2211Ann Thorac Surg CASE REPORT KASHIMURA ET AL2014;98:2211–3 CERVICOTHORACIC ESOPHAGUS RECONSTRUCTION

FEATUREARTIC

LES

Esophageal necrosis after TEVAR may go unrecog-nized, or the diagnosis may be delayed because of its rareoccurrence. Thus, a high level of suspicion for thiscomplication should exist for patients with dysphagia anda previous thoracic aortic repair; however, the diagnosisis generally not considered until the onset of sepsis. In allreports (including the present), the patients were either ina septic state due to delays in diagnosis or had severedebilitation due to underlying premorbid conditions. Inthis setting, 2 of the 3 previously reported patients havebeen deemed unable to tolerate an extensive proceduresuch as esophagectomy [1, 2].

The cause of the observed necrosis may have beenrelated to extrinsic esophageal compression from amediastinal hematoma. Unlike open repair, TEVAR doesnot allow for removal of the associated hematoma.Whether early removal of the hematoma could haveprevented this complication remains speculative, how-ever, as esophageal necrosis has also been describedafter open repair [4]. In addition, this complication mayalso be related to aortic graft infection. In another case,graft resection and extraanatomic bypass wasperformed; however, the outcome was fatal [3]. It hasbeen suggested that graft removal is not alwaysrequired in the treatment of an infection surroundingan aortic graft [5]. The present case also suggeststhat it may be more appropriate to only performesophagectomy for the treatment of esophagealnecrosis. Our patient is still alive 5 months after theesophagectomy; however, long-term follow-up will benecessary to confirm this initial success. The permanentparaplegia of our patient likely resulted from hypo-perfusion of the anterior spinal artery due to the aorticdissection as preoperative magnetic resonance imagingwas already consistent with spinal cord ischemia.Moreover, additional segmental artery sacrifice duringthe TEVAR extension and intraoperative hypoperfusionduring cardiac arrest may have contributed to irrevers-ible spinal cord injury.

Accepted for publication Jan 14, 2014.

Address correspondence to Dr Kashimura, Department of Plastic andReconstructive Surgery, Nihon University School of Medicine, 30-1Ooyaguchikami-cho, Itabashi-ku, Tokyo 173-8610, Japan; e-mail:kasi2106@live.jp.

References

1. Rascanu C, Weis-M€uller BT, F€urst G, Grotemeyer D,Sandmann W. [Esophageal necrosis following endovasculartreatment of a ruptured thoracal aortic aneurysm: caused bymediastinal compartment syndrome]. Chirurg 2009;80:544–8.

2. De Praetere H, Lerut P, Johan M, et al. Esophageal necrosisafter endoprosthesis for ruptured thoracoabdominal aneu-rysm type I: can long-segment stent grafting of the thor-acoabdominal aorta induce transmural necrosis? Ann VascSurg 2010;24:1137.e7–e12.

3. Porcu P, Chavanon O, Sessa C, Thony F, Aubert A, Blin D.Esophageal fistula after endovascular treatment in a type Baortic dissection of the descending thoracic aorta. J Vasc Surg2005;41:708–11.

4. Kaneda T, Onoe M, Asai T, Mohri Y, Saga T. Delayedesophageal necrosis and perforation secondary to thoracicaortic rupture: a case report and review of the literature.Thorac Cardiovasc Surg 2005;53:380–2.

5. Kaneda T, Iemura J, Oka H, et al. Treatment of deep infectionfollowing thoracic aorta graft replacement without graftremoval. Ann Vasc Surg 2001;15:430–4.

� 2014 by The Society of Thoracic SurgeonsPublished by Elsevier

Successful Management of the

Cervicothoracic EsophagusReconstruction by ExpandedSkin FlapTsutomu Kashimura, MD, Hiroaki Nakazawa, MD,Katsumi Shimoda, MD, Kazutaka Soejima, MD,Mitsugu Kochi, MD, and Tadatoshi Takayama, MD

Departments of Plastic and Reconstructive Surgery, andDigestive Surgery, Nihon University School of Medicine,Itabashi-ku, Tokyo, Japan

The limited availability of reconstruction materials canoften make it difficult to treat defects in the esophaguscaused by necrosis of the transplanted intestinal tissueafter cervicothoracic esophagus reconstruction. We wereforced to perform flap reconstruction on a patient whosuffered necrosis due to impeded blood flow of thetransplanted intestinal tract after twice conducting cervi-cothoracic esophagus reconstruction with an intestinaltract flap. The procedure we performed was esophagusreconstruction using a pectoralis major myocutaneousflap that had been expanded with a tissue expanderdue to the small volume of tissue available to performthe reconstruction. This case suggested that esophagusreconstruction with a skin flap using a tissue expandershould be considered as a possible treatment choicewhen performing reconstruction of the cervicothoracicesophagus, which requires stable blood flow and a largeamount of tissue.

