HSC Section 8_April 2017

relative risk (RR range 0.35 to 4.4) varied greatly between studies 4,6,7,15,18 (Table II). This indicates that studies showed different results on which surgical proce- dure would result in the lowest disease recidivism per- centage. Nyrop and Bonding 7 reported the largest absolute risk reduction: 61% (46% to 76%) in favor of the CWD procedure. We identified three articles that made a distinction between residual and recurrent disease 4,15,18 (Table III). Roden et al. 18 found a significant higher risk of residual disease in the CWU group (20.4% vs. 4.7%, P 5 .03). Recurrence risks between both techniques were equal (Table III). Similarly, Declerck found more residual dis- ease after the CWU procedure (17.6% vs. 0%). 4 However, also more recurrent disease was found after the CWU procedure (8.1% vs. 0%) (Table III). 4 Stankovic showed opposite results: less residual (3.1% vs. 7.0%) and recur- rent disease (4.7% vs. 15.1%) occurred in the CWU group (Table III). 15 We used data from the latter three studies to calculate the risk difference and relative risk for cho- lesteatoma residual and recurrent disease (Table III). Two studies showed a risk difference in favor of CWD for residual disease: 16% and 18%, respectively. 4,18 In addi- tion, recurrent disease risk difference showed to be in favor of CWD in the studies of Declerck 4 (8%) and Roden et al. 18 (4.4%) (Table III). Contrarily, Stankovic 15 found a risk difference in favor of CWU for both residual (risk reduction 2 4%; relative risk 0.43) and recurrent disease (risk reduction 2 10%; relative risk 0.31) (Table III). DISCUSSION We reviewed the literature to assess which surgical removal technique for adult acquired cholesteatoma (CWU or CWD) provided the lowest rate of disease recid- ivism. The number of available studies was substantial. The seven included studies were all of moderate or high relavance. However, the validity of included studies ranged from low to moderate. The majority of included studies showed that adult cholesteatoma patients suffer from a higher recidivism risk after the CWU procedure as compared to the CWD proce- dure. Three studies 4,15,18 reported on residual and recurrent disease rates independently. Canal wall up recidivism was more likely to be residual disease, whereas CWD recidivism tended to be recurrent disease. Because CWU is not per- formed in severe cases in common practice, and the feasibil- ity of performing CWU or CWD is related to the severity of the disease, confounding by indication might have masked even higher disease recidivism percentages. 23 This con- founding could explain heterogeneity in our results: Stan- kovic 15 showed significant results in favor of CWU. In this study, severity of the disease was different between study populations; extensive disease, a small mastoid and a dam- aged posterior wall were indications for CWD surgery. Only Nyrop and Bonding’s study 7 was unaffected by confounding by indication. Authors described CWU and CWD groups to be similar regarding the extent of cholesteatoma disease. The latter results showed higher CWU recidivism rates compared to results from the CWD technique ( P < .0001). Because Nyrop and Bonding’s study 7 provided the most

unbiased insight according to our CAT, we recommend that the CWD procedure should be performed in adult patients with acquired cholesteatoma. Tomlin et al. 5 performed a meta-analysis on the risk of cholesteatoma recidivism after CWU and CWD. Authors reported a lower percentage of residual and recurrent disease after CWD (range 5% to 17%) com- pared to CWU surgery (range 9% to 70%). The relative risk of recurrent or residual disease was 2.87 (95% confi- dence interval: 2.45 to 3.37) after CWU compared to CWD. The results of Tomlin et al. 5 are in line with our findings, although the inclusion of Stankovic’s article 15 in the current review introduces new uncertainty about the generalizability of reported surgical outcomes. Tom- lin et al. 5 included 13 articles, of which three were included in our study. 6,7,18 In contrast to Tomlin et al., 5 results from our selected studies could not be pooled in meta-analysis, a difference that marks a more heteroge- neous and different retrieval of literature. Similar to their included studies, 5 none of our included studies mentioned the use of magnetic resonance imaging (MRI) in detecting cholesteatoma recidivism in CWU patients. This could be explained by recent MRI use in CWU follow-up, whereas the majority of included studies was performed before 2006 and consisted of retrospective case series of earlier performed surgeries. Compared to Tomlin’s study, we provide additional insight because we excluded children, and in addition calculated relative risks and absolute risk reductions (both significant in five out of seven selected studies 4,6,7,15,18 ). Risk of Bias In interpreting the findings, the following consider- ations need to be taken into account. Firstly, included studies differed in their methods of follow-up. Studies with a shorter duration of follow-up might have been subjected to bias in favor of CWU. 4,7,15,18 Secondly, in the detection of residual or recurrent cholesteatoma, second-look surgery is a more sensitive method compared to clinical examination in CWU patients. 24 Therefore, results of the study that applied second-look surgery 4 might be influenced by the higher sensitivity of confirmation of disease recidivism. How- ever, diffusion-weighted MRI is expected to replace second-look surgery completely to detect cholesteatoma recidivism in the near future. 5 Diffusion-weighted imag- ing has shown to have high sensitivity and specificity for detecting cholesteatoma, especially nonecho planar diffusion-weighted MRI. 25 Lastly, surgical experience needs to be taken into consideration, because differences in skills can lead to different outcomes. However, none of our included studies mentioned the experience of the surgeons. Therefore, we cannot exclude the possibility of one surgeon achieving superior results when applying either one of the techniques.

Surgical Considerations Several factors need to be taken into account when opting for a CWU or CWD operation: hearing outcome,

Laryngoscope 126: April 2016

Kerckhoffs et al.: A Review on Cholesteatoma Recidivism After CWU and CWD

110