From the Departments of Otolaryngology (Drs Lavertu, Bonafede, Strome, Eliachar, and Wood), Hematology and Medical Oncology (Dr Adelstein), and Radiation Oncology (Dr Saxton), Cleveland Clinic Foundation, Cleveland, Ohio, and Central New York Ear Nose and Throat Consultants, Syracuse (Dr Wanamaker).
To determine the incidence of minor and major complications in patients with squamous cell carcinoma of the upper aerodigestive tract who require surgical salvage or planned neck dissection after an initial treatment regimen with radiotherapy or concurrent chemoradiotherapy for organ preservation.
The medical records of 100 patients treated in a phase 3 trial comparing radiotherapy alone with concurrent chemoradiotherapy for stage III and IV head and neck squamous cell carcinoma were reviewed. Fifty-four patients underwent 59 surgical procedures. Twenty-nine planned neck dissections were performed for persistent neck disease or initial stage N2 or greater. For persistent or recurrent disease at the primary site, 30 salvage operations were performed.
Academic tertiary care referral center.
Complications occurred in 15 (46%) of the 33 procedures in the radiation-only group and 12 (46%) of the 26 procedures in the chemoradiotherapy group. Major complications occurred in 4 (12%) of the procedures in the radiation-only group and 3 (12%) of the procedures in the chemoradiotherapy group. The incidence of minor complications was 33% and 35% in the radiation-only and chemoradiotherapy groups, respectively. The major complication rate for salvage operations did not differ between the radiation-only and chemoradiotherapy groups (16% and 27%, respectively; P=.79 by χ2test). The incidence of major complications in planned neck dissections was 7% of the radiation-only group and 0% of the chemoradiotherapy group.
After radiation or concurrent chemoradiotherapy, surgery can be performed with an acceptable rate of major complications. Adding chemotherapy did not increase the incidence of surgical complications. These results differ from other reports in the literature.
COMBINATIONS of surgery and radiation therapy are standard treatments for patients with advanced local or regional squamous cell carcinoma of the head and neck.1 Adjuvant and neoadjuvant chemotherapy has been added to various multimodality approaches.1- 7Concurrent chemotherapy and radiation therapy has also been proposed.8- 10 Advantages of this approach include the independent antineoplastic activity of both treatments, the radiotherapeutic potentiation exerted by several chemotherapeutic agents, the possibility that systemic chemotherapy may decrease micrometastatic spread, and the relatively short duration of therapy.8
Although these aggressive organ preservation protocols have produced encouraging results, an increased incidence of surgical complications in patients who require surgery for persistent or recurrent disease has been reported. Curative radiotherapy and induction chemotherapy followed by definitive radiotherapy result in complication rates for salvage operations of 37% to 74%2- 5,11and 46% to 100%,6,7,12,13respectively.
Induction chemotherapy and chemoradiotherapy for organ preservation are used with the primary goal being the elimination of extensive surgery in hopes of maintaining normal speech and swallowing. Because 31% to 61%1,11,14 of patients in these protocols will require either a prophylactic neck dissection or a salvage operation, the safety of these subsequent surgeries must be assessed.
Between April 23, 1990, and August 1, 1995, 100 patients with resectable stage III or IV squamous cell head and neck cancer (excluding nasopharynx, salivary gland, and paranasal sinus) were stratified by nodal status and randomly assigned to receive either radiation therapy alone, 68 to 72 Gy at 1.8 to 2 Gy/d, or identical radiation therapy with concurrent chemotherapy. Chemotherapy consisted of fluorouracil, 1000 mg/m2per day, and cisplatin, 20 mg/m2 per day, both given as continuous intravenous infusions over 4 days beginning on day 1 and day 22 of the radiation treatment (Figure 1). Megavoltage radiation therapy was generated by a 6-MV linear accelerator. In general, opposed lateral fields were used, with an electron beam boost given to selected nodal regions as indicated.
Treatment schema. CR indicates complete response; PR, partial response; NR, no response; PD, progression of disease; and RT, radiotherapy.
After receiving approximately 55 Gy, patients were clinically reassessed. Those patients with evidence of a clinical response completed radiation therapy to a total dose between 68 and 72 Gy. Radiation was discontinued, however, for those who did not respond, or those with progressive disease, and surgical resection was performed after a 3- to 6-week recovery period.
