Effective delivery of palliative radiotherapy: A prospective study.

PURPOSE: The main goal of palliative radiotherapy is to reduce patient's discomfort. But sometimes patients do not receive any benefits from this treatment because of rapid worsening of their general condition. This prospective monocentric study assessed the effective delivery of palliative radiotherapy. MATERIALS AND METHODS: From 1st December 2015 to 29th February 2016, all consecutive patients receiving palliative radiotherapy in our hospital were included. The primary endpoint was the effective delivery of palliative radiotherapy according to the initial prescription (total dose, overall treatment time and fractionation). The secondary endpoints were the number of treatment breaks, the clinical benefit, the number of deaths and the delays for admission in the palliative care unit. RESULTS: Fifty-nine patients were included and 64 treatments were analysed. The treatment sites were: bone (70.3%) and brain (21.9%). The treatment goals were: pain control only (43.8%), decompression only (21.9%), pain control and decompression (32.8%), haemostatic aim (1.6%). Palliative treatment was achieved in 57 cases (89%). Temporary interruption of the radiotherapy treatment was necessary in six cases (9.4%; three for medical reason, three for logistic reason). The main reason of permanent interruption was worsening of performance status (seven cases). Palliation of symptoms (complete or partial responses) was obtained in 44 cases (68.8%). Seven patients (11.9%) died during the month after the end of the treatment. No delay or cancellation for admission in the palliative care unit were observed. CONCLUSION: Palliative radiotherapy was completed as originally planned in 51 cases (79.9%) with a clinical benefit for 44 cases (68.8%). Radiation therapy must not be neglected as a palliative treatment at the end-of-life.

Multimodal management of primary adenocarcinoma of the female urethra: About four cases.

PURPOSE: To retrospectively analyse female patients treated for urethral adenocarcinoma, modalities of treatment and long-term outcomes. PATIENTS: Four cases of primary female urethral adenocarcinoma were treated in the departments of urology and radiation-oncology at Georges-Pompidou and Necker hospitals (France) over a 7-year period. RESULTS: All of them underwent surgery, with three presenting stage pT3-pT4 and one a positive cytology on inguinal node. Three patients received adjuvant cisplatin-based chemoradiotherapy up to 60Gy, and one preoperative chemoradiotherapy at 45Gy. Two recurrences were observed: one local relapse occurred at 9months from the diagnosis and was treated by anterior pelvic exenteration followed by chemoradiotherapy, with no recurrence. One tumour relapsed both at the local level and on distant metastatic sites at 9months from the diagnosis, and died 21months after this progression. Median survival and progression-free survival are respectively 4.2years and 13months. Three patients are alive at 7, 4.5 and 3years from diagnosis. CONCLUSION: Female urethral adenocarcinoma is a very rare entity and often present in locally advanced stages. Initial extensive surgery with pelvic exenteration should be considered, followed by chemoradiotherapy according to the surgical margins and lymph nodes involvement.

[Overall treatment time assessment of radiotherapy delivered in a curative intent: A prospective study]. / Respect de l'étalement de la radiothérapie à visée curative : étude prospective.

PURPOSE: To assess the overall treatment time of radiotherapy delivered with curative intent in a cohort of 185 consecutive patients and the causes of this possible delay. If delay, to propose corrective actions. MATERIALS AND METHODS: We report a single-center prospective study including all consecutive patients receiving a radiation therapy with curative intent, from 1st December 2013 to 28th February 2014, on the three linear accelerators of the radiotherapy department. For each fraction missed, the causes of non-completion were prospectively collected. This analysis took into account the following parameters: age, sex, occupation, transport type and duration, tumour localization, radiation dose, concomitant chemotherapy, hospitalization, type of linear accelerator. RESULTS: One hundred and fifty-five patients were included in the study (183 evaluable, two did not complete treatment). The overall treatment time was respected for 31 patients (17%). It was lengthened on 4.6 days (d) (0-29 d; median: 3d). The mean number of delayed fractions was 3.4 (0-17; median: 2). The reasons of delay were: breakdown 32.2%; maintenance 29.3%; holiday 11%; treatment toxicity 9.4%; inadequate planning 8.6%; other disease 3.9%; treatment refusal 2.1%; unspecified personal reasons 1.5%; refusal to wait 0.8%; transportation 0.3%; error of communication 0.3%; other 0.6%. Two parameters had a significant impact on the overall treatment time: the type of linear accelerator in relation to breakdown occurrence (P<0.001) and transportation duration (more or less than 40 min, P=0.022). One hundred and thirty-eight patients (75%) and 79 patients (43%) were treated on one or more than two Saturdays. Treatment on Saturday significantly reduced the overall treatment time (median, 1.9 d; P<0.001). CONCLUSION: The overall treatment time was globally respected excluding break down and maintenance. Their impact on the overall treatment time is minimized by the systematic opening of the department on Saturdays.

