Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Clinical-dosimetric analysis of measures of dysphagia including gastrostomy-tube dependence among head and neck cancer patients treated definitively by intensity-modulated radiotherapy with concurrent chemotherapy

Clinical-dosimetric analysis of measures of dysphagia including gastrostomy-tube dependence among... Purpose: To investigate the association between dose to various anatomical structures and dysphagia among patients with head and neck cancer treated by definitive intensity-modulated radiotherapy (IMRT) and concurrent chemotherapy. Methods and materials: Thirty-nine patients with squamous cancer of the head and neck were treated by definitive concurrent chemotherapy and IMRT to a median dose of 70 Gy (range, 68 to 72). In each patient, a gastrostomy tube (GT) was prophylacticly placed prior to starting treatment. Prolonged GT dependence was defined as exceeding the median GT duration of 192 days. Dysphagia was scored using standardized quality-of-life instruments. Dose-volume histogram (DVH) data incorporating the superior/middle pharyngeal constrictors (SMPC), inferior pharyngeal constrictor (IPC), cricoid pharyngeal inlet (CPI), and cervical esophagus (CE) were analyzed in relation to prolonged GT dependence, dysphagia, and weight loss. Results: At 3 months and 6 months after treatment, 87% and 44% of patients, respectively, were GT dependent. Spearman's ρ analysis identified statistical correlations (p < 0.05) between prolonged GT dependence or high grade dysphagia with IPC V65, IPC V60, IPC Dmean, and CPI Dmax. Logistic regression model showed that IPC V65 > 30%, IPC V60 > 60%, IPC Dmean > 60 Gy, and CPI Dmax > 62 Gy predicted for greater than 50% probability of prolonged GT dependence. Page 1 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 Conclusion: Our analysis suggests that adhering to the following parameters may decrease the risk of prolonged GT dependence and dysphagia: IPC V65 < 15%, IPC V60 < 40%, IPC Dmean < 55 Gy, and CPI Dmax < 60 Gy. placement of a GT were also excluded. The remaining 39 Introduction Concurrent chemoradiation therapy using intensity-mod- patients included in the study. The median follow up was ulated radiotherapy (IMRT) has gained widespread 15.6 months (range, 4.5 to 52 months), with 27 patients acceptance as a definitive treatment for locally advanced followed greater than 1 year. All patients received prophy- head and neck cancer due to significant improvement in lactic placement of a GT prior to starting treatment. The tumor control and organ preservation with the addition GT was subsequently removed upon resolution of high of chemotherapy, and promising advantage of increasing grade dysphagia and stabilization of weight after treat- therapeutic gain using IMRT technique [1-4]. However, it ment. Physician judgment if GT needed to be maintained is becoming increasingly clear that chemoradiation strat- was based on the criteria that 1) the patient's weight could egy is associated with an increased incidence and severity not be maintained with less than two cans of supplemen- of swallowing-related toxicities, including high-grade dys- tal feeding per day, or 2) the patient could not tolerate phagia, severe weight loss, and prolonged dependence on solid food without complaints of dysphagia, gastrostomy tube (GT) for fluid and nutritional support [5-7]. Table 1: Patient and tumor characteristics. Variable No. patients % Indwelling GT has been shown to compromise quality of life because it may cause infection and physical discom- Age Mean 56, fort, distort patient's self-esteem, and induce anxiety, Continuous range 32-77 depression, and social isolation [8]. Presently there is a Gender lack of data associating GT dependence and dosimetric Male 32 82 parameters among patients undergoing definitive chemo- Female 7 18 radiotherapy using IMRT for head and neck cancer [9,10]. Active smoking* This is of practical significance since, as a result of IMRT Yes 30 77 No 9 13 optimization, radiation doses can potentially be Alcohol use "dumped" to unspecified anatomical areas including Heavy** 11 28 those related to dysphagia that have not yet been rigor- Others 28 72 ously investigated [11]. In a prospective trial using IMRT, KPS Feng et al demonstrated the importance of monitoring 80-100 25 64 dose to the pharyngeal constrictor muscles, the cervical 60-70 14 36 Primary site esophagus (CE), and the glottic and supraglottic larynx Oral cavity 2 5 (GSL) [12]. The purpose of the present study was to inves- Oropharynx 25 64 tigate the potential association between radiation dose to Larynx 6 15 these structures vital for swallowing and severity of dys- Hypopharynx 3 8 phagia, notably prolonged GT dependence, among a Unknown primary 3 8 cohort of patients undergoing definitive IMRT chemoradi- T stage ation for locally advanced head and neck cancer. T0, 1, 2 25 64 T3, 4 14 36 N stage Methods and materials N0/N1 16 41 Patient characteristics N2 19 49 This was a retrospective study approved by the Institu- N3 4 10 tional Review Board at the University of California, Davis Chemo regimen (UCD). Between January 2003 and January 2007, forty- CDDP-based 33 85 eight patients with newly diagnosed squamous cell carci- Others 6 15 Post RT neck dissection noma involving the oral cavity, oropharynx, larynx or Yes 5 13% hypopharynx were treated with definitive chemoradiation No 34 87% consisting of IMRT and cisplatin at the UCD Cancer Center. Seven patients who either developed locoregional *: currently smoking or smoking history within one year. recurrence or were lost during follow up were excluded **: self reported active heavy alcohol drinking or more than one 6- from the study. Two patients who refused prophylactic pack of beers per day. Page 2 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 odynophagia or aspiration. None of the patients required ing neighboring critical structures. The prescribed dose to GT reinsertion once the GT was initially removed after PTV3 was 54 to 56 Gy. Dose to PTV2 ranged from 59.4 to completion of radiation therapy. Table 1 shows patient 63 Gy (median, 60 Gy). For critical normal structures, characteristics of the study population. dose constraints were designed to limit the maximum dose, whenever possible, to 1% of the volume to 54 Gy for Target volume delineation the brainstem and optic nerves, 45 Gy for the spinal cord The gross tumor volume (GTV) was specified as the gross and optic chiasm, 60 Gy for the temporal lobes, and 30 Gy extent of tumor as demonstrated by preoperative imaging to 50% of the contralateral parotid gland. Treatment was and physical examination including endoscopy. Grossly by continuous-course IMRT with once-a-day treatment. positive lymph nodes were defined as any lymph nodes Because our goal was to prescribe 1.8 Gy per fraction to greater than 1 cm or those with a necrotic center. The the PTV2 daily, the PTV1 received a higher dose per frac- high-risk clinical target volume (CTV1) was defined as the tion, typically 2.0 Gy or 2.12 Gy per fraction, and PTV3 GTV plus a margin of 1-2 cm to account for microscopic typically 1.6-1.7 Gy per fraction. disease spread. The CTV2 generally included the prophy- Delineation of swallowing structures lactically treated cervical and supraclavicular neck. A CTV3 was also created to designate an area at lowest risk within The IMRT treatment plans of all 39 patients treated by the prophylactically treated low neck. The low neck was definitive chemoradiation were retrieved from archival encompassed within the IMRT plan in all cases, and thus records. With the help of a board-certified head and neck a separate anterior low-neck field was not used. Depend- surgeon, the swallowing structures were contoured on ing on disease site, the planning target volume (PTV) con- axial CT slides as previously described [11-14] (Fig. 1). tained an automated 0.5 cm expansion of the CTV Briefly, the pharyngeal constrictor (PC) was outlined as a surfaces to account for patient setup error to create PTV1, single structure for which the cranial-most extent was the PTV2, and PTV3, if necessary. The tumor volumes and sen- caudal tips of the pterygoid plates and the caudal-most sitive normal structures were delineated on serial treat- extent was the inferior border of the cricoid cartilage. For ment planning CT images. Structures considered to be purposes of analysis, the constrictors were considered as critically at risk included the spinal cord, optic nerves, one structure and were also schematically divided into optic chiasm, orbits, lens, brainstem, and parotid glands. two parts: the superior and middle PC (SMPC) was No overlap between CTVs and uninvolved critical adja- defined from the caudal tips of the pterygoid plates cent tissues was permitted for optimization purposes. through the lower edge of the hyoid, at the level of C2, C3 and upper C4. The inferior PC (IPC) was defined from Dose specification below the hyoid through the inferior edge of the cricoid, For patients receiving definitive radiation therapy, treat- with attachment to the inferior horn of thyroid cartilage, ment plans were designed to provide a dose of 68 to 72 Gy at the level of lower C4, C5 and upper C6. On non-con- (median, 70 Gy) to 95% or greater of the PTV1 while spar- trast CT images, IPC can be identified as a structure with faint enhancement of mucosa surrounded by a thin intra- mural fat plate which facilitates the exclusion of the pos- terior cricoarytenoid muscle. The cricopharyngeal inlet (CPI) was defined as an oval structure of 1 cm in length, with lack of intramural fat plate. It extends from the cau- dal cricoid to the first tracheal ring, and is located at the level of lower C6. The CE was contoured as a round struc- ture, caudal to the CPI, with its caudal-most extent corre- sponding to the thoracic inlet. With the above structures delineated on the axial CT slices, tabular differential dose- volume histogram (DVH) data for all the structures were re-computed, taking into consideration the dose actually delivered. Chemotherapy regimens Delineation of the swallowing sim Figure 1 ulation CT and 3D reconstructed image structures on axial slices from The majority (85%) of the patients received bolus cispla- Delineation of the swallowing structures on axial tin (100 mg/m ) given every 3 weeks on days 1 and 22. slices from simulation CT and 3D reconstructed The remaining patients received either weekly carboplatin image. SMPC = superior and middle pharyngeal constrictor; (AUC = 2) or weekly paclitaxel (50 mg/m ) for 6 weeks. IPC = inferior pharyngeal constrictors; CPI = cricoid pharyn- geal inlet; and CE = cervical esophagus. Cetuximab was not used among any of the patients in the Page 3 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 study. None of the patients received sequential induction defined high grade dysphagia. Mucositis and xerostomia or consolidation therapy. was evaluated weekly during treatment, and at follow-up, based on Common Terminology Criteria for Advanced GT management Events (CTCAE), version 2.0. Accordingly, high grade The GT was inserted by the Department of Interventional mucositis was defined as confluent pseudomembranous Radiology at UCD, and was changed every three months. reaction with continuous patches > 1.5 cm (grade 3) or In 3 patients (8%), additional GT changes were performed necrosis or deep ulceration; this may include bleeding not due to complications such as infection or obstruction. induced by minor trauma or abrasion (grade 4). Patients were encouraged to undergo feeding by mouth for as long as it was tolerable. Body weight and toxicity Transnasal esophagoscopy (TNE), flexible endoscopic evaluation of swallow (FEES), and aspiration pneumonia (dysphagia, xerostomia, mucositis, nausea, vomiting, constipation, diarrhea, dysguesia, difficulties chewing) work up were assessed and addressed with patients weekly. We