Version May 2024
INTRODUCTION — Toxicity from cancer therapy is classified as acute or late based upon when it develops relative to treatment. Acute toxicity develops during or shortly after the completion of treatment and is usually temporary. Late toxicity presents months to years after the completion of treatment and is often permanent. The term "complication" is used for a treatment toxicity that causes an important medical problem.
This topic will review the late complications of treatment for head and neck cancer. The care of patients with head and neck cancer during initial therapy, both to treat acute toxicity and to prevent late complications, is discussed separately. (See "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Amifostine' and "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Submandibular gland transfer'.)
SALIVARY GLAND DAMAGE AND XEROSTOMIA — The most common long-term complication of radiation therapy (RT) and chemoradiotherapy for head and neck cancer is xerostomia, which is the result of damage to the salivary glands.
The magnitude of this damage is dose dependent. Parotid dysfunction can be detectable at a 10 to 15 Gy mean dose, and administration of an approximately 40 to 50 Gy mean dose to a parotid gland typically causes a >75 percent reduction in function [1].
Although xerostomia often improves with time [2], it is a long-lasting and frequently permanent problem that adversely impacts quality of life. The management of patients with established xerostomia includes multiple maneuvers to provide alternative wetting agents and maximize residual function of the salivary glands.
Prevention
Conformal radiation therapy techniques — The use of contemporary conformal RT techniques to minimize exposure of the salivary glands to radiation is the most important factor in the prevention of permanent salivary gland damage [3].
Amifostine and submandibular salivary gland transfer — The role of amifostine and submandibular salivary gland transfer in preventing late xerostomia are discussed separately. (See "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Prevention'.)
Lifestyle modification — Some degree of xerostomia is expected in almost all patients who receive head and neck RT. Significant beneficial interventions can be achieved by the patient independent of his/her clinicians. Dietary change from dry, tough food to moist, softer food can greatly improve nutritional status and quality of life. Use of a room humidifier in the bedroom can also provide some benefit [4].
Saliva substitutes and mucosal lubricants — Commercially available salivary substitutes or artificial saliva (oral rinses containing hyetellose, hyprolose, or carmellose) relieve the discomfort of xerostomia by wetting the oral mucosa.
Although these agents may provide temporary relief, many patients need frequent sips of water to remain comfortable. In addition to being inconvenient, this can lead to secondary problems, such as nocturia from late night fluid intake in men with prostatic hypertrophy and in men and women with small bladder capacity. (See "Treatment of dry mouth and other non-ocular sicca symptoms in Sjögren’s disease".)
Stimulation of existing salivary flow — Various gustatory, tactile, or pharmacologic agents are used to stimulate the flow of saliva from residual salivary gland tissue.
Nonpharmacologic stimuli — Gustatory stimuli, such as acidic or bitter substances, are most effective at stimulating salivary flow [5]. Sweet substances, such as sugar-free hard candy, also stimulate the flow of saliva but to a lesser extent. Chewing sugarless gum can provide both gustatory and tactile stimuli to salivary flow.
Pharmacologic stimuli — Parasympathomimetic agents that target the M3 muscarinic receptor are used to stimulate saliva production by residual glandular tissue. Pilocarpine has been the most extensively investigated, but cevimeline may offer advantages in terms of an improved side effects profile.
Pilocarpine — A systematic review of the literature identified nine randomized trials that studied pilocarpine in the postradiation setting [6]. The review concluded that approximately 50 percent of patients with xerostomia benefit from oral pilocarpine following RT. Continuous treatment with doses greater than 2.5 mg three times per day is required for optimal response, with some patients needing up to 12 weeks to experience benefit. Lifelong therapy is required.
Pilocarpine causes frequent side effects, most of which are the result of generalized parasympathomimetic stimulation [7]. These include sweating, headaches, tachycardia, hypertension, flushing, and increased bowel and bladder motility. Pilocarpine should be avoided in patients with uncontrolled asthma, acute angle glaucoma, or known hypersensitivity to pilocarpine.
Cevimeline — Cevimeline is an acetylcholine analog that has a high affinity for the M3 muscarinic receptors of the salivary gland but a low affinity for the M2 muscarinic receptors of heart and lung tissue.
In two identically designed trials, a total of 570 patients who had received at least 40 Gy of radiation to fields that included the major salivary glands were randomly assigned to cevimeline (30 mg three times daily) or placebo for 12 weeks [8]. Cevimeline could be increased to 45 mg per dose starting at week six. Cevimeline-treated subjects had significantly greater increases in unstimulated salivary flow but no difference in the stimulated salivary flow. There was no significant difference in the subjective evaluation of oral dryness.
Adverse effects are frequent with cevimeline, even though its side effect profile may be more favorable than that of pilocarpine. In an open-label study of long-term cevimeline use (45 mg three times daily for 52 weeks) in 255 adult patients with postirradiation xerostomia, 175 (69 percent) experienced an adverse effect, mostly mild to moderate in severity [9]. The most frequent effects were sweating (48 percent), dyspepsia (9 percent), nausea (8 percent), and diarrhea (6 percent). At the final visit, subjective improvement of dry mouth was reported by 59 percent of patients.
Acupuncture — Acupuncture is one treatment option for patients with chronic xerostomia when initial treatments are not effective or well tolerated. Although available data are limited and have mixed results, there is low risk of harm to patients who desire this therapy. Further studies are needed to confirm the duration of benefit and validate the optimal acupuncture technique.
For patients with xerostomia who have at least some residual salivary gland function, most randomized trials suggest that acupuncture provides meaningful palliation of symptoms of xerostomia (eg, dry mouth, sticky saliva, increased need for oral hydration) with minimal toxicity, compared with standard oral care [10-15]. In contrast, other studies failed to demonstrate a durable benefit, including one randomized trial that compared acupuncture-like transcutaneous electrical nerve stimulation (ALTENS) with pharmacologic stimuli such as pilocarpine [16]. However, ALTENS was better tolerated than pilocarpine, which can cause cholinergic side effects. (See 'Pilocarpine' above.)