[New techniques in thoracic radiation therapy]. / Apport des nouvelles techniques en radiothérapie thoracique.

Advanced technologies have led to an improvement of target volume delineation and a higher accuracy in dose delivery. Stereotactic body radiotherapy, intensity-modulated radiotherapy and respiratory gating allow new therapeutic perspectives along with an improvement of the therapeutic ratio. Ongoing trials aim to show the magnitude of gains in patient care with technical improvements.

[Tolerance and efficacy of preoperative radiation therapy for elderly patients treated for rectal cancer]. / Tolérance et efficacité de la radiothérapie préopératoire chez les patients âgés atteints d'un cancer du rectum.

PURPOSE: To retrospectively assess the impact of age on tolerance and oncologic outcomes treated by neoadjuvant treatment for patients of 70 years old or above with locally advanced rectal cancer. PATIENTS AND METHODS: Ninety-one consecutive patients were divided into three groups: group 1 from 70 to 75 years (n=31); group 2: 76 to 79 years (n=31) and group 3, patients aged 80 years or above (n=29). Radiation therapy was delivered according two schemes: 25Gy in five fractions (short scheme) or 45 to 50Gy with a classical fractionation (long scheme). Long scheme patients received a concomitant chemotherapy with 5-fluoro-uracile alone or associated with oxaliplatin. RESULTS: The three groups were comparable for performance status, Charlson's score and T staging. Long scheme radiation therapy and chemotherapy were performed in 77.5, 74.5 and 48.3% of patients (P=0.03) and 77.4, 71 and 41.4% (P=0.006) in the groups 1, 2 and 3, respectively. All patients treated with the short scheme irradiation received the treatment without any acute toxicity. In the long scheme group, 65% of patients received the treatment on time and grade 3 or above toxicity was observed in 12% of patients who did not receive oxaliplatin and in 48% of patients who received oxaliplatin. The overall survival rate at 3 and 5 years was 66.9% and 60.8% in the group 1, 90.5% and 75.9% in the group 2 and 80.5% and 73.8% in the group 3 (P=0.15). CONCLUSION: Neoadjuvant treatment is feasible with encouraging survival rates for patients aged 70 years and older. Short scheme radiation therapy seems to be an interesting option in this population.

[Quality control of dose delivered by in vivo dosimetry: can one tolerance be used for all localizations?]. / Contrôle qualité de la dose délivrée par dosimétrie in vivo : un critère de tolérance unique peut-il satisfaire toutes les localisations ?

PURPOSE: In vivo dosimetry measurements are accepted when the difference between measured and calculated dose is under 5%. A statistical analysis has been conducted to determine whether this tolerance matched the clinical practice for the studied localizations: pelvis, thorax, head and neck, breast. MATERIALS AND METHODS: The technical characteristics of the detectors were checked before being used in clinical practice. Then an automatic statistical analysis was implemented using the 2450 in vivo dosimetry measurements obtained during 1 year. MAIN RESULTS: The global average is 1.10%, the standard deviation 2.46% and the percentage of out of level measurements 4.09%. By distinguishing the localizations, the 5% tolerance appeared to be too narrow for the breast localization. DISCUSSION/CONCLUSION: Several investigations were initiated to justify the modification of the tolerance for the breast localization. They highlighted an underestimation of the calculated dose when high beam angles are set: a new correction factor was defined to take account this error. A specific tolerance was also specified for the breast localization.

[Chemoradiation in bladder cancer]. / Chimioradiothérapie concomitante des carcinomes de la vessie.

