ORIGINAL ARTICLE
Technical Challenges of Total Esophageal Reconstruction Using
A Supercharged Jejunal Flap
Melissa Poh, MD,∗ Jesse C. Selber, MD, MPH,∗ Roman Skoracki, MD,∗ Garrett L. Walsh, MD,† and Peirong Yu, MD
Objective: To review our experience and technique of the supercharged jeju-
nal flap for total esophageal reconstruction.
Background: A gastric pull-up is the first choice for total esophageal recon-
struction. When this fails or when the stomach is unavailable, a supercharged
jejunal flap may reestablish alimentary tract continuity.
Methods: We performed a retrospective review of 51 patients who under-
went a supercharged jejunal flap for total esophageal reconstruction between
March 2000 and September 2009 at a single institution. Patient characteristics,
technical details, and outcomes were analyzed.
Results: Thirty-six men and 15 women patients were included with a mean
age of 55 (28–74) years. An immediate reconstruction was performed in 34
(67%) patients and delayed in 17 patients. The jejunal conduit was passed
through a substernal route in 31 (60%) patients and a retrocardiac route in
20 patients. Most common recipient arteries were the internal mammary and
transverse cervical. Most common recipient veins were the internal mammary
and internal jugular. The overall success rate was 94% with 3 flap failures.
A total of 33 patients experienced 1 or more complications with abdominal
wound infection and pulmonary complications being the 2 most frequent.
Mean length of hospital stay was 21.5 ± 14.0 days. Forty-four (90%) patients
were able to achieve a regular diet and 39 (80%) patients discontinued their
tube feeds.
Conclusion: This technically challenging operation requires a multidisci-
plinary approach and careful planning, yet can be successfully performed with
good long-term function and acceptable morbidity. An algorithm delineating
the operative strategy is presented.
(Ann Surg 2011;253:1122–1129)
T otal esophageal reconstruction is a demanding procedure for both
patient and surgeon alike. A gastric conduit is the first choice
for reconstruction because of its robust blood supply and need for
only a single bowel anastomosis to reestablish continuity with good
results.1,2 However, in certain situations, the stomach may be un-
available. Colon interposition has been employed for several decades
as an alternative to the gastric conduit.3–7 However, patients with
atherosclerosis, inappropriate vascular anatomy, previous colon re-
section or intrinsic colon disease such as inflammatory bowel disease,
cancer or diverticulosis, are not candidates.
The jejunum is a useful alternative for esophageal reconstruc-
tion because it closely approximates the size of the native esophagus,
possesses peristaltic activity, and is typically free of intrinsic diseases.
Free jejunal flaps, which are often utilized in cases of hypopharyn-
geal and cervical esophageal carcinoma,7–12 are limited to partial
esophageal defects due to the segmental blood supply. However, the
addition of microsurgical techniques has allowed for the replacement
of the entire esophagus using longer conduits through “supercharg-
From the Departments of ∗ Plastic Surgery and †Thoracic and Cardiovascular
Surgery, MD Anderson Cancer Center, The University of Texas, Houston TX.
No intramural or extramural funding supported any aspect of this work.
Reprints: Peirong Yu, MD, 1515 Holcombe Blvd, No. 443, Houston, TX 77030.
Copyright C 2011 Wolters Kluwer Health | Lippincott Williams & Wilkins
ISSN: 0003-4932/11/25306-1122
DOI: 10.1097/SLA.0b013e318217e875
1122 | www.annalsofsurgery.com
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
ing.” Longmire13 described the first supercharged jejunal flap for
esophageal reconstruction in 1947. Since then, the field of micro-
surgery has advanced significantly and the use of these techniques
has become more widespread, making the supercharged jejunal flap
(SCJF) a viable option for total esophageal reconstruction when the
stomach is unavailable.
In the following report, we present our series of 51 patients who
underwent supercharged jejunal transfer for total esophageal recon-
struction between March 2000 and September 2009 at MD Anderson
Cancer Center.
METHODS
A retrospective review of a prospectively maintained database
was performed. Patients who underwent a SCJF for total esophageal
reconstruction from March 2000 through September 2009 at MD
Anderson Cancer Center were included in the study. The study was
