren HU, Muller JM, Pichlmaler H Surgical Therapy of esophageal cancer: The influence of surgical approach and esophageal resection on cardio pulmonary function. Eur J Cardio thoracSurg 1997 ; 11 :32 – 7


    • 12) Goldminc M, Maddern G, Leprise E etalOesophagectomy by TH approach or thoracotomy : a irrespective RCT Bt J Surg 1993 : 80 : 367 – 70


    • 13) Orringer MB, Marshall B, Stirling MC THE for benign and malignant disease J thoraccardiovascSurg 1993 ; 105: 265 – 76


    • 14) Akiyama H, Tsurumaru M, V dagawa H etal Radical LND for cancers of thoracic esophagus Ann Surg 1994 ; 220 : 364 – 374


    • 15) Isonok, Sato H, Nakayamak. Results of a nationwide study on 3FLND of esophageal cancer. Oncology 1991 ; 48 : 411 – 429


    • 16) Altorki, KentM, Ferraracetal 3 FLND for SCC and AC of esophagus Ann Surg 2002; 236 : 177 – 183


    • 17) Lerut T, Coosemans W, De Leyn P, Decker G etal Department of thoracic Surgery, Catholic University Leuven, Belgium Eur J Cardio thoracSurg, 1999; 16 Suppl 1 : S 447


    • 18) Lerut T, Coosemans W, DeLey P, DeneffeGetalDepartment of thoracic Surgery, University Hospital Gasthuisberg, Belgium – Reflection on 3 FLND in cancer of esophagus and GE junction Hepato gastroenterology, 1999, March – Apr 46(26) : 717 – 25


    • 19) Kato H, WatanabeH, Tachinori Y, Tizukat Evaluation of the neck LN dissection for thoracic esophageal carcinoma Ann ThoracSurg 1991, 51 : 931 – 5


    • 20) Nishihira T, Hirayama K, Moris. A prospective randomised trial of extended cervical and superior mediastinal lymphadenectomy for cancers of the thoracic esophagus. Am J Surg, 1998; 175 : 17 – 51


    • 21) Akiyama H, Tsurumaru M, Vdagava H Radical LND for cancers of esophagus Ann Surg 1994 ; 220 : 364 – 73


    • 22) Nishlmaki, Suzuki T, Kanda T etal Extended radical esophagectomy for superficially invasive cancers of the esophagus surgery 1999; 125 : 142 – 147


    • 23) Altorki N, Kent M, Ferrava C, Port J 3 FLND for SCC and AC of the esophagus Ann Surg 2002 Aug : 236 : 177 – 83


    • 24) VandeVenC, De Leyn P, Coosemans W etal 2 FLND and pattern of LN spread in T3 adenocarcinoma of the distal esophagus and GE junction Eur J Cardio thorac Surg. 1999; June 15 : 769 – 73


    • 25) Nishimaki T, Suzuki T, Tranakayetal Intramural metastasis from thoracic esophageal cancer : local indicators of advanced disease World J Surg 1996; 20: 32 – 7


    • 26) Roder JD, Rusch R, Stein HJ, Fink V etal Ratio of invaded to removed LN as a predictor of survival in SCC of esophagus Br J Surg 1994; 81 : 410 – 3


    • 27) Atkin B, Shah A, Hutecheson K etal Reducing hospital mobility and mortality following esophagectomy. Ann ThoracSurg 2004; 78 : 1170 – 6


    • 28) Nguyen NT, Follette DM, Wolfe BM etal Comparisons of MIE with TT and THE Arch Surg 2000; 135 : 920 – 955


    • 29) Luketich JD, Alvelo – Rivera M, Buenaventuta PO etalMIE : outcome in 222 patients. Ann Surg 2003; 238 : 486 – 494


    • 30) Gemmill Eh, Mc Culloch P Systematic review of MIE Br J Surg 2007; 94 : 1461 – 7


    • 31) Aviatl S, Zyndel N, Sz Om steni S etal Laparoscopic TME for esophageal cancer Arm J Surg 2005; 190 : 69 – 74


  • 32) Vokes EE, Brizel DM, Lawrence TS Concomitant chemo radio therapy in esophageal cancer J ClinOncol 2007; 25 : 4066 – 4074