Eur Arch Otorhinolaryngol DOI 10.1007/s00405-014-3013-7
Rhinology
Radionecrosis of the frontal lobe as a consequence of malignant ethmoid tumor management: incidence, diagnosis, risk factors, prevention and management N. Oker · P. Lang · D. Bresson · B. George · J‑P Guichard · M. Wassef · E. Sauvaget · S. Froelich · R. Kania · P. Herman
Received: 24 November 2013 / Accepted: 16 March 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Malignant ethmoid tumors are treated by surgery followed by radiotherapy. This study aimed to evaluate the incidence, risk factors and outcome of radionecrosis of frontal lobe and determine preventive measures. Retrospective study of ethmoid malignancies treated from 2000 to 2011. All patients underwent surgery with/without anterior skull base resection using endoscopic or external approaches followed by irradiation (mean dose 64 Gy). Median follow-up was 50 months. Eight of 50 patients (16 %) presented with fronto-basal radionecrosis, connected to duraplasty, with a latent interval of 18.5 months. Although asymptomatic in six, radionecrosis triggered seizures and required surgery in two cases. Survival was not impacted. Risk factors included dyslipidemia, occurrence of epilepsy and dural resection. Radionecrosis may result
from the combination of anterior skull base resection and radiotherapy for the treatment of ethmoid malignancies. Preventive measures rely on improving the duraplasty and optimization of the Gy-dose delivery.
N. Oker (*) · E. Sauvaget · R. Kania · P. Herman Head and Neck Surgery Department (Service d’Oto‑Rhino‑Laryngologie et de Chirurgie de la Face et du Cou), Hôpital Lariboisière, AP-HP, 2 Rue Ambroise Paré, 75010 Paris, France e-mail:
[email protected] D. Bresson · B. George · S. Froelich Neurosurgery Department, Hôpital Lariboisière, AP-HP, Paris, France e-mail:
[email protected] E. Sauvaget e-mail:
[email protected] R. Kania e-mail:
[email protected] P. Herman e-mail:
[email protected] N. Oker University Paris VII Diderot, Paris, France P. Lang Radiotherapy Department, Hôpital Pitié-Salpêtrière, AP-HP, Paris, France e-mail:
[email protected] Keywords Malignant ethmoid tumor · Cerebral radionecrosis of frontal lobe · Endoscopic anterior skull base surgery · Adenocarcinoma · Radiotherapy · Dural plasty
Introduction In the majority of cases, the treatment of malignant ethmoid tumors is based on surgery, extended to the skull
B. George e-mail:
[email protected] S. Froelich e-mail:
[email protected] J.-P. Guichard Neuroradiology Department, Hôpital Lariboisière, AP-HP, Paris, France e-mail: j‑
[email protected] M. Wassef Anatomopathology Department, Hôpital Lariboisière, AP-HP, Paris, France e-mail:
[email protected] 13
base, followed by radiotherapy at doses of 60–70 Gy and eventually chemotherapy. The proximity of ethmoid tumors to the skull base and the cerebral tissue places the patient at risk of cerebral radionecrosis (frontal lobe) or osteoradionecrosis of the anterior skull base. Several studies have described cerebral radionecrosis, mainly in cases of nasopharyngeal carcinoma [1–4] and brain tumors [5, 6] that were treated exclusively by irradiation. In the cases of nasopharyngeal carcinoma, the cerebral radionecrosis was located in the temporal lobe. Moreover, radionecrosis is underestimated because of the aspecific symptoms and the lack of systematic radiologic follow-up. Few studies have been performed on anterior skull base radionecrosis or cerebral radionecrosis of the frontal lobe in cases of ethmoid tumors treated with the combination of surgery and subsequent radiotherapy; only one specific series has been reported [7]. The purpose of this study was to investigate the incidence, latent intervals and risk factors of radionecrosis affecting the frontal lobe, the anterior skull base or the dural plasty of the resected anterior skull base in patients with ethmoid tumors treated by surgery and radiotherapy. The second aim was to describe our experience with the diagnosis, localization and management of these types of radionecrosis. Finally, we propose preventive measures for the therapeutic approach in cases of ethmoid malignancy.
Materials and methods Population The records of patients who presented in our department from 2000 to 2011 were reviewed. The inclusion criteria were as follows: patients who had been diagnosed with a malignant ethmoid tumor that was treated, according to the gold standard, by surgery and external radiation therapy [8]. All of the patients were followed until death or at least 500 days after the end of treatment. For every patient, the professional risk factors, medical history and cardiovascular risk factors, such as diabetes, hypertension, dyslipidemia, excessive weight, tobacco use and anticoagulant treatment, were noted. A history of coronary artery disease, the need for vascular or cardiac bypass, stroke, diabetes with complications or lower-extremity atherosclerotic arterial disease were considered a severe cardiovascular medical history. This study received the approval of the local Ethic Committee and consents for medical research were systematically obtained.
13
Eur Arch Otorhinolaryngol
Diagnosis and medical support The diagnosis of tumor stage was based on image analysis according to the TNM classification for ethmoid sinus tumors (UICC 2002) [9]. A biopsy for histopathological examination was performed under local anesthesia during medical consultation. All of the patients received tumor resection surgery at the same center. The following details concerning the type of surgery were recorded: primary surgery or salvage surgery; with or without resection of the anterior skull base; and the use of the endoscopic, transfacial or transbasal approach. The duraplasty consisted mainly of two layers of the fascia lata (one intradural and one extradural) for endoscopic resection, or pericranium for craniofacial resection. The subsequent treatment consisted of external irradiation therapy. We reviewed the delay between surgery and the beginning of irradiation, irradiation mode [2D, 3D conformational radiotherapy or intensity-modulated radiation therapy (IMRT)], total delivered dose, fraction size, irradiation duration, presence of chemotherapy and order of the therapeutic sequence (surgery followed by radiotherapy). This sequence of treatment was followed for 80 % of patients, but in the other cases, the patients were directed to our center for a second surgery (after a previous incomplete surgery) or salvage surgery (after previous surgery and radiotherapy). All of the patients were followed for at least 500 days after the end of the irradiation treatment, based on endoscopic examination and repeated MRI (every 4 months in the first year and every 6 months in the following years). The follow-up analysis recorded recurrences, death and clinical or radiological complications. Patients presenting a radionecrosis of the frontal lobe, eventually combined with necrosis of the duraplasty of the anterior skull base, were included. Osteoradionecrosis of the anterior skull base was not concerned inasmuch the anterior skull base had been resected at the time of surgery. Local necrosis of the frontal bone flap used in the transfacial or subfrontal approach was excluded. The diagnosis of radionecrosis relied on a combination of clinical, radiological and dosimetric criteria, as well as on the slow progression of MRI features (as opposed to recurrence) and exceptionally on the histopathology of necrosis. In our study, radionecrosis was diagnosed by MRI and analyzed by two independent radiologists, as MRI rather than CT is recommended for diagnosing radionecrosis [10–12]. In addition, the lesion had to be noted on 2 consecutive MRIs taken at an interval of 3 months or confirmed by pathology.
Eur Arch Otorhinolaryngol
Symptomatic patients exhibited neurological symptoms and were treated with corticosteroids, antiepileptic medication or surgery. Otherwise, patients were scored as having asymptomatic necrosis. Statistical methods All of the data were collected retrospectively. The statistical methods applied for qualitative parameters included Fisher’s exact test and the Kruskal–Wallis test with significance set at p