Langenbeck's Archives of Surgery

, Volume 402, Issue 2, pp 323–331 | Cite as

Hybrid minimally invasive esophagectomy vs. open esophagectomy: a matched case analysis in 120 patients

  • Torben Glatz
  • Goran Marjanovic
  • Birte Kulemann
  • Olivia Sick
  • Ulrich Theodor Hopt
  • Jens Hoeppner



In esophageal surgery, total minimally invasive techniques compete with hybrid and robot-assisted procedures. The benefit of the individual techniques for the patient remains vague. At our institution, the hybrid minimally invasive laparoscopic-thoracotomic esophagectomy (HMIE) has been routinely applied since 2013. We conducted this retrospective study to analyze the perioperative outcome.


Since 2013, 60 patients were operated in HMIE technique for esophageal cancer. Each of these patients was paired according to the criteria of gender, BMI, age, tumor histology, pulmonary preexisting conditions, and a history of smoking with a patient treated by open esophagectomy (OE). Perioperative parameters were extracted from our prospectively maintained database and compared among the groups.


The HMIE and OE groups were homogeneous in terms of patient- and tumor-related data. There was no difference in lymph nodes harvested (22 vs. 20, p = 0.459) and R0-resection rate (95 vs. 93%, p = 0.500). The operation time for the HMIE was significantly shorter (329 vs. 407 min, p < 0.001). There was no difference between the groups with respect to surgical complications (37 vs. 37%, p = 0.575), but the patients undergoing hybrid technique showed more delayed gastric emptying (23 vs. 10%, p = 0.042). Pulmonary morbidity was significantly reduced after HMIE (20 vs. 42%, p = 0.009). This affected both the occurrence of pneumonia and pleural effusions. The difference in the overall complication rate was not significant (50 vs. 60%, p = 0.179), but life-threatening complications (Clavien/Dindo 4/5) were less frequent (2 vs. 12%, p = 0.031). Overall, there was significantly less need for transfusion after HMIE (18 vs. 50%, p < 0.001), and hospital (and IMC) stay was significantly shorter (14 (6) vs. 18 (7) days, p = 0.002 (0.003)). The multivariate analysis confirms the surgical procedure as an independent risk factor for the development of pulmonary complications (OR 3.2, p = 0.011). Furthermore, preexisting pulmonary conditions were identified as a risk factor (OR 3.6, p = 0.006).


Our retrospective analysis shows that reduction of postoperative pulmonary morbidity, perioperative blood loss, and shortening of hospital stay can be achieved by HMIE. The procedure is safe, and the rate of surgical complications and oncological radicality is comparable to the conventional procedure.


Esophageal cancer Esophagectomy Minimally invasive surgery Gastric pull-up Postoperative morbidity Outcome 


The advent of minimally invasive techniques in esophageal surgery has caused a controversy about the optimal approach. A variety of techniques compete on the same level:
  • Conventional open thoracoabdominal or transhiatal esophagectomy

  • Transhiatal laparoscopic esophagectomy

  • Total minimally invasive thoracoabdominal approach with both laparoscopic and thoracoscopic preparation in lateral or prone position

  • Laparoscopic-thoracotomic hybrid procedure with laparoscopic gastric pull-up and conventional dissection of the esophagus via thoracotomy

  • Thoracoscopic-laparotomic hybrid procedure with thoracoscopic dissection and conventional abdominal preparation via laparotomy

  • Variation of techniques for intrathoracic or cervical anastomosis

  • Robot-assisted procedures

Up till today, very little evidence exists supporting the superiority of minimally invasive procedures over conventional esophagectomy, let alone the superiority of one minimally invasive procedure over the other. The technique applied depends mostly on the experience of the surgeon and the equipment and expertise of the medical center. Numerous retrospective series report very different results for the various techniques while there are no prospective randomized controlled trials comparing one minimally invasive technique to the other, and so far, only the results of the TIME trial comparing total minimally invasive to open esophagectomy (OE) have been fully published [1]. The results of the MIRO trial (NCT00937456) [2] comparing OE to hybrid minimally invasive laparoscopic-thoracotomic esophagectomy (HMIE) are still pending. Total minimally invasive esophagectomy is a promising procedure, but technical difficulties, long operating time, and lack of experience make the procedure difficult to adopt for many hospitals. HMIE represents a more practicable alternative with a potentially comparable benefit for the patient [3, 4]. At the Medical Center of the University of Freiburg, HMIE has been routinely performed whenever possible since 2013. We conducted this matched pair analysis to compare the results to those of conventional Ivor Lewis OE.

Patients and methods


This study evaluates the outcomes of 60 consecutive patients undergoing HMIE for esophageal cancer at a high-volume tertiary referral center between October 2013 and January 2016. Each of these patients was paired—in ranking order—with a patient treated by OE according to the criteria of preexisting pulmonary conditions, history of smoking, tumor histology, gender, BMI, and age. Only patients operated between June 2006 and January 2016 with a thoracoabdominal approach, reconstruction with gastric pull-up, and intrathoracic anastomosis were eligible for matching. Perioperative parameters were extracted from our prospectively maintained database and compared among the groups. Informed consent was obtained from all patients before their inclusion in the cancer registry. The study was approved by the Medical Ethics Committee of the University of Freiburg (file number: EK-Freiburg 569/14).

