Outcomes of Operative Management of Cholesteatoma in the Elderly: A 15‐Year Retrospective Study

It is well established that the treatment of cholesteatoma is surgical; there is currently no non-surgical treatment available capable of eradicating the disease [1]. The frequency and severity of ear infections, rate of growth of cholesteatoma, proximity to important structures, and complications related to cholesteatoma are some of the disease factors influencing the clinical decision to offer surgery. However, the decision to perform mastoid surgery can be complex in older patients as the peri-operative management of these patients presents unique challenges to the peri-operative team. Ageing is associated with frailty, a multisystem clinical syndrome, characterised by reduced physiological reserve and vulnerability to external stressors. Peri-operative complications such as cardiorespiratory complications, cognitive dysfunction and falls are known to happen in the elderly and should be considered when planning surgery [2]. In addition, discussions are necessary regarding potential progression of the cholesteatoma and related complications, if left untreated, within the lifetime remaining to the patient. Comprehensive and considerate pre-operative discussions provide an opportunity for shared decision making with the patient and surgical team where the benefits, risks and alternatives are discussed with the patient and their friends and family [3]. While surgical challenges can be deliberated in the Otology clinic, review in a dedicated pre-operative clinic with an anaesthetist allows discussion on anaesthetic and medical challenges prior to major surgery. The pre-operative clinic also provides an opportunity to optimise pre-existing comorbidities, such as congestive cardiac failure (CCF), chronic obstructive pulmonary disease (COPD) or diabetes mellitus (DM) and to ensure that peri-operative protocols for specific medications are followed and referral to specialist teams undertaken if indicated. Risk prediction tools can be used to support the peri-operative physician's assessment of peri-operative morbidity and mortality, such as Surgical Outcome Risk Tool (SORT), American College of Surgeons National Quality Improvement Programme (ACS-NSQIP) calculator and the Modified Duke Activity Status Index (M-DASI). As the population ages, the number of operations performed in the elderly population is increasing [4]; this is true of cholesteatoma surgery as well. Therefore, clinicians are faced with complex discussions around surgery in the elderly more commonly. Outcomes of otological procedures such as myringoplasty and cochlear implants have been reported in the elderly, and the older patient is not found to be disadvantaged with regard to the outcomes [5, 6]. Cholesteatoma, however, presents a unique challenge as the disease can be quite extensive, requiring longer operating times with a higher risk of intra-and post-operative challenges. Little information is available in the literature to counsel the elderly population of patients on the specific surgical and anaesthetic risks of cholesteatoma surgery. We have therefore examined the outcomes of mastoid surgeries for cholesteatoma in the elderly in our institution to inform our current practice. The aim of this study was to identify the surgical and non-surgical complications associated with surgery for cholesteatoma in an elderly population. This study was approved by the trust audit committee (reference number: 11175). It was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines and has been reported in line with the Strengthening the Reporting of Observation Studies in Epidemiology (STROBE) guidelines [7]. This study was performed in a tertiary Ear Nose and Throat Unit at a university hospital in the United Kingdom. All patients aged 70 or over with cholesteatoma and who received general anaesthesia for mastoid surgery between 2005 and 2020 under the care of the senior authors (A.B., P.R., P.B.) were included. A retrospective notes review was undertaken. Data were extracted from electronic and paper records. In our centre, patients had a pre-operative anaesthetic assessment by specialist nurses prior to surgery, with an anaesthetic consultant review where needed. Pre-existing co-morbidities are optimised and further investigations, such as echocardiography, or referral to specialist teams, are performed if required by the anaesthetic team. Robust institutional guidelines that are in place as advised by the Department of Haematology for the cessation or continuation of specific medications to ensure minimal interruption were followed. This includes guidelines for patients taking vitamin K antagonists (warfarin), direct oral anticoagulants (DOACs) and antiplatelet agents; heparin bridging in patients with a high thrombotic risk and contacting haematologists when needed for complex decision making. Primary outcomes were peri-operative complications. Secondary outcomes were surgical complications and cholesteatoma recurrence. Thirty-seven patients were included, on whom 41 operations were performed (see Table 1). Of the 41 operations, 38 were completed and 3 were abandoned due to medical complications. Of the 38 completed procedures, 34 were primary operations and 4 were revisions. Three of the 37 patients fit the inclusion criteria at both primary and revision surgery, while two patients fit the inclusion criteria at their revision surgery only. The mean age of patients was 75.3 years (range 70–87 years). 