Introduction
The incidence of chronic suppurative otitis media (CSOM), with or without complications, is generally on the decline; however, this condition continues to be a common ear disease in developing countries. The complications of CSOM have a unique set of clinical signs and symptoms and carry high morbidity and mortality rates. Therefore, it is important to identify ears at risk of complications at an early stage, and try to prevent such complications.
The aim of this study was to compare the clinical and epidemiological profiles of cases of complicated and uncomplicated CSOM, based on their prognostic factors.
Materials and methods
This was a prospective, cross-sectional study undertaken in a tertiary care hospital from July 2003 to December 2005. The hospital catered for a predominantly rural population which had a relatively high incidence of CSOM.
Patients were selected by a multiphase, random sampling process from the ENT out-patient clinic. In the first phase (universe of the study, which is the total number of patients taken as an unit from which other subsets are derived), all patients attending the ENT out-patient clinic, with various complaints, were screened (7210 patients). The second phase included those patients from the first phase who had chronic otitis media (1257 ears). The third phase included those consecutive ears from the second phase which harboured granulation tissue, polyps or cholesteatoma (187 ears). This included 13 patients who had bilateral attico-antral ear disease. Out of the 187 ears, 11 had tubo-tympanic disease in the other ear.
The study included cases of both sexes and patients aged more than eight years.
The ears in the third phase were divided into two groups – complicated and uncomplicated CSOM – depending on the presence or absence of complications. One hundred and eighty-seven ears were studied in all, out of which 62 had complications. Among the ears with complications, five had attico-antral disease, while eight had tubotympanic disease in the other ear.
The inclusion criteria were as follows. The complicated CSOM group included ears with attico-antral CSOM with its complications (extracranial, intracranial or both). The uncomplicated CSOM group included ears with attico-antral CSOM and harbouring granulation tissue, polyps or cholesteatoma but with no complications.
Each patient underwent investigations like otoendoscopy, diagnostic nasal endoscopy and ear swab for microbiological analysis. High resolution computed tomography (CT) scanning of the temporal bone or CT scanning of the head was performed wherever necessary. Histopathological analysis of middle-ear granulation tissue was also performed in all cases undergoing surgery. Pure tone audiometry (PTA) with masking was performed in patients well enough to undergo testing.
The majority of the patients underwent canal wall down mastoidectomy. Other surgical procedures (e.g. incision and drainage of abscess, burr hole craniotomy) were performed according to the requirements of each individual case. Predisposing disease foci in the nose or throat were addressed medically or surgically at the same time.
Results
In our study, it was observed that all patients with complications invariably had unhealthy granulation tissue and/or cholesteatoma, with or without polyps, in their ears. There were no complications in cases with tubo-tympanic CSOM in the second phase. Forty-two ears had extracranial complications only, eight had intracranial complications only and 12 had both.Reference Bhat, Udayashankar, Naseeruddin, Venkatesha, Manjunath and Itgimath1 Multiple complications (two or more) were present in 21 ears.Reference Bhat, Udayashankar, Naseeruddin, Venkatesha, Manjunath and Itgimath1 Only one patient had complications in both ears. The details of patients' complications are given in Table I.
Table I CSOM complications detectedFootnote *
* In the complicated chronic suppurative otitis media (CSOM) group.
We compared the two groups with respect to the following prognostic variables: age, sex, domicile, literacy status, duration of ear discharge, hearing loss, predisposing disease focus in the nose or throat, ear pathology, and ear swab microbiology.
Age distribution
Patients' ages ranged from 8 to 60 years (Average age, 19.32 years) in the complicated CSOM group and 6 to 50 years (Average age, 20.06 years) in the uncomplicated CSOM group. Most of the complications occurred before the age of 25 years. Patient's age distribution is shown in Table II. There were more paediatric patients in the complicated CSOM group (43.54%) than in the uncomplicated CSOM group (36.8%). A two-sample ‘z’ test of the raw data did not show any statistically significant difference between the two groups at 95% confidence intervals (CI) (p = 0.630).
Table II Age distribution in complicated and uncomplicated CSOM groups
CSOM = chronic suppurative otitis media; yrs = years; CG = complicated CSOM group; UG = uncomplicated CSOM group; M=male; F = female
Sex
The complicated CSOM group showed a higher male predominance, with a male:female ratio of 2.4:1, compared with 1.8:1 in the uncomplicated CSOM group.
Domicile
In both the groups, the majority of the patients were from rural areas (Table III). There were more rural patients in the complicated CSOM group. There was a strong association between rural domicile and complications at 95% CI (Fisher's exact test; p = 0.0129). Calculation of relative risk and odds ratio revealed that rural patients had nearly twice the risk of developing complications, compared with urban patients (relative risk = 2.091, odds ratio = 2.771).
