- Email: [email protected]
The best salvage operation method after total necrosis of a free jejunal graft? Transfer of a second free jejunal graft
Journal of Plastic, Reconstructive & Aesthetic Surgery (2011) 64, 1030e1035
The best salvage operation method after total necrosis of a free jejunal graft? Transfer of a second free jejunal graft Satoshi Onoda a,*, Yoshihiro Kimata a, Kiyoshi Yamada a, Narushi Sugiyama a, Minoru Sakuraba b, Ryuichi Hayashi c a
Department of Plastic and Reconstructive Surgery, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Science, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan b Division of Plastic and Reconstructive Surgery, National Cancer Center Hospital East, 6-5-1 Kashiwanoha, Kashiwa, Chiba 277-8577, Japan c Division of Head and Neck Surgery, National Cancer Center Hospital East, 6-5-1 Kashiwanoha, Kashiwa, Chiba 277-8577, Japan Received 10 August 2010; accepted 4 February 2011
KEYWORDS Free jejunum transfer; Total necrosis; Appropriate salvage operation; Secondary reconstruction
Summary Aim: Transfer of a free jejunal graft is the first choice for reconstruction after total laryngopharyngo-oesophagectomy (TPLE). After total necrosis of a jejunal graft, possible salvage procedures include temporary external fistula formation and transfer of a second free jejunal graft. The present study determines the most appropriate salvage method. Patients and methods: We have transferred over 600 vascularised free jejunal grafts during the past 22 years for reconstruction, immediately after TPLE, either at the National Cancer Center Hospital or at Okayama University Hospital. A second free jejunal graft was transferred to treat the first vascularised free jejunal graft that had undergone total necrosis in five of these patients. We reviewed the total number of operations, the interval between the operation and the start of oral feeding, the outcomes and the follow-up periods of the five patients. Results: Each of the second free jejunal grafts was positioned without complications. All patients resumed postoperative oral food intake after a mean interval of 20.4 days. Four of the five patients remain free of tumour recurrence and in good health. Conclusion: Our results suggest that the best salvage method after total necrosis of an initial free jejunal graft is to transfer a second jejunal graft. Therefore, the severity of contamination of the neck due to jejunal graft necrosis must be minimised at re-operation to transfer a second free jejunal graft using microvascular anastomosis. ª 2011 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ81 086 223 7151; fax: þ81 086 235 7210. E-mail address: [email protected] (S. Onoda). 1748-6815/$ - see front matter ª 2011 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2011.02.005
Transfer of a second free jejunal graft
1031
Table 1
Patients’ characteristics.
Patient
1
2
3
4
5
Age (y)/sex Primary disease
69/M Hypopharyngeal ca. T4N2b 18 Gy Lt.STA/Lt.IJV Venous thrombosis 1 1
64/F Hypopharyngeal ca. T4N0 None Rt.STA/Rt. STA Venous thrombosis 1 1
52/F Hypopharyngeal ca. T3N0 None Lt.SCA/Lt.IJV Venous thrombosis 2 3
64/M Hypopharyngeal ca. Recurrence 70 Gy Rt.STA/Rt.IJV Fistula＋venous thrombosis 7 7
51/F Cervical-esophagus ca. T4N2b 60 Gy Lt.STA/Lt.IJV Arterial thrombosis 10 11
Clinical staging Radiation therapy Recipient vessels Cause of flap necrosis Flap necrosis POD Salvage operation POD
IJV, internal jugular vein; POD, postoperative day; SCA, superficial cervical artery; STA, superior thyroid artery; STV, superior thyroid vein.
Methods of pharyngo-oesophageal reconstruction after total laryngopharyngo-oesophagectomy (TLPE) include gastric pull-up, colon or jejunal interposition1e3 and transfer of a free radial forearm flap4e6 or a free anterolateral thigh flap.7 The most popular method at our institution is transfer of a free jejunal graft.8e10 The advantage of this procedure is a decreased risk of fistula because the pharyngo-oesophagus is physiologically reconstructed with a pipe-shaped section of the bowel and jejunum, and unlike other flaps, a three-point suture is not required. Cervical dead space can be filled with the abundantly vascularised mesentery. Furthermore, because of flap atrophy over the long term, a wide flap (>9 cm) should be harvested, if dermal flaps are used for reconstruction. Thus, skin grafting is often needed to repair the flap donor site. This is a factor in the selection of free jejunum grafts as the flap of first choice for pharyngo-oesophageal reconstruction. A jejunal graft can also be used for reconstruction after partial hypopharyngeal resection.11,12 On the other hand, the disadvantage of free-tissue transfer is the risk of necrosis of the transferred tissue, due to impaired blood supply.13e15 Compared with other types of flaps, jejunal flaps are more prone to blood-flow disorders, such as ischaemia and congestion. Thus, salvaging a thrombotic jejunal graft through microvascular re-anastomosis is difficult, even when a blood-flow disorder is detected via a monitoring flap16e18 or by measuring venous pressure.19 For salvage surgery after total jejunal necrosis, the most common practice is conservative treatment, while waiting for the cervical inflammation and the infectious state to regress after external fistula formation, and then performing a secondary closure operation. However, such secondary closure after external fistula formation delays adjuvant treatment, such as chemotherapy or radiation. These situations adversely affect the overall status of the patient, induce mental anguish, delay postoperative recovery of eating and speech functions and often necessitate prolonged hospitalisation. Therefore, an alternative salvage method should be considered.