Bladder cancer is the second most common urologic tumor after prostate cancer. Radical cystectomy is the standard treatment of localized muscle-invasive tumors. However, urinary diversion (using a conduit or continent diversion) following radical cystectomy can be debilitating. Moreover, delayed metastases appear frequently in spite of radical surgery. So, for selected patients, chemoradiotherapy is a valid therapeutic alternative to cystectomy. Cisplatin or derivatives are administered concurrently to radiation therapy up to 60-65 Gy. Patients undergo control cystoscopy at mid-course in treatment in order to select responders from non responders. This review summarizes the main published series of radiochemotherapy in invasive bladder cancer. Results for local control, survivals, bladder preservation rates and toxicity are presented.

[State of the art and advances in radiotherapy for bladder cancer]. / Techniques et innovations en radiothérapie pour le traitement conservateur des cancers infiltrants localisés de vessie.

Radical cystectomy is the treatment of choice for nonmetastatic, muscle-infiltrating bladder cancer. However, bladder-sparing approaches can be discussed in carefully selected patients. Bladder-preservation protocols aim to guaranty local control and survival with a functional bladder and a good quality of life. Such strategies include combinations of transurethral resection and radiochemotherapy, partial cystectomy and brachytherapy, radiotherapy-cystotomy and electrontherapy. Strict selection criteria and close follow-up are mandatory. New irradiation techniques hold the promise to improve local control by selectively boosting the dose to the tumor while better sparing the organs at risk. Such advances include the use of multimodal imaging, image-guided radiotherapy, concomitant boost with conformal irradiation+/-intensity modulated radiation therapy. Brachytherapy, either high-dose or pulsed-rate, is a promising technique for selected cases. Highly-conformal irradiation with tumor tracking using the Cyberknifetrade mark technology may also provide opportunities to boost the tumor while reducing toxicities. Specific innovative irradiation techniques are discussed.

Skin metastasis of head and neck carcinoma predictive for dismal outcome.

A 64-year-old female with locally advanced oropharyngeal carcinoma presented with an innocuous appearing macule on the abdomen. The lesion rapidly enlarged over 2 weeks into an inflammatory 5 cm fleshy nodule that was diagnosed as squamous cell carcinoma (SCC) and was found to overexpress epidermal growth factor receptor (EGFR). A fatal outcome occurred 3 months after the initial diagnosis of cancer, in spite of chemotherapy and treatment with EGFR inhibitors (cetuximab). Cutaneous metastases occur in 10 percent of squamous cell carcinomas of the head and neck. Contiguous cutaneous metastases in the head and neck areas are by far the most common. Conversely, isolated infradiaphragmatic cutaneous metastases are exceedingly rare and are associated with an aggressive clinical course. In a patient with cancer, the possibility of distant skin metastasis should be considered whenever new cutaneous nodules appear.

[Postoperative treatment of malignant tumors of the parotid gland: radiotherapy, concomitant chemotherapy and radiation therapy?]. / Tumeurs malignes de la parotide: prise en charge multidisciplinaire, rôle de la radiothérapie.

The low incidence, the wide histological spectrum and the natural, sometimes slow, evolution of malignant parotid gland tumours do not allow to easily establish the impact of their treatments. At present, clinical data come from retrospective cohort analysis, whereas randomized phase II-III trials are confidential. Moreover conclusions from these studies are controversial. We present here the therapeutic highlights of parotid cancers: surgery, with the particular concern of facial nerve preservation; and adjuvant treatment essentially based on radiotherapy. This review focused on adjuvant care explores the indications and the technical aspects of radiation, as well as the role of concurrent chemotherapy.

[Clinical and dosimetric analysis of 469 prostate cancer patients treated in France in 2005 by permanent implant brachytherapy using the Iodin 125 seeds IsoSeed Bebig: report to the French Economic committee of health products (CEPS)]. / Analyse clinique et dosimétrique de 469 patients traités en France en 2005 par implantation prostatique de grains d'iode 125 IsoSeed Bebig: rapport pour le Comité économique des produits de santé (CEPS).

A French decree of February 3rd 2005, allowed the Iodin 125 seeds from several companies to be reimbursed after a permanent implantation brachytherapy for a prostate cancer. Within this frame, the French "Comité économique des produits de santé" (CEPS; Economic committee for health products) made mandatory the annual writing and publication of a follow-up study with three main aims; make sure that the seeds were used for prostate cancer patients with criterias corresponding to the national recommendations, analyze the quality of the dosimetric data, and report all side effects, complications and possible accidents. We therefore report here a clinical and dosimetric analysis of 469 patient cases treated in France in nine centers in 2005 with the Iodin 125 IsoSeed Bebig. This analysis shows that: 1) The national recommendations for selecting patients for exclusive prostate brachytherapy have been taken into account in 97% of the cases; 2) The dosimetric quality criterias totally fulfilled the recommendations in a large majority of cases; the intra-operative D90 was found to be superior to 145 Gy in 98% of the patients, and the intra-operative V100 was superior to 95% in 96% of the cases; 3) The early toxicity (mainly urinary) was found to be at the lower range of what is reported in the literature, with in particular a retention rate of 2.4%.