Pretherapeutic work-up and multimodal treatment

Patients were seen by the operating surgeon in the outpatient setting, and the decision for an HMIE approach was made based on preoperative contrast-enhanced computed tomography imaging, comorbidities, and informed consent. When we started to perform HMIE in 2013, the first few cases were selected patients with smaller tumors and without additional risk factors like visceral adiposity and previous abdominal operations. Afterward, HMIE was performed routinely without restrictions for tumor size and BMI. Contraindications were only previous extensive surgery in the epigastrium and bulky tumorous abdominal lymph nodes. Comorbidities were recorded and pulmonary and cardiac check-up was routinely performed in risk patients. Preexisting pulmonary or cardiac condition was defined as any ongoing pulmonary or cardiac disease requiring medical treatment or restricting patients’ physical performance.

Patients were compelled to stop smoking (when applicable) and underwent instruction to perform breathing and physical exercise. Pretherapeutic diagnostics included endoscopy with biopsies and thoracoabdominal computerized tomography in all patients. Endoscopic ultrasound was used routinely for staging if technically possible. All patients were discussed in our interdisciplinary cancer conference, and decision for neoadjuvant treatment was made when the T stage was T3 or T4 and/or lymph nodes were suspected to be positive, and patients had no contraindications for multimodal treatment. Neoadjuvant chemoradiation was performed for squamous cell tumors and adenocarcinoma without infiltration of the cardia according to the protocol suggested by Naunheim et al. or the CROSS protocol [5, 6]. Neoadjuvant chemotherapy was performed for AC of the distal esophagus and esophagogastric junction according to the protocol suggested by Cunningham et al. or the FLOT protocol [7, 8]. All neoadjuvant chemotherapy protocols were scheduled for postoperative continuance starting 4–8 weeks after the operation with the same drug composition and dosage as preoperatively. After neoadjuvant chemoradiation or neoadjuvant chemotherapy, the patients were restaged by endoscopy and computerized tomography, and resection was performed approximately 6 weeks after the end of neoadjuvant treatment.

Surgical technique

The operative procedure was chosen according to tumor location. For OE, patients were operated by an Ivor Lewis thoracoabdominal approach (right-sided thoracotomy, median laparotomy, and intrathoracic-stapled anastomosis) [9]. Reconstruction was performed by a gastric tube formation and gastric pull-up in all cases. Pyloric dilation was carried out via a small gastrotomy located at the tip of the gastric tube by manual dilatation of the pylorus by an intraluminal dilatation with a large clamp. Patients requiring reconstruction by a colonic interposition or cervical anastomosis were excluded from this analysis. We routinely performed two-field lymphadenectomy. Before 2013, all OE patients received a jejunostomy catheter for postoperative enteral feeding. Due to reduced complication rate during the last decade with a very low rate of anastomotic leaks, we have omitted the routine use simultaneously with the introduction of HMIE. Feeding jejunostomy is now restricted to very cachectic patients who might benefit from additional nutritional support postoperatively.

The details of our HMIE procedure have been published previously [10]. In brief, the patient is placed in a thoracoabdominal position with the thorax rotated toward the left side and the pelvis remaining in horizontal position. Five trocars are placed in a semicircular row in the upper abdomen. Dissection of the gastric ligaments, mobilization of the esophagus, and lower mediastinum are followed by lymphadenectomy at the hepatic artery, celiac trunk, and splenic artery. After identification of the arteria hepatica communis, the lymphadenectomy is started and guided along the upper edge of the artery toward the celiac trunk. After lymphadenectomy on the celiac trunk and on the splenic artery, the vena coronary ventriculi as well as the left gastric artery are dissected and deposited with PDS clips. The lymphadenectomy of the celiac trunk is thus completed. A laparoscopic colopexy is carried out to avoid postoperative hiatal large bowel herniation by suturing the mobile greater omentum with the attached transverse colon directly to the peritoneum parietale of the left-sided ventral abdominal wall with interrupted absorbable sutures. Endoscopic pyloric dilatation is routinely performed intraoperatively after mobilization of the stomach with a controlled radial expansion esophageal balloon dilatator (20 mm, 2 min). Right-sided thoracotomy and en bloc dissection of the esophagus with mediastinal lymphadenectomy is performed. After transection of the esophagus at the level of the arch of the azygos vein (adenocarcinoma) or upper thoracic aperture (squamous cell carcinoma), the stomach is pulled up and gastric tube is created by linear stapling. The esophagogastrostomy is established using a 25- or 28-mm circular stapler. After placement of a nasogastric tube, two right-sided chest tubes are placed and the wound is closed. A left-sided chest tube is routinely placed via a separate incision after both OE and HMIE to prevent postoperative pneumothorax and pleural effusion requiring an additional procedure.

