Analysis of pharyngocutaneous fistula following free jejunal transfer for total laryngopharyngectomy

The development of a pharyngocutaneous fistula is the most common and troublesome complication in the early postoperative period following free jejunal transfer for total laryngopharyngectomy. However, many aspects of this complication remain unclear. In this study, the authors analyzed their experience with the pharyngocutaneous fistula formation following free jejunal transfers to evaluate its clinical behavior, determine the significance of the anastomotic technique used, and evaluate the role of preoperative radiation therapy on its formation and management. Of 168 patients who underwent free jejunal transfers following total laryngopharyngectomy at the authors' institution between July of 1988 and March of 2000, 23 patients (13.7 percent) with postoperative fistulas were identified. The mean onset of fistula formation was 16 days. Of the 23 fistulas, 13 (56.5 percent) occurred at the proximal and 10 (43.5 percent) at the distal anastomoses. Whereas the majority of the proximal fistulas (69.2 percent) developed near the mesenteric side of the jejunal flap, most of the distal fistulas (90 percent) were located anteriorly. The incidence of proximal fistula formation was higher in patients with a single-layer repair than in patients with a two-layer repair of a proximal anastomosis (80 percent versus 38.5 percent, p = 0.09). The incidence of fistula formation was greater in patients who received preoperative radiation therapy than in those who did not (16.3 percent versus 11.4 percent, p = 0.36). In addition, whereas a majority of fistulas (80 percent) occurred at the proximal anastomosis in patients who did not receive preoperative radiation therapy, most fistulas (61.5 percent) occurred at the distal anastomosis in patients who did receive radiation therapy (p = 0.09). The fistulas closed spontaneously in 15 patients (65 percent). On average, spontaneous closure occurred in 7.4 weeks. Proximal fistulas had a significantly higher rate of spontaneous closure compared with distal fistulas (85 percent versus 40 percent, p = 0.04). The rate of spontaneous fistula closure was higher in patients who had not received preoperative radiation therapy than in those who had (90 percent versus 46 percent, p = 0.07). Surgical closure of the fistula was required in five patients. The fistulas were not repaired in three patients because of recurrent tumor. Twenty patients (87 percent) resumed oral feeding after the closure of the fistula, with 17 (85 percent) of 20 patients tolerating a regular diet and three (15 percent) of 20 a liquid diet only.In conclusion, most fistulas occur at the proximal anastomosis and near the mesenteric side of the jejunal flap, and the use of a two-layer anastomotic technique seems to be associated with a lower incidence of fistula formation at the proximal suture line. Most fistulas close spontaneously, especially ones that occur proximally. Preoperative radiotherapy does seem to increase the risk of fistula formation, especially at the distal anastomotic site and make subsequent resolution of the fistulas more difficult. Most patients are able to resume oral feeding once the fistula is closed.     

Repair of pharyngocutaneous fistulas with the submental artery island flap

Pharyngocutaneous fistulas after total laryngectomy are difficult to manage and are a cause for significant morbidity to the patient. When fistulas fail to close with conservative measures, debridement and flap closure are indicated. Although a number of techniques to repair pharyngocutaneous fistulas are described, each of these procedures has its drawbacks. The authors have used the submental island flap to close postoperative pharyngocutaneous fistulas in nine male patients during the past 4 years. The mean patient age was 65 years (range, 57 to 75 years). The submental island flap is based on the submental artery, a branch of the facial artery. The inner aspect of the fistula was initially formed using hinge flaps on the skin around the fistula. Once a watertight closure of inner side was created, the skin defect was closed with the submental island flap. The maximum flap size was 6 x 3 cm and the minimum size was 4 x 2 cm (average, 4.8 x 2.7 cm) in this series. Direct closure was achieved at all donor sites. Patients were followed for 6 months to 4 years. No major complication was noted in the postoperative period. All patients have successfully recovered their swallowing function. The submental island flap is safe, rapid, and simple to elevate and leaves minimal donor-site morbidity. The authors believe that this technique is a good alternative in the reconstruction of pharyngocutaneous fistulas. Application of the technique and results are discussed.     

