Disclaimer: The information contained within the Grand Rounds Archive is intended for use by doctors and other health care professionals. These documents were prepared by resident physicians for presentation and discussion at a conference held at Baylor College of Medicine in Houston, Texas. No guarantees are made with respect to accuracy or timeliness of this material. This material should not be used as a basis for treatment decisions, and is not a substitute for professional consultation and/or peer-reviewed medical literature.

Pharyngocutaneous Fistula after Laryngectomy
Allen Lue, M.D.
August 3, 2000

Pharyngocutaneous fistula after laryngectomy is the most common complication that we see after this procedure and it increases the morbidity, length of hospitalization and also the possibility of mortality. Here is a chart that was taken out of an article in the White Journal--it compared the postop courses of two post-laryngectomy patients; one did fine and the other developed a fistula. As you can see, the hospital stay was increased from 10 days to 26 days and the total cost of care was increased from $26,000 to $61,000. So, this is certainly a complication to be avoided if at all possible.

Billroth performed the first laryngectomy on New Year’s Eve in 1873 in Vienna. He gave this patient a controlled pharyngostome. He did not close the pharynx and used a specially modified tube with a pharyngeal lumen and a tracheal lumen to decompress the pharyngeal suture line. This patient died seven months postop and although he did not develop a fistula, Billroth was the first person to report pharyngocutaneous fistula as a complication. Fistula rates eventually increased with Gluck, who was the first to describe a single stage laryngectomy with primary closure of the pharynx putting more pressure on the pharyngeal suture line. In the 1950’s when Cobalt 60 radiation was developed, it also increased fistula rates for patients that required salvage surgery. So, moving on to incidence, here is a nice chart outlining a lot of the literature almost to date. As you can see, the rate of fistula occurrence varies widely depending on the studies anywhere from 7.0% by Dedo to up to 50% and certainly there is a bias depending on patient selection and on the surgical technique as well. But again, most of the studies are somewhere in between 50% and 7%.

A fistula is an abnormal communication between two epithelialized surfaces and in our case, these two surfaces are the pharyngeal mucosa and the skin. With a pharyngocutaneous fistula, what happens is a salivary leak develops from the pharyngeal closure to the skin, which implies a breakdown of the pharyngeal suture line. In contrast to other fistulas in other parts of the body, this is a low output fistula and there generally are not any metabolic complications from the fistula output. This is a nice chart that shows some of the favorable and unfavorable factors affecting spontaneous closure of fistulas throughout the body. And the things that are relevant to us are an associated abscess certainly makes things unfavorable. Healthy adjacent tissue is very important in terms of fistula closure. Esophageal versus gastric esophageal is more favorable and the length of the tract if it is greater than 2.0-cm, it is favorable, as well as a defect less than 1.0-cm. Experimentally, there has been one model that has been reported by Weiss in 1993. This person studied New Zealand white rabbits, removed a section of the thyrohyoid membrane and fed them oral gentian violet over a period of time to assess wound breakdown. She stated that this would be a good model to study factors such as antibiotics and the timing of feeding although I have not really seen any literature from this group after this first study.

In terms of diagnosis, fistulas usually develop five to fifteen days after the procedure and it certainly correlates to when feeding started. There are symptoms and signs, actually only signs I have listed here. Fever is very common although it is very nonspecific. A study done in 1999 showed that a temperature elevation in the first 48 hours postop highly correlated with fistula formation. Out of the 34 patients that he studied who developed a postop fever, 24 of them or 71% developed a fistula. Also interestingly, out of the hundred patients that were in his study who did not develop a postop fever, only four or 4.0% developed a fistula. Wound erythema and edema are also things to look for and things that you should be vigilant about and an elevated turbid drain output. An elevated amylase in your wound drainage can be found as early as postoperative day two in patients who will eventually develop a fistula. Contrast studies have also been written about. You can do serial studies to decide on the start of feeding. A sinus tract of 2.0-cm is supposedly predictive of eventual fistula formation. There is a growing opinion that there is really no need to do contrast studies in patients with signs or symptoms of impending fistula because these signs are fairly sensitive and specific. So, moving on to risk factors. There is really no general agreement on factors related to fistula formation. As I have said before, there is a multitude of retrospective studies. Dedo was the first person to publish these factors in 1975 and he just listed the risk factors that the patients had who eventually developed the fistula: high dose of radiation, persistent carcinoma, delayed extension to the posterior pharyngeal wall, systemic disease, foreign body granuloma and post-surgical persistence. You can really break the variables down in the literature to three categories. The first is those variables related to the patient, those related to the tumor and those related to treatment. We will discuss each of these categories separately.

