Introduction

Chronic otitis media is the term used to describe any chronic inflammatory pathology of middle ear. It has been an important cause of middle ear disease since prehistoric times. Despite the valuable contribution of antibiotics, chronic otitis media remains a common disease and its complications challenge both otologist and radiologists [1, 2]

  1. 1.

    Otitis media and resultant hearing loss remain a significant international health problem in terms of prevalence, economics, and sequelae. Short term and long term sequel of otitis media may be devastating. It can be avoided if recognised and treated properly [3, 4].

  2. 2.

    Radiological examination of temporal bone helps us to achieve this objective. It is the computed tomography which has made the most important contribution to radiology in otolaryngology. Though otitis media is essentially a clinical diagnosis but high resolution computed tomography (HRCT) is useful for showing evidence of ossicular erosion in acute and chronic mastoiditis, extent of pneumatisation of temporal bone and relationship of pathology of adjacent critical neurovascular structure such as dura, internal carotid artery, facial nerve and lateral sinus. It is now being claimed that a cholesteatoma as small as 3 mm in size can be diagnosed much earlier by the use of HRCT and also for a cholesteatoma behind an intact tympanic membrane imaging is important [5].

The main benefit of HRCT in cases where the cholestetoma with scarring, granulation tissue, or post-surgical changes in which the resulting soft tissue masses were indistinguishable. HRCT has a significant impact on the medical and surgical management of individuals with middle ear disease. It confirms and expands upon otoscopic findings, resolves clinical doubts and in many circumstances, play a significant role in determining surgical efficacy. However routine HRCT scanning prior to all surgeries of cholesteatoma can only be justified if it can be shown to influence clinical management [6].

HRCT has not gained wide acceptance as an essential aid to plan a surgery, main objective of current study was to evaluate sensitivity and specificity of HRCT temporal bone in identifying ossicular necrosis preoperatively as compared with intra operative findings in chronic otitis media cases.

Methods and Materials

Present analytical observational study was carried out from January 2015 to December 2017 on consecutive 60 patients of chronic otitis media from inpatient department of ENT at Sri Guru Ram Das Medical College, Sri Amritsar. All patients were admitted in ENT department. Necessary permission and approval from ethics committee and authority, prior to starting the study was taken. Informed consent and assent was taken from patients according to protocol of ethics committee of our institution. Patient confidentiality was maintained at all the times during course of study.

Following were inclusion and exclusion criteria

Inclusion criteria:

  1. 1.

    Age group: 8–70 years

  2. 2.

    Chronic otitis media with hearing loss without complication

Exclusion criteria:

  1. 1.

    Chronic otitis media with complication

After selection of patients detailed history with regard to otorrhoea, deafness, otalgia, vertigo, tinnitus was taken and recorded in a systematic manner.

After general physical examination, a thorough and detailed otorhinolaryngological examination and routine physical examination, these patients were subjected to a pre- operative HRCT Temporal bone.

The HRCT was done using a SIEMENS SOMATOM EMOTION 6 to get images in the axial and coronal plane, reformatting was done to a slice thickness of 0.6. Axial projection was obtained by serial 1–2 mm thin sections of the temporal bone with the line joining the infraorbital rim and external auditory meatus perpendicular to the table. The pre-operative scans were evaluated by a single experienced radiologist for the following main areas of interest

  1. 1.

    Malleus necrosis

  2. 2.

    Incus necrosis

  3. 3.

    Stapes necrosis

Although other parameters like bony erosions, soft tissue mass and facial nerve dehiscence were included in reporting but our main areas of evaluation for this study were concentrated on above said three findings. After thorough review of clinical findings, radiological findings and pathology of the disease, a surgical plan was formulated. General anesthesia fitness was taken and then surgery was performed. Canal wall up and canal wall down procedures were performed depending upon extent of the disease. Operative findings were recorded on a standard proforma under same parameters as above.

The pre-operative HRCT findings and intraoperative findings were then compared for each patient. A thorough Statistical evaluation was done, comparative tables were formulated for specific parameters to be studied.

McNemar test was used to formulate cross tabs comparing the pre-operative HRCT findings and intraoperative findings and hence, true positive, false negative, false negative, false positive and true negative patients were calculated for each of the malleus necrosis, incus necrosis, stapes necrosis. Using this data P values were calculated and so were the sensitivity, specificity, postive predictive values and negative predictive values of HRCT temporal bone in diagnosing the ossicular necrosis that was confirmed intraoperatively, considering intra operative findings the gold standard for comparison.