Here, we review patients in whom a second free jejunal graft was transferred after total necrosis of an initial graft, and examine outcomes.
Patients and methods Among over 600 patients in whom vascularised free jejunal grafts were used for reconstruction immediately after TPLE either at the National Cancer Center Hospital or at the Okayama University Hospital over the past 22 years, 11 developed total necrosis. A second free jejunal graft was transferred in five (re-jejunum group) of these 11 patients, and an external fistula at the time of re-operation was formed in the other six patients (fistula group). We reviewed the total number of operations, the interval between surgery and the start of oral feeding, the outcome and the follow-up period of the re-jejunum group. The re-jejunum group comprised two men and three women aged 51e69 (mean, 60.0) years at the time of surgery. The primary diseases, including recurrence, were hypopharyngeal cancer in four patients and cervical oesophageal cancer in one. One patient had undergone preoperative chemotherapy and three had received radiation therapy. We performed TPLE and free jejunal transfer for all five of these patients. The superior thyroid and internal jugular were the recipient veins in one and four patients, respectively, and the superior thyroid and transverse cervical arteries were the recipients in three and two patients, respectively. The cause of the jejunal graft necrosis was arterial and venous thrombosis in one and three patients, respectively. One patient had both venous thrombosis and a bowel fistula. A jejunal graft was harvested from each patient in the re-jejunum group with the second jejunal artery and vein as vascular pedicles at the initial operation. The length of the harvested jejunum was w20 cm. In addition, a jejunal graft was harvested using the third jejunal artery and vein as vascular pedicles at the secondary operation. The rejejunum group was treated by debridement of the necrotic initial graft and transfer of a second free jejunal graft
1032 Table 2
S. Onoda et al. Numbers of and course after salvage operations for re-jejunum group.
Patient
Number of re-operations
Type of re-operation
Recipient vessels
Diet start (POD)
Outcome
Follow-up period
1 2 3 4 5
1 1 1 1 1
Re-FJ Re-FJ Re-FJ Re-FJ Re-FJ
Lt.CA/Lt.EJV Lt.STA/Lt.CFV Rt.SCA/Rt.IJV Rt.SCA/Rt.SCV Lt.SCA/Lt.EJV
14 14 24 24 26
Alive Dead Alive Alive Alive
17 years 11 months 1 year 8 months 10 years 10 months 7 years 2 months 5 years 5 months
CA, carotid artery; CFV, common facial vein; EJV, external jugular vein; IJV, internal jugular vein; POD, postoperative day; Re-FJ, second free jejunum transfer; SCA, superficial cervical artery; SCV, superficial cervical vein; STA, superior thyroid artery.
during re-operation within 24 h of the necrosis being detected (Table 1).
for an average of 210 days, and they underwent from two to seven (mean, 3.3) re-operations.
Results
Representative patients
Each of the re-jejunum transfers was completed without complications. The recipient veins at re-operation were the external jugular, the internal jugular, the common facial and the superior cervical vein in two, one, one and one patient, respectively. The recipient arteries were the carotid, the superior thyroid and the superior cervical in one, one and three patients, respectively. We used vascular pedicles that differed from those used in the primary operation for all surgeries. We harvested 25- to 30-cm lengths of jejuna to fill the dead space with richly vascularised mesentery and prevent postoperative infection. Because the interval from primary operation to re-operation was more than 1 week for patients #4 and #5, minor intestinal adhesions were present by the time the second jejunal flaps were harvested; however, the procedures were otherwise uncomplicated. More severe abdominal complications, such as ileus and hernia, did not develop despite jejunal grafts having been harvested at both the first and second operations. None of the patients required another operation after transfer of the second jejunal graft. Postoperative cervical fluoroscopy revealed the absence of leakage or stenosis. All patients resumed oral food intake after a mean postoperative interval of 20.4 days, and were discharged from the hospital within 2 months. Four of the five patients have remained free of tumour recurrence and are in good health. The mean duration of follow-up was 8 years 7 months (range, 1 year 8 monthse17 years 11 months; Table 2). Six patients in the fistula group were treated by graft debridement, and an external fistula was formed from pedicle or local flaps after the initial jejunal flap underwent total necrosis. Either inflammation due to cervical site infection was very severe or no suitable recipient vessels were available for transfer of a second jejunal graft in these patients. Three of them started oral feeding at 65, 84 and 509 days after the initial operation. One each of these three patients underwent pharyngo-oesophageal reconstruction using free jejunum, a free anterolateral thigh flap and a pedicled colon after cervical infection regressed. The other three patients died of local tumour recurrence or overall physical deterioration before their cervical wounds had healed. The fistula group were hospitalised
Patient #1 was a 69-year-old man with T4N2bM0 hypopharyngeal cancer. Ablative surgery was performed with TPLE and bilateral neck dissection (Figure 1(a)). Reconstructive surgery was performed with transfer of a free jejunal flap (Figure 1(b) and (c)). However, emergency surgery was performed because a monitoring flap indicated congestion on postoperative day 1, and total necrosis of the free jejunum was identified during re-operation (Figure 1(d)). After sufficient debridement and lavage, a second jejunal flap was transferred because the cervical recipient site was reasonably clean (Figure 1(e). The postoperative course was uneventful. Fluoroscopy on postoperative day 14 indicated no abnormalities, and oral intake was started on the same day. This patient can ingest food orally without limitation and has remained healthy with no signs of tumour recurrence for 17 years.