granted approval by our institutional review board before patient
accrual or data collection.
Operative Technique
Meticulous operative planning and intraoperative execution are
essential. We utilize a 3-team approach in most cases. The esophagec-
tomy and gastrectomy, when necessary, are performed by the thoracic
and/or general surgeons. The surgical approach is either through 2
incisions (abdominal and cervical) as in a transhiatal esophagectomy,
or 3 incisions with the addition of a right thoracotomy.
Once resection is complete, the ligament of Treitz is identified
and the small bowel with its mesentery is explored to ensure no
intrinsic abnormalities or iatrogenic injuries exist. The length of the
esophageal defect is measured to estimate the length of the conduit
needed (Fig. 1). The mesentery of the proximal and midjejunum are
closely examined using a fiberoptic light to transilluminate the tissues
and elucidate the vascular anatomy (Fig. 2). The first mesenteric
branch beyond the ligament of Treitz is identified and preserved to
maintain blood supply to the distal duodenum and most proximal
portion of jejunum, which is used to reestablish enteric continuity.
Typically, the second mesenteric branch is dissected down to
the level of the superior mesenteric artery and vein, and ultimately
divided to serve as the vascular pedicle for supercharging the proximal
flap. In the simplest cases, the mesentery between the second and
third branches is divided up to the serosal border, allowing the jejunal
segment to unfurl (Fig. 3). This step helps straighten the natural
sinusoidal properties of the small bowel and reduce redundancy. If
more length is required, the third mesenteric branch is ligated and
divided (Fig. 4). Secondary arcade vessels between the third and
fourth mesenteric branches are preserved so that the segment of bowel
normally supplied by the third branch is now supplied by the pedicled
fourth branch through intact arcade vessels. This approach is the most
common set-up.
If still greater length is required, such as in patients with a
very long torso or a concomitant total laryngopharyngectomy, the
fourth mesenteric branch can also be ligated. In these cases, only
the mesentery between the third and fourth branches is divided to the
serosal border while the arcade connections are preserved between the
second and third, and between the fourth and fifth branches (Fig. 5).
Annals of Surgery r Volume 253, Number 6, June 2011
Annals of Surgery r Volume 253, Number 6, June 2011
FIGURE 1. The length of a jejunal conduit is estimated by
measuring the distance from below the xiphoid to the lower
neck.
The third segment receives perfusion from the supercharged second
segment and the fourth segment receives perfusion from the pedicled
fifth branch. Once adequate length is achieved, the jejunum is divided
proximally with a linear cutting stapler, typically 30 to 40 cm distal
to the ligament of Treitz.
There are 2 potential routes for transferring the jejunal flap
to the neck: retrocardiac, which is the orthotopic route, or subster-
nal, which is the heterotopic route (Fig. 6). The former is utilized
in patients undergoing immediate reconstruction. The latter is usu-
ally reserved for patients undergoing a delayed reconstruction after
previous failed esophageal reconstruction. In these cases, a portion
of the manubrium, clavicular head, and first rib are removed to en-
large the thoracic inlet and avoid constriction on the jejunal conduit.
Whether the retrocardiac or substernal route is selected, transfer is
accomplished using a sterile laparoscopic camera bag to protect the
conduit and prevent traction and shearing forces on the delicate arcade
vessels.
Recipient vessels are prepared before dividing the vascular
pedicle and passing the jejunal conduit. The transverse cervical ves-
sels, the external carotid system and jugular veins, and the internal
mammary vessels are common recipient vessels depending on the
clinical scenario. Once the jejunal flap is passed, arterial anastomoses
are performed under the operating microscope using 9-0 nylon su-
tures. Venous anastomoses are commonly completed using a venous
coupling device (Synovis Surgical Innovations, St Paul, MN). A vein
graft is used in some cases to bridge the gap between the jejunal
vascular pedicle and the recipient vessels.
Once the proximal jejunum is revascularized, any excess length
of jejunal conduit is removed to minimize redundancy in the neck.
The proximal 3 to 5 cm of jejunum based on 1 or 2 terminal arcade

