Complications after laryngectomy

Complications After Laryngectomy Daniel N. Weingrad, MD, New York, New York Ronald H. Spiro, MD, New York, New York

Total laryngectomy can be a morbid procedure with significant physiologic sequellae. Frequently patients undergo laryngectomy after an unsuccessful attempt to preserve the larynx with partial resection or radiation therapy. In this setting, complications may be devastating for the patient whose life is already limited by his disease. Multistage reconstructions with their own inherent complications can consume much of the patient’s survival time. Despite many studies, there is still disagreement as to which factors most predispose to complications after laryngectomy. We have reviewed our experience with 100 consecutive laryngectomies in order to identify the factors that contribute to the occurrence of complications, particularly pharyngocutaneous fistula, after laryngectomy. Material and Methods The records of the Memorial Sloan-Kettering Cancer Center tumor registry were reviewed to identify 100 consecutive patients who underwent total laryngectomy from March 1976 through December 1978. Laryngectomy was performed for epidermoid carcinoma in 94 patients and for laryngeal incompetence in 6. Of 25 patients with persistent or recurrent cancer, 23 had previous radiation therapy and two partial laryngectomy. Of the six patients with laryngeal incompetence, five had received radiation therapy and one underwent partial laryngectomy. The remaining 69 patients underwent laryngectomy as primary treatment for squamous cell carcinoma in the larynx or pharynx. In patients who were treated for carcinoma, the primary site involved the hypopharynx in 30, the glottis in 20, the supraglottis in 23, the base of the tongue in 12, the subglottis in 3, and the cervical esophagus in 1. Thirty-three patients had received radiation therapy to the larynx, often to the neck as well, from 1 month to more than 5 years before laryngectomy. All but one patient had received in excess of 4,000 rads, and the total dose exceeded 6,000 rads in 25 patients. Tracheostomy had been performed from 1 week to 6 months before From tha Head and Neck Service, Memorial Sloan-Kettering Cancer Center, New York, New York. Requests for reprints should be addressed to Ronald H. Spiro, MD, Memorial Hospital, 1275 York Avenue, New York, New York 10021. Presented at the 29th Annual Meeting of the Society of Head and Neck Surgeons, New Orleans, Louisiana, May 4-7, 1983.

Volume 146, October 1983

operation in 34 patients. Eleven patients had previous radical neck dissections, including three patients with bilateral radical neck dissections. Chemotherapy was given preoperatively to 10 patients. The operation performed was total laryngectomy (simple or wide field) in 48 patients, a partial or circumferential pharyngectomy was performed in addition to laryngectomy in an additional 40 and 12 patients, respectively. Radical neck dissection was performed with total laryngectomy in 13 of 48 patients, with laryngectomy and partial pharyngectomy in 29 of 40 and circumferential laryngopharyngectomy in 12 of 12 patients. A variety of techniques of skin incision and pharyngeal closure and suture materials were employed. The type of skin incision, tissues used for pharyngeal reconstruction (primary closure, flap repair, gastric transposition, or pharyngostomy), suture material used for closure, number of layers of closure, and techniques of closure (running versus interrupted) were analyzed. Antibiotics, usually penicillin or a cephalosporin, were given preoperatively or intraoperatively to 99 patients. All operative wounds were drained; closed suction drainage preferred (97 patients). All patients had nasoesophageal feeding tubes inserted before pharyngeal closure. Oral feeding was usually resumed by the ninth postoperative day. All charts were reviewed to determine patient characteristics (age, sex, and medical and nutritional status), tumor stage and extent, and treatment details (including units of blood transfused and operative time). Complications were considered major if reoperation was required and minor if no further surgery was necessary. Data were encoded on computer punch cards and analyzed using the chi-square test and the Wilcoxon rank sum test [I].