Patients with grade 2 or greater dysphagia beyond 3 used the American Dietetic Association Medical Nutrition months after radiation were referred for TNE. The TNE Therapy (MNT) Protocol for Cancer (Radiation Oncol- technique has previously been described [15,16]. At the ogy) and the UCD Enteral Nutrition Guidelines. Deci- discretion of the physician, a FEES or aspiration pneumo- sions to wean off enteral feeds were based on individual nia work up (bacterial culture and chest X-ray) was per- patient and chemoradiation-induced toxicities with spe- formed. FEES allows direct visual assessment of many cific emphasis placed on inability to consume adequate swallowing functions including muscular function, pre- oral nutrition and fluid, dysphagia, and prevent uncon- mature spillage, pooling, laryngeal penetration, and pres- trolled involuntary weight loss. Patients were weaned off ence of aspiration. In brief, the patients were examined enteral nutrition support when 1) the patient's weight seated upright without anesthesia. Liquid (colored water), could be maintained with less than two cans of supple- pureed food (yogurt), and chewable food (bread) were mental feed per day, and 2) the patient could have certain ingested while the hypopharynx and laryngeal contents solid food without complaints of dysphagia, were viewed with the fiberscope. The results were scored odynophagia or aspiration. as "little", "moderate", or "severe" using the following var- iables: residue, penetration, and aspiration of three differ- Follow-up evaluation ent types of diet (water, yogurt, and bread), and mucus Patients were typically seen 2 to 3 weeks after completion stases. Aspiration pneumonia was defined as cultured bac- of radiation therapy and then every 3 months thereafter terial pneumonia with radiographic evidence of infiltra- for the first year, every 6 months for the second and third tion. year, and then annually. The mean follow up time was 16.2 months (range 4.5-52 months). If a persistent neck Statistical analysis node was found on physical examination after comple- Data analysis and graphs were completed using the R soft- tion of IMRT and/or was positive on PET/CT at 2 months ware program (R Development Core Team, 2006; R Foun- follow-up, salvage neck dissection was performed. Dura- dation for Statistical Computing, Vienna, Austria). tion of GT was defined as the interval between RT comple- Spearman's ρ and univariate regression were used to cal- tion and the date of its removal, or until the date of last culate the correlation of each of these identified DVH follow-up or death if the GT was still present. Prolonged parameters and individual dependent binary variable GT dependence was defined as GT more than the median (absence or presence of prolonged GT days, grade 3+ dys- GT duration. Two patients whose follow-up time were less phagia, and severe weight loss). A logistic model, p = 1/ than the mean GT duration were excluded from further {1+exp [-(α +β *dose or volume of structure)]}, was used analysis of clinical-dosimetric association. Body weight to calculate the probability of developing prolonged GT and patient-reported dysphagia were recorded during days, grade 3+ dysphagia, or severe weight loss. The each follow up. The lowest body weight during the follow unknown parameters α and β were estimated with the up period was used. Severe weight loss was defined as maximum likelihood method. A test was also performed more than 15% weight loss. Patient-reported dysphagia whether the hypothesis β = 0 can be rejected. A p value of was assessed with the validated UWQOL questionnaires < 0.05 was interpreted as being statistically significant given to patients during each follow-up visit. It contained from zero. Confidence intervals (95%) were determined. one swallowing question with five possible answers ("I Multivariate regression was not used due to the model swallow normally", (grade 0); "I cannot swallow certain instability caused by co-linearity between DVH parame- solid food", (grade 1); "I can only swallow soft food", ters (V40, V50, V60, V65, Dmax, Dmean). Wilcoxon rank- (grade 2); "I can only swallow liquid food", (grade 3); and sum analysis was preformed to identify DVH parameters "I cannot swallow", (grade 4). Grade 3 and grade 4 that statistically correlated with esophageal stricture. Page 4 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 within one year) was identified as the only significant fac- Results Swallowing outcomes after treatment tor predictive for prolonged GT dependence (p = 0.03). At 3 months and 6 months after treatment, 87% and 44% Other clinical factors, including age, gender, history of of patients, respectively, were GT dependent (Table 2). alcohol use, KPS, tumor site, T stage, N stage, and type of The results of physician-assessed high grade dysphagia chemotherapy regime are not associated with prolonged were consistent with that of GT dependence, given that GT dependence. Similar analysis of high grade dysphagia majority of high grade dysphagia patients were grade 3 revealed active smoking (p = 0.03) and T stage (p = 0.04) with GT dependence. Due to data redundancy, results of as significant factors. No other predisposing parameter observer-assessed high grade dysphagia were not was found to be statistically significant. In terms of severe reported. Using the UWQOL instrument, 17 patients weight loss, no predisposing parameter was identified to (44%) reported high grade dysphagia at any point during be statistically significant (data not shown). A total of 5 or after treatment. The median percent of maximum patients underwent post-treatment neck dissection. Both weight loss was 12% (range, -4% to 21%). univariate and multivariate analysis did not reveal neck dissection as a significant factor for prolonged GT depend- More than half (54%) of the patients had Grade 3 or 4 ence, high grade dysphagia, or severe weight loss. mucositis at some point after radiation, with 23% and 5% having severe mucositis at 3-month and 6-month follow DVH analysis for prolonged GT dependence up evaluation respectively (Table 2). However, GT The DVH parameters for all the swallowing structures dependence did not improve as rapidly and still persisted (SMPC, IPC, CPI, and CE) were listed in Table 3. Signifi- in 87% and 44% of patients at 3 months and 6 months, cant factors (p < 0.05) for prolonged GT dependence were respectively. There was a lack of temporal association revealed using Spearman's ρ test and subsequent univari- between high grade mucositis and prolonged GT depend- ate logistic regression in an attempt to identify dose-vol- ence of more than 192 days on statistical analysis (p > ume effect for GT duration longer than 192 days versus 0.05). Grade 2 or higher xerostomia was found in 43% less than 192 days. These factors are IPC V65 (p = 0.003), and 36% at 3-month and 6-month follow up evaluation IPC V60 (p = 0.002), IPC V50 (p = 0.042), IPC Dmean (p respectively, and persisted in 31% at the last follow-up = 0.016), and CPI Dmax (p = 0.011). CPI V60 has p value (Table 2). of 0.050. DVH analysis was also performed on a com- bined structure (IPC, CPI and CE). No statistically signifi- Twelve of the patients with high grade dysphagia under- cant factor was identified (p > 0.05). went TNE. Five of them developed stricture at the upper esophageal sphincter at the level of the cricopharyngeus The results of dose-response relationships and volume- muscle, including one with complete luminal stenosis. All response relationships for prolonged GT dependence are of them underwent dilatation at the time of TNE to relieve presented in Fig. 2. IPC V65 more than 30%, IPC V60 any physical obstruction. Four out of the seven patients more than 60%, IPC Dmean more than 60 Gy, and CPI who underwent FEES had finding of moderate or severe Dmax more than 62 Gy predicted for a greater than 50% aspiration to one of the diets. Eight patients had aspira- probability of developing prolonged GT dependence. For tion pneumonia work up, and only one of them was diag- IPC V50, the dose/volume-response relationships results nosed. were not clinically meaningful. The clinical factors listed in Table 1 were included in both In view of the strong dosimetric-clinical correlations for univariate and multivariate analysis of prolonged GT CPI and IPC, we repeated the above analysis with exclu- dependence. Smoking (active smoking or smoking history sion of three patients whose primary disease overlapped with the relevant structures (CPI and IPC). The same DVH Table 2: Toxicity after treatment parameters were observed as significant factors for pro- longed GT dependence. 3 month (%) 6 month (%) GT dependence 87% 44% DVH analysis for high grade dysphagia, severe weight loss, and stricture Self-reported dysphagia grade 3 33% 21% Spearman's ρ test and subsequent univariate logistic regression analysis revealed significant associations Mucositis 23% 5% between several dosimetric parameters and grade 3+ grade 3 patient-reported dysphagia. These factors are IPC V65 (p = 0.040), CPI Dmax (p = 0.037), and CPI V60 (p = 0.046). Xerostomia grade 2 43% 36% Further analysis of dose-response relationships and vol- ume-response relationships revealed that IPC V65 more Abbreviation: GT = gastrostomy tube Page 5 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 Table 3: Swallowing structure DVH parameters (median value and range) and p values for association with GT dependence V40 V50 V60 V65 Dmax Dmean (%) (%) (%) (%) (Gy) (Gy) CE GT > 192 d 57 (1-86) 24 (0-79) 0 (0-33) 0 (0-1) 61 (45-77) 36 (20-53) GT 192 d 65 (17-98) 27 (0-86) 0 (0-13) 0 (0-0) 57 (44-76) 36 (21-56) p = 2.671 p = 0.524 p = 0.173 P = 0.169 p = 0.238 p = 0.383 CPI GT > 192 d 100 (60-100) 100 (4-100) 8 (0-100) 0 (0-100) 64 (56-78) 57 (42-69) GT 192 d 100 (93-100) 95 (3-100) 0 (0-73) 0 (0-27) 58 (53-67) 53 (43-62) p = 0.512 p = 0.069 p = 0.050 P = 0.062 p = 0.011 p = 0.083 IPC GT > 192 d 100 (97-100) 100 (76-100) 81 (40-100) 42 (21-100) 74 (66-79) 64 (54-70) GT 192 d 100 (95-100) 92 (40-100) 37 (10-84) 15 (0-60) 72 (68-77) 55 (46-68) p = 0.367 p = 0.042 p = 0.002 P = 0.003 p = 0.057 p = 0.016 SMPC GT > 192 d 100 (60-100) 100 (56-100) 94 (45-100) 72 (21-93) 76 (66-79) 67 (38-72) GT 192 d 100 (84-100) 97 (70-100) 90 (25-100) 60 (0-95) 76 (70-79) 65 (50-71) p = 0.378 p = 0.072 p = 0.063 P = 0.091 p = 0.252 p = 0.086 Abbreviation: DVH = Dose-volume histogram; GT = gastrostomy tube; SMPC = superior and middle pharyngeal constrictor; IPC = inferior pharyngeal constrictors; CPI = cricoid pharyngeal inlet; and CE = cervical esophagus; Dmax = maximum dose; Dmean = mean dose. than 65%, CPI V60 more than 78%, CPI Dmax more than aspiration or stricture development [17]. Levendag et al 70 Gy were associated with more than 50% probability of identified dose-response relationship between dysphagia developing high grade dysphagia. Similar analysis did not for solids (p < 0.02) or aspiration episodes (p < 0.02) and reveal statistically significant DVH predictors for severe mean dose to IPC. A mean dose of 33 Gy to IPC was esti- weight loss (data not shown). Wilcoxon rank-sum analy- mated as the threshold for 20% risk of dysphagia for sol- sis revealed significant associations between stricture and ids [14]. Furthermore, Dornfeld et al reported that a more two dosimetric parameters (CPI V65, CPI Dmax). restrictive diet one year after treatment is significantly cor- related with higher average dose delivered to the constric- Discussion tor muscles (lateral pharygeal wall) at the level of false It has been a common observation that a correlation exists vocal cord [18]. Jensen et al demonstrated that dose above between dysphagia and radiation doses to the anatomic 60 Gy to the upper esophageal sphincter could result in structures responsible for swallowing in patients undergo- higher risk of late swallowing dysfunction [19]. This well ing definitive chemoradiation for head and