The American Journal of Surgery xxx (2017) 1e4

Contents lists available at ScienceDirect

The American Journal of Surgery

journal homepage: www.americanjournalofsurgery.com

Near infrared perfusion assessment of gastric conduit during

minimally invasive Ivor Lewis esophagectomy
Brian G.A. Dalton, Abubaker A. Ali, Marie Crandall, Ziad T. Awad*
Department of Surgery, University of Florida Health- Jacksonville, United States

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: Anastomotic leak and conduit necrosis are devastating complications following Ivor Lewis
Received 27 September 2017 esophagectomy. Near infrared imaging (NIR) using IndoCyanine Green allows for real time tissue
Received in revised form perfusion assessment which may reduce anastomotic leak during minimally invasive Ivor Lewis
31 October 2017
esophagectomy (MIE).
Accepted 6 November 2017
Methods: Forty consecutive MIE were performed by a single surgeon at a tertiary referral center. The first
20 were assessed for gastric conduit perfusion by clinical criteria (Group 1). The second 20 were also
Keywords:
assessed using NIR laparoscopic system (Group 2).
Near infrared imaging
Ivor Lewis esophagectomy
Results: Comparing Group 1 to Group 2, no significant differences were found in overall complication
Minimally invasive esophagectomy rate, readmission or reoperation rate. NIR resulted in resection of the non perfused proximal portion of
Indocyanine green angiography the conduit in 30% (6/20). Two patients in group 2 group developed anastomotic leak (2/20) compared to
Anastomotic leak 0 in Group 1 (p ¼ 0.49). Graft necrosis led to one mortality in Group 1, while there were 0 mortalities in
Graft necrosis Group 2. (p ¼ 1.0).
Conclusion: Although NIR plays a role in assessment of tissue perfusion, in our study its use did not result
in reduction of anastomotic leak rate.
© 2017 Elsevier Inc. All rights reserved.

1. Introduction stomach and small vessels in the omentum along the greater
curvature.6
Esophagectomy is one of most invasive procedures in gastro- Historically, evaluation of the blood flow of the gastric conduit
intestinal surgery and despite improvements in surgical techniques has been a challenge to most surgeons as relying on inspection of
and perioperative care, the mortality rate remains high.1 A study the color of gastric serosa is often inaccurate and misleading.
using the National Surgery Quality Improvement Program database Indocyanine green (ICG) fluorescence angiography and Doppler
(NSQIP) between 2006 and 2011 reported that the morbidity of examination have recently been used to assess blood supply and
patients with esophagectomy was 43.8% and mortality was 3.8%.2 assist in conduit construction. These modalities have been
In particular, anastomotic leak after esophagectomy remains an employed in an effort to reduce anastomotic complications after
important cause of patient morbidity and impaired quality of esophagectomy.6,7
life.3e5 In the majority of cases, reconstruction after esophagectomy The Near Infrared (NIR) laparoscopic system (PINPOINT Endo-
is done using a gastric conduit that is perfused by the right gas- scopic Fluorescence Imaging System, NOVADAQ, Mississauga, ON,
troepiploic arcade. Perfusion at the proximal portion of the graft, in Canada) is used to provide high definition (HD) white light imaging
the area where the anastomosis is typically created, is variable but during minimally invasive surgery (MIS), as well as NIR irradiation
often tenuous because the gastroepiploic arcade rarely reaches the and ICG fluorescence emission imaging. This system provides a
tip of the graft. Instead, the most proximal portion of the graft is real-time false-color superimposition of ICG fluorescence signal on
typically perfused by intramural capillaries within the wall of the the HD white light view. This overlay allows simultaneous appre-
ciation of dynamic perfusion and visualization of the conventional
MIS video image.
* Corresponding author. Division of General and Gastrointestinal Surgery, In theory, the PINPOINT system might provide valuable real-time
University of Florida Health- Jacksonville, 653 West 8th Street, Jacksonville, FL information about perfusion of the gastric graft during minimally
32209, United States. invasive Ivor Lewis esophagectomy (MIE) and this might influence
E-mail address: ziad.awad@jax.ufl.edu (Z.T. Awad).

https://doi.org/10.1016/j.amjsurg.2017.11.026
0002-9610/© 2017 Elsevier Inc. All rights reserved.