Discussion Because free jejunal flaps contain resident bowel flora unlike other dermal flaps, cervical contamination can easily become severe and lead to total flap necrosis. Further, unlike a necrotic intraoral flap, a necrotic jejunal flap transferred to the neck is a life-threatening complication because it is adjacent to great vessels, such as the carotid arteries and the jugular veins that can become infected by necrotic tissue. For this reason, when a flap transferred to the neck undergoes total necrosis because of disordered blood flow, both early detection and an early secondary operation are critical. Reconstruction methods after total necrosis of a free jejunal flap have been described.20,21 At a secondary operation after total necrosis of a free jejunal flap, sufficient debridement of necrotic tissue and appropriate treatment of saliva flowing from the oral cavity are required. Thus, external fistula formation with a local flap or secondary reconstruction of the pharyngo-oesophagus with a free jejunal graft or other flap is required. Oki et al. have suggested that surgeons should become familiar with the indications and limitations of free jejunal transfer and be prepared to select other reconstructive options at salvage surgery.22
Transfer of a second free jejunal graft
1033
Figure 1 a: Laryngopharyngoesophagectomy and bilateral neck dissection were performed. b: A jejunal graft approximately 20 cm long was harvested with the second jejunal artery and vein as vascular pedicles via laparotomy. c: The first jejunal tranfer was performed. Intraoperative findings were not markedly abnormal. d: Findings during re-operation. The transferred jejunal flap had undergone complete necrosis. The necrosis was attributed to venous thrombosis.e: After the necrotic tissue had been removed and the site had been sufficiently washed, a second free jejunal flap was transferred. The carotid artery and external jugular vein were used as recipient vessels.
Five of our patients underwent transfer of a second free jejunal flap because the cervical recipient site was relatively clean after necrotic tissue debridement and blood vessels suitable for microsurgical anastomosis were identified at the time of re-operation. Jejunal graft necrosis was discovered within 48 h of receiving a second jejunal graft in three of the five patients. In addition, jejunal necrosis was discovered in two patients at 1 week after surgery. Keereweer et al. have reported that oral bleeding is the most important sign of jejunal graft necrosis.21 The late jejunal necrosis in our patients was attributed to arterial thrombosis and an oropharyngeal fistula. Necrosis tends to be discovered late under such circumstances because clinical signs, such as oral bleeding, differ little from signs of venous thrombosis. The inflammatory effects at re-operation were local for these two patients, and a second jejunal graft could be
transferred because appropriate recipient vessels were identified. All five patients, who received a second jejunal graft, required only a single re-operation. For this to be possible, bowel anastomosis must be established without leakage at re-operation, and any cervical infection must be no more than mild. When a jejunal graft is used for re-operation, the jejunal mesentery, which is richly vascularised, can be used to fill cervical dead space and cover cervical blood vessels damaged by infection. All patients resumed oral intake within 1 month of the initial operation. When complications are absent, we examine the neck by fluoroscopy 1 week after jejunal transfer. If fluoroscopy of the neck at 2 weeks after a second free jejunal transfer reveals no abnormalities, such as fistulation or stenosis, then oral intake of both food and water is permitted.
1034 We use vascular pedicles that differ from those of the primary operation for all transfers of second jejunal grafts. The vascular pedicle of the free jejunum has sufficient length with the mesentery, and we can select suitable recipient vessels among the various blood vessels in the cervical region. The limited number of suitable recipient vessels at re-operation is a good reason for transferring a second free jejunal graft. Our results suggest that transfer of a second free jejunal graft is the optimal method of salvage after total necrosis of an initial jejunal graft. Therefore, the severity of contamination of the neck due to necrosis of the jejunal graft must be minimised at re-operation to allow microvascular anastomosis. For this reason, cervical abnormalities must be discovered as soon as possible, using various methods of monitoring. Disorders of the transferred jejunum were discovered early in our patients via an exposed section of the jejunal graft that had been used as a monitoring flap, thus allowing early salvage surgery with minimal cervical contamination. The need for reoperation was based on changes in the colour of monitoring flaps, cervical findings, physical signs and imaging findings.
