Early Wound Complications in Advanced Head and Neck Cancer Treated With Surgery and Ir¹⁹² Brachytherapy^†‡

INTRODUCTION

Advanced or recurrent squamous cell carcinoma of the head and neck continues to have a poor prognosis and present a challenge to the head and neck surgeon. Efforts to improve survival in patients with close or positive margins at resection have included the use of brachytherapy, either temporary, via afterloading catheters,^1-5 or with permanent seed implants.^6-9 The efficacy of this technique has been demonstrated by improved survival,^{3, 4, 7, 8} but the consequences of such treatment, including wound complications and long-term sequelae, have not been well documented.

Wound complications have been demonstrated in 36% to 50% of patients after brachytherapy.^{1, 2, 7} This rate is higher than that seen with re-irradiation by external radiotherapy alone³ and seems primarily related to wound breakdown or necrosis. The most frequently noted early complications have been flap loss, wound breakdown, suture line dehiscence, and wound infection, whereas late complications have included osteoradionecrosis, carotid blowout, and fibrosis. Animal studies have also demonstrated high tolerance of the carotid artery to interstitial brachytherapy, despite long-term changes.¹⁰ Flap reconstruction has decreased the incidence of complications in several studies using either pedicled flaps^{2, 4} or free flaps^{11, 12} for reconstruction. Pectoralis major myocutaneous (PMC) flap reconstruction has been an effective means of head and neck reconstruction in the previously irradiated patient as well as those who have not received radiotherapy. Complication rates as high as 70% have been reported after PMC flap reconstruction,¹³ although the most were minor. However, other PMC flap studies have noted a lower overall complication rate-54%-with a much higher incidence of major complications and flap loss.¹⁴ The current study was undertaken to assess the incidence and severity of complications after postoperative adjuvant brachytherapy for advanced or recurrent squamous cell carcinoma of the head and neck.

MATERIALS AND METHODS

A retrospective, comprehensive chart review was conducted on all patients receiving iridium 192 (Ir¹⁹²) brachytherapy as a surgical adjunct for head and neck cancer at the Montefiore Medical Center, Bronx, NY, from December 1994 through December 1998. Nine patients were included in this study, the characteristics of whom are presented in Table I. All patients had squamous cell carcinoma and all initially presented with non-metastatic AJCC stage IV disease. There were six men and three women treated, aged 48 to 69 years (mean, 57.8 y). Six patients were treated as part of their primary therapy and three were treated as salvage therapy for recurrence. Treatment received by these patients is detailed in Table II.

Table TABLE I.. Patient Characteristics.

Table TABLE II.. Therapeutic Intervention and Complications.

The Ir¹⁹² afterloading catheters were placed at the time of extirpative surgery if the margin was considered to be at high risk of residual microscopic disease. Such disease was primarily related to surrounding structures that were thought to be nonresectable, i.e., skull base, carotid artery, cervical spine, and so forth. Additionally, three patients^{1, 3, 7} were treated with this afterloading technique because they had received prior external beam radiotherapy to the area for unassociated tumors. Delivery of the brachytherapy was suspended until final pathological analysis of the resection specimen was completed. If the specimen was found to have adequate margins, the catheters were removed without delivering the radiotherapy. High-dose therapy was delivered in 200-cGy fractions in the radiation oncology suite twice daily for 5 consecutive days in eight patients, and for 10 consecutive days in one (patient 7), with treatment beginning 3 to 8 days after surgery. Total doses ranged from 15 to 40 Gy (median, 20 Gy). Catheters were placed most frequently on or along the carotid artery (six of nine patients, 67%), followed by the neck (five of nine patients, 55%), skull base (two of nine patients, 22%), tongue base (two of nine patients, 22%), and vallecula or posterior pharynx (one of nine patients each, 11%). Intraoperative placement afforded precise localization of the catheters, which allowed them to be fixed in place with absorbable sutures maintaining a 1-cm distance between catheters throughout their course. They were brought out through a separate stab incision and secured to the skin with sutured and crimped metal buttons and methylmethacrylate. With the exception of tongue base implants, catheters were removed in the radiation oncology department without complication after the last scheduled treatment.

To exclude those complications that could be related to the long-term biological effects of the additional radiotherapy, early complications were defined as those occurring within 6 weeks of the surgical therapy and initial catheter placement. Complications were analyzed with respect to patient age, sex, tumor site, stage, implant location, prior radiotherapy, brachytherapy dose, and target volume. Statistical analysis was performed using Fisher's Exact Test, χ² analysis, Student t test, and a Wilcoxon 2-sample test.