The first is patient-related risk factors and in particular, the general condition of the patient, which can impair wound healing. Preoperative weight loss and poor nutrition certainly contributes to fistula formation. However, serum albumin was not shown to be predictive in any of the retrospective studies that I looked at. Systemic disease certainly is a risk factor. Dedo first showed that. We saw that in the previous chart. This study out of Athens showed in 314 patients that there was a 4.0% versus 19.0% fistula rate with patients who had serious medical problems, in particular coronary artery disease and pulmonary disease. Dezennes in an Italian study of 214 patients showed that congestive heart failure was an independently significant variable. Anemia is also a very interesting risk factor. Postoperative hemoglobin of less than 12.5 one week postoperatively was shown by Lavelle and Mahr in a London series to increased fistula rates from 29% to 74%. Dezennes in his Italian study also confirmed that showing a nine-fold increase in patients who had postop anemia. The need for intraoperative blood transfusion has also been shown to be a risk factor. Ayre in 1993 showed an increased fistula rate from 7.0% to 28.0% in patients who required an intraoperative transfusion. Could there be an immunosuppressive effect? It certainly has been shown in the colon cancer literature as far as recurrence goes but that has yet to be proven. However, this Turkish study showed that out of 110 patients, there was no statistically significant difference in patients who required transfusion.

So, moving to tumor-related variables, late stage is certainly a variable that affects your likelihood of fistula formation. The size of tumor is a factor, and any piriform sinus tumors requiring partial pharyngectomy. (The patient in the case presentation was also part of this group; 67% of those patients eventually developed a fistula.) An extended laryngectomy doubled the risk in another study and one can only assume that it is because of increased tension on the pharyngeal closure, which is required by a larger lesion.

As far as treatment-related risk factors go, there are sort of three classifications of the variable study. Certainly, previous XRT, surgical technique and also any associated procedures that you do along with your laryngectomy. Generally, the conventional wisdom is that radiation therapy does indeed increase your risk of postoperative fistula. However, there had been a few large studies that have shown that there was no correlation. This is really the largest retrospective series that I have found to date directly looking at XRT and fistula formation. Out of 167 patients in Liverpool who did not receive XRT, only 10 developed a fistula or 4.0%. After XRT, however, out of 190 patients, 74 developed fistula, which is 39%. The authors also noted that the fistulas, which occurred after primary surgery, that is without XRT, healed sooner. And here is really the strongest data to show that XRT causes fistulas. This is a study out of Denmark where they radiate all laryngeal cancers and so all the surgeries are essentially post-XRT. Because of this, they were able to develop a dose response actually to the radiation. This is a nice little dose response curve here and it shows that as you increase the total dose of radiation, you increase the rate of fistulas. Something else, which I will go into more detail later, is that giving Flagyl to these patients decreased their fistula rate from the 50’s to about 13%. So, here is the good part. Surgical technique can certainly alter your likelihood of fistula formation and certainly a tenet you want to follow is that you want to have minimal wound tension and tissue strangulation. One thing to consider is that a vertical closure of the pharynx is better than a T-closure since a trifurcation is structurally weaker than a simple vertical closure. You also want to avoid a tight hypopharyngeal closure. Here is a very interesting study out of Taiwan in which Wang decided not to close the constrictors over his pharyngeal suture line and that reduced the rate of fistula from 10 to 3.0% in 60 patients. Solle reported that closing with Vicryl is better than catgut as well. Associated procedures have also been shown to increase your rate of fistula formation. Preoperative tracheostomy increased your fistula rate from 34% to 53%. One could assume that has to deal with an increased likelihood of infection. Stell and Cooney advocated using a separate skin incision for the stoma to try to decrease your fistula rate. A simultaneous radical neck dissection also increased fistula rates in the Lavelle and Mahr study from 22% to 53%. Primary TEP has not been shown to increase fistula rates in this study.

So, I have been talking more about factors that you really cannot do anything about, whether the size of the tumor or patient factors in terms of their health. Here are some things that you can do to prevent fistula. One of the most important is a very meticulous watertight repair of your pharyngeal mucosa. You can use any number of stitches and one of the most popular is the Connel stitch using a vertical versus a T-closure. Placing a free dermal graft over the pharyngeal suture line was not shown to protect you from fistula. However, if you have a high suspicion of a fistula, you can use that graft and place it directly over the carotid to afford you some degree of protection in case you do develop a fistula. Give patients antibiotics after their laryngectomy. Flagyl given preop and 10 days postop reduced the fistula rate from 60% to 13%; that was that Danish study that I had shown previously. Reflux prophylaxis has also been shown to be important. Kaufman in a very well known study showed that 71% of patients with laryngeal cancer have reflux disease. And Sikley built on that and showed that ranitidine and Reglan given to 17 patients resulted in no fistulas and reduced hospital stay compared to 29 controlled patients. Your timing of feeding is certainly important in terms of the propensity for fistula formation. In a survey in 1989, about 85% of surgeons waited about seven days before feeding. There has lately though been a trend to feed sooner. Bois show that there was no difference in fistula rate in a retrospective series of patients who were fed between three and seven days.