Perioperative management

All patients included in the study received identical anesthesiological and perioperative management regardless of the surgical approach: Patients received an epidural catheter for postoperative pain management whenever possible, all patients underwent single-lung ventilation for the time of thoracotomy, patients were extubated immediately after the operation, postoperative infusion therapy was standardized, and all patients were kept on an intermediate care unit for the first 4 postoperative days. All patients intraoperatively received a nasogastric tube and were restricted to oral fluids until postoperative day (POD) 5. After a routine contrast esophagogram ruled out pyloric spasm, the nasogastric tube was removed and patients were started on solids. In patients with a jejunostomy catheter, enteral feeding was started 6 h after surgery via the catheter and was continued till patients were able to consume sufficient solid food. Patients without a jejunostomy catheter received total parenteral nutrition instead. Blood product transfusions were generally initiated when hemoglobin levels were less than 8 g/dl in patients without coronary artery disease or less than 10 g/dl in patients with coronary artery disease.


Perioperative complications were recorded up to 30 days after surgery and were graded according to the Clavien/Dindo classification [11]. Complications graded as grade I (minor deviation) were not recorded. Major surgical and pulmonary complications were defined according to the criteria stated by the “Esophagectomy Complications Consensus Group” [12] as follows:
  • Surgical complications:
    • Anastomotic leakage, defined as full thickness GI defect involving esophagus, anastomosis, staple line, or conduit irrespective of presentation or method of identification

    • Gastric conduit necrosis; endoscopically identified focal or extensive necrosis

    • Delayed gastric conduit emptying requiring intervention or delaying discharge or requiring maintenance of nasogastric drainage >7 days postoperatively

    • Surgical wound infection requiring opening wound or antibiotics

    • Vocal cord injury/palsy; vocal cord dysfunction post-resection. Confirmation and assessment by laryngoscopic direct examination.

    • Chyle leak

    • Secondary hemorrhage or other complication requiring additional surgical or endoscopic intervention

  • Pulmonary complications:
    • Atelectasis mucous plugging requiring bronchoscopy

    • Pneumonia, defined as alveolo-interstitial radiologic infiltration with the presence of at least two of the following criteria: purulent sputum, temperature >38.5 or <35 °C or leukocytes >10,000 or <1500/mm3

    • Respiratory insufficiency requiring reintubation and assisted ventilation;

    • Acute respiratory distress syndrome (ARDS), defined as severe hypoxia (PaO2/FiO2 <200) with diffuse bilateral pulmonary infiltration

    • Pleural effusion requiring additional drainage procedure

    • Pneumothorax requiring treatment

  • Cardiac complications
    • Cardiac arrest requiring CPR

    • Myocardial infarction

    • Atrial dysrhythmia requiring treatment

    • Dysrhythmia ventricular requiring treatment

    • Congestive heart failure requiring treatment

    • Pericarditis requiring treatment

Statistical analysis

All statistical calculations were performed with IBM SPSS® software version 23. The results of our study were gained by retrospective analysis of our prospectively maintained esophagogastric database. Case matching of HMIE with OE patients was performed using the duplicate finding and sorting algorithm of SPSS and manual selection of the best match for each individual case. Scale variables were expressed as median and range, and ordinal and nominal parameters as absolute numbers, and percent. Mann-Whitney’s test and Fisher’s exact test were used for statistical testing. Multivariate logistic regression analysis with forward stepwise selection strategy using a likelihood ratio, including the report of odds ratio and their 95% confidential intervals, was used to identify independent risk factors for development of pulmonary complications. A p value <0.05 was considered statistically significant.


Patient and tumor characteristics

Between June 2006 and January 2016, we performed 224 Ivor Lewis esophagectomies for esophageal cancer: 164 open and 60 minimally invasive. After we started performing HMIE in October 2013, OE was only carried out in 17 patients until January 2016. The demographic data of the overall collective eligible for matching is displayed in Table 1.
Table 1

Association of clinicopathological factors with Surgical Technique


Overall collective

All matches




n = 224

n = 120

n = 60

n = 60

Male gender (n, %)

85% (191)

84% (101)

82% (49)

87% (52)


Age (median [range], years)

62 (31–92)

61 [42–92]

61 [42–92]

61 [44–84]


BMI (median [range], kg/m2)

26 (17–40)

27 [17–40]

27 [19–40]

26 [17–38]


ASA score (n, %)


57% (128)

57% (69)

60% (36)

55% (33)



43% (96)

43% (51)

40% (24)

45% (27)


Comorbidity (n, %)

 History of smoking

43% (96)

42% (50)

40% (24)

43% (26)



17% (38)

25% (30)

25% (15)

25% (15)



27% (60)

26% (31)

23% (14)

28% (17)


Tumor location (n, %)

 Middle 1/3

15% (25)

13% (16)

13% (8)

13% (8)


 Lower 1/3

85% (191)

87% (104)

87% (52)

87% (52)


Tumor histology (n, %)


70% (156)

78% (93)

77% (46)