73% of patients had at least 1 co-morbidity (see Table 2). The most common co-morbidities were hypertension (56.8%), ischaemic heart disease (IHD) (27%) and DM (18.9%). The mean duration of surgery was 3.08 h (range 1–5 h). The mean inpatient stay was 1.34 days (range 0–11 days). The type of mastoid surgery in all completed mastoidectomy operations is presented in Table 3. Anatomical factors encountered intra-operatively are presented in Table 4. Thirteen patients had peri-operative complications; 2 patients had 2 complications each (see Table 5). Peri-operative complications were encountered in 15 of the 41 procedures performed (36.6%). Follow-up periods and long-term outcomes for each patient are presented in Table 1. Non-surgical intra-operative complications were encountered in five operations (12.2%). Three operations were abandoned due to cardiovascular instability, where there was difficulty controlling blood pressure on induction of anaesthesia and the patients became either significantly hyper- or hypotensive. These patients subsequently had cardiology assessment and optimisation prior to decisions regarding further surgery. One of these patients, with a background of myocardial infarction (MI) and intraoperative hypotension at the first surgery, had uncomplicated revision surgery. Two patients were deemed not fit for further surgery (one with aortic regurgitation and another with multiple comorbidities including DM, hypertension and chronic kidney disease (CKD) were managed conservatively). A 4th patient with a history of IHD required several boluses of vasopressors during anaesthesia to maintain blood pressure, but the patient stabilised and their operation was completed. Post-operatively, this patient was monitored in the post-operative acute care unit (PACU). In the 5th patient, who had a background of insulin dependent type I DM, there was difficulty controlling blood glucose levels. This patient was admitted acutely, presenting with acute mastoiditis with subperiosteal abscess and Grade V facial nerve palsy on a background of infected cholesteatoma. A variable rate insulin infusion was required intra- and post-operatively. Intra-operative surgical complications occurred in three operations and included cerebrospinal fluid (CSF) leak with two operations (4.8%) and one internal jugular vein (IJV) bleed at the jugular bulb (2.4%). The CSF leaks occurred in the context of exposed dura and extensive disease. They were repaired using Obex NeuroFilm and a large temporalis fascia graft. The patients were admitted post-operatively and recovered uneventfully. A small IJV bleed occurred during one operation whilst attempting to remove deep-seated cholesteatoma adherent to the jugular bulb. Intra-operative bleeding was arrested with the use of bone wax and a large temporalis fascia graft. The patient recovered uneventfully. Four patients had post-operative medical complications (9.6%) including urinary retention, readmission with vertigo, and readmission with post-operative cognitive dysfunction (POCD). The patient readmitted with POCD was the patient whose blood pressure had to be maintained intra-operatively by repeated boluses of vasopressors. Three patients had post-operative surgical complications (7.2%). One patient developed a post-operative wound infection which was managed with oral antibiotics. One patient presented with a delayed CSF leak occurring 3 years after the primary mastoid surgery, which was repaired surgically. One patient had recurrent ear infections and, 4 years post-operatively, on a clinical basis, was treated for a presumed Pseudomonas spp necrotising otitis externa with long-term antibiotics, despite no radiological evidence. They made a complete recovery. Although these latter two complications occurred 30 days following surgery, we have included them as they occurred during the period of follow-up. Cholesteatoma recurred in four patients (10.8%). Two of these patients had revision surgery within 3 years of the primary surgery, and cholesteatoma was found intraoperatively. One patient presented acutely 13 years after their primary surgery with facial nerve palsy. High resolution computerised tomography (CT) and Diffusion Weighted Magnetic Resonance Imaging (DW-MRI) revealed a lesion within the left temporal bone, involving the petrous apex. The walls of the left internal auditory canal were dehiscent, and the lesion bulged into the canal, encroached on the 7th and 8th cranial nerves and abutted the internal carotid artery. The cochlea, vestibule and parts of the semi-circular canals were invaded by the lesion as well. The patient was deemed not suitable for a revision operation but was closely followed up for regular aural toilet until the coronavirus pandemic, when they were lost to follow up. A fourth patient underwent revision surgery for suspected recurrent cholesteatoma, but only granulation tissue was encountered intraoperatively. There is little data available on the complications and outcomes of the management of cholesteatoma in elderly patients. There is limited information on whether the progression of disease or complications are comparable