Results Complications occurred in 53 patients in the immediate postoperative period (Table I). The interval from operation to discharge was increased in 48 of these patients, averaging 29 days as compared with 16 days for patients with no complications recorded (p = 0.001). One patient who required a laryngopharyngectomy and a radical neck dissection for recurrent carcinoma after radiation therapy died postoperatively as a result of carotid hemorrhage after the occurrence of a pharyngocutaneous fistula. The overall complication rate with partial or cir-

517

Weingrad and Spiro

TABLE I

Complications in 53 Patients Wide-Field Laryngectomy

Fistula Closed spontaneously Simple closure Flap repair InfectionXellulitis Nonoperative drainage Operative drainage, debridement Hematoma Nonoperative treatment Operative drainage Carotid blowout Flap necrosis Nonoperative debridement Operative debridement Seroma Nonoperative Chylous fistula, nonoperative Chylous fistula, operative ligation Total

pharyngocutaneous fistula developed in only 2 (4 percent). After partial or circumferential laryngopharyngectomy in 52 patients, fistulas developed in 19 (37 percent). Neither previous radiation therapy nor the concomitant performance of radical neck dissection had a significant impact on the rate of fistula formation. Spontaneous closure (median time 14 days) was noted in 13 of the 21 patients in whom fistulas developed. Although previous radiation therapy was not associated with a higher incidence of fistula formation, the mean time to fistula closure was 37 days in irradiated patients compared with 10 days in those who had not received irradiation (p <0.05). Operative closure was necessary in eight patients (seven of whom had required a laryngopharyngectomy), consisting of four simple closures and four major flap reconstructions. Sepsis was recorded in 31 patients, only 3 of whom required operative drainage or wound debridement. In 28 patients, the infections were minor and responded to aggressive local care and antibiotics. Infections were more likely to occur in those patients who received previous radiation therapy (p <0.05). In the 52 patients who had extended pharyngectomy, the complication rate, the time to swallowing, and the duration of hospitalization depended on the type of pharyngeal repair (Table II). Total laryngectomy without concomitant pharyngectomy was associated with low morbidity and few major complications. After a partial laryngopharyngectomy with closure of local tissues, the complication rate was high; however, most complications were minor with a relatively low mean time to first oral feeding and postoperative days to discharge. Staged reconstruction with flaps or a temporary pharyngostomy, whether for partial or circumferential laryngopharyngectomy, was associated with a high complication rate and substantial morbidity when compared with gastric transposition.

Extended Laryngectomy Total

2 1 1 0 12 11 1

19 12 3 4 19 17 2

2 0 2

6 3 3 1’ 13 8 5 8 6 2

... 4 4 0 2 1 0 1 22

21

. .. 31

.. ... 8 ... ... ‘17

... ‘lo

.. 65

87

* The patient died

cumferential pharyngolaryngectomy was 77 percent (40 of 52 patients) as compared with 27 percent (13 of 48 patients) in those who had a simple laryngectomy. Complications were considered major in 20 patients and minor in 33. Fifteen of the 20 patients who sustained major complications were among the patients who underwent laryngopharyngectomy. The fistulas were apparent by the eighth postoperative day in half of these patients. All but one were diagnosed by the 13th postoperative day. Multiple factors were analyzed for their predisposition to fistula formation. The only significant association was the extent of surgery (p = 0.0001). Of 48 patients who underwent simple laryngectomy,

TABLE II

Relation of Complications and Morbidity to Type of Repair After Laryngopharyngectomy‘

Patients (n) Laryngectomy with partial pharyngectomy Closure, local tissues

Complications n %

32

19

59

Pharyngostomy

4

3

75

Staged flap

4

3

75

8 1 3

7

88

...