(Ann Thorac Surg 2014;98:2211–3)� 2014 by The Society of Thoracic Surgeons

hen reconstruction using the alimentary tract is

Wimpossible, esophagus reconstruction must beperformed using a skin flap. We treated a patient whosuffered necrosis of the transplanted intestinal tissue after2 procedures for reconstruction of the cervicothoracicesophagus with an intestinal patch and were able toachieve a favorable outcome by using an expanded skinflap with limited materials for reconstruction available.

A 67-year-old male had undergone right transthoracicsubtotal esophagectomy for cervicothoracic esophagealcarcinoma. The patient had a history of pyloric sidegastric resection for stomach cancer so reconstruction wasperformed with the right colon in front of the sternum.On the 14th postoperative day, the reconstructed colondeveloped necrosis and debridement was performed.Two months after the initial surgical procedure, esoph-agus reconstruction was performed with free jejunaltransfer. Intraoperatively, thrombus formation meant that

0003-4975/$36.00http://dx.doi.org/10.1016/j.athoracsur.2014.01.073

Fig 1. (A) After debridement of the transferred jejunum (top arrow ¼esophageal stoma; bottom arrow ¼ colon stoma.) (B) Insertion of thetissue expander (arrow ¼ tissue expander. (C) Esophagus recon-struction with PMMC flap (arrow ¼ expanded skin.) (D) Computedtomographic findings after expansion (arrow ¼ tissue expander.)

Fig 2. Postreconstruction appearance (top arrow ¼ pectoralis majormyocutaneous; bottom arrow ¼ deltopectoral flap.)

2212 CASE REPORT KASHIMURA ET AL Ann Thorac SurgCERVICOTHORACIC ESOPHAGUS RECONSTRUCTION 2014;98:2211–3

FEATUREARTIC

LES

left and right internal thoracic arterieswere used to performmultiple vascular anastomosis procedures. However, asthrombus formation then led to necrosis of the transferredjejunum, debridement was performed and stomas wereopened in the esophagus stoma and colon (Fig 1A).

We decided to perform esophageal reconstruction with askin flap. First, we conducted a split thickness skin graft onthe skin ulcer in the sternum region. A tissue expanderwas placed below the left pectoralis major muscle to ac-quire a large volume of tissue with long vascular pedicleand stable blood flow (Fig 1B). Seventy days afterplacement of the tissue expander 850 mL saline wasinflated, we created a rolled PMMC flap, and thenreversed and sutured the skin around the esophagus andcolon stomas (Figs 1C, 1D). Split thickness skin graft wasperformed on the surface of the flap. Postoperatively, theanal side suture area gradually separated open to createa tissue defect area. We covered the area with a rightdeltopectoral skin flap. As small fistulae then appeared,wound closure was performed twice.

Contrast imaging taken 1 month after the final surgicalprocedure revealed an obstruction in the colon stump.Endoscopy indicated a blockage but no stenosis in theanastomosis region. Therefore, we placed an esophageal

stent in the colon stoma region. The obstructionimproved and, 502 days after the initial surgical proce-dure, the patient was able to orally ingest food (Fig 2).Currently, 2 years and 3 months since the final surgicalprocedure, the patient is able to orally ingest food.

Comment

The first choice for reconstruction of the esophagusafter resection due to esophageal cancer is reconstruc-tion using the alimentary canal. Reconstruction usingstomach, small intestine, and colon tissue is performedfor cervicothoracic esophagus defects [1–3]. Recon-struction with a skin flap is known to lead to moreinstances of fistula formation and stenosis than recon-struction using tissue from the alimentary canal [4].Therefore, reconstruction of the esophagus using askin flap is usually only selected when reconstructioncannot be performed using the alimentary canal [5]. Inthe present case, after multiple abdominal surgicalprocedures, reconstruction using the intestinal tractwas judged to be too difficult, leading us to selectreconstruction using a skin flap.There are few reports in the literature on reconstruction

of the esophagus with a skin flap for cervicothoracicesophagus defects. Reports that have dealt with this topic

Fig 3. Available vascular pedicle for cervico-thoracic esophagusreconstruction. (ITA ¼ internal thoracic artery; SEA ¼ superiorepigastric artery; SG ¼ skin graft; TAA ¼ thoracoacromial artery.)