In patients who completed full-course radiation therapy, 6 to 12 weeks was allowed for mucosal recovery before formal response assessment and evaluation for residual primary tumor were performed. Biopsies were performed only if residual disease was suspected. Surgery at the primary site was performed for persistent or recurrent disease; this was considered a salvage operation. Neck dissection was performed in patients with any clinically palpable remaining lymph nodes and in most patients with N2 or N3 disease at initial examination irrespective of clinical response; this was considered a planned neck dissection. All surgeries were performed by faculty or residents under faculty supervision at The Cleveland Clinic Foundation, Cleveland, Ohio.
The medical records of all patients who required surgery were retrospectively reviewed and the following information was extracted: patient age, TNM stage, site of primary disease, radiation dose, time between completion of therapy and surgery, type of surgical procedure, blood loss, duration of surgery, length of hospital stay, and perioperative surgical complications. Major complications were fistulas and those complications requiring operative intervention or intensive care unit monitoring. Minor complications were those managed nonoperatively.
Student t tests and exact χ2tests were used to compare the radiation and chemoradiotherapy groups on patient and surgical variables and complications. Exact χ2 test was used to compare complications between the planned and salvage surgical groups. Differences were considered significant at P<.05.
Of the 100 patients enrolled and assigned to treatment, 54 underwent subsequent planned neck dissections or salvage surgery. Patient demographic features are outlined in Table 1. The treatment arms did not significantly differ on the following variables: age (P=.54), location of primary tumor (P=.78), and N stage (P=.09). The radiation-only arm had significantly more patients with T1 tumors (P=.02). This difference was also noted in the original cohort of 100 patients. Eight patients with T1 lesions were assigned to radiation-only therapy, as opposed to no patients in the chemoradiation group.10
Fifty-nine surgical procedures were performed. These were equally distributed between the 2 arms of the study (P=.24) (Table 2). Five patients underwent 2 procedures. Two patients (1 in each arm) required total laryngectomy and partial pharyngectomy 16 and 36 weeks after planned modified radical neck dissection. Two patients (1 in each arm) underwent modified radical neck dissection for delayed contralateral neck metastases 46 and 102 weeks after total laryngectomy and unilateral neck dissection. One patient receiving radiation alone required a supraglottic laryngectomy 21 weeks after a planned modified radical neck dissection. Of all 11 flaps used for reconstruction, 2 were free flaps, both in the radiation-only group. All others were pectoralis major myocutaneous flaps.
Surgical complications are compared between the 2 study arms in Table 3. The overall complication rate did not differ between the radiation-only and the chemoradiotherapy groups (46%; P=1.00). Minor and major complication rates did not differ significantly between arms (P=1.00).
The complication rates did not differ between arms for planned neck dissections (P=1.00). Minor and major complication rates did not differ significantly between arms (P=.68). There were no major complications in the chemoradiotherapy group. One major complication occurred in the radiation-only group involving a 45-year-old man with a T1N2b base-of-tongue lesion. A chyle leak that occurred after a planned left modified radical neck dissection did not respond to conservative management and required 3 subsequent surgical procedures. Table 3 describes the types of complications occurring in patients with planned or salvage surgery. All skin incision complications responded to conservative outpatient management.
Salvage procedures had an overall complication rate of 60%, which differed significantly from the 31% rate for planned neck dissections (P=.04). The major complication rate was significantly greater for salvage than for planned neck dissections (20% vs 3.4%, respectively; P=.04). The overall, minor, or major complication rates in salvage operations did not differ between the radiation-only and chemoradiotherapy group (P=1.00 and .79, respectively). Table 3 describes the complications. One pharyngocutaneous fistula occurred in a 63-year-old man with a T3N0 laryngeal lesion who received radiation alone 32 days after a total laryngectomy and unilateral selective neck dissection. The fistula closed with conservative management after 7 months, and the patient had no evidence of disease at 30 months of follow-up. A carotid artery rupture occurred in a 64-year-old patient with a persistent T3N1 hypopharyngeal lesion after radiation. After a total laryngectomy and partial pharyngectomy with radial forearm flap reconstruction, a small wound breakdown, without fistula, occurred over the carotid artery, resulting in rupture on postoperative day 25. This event required ligation of the common carotid artery, sacrifice of the radial forearm free flap, and placement of a pectoralis myocutaneous flap. The patient died of unrelated causes 8 months after the primary surgery. Also, a small-bowel obstruction occurred in a 61-year-old man after a total laryngopharyngectomy and jejunal free flap reconstruction for a persistent T2N2b hypopharyngeal lesion. He was readmitted to the hospital on postoperative day 10 with emesis. The obstruction responded to intravenous hydration and nasogastric decompression.
