Before we can discuss the possible treatment options for patients that have failed chemoradiation, we must understand the underlying tissue consequences of the initial treatment.

The role of chemotherapy in the treatment of head and neck oncology is well discussed in a previous chapter in this textbook. There are many agents available for treatment in the primary setting. Investigational agents have been used in Phase I and Phase II trials, and there is some question about whether, in Phase III trials, chemotherapy as a single modality treatment may be efficacious. Chemotherapy in the majority of head and neck tumors is combined with radiation therapy in a protocol known as chemoradiation. Here, the chemotherapy may be used as a sensitizing agent or as a direct combination therapy with the radiation to kill the tumor cells. Consequently, the majority of patients who present for salvage treatment, who have had chemotherapy, will have had this as a combined modality treatment with radiation therapy.

Chemotherapy has been demonstrated to not only enhance the tumor kill effectiveness of the radiation therapy but also to increase the postoperative complications in
patients who require salvage surgery. The effects of radiation therapy on the tissues have been described both from a short- and long-term toxicity perspective. Chemotherapy will adversely affect patients in the short term as well as in the long term. Patients who recur and who require salvage treatment in the close term will oftentimes be debilitated, have lost significant amounts of body weight, and still be suffering from some of the acute toxicity, such as mucositis and soft tissue necrosis. These patients will have underlying metabolic disorders and will be in a high catabolic state. Their ability to heal their wounds will be less than patients who have not undergone this treatment. Consideration of these issues should be undertaken with nutritional supplementation, and g-tube placement will need to be considered in these patients.

The effects on soft tissue of chemotherapy and chemoradiation will be greater than that of just radiation alone. The soft tissue fibrosis and long-term scarring will be more extensive. Combined with the patient’s inability to heal well, the patients who undergo salvage surgery following chemoradiation are more prone to develop local morbidities and complications from poor tissue and wound healing. Not only must the primary closure site of the ablated defect be considered, but the prophylactic reconstruction of the soft tissues that have been affected by the previous treatments must also be considered if one is to rehabilitate these patients and allow them to heal the arterial response to radiation that occurs most severely within the capillary endothelium and is the mechanism of late tissue manifestations of radiation therapy (XRT). The results of histologic changes to the capillaries lead to tissue anoxia and dilation of remaining capillaries (26,27). Larger arteries have different responses to XRT based on their sizes; small (less than 100 µm) to medium (100 to 500 µm) vessels suffer adventitial fibrosis, medial hyalinization, and intimal foam cell accumulation, whereas large arteries (greater than 500 µm) tend to be more resistant to these changes (28,29,30). Large arteries can suffer complications associated with neointimal proliferation, atheromatosis, thrombosis, and rupture, whereas the effect of chronic radiation on medium to large vessels is caused by injury to the vasa vasorum owing to a greater incidence of atherosclerosis within these vessels—a major concern in the use of microvascular free flaps (Fig. 112.1). This phenomenon is well documented in studies of irradiated carotid arteries, to which most head and neck reconstructive free flaps are anastamosed (31). Finally, the vascular physiology itself is altered by radiation therapy such that less blood flow occurs due to increased smooth muscle vasoconstriction and increased pressure in the capillary system secondary to tissue fibrosis of surrounding tissues (32).

Additional tissue-specific treatment effects of particular interest to oncologic surgeons include late toxicity to the skin, mucosa, salivary tissue, mandible, and laryngotracheal cartilages. It is well established that when exposed to radiation therapy, mucocutaneous tissue changes that include increases in vascular permeability, fibrin deposition, collagen formation, and ultimately, tissue fibrosis occur (33). Salivary tissue is exquisitely sensitive to XRT causing marked degeneration and loss of saliva. In a series of studies on XRT-related damage to the mandible, Marx (34) demonstrated that the triad of hypoxia, hypocellularity, and hypovascularity in posttreatment bone leads to tissue breakdown and chronic nonhealing wounds associated with osteoradionecrosis (ORN) of the jaw. It is thought that this process occurs in up to 56% of patients that undergo XRT to the head and neck, but it is unclear how many will manifest clinical symptoms; however, as many as half of those that become clinically apparent will require surgical interventions (35) (Fig. 112.2). A similar process is plausible in the cartilaginous structures of the
upper aerodigestive tract leading to chondroradionecrosis (CRN) with less frequency than mandible necrosis—5% to 12% of radiated laryngeal cancers (36). These latter two tissue effects can mimic cancer recurrence and thus often require a thorough workup for recurrence. In addition, the effects on bone and cartilage may be severe enough to warrant ablative surgery of the structure in question with reconstruction, even in the absence of recurrent disease, to improve pain and function.