C 2011 Lippincott Williams & Wilkins
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Suprecharged Jejunal Flap
FIGURE 2. The mesentery of the jejunum is transilluminated
with a fiberoptic light to identify the mesentery branches
and arcade vessels. The jejunal segment with the widest span
of mesentery, usually located between the second and third
mesentery branch vessels, is selected.
vessels is exteriorized as a postoperative monitoring segment (Fig. 7).
The esophagojejunal anastomosis is then performed in the neck using
a single layer of 3-0 vicryl or polydioxanone suture in an end-to-end
fashion. In patients with a total laryngopharyngectomy, the proximal
pharynx or base of tongue is larger than the jejunal diameter, so
an end-to-side anastomosis is performed to accommodate the size
discrepancy.
Intestinal continuity is reestablished in the abdomen by the
thoracic or general surgical team through a gastrojejunal anastomosis
using the posterior wall of the stomach, or through a Roux-en-Y
jejunojejunal anastomosis if a gastrectomy was performed. A feeding
jejunostomy tube is routinely placed. Initially, nasogastric tubes were
passed through the conduit and into the gastric remnant; however,
the difficulty in placing the tube and the incidence of conduit injury
secondary to the tube led to its abandonment.
The externalized monitoring segment of the jejunum is ob-
served for 7 to 10 days postoperatively at which time its mesentery
is ligated and the segment removed. Patients are recovered first in the
intensive care unit and then transferred to the ward. If the patient’s
postoperative course is uneventful, a modified barium swallow is per-
formed 7 to 14 days postprocedure. If no leak is detected, the patient
is started on a liquid diet and slowly advanced to a postgastrectomy
diet, as tolerated.
RESULTS
From March 2000 to September 2009, 51 patients underwent
supercharged jejunal flaps for total esophageal reconstruction by 4 dif-
ferent microsurgeons. The patient characteristics are shown in Table 1.
The majority of patients (76%) had one or more comorbidity with hy-
pertension, malignancy other than esophageal or gastric carcinoma,
and coronary artery disease as the 3 most common conditions. Pa-
tients who received preoperative chemotherapy and radiation include
those that underwent these treatments in the past for esophageal can-
cer and then presented with a recurrence and those that presented to
our institution with an initial diagnosis of esophageal cancer and were
in the process of receiving triple-modality therapy.
The primary diagnosis was esophageal carcinoma in 38 pa-
tients (75%; Table 2) including 26 adenocarcinomas and 12 squamous
cell carcinomas. Eleven of the 38 patients had recurrent cancer, in-
cluding 7 who developed a recurrence after a previous esophagectomy
www.annalsofsurgery.com | 1123
Poh et al
FIGURE 3. The mesentery between the second and third branch vessels is divided up to the serosal border, allowing the jejunal
segment to unfurl. The second mesentery vessels are used for supercharging and the third set is left intact.
with gastric conduit reconstruction and 4 patients who developed a
recurrence after definitive chemoradiation treatment. These 4 patients
were reconstructed with a SCJF instead of a gastric pull-up because
of radiation induced compromise to the stomach’s vasculature. Three
patients underwent surgery for high-grade dysplasia in the setting
of Barrett’s esophagus. One such patient experienced postoperative
ischemia of the gastric conduit, resulting in gastrectomy and diver-
sion, followed by a delayed SCJF reconstruction. Another patient
had previous fundoplication and the stomach was deemed inade-
quate. The other, young and otherwise healthy, patient with severe
reflux symptoms underwent a SCJF for esophageal reconstruction
based on preoperative discussions between the patient and plastic
and thoracic surgeons. Nine patients had an esophagectomy for other
tumors including a gastrointestinal stromal tumor in 3 patients, lym-
phoma leading to a tracheoesophageal fistula in 3 patients, submu-
cosal Histoplasmosis lesion in 1, papillary thyroid cancer involving
the esophagus with perforation in 1, and laryngeal cancer extending
to the esophagus in 1 patient. One patient experienced an esophageal
perforation from Boerhaave’s syndrome with attempted repair at an
outside hospital that lead to an esophagopleural fistula.
An immediate reconstruction was performed in 34 (67%) pa-
tients and the reasons a jejunal flap was used are listed in Table 3.
The most common indication was the need for a total or subtotal
gastrectomy (9). Recurrence (5) or complications (2) after a previous
esophagectomy was another common indication. Six patients had a
1124 | www.annalsofsurgery.com
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 253, Number 6, June 2011
history of gastric surgery, including an atypical gastric “wrap” with
polypropylene mesh for obesity (1), a partial gastrectomy for ulcer or
adenomatous polyps (2), Nissen fundoplication with gastrostomy and
significant scarring (2), and a pyloroplasty plus gastrojejunostomy for
congenital pyloric stricture. The SCJF was chosen in these patients
because the remaining stomach was felt to be of insufficient size,
or high risk for complications with a gastric pull-up. Five patients
underwent an immediate SCJF because of a history of radiation to
the stomach. Four patients required a total laryngopharyngectomy for
recurrent laryngeal cancer (1), tracheoesophageal fistula 4 cm below
the vocal cords (1), and for proximal esophageal cancer (2). After
careful intraoperative evaluation, it was felt that a gastric pull-up was
not a good option due to the high lesion and long torso. Two patients
underwent a SCJF for individual reasons including previous use of
the omentum as a flap, rendering the gastroepiploic vessels unavail-
able, and the presence of a large pancreatic cyst, which concerned
the surgeon about a potential pancreatic fistula if a gastric conduit
was used. One patient with severe reflux symptoms preoperatively,
wished not to have a gastric pull-up after discussing the options with
the surgeons.
A delayed reconstruction was performed in 17 patients
(Table 4), most commonly after a damage control procedure such
as debridement and diversion for a failed gastric pull-up (9 pa-
tients). Four patients presented with a tracheoesophageal fistula due
to (1) external beam radiation for recurrent papillary carcinoma and

