Unfortunately, it is a well-recognized aphorism that to have a busy surgical otology practice, one needs only 10 children with cholesteatoma as patients. This highlights the high rates of recidivism (a term used to describe both recurrent and residual cholesteatoma) of this condition after surgical treatment. A meta-analysis by Yasser Shewel, MD, and colleagues in 2020 showed recidivism rates ranged from 5.26% to 80% (average of 27.6%) in individuals with canal wall up (CWU) mastoidectomies and from 0% to 48% (average of 17.4%) when the canal wall was down (CWD) (Egyptian J Otolaryngol. 2020. doi.org/10.1186/s43163-020-00043-z). Even with endoscopes, the reported rates of recidivism are still high, ranging from 18%-24% (Otolaryngol Head Neck Surg. 2017. doi: 10.1177/0194599817729136; Otolaryngol Head Neck Surg. 2016. doi: 10.1177/0194599815622441).
Explore This Issue
December 2024The goals of cholesteatoma surgery in all ages are:
- Minimize the number of operations to create a safe, dry, hearing ear;
- Minimize anesthetic exposure for surgery and imaging;
- Minimize radiation exposure;
- Avoid delay in necessary surgery;
- Avoid unnecessary surgery;
- Avoid future complications; and
- Prevent the child (or adult) from developing a fear of medical visits.
With this in mind, the matter up for discussion here is HOW do we monitor patients’ ears most effectively—via imaging with computed tomography/magnetic resonance imaging (CT/MRI) or via second-look surgery? This remains a Great Debate.
Imaging
There are different imaging options for cholesteatoma surveillance. Non-contrast CT scan of the temporal bones is often used to delineate disease and bony involvement, as well as to look for the status of ossicles, tegmen, inner ear, and facial nerve canal. Modern techniques have dramatically reduced the risk of radiation exposure to a minimal level; however, the use of CT, even temporal subtraction CT, in monitoring for cholesteatoma is fraught with difficulties in differentiating recurrent/residual lesions versus inflammatory tissue or granulation (Jpn J Radiol. 2022. doi: 10.1007/s11604-021-01209-2).
MRI scans of the temporal bones are increasingly used in cholesteatoma surveillance as the techniques advance, with increasing accuracy and the ability to avoid radiation exposure. Various techniques of MR imaging can be used for this purpose:
- Echoplanar (EP) Diffusion-Weighted Imaging (DWI)
– Rapid acquisition—”single shot”
– Prone to artifacts/distortion
– Poor detection if the disease is less than 5 mm
- RESOLVE Echoplanar DWI (RS-DWI)
– Better resolution
– 2-3 mm slice thickness
– Added coronal images
– Cholesteatoma detection: sensitivity is 88%; specificity is 96%
- Non-Echoplanar DWI
– 20-30 minutes in the scanner
– Multiple excitation pulses, longer echo time, higher signal-to-noise ratio (SNR)
– Fewer distortions/artifacts
– Better special resolution
– Thinner slices
– In terms of magnet strength, 3T is not measurably better than 1.5T
– Specificity is 94.6%; positive predictive value (PPV) is 96.5%; negative predictive value (NPV) is 80.5%
- Fusion Imaging—CT plus DWI
– Helps delineate cholesteatoma from other chronic otitis media
– Risk of false negative findings, particularly with small recurrences of about 3 mm
– Risk of false positive findings that can arise from inflammation.
There are benefits to using imaging to monitor for surveillance and avoid second-look surgeries. In studies of patients who were considered to have complete removal of disease, the rate of disease is low, ranging from 3% to 4.4% (Otolaryngol Head Neck Surg. 2011. doi.org/10.1177/0194599810396624; Laryngoscope. 2014. doi.org/10.1002/lary.24633).
The sensitivity and specificity of imaging are reasonable considering that pooled sensitivity in a meta-analysis was 72% to 90% for EPI to non-EPI, respectively. A single large study of RS-DWI (EPI) versus non-EPI showed PPV at 93% versus 92.5% and NPV at 70% versus 80%. This suggests that in patients considered to have complete extirpation of disease, with a low expected rate of recurrence, imaging can be a reasonable option instead of second-look surgeries that add risk and morbidity.
In addition, ossicular chain reconstruction (OCR) historically deferred to second-look procedures, which were routinely done for most cases. A pioneer in ear surgery, Simon Parisier, MD, of New York, changed the paradigm by performing primary OCR and ruling against second-look surgery in a significant percentage of cases. Primary OCR has been suggested to have improved hearing outcomes in previous studies (Eur Arch Otorhinolaryngol. 2023. doi: 10.1007/s00405-023-07835-y; Lin Chuang Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2014. PMID: 25129971).