neck cancer. documented association between high dose to IPC or CPI However, the present study is the first to document a rela- and prolonged GT dependence was also supported by two tionship between various dosimetric parameters and pro- earlier reports showing that patients were more likely to longed GT dependence. Notably, we were able to identify have prolonged GT dependence and high grade dysphagia DVH parameters which were significantly associated with when treated with extended-field IMRT rather than being prolonged GT dependence, including V65 of the IPC, V60 treated with an upper IMRT fields junctioned with an of the IPC, mean dose to the IPC, and maximum dose to anterior neck field. This is thought to be due to the pres- the CPI. Based on these dose/volume-response relation- ence of midline block in an anterior neck field to prevent ships, we currently recommend IPC V65 less than 15%, unanticipated high dose radiation to structures including IPC V60 less than 40%, IPC Dmean less than 55 Gy, and larynx, IPC, CPI and CE [20,21]. CPI Dmax less than 60 Gy as potentially important DVH constraints to guide IMRT planning in an attempt to sig- The significant dose-volume effect relationships regarding nificantly reduce the risk of swallowing dysfunction and prolonged GT dependence for IPC and CPI could be prolonged GT dependence. explained by the role of the upper esophageal sphincter (UES) in the normal swallowing process. The UES is a Our findings demonstrate the importance of IPC and CPI functional entity that is composed of three muscles: the dosimetric parameters for developing swallowing dys- IPC muscle, the CPI muscle, and the upper esophageal function and are consistent with those from several muscle. The UES opens by relaxation of the three closing recently published studies. Caglar et al showed that a muscles, traction by IPC and other muscles that attached mean dose to the IPC of more than 54 Gy and IPC V50 of to the hyoid bone and thyroid cartilage, anterior move- more than 50% were the most significant predictors for ment of the larynx, and pulsion of the bolus. The various Page 6 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 Volume-response or dose-r u CPI (D) Figure 2 me of the IPC receiving mo espon re tha sn e relationsh 6500 cGy (A), ip for the average or 6000 cGy (B prob ), or the mean dose to the ability of having prolonged GT dependence and the vol- IPC (C), or maximum dose to Volume-response or dose-response relationship for the average probability of having prolonged GT depend- ence and the volume of the IPC receiving more than 6500 cGy (A), or 6000 cGy (B), or the mean dose to the IPC (C), or maximum dose to CPI (D). GT = gastrostomy tube; IPC = inferior pharyngeal constrictors; CPI = cricoid pha- ryngeal inlet. The š lines plot the mean risk; the - lines plot the estimated upper and lower limits of 95% confidence interval. The ♦ points depict the observed values. muscles of the UES behave differently during its many tified. A cause of dysphagia could also be attributed in dynamic states, so that similar functions are accomplished part to the failure of sensation and timely response to the by different muscles. Any impairment of the CPI and IPC bolus passing through this region. The importance of IPC could result in dysphagia. In addition, UES is considered and CPI is validated by our finding that patient-reported a high pressure zone, with the highest pressure at the dysphagia was highly correlated with the dose to the two region around IPC where proprioceptive units were iden- structures. Page 7 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 In addition to the IPC and CPI, several other anatomic with both high grade dysphagia and stricture formation. structures have been reported as dysphagia/aspiration In spite of this high incidence among patients with high related with significant dose-volume relationship. These grade dysphagia, the overall incidence of stricture in all structures include GSL and PC, with superior PC having patients is 12% (5/41) in our study, consistent with stric- the strongest dose-response association [11,12]. The ture rate of 17%-37% in other studies [17,26-28]. importance of superior and middle PC for swallowing after radiation therapy was also shown by Teguh et al It is important to note that this study was retrospective [14,22,23]. Although our study failed to find a significant with inherent limitations. First, the lack of systematic eval- a correlation between GT dependence and dose to the uation of some of the major end points of late dysphagia GSL, SMPC, or PC as a whole, this could potentially be using TNE or FEES prevented more robust analysis using explained by the differences in patient characteristics. In more objective endpoints. Another limitation of the study the above mentioned studies, only oropharynx and was the relatively short follow-up with a median duration nasopharynx patients were included. As such, our results of 15.6 months. However, most of the endpoint events are consistent with those from Caglar et al that the mean occurred less than 1 year after treatment. Given the small dose or V50 to IPC, not the superior PC, were significant number of events, we conjecture that a sub-analysis of predictors for aspiration or stricture development [17]. swallowing function assessed at greater than 1 year after treatment would likely not change our findings. Nonethe- Prolonged GT dependence is regarded by most head and less, we do acknowledge that further studies with more neck cancer patients as contributing to compromised comprehensive objective endpoints with prolonged fol- quality of life because it may cause infection and physical low-up may be necessary to yield a more thorough evalu- discomfort, distort patient's self-esteem, and induce anxi- ation. This is well exemplified in a recently published ety, depression and social isolation [8]. This is of increas- retrospective study of patients with more than 1 year fol- ing concern in recent years when concurrent low-up, where a composite of 3 objective endpoints (GT chemoradiation for tumor control and organ preservation dependence, aspiration, and pharyngoesophageal stric- has gained widespread practice but is associated with high ture) were successfully used as surrogates for severe long- rate of severe late dysphagia, including prolonged GT term dysphagia [26]. dependence [5]. Multiple large randomized trials testing intensified chemoradiation regimens reported GT rates of It must also be recognized that the majority of patients in about 70%, and chronic tube dependence of 10-20% the present study presented with oropharynx cancer, and [3,6,7]. In a recent study where 95% of the chemoradia- this fact may have biased our findings. Teguh et al, for tion patients had prophylactic feeding tubes placed before instance, demonstrated that patients with base of tongue treatment, Caglar et al reported prolonged GT dependence disease experienced more severe dysphagia than those in 37% of the patients, with a median GT duration of 112 with tumors at other sites [22]. In addition to oropharynx days after radiation completion [17]. Notably, we also primary, the larynx, hypopharynx and pharyngeal wall identified smoking as a risk factor for GT dependence. The were also found to predispose to dysphagia more so than etiology for smoking induced dysphagia is likely multifac- other regions of the head and neck [26-29]. In contrast, torial and related to prolonged tissue recovery secondary however, Logemann et al showed no differences in the fre- to nicotine induced hypoxia, the appetite reducing effects quency of dysphagia across different head and neck dis- of nicotine, or mucosal irritation. Multiple previous works ease sites [30], which is further supported by a large have similarly associated smoking with higher rates of prospective study that excluded disease site as a statisti- toxicity including aspiration and esophageal stricture after cally significant factor for quality of life changes among radiation therapy [24,25]. head and neck cancer treated with radiation therapy [31]. Notably, 5 out of the 12 (42%) patients with high grade Lastly, we were unable to control for potentially con- dysphagia developed upper esophageal stricture in this founding factors which may have also predisposed to study. This high incidence could have resulted from detec- swallowing dysfunction including severe mucositis, pre- tion bias, small patient number, or most likely, patient treatment dysphagia and post-treatment xerostomia. We over-reliance on a GT which led to less swallowing and acknowledge that the endpoints in this study, such as high allowing scar and stricture formation. The last possibility grade dysphagia and prolonged GT dependence may have is supported by results from Caudell et al who demon- been confounded by the development of severe acute strated a trend toward an association (p = 0.09) between mucositis or its consequential late effects, such as submu- GT dependence and pharyngeal stricture or stenosis [26]. cosal edema, fibrosis, scarring, soft tissue necrosis, Another explanation for this high incidence of stricture impaired sensory or motor function, and loss of mucosal among patients with high grade dysphagia could be its rel- compliance. This is based on the rationale that high dose atation with CPI Dmax, which was significantly associated radiation to a large volume of the constrictor muscles Page 8 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 (resulting in high values of V60 and V65) also results in ric constraints should not replace the effort of early swal- high dose to large volumes of mucosal surface which is lowing therapy and exercises which resulted in maximal believed to lead to more severe mucositis [32-34]. How- swallowing recovery in several studies [5,40]. ever, since mucositis tends to be self-limiting and an acute, rather than late side-effect of radiation therapy, we Competing interests believe that the potential confounding effects are mini- The authors declare that they have no competing interests. mal. Furthermore, the present study demonstrated a lack of significant association between acute mucositis and Authors' contributions prolonged GT dependence, which is consistent with the BL and AMC conceived of the study, and participated in its findings of a dissociation between acute mucositis and design, carried out data collection, data analysis, manu- dysphagia by Mekhail et al [27] and it is supported by script writing, and coordination. DL and DMR performed recent data from Anand et al showing no correlation statistics analysis. DHL performed data collection regard- between long-term dysphagia and acute mucositis (Grade ing chemotherapy regimens and participated in manu- 3, 4) in spite of the severe mucosities that developed in script writing. DGF and QL performed dysphagia data 53% of locally advanced head neck cancer patients treated collection including TNE and FEES, and delineation of with IMRT [35]. In addition, we were unable to com- swallowing structures. KN and JC performed data collec- pletely exclude the possibility of other confounding fac- tion regarding GT management. JAP performed physics tors such as pre-treatment dysphagia or post-treatment consult on re-computation of DVH. All authors read and severe xerostomia as confounding factors. Multiple stud- approved the final manuscript. ies suggested that very few patients with newly diagnosed head and neck cancer have severe dysphagia or aspiration Acknowledgements This study was presented in abstract form at the 2008 annual meeting of prior to definitive treatment [29,36,37]. Moreover, we the American Society for Therapeutic Radiology and Oncology (ASTRO) in could not completely rule out xerostomia as a confound- Boston, Massachusetts. ing factor in spite of the reduction in the risk of this symp- tom associated with parotid gland sparing IMRT. References Xerostomia secondary to chemotherapy or radiation ther- 1. Adelstein DJ, Li Y, Adams GL, Wagner H Jr, Kish JA, Ensley JF, apy has been suggested to significantly affect food bolus Schuller DE, Forastiere AA: An Intergroup phase III comparis- ion of standard radiation therapy and two schedules of con- formation and swallowing function, and contribute sig- current chemoradiotherapy in patients with unresectable nificantly to dysphagia after chemoradiation [38,39]. In squamous cell head and neck cancer. J Clin Oncol 2003, particular, Teguh et al demonstrated a strong correlation 21:92-98. 