C 2011 Lippincott Williams & Wilkins
Annals of Surgery r Volume 253, Number 6, June 2011 Suprecharged Jejunal Flap

FIGURE 4. If more length is required, the third mesenteric branch is ligated and divided. Secondary arcade vessels between the
3rd and 4th mesenteric branches are preserved so that the segment of bowel normally supplied by the 3rd branch can now be
supplied by the pedicled 4th branch via arcade connections.

FIGURE 5. If still greater length is required, the fourth mesenteric branch can also be ligated. In these cases, only the mesentery
between the third and fourth branches is divided to the serosal border while arcade connections are preserved between the 2nd
and 3rd, and between the 4th and 5th branches.

esophageal stenting, who required a concomitant total laryngopha-
ryngectomy; (2) esophageal cancer invading the trachea; (3) mul-
tiple courses of chemotherapy for recurrent episodes of Hodgkin’s
lymphoma involving the stomach; and (4) chronic abscess from
microperforation of midesophagus secondary to pesticide ingestion
and previous esophagogastrectomy. These patients were first treated
with esophagectomy and diversion, followed by a delayed SCJF for
definitive reconstruction. Gastric pull-up was not chosen because of
the unavailability (2), prior surgery and radiation (1), and a concomi-
tant total laryngopharyngectomy in a patient with a very long torso.
Two patients developed unexplained gastric necrosis after staging la-
paroscopy and feeding jejunostomy for newly diagnosed esophageal
cancer, requiring a diversion and delayed SCJF. One patient with
Boerhaave’s syndrome failed a repair with Nissen fundoplication,


C 2011 Lippincott Williams & Wilkins www.annalsofsurgery.com | 1125

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Poh et al Annals of Surgery r Volume 253, Number 6, June 2011

TABLE 1. Cohort Demographics

Mean age (years) 55 (28–74)
Sex
Men 36
Women 15
Body mass index (BMI) 25 (12–42)
Smokers (patients)
History of tobacco use 32
Active smoking 6
Comorbidities (patients) 39 (76%)
Hypertension 22
Other malignancy 10
Coronary artery disease 7
Diabetes mellitus 5
COPD/asthma 5
Rheumatological disease 4
Peripheral vascular disease 3
Cerebral vascular accident 3
Preoperative treatments (patients)
Chemotherapy 33
Radiation 34
Both 31
COPD indicates chronic obstructive pulmonary disease.
FIGURE 6. Drawing showing the retrocardiac (left) and sub-
sternal (right) pathways to pull the jejunal conduit to the neck.
TABLE 2. Primary Diagnosis
Diagnosis Patients
Esophageal cancer 38 (75%)
Distal/Gastroesophageal junction 23
Recurrent 11
Proximal/Midesophagus 4
Barrett’s esophagus with high-grade dysplasia 3
Other tumors 9
Esophageal perforation 1
Total 51

