Abstract
Purpose
To assess myringotomy plus tympanostomy tube (MTT) complication rates in head and neck cancer (HNC) patients with otitis media (OM).
Materials and methods
We performed a retrospective review of 182 HNC patients with OM treated between January 2000 and October 2007 for demographic data and data about MTT-related complications, tumor type and outcomes following MTT.
Results
We identified OM in 35 nasopharyngeal (NP), 34 paranasal sinus (PNS), and 24 larynx cancer patient ears; of these, 29 (83%), 31 (91%), and 22 (92%), respectively, were treated with MTT. Of the 29 NP cancer patient ears treated with MTT, 13 (45%) received MTT before radiotherapy; complications included otorrhea in 11 ears (38%), otorrhea with perforation in 3 ears (10%), and cholesteatoma in 1 ear (3%). Of the 31 PNS cancer patient ears treated with MTT, 17 (55%) received MTT before radiotherapy; complications included otorrhea in 10 ears (32%) and otorrhea with perforation in 3 ears (10%). All 22 laryngeal cancer patient ears were treated with MTT before radiotherapy; 5 ears (23%) developed chronic otorrhea. Patients with pre-existing eustachian tube dysfunction had significantly higher rates of tympanostomy tube otorrhea (p = .009).
Conclusions
The complication rates of OM in the setting of NP or PNS cancer were not significantly different regardless of intervention timing in relation to radiotherapy. Laryngectomy patients had a high rate of tympanostomy sequelae after radiotherapy. MTT for OM has high complication rates in HNC patients.
1
Introduction
The management of otitis media (OM) in the setting of head and neck cancer (HNC) is a topic of controversy. In cancer patients, OM can result from the malignancy or its treatments and can occur before, during, or after therapy. It can also occur prior to diagnosis; for example, up to 50% of nasopharyngeal (NP) cancer patients initially present with conductive hearing loss due to OM .
Surgery and radiotherapy can affect eustachian tube (ET) function and predict OM . Surgery can alter the musculature of the ET, and radiotherapy can cause considerable ototoxicity, including sensorineural hearing loss, otitis externa, and OM. Both surgery and radiotherapy can cause lasting changes to the physiologic function of the ET, middle ear, tympanic membrane, and ear canal . Ho et al. found that the hearing outcomes of patients treated with MTT before and treated with MTT after radiotherapy were similar .
OM in HNC patients typically does not start as acute suppurative OM, as in the pediatric population, but rather as serous OM. Inflammation of the middle ear mucosa secondary to upper respiratory infection, post-surgical ET dysfunction (ETD), or radiation can cause secretory serous OM, whose symptoms include conductive hearing loss, aural fullness, autophony, slight imbalance, and a heartbeat sensation in affected ears. In HNC patients, serous OM often persists for more than 6 weeks and is then considered chronic . Left untreated, chronic OM can cause spontaneous tympanic membrane perforations or progress to suppurative OM. Chronic perforations or suppurative OM; in some HNC patients with delay can lead to persistent hearing loss, ossicular deformities, complicated infections of the temporal bone and cholesteatoma formation .
Common treatment options for OM include observation, myringotomy alone, and myringotomy with tympanostomy tube placement (MTT). Each otologic intervention carries a high risk of complication, effects of which can be more pronounced and detrimental in HNC patients than in the general population. There is no clear treatment protocol for OM in HNC patients, and few studies have investigated the complication rates of MTT in the setting of HNC. To elucidate the management of OM in HNC patients, we conducted a retrospective study to investigate the types and timing of OM treatments and subsequent outcomes in patients with a spectrum of HNCs, including NP, paranasal sinus (PNS), and laryngeal cancers.
2
Patients and methods
2.1
Patients
The University of Texas MD Anderson Cancer Center’s Institutional Review Board approved the study. We retrospectively reviewed patients with biopsy-proven HNC seen at our institution between January 2000 and October 2007. We queried the Department of Head and Neck Surgery’s billing records database for International Classification of Diseases—9th Revision codes 381.XX and 382.XX (i.e., non suppurative OM, suppurative and unspecified OM) Only patients with OM and a primary HNC were included. The identified patients’ cancers included skin, nasopharynx, larynx, oral cavity, oropharynx, hypopharynx, salivary gland, lung, breast, heart, gastrointestinal, gynecological, genitourinary, renal, brain, and PNS cancers and unknown primary tumors.
We reviewed medical records of identified patients to obtain demographic data and information about each patient’s cancer, cancer status, treatments (past and/or current), and OM interventions and complications thereof. Patients whose cancer or treatment data were unavailable were excluded from the study.