Following formulae were used

$${\text{Sensitivity }} = {\text{ True positive }}\left( {\text{TP}} \right)/{\text{True positive }}\left( {\text{TP}} \right) + {\text{False negative }}\left( {\text{FN}} \right) \times 100$$
$${\text{Specificity }} = {\text{True negative }}\left( {\text{TN}} \right)/{\text{False positive }}\left( {\text{FP}} \right) + {\text{True negative}} \times 100$$
$${\text{Positive predictive values }}\left( {\text{PPV}} \right) = {\text{TP}}/\left( {{\text{TP}} + {\text{FP}}} \right) \times 100$$
$${\text{Negative predictive values }}\left( {\text{NPV}} \right):{\text{ TN}}/\left( {{\text{FN}} + {\text{TN}}} \right) \times 100$$

Results

HRCT showed incidence of malleus erosion in about 29 out of 60 patients. Out of these only 22 confirmed malleus erosion intraoperatively. HRCT didn’t detect malleus erosion in 6 patients. The sensitivity was 78.5% and specificity 78.1% (Table 1; Fig. 1). HRCT detected incus erosion in 38 patients out of which only 30 were confirmed to have incus erosion on surgery. It had a sensitivity and specificity of 73.1 and 57.8% respectively (Table 2; Fig. 2). Stapes was found to be eroded in 28 patients on HRCT out of which only 13 were confirmed on surgery with sensitivity of 52% and specificity of 57.1% (Table 3; Fig. 3). P value was 0.05.

Table 1 Malleus erosion on HRCT in comparison to intraoperative findings
Full size table
Fig. 1
figure 1

Depicting various parameters of HRCT in evaluating malleus erosion

Full size image
Table 2 Incus erosion on HRCT in comparison to intraoperative findings
Full size table
Fig. 2
figure 2

Depicting various parameters of HRCT in evaluating incus erosion

Full size image
Table 3 Stapes erosion on HRCT as compared to intraoperative finding
Full size table
Fig. 3
figure 3

Depicting various parameters of HRCT in evaluating stapes erosion

Full size image

Discussion

The advent of HRCT in 1980 has allowed excellent pre-operative imaging of anatomy, evidence of extent of disease and screen for asymptomatic complications. It had not gained wide acceptance as an essential aid in planning surgery as most clinicians preoperative scanning for selected cases such as complication, revision surgeries or with congenital anomalies. Routine HRCT can be justified if it is shown to influence the clinical management, thus converting surgery into planned surgery.

The present study was carried out to find out specificity and sensitivity of HRCT temporal bone in determining the status of ossicular chain preoperatively. According to this diagnostic Observational study done in 60 patients, with preoperative scanning being done in every patient posted for tympanomastoid surgery and findings being confirmed intraoperatively it documented particularly that HRCT has high specificity and sensitivity in cases of ossicular erosion.

Although present study was limited to cases of chronic otitis media but some cases of unsafe pathology were also included. It was found that HRCT was effective in preoperative diagnosis of ossicular necrosis in such cases. Though the strength of this study was that it was a prospective study with a well balanced cohort and accurate data collection by same radiologist preoperatively and same surgeon intraoperatively in all cases but the drawback was that randomization was not done. Therapeutic plan was well framed and executed for all cases keeping surgical findings as gold standard for deciding extent of intervention in all cases. Study parameters were kept same for preoperative HRCT findings and intraoperative findings for balanced comparison, sensitivity, specificity, positive predictive negative predictive values were calculated according to standard formulas and P values was sorted out for each observation.

In present study of 60 cases, the sensitivity of HRCT in malleus erosion was 78.5% and specificity was 78.1%, and this was high as compared to sensitivity and specificity of stapes erosion which was 52.3 and 57.1% respectively, however P value for both malleus erosion and stapes necrosis was significant, giving evidence that HRCT is good diagnostic modality for preoperative identification of malleus and stapes necrosis. However reason for low sensitivity in cases of stapes erosion can be due to non consistent visualisation and when seen it appears as a soft tissue density in oval window niche and for these reasons it was not possible to distinguish between destruction of stapes and its mere entrapment by soft tissue.

In a study by conducted Rogha et al. [7] and Garg et al. [8] have reported that sensitivity of malleus erosion in both studies were 82.4 and 90% and specificity of 87 and 89.47% for malleus erosion, sensitivity and specificity for stapes necrosis came out to be 61.9, 66.7, 40 and 26.67% respectively. These studies correlated with the evidence suggested by present study.

However preoperative erosion of incus was seen in 63% cases (n = 60) while it was confirmed in 50% cases, there were 11 false negative cases, hence bringing sensitivity to 73.1% while specificity of HRCT dropped to 57.8%, positive predictive value 78.9% and negative predictive value was 50%, calculated P value was significant inferring that HRCT is good investigation for diagnosing incus necrosis preoperatively.

Present study correlated for diagnosing incus erosion in studies conducted by Shah et al., Zhang et al., Chee and Tan, Gomma et al., Datta et al., Rogha et al., O ‘ Donoghue et al., which showed sensitivity 90, 85, 85, 85.7, 87, 90.6, 81.4% and specificity of Shah et al. and Rogha et al. was 66.7 and 50% respectively [7, 9,10,11,12,13,14]. After correlation with above mentioned national and international literature it is our impression HRCT has a role in in evaluation of chronic otitis media cases preopratively.

Conclusion

HRCT is good preoperative diagnostic modality for identifying malleus, incus and stapes necrosis preoperatively and its use by otologist is to be encouraged for better adjunctive preoperative assessment and better surgical outcome.