Conclusion We reviewed patients in whom an initial vascularised free jejunal graft underwent total necrosis and necessitated the transfer of a second free jejunal graft. The outcomes suggest that this is an optimal method of salvage after total necrosis of an initial jejunal graft. Therefore, the severity of neck contamination due to necrosis of the jejunal graft must be minimised at re-operation to allow microvascular anastomosis.
Conflict of interest None.
Funding None.
References 1. Maier A, Pinter H, Tomaselli F, et al. Retrosternal pedicled jejunum interposition: an alternative for reconstruction after total esophagogastrectomy. Eur J Cardiothorac Surg 2002;22: 661e5. 2. Cseke L, Horvath OP. Indications, new surgical technique and results of colon interposition or bypass in esophageal surgery. Acta Chir Hung 1997;36:49e50. 3. Davis PA, Law S, Wong J. Colonic interposition after esophagectomy for cancer. Arch Surg 2003;138:303e8. 4. Anthony JP, Singer MI, Mathes SJ. Pharyngoesophageal reconstruction using the tubed free radial forearm flap. Clin Plast Surg 1994;21:137e41.
S. Onoda et al. 5. Scharpf J, Esclamado RM. Reconstruction with radial forearm flaps after ablative surgery for hypopharyngeal cancer. Head Neck 2003;25:261e6. 6. Nakatsuka T, Harii K, Asato H, Ebihara S, et al. Comparative evaluation in pharyngo-oesohageal reconstruction: radial forearm flap compared with jejunal flap. A 10-year experience. Scand J Plast Reconstr Surg Hand Surg 1998;32(3):307e10. 7. Yu P, Lewin JS, Reece GP, et al. Comparison of clinical and functional outcomes and hospital costs following pharyngoesophageal reconstruction with the anterolateral thigh free flap versus the jejunal flap. Plast Reconstr Surg 2006;117(3): 968e74. 8. Seidenberg B, Rosenak SS, Hurwitt ES, et al. Immediate reconstruction of the cervical esophagus by a revascularized isolated jejunal segment. Ann Surg 1959;149(2):162e71. 9. McKee DM, Peters CR. Reconstruction of the hypopharynx and cervical esophagus with microvascular jejunal transplant. Clin Plast Surg 1978;5(2):305e12. 10. Reece GP, Bengtson BP, Schusterman MA, et al. Reconstruction of the pharynx and cervical esophagus using free jejunal transfer. Clin Plast Surg 1994;21(1):125e36. 11. Kadota H, Sakuraba M, Kimata Y, et al. Larynx-preserving esophagectomy and jejunal transfer for cervical esophageal carcinoma. Laryngoscope 2009;119(7):1274e80. 12. Nakatsuka T, Harii K, Ueda K, et al. Preservation of the larynx after resection of a carcinoma of the posterior wall of the hypopharynx: versatility of a free flap patch graft. Head Neck 1997;19(2):137e42. 13. Okazaki M, Asato H, Takushima A, et al. Analysis of salvage treatments following the failure of free flap transfer caused by vascular thrombosis in reconstruction for head and neck cancer. Plast Reconstr Surg 2007;119:1223e32. 14. Suh JD, Sercarz JA, Abemayor E, et al. Analysis of outcome and complications in 400 cases of microvascular head and neck reconstruction. Arch Otolaryngol Head Neck Surg 2004; 130(8):962e6. 15. Pohlenz P, Blessmann M, Heiland M, Blake F, Schmelzle R, Li L. Postoperative complications in 202 cases of microvascular head and neck reconstruction. J Craniomaxillofac Surg 2007; 35(6e7):311e5. 16. Bafitis H, Stallings JO, Ban J. A reliable method for monitoring the microvascular patency of free jejunal transfers in reconstructing the pharynx and cervical esophagus. Plast Reconstr Surg 1989;83(5):896e8. 17. Katsaros J, Banis JC, Acland RD, et al. Monitoring free vascularised jejunum grafts. Br J Plast Surg 1985;38(2):220e2. 18. Disa JJ, Cordeiro PG, Hidalgo DA. Efficacy of conventional monitoring techniques in free tissue transfer: an 11-year experience in 750 consecutive cases. Plast Reconstr Surg 1999;104: 97e101. 19. Sakurai H, Nozaki M, Takeuchi M, et al. Monitoring the changes in intraparenchymatous venous pressure to ascertain flap viability. Plast Reconstr Surg 2007;119(7):2111e7. 20. Bertino G, Benazzo M, Occhini A, et al. Reconstruction of the hypopharynx after free jejunum flap failure: is a second free jejunum transfer feasible? Oral Oncol 2008;44(1):61e4. 21. Keereweer S, Sewnaik A, Kerrebijn J, et al. Salvage or what follows the failure of a free jejunum transfer for reconstruction of the hypopharynx? J Plast Reconstr Aesthet Surg 2010; 63(6):976e80. 22. Oki M, Asato H, Suzuki Y, et al. Salvage reconstruction of the oesophagus: a retrospective study of 15 cases. J Plast Reconstr Aesthet Surg 2010;63(4):589e97.