It is now very common to offer primary OCR unless there is too much infection and inflammation at the first surgery. In the case of primary cholesteatoma with ossicular chain erosion, in which the surgeon feels confident of complete extirpation and performs primary OCR, it is reasonable to employ imaging as the second look. (Endoscopes can help with this, but that is a debate for another day.) When this issue was discussed at the European Society of Pediatric Otorhinolaryngology in 2023, imaging won handily over second-look surgery. Overall, in certain cases, imaging seems to be a reasonable option to avoid the risks and comorbidity that come with second-look surgeries.
Second-Look Surgery
Second-look surgeries are the gold standard in the management of cholesteatoma and are used to evaluate for recurrence/recidivism and perform ossicular chain reconstruction if needed. When the surgery is staged, the second look is ideally performed at four to 12 months, depending on the extent and location of the disease and clinical suspicion. It is well accepted that childhood cholesteatoma is more aggressive, and rates of cholesteatoma detected in second-look surgeries range from 23% to 53% (Otolaryngol Head Neck Surg. 2017. doi: 10.1177/0194599817729136; JAMA Otolaryngol Head Neck Surg. 2022. doi:10.1001/jamaoto.2021.3785; Otolaryngol Head Neck Surg. 2011. doi.org/10.1177/0194599810396624).
Concern about the need for anesthesia to perform imaging studies is particularly notable in young children. In addition, false-positive imaging may result in unnecessary surgical intervention in which cholesteatoma is not identified. On the other hand, a false negative study, when the imaging erroneously suggests cholesteatoma is not present, can delay necessary surgical intervention. The percentages listed above for PPV and NPV come from large institutions with highly experienced neuroradiologists. Sensitivity and specificity can vary significantly depending on the institution and expertise. Based on a systematic review of the utility of DW MRI in the detection of cholesteatoma, in order for recurrence to be detected on imaging, the size of the lesion must be 3 mm at minimum (Int J Pediatr Otorhinolaryngol. 2019. doi: 10.1016/j.ijporl.2018.12.031).
For this size cholesteatoma to be seen, it is suggested that imaging be delayed for 18 months after surgery (Otol Neurotol. 2020. doi:10.1097/MAO.0000000000002521). Knowing the aggressive nature of cholesteatoma and high rates of recidivism in children, however, an 18-month delay places the child at significant risk of disease that may surpass the extent of the original disease if there is a false-negative MRI scan.
Imaging is also presumed to be a more cost-effective alternative to second-look surgeries; however, in a study that compared these costs, planned second-look surgery ($11,537, 17.30 QALY) and imaging surveillance with non-EP DW MRI ($10,439, 17.26 QALY) had similar costs (Ann Otol Rhinol Laryngol. 2024. doi: 10.1177/00034894241250253; J Otolaryngol Head Neck Surg. 2019. doi: 10.1186/s40463-019-0384-1).
These findings were congruent with findings in a study by Choi and colleagues, which also showed similar costs with (EPI) DW MRI. It’s also important to note the locations within the tympanic cavity in which recurrence is poorly detected, such as mesotympanic and retrotympanic areas, as shown in one study. (JAMA Otolaryngol Head Neck Surg. 2022. doi: 10.1001/jamaoto.2021.3785).
Additionally, MRIs, with their attendant need for repeat sedation in young children, must be repeated for up to five years to complete surveillance for recurrence (Otol Neurotol. 2019. doi: 10.1097/MAO.0000000000002188). These drawbacks would suggest that second-look surgeries are more reasonable than MRI when there is high clinical suspicion for recidivism based on peri-operative findings.
Conclusion
As in many debates, this one has no clear winner. The decision to monitor patients with imaging versus proceeding with second-look surgeries is nuanced and depends on numerous factors, including the extent of the disease, the physician’s confidence in total extirpation of the disease, its location, the surgeon’s experience, the patient’s age, and other factors. DW MRI can be appropriately used for surveillance when the disease is limited, there is confidence in complete removal, the cholesteatoma was removed without interruption of matrix, and as a way to monitor ears when the exam is limited. Second-look surgeries continue to be the main tool in management when the disease is extensive, complete extirpation cannot be achieved or cannot be guaranteed, or the extent of disease, inflammation, and infection prevented primary ossicular chain reconstruction. Ultimately, this lends itself to shared decision making between the patient/caregiver and
the physician.
Dr. Chandrasekhar is a leading otologist and neurotologist at ENT and Allergy Associates, LLP, with offices in Manhattan, NYC, and Parsippany, N.J. She is a clinical professor of otolaryngology at the Zucker School of Medicine, Hofstra-Northwell in Hempstead, N.Y. Dr. Anne is a professor at Cleveland Clinic Head and Neck Institute and serves as the medical director of pediatric ear and hearing disorders. She has authored more than 100 papers, has published a book on pediatric sensorineural hearing loss, and has presented internationally numerous times on this topic.