Cleft palate fistulas: a multivariate statistical analysis of prevalence, etiology, and surgical management

A retrospective, multivariate statistical analysis of 129 consecutive nonsyndromic patients undergoing cleft palate repair was performed to document the incidence of postoperative fistulas, to determine their cause, and to review methods of surgical management. Nasal-alveolar fistulas and/or anterior palatal fistulas that were intentionally not repaired were excluded from study. Cleft palate fistulas (CPFs) occurred in 30 of 129 patients (23 percent), although nearly a half were 1 to 2 mm in size. Extent of clefting, as estimated by the Veau classification, was significantly more severe in those patients who developed cleft palate fistula. Type of palate closure also influenced the frequency of cleft palate fistula. Forty-three percent of patients undergoing Wardill-type closures developed cleft palate fistula versus 10, 22, and 0 percent for Furlow, von Langenbeck, and Dorrance style closures, respectively. The fistula rate was similar in patients with (30 percent) and without (25 percent) intravelar veloplasty. Age at palate closure did not significantly affect the rate of fistulization; however, the surgeon performing the initial closure did not have an effect. Thirty-seven percent of patients developed recurrent cleft palate fistulas following initial fistula repair. Recurrence of cleft palate fistulas was not influenced by severity of cleft or type of original palate repair. Following end-stage management, a second cleft palate fistula recurrence occurred in 25 percent of patients. Continued open discussion of results of cleft palate repair is recommended.     

Management and timing of cleft palate fistula repair

This study reviewed 199 cleft palate repairs resulting in 22 percent fistula formation. Of these, 49 percent were judged to be symptomatic. Of 44 fistulas, 21 required treatment, of which 14 had conventional type surgical closure with an overall success rate of 35 percent. Good surgical technique and good surgical judgment were felt to be important factors both in preventing postoperative fistula and in the success of their repair. Conventional methods of surgical repair of hard palate fistulas were seen to result in a very poor success rate. Orthodontic movement of maxillary segments was seen to contribute to late postoperative fistula formation. Therefore, orthodontic movement should be completed before undertaking surgical repair of anterior palatal fistulas. Finally, the success rate of anterior fistula repair has been dramatically improved by the addition of free periosteal grafts and cancellous bone grafts.     

Have we found an ideal soft-tissue flap? An experience with 672 anterolateral thigh flaps

The free anterolateral thigh flap is becoming one of the most preferred options for soft-tissue reconstruction. Between June of 1996 and August of 2000, 672 anterolateral thigh flaps were used in 660 patients at Chang Gung Memorial Hospital. Four hundred eighty-four anterolateral thigh flaps were used for head and neck region recontruction in 475 patients, 58 flaps were used for upper extremity reconstruction in 58 patients, 121 flaps were used for lower extremity reconstruction in 119 patients, and nine flaps were used for trunk reconstruction in nine patients. Of the 672 flaps used in total, a majority (439) were musculocutaneous perforator flaps. Sixty-five were septocutaneous vessel flaps. Of these 504 flaps, 350 were fasciocutaneous and 154 were cutaneous flaps. Of the remaining 168 flaps, 95 were musculocutaneous flaps, 63 were chimeric flaps, and the remaining ten were composite musculocutaneous perforator flaps with the tensor fasciae latae. Total flap failure occurred in 12 patients (1.79 percent of the flaps) and partial failure occurred in 17 patients (2.53 percent of the flaps). Of the 12 flaps that failed completely, five were reconstructed with second anterolateral thigh flaps, four with pedicled flaps, one with a free radial forearm flap, one with skin grafting, and one with primary closure. Of the 17 flaps that failed partially, three were reconstructed with anterolateral thigh flaps, one with a free radial forearm flap, five with pedicled flaps, and eight with primary suture, skin grafting, and conservative methods.In this large series, a consistent anatomy of the main pedicle of the anterolateral thigh flap was observed. In cutaneous and fasciocutaneous flaps, the skin vessels (musculocutaneous perforators or septocutaneous vessels) were found and followed until they reached the main pedicle, regardless of the anatomic position. There were only six cases in this series in which no skin vessels were identified during the harvesting of cutaneous or fasciocutaneous anterolateral thigh flaps. In 87.1 percent of the cutaneous or fasciocutaneous flaps, the skin vessels were found to be musculocutaneous perforators; in 12.9 percent, they were found as septocutaneous vessels. The anterolateral thigh flap is a reliable flap that supplies a large area of skin. This flap can be harvested irrespective of whether the skin vessels are septocutaneous or musculocutaneous. It is a versatile soft-tissue flap in which thickness and volume can be adjusted for the extent of the defect, and it can replace most soft-tissue free flaps in most clinical situations.     