Conformal radiotherapy (CRT) planning for lung cancer: analysis of intrathoracic organ motion during extreme phases of breathing.

PURPOSE: Conformal radiotherapy beams are defined on the basis of static computed tomography acquisitions by taking into account setup errors and organ/tumor motion during breathing. In the absence of precise data, the size of the margins is estimated arbitrarily. The objective of this study was to evaluate the amplitude of maximum intrathoracic organ motion during breathing. METHODS AND MATERIALS: Twenty patients treated for non-small-cell lung cancer were included in the study: 10 patients at the Institut Curie with a personalized alpha cradle immobilization and 10 patients at Tenon Hospital with just the Posirest device below their arms. Three computed tomography acquisitions were performed in the treatment position: the first during free breathing and the other two during deep breath-hold inspiration and expiration. For each acquisition, the displacements of the various intrathoracic structures were measured in three dimensions. RESULTS: Patients from the two centers were comparable in terms of age, weight, height, tumor site, and stage. In the overall population, the greatest displacements were observed for the diaphragm, and the smallest displacements were observed for the lung apices and carina. The relative amplitude of motion was comparable between the two centers. The use of a personalized immobilization device reduced lateral thoracic movements (p < 0.02) and lung apex movements (p < 0.02). CONCLUSION: Intrathoracic organ movements during extreme phases of breathing are considerable. Quantification of organ motion is necessary for definition of the safety margins. A personalized immobilization device appears to effectively reduce apical and lateral displacement.

Adenocarcinoma of the endometrium treated with combined irradiation and surgery: study of 437 patients.

PURPOSE: To identify prognostic factors and treatment toxicity in a series of operable endometrial adenocarcinomas. METHODS AND MATERIALS: Between November 1971 and October 1992, 437 patients (pts) with endometrial carcinoma, staged according to the 1988 FIGO staging system (225 Stage IB, 107 Stage IC, 4 Stage IIA, 35 Stage IIB, 30 Stage IIIA, 6 Stage IIIB, and 30 Stage IIIC), underwent total abdominal hysterectomy and bilateral salpingo-oophorectomy without (n = 140) or with (n = 297) pelvic lymph node dissection. The chronology of adjuvant RT was not randomized and depended on the usual practices of the surgical teams. Seventy-nine pts (Group I) received preoperative low-dose-rate uterovaginal brachytherapy (mean dose [MD]: 57 Gy). Three hundred fifty-eight pts (Group II) received postoperative RT. One hundred ninety-six pts received low-dose-rate vaginal brachytherapy alone (MD: 50 Gy). One hundred fifty-eight pts had external beam pelvic RT (MD: 46 Gy) followed by low-dose-rate vaginal brachytherapy (MD: 17 Gy). Four pts had external beam pelvic RT alone (MD: 47 Gy). The mean follow-up from the beginning of treatment was 128 months. RESULTS: The 10-year disease-free survival rate was 86%. From 57 recurrences, only 12 were isolated locoregional recurrences. The independent factors decreasing the probability of disease-free survival were as follows: histologic type (clear-cell carcinoma, p = 0.038), largest histologic tumor diameter >3 cm (p = 0.015), histologic grade (p = 0.008), myometrial invasion > 1/2 (p = 0.005), and 1988 FIGO staging system (p = 9.10(-8)). In Group II, the addition of external beam pelvic RT did not seem to independently improve vaginal or pelvic control. The postoperative complication rate was 7%. The independent factors increasing the risk of postoperative complications were stage FIGO (p = 0.02) and pelvic lymph node dissection (p = 0.011). The 10-year rate for Grade 3 and 4 late radiation complications according to the LENT-SOMA scoring system was 3.1%. External beam pelvic RT independently increased the rate for Grade 3 and 4 late complication (RR: 5.6, p = 0.0096). CONCLUSION: Postoperative external beam pelvic RT increases the risk of late radiation complications. After surgical and histopathologic staging with pelvic lymph node dissection, in subgroup of "intermediate-risk" patients (Stage IA Grade 3, IB-C and II), postoperative vaginal brachytherapy alone is probably sufficient to obtain a good therapeutic index. Results for patients with Stage III tumor are not satisfactory.