2. Denis F, Garaud P, Bardet E, Alfonsi M, Sire C, Germain T, Bergerot between dry mouth and sticky saliva with dysphagia- P, Rhein B, Tortochaux J, Calais G: Final results of the 94-01 related quality of life such as normalcy of diet and French Head and Neck Oncology and Radiotherapy Group randomized trial comparing radiotherapy alone with con- odynophagia [22]. As a result of these potential con- comitant radiochemotherapy in advanced-stage oropharynx founding factors, it remains difficult to definitively estab- carcinoma. J Clin Oncol 2004, 22:69-76. lish a cause-effect association in spite of the significant 3. Forastiere AA, Goepfert H, Maor M, Pajak TF, Weber R, Morrison W, Glisson B, Trotti A, Ridge JA, Chao C, Peters G, Lee DJ, Leaf A, dose/volume-response relationship between some key Ensley J, Cooper J: Concurrent chemotherapy and radiother- DVH parameters and risk of prolonged GT dependence. A apy for organ preservation in advanced laryngeal cancer. N larger prospective study is required in the future to further Engl J Med 2003, 349:2091-8. 4. Grégoire V, De Neve W, Eisbruch A, Lee N, Weyngaert D Van den, investigate theses associations. Van Gestel D: Intensity-modulated radiation therapy for head and neck carcinoma. Oncologist 2007, 12:565-8. 5. Rosenthal DI, Lewin JS, Eisbruch A: Prevention and treatment of Conclusion dysphagia and aspiration after chemoradiation for head and We provided evidence that prolonged GT is correlated neck cancer. J Clin Oncol 2006, 24:2636-43. with DVH parameters for patients with locally advanced 6. Garden AS, Harris J, Vokes EE, Forastiere AA, Ridge JA, Jones C, Hor- witz EM, Glisson BS, Nabell L, Cooper JS, Demas W, Gore E: Pre- head and neck cancer treated with definitive concurrent liminary results of Radiation Therapy Oncology Group 97- chemotherapy and IMRT. To minimize the risk of pro- 03: a randomized phase II trial of concurrent radiation and chemotherapy for advanced squamous cell carcinomas of longed GT dependence, we currently strive to keep IPC the head and neck. J Clin Oncol 2004, 22:2856-64. V65 less than 15%, IPC V60 less than 40%, and maintain 7. Adelstein DJ, Lavertu P, Saxton JP, Secic M, Wood BG, Wanamaker IPC Dmean less than 55 Gy, and CPI Dmax less than 60 JR, Eliachar I, Strome M, Larto MA: Mature results of a phase III randomized trial comparing concurrent chemoradiotherapy Gy during IMRT planning in an attempt to decrease the with radiation therapy alone in patients with stage III and IV risk of prolonged GT dependence. It should be noted that squamous cell carcinoma of the head and neck. Cancer 2000, these guidelines are implemented on a case-by-case basis 88:876-83. 8. Nguyen NP, Frank C, Moltz CC, Vos P, Smith HJ, Karlsson U, Dutta considering such factors as tumor extent and location. S, Midyett A, Barloon J, Sallah S: Impact of dysphagia on quality Future directions include large-scale prospective trials of life after treatment of head-and-neck cancer. Int J RadiatOn- col Biol Phys 2005, 61:772-8. aiming to assess the clinical benefits gained by applying 9. Poulsen MG, Riddle B, Keller J, Porceddu SV, Tripcony L: Predictors these dosimetric strategies. Lastly, the proposed dosimet- of acute grade 4 swallowing toxicity in patients with stages Page 9 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 III and IV squamous carcinoma of the head and neck treated locally advanced head-and-neck cnacer. Int J Radiat Oncol Biol with radiotherapy alone. Radiother Oncol 2008, 87:253-9. Phys, Int J Radiat Oncol Biol Phys 2009, 73:410-5. 10. Mendenhall WM, Morris CG, Amdur RJ, Hinerman RW, Mancuso AA: 27. Mekhail TM, Adelstein DJ, Rybicki LA, Larto MA, Saxton JP, Lavertu Parameters that predict local control after definitive radio- P: Enteral nutrition during the treatment of head and neck therapy for squamous cell carcinoma of the head and neck. carcinoma: is a percutaneous endoscopic gastrostomy tube Head Neck 2003, 25:535-42. preferable to a nasogastric tube? Cancer 2001, 91:1785-90. 11. Eisbruch A, Schwartz M, Rasch C, Vineberg K, Damen E, Van As CJ, 28. Lee WT, Akst LM, Adelstein DJ, Saxton JP, Wood BG, Strome M, But- Marsh R, Pameijer FA, Balm AJ: Dysphagia and aspiration after ler RS, Esclamado RM: Risk factors for hypopharyngeal/upper chemoradiotherapy for head-and-neck cancer: which ana- esophageal stricture formation after concurrent chemoradi- tomic structures are affected and can they be spared by ation. Head Neck 2006, 28:808-12. IMRT? Int J Radiat Oncol Biol Phys 2004, 60:1425-39. 29. Machtay M, Moughan J, Trotti A, Gargen AS, Weber RS, Cooper JS, 12. Feng FY, Kim HM, Lyden TH, Haxer MJ, Feng M, Worden FP, Chep- Forastiere A, Ang KK: Factors associated with severe late tox- eha DB, Eisbruch A: Intensity-modulated radiotherapy of head icity after concurrent chemoradiation for locally advanced and neck cancer aiming to reduce dysphagia: early dose- head and neck cancer: an RTOG analysis. J Clin Oncol 2008, effect relationships for the swallowing structures. Int J Radiat 26:3581-88. Oncol Biol Phys 2007, 68:1289-98. 30. Logeman JA, Rademaker AW, Pauloski BR, Lazarus CL, Mittal BB, 13. Schmalfuss IM, Mancuso AA, Tart RP: Postcricoid region and cer- Brockstein B, MacCracken E, Haraf DJ, Vokes EE, Newman LA, Liu D: vical esophagus: normal appearance at CT and MR imaging. Site of disease and treatment protocol as correlated of swal- Radiology 2000, 214:237-46. lowing function in patients with head and neck cancer 14. Levendag PC, Teguh DN, Voet P, Est H van der, Noever I, de Kruijf treated with chemoradiaiton. Head Neck 2006, 28:64-73. WJ, Kolkman-Deurloo IK, Prevost JB, Poll J, Schmitz PI, Heijmen BJ: 31. Langendijk JA, Doornaert P, Verdonck-de Leeuw IM, Leemans CR, Dysphagia disorders in patients with cancer of the orophar- Aaronson NK, Slotman BJ: Impact of late treatment-related ynx are significantly affected by the radiation therapy dose to toxicity on quality of life among patients with head and neck the superior and middle constrictor muscle: a dose-effect cancer treated with radiotherapy. J Clin Oncol 3776, relationship. Radiother Oncol 2007, 85:64-73. 26:3770-2008. 15. Postma GN, Cohen JT, Belafsky PC, Halum SL, Gupta SK, Bach KK, 32. Rosenthal DI, Trotti A: Strategies for managing radiation- Koufman JA: Transnasal esophagoscopy: revisited (over 700 induced mucositis in head and neck cancer. Semin Radiat Oncol consecutive cases). Laryngoscope 2005, 115:321-323. 2009, 19:29-34. 16. Langmore SE, Schatz K, Olsen N: Fiberoptic endoscopic exami- 33. Dorr W, Hendry JH: Consequential late effects in normal tis- nation of swallowing safety: a new procedure. Dysphagia 1998, sues. Radiother Oncol 2001, 61:223-231. 2:216-9. 34. Trotti A, Bellm LA, Esptein JB, Frame D, Fuchs HJ, Gwede CK, 17. Caglar HB, Tishler RB, Othus M, Burke E, Li Y, Goguen L, Wirth LJ, Komaroff E, Nalysnyk L, Zilberberg MD: Mucositis incidence, Haddad RI, Norris CM, Court LE, Aninno DJ, Posner MR, Allen AM: severity and associated outcomes in patients with head and Dose to Larynx Predicts for Swallowing Complications After neck cancer receiving radiotherapy with or without chemo- Intensity-Modulated Radiotherapy. Int J Radiat Oncol Biol Phys therapy: a systemic literature review. Radiother Oncol 2003, 1118, 72:1110-2008. 66:253-262. 18. Dornfeld K, Simmons JR, Karnell L, Karnell M, Funk G, Yao M, Wacha 35. Anand A, Chaudhoory A, Shukla A, Negi PS, Sinha SN, Babu AA, J, Zimmerman B, Buatti JM: Radiation doses to structures within Munjal RK, Dewan AK, Kumar K, Doval DC, Vaid AK: Favourable and adjacent to the larynx are correlated with long-term impact of intensity-modulated radiation therapy on chronic diet- and speech-related quality of life. Int J Radiat Oncol Biol Phys dysphagia in patients with head and neck cnacer. Br J Radiol 2007, 68:750-7. 2008, 81:865-71. 19. Jensen K, Lambertsen K, Grau C: Late swallowing dysfunction 36. Pauliski BR, Rademake AW, Logemann JA, Sten D, Beery Q, Newman and dysphagia after radiotherapy for pharynx cancer: fre- L, Hanchett C, Tusant S, MacCracken E: Pretreatment swallowing quency, intensity and correlation with dose and volume function in patients with head and neck cancer. Head Neck parameters. Radiother Oncol 2007, 85:74-82. 2000, 22:474-82. 20. Fua TF, Corry J, Milner AD, Cramb J, Walsham SF, Peters LJ: Inten- 37. Stenson KM, MacCracken E, List M, Haraf DJ, Brockstein B, Weich- sity-modulated radiotherapy for nasopharyngeal carcinoma: selbaum R, Vokes EE: Swallowing function in patients with head clinical correlation of dose to the pharyngo-esophageal axis and neck cancer prior to treatment. Arch Otolaryngol Head Neck and dysphagia. Int J Radiat OncolBiol Phys 2007, 67:976-81. Surg 2000, 126:371-7. 21. Lee NY, O'Meara W, Chan K, Della-Bianca C, Mechalakos JG, Zhung 38. Logemann JA, Pauloski BR, Rademaker AW, Lazarus CL, Mittal B, J, Wolden SL, Narayana A, Kraus D, Shah JP, Pfister DG: Concur- Gaziano J, Stachowiak L, MacCracken E, Newman LA: Xerostomia: rent chemotherapy and intensity-modulated radiotherapy 12-month changes in saliva production and its relationship to for locoregionally advanced laryngeal and hypopharyngeal perception and performance of swallow function, oral cancers. Int J RadiatOncol Biol Phys 2007, 69:459-68. intake, and diet after chemoradiation. Head Neck 2003, 22. Teguh DN, Levendag PC, Noever I, van Rooij P, Voet P, Est H van der, 25:432-7. Sipkema D, Sewnaik A, Baatenburg de Jong RJ, de la Bije D, Schmitz 39. Hamlet S, Faull J, Klein B, Aref A, Fontanesi J, Stachler R, Shamsa F, PI: Treatment Techniques and Site Considerations Regard- Jones L, Simpson M: Mastification and swallowing in patients ing Dysphagia-Related Quality of Life in Cancer of the with postirradiation xerostomia. Int J Radiat Oncol Biol Phys 1997, Oropharynx and Nasopharynx. Int J Radiat Oncol Biol Phys 2008, 37:789-96. 72:1119-27. 40. Shaker R, Easterling C, Kern M, Nitschke T, Massey B, Daniels S, 23. Teguh DN, Levendag PC, Sewnaik A, Hakkesteegt MM, Noever I, Grande B, Kazandjian M, Dikeman K: Rehabilitation of swallowing Voet P, Est H van der, Sipkema D, van Rooij P, Baatenburg de Jong RJ, by exercise in tube-fed patients with pharyngeal dysphagia Schmitz PI: Results of fibroptic endoscopic evaluation of swal- secondary to abnormal UES opening. Gastroenterology 2002, lowing vs. radiation dose in the swallowing muscles after 122:1314-21. radiotherapy of cancer in the oropharynx. Radiother Oncol 2008, 89:57-63. 24. Jensen K, Jensen AB, Grau C: Smoking has a negative impact upon health related quality of life after treatment for head and neck cancer. Oral Oncol 2007, 43:187-92. 25. Mangar S, Slevin N, Mais K, Sykes A: Evaluating predictive factors for determining enteral nutrition in patients receiving radi- cal radiotherapy for head and neck cancer: A retrospective review. Radiother Oncol 2006, 78:152-158. 26. Caudell J, Shaner P, Meredith R, Bonner J, Locher JL, Nabell LM, Car- roll WR, Magnuson JS, Spencer SA, Bonner JA: Factors associated with long-term dysphagia after definitive radiotherapy for Page 10 of 10 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Radiation Oncology Springer Journals

Clinical-dosimetric analysis of measures of dysphagia including gastrostomy-tube dependence among head and neck cancer patients treated definitively by intensity-modulated radiotherapy with concurrent chemotherapy

Loading next page...