TABLE 3. Reasons for Immediate Supercharged Jejunal

Reconstruction
Reasons Patients
Gastrectomy required 9
Recurrence or complications after esophagectomy 7
Prior gastric surgery 6
FIGURE 7. A short segment of proximal jejunum supplied by Radiated stomach 5
Defects to base of tongue 4
1 or 2 terminal arcade vessels is externalized to monitor per-
Prior omental flap 1
fusion. A silk tie is placed loosely around the terminal arcade Large pancreatic cyst 1
vessels during surgery. When it is ready to remove the mon- Severe reflux symptoms 1
itoring segment, one can simply tie down the preplaced silk Total 34
ties.

also requiring a diversion and delayed SCJF. One patient who had a TABLE 4. Indications for Delayed Supercharged Jejunal
previous Ivor-Lewis esophagectomy with a gastric conduit for recon- Reconstruction
struction developed a severe stricture at the anastomosis leading to Reasons Patients
near complete obstruction.
The conduit was passed through a substernal route in 31 (61%) Gastric necrosis or leak after esophagectomy 9
patients and a retrocardiac route in 20 (39%) patients. Because the Tracheoesophageal fistula 4
Gastric necrosis 2
substernal pathway was chosen in the majority of cases, it follows that
Boerhaave’s and failed Nissen repair 1
the internal mammary artery and vein were predominantly used as Esophageal obstruction following esophagectomy 1
the recipient vessels because of their close proximity to the proximal Total 17
flap when transposed to the neck (Table 5). Other recipient vessel

1126 | www.annalsofsurgery.com 
C 2011 Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 253, Number 6, June 2011 Suprecharged Jejunal Flap