2.2
OM diagnosis and treatment
OM was diagnosed by microscopic otoscopy, audiometry, tympanometry, and a review of computed tomography or magnetic resonance imaging studies obtained for cancer treatment management. Otologic interventions offered to patients included MTT or myringotomy alone while other patients opted for observation. Patients self-selected treatment based on counseling from the physician. Details of each intervention and subsequent risks were described to patients to allow for educated decisions.
OM interventions include myringotomy with our without tube placement and each was performed using topical phenol anesthetic in an outpatient clinic. Myringotomies were performed in the posterior–inferior quadrant by the attending surgeon. All patients who underwent either myringotomy alone or MTT were given 5 drops of ofloxacin in the affected ear(s) twice daily for 3 days following the procedure. Outcomes for each ear were recorded. Pope tympanostomy tubes were used in all patients. Tube otorrhea was defined as persistent drainage from a tube for greater than 2 months. Otorrhea with perforation was defined as persistent drainage without a tube. Resolution of OM was defined as an intact tympanic membrane and an aerated middle ear.
2.3
Statistical analyses
Descriptive statistics for scaled values and frequencies of study patients within the categories for each of the parameters of interest were calculated using commercial statistical software. Between-group differences were assessed using the Student t-test. Correlations between parameters and endpoints were assessed using the Pearson chi-square test or, if fewer than 10 patients were included in any cell of a 2 × 2 cell grid, the 2-tailed Fisher exact test. P values of < .05 were considered statistically significant. Statistical analyses were performed using Statistica version 12 (Stat Soft, Inc., Tulsa, OK) and SPSS version 21 (IBM Corporation, Armonk, NY) statistical software applications.
2
Patients and methods
2.1
Patients
The University of Texas MD Anderson Cancer Center’s Institutional Review Board approved the study. We retrospectively reviewed patients with biopsy-proven HNC seen at our institution between January 2000 and October 2007. We queried the Department of Head and Neck Surgery’s billing records database for International Classification of Diseases—9th Revision codes 381.XX and 382.XX (i.e., non suppurative OM, suppurative and unspecified OM) Only patients with OM and a primary HNC were included. The identified patients’ cancers included skin, nasopharynx, larynx, oral cavity, oropharynx, hypopharynx, salivary gland, lung, breast, heart, gastrointestinal, gynecological, genitourinary, renal, brain, and PNS cancers and unknown primary tumors.
We reviewed medical records of identified patients to obtain demographic data and information about each patient’s cancer, cancer status, treatments (past and/or current), and OM interventions and complications thereof. Patients whose cancer or treatment data were unavailable were excluded from the study.
2.2
OM diagnosis and treatment
OM was diagnosed by microscopic otoscopy, audiometry, tympanometry, and a review of computed tomography or magnetic resonance imaging studies obtained for cancer treatment management. Otologic interventions offered to patients included MTT or myringotomy alone while other patients opted for observation. Patients self-selected treatment based on counseling from the physician. Details of each intervention and subsequent risks were described to patients to allow for educated decisions.
OM interventions include myringotomy with our without tube placement and each was performed using topical phenol anesthetic in an outpatient clinic. Myringotomies were performed in the posterior–inferior quadrant by the attending surgeon. All patients who underwent either myringotomy alone or MTT were given 5 drops of ofloxacin in the affected ear(s) twice daily for 3 days following the procedure. Outcomes for each ear were recorded. Pope tympanostomy tubes were used in all patients. Tube otorrhea was defined as persistent drainage from a tube for greater than 2 months. Otorrhea with perforation was defined as persistent drainage without a tube. Resolution of OM was defined as an intact tympanic membrane and an aerated middle ear.
2.3
Statistical analyses
Descriptive statistics for scaled values and frequencies of study patients within the categories for each of the parameters of interest were calculated using commercial statistical software. Between-group differences were assessed using the Student t-test. Correlations between parameters and endpoints were assessed using the Pearson chi-square test or, if fewer than 10 patients were included in any cell of a 2 × 2 cell grid, the 2-tailed Fisher exact test. P values of < .05 were considered statistically significant. Statistical analyses were performed using Statistica version 12 (Stat Soft, Inc., Tulsa, OK) and SPSS version 21 (IBM Corporation, Armonk, NY) statistical software applications.