 
/lp/springer-journals/clinical-dosimetric-analysis-of-measures-of-dysphagia-including-yukFGKOvhj
Publisher
Springer Journals
Copyright
Copyright © 2009 by Li et al; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; Oncology; Radiotherapy
eISSN
1748-717X
DOI
10.1186/1748-717X-4-52
pmid
19909531
Publisher site
See Article on Publisher Site

Abstract

Purpose: To investigate the association between dose to various anatomical structures and dysphagia among patients with head and neck cancer treated by definitive intensity-modulated radiotherapy (IMRT) and concurrent chemotherapy. Methods and materials: Thirty-nine patients with squamous cancer of the head and neck were treated by definitive concurrent chemotherapy and IMRT to a median dose of 70 Gy (range, 68 to 72). In each patient, a gastrostomy tube (GT) was prophylacticly placed prior to starting treatment. Prolonged GT dependence was defined as exceeding the median GT duration of 192 days. Dysphagia was scored using standardized quality-of-life instruments. Dose-volume histogram (DVH) data incorporating the superior/middle pharyngeal constrictors (SMPC), inferior pharyngeal constrictor (IPC), cricoid pharyngeal inlet (CPI), and cervical esophagus (CE) were analyzed in relation to prolonged GT dependence, dysphagia, and weight loss. Results: At 3 months and 6 months after treatment, 87% and 44% of patients, respectively, were GT dependent. Spearman's ρ analysis identified statistical correlations (p < 0.05) between prolonged GT dependence or high grade dysphagia with IPC V65, IPC V60, IPC Dmean, and CPI Dmax. Logistic regression model showed that IPC V65 > 30%, IPC V60 > 60%, IPC Dmean > 60 Gy, and CPI Dmax > 62 Gy predicted for greater than 50% probability of prolonged GT dependence. Page 1 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 Conclusion: Our analysis suggests that adhering to the following parameters may decrease the risk of prolonged GT dependence and dysphagia: IPC V65 < 15%, IPC V60 < 40%, IPC Dmean < 55 Gy, and CPI Dmax < 60 Gy. placement of a GT were also excluded. The remaining 39 Introduction Concurrent chemoradiation therapy using intensity-mod- patients included in the study. The median follow up was ulated radiotherapy (IMRT) has gained widespread 15.6 months (range, 4.5 to 52 months), with 27 patients acceptance as a definitive treatment for locally advanced followed greater than 1 year. All patients received prophy- head and neck cancer due to significant improvement in lactic placement of a GT prior to starting treatment. The tumor control and organ preservation with the addition GT was subsequently removed upon resolution of high of chemotherapy, and promising advantage of increasing grade dysphagia and stabilization of weight after treat- therapeutic gain using IMRT technique [1-4]. However, it ment. Physician judgment if GT needed to be maintained is becoming increasingly clear that chemoradiation strat- was based on the criteria that 1) the patient's weight could egy is associated with an increased incidence and severity not be maintained with less than two cans of supplemen- of swallowing-related toxicities, including high-grade dys- tal feeding per day, or 2) the patient could not tolerate phagia, severe weight loss, and prolonged dependence on solid food without complaints of dysphagia, gastrostomy tube (GT) for fluid and nutritional support [5-7]. Table 1: Patient and tumor characteristics. Variable No. patients % Indwelling GT has been shown to compromise quality of life because it may cause infection and physical discom- Age Mean 56, fort, distort patient's self-esteem, and induce anxiety, Continuous range 32-77 depression, and social isolation [8]. Presently there is a Gender lack of data associating GT dependence and dosimetric Male 32 82 parameters among patients undergoing definitive chemo- Female 7 18 radiotherapy using IMRT for head and neck cancer [9,10]. Active smoking* This is of practical significance since, as a result of IMRT Yes 30 77 No 9 13 optimization, radiation doses can potentially be Alcohol use "dumped" to unspecified anatomical areas including Heavy** 11 28 those related to dysphagia that have not yet been rigor- Others 28 72 ously investigated [11]. In a prospective trial using IMRT, KPS Feng et al demonstrated the importance of monitoring 80-100 25 64 dose to the pharyngeal constrictor muscles, the cervical 60-70 14 36 Primary site esophagus (CE), and the glottic and supraglottic larynx Oral cavity 2 5 (GSL) [12]. The purpose of the present study was to inves- Oropharynx 25 64 tigate the potential association between radiation dose to Larynx 6 15 these structures vital for swallowing and severity of dys- Hypopharynx 3 8 phagia, notably prolonged GT dependence, among a Unknown primary 3 8 cohort of patients undergoing definitive IMRT chemoradi- T stage ation for locally advanced head and neck cancer. T0, 1, 2 25 64 T3, 4 14 36 N stage Methods and materials N0/N1 16 41 Patient characteristics N2 19 49 This was a retrospective study approved by the Institu- N3 4 10 tional Review Board at the University of California, Davis Chemo regimen (UCD). Between January 2003 and January 2007, forty- CDDP-based 33 85 eight patients with newly diagnosed squamous cell carci- Others 6 15 Post RT neck dissection noma involving the oral cavity, oropharynx, larynx or Yes 5 13% hypopharynx were treated with definitive chemoradiation No 34 87% consisting of IMRT and cisplatin at the UCD Cancer Center. Seven patients who either developed locoregional *: currently smoking or smoking history within one year. recurrence or were lost during follow up were excluded **: self reported active heavy alcohol drinking or more than one 6- from the study. Two patients who refused prophylactic pack of beers per day. Page 2 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 odynophagia or aspiration. None of the patients required ing neighboring critical structures. The prescribed dose to GT reinsertion once the GT was initially removed after PTV3 was 54 to 56 Gy. Dose to PTV2 ranged from 59.4 to completion of radiation therapy. Table 1 shows patient 63 Gy (median, 60 Gy). For critical normal structures, characteristics of the study population. dose constraints were designed to limit the maximum dose, whenever possible, to 1% of the volume to 54 Gy for Target volume delineation the brainstem and optic nerves, 45 Gy for the spinal cord The gross tumor volume (GTV) was specified as the gross and optic chiasm, 60 Gy for the temporal lobes, and 30 Gy extent of tumor as demonstrated by preoperative imaging to 50% of the contralateral parotid gland. Treatment was and physical examination including endoscopy. Grossly by continuous-course IMRT with once-a-day treatment. positive lymph nodes were defined as any lymph nodes Because our goal was to prescribe 1.8 Gy per fraction to greater than 1 cm or those with a necrotic center. The the PTV2 daily, the PTV1 received a higher dose per frac- high-risk clinical target volume (CTV1) was defined as the tion, typically 2.0 Gy or 2.12 Gy per fraction, and PTV3 GTV plus a margin of 1-2 cm to account for microscopic typically 1.6-1.7 Gy per fraction. disease spread. The CTV2 generally included the prophy- Delineation of swallowing structures lactically treated cervical and supraclavicular neck. A CTV3 was also created to designate an area at lowest risk within The IMRT treatment plans of all 39 patients treated by the prophylactically treated low neck. The low neck was definitive chemoradiation were retrieved from archival encompassed within the IMRT plan in all cases, and thus records. With the help of a board-certified head and neck a separate anterior low-neck field was not used. Depend- surgeon, the swallowing structures were contoured on ing on disease site, the planning target volume (PTV) con- axial CT slides as previously described [11-14] (Fig. 1). tained an automated 0.5 cm expansion of the CTV Briefly, the pharyngeal constrictor (PC) was outlined as a surfaces to account for patient setup error to create PTV1, single structure for which the cranial-most extent was the PTV2, and PTV3, if necessary. The tumor volumes and sen- caudal tips of the pterygoid plates and the caudal-most sitive normal structures were delineated on serial treat- extent was the inferior border of the cricoid cartilage. For ment planning CT images. Structures considered to be purposes of analysis, the constrictors were considered as critically at risk included the spinal cord, optic nerves, one structure and were also schematically divided into optic chiasm, orbits, lens, brainstem, and parotid glands. two parts: the superior and middle PC (SMPC) was No overlap between CTVs and uninvolved critical adja- defined from the caudal tips of the pterygoid plates cent tissues was permitted for optimization purposes. through the lower edge of the hyoid, at the level of C2, C3 and upper C4. The inferior PC (IPC) was defined from Dose specification below the hyoid through the inferior edge of the cricoid, For patients receiving definitive radiation therapy, treat- with attachment to the inferior horn of thyroid cartilage, ment plans were designed to provide a dose of 68 to 72 Gy at the level of lower C4, C5 and upper C6. On non-con- (median, 70 Gy) to 95% or greater of the PTV1 while spar- trast CT images, IPC can be identified as a structure with faint enhancement of mucosa surrounded by a thin intra- mural fat plate which facilitates the exclusion of the pos- terior cricoarytenoid muscle. The cricopharyngeal inlet (CPI) was defined as an oval structure of 1 cm in length, with lack of intramural fat plate. It extends from the cau- dal cricoid to the first tracheal ring, and is located at the level of lower C6. The CE was contoured as a round struc- ture, caudal to the CPI, with its caudal-most extent corre- sponding to the thoracic inlet. With the above structures delineated on the axial CT slices, tabular differential dose- volume histogram (DVH) data for all the structures were re-computed, taking into consideration the dose actually delivered. Chemotherapy regimens Delineation of the swallowing sim Figure 1 ulation CT and 3D reconstructed image structures on axial slices from The majority (85%) of the patients received bolus cispla- Delineation of the swallowing structures on axial tin (100 mg/m ) given every 3 weeks on days 1 and 22. slices from simulation CT and 3D reconstructed The remaining patients received either weekly carboplatin image. SMPC = superior and middle pharyngeal constrictor; (AUC = 2) or weekly paclitaxel (50 mg/m ) for 6 weeks. IPC = inferior pharyngeal constrictors; CPI = cricoid pharyn- geal inlet; and CE = cervical esophagus. Cetuximab was not used among any of the patients in the Page 3 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 study. None of the patients received sequential induction defined high grade dysphagia. Mucositis and xerostomia or consolidation therapy. was evaluated weekly during treatment, and at follow-up, based on Common Terminology Criteria for Advanced GT management Events (CTCAE), version 2.0. Accordingly, high grade The GT was inserted by the Department of Interventional mucositis was defined as confluent pseudomembranous Radiology at UCD, and was changed every three months. reaction with continuous patches > 1.5 cm (grade 3) or In 3 patients (8%), additional GT changes were performed necrosis or deep ulceration; this may include bleeding not due to complications such as infection or obstruction. induced by minor trauma or abrasion (grade 4). Patients were encouraged to undergo feeding by mouth for as long as it was tolerable. Body weight and toxicity Transnasal esophagoscopy (TNE), flexible endoscopic evaluation of swallow (FEES), and aspiration pneumonia (dysphagia, xerostomia, mucositis, nausea, vomiting, constipation, diarrhea, dysguesia, difficulties chewing) work up were assessed and addressed