a jejunojejunostomy. An attempted free jejunal transfer on another

TABLE 5. Recipient Vessels Used for Supercharging
proximal stricture was unsuccessful. Six patients (12%) developed a
Artery Patients Vein Patients ventral hernia requiring surgical repair. Three patients were noted to
have minor dumping syndrome. Eight patients (16%) noted a neck
Internal mammary 28 Internal mammary 20
bulging during meals.
Transverse cervical 8 Internal jugular 20
External carotid 5 External jugular 7
Common carotid 5 Transverse cervical 3 DISCUSSION
Superior thyroid 3 Innominate 1 Free jejunal transfer for pharyngoesophageal defects is a
Lingual 2 well-accepted, safe technique that provides a reliable functional
reconstruction.8–12 The jejunal segment that is needed to complete
the reconstruction is typically less than 15 cm. Once the esophageal
TABLE 6. Complications defect is greater than 20 cm, another reconstructive option must be
used because each mesenteric vessel only supplies up to 20 cm of
Medical Cases Surgical Cases jejunum.14–16 Hence, either a double-pedicle, free jejunal transfer, or
Respiratory failure 11 (21%) Fistula 7 (14%) a pedicled jejunum with supercharging must be used. Both options
Pneumonia 8 (16%) Radiographic leak 6 (12%) can provide sufficient length for a total esophageal reconstruction;
Arrhythmia 6 (12%) Stricture 5 (10%) but the latter avoids the need to perform a second microvascular
Multiorgan failure 3 (6%) Chest complications∗ 10 (20%) anastomosis.
Pericardial effusion 2 (4%) Abdominal wound infection 11 (21%) Supercharged jejunal transfer is a technically demanding
Visceral perforation 2 (4%) Ventral hernia repair 6 (12%) surgery yet reasonable functional outcomes can be achieved.17–21 The
Congestive heart failure 1 (2%) Ileus 3 (6%)
present study is the largest series of supercharged jejunal flaps for
C. Diff colitis 1 (2%) Neck bulging 8 (16%)
total esophageal reconstruction to date. The majority of our patients
∗
Chest complications included pleural effusion, empyema, and mediastinal infec- were able to achieve a regular diet (90%) and discontinue their tube
tion.
feeds (80%), which has a tremendous psychological and physical
benefit to the patient. Reflux was infrequent and mild.
Our overall success rate using a SCJF for total esophageal re-
choices are shown in Table 5. Eleven patients (22%) required the use construction was 94%, with 3 flap failures in our early experience.
of vein grafts. Two patients died during their hospitalization and medical and surgi-
Mean ischemia time was 116 ± 29 minutes. Mean length of cal complications occurred in 65% of patients, reflecting the fact that
hospital stay was 21.5 ± 14.0 days with a mean intensive care unit the SCJF is a challenging and complex procedure. Patients often have
(ICU) stay of 9.0 ± 11.7 days. Five patients did not require ICU stay. significant comorbid diseases and therefore need be nutritionally and
Two flaps were returned to the operating room—one for bleeding and medically optimized before surgery. Major postoperative respiratory
the other for evidence of arterial thrombosis—both were salvaged. complications are particularly common and difficult to manage be-
Three flaps were lost (5.9%). The reasons for flap failure include cause of the involvement of multiple organs and body cavities, major
abdominal compartment syndrome and multiorgan failure, anasto- fluid shifts, frequent need for fluid resuscitation, and the relative con-
motic dehiscence with midjejunum ischemic stricture, and anasto- traindications to using vasopressors in the setting of microvascular
motic leak with tracheal necrosis and subsequent multiorgan failure. reconstruction.
The mean time to first oral intake was 25.5 ± 20.5 days. There was Our conduit-related complications are comparable to those re-
no 30-day mortality; however, 2 patients died during their hospital- ported in the literature18,19 and the majority of those occurred in
ization. One patient succumbed to respiratory and multiorgan failure our early experience. All subclinical radiographic leaks healed spon-
due to tracheal necrosis after takedown of a radiation-induced tra- taneously within 2 weeks. Once a leak is identified on the initial
cheoesophageal fistula at the time of the esophagectomy and SCJF modified barium swallow, the patients are kept Nil per os for ad-
reconstruction. The second patient was an elderly man with multiple ditional 2 weeks and then a modified barium swallow is repeated.
cardiovascular and pulmonary comorbidities who developed conges- Most clinical fistulas also heal with conservative management. Major
tive heart failure postoperatively and ultimately died from respiratory leaks or anastomotic breakdowns should be surgically explored and
failure secondary to pneumonia. possibly reconstructed to prevent catastrophic complications. Anas-
Forty-four patients (90%) were able to achieve a regular diet tomotic strictures are typically managed with endoscopic dilatation
and 39 (80%) patients were able to discontinue their tube feeds. The and repeated dilatations may be required. Two patients in our series
mean length of time to discontinuing tube feeds was 103 ± 81 days. developed stricture in the watershed jejunal segment in the middle
The mean length of follow-up was 21.9 months (2–80 months). of the conduit requiring surgical intervention, which highlights the
A total of 33 patients experienced 1 or more complications importance of selecting the appropriate jejunal segment based on the
during the study period (Table 6). Six patients (12%) demonstrated mesenteric anatomy and understanding the extent to which arcade
a leak on barium swallow without a clinical fistula, which all healed vessels can be divided.
spontaneously. Seven patients (14%) developed a esophagocutaneous After passing the bowel through the chest and into the neck, ten-
fistula in the neck due to anastomotic leak (4), erosion from a Blake sions on the mesentery may compromise the perfusion of the middle
drain (2), and chest compression for cardiac arrest (1). Three of the jejunal segment. This segment should be carefully inspected before
fistulas healed spontaneously. One drain injury was repaired primarily abdominal closure. If the perfusion is questionable, it can be revascu-
and covered with a sternocleidomastoid muscle flap. Another was larized by anastomosing its mesenteric branch to the gastroepiploic,
repaired 6 months later with direct excision and reanastomosis. One right gastric, mesocolonic vessels, or even the stump of the superior
was left unrepaired after several unsuccessful attempts. The other mesenteric vessels with or without a vein graft. Revascularization of
patient died from tracheal necrosis. Five patients (10%) developed the middle segment was successfully performed in 4 patients in our
strictures, 3 were located at the proximal anastomosis and 2 in the study.
middle segment of the conduit. Endoscopic dilatations were employed Typically, 3 surgical services are involved in this type of cases:
in 3 patients. One proximal stricture was successfully managed with thoracic surgery, general surgery, and plastic surgery. To minimize