Few studies also include a control arm that includes sham acupuncture [14], which limits interpretation of the data. In the setting of chronic pain, both acupuncture and sham acupuncture are more effective than no treatment, suggesting a strong placebo effect. (See "Subacute and chronic low back pain: Nonpharmacologic and pharmacologic treatment", section on 'Acupuncture'.)
The optimal acupuncture technique is also not established and treatment protocols are heterogeneous [17]. Although one study suggests potential benefit for combining auricular plus traditional acupuncture, the control group did not receive either intervention or a sham procedure [15].
Hyperbaric oxygen — Preliminary evidence suggests that hyperbaric oxygen (HBO) may have a beneficial effect on xerostomia, but these results must be confirmed on a larger scale before such therapy can be recommended [18,19]. As an example, a pilot study evaluated the salivary effects of HBO in a group of 80 patients, 45 of whom had hyposalivation. Patient self-assessment of xerostomia, and unstimulated and stimulated whole saliva flow rates all increased after 30 sessions of HBO [18].
DENTAL ISSUES — Dental status has a significant effect on posttreatment quality of life among patients with head and neck cancer [20,21]. Patients with head and neck cancer often have poor preexisting dentition and dental health. Poor dentition increases the risk of late complications, including osteoradionecrosis and infection. (See "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Dental issues'.)
Reduced quantity and altered composition of saliva is known to increase the risk of dental caries. Thus, patients should have a pretreatment dental evaluation that includes the initiation of fluoride treatments, and other oral and dental hygiene, which should be continued on a lifelong basis. Long-term follow-up by an experienced dentist is necessary. (See "Overview of approach to long-term survivors of head and neck cancer", section on 'Dental complications and oral health'.)
Patients who require dental extractions in a previously irradiated area of mandible or maxilla are at risk to develop osteoradionecrosis. (See 'Osteoradionecrosis and soft tissue necrosis' below.)
OSTEORADIONECROSIS AND SOFT TISSUE NECROSIS
Clinical manifestations — Osteoradionecrosis is defined as exposed, irradiated bone in the absence of recurrent or residual tumor. Osteoradionecrosis is a complication of radiation therapy (RT) due to vascular obliteration and decreased vascular supply of the irradiated tissues [22]. This leads to hypovascular areas with associated tissue hypoxia. Osteoradionecrosis is commonly precipitated by an injury (surgery, dental extractions, poor dentition, infection) to hypoxic bone tissue.
Symptoms of osteoradionecrosis can include pain, bad breath, dysgeusia, dysesthesia or anesthesia, trismus, difficulty with chewing and swallowing, speech difficulties, fistula formation, pathologic fracture, and infection. The radiologic interpretation of osteomyelitis of the mandible in a patient who has been treated with RT for head and neck cancer should be interpreted as osteomyelitis superimposed on osteoradionecrosis until proven otherwise.
In some cases, osteoradionecrosis is manifested by asymptomatic exposure of a small area of bone that remains stable for months to years and heals with conservative management. In other cases, osteoradionecrosis may gradually progress and be associated with skin fistulas or infections. Severe necrosis may require surgical intervention and reconstruction [23,24].
The time to onset of osteoradionecrosis is quite variable. In some cases, it may be diagnosed shortly after completion of RT, while in other patients it may not be diagnosed for years after the original cancer treatment [25,26].
The mandible is the most frequently affected bone, because in many patients treated for head and neck cancer, a large portion of the mandible is exposed to high doses of radiation [25]. Maxillary osteoradionecrosis is rare and seen most often in the setting of irradiation for nasopharyngeal cancer [27]. The vascular supply to the mandible is not as robust as the maxilla, which also predisposes the mandible to injury.
Epidemiology and risk factors — Factors that affect the risk of developing osteoradionecrosis include tumor location and size, radiation technique and dose, and acute and chronic trauma to the mandible and/or maxilla. Smoking is also associated with an increased risk of osteoradionecrosis [26].
RT to oral cavity targets generally administers higher doses of radiation to larger volumes of the mandible than treatment of targets in the pharynx and larynx. As a consequence, osteoradionecrosis is most common in patients treated for tumors of the oral cavity [24].
The incidence of osteoradionecrosis ranges from 3 to 7 percent regardless of treatment technique, including conventional RT, intensity-modulated radiation therapy (IMRT), chemoradiotherapy (generally not using IMRT), or brachytherapy (with or without an external beam component) [23,28]. One observational series of fast neutron therapy for targets above the clavicles demonstrated a relatively low rate of osteoradionecrosis of the mandible or facial bones (crude incidence rate of 2 percent) [29]. The crude incidence of osteoradionecrosis with reirradiation is only slightly higher than that seen during the initial course of RT likely secondary to small expansions in the reirradiation setting [30,31].
However, the significance of RT technique should be emphasized. When using IMRT, failure to limit "hot spots" in the mandible can deposit excess dose and lead to osteoradionecrosis [32-34]. Meticulous attention to minimizing the volume of mandible receiving high doses of RT has the potential to minimize the risk [35]. When using brachytherapy, differences in the relative contribution of the external beam and brachytherapy components of treatment can cause widely different crude incidence rates of osteoradionecrosis [36]. Finally, these comprehensive analyses include relatively few patients treated with IMRT and concurrent chemotherapy, the current standard of care for advanced head and neck cancer.
Unnecessary trauma to the irradiated mandible/maxilla is to be avoided. This is at times difficult; a comprehensive review of dental disease in patients undergoing cancer therapy found that post-RT patients had the highest incidence of decayed, missing, and filled teeth of all cancer patients surveyed [37]. The association of osteoradionecrosis with dental factors has led to the recommendations for dental management during and after the initial treatment of patients with head and neck cancer. (See "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Dental issues'.)