The best salvage operation method after total necrosis of a free jejunal graft? Transfer of a second free jejunal graft Satoshi Onoda a,*, Yoshihiro Kimata a, Kiyoshi Yamada a, Narushi Sugiyama a, Minoru Sakuraba b, Ryuichi Hayashi c a
Department of Plastic and Reconstructive Surgery, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Science, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan b Division of Plastic and Reconstructive Surgery, National Cancer Center Hospital East, 6-5-1 Kashiwanoha, Kashiwa, Chiba 277-8577, Japan c Division of Head and Neck Surgery, National Cancer Center Hospital East, 6-5-1 Kashiwanoha, Kashiwa, Chiba 277-8577, Japan Received 10 August 2010; accepted 4 February 2011
KEYWORDS Free jejunum transfer; Total necrosis; Appropriate salvage operation; Secondary reconstruction
Summary Aim: Transfer of a free jejunal graft is the first choice for reconstruction after total laryngopharyngo-oesophagectomy (TPLE). After total necrosis of a jejunal graft, possible salvage procedures include temporary external fistula formation and transfer of a second free jejunal graft. The present study determines the most appropriate salvage method. Patients and methods: We have transferred over 600 vascularised free jejunal grafts during the past 22 years for reconstruction, immediately after TPLE, either at the National Cancer Center Hospital or at Okayama University Hospital. A second free jejunal graft was transferred to treat the first vascularised free jejunal graft that had undergone total necrosis in five of these patients. We reviewed the total number of operations, the interval between the operation and the start of oral feeding, the outcomes and the follow-up periods of the five patients. Results: Each of the second free jejunal grafts was positioned without complications. All patients resumed postoperative oral food intake after a mean interval of 20.4 days. Four of the five patients remain free of tumour recurrence and in good health. Conclusion: Our results suggest that the best salvage method after total necrosis of an initial free jejunal graft is to transfer a second jejunal graft. Therefore, the severity of contamination of the neck due to jejunal graft necrosis must be minimised at re-operation to transfer a second free jejunal graft using microvascular anastomosis. ª 2011 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ81 086 223 7151; fax: þ81 086 235 7210. E-mail address: [email protected] (S. Onoda). 1748-6815/$ - see front matter ª 2011 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2011.02.005
Transfer of a second free jejunal graft
1031
Table 1
Patients’ characteristics.
Patient
1
2
3
4
5
Age (y)/sex Primary disease
69/M Hypopharyngeal ca. T4N2b 18 Gy Lt.STA/Lt.IJV Venous thrombosis 1 1
64/F Hypopharyngeal ca. T4N0 None Rt.STA/Rt. STA Venous thrombosis 1 1
52/F Hypopharyngeal ca. T3N0 None Lt.SCA/Lt.IJV Venous thrombosis 2 3
64/M Hypopharyngeal ca. Recurrence 70 Gy Rt.STA/Rt.IJV Fistula＋venous thrombosis 7 7
51/F Cervical-esophagus ca. T4N2b 60 Gy Lt.STA/Lt.IJV Arterial thrombosis 10 11
Clinical staging Radiation therapy Recipient vessels Cause of flap necrosis Flap necrosis POD Salvage operation POD
IJV, internal jugular vein; POD, postoperative day; SCA, superficial cervical artery; STA, superior thyroid artery; STV, superior thyroid vein.
Methods of pharyngo-oesophageal reconstruction after total laryngopharyngo-oesophagectomy (TLPE) include gastric pull-up, colon or jejunal interposition1e3 and transfer of a free radial forearm flap4e6 or a free anterolateral thigh flap.7 The most popular method at our institution is transfer of a free jejunal graft.8e10 The advantage of this procedure is a decreased risk of fistula because the pharyngo-oesophagus is physiologically reconstructed with a pipe-shaped section of the bowel and jejunum, and unlike other flaps, a three-point suture is not required. Cervical dead space can be filled with the abundantly vascularised mesentery. Furthermore, because of flap atrophy over the long term, a wide flap (>9 cm) should be harvested, if dermal flaps are used for reconstruction. Thus, skin grafting is often needed to repair the flap donor site. This is a factor in the selection of free jejunum grafts as the flap of first choice for pharyngo-oesophageal reconstruction. A jejunal graft can also be used for reconstruction after partial hypopharyngeal resection.11,12 On the other hand, the disadvantage of free-tissue transfer is the risk of necrosis of the transferred tissue, due to impaired blood supply.13e15 Compared with other types of flaps, jejunal flaps are more prone to blood-flow disorders, such as ischaemia and congestion. Thus, salvaging a thrombotic jejunal graft through microvascular re-anastomosis is difficult, even when a blood-flow disorder is detected via a monitoring flap16e18 or by measuring venous pressure.19 For salvage surgery after total jejunal necrosis, the most common practice is conservative treatment, while waiting for the cervical inflammation and the infectious state to regress after external fistula formation, and then performing a secondary closure operation. However, such secondary closure after external fistula formation delays adjuvant treatment, such as chemotherapy or radiation. These situations adversely affect the overall status of the patient, induce mental anguish, delay postoperative recovery of eating and speech functions and often necessitate prolonged hospitalisation. Therefore, an alternative salvage method should be considered.