What do you do for a fistula when it does occur? Most fistulas do respond well to conservative management, about 60-80% of them, you start them on tube feeds and make them NPO. You certainly give them antibiotics and you start local wound care. Ideally, you pack the fistula with gauze to encourage healing. Another very important thing is to protect the airway. If you are fortunate enough to still have your suction drain, you can use it to divert the fistula away from your stoma so that they are not dripping into your lungs and causing an aspiration pneumonitis. A cuff tracheostomy tube is also useful in that regard. You go to a surgical repair after your conservative measures fail. Timing varies. It has been reported that some people wait two weeks for non-radiated tissue. They operate if the fistula has not healed and they wait three weeks for an operation in radiated tissue. Cohen treated a series of patients aggressively closing their fistulas on average postoperative day 12 that they occurred and he showed that it reduced hospital stay and the need for wound care. Certainly, if there is carotid exposure, your need for coverage is fairly urgent. So, surgical repair is generally used for larger fistulas and one thing to know is that you have to close your epithelial layers into the skin and the mucosa. This is an article from Stelle and Cooney in 1974. It classifies fistulas in terms of whether or not tissue can be obtained locally or at a distance. Locally, if there is a healthy tissue around the fistula then you can get away with getting tissue from close by. Certainly it would be easiest to directly suture the fistula closed. However, usually you need to bring in some surrounding well-vascularized tissue. You may need to use a rotational flap to bring vascularized tissue near differential defect. You can get distant tissue as well if local tissue is not good enough. One of the most popular flaps to use is a pectoralis major flap; it has a very abundant blood supply but the problem is that it can sometimes be bulky and can give you problems with dysphagia. Deltopectoral flaps and trapezius flaps have also been described. Free flaps have also been described for extensive or circumferential defects. The problem with the free flap though is that you are not going to want to put that free flap into an infected recipient site. Postoperative radiation of fistula patients is a very special problem. Vichram in a landmark study showed that any delay of greater than six weeks postop increases your neck recurrence from 2.0% to 29.0% so it is important to try to start your therapy within this window. However, Isaacs did show that radiated fistulas can eventually heal. Sixty seven percent of these fistulas healed despite getting 60-70 Gy. Fifteen percent of those fistulas were closed surgically and 18% of these patients died with their fistula or were lost to follow up.

These are the complications of fistula. Here is a picture of a carotid blowout. Some of the more minor complications though are certainly wound dehiscence, aspiration pneumonitis, mediastinitis can occur from fistula drainage and certainly sepsis as well and again one of the most important complications to avoid is carotid blowout and preserving the SEM can help to some extent. So, in conclusion, pharyngocutaneous fistula is one of the most common and troublesome complications after laryngectomy. There are a lot of factors that contribute to fistula formation and some are avoidable and certainly those include vigilance of the wound for signs, meticulous closure, antibiotics, reflux precautions and careful timing of oral feeding. Most fistulas do close with conservative management although surgical closure may be required to prevent catastrophic complications.

Case Presentation:

Bibliography:

Applebaum E, Levine H. Pharyngeal reconstruction after laryngectomy. Laryngoscope 1997;87:1884-1890.

Berry SM, Fischer JE. Classification and pathophysiology of enterocutaneous fistulas. Surg Clin North Am 1996;76:1009-1018.

Boyce S, Myers A. Oral feeding after total laryngectomy. Head Neck Surg 1989;11:269-273.

Bresson K, Rasmusson H, Rasmussan P. Pharyngocutaneous fistulae in totally laryngectomized patients. J Laryngol Otol 1974;88:835-842.

Briant TDR. Spontaneous pharyngeal fistula and wound infection following laryngectomy. Laryngoscope 1975;85:829.

Celikkanat S, Koc C, Akyol MU, Ozdem C. Effect of blood transfusion on tumor recurrence and post-operative pharyngocutaneous fistula formation in patients subjected to total laryngectomy. Acta Otolaryngol 1995;115:566-568.

Cohen M, Marschall MA, Greager J. Early, aggressive management of postoperative oropharyngocutaneous fistulas. Plast Reconstr Surg 1992;89:56-61.

De Jong P, Struben W. Pharyngeal fistulae after laryngectomy. J Laryngol Otol 1970;84:897-903.

De Zinis L, Ferrari L, Tomenzoli D, Premoli G, Parrinello G, Nicolai P. Postlaryngectomy pharyngocutaneous fistula: incidence, predisposing factors and therapy. Head Neck 1999:21:131-138.

Dedo D, Alonzo W, Ogura J. Incidence, predisposing factors and outcome of pharyngocutaneous fistulas complicating head and neck surgery. Ann Otol Rhinol Laryngol 1975;84:833-840.