Treatment — The management of osteoradionecrosis and soft tissue necrosis is complex. For mild cases, conservative debridement plus antibiotics is usually successful [38]. However, when bone and soft tissue necrosis are extensive, resection of the mandible with immediate microvascular reconstruction may provide better results [39-41]. Recurrent cancer must always be considered in the differential diagnosis of exposed mandible or maxilla especially if the exposure is in the same general anatomic area as the previously treated tumor.
Hyperbaric oxygen — Many oral surgeons and radiation oncologists prescribe hyperbaric oxygen (HBO) to prevent and/or treat osteoradionecrosis in patients with previous significant RT exposure to the mandible. Although data on its efficacy are conflicting (with some studies suggesting limited benefit), we offer HBO to these patients on a case-by-case basis to avoid or minimize the debilitating long-term toxicities of osteoradionecrosis [42,43]. Additionally, while one randomized trial evaluating preventative HBO reported a relatively low incidence of osteoradionecrosis with short-term follow-up (approximately 6 percent at six months) [44], we have observed higher clinical rates of osteoradionecrosis, particularly with longer follow-up.
For prevention of osteoradionecrosis, we offer HBO to patients with no evidence of local tumor recurrence who are undergoing dental interventions to the mandible and who have risk factors for developing osteoradionecrosis (eg, ≥50 Gy RT exposure to the mandible, exposed bone and/or necrosis postoperatively). For treatment of osteoradionecrosis, we offer HBO to patients with ≥50 Gy RT exposure to the mandible and no response to more conservative measures, such as debridement and/or antibiotics. Our decision as to whether to use HBO also takes into account previous cancer site (ie, those with oral cavity cancers or tonsillar cancers ipsilateral to the mandibular molars being treated), oral mucosal health, patient performance status, and invasiveness of the planned dental intervention.
HBO is thought to act by stimulating monocyte and fibroblast function, thereby increasing vascular density and improving circulation [45]. HBO is typically administered in a wound care center by a clinician certified in hyperbaric medicine. HBO is expensive and time consuming. Treatment typically requires 30 to 40 sessions ("dives") delivered once or twice daily, generally for five days a week and thus lasting up to eight weeks [46]. The most common side effect is reversible myopia.
Available data on the efficacy of preventative or therapeutic HBO are as follows:
●Prevention of osteoradionecrosis – In patients undergoing interventions to the mandible (eg, dental repair, tooth extractions, and/or implants) who have a previous history of RT, HBO may be administered perioperatively in an effort to prevent osteoradionecrosis.
In observational studies and a randomized trial, the incidence of osteoradionecrosis could be reduced using this technique [47,48]. In a seminal trial, 74 patients previously irradiated with a two-dimensional technique who needed tooth extractions were randomly assigned to perioperative antibiotics (penicillin) versus perioperative antibiotics plus HBO therapy [47]. The addition of HBO to antibiotics decreased the incidence of osteoradionecrosis (30 versus 5 percent). However, it should be noted that this trial was performed with less contemporary RT techniques and prior to the use of IMRT, a technique that is effective at limiting RT exposure to the mandible.
In contrast, subsequent studies have suggested a lower baseline rate of osteoradionecrosis in this population and a limited benefit from HBO in the prevention of osteoradionecrosis. As an example [44,49], in a randomized phase III trial (Hyperbaric Oxygen for the Prevention of Osteoradionecrosis [HOPON]) of 144 portions of mandibles previously irradiated to ≥50 Gy with more-contemporary techniques and undergoing mandible extractions or implants, the addition of perioperative HBO to standard supportive care (eg, chlorhexidine mouthwashes and antibiotics) did not reduce the incidence of jaw osteoradionecrosis (6.4 versus 5.7 percent) at six-month follow-up [44]. Although HBO reduced acute postoperative symptoms (eg, pain, swelling, bleeding, and ability to open the mouth) compared with supportive care, it did not improve longer-term outcomes (eg, chronic treatment-related pain or quality of life).
●Treatment of osteoradionecrosis – The use of HBO for treatment refers to the situation where certain clinical symptoms (pain, dysesthesia in the distribution of the inferior alveolar nerve, areas of bone exposure, trismus, fistula), radiographic findings (increased density, periosteal thickening, diffuse radiolucency, mottled areas of osteoporosis, sclerosis sequestration), or physical examination findings (exposed bone) consistent with osteoradionecrosis are present.
A randomized trial of the efficacy of HBO in cases of overt osteoradionecrosis was closed prematurely secondary to a trend suggesting that HBO was associated with worse outcomes than placebo [50]. It is unclear why the placebo arm of this trial performed much better than expected (10 percent recovery expected, 32 percent recovery recorded), although there were some imbalances in the randomization of the two treatment arms. By contrast, a retrospective study using a questionnaire administered to patients after completion of hyperbaric therapy reported that 75 percent of patients noted an improvement in their principal presenting symptom [51].
Pentoxifylline and vitamin E — A single institution phase II study using pentoxifylline and vitamin E in patients with refractory osteoradionecrosis resulted in all 54 patients experiencing a complete recovery in a median of nine months [52]. This finding needs to be confirmed in a randomized trial. (See "Clinical manifestations, prevention, and treatment of radiation-induced fibrosis", section on 'Pentoxifylline plus tocopherol'.)
LYMPHEDEMA AND FIBROSIS — Lymphedema and fibrosis are underappreciated long-term effects from head and neck cancer and its therapy. Lymphedema can be categorized as either internal or external; internal and external lymphedema are not mutually exclusive.
●Internal lymphedema involves the mucosa and underlying soft tissue. It may result in hoarseness, airway compromise, and dysphagia.