Oronasal fistula repair with three layers

We present an innovative method for closure of oronasal fistulas involving a three-layer repair, consisting of septal mucosa flap, bone or cartilage graft, and palatal mucosa flap. The septal mucosa flap closes the nasal side of the defect. This is an inferiorly based flap along the nasal floor and consists of septal mucosa from the side opposite the oronasal fistula. A slit is created in the remaining layers of the nasal septum, allowing the flap to be delivered into the defect. When the septal flap is folded down in this fashion, it exposes nasal septal bone and cartilage. The bone and cartilage are harvested and are used to create the middle layer of the three-layer fistula repair. The oral layer of the repair is provided by a palatal mucosa transposition flap. This method allows the bone/cartilage graft to be sandwiched between two vascular layers. We have successfully used the three-layer repair on three patients. All of the oronasal defects were 2 cm in size. All patients are at least 1 year after repair with 100 percent closure; thus, no oronasal leakage. The flaps both septal and palatal resulted in no morbidity once healed. Specifically, the surgically created slit in the nasal septum is well mucosalized and barely discernible. Also, no nasal obstruction occurs from the septal flap on the floor of the nose. We perform the procedure on an outpatient basis. The three-layer repair can be used in adult patients with oronasal fistulas of the middle and posterior hard palate up to 3 cm in size. This technique is not recommended for children.     

Incidence of cleft palate fistula: an institutional experience with two-stage palatal repair.

The purpose of this study was to determine the incidence of cleft palatal fistula in a series of nonsyndromic children treated at the authors' institution. This retrospective analysis of 103 patients with cleft palate treated by five surgeons between 1982 and 1995 includes 60 boys and 33 girls, whose median age was 18.4 months at the time of surgery. The median length of follow-up was 4.9 years after primary palatoplasty. Cleft palatal fistula was defined as a failure of healing or a breakdown in the primary surgical repair of the palate. Intentionally unrepaired fistulas of the primary and secondary palate were excluded. Extent of clefting was described according to the Veau classification. Statistical examination of multiple variables was performed using contingency table analysis, multivariate logistic regression, and the Wilcoxon rank sum test. The incidence of cleft palatal fistula in this series was 8.7 percent. All of these fistulas were clinically significant. The rate of fistula recurrence was 33 percent. The incidence of cleft palatal fistula when compared by Veau classification was statistically significant, with nine fistulas occurring in patients with Veau 3 and 4 clefts and no fistulas occurring in patients with Veau 1 and 2 clefts (p = 0.0441). No significant differences between patients with and without fistulas were identified with respect to operating surgeon, patient sex, patient age at palatoplasty, type of palatoplasty, and use of presurgical orthopedics or palatal expansion. All three recurrent fistulas occurred in the anterior palate, two in patients with Veau class 3 clefts and one in a patient with a Veau class 4 cleft. The low rate of clinically significant fistula was attributed to early delayed primary closure, with smaller secondary clefts allowing repair with a minimum of dissection and disruption of vascularity.     

Management of recalcitrant bronchopleural fistulas with muscle flap obliteration

Use of muscle flap obliteration for bronchopleural fistulas appears to be indicated with (1) failure of a previous thoracoplasty, (2) anticipated failure of a thoracoplasty alone, and (3) the need to obviate a formal debilitating thoracoplasty. With the use of well-vascularized muscle flaps to fully obliterate the densely scarred cavities associated with persistent bronchopleural fistulas, we may hope to see improved healing in the bronchial stump and, in cases of residual infection, better resistance of the flap to necrosis, as well as improved delivery of chemotherapeutic agents to the local tissues. These factors may confer improved cure rates for bronchopleural fistulas similar to those seen in lower extremity salvage surgery for osteomyelitis following the introduction of vascularized pedicle and free muscle flaps. In this article we have described the versatility of the island pedicle latissimus dorsi muscle flap for closure of recalcitrant bronchopleural fistulas and associated empyema cavities. Utilizing either the dominant thoracodorsal or the minor paraspinal pedicle(s), it can reach any intrathoracic cavity by means of the appropriate thoracotomy incision.     