Table III Domicile in complicated and uncomplicated CSOM groups
Literacy status
In both groups, there were more illiterate patients than literate patients, although the percentage of illiterate patients was higher in the complicated CSOM group (Table IV). The association between literacy status and complications was not statistically significant at 95% CI (Fisher's exact test; p = 0.4415). However, calculation of relative risk (1.179) and odds ratio (1.278) showed an increased risk of complications in illiterate patients.
Table IV Literacy status in complicated and uncomplicated CSOM groups
Ear discharge duration
Table V shows the approximate duration of ear discharge at the time of diagnosis for the two groups. There was little difference between the two groups; the average ear discharge duration at presentation was 7.74 years (Range, 15 days to 23 years) for the complicated CSOM group and 7.32 years (range, 8 days to 32 years) for the uncomplicated CSOM group. This difference was not statistically significant, using a two-sample unpaired t-test at 95% CI (p = 0.661).
Table V Ear discharge duration at presentation, in complicated and uncomplicated CSOM groups
Ear pathology
Patients' ears were examined by otoendoscopy and the presence or absence of macroscopic cholesteatoma, granulation tissue or polyps was noted. Findings were further verified during definitive surgery, if undertaken. Granulation tissue from the middle ear was sent for histopathological examination and was analysed for keratin, cholesterol crystals, giant cells, macrophages etc. Both cholesteatoma and granulation tissue were found to be commoner in the complicated CSOM group.
The association between cholesteatoma presence and CSOM complications was found to be very significant at 99% CI (Fisher's exact test, p = 0.0069) (Table VI). Calculation of relative risk (2.085) and odds ratio (2.794) showed an increased risk of complications in the presence of cholesteatoma.
Table VI Presence of cholesteatoma in complicated and uncomplicated CSOM groups
Similarly, there was a statistically significant association between granulation tissue presence and complications at 95% CI (Fisher's exact test, p = 0.011) (Table VII). Calculation of relative risk (2.202) and odds ratio (2.948) showed that the risk of complications was increased more than twice in the presence of granulation tissue.
Table VII Presence of granulation tissue in complicated and uncomplicated CSOM groups
Cholesteatoma and granulation tissue were simultaneously present in 69.35% of cases in the complicated CSOM group, as compared with 34.4% in the uncomplicated group. There was a very strong association between such double pathology and complications at 99.9% CI (Fisher's exact test p < 0.0001), (Table VIII). Calculation of relative risk (2.658) and odds ratio (4.3157) showed that such double pathology increased the risk of complications by more than 2.5 times, compared with the effect of either cholesteatoma or granulation tissue when present separately.
Table VIII Presence of cholesteatoma and granulation tissue in complicated and uncomplicated CSOM groups
* Patients with cholesteatoma (Chol) and granulation tissue (GT), plus those with Chol or GT.
During surgical exploration of the mastoid and middle ear in complicated cases, we found relatively larger quantities of cholesteatoma and granulation tissue, especially in patients with multiple complications.
Predisposing disease focus in nose or throat
A diagnostic nasal endoscopy was performed to look for a disease focus in the nose or throat, such as persistent adenoids, tubal tonsillitis, hypertrophied turbinates, nasal polyp, sinusitis, atrophic rhinitis or grossly deviated nasal septum. A contributing disease focus in the nose or throat was found in 70.96% of complicated CSOM cases and in 82.4% of uncomplicated CSOM cases. The age distribution of patients with a predisposing nose or throat disease focus is shown in Table IX. There was no statistically significant association between such disease foci and CSOM complications at 95% CI (Fisher's exact test, p = 0.0886). Calculation of relative risk (0.6652) and odds ratio (0.5221) showed that the presence of nose or throat disease foci did not increase patients' risk of CSOM complications.
Table IX Presence of predisposing disease focus in nose or throat, by age
Ear swab microbiology
On microbiological analysis of ear swabs, both groups showed a similar pattern. Pseudomonas aeruginosa was the commonest organism found in both the groups (Table X).
Table X Microbiological organisms found in complicated and uncomplicated CSOM groups
Hearing loss
In the complicated CSOM group, all patients who underwent PTA had pure conductive hearing loss, while 85.71% had mixed hearing loss. In the uncomplicated CSOM group, of those who underwent PTA, 93.18% had pure conductive hearing loss, while 47.72% had mixed hearing loss. There was a statistically significant difference between hearing levels in the two groups at 99% CI, by two-sample unpaired t-test of raw data (Pure tone average in dB) (p < 0.001).