Here, we review patients in whom a second free jejunal graft was transferred after total necrosis of an initial graft, and examine outcomes.
Patients and methods Among over 600 patients in whom vascularised free jejunal grafts were used for reconstruction immediately after TPLE either at the National Cancer Center Hospital or at the Okayama University Hospital over the past 22 years, 11 developed total necrosis. A second free jejunal graft was transferred in five (re-jejunum group) of these 11 patients, and an external fistula at the time of re-operation was formed in the other six patients (fistula group). We reviewed the total number of operations, the interval between surgery and the start of oral feeding, the outcome and the follow-up period of the re-jejunum group. The re-jejunum group comprised two men and three women aged 51e69 (mean, 60.0) years at the time of surgery. The primary diseases, including recurrence, were hypopharyngeal cancer in four patients and cervical oesophageal cancer in one. One patient had undergone preoperative chemotherapy and three had received radiation therapy. We performed TPLE and free jejunal transfer for all five of these patients. The superior thyroid and internal jugular were the recipient veins in one and four patients, respectively, and the superior thyroid and transverse cervical arteries were the recipients in three and two patients, respectively. The cause of the jejunal graft necrosis was arterial and venous thrombosis in one and three patients, respectively. One patient had both venous thrombosis and a bowel fistula. A jejunal graft was harvested from each patient in the re-jejunum group with the second jejunal artery and vein as vascular pedicles at the initial operation. The length of the harvested jejunum was w20 cm. In addition, a jejunal graft was harvested using the third jejunal artery and vein as vascular pedicles at the secondary operation. The rejejunum group was treated by debridement of the necrotic initial graft and transfer of a second free jejunal graft
1032 Table 2
S. Onoda et al. Numbers of and course after salvage operations for re-jejunum group.
Patient
Number of re-operations
Type of re-operation
Recipient vessels
Diet start (POD)
Outcome
Follow-up period
1 2 3 4 5
1 1 1 1 1
Re-FJ Re-FJ Re-FJ Re-FJ Re-FJ
Lt.CA/Lt.EJV Lt.STA/Lt.CFV Rt.SCA/Rt.IJV Rt.SCA/Rt.SCV Lt.SCA/Lt.EJV
14 14 24 24 26
Alive Dead Alive Alive Alive
17 years 11 months 1 year 8 months 10 years 10 months 7 years 2 months 5 years 5 months
CA, carotid artery; CFV, common facial vein; EJV, external jugular vein; IJV, internal jugular vein; POD, postoperative day; Re-FJ, second free jejunum transfer; SCA, superficial cervical artery; SCV, superficial cervical vein; STA, superior thyroid artery.
during re-operation within 24 h of the necrosis being detected (Table 1).
for an average of 210 days, and they underwent from two to seven (mean, 3.3) re-operations.
Results
Representative patients
Each of the re-jejunum transfers was completed without complications. The recipient veins at re-operation were the external jugular, the internal jugular, the common facial and the superior cervical vein in two, one, one and one patient, respectively. The recipient arteries were the carotid, the superior thyroid and the superior cervical in one, one and three patients, respectively. We used vascular pedicles that differed from those used in the primary operation for all surgeries. We harvested 25- to 30-cm lengths of jejuna to fill the dead space with richly vascularised mesentery and prevent postoperative infection. Because the interval from primary operation to re-operation was more than 1 week for patients #4 and #5, minor intestinal adhesions were present by the time the second jejunal flaps were harvested; however, the procedures were otherwise uncomplicated. More severe abdominal complications, such as ileus and hernia, did not develop despite jejunal grafts having been harvested at both the first and second operations. None of the patients required another operation after transfer of the second jejunal graft. Postoperative cervical fluoroscopy revealed the absence of leakage or stenosis. All patients resumed oral food intake after a mean postoperative interval of 20.4 days, and were discharged from the hospital within 2 months. Four of the five patients have remained free of tumour recurrence and are in good health. The mean duration of follow-up was 8 years 7 months (range, 1 year 8 monthse17 years 11 months; Table 2). Six patients in the fistula group were treated by graft debridement, and an external fistula was formed from pedicle or local flaps after the initial jejunal flap underwent total necrosis. Either inflammation due to cervical site infection was very severe or no suitable recipient vessels were available for transfer of a second jejunal graft in these patients. Three of them started oral feeding at 65, 84 and 509 days after the initial operation. One each of these three patients underwent pharyngo-oesophageal reconstruction using free jejunum, a free anterolateral thigh flap and a pedicled colon after cervical infection regressed. The other three patients died of local tumour recurrence or overall physical deterioration before their cervical wounds had healed. The fistula group were hospitalised
Patient #1 was a 69-year-old man with T4N2bM0 hypopharyngeal cancer. Ablative surgery was performed with TPLE and bilateral neck dissection (Figure 1(a)). Reconstructive surgery was performed with transfer of a free jejunal flap (Figure 1(b) and (c)). However, emergency surgery was performed because a monitoring flap indicated congestion on postoperative day 1, and total necrosis of the free jejunum was identified during re-operation (Figure 1(d)). After sufficient debridement and lavage, a second jejunal flap was transferred because the cervical recipient site was reasonably clean (Figure 1(e). The postoperative course was uneventful. Fluoroscopy on postoperative day 14 indicated no abnormalities, and oral intake was started on the same day. This patient can ingest food orally without limitation and has remained healthy with no signs of tumour recurrence for 17 years.