●External lymphedema involves tissues of the face, neck, and shoulders. It may result in swelling, tightness, and decreased range of motion with associated decreased function and discomfort.
It is likely that over half of patients who are treated for head and neck cancer (either with surgery, radiation, or both) will experience some component of lymphedema.
Development of both internal and external lymphedema is multifactorial. Factors associated with the development of internal lymphedema include surgery, radiation, and increasing number of treatment modalities utilized. Factors associated with the development of external lymphedema included location of primary tumor, duration since completion of head and neck cancer treatment, and increasing number of treatment modalities utilized. Total dose of radiation and increasing treatment time have also been associated with combined lymphedema [53].
Lymphedema correlated with symptom burden, functional status, and long-term quality of life in patients treated for head and neck cancer [54]. In severe cases, in particular when accompanied by neck dissection, or in the case of reirradiation, neck and facial edema may be extremely severe and cosmetically significant.
Over 50 percent of patients respond to head and neck cancer-specific complete decongestive therapy, using techniques that are distinct from the management of lymphedema in other body sites. Treatment adherence to recommended regimens improves the response rate [55]. (See "Clinical staging and conservative management of peripheral lymphedema", section on 'Complete decongestive therapy'.)
Surgery to the neck can also result in hardening of the tissues of the neck or face, which can cause facial, submental, or, rarely, cerebral edema due to venous and lymphatic compromise [56]. This is generally worse with bilateral neck dissection, with neck dissections that sacrifice the jugular vein, or when neck dissection follows or precedes RT. Contractures or loss of normal range of motion of the neck can occur, especially if neck dissection is combined with RT.
Treatments that have been reported to be useful for neck stiffness/soft tissue fibrosis in the neck include pentoxifylline [57,58], massage therapy/physical therapy [59], and botulinum toxin injections, particularly in setting of neck muscle spasm [60].
CAROTID ARTERY INJURY
Stroke — Ischemic stroke can be a late complication of neck irradiation [61-63]. In a Canadian cancer registry study of 14,069 patients treated for cure between 1990 and 2010, 6 percent had an ischemic stroke [63]. Compared with patients treated with surgery alone, the risk was significantly greater in those whose treatment incorporated any radiation therapy (RT; adjusted hazard ratio 1.46, 95% CI 1.23-1.73, increase from 5 to 6 percent). Another retrospective series of 367 patients who were treated with RT before the age of 60 years found that the cumulative risk of stroke was 12 percent at 15 years and that the relative risk of an ischemic stroke was significantly increased compared with the general population [64].
Multiple factors may contribute to this increased risk in patients with head and neck cancer, including carotid artery stenosis (CAS) [65] and increased deposition of plaque [66], as well as other preexisting risk factors for cerebrovascular disease, particularly smoking. (See "Stroke: Etiology, classification, and epidemiology".)
There is currently no standard screening mechanism for survivors. Screening ultrasound within the first 12 to 18 months since completion of RT, followed by repeat ultrasounds, demonstrated CAS (defined as >50 percent occlusion, stroke, or transient ischemic attack) in 29 percent of patients at eight years. On multivariate analysis, diabetes was predictive of time to CAS, but no radiation dose parameters were found to be significantly associated with development of CAS. It is unclear what percentage of patients had stenosis prior to the initiation of therapy [67,68].
Carotid artery rupture — Carotid artery rupture (sometimes called "carotid blowout syndrome") is a medical emergency. Mortality is high, and emergency surgery is indicated [69]. Endovascular stenting may have a role in managing patients with impending carotid artery rupture, in preference to surgical ligation [70].
Rupture of the carotid artery is most common in the setting of a surgically exposed carotid artery with limited residual normal tissue to cover it, recurrent tumor, infection, or severe radiation damage [71]. Carotid blowout syndrome usually occurs proximal to the carotid bifurcation and is commonly associated with soft tissue necrosis in the neck and mucocutaneous fistulas. In a case series and review of the literature that identified 140 patients with carotid artery events, 60 percent of patients had life-threatening hemorrhage that required emergency intervention [70]. A review of the incidence of carotid blowout in the setting of reirradiation suggested that the syndrome was more common in accelerated fractionation reirradiation techniques [72].
TRISMUS — Trismus is a condition characterized by limited ability to open the jaw, which is generally caused by a combination of spasm, fibrosis, and contraction of the muscles responsible for movement at the temporomandibular joint [73]. The inability to fully open the jaw has a variety of functional consequences, including impaired nutrition, speech, and oral hygiene.
Some limitation of maximal mouth opening is seen in almost all patients treated for oral cavity and oropharyngeal cancers, regardless of the treatment modality [74]. However, the changes are more severe in those managed with RT or surgery plus RT rather than surgery alone. Although there is no absolute interincisor distance that defines trismus, distances of 20 to 40 mm have been suggested as indicative of trismus [75].
The incidence of trismus has varied widely in patients who received radiation therapy (RT) for head and neck cancer. A systematic review of the literature found that the mean prevalence of trismus was approximately 25 percent using older RT techniques [73]. However, newer techniques that minimize the dose of radiation to the muscles of mastication appear to significantly reduce the incidence of trismus. At least three reports using intensity-modulated radiation therapy (IMRT) have found that the incidence of trismus was approximately 5 percent [76-78]. Associated surgery can also significantly affect the incidence of trismus, both from the standpoint of primary site resection and surgical approaches such as mandibular split procedures. (See "General principles of radiation therapy for head and neck cancer", section on 'Three-dimensional conformal RT'.)
Early treatment is important to prevent the development of severe, irreversible contractures. Although evidence is limited, the use of jaw exercises, passive motion devices, and splinting may be useful [79-82], and these can be used in the early postoperative period [83]. (See "Physical rehabilitation for cancer survivors", section on 'Trismus'.)