Surgical variables are compared between the 2 arms in Table 4. Information for operative blood loss, duration of surgery, and length of hospital stay was not available for all procedures. No significant differences were seen in overall length of hospital stay (P=.59), operative blood loss (P=.90), or length of time between completion of therapy and surgery (P=.61). Planned neck dissections and salvage groups were separated into those with complications (minor or major) and those without complications and were compared on the same variables.
Planned neck dissections, regardless of complication status, did not differ significantly between the 2 arms for all variables. In the salvage groups, however, some trends were noted. Those receiving chemoradiation and salvage surgery who had complications compared with those without complications had greater mean operative blood loss (877 mL and 388 mL, respectively), mean duration of surgery (613 and 341 minutes, respectively), and length of stay (15.3 and 7.3 days, respectively). The mean length of time between completion of therapy and surgery was less in the planned group (range, 6.8 to 8.4 weeks) than in the salvage group (18.3 to 64.5 weeks); the interval in the salvage group with complications in the chemoradiation arm was shorter than that in the group without complications (18.3 and 64.5 weeks, respectively).
Table 5 compares duration of procedure, estimated blood loss, and length of hospital stay by specific surgical procedure. Total laryngopharyngectomy plus flap reconstruction and total glossectomy, total laryngectomy plus flap reconstruction were combined because of the similar complexity of the procedures. No statistical analysis was performed because of the small group sizes; however, certain trends existed. Total laryngectomies were approximately 90 minutes shorter in mean duration (500 vs 590 minutes) and modified radical neck dissections were approximately 90 minutes longer in mean duration (318 vs 233 minutes) in the radiation-only group as opposed to the chemoradiotherapy group.
The mean estimated blood loss and mean hospital stay were similar between the arms for surgical procedures except for the total laryngectomy and partial pharyngectomy group. In this group, the mean hospital stay was substantially longer in the chemoradiotherapy group (29 days) than the radiotherapy group (11.3 days). This difference can be explained by 1 patient with a 48-day hospital stay; this was a 66-year-old man with a T3N2b laryngeal lesion who had a cerebrovascular accident and required a carotid endarterectomy, systemic anticoagulation, and 3 weeks in an intensive care unit for pulmonary complications. Widespread locoregional recurrence developed, and the patient died on postoperative day 48.
Preoperative radiotherapy for advanced squamous cell carcinoma of the head and neck has been reported to increase surgical complication rates.2- 4 Complication rates for surgical salvage operations of 37%, 43%, 58%, and 74% have been reported.2- 5 Radiation dose of more than 50 Gy has been implicated, and some recent reports suggest that doses of 66.5 Gy and higher increase complication rates.2When higher doses of preoperative radiotherapy (50-70 Gy) are administered, the rate of pharyngocutaneous fistula alone is approximately 31% to 34%.15
Most studies addressing surgical complications in organ preservation protocols have involved induction chemotherapy with planned surgery and postoperative radiotherapy schedules. Schuller et al12 reported that 28 (46%) of 61 patients developed surgical complications after 3 courses of induction with cisplatin, methotrexate, and bleomycin sulfate followed by surgery and radiotherapy. Six patients developed severe complications and 8 developed fistulas. Corey et al7 reported a 52% surgical complication rate and a 22% fistula rate in 23 patients treated with 3 cycles of induction methotrexate and leucovorin calcium rescue; the majority of patients received postoperative radiotherapy.
Organ-preservation protocols that include induction chemotherapy and radiation before surgery have produced even higher complication rates. Sassler et al6 reported a 66% overall complication rate and a 50% fistula rate in 18 patients who required salvage surgery after 3 cycles of induction chemotherapy (cisplatin and fluorouracil) and a 66- to 74-Gy radiation dose delivered to the primary tumor. Complications were reported in 6 of 17 patients by Newman et al.14 This treatment protocol consisted of 2 cycles of induction chemotherapy (cisplatin and fluorouracil) followed by simultaneous administration of 2 cycles of combination chemotherapy and standard fractionated radiotherapy to a total dose of 66 to 75 Gy. Five of the 6 patients had undergone pharyngeal procedures, 1 of whom then experienced a pharyngocutaneous fistula.