Discussion Because free jejunal flaps contain resident bowel flora unlike other dermal flaps, cervical contamination can easily become severe and lead to total flap necrosis. Further, unlike a necrotic intraoral flap, a necrotic jejunal flap transferred to the neck is a life-threatening complication because it is adjacent to great vessels, such as the carotid arteries and the jugular veins that can become infected by necrotic tissue. For this reason, when a flap transferred to the neck undergoes total necrosis because of disordered blood flow, both early detection and an early secondary operation are critical. Reconstruction methods after total necrosis of a free jejunal flap have been described.20,21 At a secondary operation after total necrosis of a free jejunal flap, sufficient debridement of necrotic tissue and appropriate treatment of saliva flowing from the oral cavity are required. Thus, external fistula formation with a local flap or secondary reconstruction of the pharyngo-oesophagus with a free jejunal graft or other flap is required. Oki et al. have suggested that surgeons should become familiar with the indications and limitations of free jejunal transfer and be prepared to select other reconstructive options at salvage surgery.22
Transfer of a second free jejunal graft
1033
Figure 1 a: Laryngopharyngoesophagectomy and bilateral neck dissection were performed. b: A jejunal graft approximately 20 cm long was harvested with the second jejunal artery and vein as vascular pedicles via laparotomy. c: The first jejunal tranfer was performed. Intraoperative findings were not markedly abnormal. d: Findings during re-operation. The transferred jejunal flap had undergone complete necrosis. The necrosis was attributed to venous thrombosis.e: After the necrotic tissue had been removed and the site had been sufficiently washed, a second free jejunal flap was transferred. The carotid artery and external jugular vein were used as recipient vessels.
Five of our patients underwent transfer of a second free jejunal flap because the cervical recipient site was relatively clean after necrotic tissue debridement and blood vessels suitable for microsurgical anastomosis were identified at the time of re-operation. Jejunal graft necrosis was discovered within 48 h of receiving a second jejunal graft in three of the five patients. In addition, jejunal necrosis was discovered in two patients at 1 week after surgery. Keereweer et al. have reported that oral bleeding is the most important sign of jejunal graft necrosis.21 The late jejunal necrosis in our patients was attributed to arterial thrombosis and an oropharyngeal fistula. Necrosis tends to be discovered late under such circumstances because clinical signs, such as oral bleeding, differ little from signs of venous thrombosis. The inflammatory effects at re-operation were local for these two patients, and a second jejunal graft could be
transferred because appropriate recipient vessels were identified. All five patients, who received a second jejunal graft, required only a single re-operation. For this to be possible, bowel anastomosis must be established without leakage at re-operation, and any cervical infection must be no more than mild. When a jejunal graft is used for re-operation, the jejunal mesentery, which is richly vascularised, can be used to fill cervical dead space and cover cervical blood vessels damaged by infection. All patients resumed oral intake within 1 month of the initial operation. When complications are absent, we examine the neck by fluoroscopy 1 week after jejunal transfer. If fluoroscopy of the neck at 2 weeks after a second free jejunal transfer reveals no abnormalities, such as fistulation or stenosis, then oral intake of both food and water is permitted.
1034 We use vascular pedicles that differ from those of the primary operation for all transfers of second jejunal grafts. The vascular pedicle of the free jejunum has sufficient length with the mesentery, and we can select suitable recipient vessels among the various blood vessels in the cervical region. The limited number of suitable recipient vessels at re-operation is a good reason for transferring a second free jejunal graft. Our results suggest that transfer of a second free jejunal graft is the optimal method of salvage after total necrosis of an initial jejunal graft. Therefore, the severity of contamination of the neck due to necrosis of the jejunal graft must be minimised at re-operation to allow microvascular anastomosis. For this reason, cervical abnormalities must be discovered as soon as possible, using various methods of monitoring. Disorders of the transferred jejunum were discovered early in our patients via an exposed section of the jejunal graft that had been used as a monitoring flap, thus allowing early salvage surgery with minimal cervical contamination. The need for reoperation was based on changes in the colour of monitoring flaps, cervical findings, physical signs and imaging findings.