Please cite this article in press as: Dalton BGA, et al., Near infrared perfusion assessment of gastric conduit during minimally invasive Ivor Lewis
esophagectomy, The American Journal of Surgery (2017), https://doi.org/10.1016/j.amjsurg.2017.11.026
2 B.G.A. Dalton et al. / The American Journal of Surgery xxx (2017) 1e4

resection margins and surgical outcomes. The hypothesis is the use

PINPOINT system during MIE esophagectomy would reduce anas-
tomotic leak rates.

2. Methods

After Institutional Review Board approval (IRB#201600651) a

retrospective analysis of prospectively collected data of all MIE per-
formed at our institution was performed. All MIE from April 2014 to
January 2016 were included. All surgeries were performed for cancer
and were performed by a single surgeon (ZA) at a tertiary referral
center. Cases were analyzed in 2 groups. Group 1, the conduit was
evaluated using standard clinical criteria by inspecting the color of the
gastric serosa. Group 2, NIR technology was used intraoperatively for Fig. 2. Gastric conduit after ICG injection on NIR with poor perfusion evident in
assessment of graft perfusion. Data is presented as mean ± standard proximal conduit. Line of perfusion demarcation (arrows) shown in PINPOINT modes.
deviation unless otherwise stated. Significance is defined as P  0.05.
Comparative analysis was performed using student t test for contin-
uous variables and Fishers exact for binary variables or Chi square test
with Pearson correlation where appropriate.

3. Technique

The abdominal portion of the procedure was performed lapa-

roscopically. Dissection began by opening the gastrocolic omentum
below the level of the pylorus, taking care to preserve the right
gastroepiploic arcade. The right gastric artery was preserved in all
patients. The greater curvature of the stomach was mobilized and
the short gastric vessels are divided. The left gastric artery was
dissected and transected using the vascular stapler after all nodal
tissue is lifted off the celiac axis. A 5 cm wide gastric conduit (Fig. 1)
was constructed using multiple firings of the laparoscopic linear Fig. 3. Gastric conduit after ICG injection showing good perfusion throughout conduit.
stapler. Pyroloplasty was performed in all patients.
The thoracic portion was performed thoracoscopically with the
patient in the left lateral decubitus position. The inferior pulmonary
When the conduit appeared well perfused throughout (Fig. 3) a
ligament was incised and the right lung was retracted. The medi-
stitch was still placed on the anterior surface of the conduit for
astinal pleura was incised and the esophagus was dissected to the
orientation purposes upon transposition into the chest.
level of the azygous vein, which was divided using the laparoscopic
All patients underwent upper GI contrast study on day 4 prior to
vascular stapler. The gastric conduit was pulled into the right chest,
oral intake. Anastomotic leak was defined as full thickness defect of
the specimen removed and end to end esophagogastric anasto-
esophagus, conduit, anastomosis or staple line.8 Oral intake was
mosis was constructed intracorporeally using the mechanical
started the day of upper GI contrast study if the study was negative
circular stapler 25 mm DST XL EEA (Covidien, Newhaven, CT, USA).
for leak.
Beginning in February 2015, NIR assessment of conduit perfu-
sion was done using the PINPOINT system - 7.5 mg of ICG is injected
intravenously and conduit perfusion assessment is done 60 s after 4. Results
injection. A marking silk stitch was placed when there was a
demarcation between the perfused and non-perfused part of the During the study period, 40 consecutive patients having
conduit (Fig. 2). The non-perfused segment was resected in the undergone MIE. Group 1, 20 consecutive patients (April 2014eJanuary
chest, and the anastomosis was constructed between the native 2015). Group 2, 20 consecutive patients (February 2015eJanuary
esophagus and perfused proximal stomach. 2016) were NIR was used. The administration of ICG caused no
adverse events in the NIR group. No differences were found between
the 2 groups with regard demographic or preoperative parameters
(Table 1).