Pentoxifylline has been used to treat radiation-induced fibrosis; at least one preliminary study suggests that this may have a modest benefit in patients with trismus [57]. Injection of botulinum toxin directly into the masseter muscles may decrease pain and muscle spasm, but has not resulted in improvement of trismus [60]. (See "Clinical manifestations, prevention, and treatment of radiation-induced fibrosis".)
DYSPHAGIA — Treatment-induced dysphagia can compromise the benefits of organ preservation. In addition to the effects of radiation, other causative factors including xerostomia, surgical disruption, and concurrent chemotherapy [84,85]. The locally destructive effects of the primary tumor prior to treatment can also contribute to dysphagia. The dose dependence of radiation therapy (RT) related dysfunction and its avoidance is complicated by need to deliver high doses radiation to eradicate tumor. When control of the tumor is not compromised, the radiation plan should be contoured to avoid the pharyngeal constrictor muscles.
There are several methods by which to evaluate treatment related dysphagia: instrumental assessment (modified barium swallow, manometry, fiberoptic endoscopic evaluation of swallowing), subjective assessment (Common Terminology Criteria for Adverse Events [CTCAE] grading), and patient reported quality of life forms (Functional Assessment of Cancer Therapy Head and Neck [FACT H&N], etc). The MD Anderson Dysphagia Index (MDADI) has been validated and is a useful tool for assessment [86,87]. It is not uncommon for different instruments to provide contradictory results for the same patient, making accurate comparisons and estimates of dysphagia incidence subject to some degree of uncertainty.
A meta-analysis of three concurrent chemotherapy trials (RTOG 91-11, RTOG 97-03, RTOG 99-14) found that feeding tube use two or more years after RT (>10 percent of evaluable patients) and pharyngeal dysfunction (27 percent of patients evaluable) were common. Older age, advanced T-stage, larynx/hypopharynx primary, and posttreatment neck dissection were found to be predictive of severe late toxicity [88].
RT technique can have an important effect on the incidence and severity of dysphagia:
●A retrospective comparison of patients treated with either whole neck intensity-modulated radiation therapy (IMRT) (entire elective lymphatic volume treated by an inverse planned IMRT technique) or split field IMRT (neck lymphatics below the thyroid notch treated by an anterior low neck field) demonstrated an increased rate of grade 3 dysphagia for patients treated with whole neck IMRT (95 versus 62.5 percent, p = 0.06). This was thought to be secondary to inadequate pharyngeal axis radiation avoidance in the whole neck IMRT group (mean dose: 55.2 versus 27.2 Gy, p <0.001) [89].
●A retrospective review of 96 patients demonstrated a 37 percent rate of pharyngeal stricture following treatment. A mean radiation dose of greater than 50 Gy to the larynx and inferior pharyngeal constrictor was found to correlate with stricture and aspiration [84].
●A proposed method to avoid RT-dose-related dysphagia is to enter the uninvolved pharyngeal constrictor muscles and larynx into the IMRT cost function as avoidance structures. This was evaluated in a prospective trial of oropharynx cancer patients that demonstrated that all measures of dysphagia worsened soon after therapy; however, the observer and patient reported scores recovered after time. After one year, only one patient was feeding tube dependent and three required a soft diet. Locoregional control was not affected by this technique (96 percent at three years) [90].
A retrospective analysis suggested that prophylactic placement of a gastrostomy tube results in both a higher percentage of feeding-tube dependent patients at one year posttreatment (21 versus 0 percent, p <0.001) and a higher rate of esophageal stricture (30 versus 6 percent), perhaps due to disuse of pharyngeal musculature during therapy. This needs to be confirmed prospectively. Placement of a feeding tube is at the discretion of the clinician [91].
Newer surgical techniques followed by postoperative RT have been postulated to cause less treatment-related dysphagia than combinations of RT and systemic therapy [92]. However, this needs to be prospectively evaluated since much larger analyses that suggested that severe complications are much more common in patients managed with primary surgery [93].
Speech pathologists are a vital part of the head and neck cancer treatment team and should evaluate all patients receiving surgery and/or high-dose RT that could potentially interfere with the ability to protect the airway or swallow. Neuromuscular electrical stimulation (NMSE) is a potentially promising improvement to traditional therapy [88]. (See "Speech and swallowing rehabilitation of the patient with head and neck cancer".)
Dysphagia can occur many years after the completion of therapy, even in patients who did not have dysphagia during treatment or in the immediate aftermath. This seems to be more common in patients treated with more intense therapy (ie, concurrent chemotherapy) [94], although it is often difficult to predict which patients will develop late swallowing complications and at what interval. In a small subset of patients, late swallowing complications may be related to cranial neuropathy [95]. In addition, many patients do not appreciate dysphagia secondary to adaptation over time and pharyngeal/laryngeal sensory neuropathy [96,97]. Patient-reported outcome questionnaires may be superior to detecting such swallowing dysfunction compared with patient perception [98], but this is not clear.
ESOPHAGEAL TOXICITY — Esophageal toxicity appears to be multifactorial in patients treated for head and neck cancer. Factors that may contribute include mucositis from RT or chemoradiotherapy, fibrosis in muscle, or changes due to alterations in bacterial flora or change in pH due to xerostomia [99]. Factors that appear to be particularly important in determining the frequency and severity of esophageal toxicity include the use of concurrent chemotherapy, duration of acute effects, dose of radiation, and primary tumors that are in close proximity to the esophageal verge (eg, laryngeal and hypopharyngeal primaries):
●In one series, the incidence of grade 3 or greater toxicity (able to swallow only liquids, dilatation required) was analyzed in 211 patients six months after treatment [100]. Toxicity was significantly more frequent in those treated with concurrent chemotherapy compared with RT alone (33 of 118 [30 percent] versus 7 of 93 [8 percent]). By contrast, multiple prospective studies evaluating concurrent chemoradiotherapy versus RT alone reported similar swallowing/esophageal toxicity in both arms [85,101,102]. The true influence of treatment intensification (ie, concurrent chemotherapy) is at times difficult to quantify as swallowing difficulty can be a late complication developing years after therapy.