C 2011 Lippincott Williams & Wilkins www.annalsofsurgery.com | 1127

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Poh et al
FIGURE 8. Proposed algorithm for choosing recipient vessels
and transthoracic pathways.
the operating time, it is important to establish a coordinated flow be-
tween the various teams. If a thoracotomy is required, the patient is
first placed in lateral decubitus position and then switched to supine.
While the ablative surgical team is operating in the abdomen, the re-
constructive surgeons can start to prepare recipient vessels in the neck.
The teams then change positions. The reconstructive surgeons now
prepare the jejunal flap while the ablative surgeons expose the proxi-
mal esophagus through the neck incision and remove the manubrium
and clavicular head if a substernal approach is desired. The jejunal
conduit is then passed through the chest and into the neck, and the
teams switch positions again. The reconstructive surgeons revascular-
ize the proximal jejunum, complete the esophagojejunal anastomosis,
and fashion the monitoring segment while the other surgical team
restores gastrointestinal continuity in the abdomen. When properly
organized, no time is wasted for all teams involved. Clear communi-
cation between surgical teams and anesthesia regarding fluid resus-
citation and avoidance of vasopressor medications is imperative as
different disciplines may manage these issues differently.
Technically, 3 key steps are required in such a complex re-
construction: selecting the appropriate jejunal segment, choosing the
optimal recipient vessels for microsurgical anastomosis, and creating
a suitable conduit passageway.
Careful consideration of the mesenteric anatomy is the most
critical component in surgical planning. If possible, the most proxi-
mal jejunum beyond the first mesenteric branch should be included
to avoid wasting of the proximal bowel. Thus, the second mesenteric
branch is often selected for supercharging. However, this selection
also depends on the anatomy of the mesentery because it is the arc
length of the mesentery between branches, not the bowel itself, that
determines the reach of the flap. Therefore, the segment of bowel
with longer mesenteric vessels and wider mesenteric intervals be-
tween the vessels should be selected so the conduit can be adequately
unfurled and lengthened. Transillumination of the mesentery can help
determine the vascular anatomy.
Choice of recipient vessels is another critical factor in the
decision-making algorithm. In general, recipient vessels lower in the
neck (transverse cervical vessels) or in the chest (internal mammary
vessels) are preferred to facilitate reach of the mesenteric pedicle. The
choice of recipient vessels also depends on conduit pathway, which
is dictated by the clinical scenario of each individual patient (eg,
immediate vs delayed reconstruction). A reconstructive algorithm is
created by the senior author (P.Y.) to assist with surgical planning
(Fig. 8).
1128 | www.annalsofsurgery.com
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 253, Number 6, June 2011
For delayed reconstruction, the substernal pathway is chosen
because the retrocardiac route is scarred from previous surgery or
infection. The left manubrium and clavicular head are resected and
the internal mammary vessels are easily exposed, making them the re-
cipient vessel of choice. The vascular anastomosis is then completed
in the upper chest in the space where the manubrium and clavicu-
lar head were removed. The left transverse cervical vessels are an
alternate option.
For immediate reconstruction, the native retrocardiac path-
way is available and preferred to avoid unnecessary resection of the
manubrium and clavicular head. This pathway is also slightly shorter
than the substernal pathway. The internal mammary vessels, which
are located anteriorly, are not a good option with this pathway. The
transverse cervical vessels lower in the neck are preferred to avoid
vein grafting to the external carotid system. The substernal pathway,
however, should be considered during an immediate reconstruction in
the following clinical circumstances: (1) the transverse cervical ves-
sels are unavailable, which may occur in 23% of the unilateral necks22 ;
(2) the patient suffers from severe kyphosis. One contraindication for
choosing the substernal pathway is in patients with previous coronary
artery bypass surgery where the internal mammary artery was utilized
as the bypass graft. Therefore, for immediate reconstruction, we now
prefer exploring the left transverse cervical vessels first. If they are
adequate, the retrocardiac pathway is chosen. If not, then we opt for
the substernal pathway and the use of the internal mammary vessels.
The second option is to proceed with the retrocardiac pathway and
use the external carotid system, such as the superior thyroid artery
and internal jugular vein as recipient vessels (Fig. 8). This option
is less desirable because a vein graft if often necessary, making an
already complex reconstruction even more so.
CONCLUSION
Total esophageal reconstruction using a supercharged jejunal
flap is technically challenging and requires a multidisciplinary ap-
proach. This procedure can be successfully performed as an alterna-
tive reconstruction when the stomach is unavailable with good long-
term function and acceptable morbidity. Thorough surgical planning,
patient optimization, and meticulous postoperative care are critical to
the success of this operation. Selecting the right segment of bowel,
the appropriate recipient vessel, and the ideal conduit pathway are
key steps in this surgical endeavor. An operative algorithm can help
guide the surgeons in the decision-making process.
REFERENCES
1. Orringer MB, Marshall B, Chang AC, et al. Two thousand transhiatal
esophagectomies. Changing trends, lessons learned. Ann Surg. 2007;246:363–
372.
2. Young MM, Deschamps C, Trastek VF, et al. Esophageal reconstruction for
benign disease: early morbidity, mortality and functional results. Ann Thorac
Surg. 2000;70:1651–1655.
3. DeMeester TR, Johansson KE, Franze I, et al. Indications, surgical technique,
and long-term functional results of colon interposition or bypass. Ann Surg.
1988;208:460–474.
4. DeMeester SR. Colon interposition following esophagectomy. Dis Esophagus.
2001;14:169–172.
5. Mansour KA, Bryan FC, Carlson GW. Bowel-interposition for esophageal
replacement: twenty-five year experience. Ann Thorac Surg. 1997;64:752–756.
6. Cense HA, Visser MR, Van Sandick JW, et al. Quality of life after colon
interposition by necessity for esophageal cancer replacement. J Surg Oncol.
2004;88:32–38.
7. Klink CD, Binnebosel M, Schneider M, et al. Operative outcome of colon
interposition in the treatment of esophageal cancer: a 20-year experience.
Surgery. 2010;147:491–496.
8. Chen HC, Tang YB. Microsurgical reconstruction of the esophagus. Semin
Surg Oncol. 2000;19:235–245.