Table 1
Demographic and preoperative date Group 1 vs Group 2.

Value Group 1 (n ¼ 20) Group 2 (n ¼ 20) P value

Age (years) 66.2 ± 8 61.8 ± 12.8 0.2

% Male 80 80 1
BMI (kg/m2) 26.3 ± 4.1 26.4 ± 4.9 0.97
% adenocarcinoma 70 90 0.24
Neoadjuvant therapy (%) 95 85 0.6
COPD (%) 20 15 1
Coronary artery disease (%) 15 25 0.69
Peroperative albumin (mg/dL) 3.9 ± 0.4 3.9 ± 0.4 0.69

Fig. 1. Tubularized gastric conduit. COPD: chronic obstructive pulmonary disease.

Please cite this article in press as: Dalton BGA, et al., Near infrared perfusion assessment of gastric conduit during minimally invasive Ivor Lewis
esophagectomy, The American Journal of Surgery (2017), https://doi.org/10.1016/j.amjsurg.2017.11.026
 B.G.A. Dalton et al. / The American Journal of Surgery xxx (2017) 1e4 3

Table 2 fluorescence angiography- SPY Elite System (LifeCell, Bridgewater,

Perioperative outcomes Group 1 and Group 2. NJ, USA) among patients undergoing Ivor Lewis esophagectomy.7
Value Group 1 Group 2 P value They adopted a quantitative perfusion cutoff of 75% as the area
(n ¼ 20) (n ¼ 20) where the esophagogastric anastomosis was performed, usually
Anastomotic leak (%) 0 10 0.49 10 cm above the pylorus. The leak rate was 0% (30 patients) when
Graft necrosis (%) 5 0 1 the anastomosis was placed proximal to the 75% relative perfusion
Length of stay (days) 8.1 ± 1.9 9.8 ± 4 0.09 line compared to 20% leak rate (60 patients) prior to the use of
Overall complication rate (%) 55 40 0.53
either technology.7 In both studies,6,7 a midline abdominal incision
Reoperation within same 5 10 1
admission (%) is required to deliver the conduit to the outside for quantitative
90 day reoperation (%) 15 15 1 perfusion assessment using SPY System or intraoperative Doppler,
90 day readmission (%) 20 10 0.66 while our approach is pure minimally invasive (laparoscopic and
90 day mortality (%) 5 0 1
thoracoscopic). We did not use laparoscopic Doppler and relied
Atrial fibrillation (%) 25 25 1
Respiratory failure 10 0 0.49
solely on qualitative assessment of gastric conduit perfusion. If
quantitative data were available with the PINPOINT system
establishing a standardized level of perfusion necessary for suc-
cessful anastomosis, this could play a role in reducing anastomotic
All patients were diagnosed with esophageal cancer. There was
leak rate. Degett et al. in a systemic literature review of ICG- FA
no difference in operative time (371 ± 90 vs 379 ± 66 min, p ¼ 0.76)
(fluorescence angiography) to assess perfusion during construc-
or estimated blood loss (159 ± 150 ml vs 338 ± 723, p ¼ 0.28)
tion of a primary gastrointestinal anastomosis showed the tech-
between the 2 groups. Mean blood loss and standard deviation (SD)
nique to be associated with reduced risk of anastomotic colorectal
in Group 2 was increased by the presence of one outlier (>2 SD
leaks compared with no ICG-FA.12 However, the anastomotic
outside the mean). This patient had an injury to an aberrant pul-
leakage rate in patients with esophageal anastomoses and intra-
monary vein branch which was controlled but the esophagogas-
operative ICG-FA assessment was 14% (n ¼ 30/214). None of the
trostomy was delayed until an operating room return 2 days later.
studies involving esophageal anastomoses had a control group
This was also the only patient in the series to receive an intra-
without ICG-FA assessment. They concluded that ICG-FA is a
operative transfusion.
feasible and promising tool. However, heterogeneous studies and
No statistically significant differences were found between the 2
the lack of high level of evidence the clinical benefit of ICG-FA is
groups in overall complication rate, anastomotic leak, graft necro-
inconclusive. Furthermore, the technique remains subjective until
sis, reoperation or length of stay (Table 2). The use of NIR resulted in
more objective cutoff levels for sufficient perfusion has been
resection of proximal non-perfused proximal part of the conduit in
established.12