●A retrospective analysis of patients treated with hyperfractionated RT and concurrent cisplatin and fluorouracil (FU) demonstrated on multivariate analysis that the duration of ≥grade 2 mucositis was predictive (p <0.01) for stricture [103].
●In another series, 24 patients with head and neck squamous cancer of unknown primary were treated with IMRT to the neck, including concurrent chemotherapy in 22 (92 percent) [104]. The median dose to the esophageal mucosa was 60 Gy. With a median follow-up of two years, 11 patients (46 percent) had required esophageal dilatation for a stricture.
●A retrospective, case control series compared 70 patients with head and neck cancer who developed an esophageal stricture following RT without concurrent chemotherapy with 66 matched controls treated in the same way [105]. The 70 patients with strictures represented approximately 3 percent of all patients treated with RT during the period of the study. Factors associated with a significantly increased risk of esophageal stricture included the need for a percutaneous endoscopic gastrostomy (PEG) tube and/or nasogastric tube during treatment, and the use of a radiation dose greater than 45 Gy to the upper esophagus.
●Advanced tumors of the hypopharynx and larynx that require surgical manipulation of the esophageal verge have increased rates of esophageal toxicity with surgery alone [106] and with surgery followed by postoperative RT [107].
Less severe esophageal toxicity may be even more frequent. In a prospective series, 100 patients who had finished all treatment at least three months earlier underwent transnasal esophagoscopy as part of their routine surveillance [99]. Overall, some pathology was identified in 87 percent of cases, including peptic esophagitis and stricture in 63 and 23 percent of cases, respectively.
Patients with esophageal stricture generally can be managed with dilatation, although frequent repeat procedures may be required, and it is not always effective. (See "Endoscopic interventions for nonmalignant esophageal strictures in adults".)
The management of swallowing disorders is discussed in detail separately. (See "Swallowing disorders and aspiration in palliative care: Assessment and strategies for management".)
THYROID DISEASE — The incidence of thyroid disease following radiation therapy (RT) for head and neck cancer varies widely but appears to be dose-dependent. In a systematic review of the literature that included 971 patients in five studies, most of whom had been radiated for head and neck cancer, the incidence of subclinical and clinical hypothyroidism varied from 26 to 53 and 11 to 33 percent, respectively [108]. (See "Diagnosis of and screening for hypothyroidism in nonpregnant adults" and "Treatment of primary hypothyroidism in adults".)
Irradiation of the low neck can cause direct damage to the thyroid gland resulting in primary hypothyroidism. This is usually detected by an elevation in serum TSH and is clinically significant in only a minority of cases. Cases of hypothyroidism that are caused by damage to the hypothalamus or pituitary region (central hypothyroidism) are significantly less common [109]. (See "Central hypothyroidism".)
RT-induced hypothyroidism develops at a median of 1.4 to 1.8 years after therapy (range 0.3 to 7.2 years) [110,111]. It is more common in patients undergoing both neck surgery and RT. Some reports have observed a higher rate of hypothyroidism in patients treated using intensity-modulated radiation therapy (IMRT); however, this increased risk can be avoided with appropriate technique that minimizes the dose of radiation to the thyroid gland [112]. The incidence of hypothyroidism does not appear to be higher in patients who are managed with chemoradiotherapy compared with RT alone. (See "Disorders that cause hypothyroidism", section on 'Iatrogenic disease'.)
For patients treated with neck RT, emphasis is generally placed on posttreatment screening and thyroid hormone replacement rather than prevention. A single institution analysis of pediatric patients suggested that TSH suppression during RT had a "protective effect" on thyroid function [113], but this needs to be confirmed on a larger scale before it can be recommended. Currently expert groups recommend that serum TSH should be checked within 12 months of completing therapy and repeated every 6 to 12 months [114,115].
MYELITIS — Radiation damage to the spinal cord is one of the most feared complications of the treatment of cancer with radiation therapy (RT). As such, cooperative group trials generally prioritize spinal cord avoidance over target coverage in intensity-modulated radiation therapy (IMRT) planning.
The syndrome of radiation myelitis usually begins 9 to 15 months after the end of RT with paresthesias and other sensory disturbances. It then progresses steadily to involve motor signs. It is fatal in approximately 50 percent of cases [116]. In an analysis of 1112 patients who were treated prior to CT-based planning, the incidence of radiation-induced myelitis for cord doses <55 Gy was 0.18 percent [117]. Current CT based planning techniques, where dose is limited to extremely small volumes, are expected to cause even lower incidences of this feared complication. There is no successful treatment. (See "Complications of spinal cord irradiation", section on 'Late radiation-induced myelopathy'.)
A transient phenomenon related to both cervical cord irradiation and platinum containing chemotherapy drugs such as cisplatin is Lhermitte sign, known as the "barber's chair sign." Lhermitte sign is characterized by an electric shock sensation that radiates down the spine and extremities on flexion of the neck. It rarely progresses to radiation myelopathy and is typically self-limited in nature. The typical latency is two to four months after the completion of RT. No specific treatment exists, although a soft cervical collar can reduce the flexion of the cervical spine and reduce symptoms [118]. (See "Complications of spinal cord irradiation", section on 'Early radiation-induced myelopathy' and "Overview of neurologic complications of platinum-based chemotherapy", section on 'Clinical and electrophysiologic manifestations'.)