Use of the vascular closure staple clip applier for microvascular anastomosis in free-flap surgery

We report our initial experience using the vascular closure staple clip applier (a nonpenetrating titanium clip applied in an interrupted, everting fashion) for microvascular anastomosis in free-flap surgery. In total, 153 anastomoses were performed in 87 free flaps (174 potential anastomoses) using the vascular closure stapler between October of 1997 and June of 1999. In 66 flaps, both the arterial and venous anastomosis were performed with the clip applier, whereas in 21 flaps only the venous anastomosis was performed using the clips. A total of 146 anastomoses were performed in an end-to-end fashion, and seven were performed end-to-side. Of the 87 flaps there were 53 TRAM flaps, seven bilateral TRAM, five latissimus dorsi, four gastrocnemius, three rectus abdominis, two radial forearm fibula, and four Rubens fat-pad flaps. Seventy flaps were used for breast reconstruction, seven flaps for lower limb reconstruction, four flaps for head and neck reconstruction, and six flaps for chest wall/trunk reconstruction. There were no postoperative anastomotic complications of bleeding, thrombosis, or need for revision (100 percent patency rate), with a significantly reduced time for completion of anastomoses. The clip applier is a safe, reliable method for performing microvascular anastomoses, allowing reduced operating time and possible cost savings in free-flap surgery.     

One-stage closure of the entire primary palate

Timing of the closure of the anterior palate and alveolus is a subject of debate. Late repair of this defect is complicated by high fistula formation and subjects the patient to the problems of palate fistula for extended periods of time. We have utilized a single procedure performed when the child is 3 months of age that completely closes the anterior hard palate and alveolus along with the cleft lip. Our series consisted of 61 consecutive patients with unilateral clefts of the primary and secondary palate. Mucosal turnover flaps from the vomer along with lateral nasal mucosal flaps provide the nasal lining. A buccal sulcus flap with a Veau flap completes the oral repair. Ninety-five percent (58 of 61) of the patients had complete and stable closure of their anterior palate and alveolus after 1 year. The incidence of fistula formation in our series (3 of 61) is much lower than that reported with the utilization of other protocols. Excellent exposure of the anterior palate and alveolar defect during lip repair, early restoration of anatomic relationships, establishment of a good nostril floor and sill, and very low fistula formation are among the benefits of this procedure. The increase in operative time is considered minimal in light of aforementioned advantages.     

Bipedicle paraspinous muscle flaps for spinal wound closure: an anatomic and clinical study

The purpose of this study was to evaluate the vascular anatomy of the paraspinous muscles and review their clinical use as bipedicled flaps in spinal wound closure. Anatomically, through cadaver dissections, lead oxide injections, and radiographic imaging, the blood supply to the paraspinous muscles was determined. Clinically, 29 consecutive patients treated with spinal wounds and exposed bone or hardware were reviewed retrospectively. Of these patients, 19 underwent closure in delayed primary fashion, whereas 10 were referred to plastic surgery for reconstruction because of the complex nature of their wounds. The cadaver study demonstrated the paraspinous muscles to possess a segmental arterial supply through medial and lateral perforators. Division of the medial perforators allowed for medial advancement of the muscles. Lead oxide injection of the lateral perforators demonstrated adequate medial muscle perfusion with ligation of the medial perforators. Ten of the 29 patients (six women, four men, 32 to 62 years of age) were reconstructed with paraspinous (eight), latissimus (one), and trapezius (one) muscle flaps. A higher complication rate was found in wounds closed in delayed primary fashion (13 of 19 patients, 68 percent) than those reconstructed with muscle flaps (2 of 10 patients, 20 percent) (p = 0.021). Follow-up of the muscle flap reconstructed patients averaged 12 months (range, 3 to 27 months). Cadaver muscle injections predicted and clinical cases confirmed that the paraspinous muscles can be raised on lateral perforators and advanced medially to close lumbar spine wounds reliably with fewer complications.     