Our overall complication rate of 45.8% (for the 59 procedures on 54 patients)compares favorably with that reported for other organ-preservation protocols.6,9,11,14Considering granulation tissue formation at the tracheostoma and hypothyroidism as minor complications (Table 3) contributed to our somewhat higher minor complication rate (33.9%). Major complication rates were acceptable for the radiation-only and chemoradiotherapy groups (12% and 11%, respectively).
Only 1 pharyngocutaneous fistula formed. This rate (4%) of fistula formation in pharyngeal procedures is considerably lower than that reported in the recent literature.2,5,7,11- 14This difference may be a result of our aggressive use of either nasogastric or percutaneous gastric tube placement during therapy to optimize nutrition and our frequent use of flap reconstruction in 11(46%) of 24 patients. The spontaneous closure rate of this fistula, after 7 months, certainly was delayed because of the previous radiotherapy.
Neither minor nor major complications differed between treatment arms. There was a significantly higher incidence of overall complications, as expected, in the salvage procedures (60%) than in the planned neck dissections (31%) (P=.04). No significant difference was found when these procedures (salvage and planned neck dissections) were compared within both treatment arms. Major complication rates in salvage operations (16% for radiotherapy and 27% for chemoradiation) compare favorably with those of other organ-preservation protocols,6,7,11,13,14although comparisons among protocols are difficult because of different treatment schemas, types of surgical procedures, and definitions of major and minor complications.
Our low rates of major complications indicate that planned neck dissections can be performed after radiotherapy or chemoradiotherapy, which agrees with findings in other series.14 We found a higher incidence of minor incisional breakdown in the chemoradiotherapy group.
Sassler et al6 found that major wound complications increased in salvage procedures performed less than 52 weeks from the completion of chemotherapy and radiation therapy (77%) vs those performed after more than 52 weeks (23%) (P=.47). We noticed a similar trend in our patients undergoing salvage surgery with complications in the chemoradiation group. In these patients (n=7) the mean time from completion of therapy to surgery was 18.3 weeks, whereas in the salvage group without complications (n=4) in the same arm the interval was 64.5 weeks. Apparently surgical morbidity decreases as the interval between completion of chemoradiotherapy and surgery increases. It is unclear whether this difference results from a selection bias in which patients with more aggressive tumors at the primary site require earlier surgery and have more extensive pharyngeal defects or if recovery from the systemic effects of chemotherapy is a factor.
To identify whether the addition of chemotherapy influenced the technical aspects of the procedure, we compared the procedure on the surgical variables between the treatment arms (Table 4and Table 5) and found the data were similar to those of other reports. Mean operative time for neck dissections was longer than we expected and may be explained by the fact that the majority of these procedures included direct laryngoscopy (with or without biopsy) to assess the primary site. Newman et al14 had mean (±SD) operative time of 246±18 minutes and a length of hospital stay of 5.0±1.7 days for neck dissections in 17 patients after induction and concurrent chemoradiotherapy. Girod et al5found a mean operative time of 589±194 minutes for major head and neck procedures in a group of 159 patients; however, only 39% had previous radiotherapy and 52.8% required flap reconstruction.
Our mean length of stay for patients with salvage procedures was low (Table 4). This reflects our use of home nursing care, excellent supporting services at our institution, and our low major complication rate. Many of these patients however, required frequent outpatient visits early in the postoperative period.
This study confirms other reports9,11,14showing acceptable surgical morbidity rates for salvage surgery performed after intensive organ-preservation protocols. Surgical complication rates do not significantly increase when concurrent chemotherapy is added to preoperative curative radiotherapy. Because only 30% of patients in this protocol required salvage surgery and only 20% of these developed major complications, the overall surgical morbidity of the organ-preservation protocol was quite acceptable.
Accepted for publication October 21, 1997.
Presented at the annual meeting of the American Society of Head and Neck Surgery, Scottsdale, Ariz, May 13, 1997.
Corresponding author: Pierre Lavertu, MD, Department of Otolaryngology and Communicative Disorders A-71, The Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195 (e-mail: email@example.com).
Thank you for submitting a comment on this article. It will be reviewed by JAMA Otolaryngology–Head & Neck Surgery editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 87
Customize your page view by dragging & repositioning the boxes below.
Care at the Close of Life: Evidence and Experience
The Rational Clinical Examination: Evidence-Based Clinical Diagnosis
Evidence Summary and Review 3
All results at
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.