78% (47)


 Squamous cell carcinoma

30% (68)

23% (27)

23% (14)

22% (13)


Neoadjuvant treatment (n, %)


18% (41)

19% (23)

22% (13)

17% (10)



56% (126)

61% (73)

58% (35)

63% (38)



25% (57)

20% (24)

20% (12)

20% (12)


Pathological UICC stage (n, %)


53% (118)

52% (62)

58% (35)

45% (27)



21% (46)

19% (23)

15% (9)

23% (14)



22% (49)

23% (28)

23% (14)

23% (14)



5% (11)

6% (7)

2% (1)

8% (5)


Scale variables are expressed as median and range, ordinal and nominal parameters as absolute numbers, and percent. Mann-Whitney and Fisher’s exact test were used for statistical testing

After the matching process, 120 patients were included in our analysis. The majority of patients was male (84%). The median age was 61 years (range 42–92) and the median BMI 27 kg/m2 (range 17–40). Forty-two percent had a history of smoking ( ≥20 PY or smoking within the last year prior to surgery), 25% had a preexisting pulmonary condition, and 26% suffered a cardiac comorbidity. Fifty-seven percent were ranked as ASA 1/2 and 43% as 3/4. Eighty-seven percent of tumors were located in the lower third of the esophagus, including tumors of the esophageal gastric junction (AEG I and II) with a 78% share of adenocarcinoma. Only 19% of our patients were operated primarily without undergoing prior neoadjuvant treatment. Twenty percent received neoadjuvant radio-chemotherapy, and 61% were treated in perioperative chemotherapy protocols. Postoperative histopathological staging revealed a UICC stage 0 or 1 in 52% of cases, stage 2 in 19%, stage 3 in 23%, and stage 4 in 6%. The HMIE and OE groups were homogeneous in terms of patient- and tumor-related data. Matching all variables 1:1 was not possible in all cases, but only minor differences occurred (Table 1).

Surgical procedure and hospital stay

Oncological radicality was comparable between the two groups. The median lymph node count was 20 in the OE group and 22 in the HMIE group (p = 0.459). In the HMIE group, 3 patients (5%) had a positive resection margin compared to 4 patients (7%) in the OE group (p = 0.500). The median operating time was 363 min. Interestingly, the median operating time was shorter in the HMIE group compared to the OE group (329 vs. 407 min, p < 0.001). Thirty-four percent of our patients had intra- or postoperative blood transfusion. HMIE resulted in significantly less need for transfusion (18 vs. 50%, p < 0.001), a median number of 1 RBCs was transfused in the OE group (range 0–29). Patients spent a median of 6 days on the intermediate care unit and left the hospital after a median of 16 days. Both intermediate care and hospital stay were significantly shorter in the HMIE group (6, respectively, 14 days vs. 7, respectively, 18 days, p = 0.003, respectively, 0.002). The surgical outcome is displayed in Table 2.
Table 2

Surgical outcome






n = 120

n = 60

n = 60

Lymph nodes harvested (n)

20 [3–57]

22 [10–46]

20 [3–57]


Negative resection margin (R0)

94% (113)

95% (57)

93% (56)


Operating time (min)

363 [255–617]

329 [255–426]

407 [264–617]


Blood transfusion required (n, %)

34% (41)

18% (11)

50% (30)


Number of rbcs applied

0 [0–29]

0 [0–15]

1 [0–29]


IMC stay (days)

7 [4–36]

6 [4–18]

7 [4–36]


Hospital stay (days)

16 [9–94]

14 [9–37]

18 [10–94]


Scale variables are expressed as median and range, ordinal and nominal parameters as absolute numbers, and percent. Mann-Whitney and Fisher’s exact test were used for statistical testing

Perioperative morbidity and mortality

Fifty-five percent of patients had at least one perioperative complication. There was no difference between the groups (50 vs. 60%, p = 0.179, Table 3). In-hospital mortality was low with 2%. There were 2 casualties in the OE group and none in the HMIE group (p = 0.248), and 90-day mortality was 7% in the OE-group and 0% in the HMIE group (p = 0.059). Grading of the complications according to the Clavien-Dindo classification showed grade II and III in 48% of patients and grade IV and V in 7%. Life-threatening complications (grade IV/V) were less frequent in the HMIE group (2 vs. 12%, p = 0.031).
Table 3

Postoperative complications






n = 120

n = 60

n = 60

Perioperative morbidity

55% (66)

50% (30)

60% (36)


Surgical morbidity

37% (44)

37% (22)

37% (22)


 Anastomotic leakage

3% (4)


7% (4)


 Gastric conduit necrosis

4% (5)

5% (3)

3% (2)


 Wound infection

11% (13)

7% (4)

15% (9)


 Delayed gastric conduit emptying

17% (20)

23% (14)

10% (6)


 Hiatal herniation

3% (3)

5% (3)



Pulmonary morbidity

31% (37)

20% (12)

42% (25)



18% (21)

8% (5)

27% (16)



21% (25)

15% (9)

27% (16)