30% (6/20) of patients. The 2 leaks in the NIR were in the group that
In our study, 6/20 (30%) of patients underwent resection of
needed proximal conduit resection, and both leaks were managed
non-perfused proximal gastric conduit using the NIR, allowing a
with endoscopic stent placement. These endoscopic stents were
well-perfused gastric conduit to be consistently used for the
placed in an elective fashion.
esophagogastric anastomosis. It is concerning, however, that
In Group 1, 3 patients required surgical intervention within 90
anastomotic leaks were observed in 2/6 (33.33%) patients in this
days: one patient for wound infection required debridement at the
group, and raises the possibility that proximal resection might have
jejunostomy site, one for cecal volvulus necessitating right colon
created undue anastomotic tension and compromised blood sup-
resection, and the third patient was explored for graft necrosis. This
ply. Although tension is difficult to assess and quantify, especially
patient underwent resection of the graft with cervical esoph-
when minimally invasive surgery is utilized, we routinely roll the
agostomy, and was the only mortality from either group. In Group
anastomosis from side to side and lift it to examine the entire staple
2, reoperation within 90 days were observed in 3 patients: one
line; this maneuver would not be possible had the anastomosis
patient with conduit obstruction as a result of redundant conduit
being under tension. Is it unknown whether those patients with
above the diaphragm, the patient with pulmonary vein branch
demarcation between the perfused and non-perfused part of the
injury described earlier, and one patient with post esophagectomy
conduit using NIR are at risk of anastomotic failure, and proximal
hiatal hernia. Ten total patients were diagnosed with postoperative
resection was a marker of a potentially higher-risk anastomosis.
atrial fibrillation. One of these patients required cardioversion
Other authors agree with our opinion that the intra-thoracic
(Group 2). One patient in the Group 1 was diagnosed with a Clos-
anastomosis provides the ability to resect the tip of the gastric
tridium Difficile infection and was treated with antibiotics. Another
conduit, and the anastomosis is not under excessive tension13 as
patient in Group 1 developed colonic pseudo-obstruction and was
the surgeon has control on the level of transection of the esophagus
treated non-operatively.
relative to the location of the esophageal cancer.
Limitations of this study include small population, retrospective
5. Discussion nature, single center and single surgeon, all of which introduce
potential bias. However, to our knowledge, the present study is the
A large database study examining all modalities of esoph- first using the PINPOINT system for graft perfusion assessment
agectomy showed overall major morbidity of esophagectomy to be among patients undergoing strict laparoscopic/thoracoscopic
33.8% and anastomotic leak rate of 12.9%.9 Anastomotic leak has esophagectomy. Future directions of research include studying this
also been linked to increased morbidity in patients undergoing modality of NIR using quantified perfusion values should they come
esophagectomy.10,11 available. We would hope to establish a level of perfusion that
Our study did not answer the question if NIR can reduce the results in a negligible anastomotic leak rate.
anastomotic leak rate. Zehetner et al. using SPY Imaging System
(Novadaq, Ontario, Canada) among patients undergoing open 6. Conclusion
transhiatal esophagectomy showed a significantly higher leak rate
(45% vs 2%, p < 0.0001) when the anastomosis was performed to an In our study, the use of NIR or interventions based on its findings
area of gastric conduit distal to the line of perfusion demarcation.6 did not result in reduction of anastomotic leak rate. Better quanti-
Campbell et al. employed both Doppler and intraoperative ICG tative integrated metrics are needed to better define tissue