C 2011 Lippincott Williams & Wilkins
Annals of Surgery r Volume 253, Number 6, June 2011
9. Disa JJ, Pusic AL, Mehrara BJ. Reconstruction of the hypopharynx with the
free jejunum transfer. J Surg Oncol. 2006;94:466–470.
10. Reece GP, Schusterman MA, Miller MJ, et al. Morbidity and functional
outcome of free jejunal transfer reconstruction for circumferential defects
of the pharynx and cervical esophagus. Plast Reconstr Surg. 1995;96:
1307–1316.
11. Ferahkose Z, Bedirli A, Kerem M, et al. Comparison of free jejunal graft with
gastric pull-up reconstruction after resection of hypopharyngeal and cervical
esophageal carcinoma. Dis Esophagus. 2008;21:340–345.
12. Disa JJ, Pusic AL, Hidalgo DA, et al. Microvascular reconstruction of the hy-
popharynx: defect classification, treatment algorithm, and functional outcome
based on 165 consecutive cases. Plast Reconstr Surg. 2003;111:652–660.
13. Longmire WP. A modification of the Roux technique for antethoracic
esophageal reconstruction. Surgery. 1947;22:94–1008.
14. Okazaki M, Asato H, Takushima A, et al. Secondary reconstruction of failed
esophageal reconstruction. Ann Plast Surg. 2005;54:530–537.
15. Dowson HMP, Strauss D, Ng R, et al. The acute management and surgical
reconstruction following failed esophagectomy in malignant disease of the
esophagus. Dis Esophagus. 2007;20:135–140.

C 2011 Lippincott Williams & Wilkins
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Suprecharged Jejunal Flap
16. Chana JS, Chen HC, Sharma R, et al. Microsurgical reconstruction of the
esophagus using supercharged pedicled jejunum flaps: special indications and
pitfalls. Plast Reconstr Surg. 2002;110:742–748.
17. Heitmiller RF, Gruber PJ, Swier P, et al. Long-segment substernal jejunal
esophageal replacement with internal mammary vascular augmentation. Dis
Esophagus. 2000;13:240–242.
18. Sekido M, Yamamoto Y, Minakawa H, et al. Use of the “supercharge” technique
in esophageal and pharyngeal reconstruction to augment microvascular blood
flow. Surgery. 2003;134:420–424.
19. Sakuraba M, Kimata Y, Hishinuma S, et al. Importance of additional microvas-
cular anastomosis in esophageal reconstruction after salvage esophagectomy.
Plast Reconstr Surg. 2004;113:1934–1939.
20. Hirabayashi S, Miyata M, Shoji M, et al. Reconstruction of the thoracic esoph-
agus with extended jejunum used as a substitute with the aid of microvascular
anastomosis. Surgery. 1993;113:515–519.
21. Huang JL, Duan ZQ, Yang L, et al. Esophageal reconstruction by jejunal
transfer. Ann Plast Surg. 1999; 42:658–661.
22. Yu P. The transverse cervical vessels as recipient vessels for previously treated
head and neck cancer patients. Plast Reconstr Surg. 2005;115:1253–1258.
www.annalsofsurgery.com | 1129