Microsurgical reconstruction of the esophagus using supercharged pedicled jejunum flaps: special indications and pitfalls

This report outlines a microsurgical technique for total esophageal reconstruction in situations in which conventional methods using stomach or colon are not available. Eleven patients with corrosive injury and one patient following tumor resection underwent total esophageal reconstruction in a two-stage procedure. In the first stage, skin flaps or free jejunal transfers were used for the cervical reconstruction. In the second stage, supercharged pedicled jejunum flaps placed subcutaneously were used for thoracic esophageal replacement. The study included one male and 10 female patients, with a mean age of 38.4 years. The mean follow-up period was 78.9 months. All patients had one or more complications that required revisional surgery. Pedicled myocutaneous flaps were used to close fistulas or chronic wounds in four patients. The cervical skin tube in two patients and the jejunum in another two patients required shortening because of redundancy. Four patients had dysphagia caused by neck contractures, which were released. Two patients developed pharyngoesophageal strictures that required further free skin flaps for release. Two patients had reflux because of blind pouches arising from the original esophagus and required thoracotomy for removal. At long-term follow-up, all patients are fully rehabilitated and have resumed an oral diet with significant weight gain. Compared with lifelong jejunostomy feeding and its associated psychosocial disadvantages, the authors' experience demonstrates that the application of microsurgical techniques to fully reconstruct the esophagus is of considerable benefit to this difficult patient group.     

Cigarette smoking and face lift: conservative versus wide undermining

The effects of cigarette smoking on the skin flaps of the face lift procedure are discussed. Reported elsewhere is a significant incidence of skin slough in smokers with use of wide undermining techniques. This complication is thought to be due to the vasoconstrictive effects of nicotine on the peripheral circulation. Our group has employed a conservative bilateral undermining technique in 407 face lifts. Of these, 32.4 percent were smokers and 67.6 percent were nonsmokers. No cases of skin slough were encountered. Our conservative undermining technique is briefly discussed. Among its advantages are shorter operative time, use of less local and/or general anesthesia, less intraoperative bleeding, adequate exposure for SMAS and platysmal surgery, and snugger skin closure without the risk of flap necrosis. As shown by our statistics, it is a safer procedure in smokers than the usually performed more radical procedure.     

Influence of palatal fistulas on velopharyngeal movements: a cineradiographic study

Ten patients with hard palate fistulas and velopharyngeal incompetence were examined cineradiographically with lateral and frontal projections during connected speech. The purpose was to study velopharyngeal function for open versus covered fistulas. Fistula length, width, and size were measured. With an open fistula, the degree of lateral pharyngeal wall activity was significantly correlated at the 0.05 level to fistula size. There was no statistically significant correlation between fistula size and the degree of velar activity. With the fistula covered, the velopharyngeal movements improved or even normalized in all patients regardless of fistula size. These findings indicate that patients with both velopharyngeal incompetence and an additional fistula will need covering of the fistula. If velopharyngeal incompetence persists after a temporary covering of the fistula, a combination of velopharyngeal flap surgery and fistula covering ought to be performed to normalize the velopharyngeal activity and speech and resonance.     

Radiation diagnosis of aortic aneurysmal ruptures with formation of anastomoses with the inferior cava and various organs

The results of radiation diagnostic techniques were analyzed in 29 patients with aortic aneurysmal ruptures with formation of aortic anastomoses. The examination and treatment of 362 patients with abdominal aortic aneurysmal ruptures revealed that 23 (6.35%) patients had anastomoses (aortocaval (n = 15), aortoduodenal (n = 6), aortogastric (n = 1), and aortoureteral (n = 1)). Six patients were observed to have secondary aortointestinal anastomoses occurring after reconstructive vascular surgery. In one patient, abdominal aortic aneurysmal rupture first led to the formation of a primary aortoduodenal anastomosis and some time after surgery a secondary aortointestinal fistula emerged. One out of 113 patients with aortic dissecting aortic aneurysms had an aortopulmonary fistula. Ultrasonography, computed tomography, and angiography were performed in 27, 19, and 14 patients, respectively. Radiation diagnostic techniques revealed fistulas in 9 (31%) out of the 29 patients. The diagnosis of aortic anastomoses presents challenges.     