with patients weekly. We Patients with grade 2 or greater dysphagia beyond 3 used the American Dietetic Association Medical Nutrition months after radiation were referred for TNE. The TNE Therapy (MNT) Protocol for Cancer (Radiation Oncol- technique has previously been described [15,16]. At the ogy) and the UCD Enteral Nutrition Guidelines. Deci- discretion of the physician, a FEES or aspiration pneumo- sions to wean off enteral feeds were based on individual nia work up (bacterial culture and chest X-ray) was per- patient and chemoradiation-induced toxicities with spe- formed. FEES allows direct visual assessment of many cific emphasis placed on inability to consume adequate swallowing functions including muscular function, pre- oral nutrition and fluid, dysphagia, and prevent uncon- mature spillage, pooling, laryngeal penetration, and pres- trolled involuntary weight loss. Patients were weaned off ence of aspiration. In brief, the patients were examined enteral nutrition support when 1) the patient's weight seated upright without anesthesia. Liquid (colored water), could be maintained with less than two cans of supple- pureed food (yogurt), and chewable food (bread) were mental feed per day, and 2) the patient could have certain ingested while the hypopharynx and laryngeal contents solid food without complaints of dysphagia, were viewed with the fiberscope. The results were scored odynophagia or aspiration. as "little", "moderate", or "severe" using the following var- iables: residue, penetration, and aspiration of three differ- Follow-up evaluation ent types of diet (water, yogurt, and bread), and mucus Patients were typically seen 2 to 3 weeks after completion stases. Aspiration pneumonia was defined as cultured bac- of radiation therapy and then every 3 months thereafter terial pneumonia with radiographic evidence of infiltra- for the first year, every 6 months for the second and third tion. year, and then annually. The mean follow up time was 16.2 months (range 4.5-52 months). If a persistent neck Statistical analysis node was found on physical examination after comple- Data analysis and graphs were completed using the R soft- tion of IMRT and/or was positive on PET/CT at 2 months ware program (R Development Core Team, 2006; R Foun- follow-up, salvage neck dissection was performed. Dura- dation for Statistical Computing, Vienna, Austria). tion of GT was defined as the interval between RT comple- Spearman's ρ and univariate regression were used to cal- tion and the date of its removal, or until the date of last culate the correlation of each of these identified DVH follow-up or death if the GT was still present. Prolonged parameters and individual dependent binary variable GT dependence was defined as GT more than the median (absence or presence of prolonged GT days, grade 3+ dys- GT duration. Two patients whose follow-up time were less phagia, and severe weight loss). A logistic model, p = 1/ than the mean GT duration were excluded from further {1+exp [-(α +β *dose or volume of structure)]}, was used analysis of clinical-dosimetric association. Body weight to calculate the probability of developing prolonged GT and patient-reported dysphagia were recorded during days, grade 3+ dysphagia, or severe weight loss. The each follow up. The lowest body weight during the follow unknown parameters α and β were estimated with the up period was used. Severe weight loss was defined as maximum likelihood method. A test was also performed more than 15% weight loss. Patient-reported dysphagia whether the hypothesis β = 0 can be rejected. A p value of was assessed with the validated UWQOL questionnaires < 0.05 was interpreted as being statistically significant given to patients during each follow-up visit. It contained from zero. Confidence intervals (95%) were determined. one swallowing question with five possible answers ("I Multivariate regression was not used due to the model swallow normally", (grade 0); "I cannot swallow certain instability caused by co-linearity between DVH parame- solid food", (grade 1); "I can only swallow soft food", ters (V40, V50, V60, V65, Dmax, Dmean). Wilcoxon rank- (grade 2); "I can only swallow liquid food", (grade 3); and sum analysis was preformed to identify DVH parameters "I cannot swallow", (grade 4). Grade 3 and grade 4 that statistically correlated with esophageal stricture. Page 4 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 within one year) was identified as the only significant fac- Results Swallowing outcomes after treatment tor predictive for prolonged GT dependence (p = 0.03). At 3 months and 6 months after treatment, 87% and 44% Other clinical factors, including age, gender, history of of patients, respectively, were GT dependent (Table 2). alcohol use, KPS, tumor site, T stage, N stage, and type of The results of physician-assessed high grade dysphagia chemotherapy regime are not associated with prolonged were consistent with that of GT dependence, given that GT dependence. Similar analysis of high grade dysphagia majority of high grade dysphagia patients were grade 3 revealed active smoking (p = 0.03) and T stage (p = 0.04) with GT dependence. Due to data redundancy, results of as significant factors. No other predisposing parameter observer-assessed high grade dysphagia were not was found to be statistically significant. In terms of severe reported. Using the UWQOL instrument, 17 patients weight loss, no predisposing parameter was identified to (44%) reported high grade dysphagia at any point during be statistically significant (data not shown). A total of 5 or after treatment. The median percent of maximum patients underwent post-treatment neck dissection. Both weight loss was 12% (range, -4% to 21%). univariate and multivariate analysis did not reveal neck dissection as a significant factor for prolonged GT depend- More than half (54%) of the patients had Grade 3 or 4 ence, high grade dysphagia, or severe weight loss. mucositis at some point after radiation, with 23% and 5% having severe mucositis at 3-month and 6-month follow DVH analysis for prolonged GT dependence up evaluation respectively (Table 2). However, GT The DVH parameters for all the swallowing structures dependence did not improve as rapidly and still persisted (SMPC, IPC, CPI, and CE) were listed in Table 3. Signifi- in 87% and 44% of patients at 3 months and 6 months, cant factors (p < 0.05) for prolonged GT dependence were respectively. There was a lack of temporal association revealed using Spearman's ρ test and subsequent univari- between high grade mucositis and prolonged GT depend- ate logistic regression in an attempt to identify dose-vol- ence of more than 192 days on statistical analysis (p > ume effect for GT duration longer than 192 days versus 0.05). Grade 2 or higher xerostomia was found in 43% less than 192 days. These factors are IPC V65 (p = 0.003), and 36% at 3-month and 6-month follow up evaluation IPC V60 (p = 0.002), IPC V50 (p = 0.042), IPC Dmean (p respectively, and persisted in 31% at the last follow-up = 0.016), and CPI Dmax (p = 0.011). CPI V60 has p value (Table 2). of 0.050. DVH analysis was also performed on a com- bined structure (IPC, CPI and CE). No statistically signifi- Twelve of the patients with high grade dysphagia under- cant factor was identified (p > 0.05). went TNE. Five of them developed stricture at the upper esophageal sphincter at the level of the cricopharyngeus The results of dose-response relationships and volume- muscle, including one with complete luminal stenosis. All response relationships for prolonged GT dependence are of them underwent dilatation at the time of TNE to relieve presented in Fig. 2. IPC V65 more than 30%, IPC V60 any physical obstruction. Four out of the seven patients more than 60%, IPC Dmean more than 60 Gy, and CPI who underwent FEES had finding of moderate or severe Dmax more than 62 Gy predicted for a greater than 50% aspiration to one of the diets. Eight patients had aspira- probability of developing prolonged GT dependence. For tion pneumonia work up, and only one of them was diag- IPC V50, the dose/volume-response relationships results nosed. were not clinically meaningful. The clinical factors listed in Table 1 were included in both In view of the strong dosimetric-clinical correlations for univariate and multivariate analysis of prolonged GT CPI and IPC, we repeated the above analysis with exclu- dependence. Smoking (active smoking or smoking history sion of three patients whose primary disease overlapped with the relevant structures (CPI and IPC). The same DVH Table 2: Toxicity after treatment parameters were observed as significant factors for pro- longed GT dependence. 3 month (%) 6 month (%) GT dependence 87% 44% DVH analysis for high grade dysphagia, severe weight loss, and stricture Self-reported dysphagia grade 3 33% 21% Spearman's ρ test and subsequent univariate logistic regression analysis revealed significant associations Mucositis 23% 5% between several dosimetric parameters and grade 3+ grade 3 patient-reported dysphagia. These factors are IPC V65 (p = 0.040), CPI Dmax (p = 0.037), and CPI V60 (p = 0.046). Xerostomia grade 2 43% 36% Further analysis of dose-response relationships and vol- ume-response relationships revealed that IPC V65 more Abbreviation: GT = gastrostomy tube Page 5 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 Table 3: Swallowing structure DVH parameters (median value and range) and p values for association with GT dependence V40 V50 V60 V65 Dmax Dmean (%) (%) (%) (%) (Gy) (Gy) CE GT > 192 d 57 (1-86) 24 (0-79) 0 (0-33) 0 (0-1) 61 (45-77) 36 (20-53) GT 192 d 65 (17-98) 27 (0-86) 0 (0-13) 0 (0-0) 57 (44-76) 36 (21-56) p = 2.671 p = 0.524 p = 0.173 P = 0.169 p = 0.238 p = 0.383 CPI GT > 192 d 100 (60-100) 100 (4-100) 8 (0-100) 0 (0-100) 64 (56-78) 57 (42-69) GT 192 d 100 (93-100) 95 (3-100) 0 (0-73) 0 (0-27) 58 (53-67) 53 (43-62) p = 0.512 p = 0.069 p = 0.050 P = 0.062 p = 0.011 p = 0.083 IPC GT > 192 d 100 (97-100) 100 (76-100) 81 (40-100) 42 (21-100) 74 (66-79) 64 (54-70) GT 192 d 100 (95-100) 92 (40-100) 37 (10-84) 15 (0-60) 72 (68-77) 55 (46-68) p = 0.367 p = 0.042 p = 0.002 P = 0.003 p = 0.057 p = 0.016 SMPC GT > 192 d 100 (60-100) 100 (56-100) 94 (45-100) 72 (21-93) 76 (66-79) 67 (38-72) GT 192 d 100 (84-100) 97 (70-100) 90 (25-100) 60 (0-95) 76 (70-79) 65 (50-71) p = 0.378 p = 0.072 p = 0.063 P = 0.091 p = 0.252 p = 0.086 Abbreviation: DVH = Dose-volume histogram; GT = gastrostomy tube; SMPC = superior and middle pharyngeal constrictor; IPC = inferior pharyngeal constrictors; CPI = cricoid pharyngeal inlet; and CE = cervical esophagus; Dmax = maximum dose; Dmean = mean dose. than 65%, CPI V60 more than 78%, CPI Dmax more than aspiration or stricture development [17]. Levendag et al 70 Gy were associated with more than 50% probability of identified dose-response relationship between dysphagia developing high grade dysphagia. Similar analysis did not for solids (p < 0.02) or aspiration episodes (p < 0.02) and reveal statistically significant DVH predictors for severe mean dose to IPC. A mean dose of 33 Gy to IPC was esti- weight loss (data not shown). Wilcoxon rank-sum analy- mated as the threshold for 20% risk of dysphagia for sol- sis revealed significant associations between stricture and ids [14]. Furthermore, Dornfeld et al reported that a more two dosimetric parameters (CPI V65, CPI Dmax). restrictive diet one year after treatment is significantly cor- related with higher average dose delivered to the constric- Discussion tor muscles (lateral pharygeal wall) at the level of false It has been a common observation that a correlation exists vocal cord [18]. Jensen et al demonstrated that dose above between dysphagia and radiation doses to the anatomic 60 Gy to the upper esophageal sphincter could result in structures responsible for swallowing in patients undergo- higher risk of late swallowing dysfunction [19]. This well ing definitive chemoradiation for head and neck cancer. documented association between high dose to IPC or CPI However, the present study is the first to document a rela- and prolonged GT dependence was also supported by two tionship between various dosimetric parameters and pro- earlier reports showing that patients were more likely to longed GT dependence. Notably, we were able to identify have prolonged GT dependence and high grade dysphagia DVH parameters which were significantly associated with when treated with extended-field IMRT rather than being prolonged GT dependence, including V65 of the IPC, V60 treated with an upper IMRT fields junctioned with an of the IPC, mean dose to the IPC, and maximum dose to anterior neck field. This is thought to be due to the pres- the CPI. Based on these dose/volume-response relation- ence of midline block in an anterior neck field to prevent ships, we currently recommend IPC V65 less than 15%, unanticipated high dose radiation to structures including IPC V60 less than 40%, IPC Dmean less than 55 Gy, and larynx, IPC, CPI and CE [20,21]. CPI Dmax less than 60 Gy as potentially important DVH constraints to guide IMRT planning in an attempt to sig- The significant dose-volume effect relationships regarding nificantly reduce the risk of swallowing dysfunction and prolonged GT dependence for IPC and CPI could be prolonged GT dependence. explained by the role of the upper esophageal sphincter (UES) in the normal swallowing process. The UES is a Our findings demonstrate the importance of IPC and CPI functional entity that is composed of three muscles: the dosimetric parameters for developing swallowing dys- IPC muscle, the CPI muscle, and the upper esophageal function and are consistent with those from several muscle. The UES opens by relaxation of the three closing recently published studies. Caglar et al showed that a muscles, traction by IPC and other muscles that attached mean dose to the IPC of more than 54 Gy and IPC V50 of to the hyoid bone and thyroid cartilage, anterior move- more than 50% were the most significant predictors for ment of the larynx, and pulsion of the bolus. The various Page 6 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 Volume-response or dose-r u CPI (D) Figure 2 me of the IPC receiving mo espon re tha sn e relationsh 6500 cGy (A), ip for the average or 6000 cGy (B prob ), or the mean dose to the ability of having prolonged GT dependence and the vol- IPC (C), or maximum dose to Volume-response or dose-response relationship for the average probability of having prolonged GT depend- ence and the volume of the IPC receiving more than 6500 cGy (A), or 6000 cGy (B), or the mean dose to the IPC (C), or maximum dose to CPI (D). GT = gastrostomy tube; IPC = inferior pharyngeal constrictors; CPI = cricoid pha- ryngeal inlet. The š lines plot the mean risk; the - lines plot the estimated upper and lower limits of 95% confidence interval. The ♦ points depict the observed values. muscles of the UES behave differently during its many tified. A cause of dysphagia could also be attributed in dynamic states, so that similar functions are accomplished part to the failure of sensation and timely response to the by different muscles. Any impairment of the CPI and IPC bolus passing through this region. The importance of IPC could result in dysphagia. In addition, UES is considered and CPI is validated by our finding that patient-reported a high pressure zone, with the highest pressure at the dysphagia was highly correlated with the dose to the two region around IPC where proprioceptive units were iden- structures. Page 7 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 In addition to the IPC and CPI, several other anatomic with both high grade dysphagia and stricture formation. structures have been reported as dysphagia/aspiration In spite of this high incidence among patients with high related with significant dose-volume relationship. These grade dysphagia, the overall incidence of stricture in all structures include GSL and PC, with superior PC having patients is 12% (5/41) in our study, consistent with stric- the strongest dose-response association [11,12]. The ture rate of 17%-37% in other studies [17,26-28]. importance of superior and middle PC for swallowing after radiation therapy was also shown by Teguh et al It is important to note that this study was retrospective [14,22,23]. Although our study failed to find a significant with inherent limitations. First, the lack of systematic eval- a correlation between GT dependence and dose to the uation of some of the major end points of late dysphagia GSL, SMPC, or PC as a whole, this could potentially be using TNE or FEES prevented more robust analysis using explained by the differences in patient characteristics. In more objective endpoints. Another limitation of the study the above mentioned studies, only oropharynx and was the relatively short follow-up with a median duration nasopharynx patients were included. As such, our results of 15.6 months. However, most of the endpoint events are consistent with those from Caglar et al that the mean occurred less than 1 year after treatment. Given the small dose or V50 to IPC, not the superior PC, were significant number of events, we conjecture that a sub-analysis of predictors for aspiration or stricture development [17]. swallowing function assessed at greater than 1 year after treatment would likely not change our findings. Nonethe- Prolonged GT dependence is regarded by most head and less, we do acknowledge that further studies with more neck cancer patients as contributing to compromised comprehensive objective endpoints with prolonged fol- quality of life because it may cause infection and physical low-up may be necessary to yield a more thorough evalu- discomfort, distort patient's self-esteem, and induce anxi- ation. This is well exemplified in a recently published ety, depression and social isolation [8]. This is of increas- retrospective study of patients with more than 1 year fol- ing concern in recent years when concurrent low-up, where a composite of 3 objective endpoints (GT chemoradiation for tumor control and organ preservation dependence, aspiration, and pharyngoesophageal stric- has gained widespread practice but is associated with high ture) were successfully used as surrogates for severe long- rate of severe late dysphagia, including prolonged GT term dysphagia [26]. dependence [5]. Multiple large randomized trials testing intensified chemoradiation regimens reported GT rates of It must also be recognized that the majority of patients in about 70%, and chronic tube dependence of 10-20% the present study presented with oropharynx cancer, and [3,6,7]. In a recent study where 95% of the chemoradia- this fact may have biased our findings. Teguh et al, for tion patients had prophylactic feeding tubes placed before instance, demonstrated that patients with base of tongue treatment, Caglar et al reported prolonged GT dependence disease experienced more severe dysphagia than those in 37% of the patients, with a median GT duration of 112 with tumors at other sites [22]. In addition to oropharynx days after radiation completion [17]. Notably, we also primary, the larynx, hypopharynx and pharyngeal wall identified smoking as a risk factor for GT dependence. The were also found to predispose to dysphagia more so than etiology for smoking induced dysphagia is likely multifac- other regions of the head and neck [26-29]. In contrast, torial and related to prolonged tissue recovery secondary however, Logemann et al showed no differences in the fre- to nicotine induced hypoxia, the appetite reducing effects quency of dysphagia across different head and neck dis- of nicotine, or mucosal irritation. Multiple previous works ease sites [30], which is further supported by a large have similarly associated smoking with higher rates of prospective study that excluded disease site as a statisti- toxicity including aspiration and esophageal stricture after cally significant factor for quality of life changes among radiation therapy [24,25]. head and neck cancer treated with radiation therapy [31]. Notably, 5 out of the 12 (42%) patients with high grade Lastly, we were unable to control for potentially con- dysphagia developed upper esophageal stricture in this founding factors which may have also predisposed to study. This high incidence could have resulted from detec- swallowing dysfunction including severe mucositis, pre- tion bias, small patient number, or most likely, patient treatment dysphagia and post-treatment xerostomia. We over-reliance on a GT which led to less swallowing and acknowledge that the endpoints in this study, such as high allowing scar and stricture formation. The last possibility grade dysphagia and prolonged GT dependence may have is supported by results from Caudell et al who demon- been confounded by the development of severe acute strated a trend toward an association (p = 0.09) between mucositis or its consequential late effects, such as submu- GT dependence and pharyngeal stricture or stenosis [26]. cosal edema, fibrosis, scarring, soft tissue necrosis, Another explanation for this high incidence of stricture impaired sensory or motor function, and loss of mucosal among patients with high grade dysphagia could be its rel- compliance. This is based on the rationale that high dose atation with CPI Dmax, which was significantly associated radiation to a large volume of the constrictor muscles Page 8 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 (resulting in high values of V60 and V65) also results in ric constraints should not replace the effort of early swal- high dose to large volumes of mucosal surface which is lowing therapy and exercises which resulted in maximal believed to lead to more severe mucositis [32-34]. How- swallowing recovery in several studies [5,40]. ever, since mucositis tends to be self-limiting and an acute, rather than late side-effect of radiation therapy, we Competing interests believe that the potential confounding effects are mini- The authors declare that they have no competing interests. mal. Furthermore, the present study demonstrated a lack of significant association between acute mucositis and Authors' contributions prolonged GT dependence, which is consistent with the BL and AMC conceived of the study, and participated in its findings of a dissociation between acute mucositis and design, carried out data collection, data analysis, manu- dysphagia by Mekhail et al [27] and it is supported by script writing, and coordination. DL and DMR performed recent data from Anand et al showing no correlation statistics analysis. DHL performed data collection regard- between long-term dysphagia and acute mucositis (Grade ing chemotherapy regimens and participated in manu- 3, 4) in spite of the severe mucosities that developed in script writing. DGF and QL performed dysphagia data 53% of locally advanced head neck cancer patients treated collection including TNE and FEES, and delineation of with IMRT [35]. In addition, we were unable to com- swallowing structures. KN and JC performed data collec- pletely exclude the possibility of other confounding fac- tion regarding GT management. JAP performed physics tors such as pre-treatment dysphagia or post-treatment consult on re-computation of DVH. All authors read and severe xerostomia as confounding factors. Multiple stud- approved the final manuscript. ies suggested that very few patients with newly diagnosed head and neck cancer have severe dysphagia or aspiration Acknowledgements This study was presented in abstract form at the 2008 annual meeting of prior to definitive treatment [29,36,37]. Moreover, we the American Society for Therapeutic Radiology and Oncology (ASTRO) in could not completely rule out xerostomia as a confound- Boston, Massachusetts. ing factor in spite of the reduction in the risk of this symp- tom associated with parotid gland sparing IMRT. References Xerostomia secondary to chemotherapy or radiation ther- 1. Adelstein DJ, Li Y, Adams GL, Wagner H Jr, Kish JA, Ensley JF, apy has been suggested to significantly affect food bolus Schuller DE, Forastiere AA: An Intergroup phase III comparis- ion of standard radiation therapy and two schedules of con- formation and swallowing function, and contribute sig- current chemoradiotherapy in patients with unresectable nificantly to dysphagia after chemoradiation [38,39]. In squamous cell head and neck cancer. J Clin Oncol 2003, particular, Teguh et al demonstrated a strong correlation 21:92-98. 