CT and (18)F-deoxyglucose (FDG) image fusion for optimization of conformal radiotherapy of lung cancers.

PURPOSE: To validate a computed tomography (CT) and (18)F-deoxyglucose (FDG) image fusion procedure and to evaluate its usefulness to facilitate target definition and treatment planning in three-dimensional conformal radiation therapy (3D-CRT) for non-small-cell lung cancer. METHODS AND MATERIALS: Twelve patients were assessed by CT and FDG-coincidence mode dual-head gamma camera (CDET) before radiotherapy. The patients were placed in a similar position during CT and FDG-CDET. Matching was achieved by minimizing the cost function by 3D translation and rotation between four landmarks drawn on the patient's skin. Virtual simulation was performed from image fusion and estimated dose-volume histograms (DVH) were calculated. RESULTS: Quantitative analysis indicated that the matching error was < 5 mm. Fusion of anatomic and metabolic data corrected staging of lymph nodes (N) for 4 patients and staging of metastases for 1 patient. In these 5 patients, DVH revealed that the lung volume irradiated at 20 Gy (Vl(20)) was decreased by an average of 22.8%, and tumor volume irradiated at the 95% isodose (V(95)) was increased by 22% and 8% for 2 patients, respectively, and was decreased by an average of 59% for 3 patients after fusion. No difference in terms of Vl(20) and V(95) was observed for the other 7 patients. CONCLUSION: We have validated CT and FDG-CDET lung image fusion to facilitate determination of lung cancer volumes, which improved the accuracy of 3D-CRT.

Evaluation of microscopic tumor extension in non-small-cell lung cancer for three-dimensional conformal radiotherapy planning.

PURPOSE: One of the most difficult steps of the three-dimensional conformal radiotherapy (3DCRT) is to define the clinical target volume (CTV) according to the degree of local microscopic extension (ME). In this study, we tried to quantify this ME in non-small-cell lung cancer (NSCLC). MATERIAL AND METHODS: Seventy NSCLC surgical resection specimens for which the border between tumor and adjacent lung parenchyma were examined on routine sections. This border was identified with the naked eye, outlined with a marker pen, and the value of the local ME outside of this border was measured with an eyepiece micrometer. The pattern of histologic spread was also determined. RESULTS: A total of 354 slides were examined, corresponding to 176 slides for adenocarcinoma (ADC) and 178 slides for squamous cell carcinoma (SCC). The mean value of ME was 2.69 mm for ADC and 1.48 mm for SCC (p = 0.01). The usual 5-mm margin covers 80% of the ME for ADC and 91% for SCC. To take into account 95% of the ME, a margin of 8 mm and 6 mm must be chosen for ADC and SCC, respectively. Aerogenous dissemination was the most frequent pattern observed for all groups, followed by lymphatic invasion for ADC and interstitial extension for SCC. CONCLUSION: The ME was different between ADC and SCC. The usual CTV margin of 5 mm appears inadequate to cover the ME for either group, and it must be increased to 8 mm and 6 mm for ADC and SCC, respectively, to cover 95% of the ME. This approach is obviously integrated into the overall 3DCRT procedure and with other margins.

[Use of radiotherapy in cardiovascular disease. Radiophysical basis, current results, indications and perspectives]. / La curiethérapie en pathologie cardiovasculaire. Bases radiophysiques, résultats actuels, indications et perspectives.

Restenosis is the main limitation of percutaneous angioplasty, especially in vessels of small diameters such as the coronary arteries, the femoro-popliteal and tibial-peroneal arteries and the arterio-venous dialysis grafts. The extensive use of tents has not entirely prevented its occurrence, whereas treating in-stent restenosis gives even more uncertain results. Endovascular radiotherapy has emerged over the past few years as a promising approach to both prevent and cure it. The analogy between the tumour-like cellular proliferations observed in post-angioplasty restenosis and tumour processes prompted pioneering works to study the effect of ionizing radiations in animal models of arterial restenosis. The demonstrated feasibility, tolerance and efficacy of this approach lead to test this strategy in humans. The results of 3 recently presented randomized double-blind trials in the treatment of coronary in-stent restenosis have been so promising that endovascular brachytherapy might now be considered the treatment of choice in this indication. Other randomized trials are currently carried out to test whether endovascular brachytherapy may prevent restenosis in coronary and femoro-popliteal arteries as well as in hemodialysis shunts. In the present review, we describe the basics of the biological effects of ionizing radiations, the technical modalities to deliver endovascular radiations, our current knowledge about their effects on the vascular wall and the restenosis mechanisms, and the results of the first clinical studies. Finally, we address the remaining problems in the use of endovascular curietherapy and question the promises and challenges of its clinical application.