2213Ann Thorac Surg CASE REPORT KASHIMURA ET AL2014;98:2211–3 CERVICOTHORACIC ESOPHAGUS RECONSTRUCTION

FEATUREARTIC

LES

have performed reconstruction using thoracic skin flapsand abdominal rectus muscle flaps [6]. When recipientblood vessels can be secured, reconstruction using ajejunal flap may be considered. However, in the presentcase, thrombi occurred in the internal thoracic arterieson both sides, making it difficult to secure a recipientblood vessel. Therefore, we chose to performreconstruction using a pedicle flap. When thoracic skinflaps are used, the internal thoracic artery perforatingbranch is separated to create the reconstruction path infront of the sternum, leading to unstable blood flow.Therefore, a thoracic skin flap was considered notsuitable for the present case. An abdominal rectusmuscle flap was also considered unsuitable becauseblood flow of the superior epigastric artery of theinternal thoracic artery bifurcation was superior (Fig 3).

The pectoralis major myocutaneous (PMMC) flapsuse the thoracoacromial artery as a vascular pedicleand are therefore the number 1 choice for cervicalesophagus reconstruction [7]. However, the lengthof the vascular pedicle and size of the skin flapare insufficient for cervicothoracic esophagusreconstruction. In cases such as these, reconstruction

can be performed after using a tissue expander toexpand the tissue. This device involves inserting asilicone balloon subcutaneously and injecting salinesolution over time. Using a tissue expander requires 2surgical procedures and time is required for theexpansion to be achieved. However, tissue expanderscan be used to not only achieve expansion of graftskin and soft tissue, they can also promote vascularpedicle elongation and increased blood flow in theexpanded tissue [8]. There are no case reports in theliterature of cervicothoracic esophageal reconstructionwith an expanded skin flap. We were able tosuccessfully perform reconstruction of thecervicothoracic esophagus with a PMMC flap forwhich a tissue expander had been used to achievesufficient tissue volume and vascular pedicle length.Stoma surgery for the alimentary canal usually in-

volves moving objects such as an artificial anus out frominside the alimentary canal. However, the colon stoma inthe present case required that an object be moved fromthe postreconstruction skin side into the alimentary ca-nal. It is likely that during this process the intestinalmucosa acted like a check valve and caused theobstruction. It therefore appears that a sufficient amountof lumen needs to be secured for stomas that requireobjects to be moved into the alimentary canal side. Anesophageal stent proved extremely effective for thepresent case, making it possible for the patient to ingestfood.We believe that tissue expanders are a useful

option when performing cervicothoracic esophagusreconstruction with a skin flap after intestinal necrosiswhen treatment is difficult due to a limited amount ofreconstruction materials.

References

1. Pesko P, Sabljak P, Bjelovic M, et al. Surgical treatment andclinical course of patients with hypopharyngeal carcinoma.Dis Esophagus 2006;19:248–53.

2. Sekido M, Yamamoto Y, Minakawa H, et al. Use of the “su-percharge” technique in esophageal and pharyngeal recon-struction to augment microvascular blood flow. Surgery2003;134:420–4.

3. Davis PA, Law S, Wong J. Colonic interposition afteresophagectomy for cancer. Arch Surg 2003;138:303–8.

4. Murray DJ, Novak CB, Neligan PC. Fasciocutaneous free flapsin pharyngolaryngo-oesophageal reconstruction: a critical re-view of the literature. J Plast Reconstr Aesthet Surg 2008;61:1148–56.

5. Oki M, Asato H, Suzuki Y, et al. Salvage reconstruction of theoesophagus: a retrospective study of 15 cases. J Plast ReconstrAesthet Surg 2010;63:589–97.

6. Hamai Y, Hihara J, Emi M, Tanabe K, Miyamoto Y,Okada M. Skin tube reconstruction for esophageal defectdue to postoperative complication. Ann Thorac Surg2009;87:1605–7.

7. Ariyan S. The pectoralis major myocutaneous flap. A versatileflap for reconstruction in the head and neck. Plast ReconstrSurg 1979;63:73–81.

8. Manders EK, Schenden MJ, Furrey JA, Hetzler PT, Davis TS,Graham WP III. Soft-tissue expansion: concepts and compli-cations. Plast Reconstr Surg 1984;74:493–507.