Laryngectomy with circumferential pharyngectomy Staged deltopectoral flaps Pharyngostomy Gastric transposition

2

67

Median Days to First Oral Feedina

Median Days to Discharge

9 (5-57) 9 FT 34 (30-38) 2FT 4 FT

18 (10-59) 27 (22-99+)

8 FT 1 FT 7 (7-10)

40 (1599+) 65 20 (13-4;)

47 (18-62)

Numbers in parentheses indicate the range. FT = feeding tube in place at time of discharge. l

518

The American Journal of Surgery

Complications

TABLE III

Incidence of Pharyngocutaneous Fistula Formation After Laryngectomy in 10 Reports and the Present Series

Reference

Extent Surgery

Bains and Spiro [ 91 Bresson et al [ 6] Dedo et al [ IO]

Gall et al [ 21 Goldman et al

[ 3]

Horgan and Dedo [ 7I] Lavelle and Maw [ 51 Fiobbins et al [ 1.21 Shaw [ 131 Thawley [ 741 Present report

l

After Laryngectomy

Circumferential laryngopharyngectomy (gastric transposition) Laryngectomy Laryngectomy Partial laryngopharyngectomy Circumferential laryngopharyngectomy Partial laryngopharyngectomy Laryngectomy with or without partial pharyngectomy Laryngectomy Laryngectomy Laryngectomy Laryngectomy with or without partial circumferential pharyngectomy Laryngectomy Partial laryngopharyngectomy Laryngectomy Partial laryngopharyngectomy Circumferential laryngopharyngectomy

Fistulas (n)

Patients (n)

Incidence

1

16

6

97 7 15

66 6 21

‘1’2 21

148 117 72 7’ 123 74

‘10 33

20 64 2 10

135 170 24 43

15 38 a 23

5 6

232 75

2 a

2 14 5

48 40 12

4 35 42

(%)

Feeding tubes

Comments Although our overall complication rate appeared to be higher than the 20 to 52 percent incidence described by others [2-41, comparisons are difficult because of different reporting criteria and the high proportion of extended laryngectomies (52 percent). Pharyngocutaneous fistulas after laryngectomy were attributed to a variety of factors. According to LaVelle and Maw [5], preoperative tracheostomy, radical neck dissection, and a low postoperative hemoglobin level were associated with an increased rate of fistula formation; others have implicated technical factors. Gall et al [2] reported an increased complication rate associated with positive resection margins. Radiation therapy was thought by Bresson et al [6] and others [ 7,8] to markedly increase the fistula formation rate. In this study, we found that previous radiation therapy has no significant impact. As with 10 other reports [2,3,5,6,9-141 reviewed (Table III), the incidence of fistula depended on the extent of associated pharyngeal resection and the type of repair employed. This is especially apparent in those reports [10,14] in which the fistula rate was calculated separately for laryngectomy and laryngopharyngectomies. In recent years, we have generally preferred primary repair rather than a planned pharyngostomy. In our experience, staged reconstruction using a deltopectoral flap had an unacceptably high complication rate and remarkably prolonged convalescence. It seems clear that the best hope for reduced morbidity after laryngopharyngectomy rests with the newer methods of pharyngeal reconstruction available. The myocutaneous flap has been particularly

Volume 146, October 1983

effective in this regard, particularly when used as a patch for partial pharyngeal reconstruction, rather than a tube for circumferential reconstruction [15-171. Although fistulas still occur with this technique, they are usually minor and can be managed nonoperatively. Our current preference for repair after circumferential pharyngectomy involves the use of the transposed stomach [9]. Pharyngogastrostomy has been associated with a fistula formation rate of less than 15 percent in our most recent experience. We have identified the extent of resection as the most significant factor predisposing the patient to the occurrence of complications and fistula formation after laryngectomy. Other factors were not significant when extent of resection was considered. For the patient undergoing partial or circumferential pharyngolaryngectomy, who is at considerable risk for the development of complications or fistula, one-stage reconstruction of the pharynx with a myocutaneous flap or gastric transposition should be considered. Summary The charts of 100 consecutive patients who underwent laryngectomy at Memorial Hospital were reviewed to assess those factors that contribute to postoperative complications. Laryngectomy was performed for epidermoid carcinoma in 94 patients and for laryngeal incompetence in 6. Total laryngectomy was performed in 48 patients and partial and circumferential pharyngectomies in addition to laryngectomy in 40 and 12 patients, respectively. Significant complications, which delayed discharge, occurred in 13 patients (27 percent) who had simple laryngectomy, including the formation of two fistulas