 Pleural effusion

13% (15)

5% (3)

20% (12)


 Respiratory insufficiency

6% (7)

2% (1)

10% (6)


Cardiac complications

11% (13)

12% (7)

10% (6)


Severity of complicationsb

 Grade 0/I

45% (54)

50% (30)

40% (24)


 Grade II/IIIa/b

48% (58)

48% (29)

48% (29)


 Grade IVa/b/V

7% (8)

2% (1)

12% (7)


In-hospital mortality

2% (2)


3% (2)


30-day mortality

1% (1)


2% (1)


90-day mortality

3% (4)


7% (4)


Scale variables are expressed as median and range, ordinal and nominal parameters as absolute numbers, and percent. Mann-Whitney and Fisher’s exact test were used for statistical testing. Complications were graded according to Clavien/Dindo classification

Surgical complications were most frequent and were diagnosed in 37% of patients. Four patients had an anastomotic leak requiring endoscopic or surgical treatment (all OE), and another 5 patients had focal conduit necrosis of the gastric pull-up, managed conservatively (2 HMIE, 3 OE). Wound infection rate was 11%, and delayed gastric emptying occurred in 17% of patients. There was no difference between the groups with respect to total number of surgical complications (37 vs. 37%, p = 0.575), but the patients undergoing HMIE showed delayed gastric conduit emptying significantly more often (23 vs. 10%, p = 0.042). Reoperation rate was 5% in the HMIE group and 15% in the OE group (p = 0.063). Additional to the perioperative complications, three patients (5%) in the HMIE group presented with diaphragmatic herniation of the colon requiring laparoscopic re-exploration, hernia reduction, and colopexia several months after surgery.

Pulmonary morbidity occurred in 31% of patients and was significantly reduced after laparoscopic gastric pull-up (20 vs. 42%, p = 0.009). Especially, pulmonary infections occurred less frequently in the HMIE group (8 vs 27%, p = 0.007) as did pleural effusions requiring replacement of thoracic drains (5 vs. 20%, p = 0.012). Reintubation was necessary in seven patients (1 HMIE, 6 OE, p = 0.057).

Risk factor assessment for development of pulmonary complications

Univariate risk factor analysis revealed the surgical approach (OE vs. HMIE) as a risk factor for the development of a postoperative pulmonary complication (42 vs. 20%, p = 0.009). To verify the actual impact of the HMIE approach, the collective was scanned for other factors having an impact on postoperative pulmonary complication rate.

Patients with a preexisting pulmonary condition had a substantially higher risk of developing a postoperative pulmonary complication (52 vs. 24%, p = 0.006). A history of smoking (39 vs. 25%, p = 0.066) and female sex (47 vs. 27%, p = 0.079) were likewise associated with a higher rate of pulmonary complications, but the difference failed to be significant. All other potential factors including age, BMI, tumor location, histology, neoadjuvant treatment, ASA score, and preexisting cardiac disease had no impact.

The multivariate analysis (multivariate logistic regression analysis with forward stepwise selection strategy using a likelihood ratio) confirms the surgical procedure as an independent risk factor for the development of pulmonary complications (odds ratio, 3.2; 95% confidence interval, 1.4–8.8; p = 0.011). Furthermore, patients with a preexisting pulmonary restriction were at higher risk for postoperative pulmonary complications (odds ratio, 3.6; 95% confidence interval, 1.3–7.0; p = 0.006). History of smoking and gender was not confirmed as independent risk factors in this setting (Table 4).
Table 4

Uni- and multivariate analysis of risk factors for developement of pulmonary complications after esophagectomy


Univariate analysis

Multivariate analysis


Confidence Interval


Pulmonary complication


Sex (female/ ale)




Age (≥65/<65 yrs)




BMI (<25/≥25 kg/m2)




Tumor location (middle/upper third)




Tumor histology (squamous cell/adeno)




Multimodal therapy (none/(R) CTX)




ASA score (3–4/1–2)




Preexisting cardiac condition (Y/N)




History of smoking (Y/N)




Pre-existing pulmonary condition (Y/N)






Surgical technique (HMIE/OE)






aFisher’s exact test

bMultivariate logistic regression analysis with forward stepwise including the report of Odds ratio (OR) and the 95% confidence intervals (CI)


We retrospectively analyzed 120 patients who underwent Ivor Lewis esophagectomy at our institution. Sixty patients received OE, and 60 patients underwent an HMIE procedure. While perioperative mortality was low with 2%, morbidity remained high with 55%. In our analysis, we were able to demonstrate a reduced postoperative pulmonary complication rate, a reduced rate of severe complications, and an enhanced recovery process after HMIE compared to OE.

To illustrate the true effect of the HMIE approach, our study design tried to eliminate as many parameters possible which might influence the surgical outcome. All patients included in the study regardless of the surgical approach received identical anesthesiological and perioperative management: All patients received an epidural catheter for postoperative pain management, all patients underwent single lung ventilation for the time of thoracotomy, patients were extubated immediately after the operation, postoperative infusion therapy was standardized, a fluid restrictive management was applied, and all patients were kept on an intermediate care unit at least for the first 4 postoperative days. Therefore, we believe that the HMIE approach actually accounts for the reduced postoperative complication rate and the faster recovery process.