RETINOPATHY AND OPTIC NEUROPATHY — Radiation-induced optic neuropathy is caused by radiation-induced ischemia to the optic nerve. It is a primary concern for treatment of tumors of the nasopharynx, nasal cavity, and paranasal sinuses that frequently require doses greater than 54 Gy to the optic apparatus (optic nerves, retina, optic chiasm). A sophisticated analysis of radiation parameters affecting radiation-induced optic neuropathy demonstrated that the incidence increases significantly with doses greater than 50 Gy. Protective measures that seem to reduce the incidence are daily fraction sizes ≤1.8 Gy or twice daily fractionation, which seems to be the most protective, particularly at high (eg, ≥ 60 Gy) doses [119]. In some instances, treatment of paranasal sinus tumors will require that portions of the retina exceed this threshold. However, CT-based treatment planning in these cases may allow dose-sparing of the macula to preserve central vision. The effect of concurrent chemotherapy is unclear, as its use was rare in this analysis. (See "Delayed complications of cranial irradiation", section on 'Ototoxicity'.)
A number of therapeutic options for radiation-induced optic neuropathy have been tried. These include corticosteroids, hyperbaric oxygen (HBO), and optic nerve sheath fenestration. It appears that corticosteroids and HBO work infrequently [120] and that optic nerve fenestration not only does not work but may actually be harmful [121].
OTOTOXICITY — Both ionizing radiation and cisplatin can cause sensorineural hearing loss regardless of other comorbidities. When combined, toxicity can be additive [122].
The inner ear apparatus lies in the petrous portion of the temporal bone and should be contoured as an avoidance structure for radiation therapy (RT) planning [119]. A detailed radiation dosimetric analysis demonstrated that the incidence of sensorineural hearing loss was related to radiation dose [123]. The cochlea is such a small structure that it is generally evaluated as a mean dose function; mean dose less than 45 Gy is typical in cooperative group trials. (See "Delayed complications of cranial irradiation", section on 'Ototoxicity'.)
The most active chemotherapy agent used in the head and neck cancer, cisplatin, is associated with significant ototoxicity independent of concurrent RT. Cisplatin ototoxicity is dependent on both cisplatin dose and scheduling. (See "Overview of neurologic complications of platinum-based chemotherapy", section on 'Ototoxicity'.)
The combination of cisplatin and radiation lowers the radiation dose that causes sensorineural hearing loss. In a retrospective analysis, patients treated with radiation alone could receive up to 40 Gy to the inner ear without measurable hearing impairment. By contrast, concomitant cisplatin caused detectable hearing loss at doses as low as 10 Gy. High dose cisplatin (100 mg/m2 every three weeks) was noted to cause significantly more ototoxicity than low dose weekly cisplatin (40 mg/m2 weekly) [124]. A more recent analysis has proposed that inner ear RT constraints should be based upon the number of cycles of cisplatin received [125].
The risk of ototoxicity is highest in patients with nasopharynx cancer and parotid cancer when the facial intratemporal portion of the facial nerve requires treatment. The cochlea and vestibular apparatus should be identified as organs at risk.
Carboplatin is a platinum containing compound that has been used concurrently with head and neck cancer. It is the standard of care for concurrent therapy after induction chemotherapy with TPF (docetaxel, cisplatin, fluorouracil [FU]). Carboplatin is associated with less ototoxicity than cisplatin [126].
Treatment induced ototoxicity can range from tinnitus to sensorineural hearing loss. Its treatment is case specific. (See "Evaluation of hearing loss in adults" and "Treatment of tinnitus".)
LACRIMAL GLAND — Dry eye syndrome is a disorder that results from deficiencies or defects in the components of lubrication that can lead to structural and functional abnormalities of the ocular surface. Mild to moderate dry eye syndrome responds to conservative measures, but severe dry eye syndrome can cause a multitude of symptoms including compromised vision, severe pain, and eye loss. Therapeutic radiation therapy (RT) doses to the entire globe can cause severe dry eye syndrome. It appears this is dose dependent. An analysis demonstrated that a mean dose <34 Gy to the lacrimal gland limits the incidence of severe dry eye syndrome to less than 5 percent [127]. (See "Delayed complications of cranial irradiation", section on 'Xerophthalmia'.)
DYSPHONIA — Dysphonia can be caused by treatment of head and neck cancer when the larynx is targeted by radiation therapy (RT) and/or surgical instruments (ie, laryngeal cancer) and when the larynx is in close proximity to targets (ie, neck lymphatics).
Dysphonia after treatment for laryngeal cancer depends on the modality used for primary treatment. Patients with early, superficial T1a larynx cancers have comparable voice quality following treatment with either surgery or RT. Tumors with more than superficial depth of invasion likely have a better functional outcome after primary RT compared with those treated with surgery [128]. Patients with advanced laryngeal cancer seem to have low rates of moderate speech impairment two years after completion of organ preservation therapy [85]. Patients who note voice impairment after treatment for early glottic carcinoma benefit from voice therapy [129].
Patients receiving either elective or therapeutic nodal irradiation should have the larynx contoured and avoided, as toxicity is dose-related [130]. With careful treatment planning, the ability to limit mean dose to the larynx is similar among different RT techniques [131].
As with all these symptoms, new or worsening dysphonia should raise suspicion for recurrent disease or a second primary cancer of the upper aerodigestive tract.
SMOKING CESSATION — All patients with head and neck cancer should receive counseling about the importance of smoking cessation. Besides being a major risk factor, smoking can also influence cancer prognosis. When smoking is continued during and after radiation therapy (RT), it can increase the severity and duration of mucosal reactions, exacerbate xerostomia, and compromise oncologic outcome [132-134]. (See "Epidemiology and risk factors for head and neck cancer".)
One study, for example, evaluated 115 patients who were treated with RT with or without fluorouracil (FU) [133]. The 53 patients who continued to smoke during RT had significantly lower rates of complete response (45 versus 74 percent) and two year survival (39 versus 66 percent) compared with the 62 patients who did not smoke or who had quit before treatment. Among nonsmoking patients, mortality rate was influenced by the duration since quitting smoking. Compared with patients who continued to smoke, the death rate was 40 percent lower in patients who had quit less than 12 weeks before and 70 percent lower in patients who had quit more than one year before diagnosis.