to a younger otologic population, and it is therefore challenging for clinicians to determine the best management for the patient, taking into consideration associated co-morbidities, potential peri- and post-operative morbidity and mortality, and future recurrence. There has been increased interest and research in anaesthesia in the elderly as the number of operations performed in the elderly increases [4]. Mortality following general anaesthesia (GA) has been shown to exponentially increase with age, and a linear relationship with surgical morbidity and increasing age has been reported [8]. In our patient cohort, several post-operative medical complications were encountered. One patient was readmitted with post-operative cognitive dysfunction (POCD), resulting in the longest hospital admission. POCD is common in the elderly, affecting 15%–53% of patients [9]. GA carries an increased risk of POCD compared to regional anaesthesia [10]. POCD has been associated with poor long-term outcomes, including increased mortality, increased length of inpatient stay, and long-term cognitive impairment. Risk factors for POCD include increasing age, post-operative pain, increased duration of anaesthesia, and hearing impairment [9, 11]. Mastoid surgery for cholesteatoma often requires a long anaesthetic, a mean of 3.08 h in our cohort. In addition, these patients may have a pre-operative conductive hearing loss, often superimposed on a degree of age-related sensorineural hearing loss. This will be exacerbated in the immediate post-operative period by the presence of a head bandage and packing the external auditory canal. Evidence-based strategies to prevent POCD include pre-operative assessment and modification of risk factors where possible, multi-component interventions which include adequate analgesia, early mobilisation, cognitive stimulation via therapeutic activities and multi-disciplinary involvement, avoidance of peri-operative benzodiazepines and Bispectral Index (BIS) targeting intra-operatively [12]. It is very important to optimise hearing, including consideration of bone conduction aids, in the post-operative period before the patient can return to conventional hearing aids. Urinary retention is a common post-operative complication and was the most common in our cohort; this is also reported by other authors [13]. The risk increases with age and is also proportional to the duration of surgery [14]. Diabetes mellitus is a common co-morbidity in the elderly. These patients require close peri-operative blood sugar monitoring. Peri-operative blood sugar control can be interrupted by pre-operative fasting, disruption to normal insulin protocols, and the physiologic stress response to surgery. Poor control of blood sugars may lead to increased post-operative morbidity, including wound infections [15] and poor healing. It is thus important to optimise blood glucose control in these patients pre-operatively. Cardiovascular complications are a significant concern in anaesthesia of the elderly and were encountered in our patient cohort. Increased incidence of cardiovascular disease in the elderly and inherent increased risks due to age-related cardiovascular changes and reduction in physiological reserve [9] predispose them to cardiovascular complications. In addition, GA can reduce sympathetic tone, leading to vasodilatation and reduction in systemic blood pressure, compromising organ perfusion. Increasing age increases the risk of post-operative cardiac complications [16]. Hypertension is common in the elderly population and was the most common co-morbidity in our cohort. Hypertension has been associated with an increase in lability of blood pressure intra-operatively and intra-operative hypotension [17]. Other factors found to increase the risk of cardiac adverse events include a long operative duration [18]. Strategies to reduce the risk of cardiovascular complications include the continuation of beta-blocker therapy and pre-operative control of hypertension [9]. In patients who are unfit for GA, mastoidectomy can be performed under local anaesthesia (LA), with or without sedation [19, 20]. This technique requires appropriate patient selection, pre-operative counselling and adequate LA infiltration [20], it is generally well tolerated. Patients' main concerns are loud noises during surgery and anxiety, with dizziness and backache due to positioning. Pain is not a significant issue intra-operatively [21]. Sedation may be used to increase patient comfort [19], this is not always suitable for patients with co-morbidities such as cardiovascular disease [22]. Surgery under LA; however, may not be suitable for extensive or complicated cholesteatoma. To date, there are no non-surgical treatments available capable of eradicating cholesteatoma. Our patients who were unfit for GA were treated with topical antibiotics and regular aural toilet. Due to evident ethical concerns, trials to compare surgical versus conservative management of diagnosed cholesteatoma would be challenging; and very little information from retrospective analyses on this is available in the literature. Surgical complications were less common than the post-operative medical issues. We encountered three CSF leaks in our patient cohort—two were intraoperative and related to extensive disease and challenging