In contrast to the reduced pulmonary complication rate, we observed a higher rate of delayed gastric emptying in the HMIE group, an effect which can likewise be attributed to the surgical technique: During OE, pyloric dilation was performed manually by intraluminal dilatation with a large clamp, a method that seems to prevent delayed gastric emptying more effective than the endoscopic dilatation performed during HMIE, but which cannot be performed safely without possibility for palpation of the pylorus during HMIE.

Minimally invasive approaches to esophagectomy have become increasingly applied and are currently widespread worldwide. In a 2014 survey, minimally invasive operations were the preferred approach of 43% of all specialized esophageal surgeons worldwide [13]. Minimally invasive techniques for esophagectomy have been reported in the literature for nearly two decades. Early descriptions of HMIE as a two-step approach with laparoscopic gastric mobilization followed by thoracotomic esophagectomy after an interval of several days dates back to 2007 [14]. Since then, the technique has evolved and its safety has been established [10]. A retrospective matched case analysis in a historic collective from the French group around Christoph Mariette was able to demonstrate reduced pulmonary morbidity after HMIE [3]. Decreased postoperative mortality after HMIE was shown in a French Nationwide Study [4]. Preliminary reports from the MIRO trial [15] confirm these results.

The results from our study are concordant with the findings of the French group in a slightly different patient collective and provide independent evidence for the benefit of the HMIE procedure. Our study certainly has limitations, the biggest being its retrospective design allowing only limited definitive conclusions. Even though our study includes only patients operated after 2006, further improvement in anesthetic experience and intensive care management as well as management of postoperative complications since 2006 might contribute to the beneficial results considering that all patients in the HMIE group were operated between 2013 and 2016 [16, 17, 18]. Additionally, patients were nonrandomized to the surgical technique, but the choice of the surgical approach depended mostly on the time period the patient presented to our clinic. Patients operated with an open approach after October 2013 had a contraindication for HMIE extensive surgery in the epigastrium and bulky abdominal lymph nodes. Finally, the limited number of patients eligible for this study did not allow for 100% matching for all parameters, and additional parameters which might influence the surgical outcome were not taken into account. On the other hand, the demographics displayed in Table 1 show very balanced groups.

We believe that our study design has been carefully chosen to deliver the best evidence supporting the advantages of HMIE over OE that a retrospective study can provide. Additional changes to the HMIE approach like the use of a muscle-sparing small-size incision for thoracotomy could make this approach even more effective.

When comparing our results to other analyses, the surgical approach has to be considered. This study only includes Ivor Lewis esophagectomy for tumors of the middle and distal esophagus, while other series include McKeown esophagectomy, often associated with higher postoperative morbidity [19]. Additionally, multimodal treatment has to be taken into account. Our study includes 20% of patients with preoperative chemoradiation, which might—especially in the past—have contributed to postoperative morbidity, as demonstrated elsewhere [20].

Simultaneous to the HMIE technique, other groups have focused on the omission of thoracotomy and have developed laparotomic-thoracoscopic or total minimally invasive techniques. Feasibility and beneficial results of the total minimally invasive technique were demonstrated by Luketich et al. in an impressive series of over 1000 patients [21]. A randomized controlled trial from the Netherlands demonstrated a decreased pulmonary complication rate and a better quality of life after minimally invasive esophagectomy [1, 22]. Reduced postoperative mortality was shown in a large databank analysis [23], and the beneficial results are supported by a systematic review [24].

The search for evidence of beneficial effects of the thoracoscopic-laparotomic technique in the literature is in vain. Significant improvement of the perioperative results was only achieved after adding laparoscopic mobilization of the gastric tube to the thoracoscopic hybrid procedure [25], emphasizing the relevance of the laparoscopic procedure in the total minimally invasive technique. A recent retrospective propensity score-matched analysis failed to demonstrate an advantage of total minimally invasive esophagectomy over the HMIE method [26]. Currently, a pilot trial with the goal to validate the feasibility of a randomized controlled trial comparing different techniques of esophagectomy (total minimally invasive esophagectomy vs. HMIE vs. OE) for esophageal cancer—the ROMIO—is running [27]. Up till today, the two techniques have to be considered very similar concerning effects on postoperative pulmonary complication rate and patient recovery.


Advantages of HMIE are doubtlessly its feasibility and safety for the patient, offering a shorter operation time, an expected comparable oncological radicality to OE, and reduced blood loss. But our study shows that this technique can additionally achieve a reduction of postoperative pulmonary complications and shortening of hospital stay. If the results of the prospective randomized controlled trial confirm these retrospective results, HMIE has to be regarded as the technical procedure with the best evidence for superior short-term outcomes and will be established as the preferable approach to esophagectomy.