RESULTS

All patients received high-dose brachytherapy in 200-cGy dosing twice daily with a 6-hour minimum interfraction interval. Of the six patients who were treated with brachytherapy for their primary, three had received prior external beam radiotherapy for distinct, separate cancers and were therefore ineligible to receive additional external radiotherapy. The remaining patients receiving brachytherapy as part of their primary treatment also received postoperative external beam radiotherapy. Early complications occurred in 55% of patients (5/9) receiving Ir¹⁹² brachytherapy after surgery. There were eight complications in these five patients, with three patients experiencing two complications (Table II). The most frequent complication was minor wound separation, occurring in three of the eight patients (38%), followed by major wound breakdown in two (25%), and in three separate patients (13%), local tissue erosion, wound infection, and bacteremia.

Wound complications were the most frequent occurrences in this population, occurring in all five patients having complications. The minor wound separation present in patient 5 occurred at the closure of the upper and lower eyelids, after maxillectomy with orbital exenteration, whereas in patient 6 it occurred on the tongue component of the suture line with the PMC flap. The cutaneous edge of the PMC flap used to reconstruct the lateral pharyngeal wall and nasopharynx in patient 7 had a 5-mm rim of epidermolysis. All three patients healed uneventfully with local wound care not requiring operative debridement. One of the two patients with major wound breakdown required a secondary procedure to debride the nonviable flap and rotate a second, contralateral PMC flap to close the defect. The initial wound separation came during an agitation episode on the evening after surgery and led to the ensuing wound care that ultimately contributed to the flap loss, given the poor healing of the surrounding tissue. Deep local recurrence was identified 3 months after surgery. The other patient with major wound breakdown, patient 8, had a sloughing of the previously irradiated cervical skin surrounding the superior aspect of the PMC flap on postoperative day 9, at the end of the brachytherapy. This was managed with local wound care, which eventually led to a stable ulceration. Two months later she was noted to have a deep recurrence at the lateral vertebral body.

The complications unrelated to wound breakdown were mild and responded to conservative management. The one case of bacteremia responded to intravenous antibiotics and correlated with a positive central venous catheter culture. The wound infection present in patient 5 occurred in the maxillectomy cavity and occurred 1 month after completion of the brachytherapy. Additionally, patient 6 had local tissue erosion of the lateral soft palate secondary to pressure from the catheter tips. This healed well after saline wound care.

The complications all occurred in patients undergoing PMC flap reconstruction. Three of the patients underwent treatment for advanced recurrent disease, and the other two had brachytherapy as part of their primary treatment. There was no relationship between flap size and wound complications. The average cutaneous flap size was 50 cm² in patients without complications, 43 cm² in those with minor complications, and 33 cm² in those two with wound dehiscence. There was a slight trend toward decreased skin paddle size in the group with complications, although this was not significant, and a large muscle flap was employed in all cases. There was no significant association between complication rate and patient age, site, prior radiotherapy, timing of therapy, number of catheters, or dosimetry.

DISCUSSION

The inclusion of brachytherapy into a comprehensive management plan for locally advanced or recurrent squamous cell carcinoma of the head and neck is well accepted. The use of this technique has translated into improved local control in patients with “inoperable” or fixed neck disease^{2, 4, 6} or those with close or microscopically positive margins.⁹ We have previously demonstrated a 2-year actuarial local control of 93% in patients implanted with I¹²⁵ seeds as a boost to treat microscopically positive margins in head and neck squamous cell carcinoma treated with primary surgery and external beam radiotherapy.⁹ This technique has also been used by Fee et al.⁶ to treat patients with tumors attached to the carotid artery to avoid sacrifice of the carotid, and they have demonstrated a 1-year local control rate of 76% using intraoperative placement of I¹²⁵ seeds. Cornes et al.⁴ used Ir¹⁹² in cases of “inoperable” neck recurrences to achieve a 1-year actuarial local control of 63%. These studies, and others, document the validity of this approach.

The 55% overall complication rate present in this series is similar to those documented in previous reports using either Ir¹⁹² or I¹²⁵ as adjuvant therapy. The current series compares favorably with that reported by Righi et al.,¹ who had a 50% overall complication rate in their patients treated with low-dose Ir¹⁹² interstitial brachytherapy for primary and salvage surgery of head and neck cancer. In our series minor wound separation and major wound breakdown occurred in 38% and 25% of the complications, respectively, whereas in their series, 50% of the complications were minor wound breakdown, with an additional 21% of complications representing soft tissue necrosis. However, only 29% of the patients with complications were being treated with salvage surgery and brachytherapy, while the remainder underwent brachytherapy of the primary without primary site surgical intervention. Park et al.⁷ report a complication rate of 36% overall, which increased to 56% after flap reconstruction, and Syed et al.⁵ report a 30% incidence of necrosis in the implant area.