anatomy while the other was delayed in presentation. The incidence of intraoperative CSF leaks (4.8%) in our series is similar to the incidence of CSF leak reported by Walker et al. [23] in a series of 285 mastoid surgeries in a patient cohort with an average age of 35 years. It is important to note, however, that the dura can be more adherent to the skull in the elderly [24] which may potentially increase the chances of dural breach if the dural plate is opened during surgery. On a similar line, loss of collagen can make the veins more fragile in the elderly and has to be considered. One patient in our cohort developed an IJV bleed on attempting to remove cholesteatoma matrix that was adherent to the vein, and the bleed was stopped with pressure, bone wax, and fascia. Bleeding is also an important consideration in the elderly as many are on aspirin and/or clopidogrel for cardiovascular issues. In our institution, we recommend withdrawal of aspirin and clopidogrel for 7 days prior to mastoid surgery to decrease the risk of bleeding and discuss with patients the increased risk of prothrombotic events as part of the consent process. Recurrence rates quoted for cholesteatoma vary widely, from 12.6%–29.5% in canal wall up procedures to 6.6%–16.5% in canal wall down procedure [25]. Four patients in our cohort (10.8%) had recurrence of their disease (three cholesteatoma and one granulation tissue) in keeping with recurrence rates for cholesteatoma in adults. It is accepted that recurrence following canal wall down procedures is lower. In our cohort, 51.2% of patients received a canal wall down procedure. There is a lower threshold in the elderly population towards canal wall down procedures to avoid revision surgery in the future. However, it has been seen that re-epithelisation of the mastoid cavity in the elderly can be delayed due to poor tissue oxygenation resulting in a chronic wound infection [26]. This would be an important factor to consider in operative planning in elderly patients with a history of peripheral vascular disease or wound healing issues. Additionally, long-term follow-up may not be feasible in the elderly. Several patients in our cohort discontinued follow-up in the ENT clinic despite continuing to attend hospital appointments in other specialities—potentially due to challenges with attending multiple appointments and non-ENT appointments taking priority. Migirov [27] reports one of the few studies discussing outcomes following mastoid surgery for cholesteatoma in the elderly. Of their 16 patients with cholesteatoma, nine patients were admitted with complications, such as intracerebral abscess, facial nerve palsy, and lateral semicircular canal fistula, requiring emergency surgery. Twelve patients underwent modified radical mastoidectomy and four patients a canal wall up procedure. Recurrence occurred in 1 patient (6.25%), comparable to our study. Ahn et al. [13] compared mastoid surgery for chronic otitis media with and without cholesteatoma between a younger and older cohort. They found no significant difference between surgical complication rates and outcomes between older and younger patients. In their elderly cohort, complications included reperforation, otitis externa and delayed facial palsy. No information regarding anaesthetic complications was reported. A similar study by Jolink et al. [28] examined post-operative success and adverse events in patients of different age groups undergoing mastoid surgery for CSOM (both with and without cholesteatoma, 26 patients with cholesteatoma > 65 years) and did not find any difference in the post-operative adverse events and recurrence rates in cholesteatoma in the adult age groups. Our study is limited by its retrospective, single centre nature. Despite this, there was a long follow-up period in the majority of patients and it is one of the few studies specifically investigating complications of surgery for cholesteatoma in the elderly; therefore, it significantly adds to the literature. Further research is needed in this population to allow clinicians to counsel elderly patients regarding the risks of surgery. In particular, research considering long-term outcomes of elderly patients whose cholesteatomas were treated without surgery would be informative. Surgical outcomes and recurrences are not significantly different for mastoid surgery for cholesteatoma in elderly patients in this study compared to those reported in the younger adult population in the literature. However, GA for mastoid surgery is associated with significant morbidity and mortality in our series. Therefore, pre-operative anaesthetic assessment and optimisation is important, as is discussion of the risks of anaesthesia and surgery versus conservative treatment. D.D. was responsible for the data collection, data analysis and co-wrote the manuscript. E.G.H. performed data analysis and co-wrote the manuscript. R.M. provided specialist anaesthetic insight for the manuscript. A.B. provided access to patients. P.R. provided access to patients, intellectual support, and edited the manuscript. P.B. designed and directed the study, provided access to patients, performed data review, provided intellectual support and edited the manuscript. This study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. The review history for this is available at