2. Denis F, Garaud P, Bardet E, Alfonsi M, Sire C, Germain T, Bergerot between dry mouth and sticky saliva with dysphagia- P, Rhein B, Tortochaux J, Calais G: Final results of the 94-01 related quality of life such as normalcy of diet and French Head and Neck Oncology and Radiotherapy Group randomized trial comparing radiotherapy alone with con- odynophagia [22]. As a result of these potential con- comitant radiochemotherapy in advanced-stage oropharynx founding factors, it remains difficult to definitively estab- carcinoma. J Clin Oncol 2004, 22:69-76. lish a cause-effect association in spite of the significant 3. Forastiere AA, Goepfert H, Maor M, Pajak TF, Weber R, Morrison W, Glisson B, Trotti A, Ridge JA, Chao C, Peters G, Lee DJ, Leaf A, dose/volume-response relationship between some key Ensley J, Cooper J: Concurrent chemotherapy and radiother- DVH parameters and risk of prolonged GT dependence. A apy for organ preservation in advanced laryngeal cancer. N larger prospective study is required in the future to further Engl J Med 2003, 349:2091-8. 4. Grégoire V, De Neve W, Eisbruch A, Lee N, Weyngaert D Van den, investigate theses associations. Van Gestel D: Intensity-modulated radiation therapy for head and neck carcinoma. Oncologist 2007, 12:565-8. 5. Rosenthal DI, Lewin JS, Eisbruch A: Prevention and treatment of Conclusion dysphagia and aspiration after chemoradiation for head and We provided evidence that prolonged GT is correlated neck cancer. J Clin Oncol 2006, 24:2636-43. with DVH parameters for patients with locally advanced 6. Garden AS, Harris J, Vokes EE, Forastiere AA, Ridge JA, Jones C, Hor- witz EM, Glisson BS, Nabell L, Cooper JS, Demas W, Gore E: Pre- head and neck cancer treated with definitive concurrent liminary results of Radiation Therapy Oncology Group 97- chemotherapy and IMRT. To minimize the risk of pro- 03: a randomized phase II trial of concurrent radiation and chemotherapy for advanced squamous cell carcinomas of longed GT dependence, we currently strive to keep IPC the head and neck. J Clin Oncol 2004, 22:2856-64. V65 less than 15%, IPC V60 less than 40%, and maintain 7. Adelstein DJ, Lavertu P, Saxton JP, Secic M, Wood BG, Wanamaker IPC Dmean less than 55 Gy, and CPI Dmax less than 60 JR, Eliachar I, Strome M, Larto MA: Mature results of a phase III randomized trial comparing concurrent chemoradiotherapy Gy during IMRT planning in an attempt to decrease the with radiation therapy alone in patients with stage III and IV risk of prolonged GT dependence. It should be noted that squamous cell carcinoma of the head and neck. Cancer 2000, these guidelines are implemented on a case-by-case basis 88:876-83. 8. Nguyen NP, Frank C, Moltz CC, Vos P, Smith HJ, Karlsson U, Dutta considering such factors as tumor extent and location. S, Midyett A, Barloon J, Sallah S: Impact of dysphagia on quality Future directions include large-scale prospective trials of life after treatment of head-and-neck cancer. Int J RadiatOn- col Biol Phys 2005, 61:772-8. aiming to assess the clinical benefits gained by applying 9. Poulsen MG, Riddle B, Keller J, Porceddu SV, Tripcony L: Predictors these dosimetric strategies. Lastly, the proposed dosimet- of acute grade 4 swallowing toxicity in patients with stages Page 9 of 10 (page number not for citation purposes) Radiation Oncology 2009, 4:52 http://www.ro-journal.com/content/4/1/52 III and IV squamous carcinoma of the head and neck treated locally advanced head-and-neck cnacer. Int J Radiat Oncol Biol with radiotherapy alone. Radiother Oncol 2008, 87:253-9. Phys, Int J Radiat Oncol Biol Phys 2009, 73:410-5. 10. Mendenhall WM, Morris CG, Amdur RJ, Hinerman RW, Mancuso AA: 27. Mekhail TM, Adelstein DJ, Rybicki LA, Larto MA, Saxton JP, Lavertu Parameters that predict local control after definitive radio- P: Enteral nutrition during the treatment of head and neck therapy for squamous cell carcinoma of the head and neck. carcinoma: is a percutaneous endoscopic gastrostomy tube Head Neck 2003, 25:535-42. preferable to a nasogastric tube? Cancer 2001, 91:1785-90. 11. Eisbruch A, Schwartz M, Rasch C, Vineberg K, Damen E, Van As CJ, 28. Lee WT, Akst LM, Adelstein DJ, Saxton JP, Wood BG, Strome M, But- Marsh R, Pameijer FA, Balm AJ: Dysphagia and aspiration after ler RS, Esclamado RM: Risk factors for hypopharyngeal/upper chemoradiotherapy for head-and-neck cancer: which ana- esophageal stricture formation after concurrent chemoradi- tomic structures are affected and can they be spared by ation. Head Neck 2006, 28:808-12. IMRT? Int J Radiat Oncol Biol Phys 2004, 60:1425-39. 29. Machtay M, Moughan J, Trotti A, Gargen AS, Weber RS, Cooper JS, 12. Feng FY, Kim HM, Lyden TH, Haxer MJ, Feng M, Worden FP, Chep- Forastiere A, Ang KK: Factors associated with severe late tox- eha DB, Eisbruch A: Intensity-modulated radiotherapy of head icity after concurrent chemoradiation for locally advanced and neck cancer aiming to reduce dysphagia: early dose- head and neck cancer: an RTOG analysis. J Clin Oncol 2008, effect relationships for the swallowing structures. Int J Radiat 26:3581-88. Oncol Biol Phys 2007, 68:1289-98. 30. Logeman JA, Rademaker AW, Pauloski BR, Lazarus CL, Mittal BB, 13. Schmalfuss IM, Mancuso AA, Tart RP: Postcricoid region and cer- Brockstein B, MacCracken E, Haraf DJ, Vokes EE, Newman LA, Liu D: vical esophagus: normal appearance at CT and MR imaging. Site of disease and treatment protocol as correlated of swal- Radiology 2000, 214:237-46. lowing function in patients with head and neck cancer 14. Levendag PC, Teguh DN, Voet P, Est H van der, Noever I, de Kruijf treated with chemoradiaiton. Head Neck 2006, 28:64-73. WJ, Kolkman-Deurloo IK, Prevost JB, Poll J, Schmitz PI, Heijmen BJ: 31. Langendijk JA, Doornaert P, Verdonck-de Leeuw IM, Leemans CR, Dysphagia disorders in patients with cancer of the orophar- Aaronson NK, Slotman BJ: Impact of late treatment-related ynx are significantly affected by the radiation therapy dose to toxicity on quality of life among patients with head and neck the superior and middle constrictor muscle: a dose-effect cancer treated with radiotherapy. J Clin Oncol 3776, relationship. Radiother Oncol 2007, 85:64-73. 26:3770-2008. 15. Postma GN, Cohen JT, Belafsky PC, Halum SL, Gupta SK, Bach KK, 32. Rosenthal DI, Trotti A: Strategies for managing radiation- Koufman JA: Transnasal esophagoscopy: revisited (over 700 induced mucositis in head and neck cancer. Semin Radiat Oncol consecutive cases). Laryngoscope 2005, 115:321-323. 2009, 19:29-34. 16. Langmore SE, Schatz K, Olsen N: Fiberoptic endoscopic exami- 33. Dorr W, Hendry JH: Consequential late effects in normal tis- nation of swallowing safety: a new procedure. Dysphagia 1998, sues. Radiother Oncol 2001, 61:223-231. 2:216-9. 34. Trotti A, Bellm LA, Esptein JB, Frame D, Fuchs HJ, Gwede CK, 17. Caglar HB, Tishler RB, Othus M, Burke E, Li Y, Goguen L, Wirth LJ, Komaroff E, Nalysnyk L, Zilberberg MD: Mucositis incidence, Haddad RI, Norris CM, Court LE, Aninno DJ, Posner MR, Allen AM: severity and associated outcomes in patients with head and Dose to Larynx Predicts for Swallowing Complications After neck cancer receiving radiotherapy with or without chemo- Intensity-Modulated Radiotherapy. Int J Radiat Oncol Biol Phys therapy: a systemic literature review. Radiother Oncol 2003, 1118, 72:1110-2008. 66:253-262. 18. Dornfeld K, Simmons JR, Karnell L, Karnell M, Funk G, Yao M, Wacha 35. Anand A, Chaudhoory A, Shukla A, Negi PS, Sinha SN, Babu AA, J, Zimmerman B, Buatti JM: Radiation doses to structures within Munjal RK, Dewan AK, Kumar K, Doval DC, Vaid AK: Favourable and adjacent to the larynx are correlated with long-term impact of intensity-modulated radiation therapy on chronic diet- and speech-related quality of life. Int J Radiat Oncol Biol Phys dysphagia in patients with head and neck cnacer. Br J Radiol 2007, 68:750-7. 2008, 81:865-71. 19. Jensen K, Lambertsen K, Grau C: Late swallowing dysfunction 36. Pauliski BR, Rademake AW, Logemann JA, Sten D, Beery Q, Newman and dysphagia after radiotherapy for pharynx cancer: fre- L, Hanchett C, Tusant S, MacCracken E: Pretreatment swallowing quency, intensity and correlation with dose and volume function in patients with head and neck cancer. Head Neck parameters. Radiother Oncol 2007, 85:74-82. 2000, 22:474-82. 20. Fua TF, Corry J, Milner AD, Cramb J, Walsham SF, Peters LJ: Inten- 37. Stenson KM, MacCracken E, List M, Haraf DJ, Brockstein B, Weich- sity-modulated radiotherapy for nasopharyngeal carcinoma: selbaum R, Vokes EE: Swallowing function in patients with head clinical correlation of dose to the pharyngo-esophageal axis and neck cancer prior to treatment. Arch Otolaryngol Head Neck and dysphagia. Int J Radiat OncolBiol Phys 2007, 67:976-81. Surg 2000, 126:371-7. 21. Lee NY, O'Meara W, Chan K, Della-Bianca C, Mechalakos JG, Zhung 38. Logemann JA, Pauloski BR, Rademaker AW, Lazarus CL, Mittal B, J, Wolden SL, Narayana A, Kraus D, Shah JP, Pfister DG: Concur- Gaziano J, Stachowiak L, MacCracken E, Newman LA: Xerostomia: rent chemotherapy and intensity-modulated radiotherapy 12-month changes in saliva production and its relationship to for locoregionally advanced laryngeal and hypopharyngeal perception and performance of swallow function, oral cancers. Int J RadiatOncol Biol Phys 2007, 69:459-68. intake, and diet after chemoradiation. Head Neck 2003, 22. Teguh DN, Levendag PC, Noever I, van Rooij P, Voet P, Est H van der, 25:432-7. Sipkema D, Sewnaik A, Baatenburg de Jong RJ, de la Bije D, Schmitz 39. Hamlet S, Faull J, Klein B, Aref A, Fontanesi J, Stachler R, Shamsa F, PI: Treatment Techniques and Site Considerations Regard- Jones L, Simpson M: Mastification and swallowing in patients ing Dysphagia-Related Quality of Life in Cancer of the with postirradiation xerostomia. Int J Radiat Oncol Biol Phys 1997, Oropharynx and Nasopharynx. Int J Radiat Oncol Biol Phys 2008, 37:789-96. 72:1119-27. 40. Shaker R, Easterling C, Kern M, Nitschke T, Massey B, Daniels S, 23. Teguh DN, Levendag PC, Sewnaik A, Hakkesteegt MM, Noever I, Grande B, Kazandjian M, Dikeman K: Rehabilitation of swallowing Voet P, Est H van der, Sipkema D, van Rooij P, Baatenburg de Jong RJ, by exercise in tube-fed patients with pharyngeal dysphagia Schmitz PI: Results of fibroptic endoscopic evaluation of swal- secondary to abnormal UES opening. Gastroenterology 2002, lowing vs. radiation dose in the swallowing muscles after 122:1314-21. radiotherapy of cancer in the oropharynx. Radiother Oncol 2008, 89:57-63. 24. Jensen K, Jensen AB, Grau C: Smoking has a negative impact upon health related quality of life after treatment for head and neck cancer. Oral Oncol 2007, 43:187-92. 25. Mangar S, Slevin N, Mais K, Sykes A: Evaluating predictive factors for determining enteral nutrition in patients receiving radi- cal radiotherapy for head and neck cancer: A retrospective review. Radiother Oncol 2006, 78:152-158. 26. Caudell J, Shaner P, Meredith R, Bonner J, Locher JL, Nabell LM, Car- roll WR, Magnuson JS, Spencer SA, Bonner JA: Factors associated with long-term dysphagia after definitive radiotherapy for Page 10 of 10 (page number not for citation purposes)

Journal

Radiation OncologySpringer Journals

Published: Nov 12, 2009

References