The addition of flap coverage has had mixed results in the reduction of complications. Righi et al.¹ showed no change in their overall complication rate of 50% with immediate flap reconstruction and catheter coverage, whereas Park et al.⁷ had a 56% complication rate after flap reconstruction compared with 19% without. Moscoso et al.¹¹ had a 40% incidence of wound complications after PMC flap reconstruction, with a decrease to 20% after free flap reconstruction. However, although Stafford and Dearnaley² reported wound complications in all their patients treated with low-dose rate Ir¹⁹² who did not receive concomitant flap reconstruction, no patient receiving immediate flap reconstruction demonstrated local complications. This finding was corroborated by Cornes et al.,⁴ who noted late side effects of ulceration and severe fibrosis in 46% of their patients without flap reconstruction, with a decrease in both the severity and frequency (11%) of complications after immediate flap reconstruction.

Pectoralis major myocutaneous flap reconstruction has been a mainstay of head and neck surgery for more than two decades. The technique, however, is not without its own incidence of complication. Biller et al.¹³ reviewed 73 PMC flap reconstructions in the head and neck, noting an overall wound complication rate of 30%, 23% with suture line separation and 7% with flap necrosis. Additionally, Mehrhof et al.¹⁴ reported partial flap necrosis in 12%, total flap necrosis in 4%, and suture line dehiscence in 12% of their 42 cases. The overall flap complication rate in these studies, without the presence of afterloading brachytherapy catheters, was only slightly lower than the 55% present in the current study. The presence of flap reconstruction in all patients with early wound complications in this series speaks to the extensive resection and reconstruction required to treat these patients with locally advanced stage IV disease, and was also required in two of four patients (50%) who did not have complications.

Radiation technique may have contributed to the relatively high complication rate seen in this series. Reirradiation of head and neck tumors with external beam therapy alone has a 20% to 38% complication rate.³ Our regimen employed the use of a high-dose-rate afterloading technique delivering 200 cGy twice daily over a short time period, with a minimum 6-hour interfraction recovery period. Fontanesi et al.,¹⁵ using conventional low-dose iridium, have demonstrated that dose rates above 42 cGy/h may be associated with severe complications, although all patients had prior external beam radiotherapy. Prior external beam radiotherapy was given in four of six patients (67%) with complications in the current study, whereas only one of three patients (33%) without prior radiotherapy had complications.

There was no association between any patient or tumor factors and the presence of complications in this series. This finding may be related to the small number of patients treated or the lack of any homogeneous patient group. The lack of association between carotid artery implantation and any complication demonstrates this structure's relative tolerance of high doses of radiotherapy. Other reasons for the higher complication rates in this series may include the use of a rigorous definition of the term complications. The wound complications reported in patients 5 and 9 were more likely related to technical and patient care issues, respectively. The minor wound breakdown at the orbital skin edge noted in patient 5 began to demarcate before the initiation of brachytherapy and was more likely related to surgical technique. The wound breakdown noted in patient 9 occurred immediately after surgery during an agitation episode in which the patient tore apart the cutaneous wound. Excluding these complications would have lowered the overall complication rate to 44%, the minor wound separation rate to 22%, and the major wound breakdown to 11%. Additionally, and perhaps most importantly, the two cases who had major wound breakdown went on to develop recurrent disease within 3 months. It is possible that residual disease was a causative factor in the wound separation, although this was not present grossly. The use of flap reconstruction (myocutaneous in these cases), is also likely to have contributed to the healing of these patients because of it unirradiated vascular supply, and may have limited the severity of the complications.

CONCLUSION

The use of high-dose adjuvant brachytherapy for the high-risk surgical margin is a safe modality to assist in the management of advanced or recurrent squamous cell carcinoma of the head and neck. The complication rate and severity is not markedly different from that reported after flap reconstruction alone or other re-treatment techniques, and is acceptable in this high-risk patient population. Both of the major wound complications heralded local recurrence and may have been related to persistent disease, although the complication occurred early after surgery. Complications, when they occur, are minor and are controlled conservatively in most cases. The judicious use of myogenous, or myocutaneous, flaps will likely continue to maximize the use of brachytherapy while minimizing the complication severity. Ultimately, analysis of survival must be coupled with complication data to identify those patients who will attain maximum benefit without excessive morbidity.