Authors’ contributions

Jens Höppner, Torben Glatz, and Ulrich Theodor Hopt partipated in the study conception and design. Torben Glatz, Olivia Sick, and Goran Marjanovic participated in the acquisition of data. Torben Glatz, Birte Kulemann, and Olivia Sick participated in the analysis and interpretation of data. Torben Glatz and Jens Höppner participated in drafting of manuscript. Ulrich Theodor Hopt, Birte Kulemann, and Goran Marjanovic participated in the critical revision of manuscript.


Compliance with ethical standards


No external funding was received.

Conflict of interest

Authors declare that they have no conflict of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.

Informed consent

Informed consent was obtained from all individual participants included in the study.


  1. 1.
    Biere SSAY, van Berge Henegouwen MI, Maas KW et al (2012) Minimally invasive versus open oesophagectomy for patients with oesophageal cancer: a multicentre, open-label, randomised controlled trial. Lancet 379(9829):1887–1892. doi:10.1016/S0140-6736(12)60516-9 CrossRefPubMedGoogle Scholar
  2. 2.
    Briez N, Piessen G, Bonnetain F et al (2011) Open versus laparoscopically-assisted oesophagectomy for cancer: a multicentre randomised controlled phase III trial—the MIRO trial. BMC Cancer 11:310. doi:10.1186/1471-2407-11-310 CrossRefPubMedPubMedCentralGoogle Scholar
  3. 3.
    Briez N, Piessen G, Torres F, Lebuffe G, Triboulet J-P, Mariette C (2012) Effects of hybrid minimally invasive oesophagectomy on major postoperative pulmonary complications. Br J Surg 99(11):1547–1553. doi:10.1002/bjs.8931 CrossRefPubMedGoogle Scholar
  4. 4.
    Messager M, Pasquer A, Duhamel A, Caranhac G, Piessen G, Mariette C (2015) Laparoscopic gastric mobilization reduces postoperative mortality after esophageal cancer surgery: a French Nationwide Study. Ann Surg 262(5):817–822 . doi:10.1097/SLA.0000000000001470discussion 822-3CrossRefPubMedGoogle Scholar
  5. 5.
    Naunheim KS, Petruska P, Roy TS, Andrus CH, Johnson FE, Schlueter JM, Baue AE (1992) Preoperative chemotherapy and radiotherapy for esophageal carcinoma. J Thorac Cardiovasc Surg 103(5):887PubMedGoogle Scholar
  6. 6.
    van Hagen P, Hulshof MCCM, van Lanschot JJB et al (2012) Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med 366(22):2074–2084. doi:10.1056/NEJMoa1112088 CrossRefPubMedGoogle Scholar
  7. 7.
    Cunningham D, Allum WH, Stenning SP et al (2006) Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 355(1):11–20. doi:10.1056/NEJMoa055531 CrossRefPubMedGoogle Scholar
  8. 8.
    Lorenzen S, Hentrich M, Haberl C, Heinemann V, Schuster T, Seroneit T, Roethling N, Peschel C, Lordick F (2007) Split-dose docetaxel, cisplatin and leucovorin/fluorouracil as first-line therapy in advanced gastric cancer and adenocarcinoma of the gastroesophageal junction: results of a phase II trial. Ann Oncol 18(10):1673–1679. doi:10.1093/annonc/mdm269 CrossRefPubMedGoogle Scholar
  9. 9.
    Allen MS (1992) Ivor Lewis esophagectomy. Semin Thorac Cardiovasc Surg 4(4):320–323PubMedGoogle Scholar
  10. 10.
    Hoeppner J, Marjanovic G, Glatz T, Kulemann B, Hopt UT (2014) Laparoskopisch-thorakotomische Ösophagusresektion mit intrathorakaler Ösophagogastrostomie als Hybridverfahren (hybrid laparoscopic thoracotomic esophagectomy with intrathoracic esophagogastric anastomosis). Chirurg 85(7):628–635. doi:10.1007/s00104-014-2783-1 CrossRefPubMedGoogle Scholar
  11. 11.
    Dindo D, Demartines N, Clavien P-A (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240(2):205–213CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Low DE, Alderson D, Cecconello I et al (2015) International consensus on standardization of data collection for complications associated with esophagectomy: Esophagectomy Complications Consensus Group (ECCG). Ann Surg 262(2):286–294. doi:10.1097/SLA.0000000000001098 CrossRefPubMedGoogle Scholar
  13. 13.
    Haverkamp L, Seesing MFJ, Ruurda JP, Boone J, Rv H (2016) Worldwide trends in surgical techniques in the treatment of esophageal and gastroesophageal junction cancer. Diseases of the Esophagus: n/a-n/a. doi:10.1111/dote.12480 Google Scholar
  14. 14.
    Hölscher AH, Schneider PM, Gutschow C, Schröder W (2007) Laparoscopic ischemic conditioning of the stomach for esophageal replacement. Ann Surg 245(2):241–246. doi:10.1097/01.sla.0000245847.40779.10 CrossRefPubMedPubMedCentralGoogle Scholar