Jejunal free autograft: analysis of complications and their resolution

Review of 101 patients who underwent 111 free jejunal autografts has demonstrated an absolute procedural failure rate of 13.5 percent. Salvage reconstruction with a second jejunum was successful in six of nine patients and one third-time jejunum was successful, giving an overall salvage rate of 70 percent. There were 33 patients experiencing pharyngocutaneous fistulas, 20 of whom had been previously irradiated. Of these patients, 15 experienced spontaneous closure and 9 others had successful surgical correction. The mortality rate was 5 percent. Eighty-three percent of patients were restored to adequate per oral alimentation. The jejunum, despite its relatively high complication rate, is an excellent method for pharyngoesophageal reconstruction, expeditiously providing return to function for patients with late-stage disease.     

Limb salvage for soft-tissue malignancies of the foot: an evaluation of free-tissue transfer.

Free flaps may safely allow meaningful ambulation, durable limb preservation, and better quality of life in patients undergoing resections of soft-tissue cancers of the foot. To prove this, the records of a series of patients at The University of Texas M. D. Anderson Cancer Center (n = 67) who underwent limb salvage following tumor-related resection (n = 71 procedures) from 1989 to 1999 were retrospectively reviewed. Eighteen patients who were not candidates for local flaps or skin grafts received a total of 20 free flaps to preserve their limbs. Most defects (mean size, 78 cm2; range, 20 to 150 cm2) were on a weight-bearing surface of the foot (nine on a weight-bearing heel, three on a plantar foot); the remainder were on a non-weight-bearing surface (six on dorsum, two on a non-weight-bearing heel). Melanoma was diagnosed in nine cases (50 percent); soft-tissue sarcoma, in seven (39 percent); and squamous cell carcinoma, in two (11 percent). Fasciocutaneous and skin-grafted muscle flaps were used on both weight-bearing and non-weight-bearing surfaces. Free-tissue transfer was successful in 17 of 20 cases (85 percent); the three flap losses occurred in two patients. Minor complications (i.e., small hematoma, partial skin graft loss, and delayed wound healing) occurred in five patients. In all cases of successful free-tissue transfer, patients began partial weight bearing at a mean of 7.4 weeks (range, 2 to 12 weeks), and all ultimately achieved full weight bearing. Sixty-seven percent still required special footwear. In one patient, an ulceration on the weight-bearing portion of the flap resolved after a footwear adjustment. Only one patient was lost to follow-up (mean, 23 months). In the 17 remaining patients, limb salvage succeeded in 15 (88 percent). Of these, nine (60 percent) were alive without evidence of disease, three (20 percent) were alive with disease, and three (20 percent) had died of disease. Local recurrence developed in two patients but was successfully treated by excision and closure. No late amputations were required for local control. Thus, it seems that free flaps help facilitate limb salvage and that they may preserve meaningful limb function in patients who undergo resection of soft-tissue malignancies of the foot.     

A simple technique for closure of a palatal fistula using a conchal cartilage graft.

A simple technique for closure of a palatal fistula using a conchal cartilage graft as a substitute for nasal lining flaps is reported. This graft simplifies the repair of the palatal fistula and protects the suture line of the oral covering flaps from recurrence.     

Modified surgical technique for treating hepatic echinococcosis]

The authors present modified operational technique in hepatic echinococcosis. Echinococcus in 14 patients was removed by either peri-cystic excision or hepatic lobe excision with the closure of blood inflow into the liver by construction the blood vessels in hepato-duodenal ligament for 20 minutes. All patients survived. Biliary fistulas lasting for a few days up to 3 weeks were the only complications encountered. Presented modified operational technique of surgery is safe; it shortens the duration of surgery and prevents major intraoperative haemorrhages.