Free tissue transfer is now accepted as the standard of care in the reconstruction of many head and neck defects, with high probability success rates (95%) (44). However, transferring free tissue to a previously treated head and neck site poses several challenges. Whether the initial treatment was surgery, radiation, chemoradiation, or a combination thereof, clear tissue effects may limit local reconstructive options as well as the ability to transfer free tissue (45).

Multiple studies document that decreased vascularity of skin results in relatively high rates of local skin flap necrosis for salvage surgery following XRT or CRT (Fig. 112.3) (10,11,22). The reconstructive surgeon must be prepared to deal with local wound complications resulting from skin changes and should counsel patients preoperatively of the potential for a protracted wound healing course. Some authors advocate the use of concomitant regional flaps for superficial coverage of important structures due to the expectation of some degree of skin loss in the heavily treated neck, particularly when single free-flap procedures will not be adequate to provide the volume of tissue necessary to reconstitute the internal defect and bolster marginal skin (19). Surgical planning is paramount to limiting skin flap loss, as prior incisions must be coordinated with planned salvage incisions in a way that limits devascularization of tenuous skin. Proper planning can limit minor skin flap loss as well as potentially lethal exposures of deeper neck structures such as the carotid arteries.

Nonetheless, free tissue transfer has proven feasible in the previously treated patient with similar rates of flap failure (15,16,17,19,22,23,24,46). In a study by Choi et al. (16) which compared patient cohorts that had no XRT to those undergoing preoperative or postoperative XRT, no difference in rate or severity of local postoperative complications was noted, but high complication rates were cited in all groups (46% to 65%). Additionally, Cohn et al. (17) examined the use of free tissue reconstruction after twice previous chemoradiation and found that wound complication was common (66%), but flap loss was not (6%). However, of the 24 patients undergoing free-flap reconstruction, four required additional free flaps for wound healing issues, and eight required additional regional flaps for the same.

In contrast, other authors have noted increased free-flap loss in addition to expected delays in wound healing and increased infections in second flap procedures (21). The largest such cohort examined results with 50 second flaps and 12 third flaps and reported 88% and 75% success, respectively, compared with 94% in primary flap surgery (45). Whether the lower success rates are related to vessel selection, flap choice, patient physiology, or other reasons remains speculative.

The emerging sentiment from these studies seems to be that even though free-flap success rates are high, the surgeon should expect some degree of minor wound complication with the potential for more severe healing problems that require second flaps. The small cohort of patients reported by Dubsky et al. (19) lends promise to the concept of double flap reconstruction; they utilized jejunumfree tissue with pectoralis major rotational flaps in the reconstruction of recurrent hypopharyngeal tumors and suffered no flap losses or wound complications requiring further surgical therapy.

The selection of recipient vessels remains a challenge in the previously treated patient; some authors report a 50% rate of contralateral vessel use when previous neck surgery had been performed, necessitating long vascular pedicles in reconstruction (47). In addition, the surgeon must be prepared to utilize out of field recipient vessels if the health of in-field vessels is suboptimal. Particularly in the previously irradiated and surgically treated neck, the internal mammary artery and transverse cervical arteries and veins should be held in reserve, and saphenous vein grafts may be necessary (17) (Fig. 112.4).

Few papers exist specifically addressing the use of free flaps in the salvage setting. Below is a summary of pertinent literature on the use of free flaps in previously treated patients. Some reference is made to larger series that include treated and treatment naïve patients in the interest of completeness.