519

Weingrad

and Spiro

(4 percent). After laryngopharyngectomy, the complication rate was 77 percent (40 of 52 patients) with pharyngocutaneous fistulas in 19 patients (37 percent). The fistula rate of formation was not increased in irradiated patients; however, the duration of time to closure of a pharyngocutaneous fistula, if it occurred, was longer. Planned pharyngostomy or staged deltopectoral flap reconstruction after extended laryngopharyngectomy was associated with excessive morbidity. Newer techniques of reconstruction utilizing flaps or gastric transposition offer the prospect of reduced morbidity after laryngopharyngectomy. Acknowledgment: We thank Sara Bretsky MD, Data Analysis Group, Clinical Information Center of Memorial Hospital, for the statistical analysis. References 1. Wilcoxon F. Individual comparisons by ranking methods. Biometrics 1945;1:50-83. 2. Gall AM, Sessions DG, Ogura JH. Complications following surgery for cancer of the larynx and hypopharynx. Cancer 1977;39:824-31. 3. Goldman JL, Silverstone SM, Roffman JD, Birken EA. High dosage preoperative radiation and surgery for carcinoma of the larynx and laryngopharynx. A 14 year program. Laryngoscope 1972;82:1869-82. 4. Thawley SE. Complications of combined radiation therapy and surgery for carcinoma of the larynx and inferior hypopharynx. Laryngoscope 1981;151:677-700. 5. Lavelle RJ, Maw AR. The etiology of postlaryngectomy phar-

520

yngocutaneous fistulae. J Laryngol Dtol 1972;86:785-93. 6. Bresson K, Rasmussen H, Rasmussen P. Pharyngocutaneous fistulae in totally laryngectomized patients. J Laryngol Dtol 1974;88:835-42. 7. Cantrell RW. Pharyngeal fistula: prevention and treatment. Laryngoscope 1978;88:1204-8. 8. Joseph DL, Shumrick DL. Risks of head and neck surgery in previously irradiated patients. Arch Otolaryngol 1973;97: 381-4. 9. Bains MS, Spiro RH. Pharyngolaryngectomy, total extrathoracic esophagectomy and gastric transposition. Surg Gynecol Obstet 1979;149:693-6. 10. Dedo DD, Alonso WA, Ogura JH. Incidence, predisposing factors and outcome of pharyngocutaneous fistulas complicating head and neck cancer surgery. Ann Otol Rhino1 Laryngol 1975;84:833-40. 11. Horgan EC, Dedo HH. Prevention of major and minor fistulae after laryngectomy. Laryngoscope 1979;89:250-60. 12. Robbins JP, Marks RM, Fitz-Hug GS, Constable WC. Immediate complications of laryngectomy following high-dose preoperative radiotherapy. Cancer 1972;30:91-6. 13. Shaw HJ. Radical surgery in cancer of the extrinsic larynx and laryngopharynx. Ann R Coil Surg Engl 1957;21:290-318. 14. Thawley SE. Complications of combined radiation therapy and surgery for carcinoma of the larynx and inferior hypopharynx. Laryngoscope 1981;91:677-700. 15. Baek S, Lawson W, Biller HF. An analysis of 133 pectoralis major myocutaneous flaps. Plast Reconstr Surg 1982;69: 460-7. 16. Strawberry CW, DeFries HO, Deeb ZE. Reconstruction of the hypopharynx and cervical esophagus with bilateral pectoralis major rnyoc&neous flaps. Head f&k Surg 1981;4~161-4. 17. Theooarai SD. Merritt WH. Acherva G. Cohen IK. The oectoralis mijor musculocutaneous island flap in single-stage reconstruction of the pharyngcesophageal region. Plast Reconstruct Surg 1980;65:267-76.

The American Journal of Surgery