Discussion
Chronic suppurative otitis media would have been just another benign ear disease, but for its potential for complications. Complications of CSOM are particularly common in developing countries, especially in rural areas with low socioeconomic status and low literacy rates. Other contributing socioeconomic factors are ignorance, poverty, overcrowding, poor personal hygiene and lack of access to medical help. Hence, illiterate patients in rural areas are particularly vulnerable to CSOM complications. Yet another problem in rural areas is the late reporting of cases at an advanced stage of disease, often with multiple, simultaneous complications, due to lack of health awareness. In the third phase of our study, some ears had complications, while the others did not. Why is this so? Relevant information is scarce; very few published, prospective studies have compared the prognostic factors for complicated and uncomplicated CSOM. No previous prospective study has compared the effect of domicile, literacy status, predisposing disease focus in the nose or throat, and hearing loss.
In the current study, age, sex and duration of ear discharge did not appear to cause any statistically significant difference between the groups. However, a few studies have reported a younger ageReference Rupa and Raman2, Reference Singh and Maharaj3 and a shorter duration of ear discharge in patients with complicated CSOM.Reference Rupa and Raman2, Reference Panda, Sreedharan, Mann and Sharma4 Many authors have found a higher male:female ratio for patients with CSOM complications.Reference Singh and Maharaj3, Reference Kangsanarak, Fooanant, Ruckphapunt, Naracharoen and Teotrakul5, Reference Osma, Curreoghu and Hosoghu6 Granulation tissue and/or cholesteatoma have been major findings in complicated CSOM cases in a few previous studies.Reference Panda, Sreedharan, Mann and Sharma4–Reference Modak, Chavan, Borade, Kotnis and Jaiswal7 Most of the cases that had multiple complications or only intracranial complications were found to harbor cholesteatoma.Reference Bhat, Udayashankar, Naseeruddin, Venkatesha, Manjunath and Itgimath1 A few authors have reported cholesteatoma as a common finding in patients with intracranial CSOM complications.Reference Migirov, Duvdevani and Kronenberg8, Reference Kurien, Job, Mathew and Chandy9
Cholesteatoma and granulation tissue were identified as independent risk factors in the current study, and it appeared that their simultaneous existence in the same ear compounded the risk of developing a CSOM complication. This compounding effect might possibly have been due to facilitation of each other's bone-eroding effects. The quantum (i.e., volume) of granulation tissue or cholesteatoma could be an additional factor for bone erosion and destruction. Hence concomitant presence of cholesteatoma and granulation tissue could represent a greater risk factor than that posed by either of them alone. This might strongly predict the development of complications in CSOM. This may mean that such patients need priority treatment and require more extensive and meticulous surgical disease clearance.
• This study compared prognostic factors for complicated and uncomplicated attico-antral chronic suppurative otitis media, within a developing country
• Rural and illiterate patients had a higher risk of developing complications
• Coexistence of cholesteatoma and granulation tissue was a potential risk factor for complications
• Ears with complications were more prone to develop sensorineural hearing loss
We also observed that the likelihood of significantly improving the hearing of patients with complicated CSOM was not very high, because of the severe mixed hearing loss which was already present.
Both our study groups had a high incidence of disease foci in the nose or throat. However, there were no such foci in complicated CSOM patients above the age of 25 years. The presence of such disease foci was not associated with either group significantly, and did not represent a risk factor for complications.
A few other authors have reported pseudomonas species as the commonest organisms found in patients with CSOM complications, in agreement with our findings.Reference Kangsanarak, Fooanant, Ruckphapunt, Naracharoen and Teotrakul5, Reference Agrawal, Husein and MacRae10
We found sensorineural hearing loss to be more strongly associated with the complicated CSOM group than the uncomplicated CSOM group. Leskinen et al. found permanent hearing loss in 26% of their complicated CSOM cases.Reference Leskinen and Jero11
Conclusion
It is important to identify cases of CSOM which are more predisposed to the development of complications. In our study, rural domicile and illiteracy significantly increased patients' risk of CSOM complications. In such underprivileged patients, early detection (by screening tests) and timely treatment of CSOM may prevent the development of life-threatening complications. We found that patients with complications were younger and had a slightly longer duration of ear discharge at presentation. Cholesteatoma and granulation tissue when present is an important prognostic indicator of complications. Ears that harbour relatively large quantities of both cholesteatoma and granulation tissue together require early surgical intervention and extensive disease clearance in order to prevent complications. Patients with complications are more vulnerable to the development of sensorineural hearing loss. Age, sex, duration of ear discharge, predisposing disease foci in the nose or throat, and ear swab microbiology were not statistically significant prognostic indicators of CSOM complications in this study.
Acknowledgements
The authors wish to express their gratitude to the following persons, who made this study a success: Dr S R Itgimath, Lecturer in Statistics, Department of Community Medicine, Karnataka Institute of Medical Sciences, Hubli, India, for her valuable assistance on biostatistics; and Dr G N Prabhakar, Professor and Head, Department of Community Medicine, Sri Dharmastala Manjunatheshwara College of Medical Sciences, Dharwad, India, for his valuable assistance on epidemiology.