3
Results
We identified 182 HNC patients with OM who were seen at our institution from January 2000 to October 2007; their malignancies are given in Table 1 . Only patients with primary HNC were included. The mean patient age was 54 years ± 14.3 years (range, 14–85 years). The mean follow-up time was 2.9 years ± 2.7 years. Otologic interventions offered to patients included MTT or myringotomy alone while other patients opted for observation. Seventy patients had NP, PNS, or larynx cancer, and these patients are the focus of this report. FLOAT NOT FOUND
Among these 70 patients, OM was identified in 93 ears, of which 82 (88%) were treated with MTT, 5 (5%) treated with myringotomy alone, and 6 (7%) observed. Of the 82 ears treated with MTT, 56 (68%) had OM resolution without complication and 26 (32%) developed tube otorrhea. Of the 5 ears treated with myringotomy alone, 3 (60%) had OM resolution, 1 (20%) developed perforation with otorrhea, and 1 (20%) developed cholesteatoma. Of the 6 ears that were observed, 4 (67%) had OM resolution and 2 (33%) developed perforation with otorrhea.
Of the 93 ears in which OM was identified, 80 (86%) were treated with 2 or fewer interventions and 7 (8%) were treated with 3 or more interventions. The complication rate of the ears treated with 2 or fewer interventions (36%) was smaller than that of the ears treated with 3 or more interventions (39%), but this difference was not significant (p = .82). Of the 89 ears of patients not lost to follow up that were treated with an otologic intervention, 26 (29%) had ETD before the HNC diagnosis while more than half of these ears (15 ears; 58%) developed complications after interventions. This complication rate was significantly higher than that of the 18 ears with complications that did not have ETD before HNC diagnosis (29%; p = .009).
3.1
NP cancer
OM was identified in 35 ears in 27 NP cancer patients. Of the 35 ears, 29 (83%) were treated with MTT of which 13 (45%) were treated prior to radiotherapy, 2 (6%) were treated with myringotomy alone, and 4 (11%) were observed. MTT complications included tube otorrhea in 11 ears (38%), otorrhea with perforation in 3 ears (10%), and cholesteatoma in 1 ear (3%); 1 patient (2 ears) was lost to follow up. Of the 2 ears treated with myringotomy alone, 1 had OM resolution and 1 developed cholesteatoma. Of the 4 ears that were observed, 2 had OM resolution and 2 developed a spontaneous perforation with chronic otorrhea. Eleven patients died of their disease during the observation period.
3.2
PNS cancer
OM was identified in 34 ears in 28 PNS cancer patients. Of the 34 ears, 31 (91%) were treated with MTT of which 17 (55%) were treated prior to radiotherapy. 2 (6%) ears were treated with myringotomy alone, and 1 (3%) was observed. Complications of MTT included tube otorrhea in 10 ears (32%) and otorrhea with perforation in 3 ears (10%); 1 patient (2 ears) was lost to follow up. The 1 ear that was observed had complete OM resolution. Of the 2 ears treated with myringotomy alone, 1 had complete OM resolution and the other developed persistent tympanic membrane perforation with otorrhea. Of the 28 PNS cancer patients, 23 underwent maxillectomy and 5 received radiotherapy with or without chemotherapy. Eight patients died of their disease during the observation period.
3.3
Laryngeal cancer
OM was identified in 24 ears in 15 laryngeal cancer patients, all of whom had undergone a total laryngectomy or laryngopharyngectomy prior to the development of OM. Of the 24 ears with OM, 22 (92%) were treated with MTT, 1 (4%) was treated with myringotomy alone, and 1 (4%) was observed. Of the 22 ears treated with MTT, 17 (77%) had OM resolution and 5 (23%) developed tube otorrhea. The 2 ears that were either treated with a myringotomy alone or observed had OM resolution. All patients received radiation after their OM intervention. Eleven patients died of their disease during the observation period.
3.4
Radiotherapy
Of the 70 patients, 66 (94%) received radiotherapy with or without surgery or chemotherapy. The mean overall radiation dose to the primary site was 64.5 Gy. Radiation was delivered to 26 of 27 NP cancer patients (mean dose, 66.3 Gy), 24 of 28 PNS cancer patients (mean dose, 62.4 Gy) and all 15 laryngeal cancer patients (mean dose, 64.4 Gy).
Seventy patients received radiotherapy; however radiotherapy treatment details were not available for 32 patients (46%). Of the remainder, the majority (54%, 38 patients resulting in 51 ears) were treated with intensity modulated radiation therapy (IMRT). Of 51 the ears that received IMRT, 16 (31%) developed tube otorrhea and 5 (10%) developed a tympanic membrane perforation with otorrhea after myringotomy alone or observation.
Of the 89 ears with OM that were not lost to follow up, 57 (64%) were treated with MTT before radiotherapy and 23 (26%) were treated with MTT after radiotherapy. The complication rate of ears treated with MTT before radiotherapy (33%; 19 of 57 ears developed tube otorrhea) was lower than that of ears treated with MTT after radiotherapy (43.5%; 10 of 23 ears developed post-tympanostomy otorrhea), but this difference was not significant (p = .62) ( Table 2 ). FLOAT NOT FOUND
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