Conclusion We reviewed patients in whom an initial vascularised free jejunal graft underwent total necrosis and necessitated the transfer of a second free jejunal graft. The outcomes suggest that this is an optimal method of salvage after total necrosis of an initial jejunal graft. Therefore, the severity of neck contamination due to necrosis of the jejunal graft must be minimised at re-operation to allow microvascular anastomosis.
Conflict of interest None.
Funding None.
References 1. Maier A, Pinter H, Tomaselli F, et al. Retrosternal pedicled jejunum interposition: an alternative for reconstruction after total esophagogastrectomy. Eur J Cardiothorac Surg 2002;22: 661e5. 2. Cseke L, Horvath OP. Indications, new surgical technique and results of colon interposition or bypass in esophageal surgery. Acta Chir Hung 1997;36:49e50. 3. Davis PA, Law S, Wong J. Colonic interposition after esophagectomy for cancer. Arch Surg 2003;138:303e8. 4. Anthony JP, Singer MI, Mathes SJ. Pharyngoesophageal reconstruction using the tubed free radial forearm flap. Clin Plast Surg 1994;21:137e41.
S. Onoda et al. 5. Scharpf J, Esclamado RM. Reconstruction with radial forearm flaps after ablative surgery for hypopharyngeal cancer. Head Neck 2003;25:261e6. 6. Nakatsuka T, Harii K, Asato H, Ebihara S, et al. Comparative evaluation in pharyngo-oesohageal reconstruction: radial forearm flap compared with jejunal flap. A 10-year experience. Scand J Plast Reconstr Surg Hand Surg 1998;32(3):307e10. 7. Yu P, Lewin JS, Reece GP, et al. Comparison of clinical and functional outcomes and hospital costs following pharyngoesophageal reconstruction with the anterolateral thigh free flap versus the jejunal flap. Plast Reconstr Surg 2006;117(3): 968e74. 8. Seidenberg B, Rosenak SS, Hurwitt ES, et al. Immediate reconstruction of the cervical esophagus by a revascularized isolated jejunal segment. Ann Surg 1959;149(2):162e71. 9. McKee DM, Peters CR. Reconstruction of the hypopharynx and cervical esophagus with microvascular jejunal transplant. Clin Plast Surg 1978;5(2):305e12. 10. Reece GP, Bengtson BP, Schusterman MA, et al. Reconstruction of the pharynx and cervical esophagus using free jejunal transfer. Clin Plast Surg 1994;21(1):125e36. 11. Kadota H, Sakuraba M, Kimata Y, et al. Larynx-preserving esophagectomy and jejunal transfer for cervical esophageal carcinoma. Laryngoscope 2009;119(7):1274e80. 12. Nakatsuka T, Harii K, Ueda K, et al. Preservation of the larynx after resection of a carcinoma of the posterior wall of the hypopharynx: versatility of a free flap patch graft. Head Neck 1997;19(2):137e42. 13. Okazaki M, Asato H, Takushima A, et al. Analysis of salvage treatments following the failure of free flap transfer caused by vascular thrombosis in reconstruction for head and neck cancer. Plast Reconstr Surg 2007;119:1223e32. 14. Suh JD, Sercarz JA, Abemayor E, et al. Analysis of outcome and complications in 400 cases of microvascular head and neck reconstruction. Arch Otolaryngol Head Neck Surg 2004; 130(8):962e6. 15. Pohlenz P, Blessmann M, Heiland M, Blake F, Schmelzle R, Li L. Postoperative complications in 202 cases of microvascular head and neck reconstruction. J Craniomaxillofac Surg 2007; 35(6e7):311e5. 16. Bafitis H, Stallings JO, Ban J. A reliable method for monitoring the microvascular patency of free jejunal transfers in reconstructing the pharynx and cervical esophagus. Plast Reconstr Surg 1989;83(5):896e8. 17. Katsaros J, Banis JC, Acland RD, et al. Monitoring free vascularised jejunum grafts. Br J Plast Surg 1985;38(2):220e2. 18. Disa JJ, Cordeiro PG, Hidalgo DA. Efficacy of conventional monitoring techniques in free tissue transfer: an 11-year experience in 750 consecutive cases. Plast Reconstr Surg 1999;104: 97e101. 19. Sakurai H, Nozaki M, Takeuchi M, et al. Monitoring the changes in intraparenchymatous venous pressure to ascertain flap viability. Plast Reconstr Surg 2007;119(7):2111e7. 20. Bertino G, Benazzo M, Occhini A, et al. Reconstruction of the hypopharynx after free jejunum flap failure: is a second free jejunum transfer feasible? Oral Oncol 2008;44(1):61e4. 21. Keereweer S, Sewnaik A, Kerrebijn J, et al. Salvage or what follows the failure of a free jejunum transfer for reconstruction of the hypopharynx? J Plast Reconstr Aesthet Surg 2010; 63(6):976e80. 22. Oki M, Asato H, Suzuki Y, et al. Salvage reconstruction of the oesophagus: a retrospective study of 15 cases. J Plast Reconstr Aesthet Surg 2010;63(4):589e97.