Please cite this article in press as: Dalton BGA, et al., Near infrared perfusion assessment of gastric conduit during minimally invasive Ivor Lewis
esophagectomy, The American Journal of Surgery (2017), https://doi.org/10.1016/j.amjsurg.2017.11.026
4 B.G.A. Dalton et al. / The American Journal of Surgery xxx (2017) 1e4

perfusion using the PINPOINT system. Further studies are war- 3. Fernandez FG, Meyers BF. Quality of life after esophagectomy. Semin Thorac
Cardiovasc Surg. 2004;16(2):152e159. Summer.
ranted to validate the use of this novel technology.
4. Schuchert MJ, Abbas G, Nason KS, et al. Impact of anastomotic leak on out-
comes after transhiatal esophagectomy. Surgery. 2010 Oct;148(4):831e838.
Conflicts of interest discussion 838e840.
5. Kassis ES, Kosinski AS, Ross Jr P, Koppes KE, Donahue JM, Daniel VC. Predictors
of anastomotic leak after esophagectomy: an analysis of the society of thoracic
No conflicts of interest were encountered during this project by surgeons general thoracic database. Ann Thorac Surg. 2013 Dec;96(6):
any of the authors. No external financial or material support was 1919e1926.
provided, and there are no commercial or financial interests related 6. Zehetner J, DeMeester SR, Alicuben ET, et al. Intraoperative assessment of
perfusion of the gastric graft and correlation with anastomotic leaks after
to the topic. esophagectomy. Ann Surg. 2015 Jul;262(1):74e78.
7. Campbell C, Reames MK, Robinson M, Symanowski J, Salo JC. Conduit vascular
Author contributions evaluation is associated with reduction in anastomotic leak after esoph-
agectomy. J Gastrointest Surg. 2015 May;19(5):806e812.
8. Low DE, Alderson D, Cecconello I, et al. International consensus on standardi-
BD: Data collection, analysis, interpretation, manuscript draft- zation of data collection for complications associated with esophagectomy:
ing, critical revision AA: Study design, data collection, critical esophagectomy complications consensus group (ECCG). Ann Surg. 2015
Aug;262(2):286e294.
revision. 9. Raymond DP, Seder CW, Wright CD, et al. Predictors of major morbidity or
MC: Interpretation, manuscript drafting, critical revision. mortality after resection for esophageal cancer: a society of thoracic surgeons
ZA: Study design, data collection, analysis, interpretation, general thoracic surgery database risk adjustment model. Ann Thorac Surg.
2016 Jul;102(1):207e214.
manuscript drafting, critical revision.
10. Ryan CE, Paniccia A, Meguid RA, McCarter MD. Transthoracic anastomotic leak
All authors agree to the manuscript its submitted form and after esophagectomy: current trends. Ann Surg Oncol. 2017 Jan;24(1):281e290.
attest to the accuracy of the data and interpretation of the data. 11. Alanezi K, Urschel JD. Mortality secondary to esophageal anastomotic leak. Ann
Thorac Cardiovasc Surg. 2004 Apr;10(2):71e75.
12. Degett TH, Andersen HS, Go €genur I. Indocyanine green fluorescence angiog-
References raphy for intraoperative assessment of gastrointestinal anastomotic perfusion:
a systematic review of clinical trials. Langenbecks Arch Surg. 2016 Sep;401(6):
1. Saeki H, Masuda T, Okada S, et al. Impact of perioperative peripheral blood 767e775.
values on postoperative complications after esophageal surgery. Surg Today. 13. Nguyen NT, Hinojosa MW, Smith BR, Chang KJ, Gray J, Hoyt D. Minimally
2010 Jul;40(7):626e631. invasive esophagectomy: lessons learned from 104 operations. Ann Surg. 2008
2. Lucas DJ, Pawlik TM. Quality improvement in gastrointestinal surgical oncology Dec;248(6):1081e1091.
with American college of surgeons national surgical quality improvement
Program. Surgery. 2014 Apr;155(4):593e601.

Please cite this article in press as: Dalton BGA, et al., Near infrared perfusion assessment of gastric conduit during minimally invasive Ivor Lewis
esophagectomy, The American Journal of Surgery (2017), https://doi.org/10.1016/j.amjsurg.2017.11.026