Fielder CP, Morton RP. Wound amylase estimation and the prediction of pharyngocutaneous fistula. Clin Otolaryngol 1989;14:101-105.

Friedman M, Venkatesan TK, Yakovlev A, Lim JW, Tanyeri HM, Caldarelli DD. Early detection and treatment of postoperative pharyngocutaneous fistula. Otolaryngol Head Neck Surg 1999;121:378-380.

Heir M, Black MJ, Lafond G. Pharyngocutaneous fistulas after total laryngectomy: incidence, etiology, and outcome analysis. J Otolaryngol 1992;22:164-166.

Johansen LV, Overgaard J, Elbrønd O. Pharyngocutaneous fistulae after laryngectomy. Influence of previous radiotherapy and prophylactic metronidazole. Cancer 1988;61:673-678.

Koltai PJ, Lipzig B. Dermal graft for protection of the pharyngeal suture line in cancer surgery of the head and neck. Otolaryngol Head Neck Surg 1981;89:260-263.

Koufman J. The otolaryngologic manifestations of gastroesophageal reflux disease. Laryngoscope 1991;101(Suppl. 53):1-78.

Krause JH, Metson R. Barium swallow is a predictor of salivary fistula following laryngectomy. Otolaryngol Head Neck Surg 1992;106:254-257.

Lavelle RJ, Maw AR. The aetiology of post-laryngectomy pharyngo-cutaneous fistulae. J Laryngol Otol 1972;86:785-793.

Lundgren J, Olofsson J. Pharyngocutaneous fistula following total laryngectomy. Clin Otolaryngol 1979;4:13-23.

Mantravadi R, Skolnik E, Applebaum E. Complications of post-operative and pre-operative radiation therapy in head and neck cancers. Arch Otolaryngol 1981;107:690-693.

McCombe A, Jones A. Radiotherapy and complications of laryngectomy. J Laryngol Otol 1993;107:130-132.

Moses B, Eisele D, Jones B. Radiologic assessment of the early postoperative total-laryngectomy patient. Laryngoscope 1993;103:1157-1160.

Papazoglou G, Doundoulakis G, Terzakis G, Dokianakis G. Pharyngocutaneous fistula after total laryngectomy: incidence, cause and treatment. Ann Otol Rhinol Laryngol 1994;103:801-805.

Parikh SR, Irish JC, Curran AJ, Gullane PJ, Brown DH, Rotstein LE. Pharyngocutaneous fistula in laryngectomy patients: the Toronto Hospital experience. J Otolaryngol 1998;27:136-140.

Raman R. Diversion as treatment of early postlaryngectomy suprastomal pharyngocutaneous fistula. Plast Recon Surg 1998;101:543.

Rees R, Asher C, Shack RB, et al. Pharyngocutaneous fistulas in advanced cancer: closure with musculocutaneous or muscle flaps. Am J Surg 1987;154:381-383.

Rubin JS. Repair of post-laryngectomy pharyngeal fistulae. J Otol Laryngol 1989;130: 302-305.

Seikaly H, Park P. Gastroesophageal reflux prophylaxis decreases the incidence of pharyngocutaneous fistula after total laryngectomy. Laryngoscope 1995;105:1220-1222.

Soylu L, Kiroglu M, Aydogan B, et al. Pharyngocutaneous fistula following laryngectomy. Head Neck 1998;20:22-25.

Stell P, Cooney T. Management of fistulae of the head and neck after radical surgery. J Laryngol Otol 1974;88:819-833.

Tomkinson A, Shone GR, Dingle A, Roblin DG, Quine S. Pharyngocutaneous fistula following total laryngectomy and post-operative vomiting. Clin Otolaryngol 1996;21:369-370.

Vikram B, Strong EW, Shah JP, Spiro R. Failure in the neck following multimodality treatment for advanced head and neck cancer. Head Neck Surg 1984;6:724-729.

Wang CP, Tseng TC, Lee RC, Chang SY. The techniques of nonmuscular closure of hypopharyngeal defect following total laryngectomy. J Laryngol Otol 1997;111:1060-1063.

Weingrad D, Spiro R. Complications after laryngectomy. Am J Surg 1983;146:517-520.

Weiss LS, Miller RH, Bates ML, Fisher E, Weiner N, Shepard D. An animal model for pharyngocutaneous fistulas. Otolaryngol Head Neck Surg 1993;108:738-742.

BCM Public | BCM Intranet | Privacy Notices | Contact BCM | BCM Site Map |

©2001-2006 Baylor College of Medicine
Bobby R. Alford Department of Otolaryngology-Head and Neck Surgery
Mail: One Baylor Plaza, NA102, Houston, TX 77030
Phone: 713-798-5906
E-mail: oto@bcm.tmc.edu

Last modified: January 3, 2006