  15. 15.
    Mariette C, Meunier B, Pezet D et al. (2015) Hybrid minimally invasive versus open oesophagectomy for patients with oesophageal cancer: a multicenter, open-label, randomized phase III controlled trial, the MIRO trial. ASCO Meeting Abstracts 33(3_suppl): 5Google Scholar
  16. 16.
    Makowiec F, Baier P, Kulemann B, Marjanovic G, Bronsert P, Zirlik K, Henke M, Hopt UT, Hoeppner J (2013) Improved long-term survival after esophagectomy for esophageal cancer: influence of epidemiologic shift and neoadjuvant therapy. J Gastrointest Surg 17(7):1193–1201. doi:10.1007/s11605-013-2212-7 CrossRefPubMedGoogle Scholar
  17. 17.
    Glatz T, Marjanovic G, Zirlik K, Brunner T, Hopt UT, Makowiec F, Hoeppner J (2015) Chirurgische Therapie des Ösophaguskarzinoms: Entwicklung Von Management und Prognose über die letzten drei Jahrzehnte (surgical treatment of esophageal cancer : evolution of management and prognosis over the last 3 decades). Chirurg 86(7):662–669. doi:10.1007/s00104-014-2877-9 CrossRefPubMedGoogle Scholar
  18. 18.
    Porteous GH, Neal JM, Slee A, Schmidt H, Low DE (2015) A standardized anesthetic and surgical clinical pathway for esophageal resection: impact on length of stay and major outcomes. Reg Anesth Pain Med 40(2):139–149. doi:10.1097/AAP.0000000000000197 CrossRefPubMedGoogle Scholar
  19. 19.
    Jafari MD, Halabi WJ, Smith BR, Nguyen VQ, Phelan MJ, Stamos MJ, Nguyen NT (2013) A decade analysis of trends and outcomes of partial versus total esophagectomy in the United States. Ann Surg 258(3):450–458. doi:10.1097/SLA.0b013e3182a1b11d CrossRefPubMedGoogle Scholar
  20. 20.
    Hoeppner J, Zirlik K, Brunner T, Bronsert P, Kulemann B, Sick O, Marjanovic G, Hopt UT, Makowiec F (2014) Multimodal treatment of locally advanced esophageal adenocarcinoma: which regimen should we choose? Outcome analysis of perioperative chemotherapy versus neoadjuvant chemoradiation in 105 patients. J Surg Oncol 109(3):287–293. doi:10.1002/jso.23498 CrossRefPubMedGoogle Scholar
  21. 21.
    Luketich JD, Pennathur A, Awais O et al (2012) Outcomes after minimally invasive esophagectomy: review of over 1000 patients. Ann Surg 256(1):95–103. doi:10.1097/SLA.0b013e3182590603 CrossRefPubMedPubMedCentralGoogle Scholar
  22. 22.
    Maas KW, Cuesta MA, van Berge Henegouwen MI et al (2015) Quality of life and late complications after minimally invasive compared to open esophagectomy: results of a randomized trial. World J Surg 39(8):1986–1993. doi:10.1007/s00268-015-3100-y CrossRefPubMedPubMedCentralGoogle Scholar
  23. 23.
    Sihag S, Kosinski AS, Gaissert HA, Wright CD, Schipper PH (2015) Minimally invasive versus open esophagectomy for esophageal cancer: a comparison of early surgical outcomes from the Society of Thoracic Surgeons National Database. Ann Thorac Surg. doi:10.1016/j.athoracsur.2015.09.095 PubMedGoogle Scholar
  24. 24.
    Xiong W-L, Li R, Lei H-K, Jiang Z-Y (2015) Comparison of outcomes between minimally invasive oesophagectomy and open oesophagectomy for oesophageal cancer. ANZ J Surg. doi:10.1111/ans.13334 PubMedGoogle Scholar
  25. 25.
    Lee J-M, Cheng J-W, Lin M-T, Huang P-M, Chen J-S, Lee Y-C (2011) Is there any benefit to incorporating a laparoscopic procedure into minimally invasive esophagectomy? The impact on perioperative results in patients with esophageal cancer. World J Surg 35(4):790–797. doi:10.1007/s00268-011-0955-4 CrossRefPubMedGoogle Scholar
  26. 26.
    Bonavina L, Scolari F, Aiolfi A, Bonitta G, Sironi A, Saino G, Asti E (2015) Early outcome of thoracoscopic and hybrid esophagectomy: propensity-matched comparative analysis. Surgery. doi:10.1016/j.surg.2015.08.019 PubMedGoogle Scholar
  27. 27.
    Avery KNL, Metcalfe C, Berrisford R et al (2014) The feasibility of a randomized controlled trial of esophagectomy for esophageal cancer—the ROMIO (Randomized Oesophagectomy: Minimally Invasive or Open) study: protocol for a randomized controlled trial. Trials 15:200. doi:10.1186/1745-6215-15-200 CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2017

Authors and Affiliations

  1. 1.Department of General and Visceral Surgery and Faculty of MedicineUniversity of FreiburgFreiburgGermany

Personalised recommendations