Although the epidemiology of head and neck cancer is changing with the emergence of HPV associated oropharyngeal cancer, the importance of smoking cessation has not changed. In an unplanned analysis of RTOG 01-29, the mortality in the HPV positive group was higher for smokers compared with nonsmokers [135]. (See "Epidemiology, staging, and clinical presentation of human papillomavirus associated head and neck cancer".)
POSTTREATMENT DEATH — Death from non-cancer causes is an important event in head and neck cancer. Improved disease control and increased toxicity have made competing mortality an increasingly important concept in both head and neck cancer treatment and protocol development. An evaluation of almost 500 patients treated with organ preserving chemotherapy, radiation therapy, and surgery on consecutive prospective clinical trials demonstrated a five-year non-cancer cumulative competing mortality incidence of 19.6 percent [136]. A similar study of competing mortality using the Surveillance, Epidemiology, and End Results (SEER) database demonstrated similar mortality rates secondary to cancer recurrence and competing risks (23.8 and 27.6 percent at five years, respectively) [137]. Attempts were made in both analyses to determine factors predictive of competing mortality; only increasing age was found to be significant in both analyses.
It is important to understand issues surrounding competing mortality in modern head and neck cancer therapy. Predictive factors have been suggested. Analysis of competing risks may become prominent in clinical trial development and risk stratification.
SECOND MALIGNANCIES — Patients with head and neck cancer are at increased risk for the development of second malignancies, both in the head and neck and elsewhere. In large part this is due to the same risk factors associated with the initial malignancy (tobacco, alcohol, human papillomavirus). In addition, radiation therapy (RT) used to treat the primary tumor may contribute to an increased risk. (See "Second primary malignancies in patients with head and neck cancers", section on 'Risk factors'.)
REHABILITATION
Speech and swallowing — Impairments of speech and swallowing have a serious impact on survivor quality of life. Functional rehabilitation is an important component of patient management and can be essential for an optimal outcome following treatment.
These issues are discussed separately:
●(See "Alaryngeal speech rehabilitation".)
●(See "Speech and swallowing rehabilitation of the patient with head and neck cancer".)
Musculoskeletal issues — A variety of musculoskeletal abnormalities occur as late complications in head and neck cancer survivors. These include neck muscle spasm, which can vary in severity from mild to debilitating; muscle atrophy; neck weakness; myofascial pain syndrome; and impaired shoulder function. Rehabilitation issues for these issues are discussed separately. (See "Physical rehabilitation for cancer survivors".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Head and neck cancer".)
SUMMARY AND RECOMMENDATIONS — The treatment of head and neck cancers can result in a wide range of late toxicities that have a significant impact on the patient's quality of life. Potential approaches to preventing the development of such toxicities that are applicable during the initial treatment are discussed separately. (See "Management and prevention of complications during initial treatment of head and neck cancer".)
●Xerostomia – The most common long-term complication of radiation therapy (RT) and chemoradiotherapy for head and neck cancer is xerostomia, which is the result of damage to the salivary glands. The use of contemporary conformal RT techniques to minimize exposure of the salivary glands to radiation is the most important factor in the prevention of permanent salivary gland damage. (See "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Prevention'.)
For patients with symptomatic xerostomia, treatment approaches include lifestyle modification (moist, soft foods; humidifiers), saliva substitutes and nonpharmacologic salivary gland stimulants, and pharmacologic stimulation with parasympathomimetic agents (pilocarpine, cevimeline). Intravenous amifostine, though not often used, may decrease the risk of permanent xerostomia. Acupuncture is also a treatment option for patients with chronic xerostomia when initial therapies are not effective or well tolerated, although data are limited and have mixed results. (See 'Salivary gland damage and xerostomia' above.)
●Osteoradionecrosis – Osteoradionecrosis is defined as the presence of a nonhealing area of exposed necrotic bone after RT in the absence of recurrent or residual tumor. The primary risk factors associated with the development of osteoradionecrosis are dental extractions and surgery. Compromised teeth in an area that will receive at least 50 Gy and teeth that are outside of the RT treatment field but have a hopeless prognosis should be extracted prior to the initiation of therapy. There is no indication to extract healthy teeth. After therapy, routine dental care and supplemental fluoride are recommended.
•For mild cases of osteoradionecrosis, treatment with conservative debridement and antibiotics is usually sufficient. For more advanced cases of bone and soft tissue necrosis, extensive resection of the mandible with immediate microvascular reconstruction may provide better results. (See 'Osteoradionecrosis and soft tissue necrosis' above.)
•We offer hyperbaric oxygen (HBO) to select patients for prevention and treatment of osteoradionecrosis. (See 'Hyperbaric oxygen' above.)
●Dysphagia and esophageal toxicity – Dysphagia and esophageal toxicity are multifactorial complications of head and neck cancer therapy. Contemporary prospective trials demonstrate long-term gastrostomy tube rates of 10 percent; however, the true incidence is likely underreported as the development of this toxicity can be many years after treatment and insidious in nature. Careful attention to minimizing radiation to pharyngeal constrictors appears to be helpful in minimizing these complications. (See 'Dysphagia' above and 'Esophageal toxicity' above.)
●Other toxicities – There is a wide range of other toxicities that can be seen depending on the specific tissues irradiated and the dose of radiation received. These can include injury to soft tissues of the head and neck (trismus, fibrosis in the neck), hypothyroidism, carotid artery injury, stroke, various forms of neurologic toxicity (myelitis, optic neuropathy and retinopathy, ototoxicity), and second malignancies. (Refer to appropriate sections above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