Ionizing radiation-induced death in bovine lens epithelial cells: mechanisms and influence of irradiation dose rate.

We recently reported, in a series of patients receiving total body irradiation before transplant, an influence of dose rate (DR) on cataract formation. The aim of our present in vitro study was to investigate the influence of DR and the mechanisms of lens cell death in a bovine model. After a single fraction of 10 Gy, delivered using low (0.05 Gy/min) or high (2 Gy/min) DR (LDR and HDR, respectively), cells were incubated in media supplemented with two different fetal calf serum (FCS) concentrations (1% and 10%). Cell proliferation was evaluated using Hoechst 33342 (HO) probe and cell viability, with neutral red probe. These fluorimetric assays used a cold light cytofluorimeter. After HO assay, stained cells were examined with fluorescence microscopy to evaluate the nuclear changes related to apoptosis. Global comparison of the mean HO fluorescent values observed with LDR/controls (c) vs. HDR/c revealed a significant difference only after 96 hr (P = 0.036). In 1% FCS conditions, the difference between HDR/c and LDR/c was also statistically significant at 96 hr (P = 0.04). Pairwise multiple comparison using values observed in 1% FCS conditions after 96 hr incubation showed significant difference between HDR vs. c (P = 0.001) and HDR vs. LDR (P = 0.007). This difference, in terms of fluorescence, was correlated to the proportion of cells with nuclear apoptotic morphology. In contrast, cell viability was not influenced by DR whatever the FCS concentration used, from 24 to 96 hr after irradiation. We conclude that our fluorimetric methodology is adapted to evaluate intracellular DNA modifications and cell viability after x-ray irradiation. We observed that a single fraction of 10 Gy induces in vitro lens epithelial cell apoptosis, which is influenced by DR. In humans, HDR is considered more cataractogenic than LDR. Thus, we speculate that lens cell apoptosis could be one of the major mechanisms of radiation-induced cataract. Further investigations are necessary to study the other possible mechanisms of cataractogenesis. Int. J. Cancer (Radiat. Oncol. Invest.) 90, 138-144 (2000).

Complications of postoperative radiation therapy after partial laryngectomy in supraglottic cancer: A long-term evaluation.

This retrospective study, based on a series of 90 patients with invasive squamous cell carcinoma of the supraglottis, was designed to document the functional outcome and complications after postoperative radiation therapy following partial laryngeal surgery. The surgical procedure was a standard supraglottic laryngectomy in 62 patients and a supracricoid partial laryngectomy in 28 patients. All of the patients had an unremarkable postoperative course and achieved locoregional control. The average dose delivered to the remaining larynx was 51.2 Gy (range 25-71 Gy). The average dose delivered to the neck was 50.6 Gy (range 22-70 Gy). The patients were treated at 180-cGy per fractions in a continuous course technique with a cobalt 60 beam. In 5 patients (5.5%) complications led to cessation of postoperative radiation therapy, and the total dose delivered to the remaining larynx and neck was less than 40 Gy. All patients were followed up for a minimum of 10 years or until death. The 5-, 10-, and 15-year actuarial survival estimates were 71. 5%, 44.3%, and 36.3%, respectively. The 5-, 10-, and 15-year actuarial severe complication estimates were all 11.2%. Overall, severe complications occurred in 15 patients. Severe complications led to death in 3 patients (3.3%), permanent gastrostomy in 3 (3.3%), and permanent tracheostomy in 1 (1.1%). A severe complication never resulted in completion of total laryngectomy. In univariate analysis, the mean dose delivered to the larynx was the only variable statistically related to the incidence of a severe complication. The mean dose delivered to the larynx was statistically higher (P = 0.014) in patients who had severe complications (60 Gy) than in patients who did not (50 Gy).