Otitis media (OM) is one of the most prevalent infectious diseases worldwide: over 80% of children have acute OM (AOM) once before age 3, and 40% have 6 or more recurrences by age 7.1 Globally, OM is the third most important cause of hearing loss (HL), with a prevalence of 30.82 per 10,000.1,2 In developed countries, OM accounts for the largest number of non-well-child doctor visits and is the primary reason for childhood antibiotic prescriptions. Children spend a mean of 90 days on antibiotics for OM in their first two years of life.3

AOM, which is generally self-limited, has an incidence of 10.9%. Chronic OM (COM) is more difficult to diagnose and treat and has an incidence of 4.8%,4–6 OM complication rates are as high as 12.5%6,7 and can be extracranial and/or intracranial. The most common extracranial complications of OM include facial paralysis, subperiosteal abscess, mastoiditis, and labyrinthitis. The most common intracranial complications (ICCs) of OM are meningitis, cerebral abscess, lateral sinus thrombosis (LST), extradural abscess, otitic hydrocephalus, and encephalitis.8–11 Reported incidences for these complications (summarized in Table 1) range widely.

Despite the introduction of effective antibiotic therapy, the rate of ICC from OM is still about 8%. Anaerobic bacteria play a significant role.12 Thirty percent of complications from OM are intracranial; of ICCs, 5–26% result in mortality.13–15 ICCs of AOM usually develop rapidly and are often recognized and treated promptly, with reasonable recovery of function.16 Conversely, early COM is often unrecognized and untreated, and it may take years to develop complications, which are often hard to diagnose and are much more severe and life-threatening.

There is persistent inaccuracy in diagnosis of AOM by primary care physicians (PCPs) – pediatricians, family physicians, general practitioners, emergency room physicians, and otolaryngologists. In a study of 590 children, only 62% were diagnosed accurately by their PCPs, more frequently in complicated rather than simple AOM.17 Using video-recorded otoscopy of AOM, OM with effusion (OME), and normal, otolaryngologists were correct 74±16% of the time; pediatricians, 51±11%; and general practitioners, 46±21%.18 There is a lack of formalized PCP resident education in OM diagnosis, despite this being the most pediatric diagnosis.19,20 Diagnostic accuracy is most important in COM, to prevent the higher likelihood of intra- and extra-cranial complications.13 However, the average age at diagnosis for even congenital cholesteatoma is 4–5 years old.21 Even after entering otolaryngologic care, there is an average of 3.2 years delay in diagnosis of pediatric cholesteatoma.22 This unrecognized, and therefore untreated, COM can cause significant complications later on.

A 15-year review of negligence claims in the United Kingdom found that otologic claims made up 137 (26%); of the 97 successful claims, 15 were related to misdiagnosis/delayed diagnosis, with COM being the condition most frequently missed, and 4 additional cases were claims for morbidity due to delayed surgery.23

Even complications are difficult to diagnose: a study of four cases of pediatric lateral sinus thrombosis (LST), 3 from OM and 1 from tonsillitis, reported on the non-specific presentation and propensity to miss early diagnosis.24 In a US tertiary care hospital, asymptomatic ICCs were found in 8 of 11 children with acute mastoiditis treated over a one-year period.25 The clinical evolution of LST is very variable and is often missed because of lack of “pathognomonic” features, such as picket fence fever, which are, in fact, infrequently seen.26

ICCs of AOM and COM represent life-threatening situations and require immediate action. Due to atypical manifestations and the possibility of hidden expression, often related to the previous use of antibiotics, the prompt identification of ICCs requires a high clinical index of suspicion and a team approach by physicians, including PCPs, infectious disease specialists, otolaryngologists, radiologists, and neurosurgeons.27–29

Given the high incidence of OM, the inaccuracies in diagnosis, and the possibility of life threatening sequelae, this is an important ongoing problem.30

The aim of this study is to describe the epidemiological and clinical aspects of ICCs resulting from OM, to analyze the progression of the condition and to determine the relevant aspects that should be evaluated to minimize morbidity and mortality.

This is a retrospective review of clinical data that was carried out in a teaching and research institution. Fifty-one cases of ICC from OM identified from one emergency room (ER) over a 22-year period, from 1990 to 2012, were analyzed. In addition, the single year 2010 was selected for more detailed analysis of the incidence of ear infections and their complications as they presented to the ER of the tertiary referral hospital with 743 inpatient beds and 366,488 emergency room (ER) visits in the representative year 2010.

The treatment team included neuroradiology, neurosurgery, pediatrics, infectious disease, medical intensive care, and otolaryngology physicians. Specific tests, such as computerized tomography (CT) scan, magnetic resonance imaging (MRI), magnetic resonance (MR) angiography, and lumbar puncture were tailored to the patient and the presentation. The outcomes considered for analysis included: death, other permanent sequelae, and deafness. Patients were stratified by gender, type of otitis, and type of intracranial complication.

Data were collected in four main categories:

• 1.

  Demographic data: age at presentation, gender.

• 2.

  Disease data: the type of otitis media – AOM or COM, which was divided into non-cholesteatomatous COM (NCCOM) and cholesteatomatous COM (CCOM); patient age at first COM diagnosis; duration of illness (time between initial symptoms of OM and the development of the ICC); time delay to ICC diagnosis (interval between presentation with initial ICC symptoms and diagnosis of the ICC); the type of ICC, the bacteriology involved, and results of other specific tests as outlined above.

• 3.

  Treatment data: the clinical and surgical treatments adopted in each case. Preoperative imaging findings and suspicions were confirmed during surgery.

• 4.

  Outcome data: the duration of hospitalization, hearing status post-treatment, neurological sequelae and mortality rate.

Statistical analysis was performed by calculating the relative risk and correlation coefficient of Yule, with subsequent calculation of the X2 test and Pearson association. Statistical significance was set at 5% alpha with 1 degree of freedom (critical value of 3.84).

This study was approved by the Institutional Ethics Committee (protocol: 0352/08).

ICC from OM is seen predominantly in male children and young adults, which agrees with other reports in the literature.31–34 CCOM is responsible for over 76% of cases of ICCs. AOM-related ICCs occur mostly below age 15, which corresponds with the highest occurrence of AOM. It is customary to think about AOM as a childhood problem; however, one cannot overlook the second “bump” of AOM in the geriatric population. This bimodal distribution of incidence skews the average age of patients with ICC from AOM misleadingly higher.

ICCs associated with COM occur predominantly between 10 and 39 years of age. Assuming the “age of onset” of COM is approximately 11 years old and the “duration of the disease” is approximately 15 years, we can postulate that indolent pediatric cases of COM, which were not properly diagnosed, treated or followed, would become more aggressive disease that would culminate with intracranial complications during adolescence and early adulthood. COM, in particular CCOM, was the most common etiology in this report, a result that is similar to reports by other authors from other developing countries. In contrast, a different scenario has been observed in developed countries, such as Finland, where AOM accounts for more than 80% of the complications, and COM and CCOM are responsible for only 12% and 8% of complications, respectively.

With the progress achieved through collective action and public health, the mortality rate from infectious diseases has been decreasing in most of the world. In the mid-twentieth century, it was even thought that the eradication of infectious diseases would be achieved imminently;35 however, this prediction failed to materialize, even in countries currently considered highly developed. In fact, what occurred in many places was a change in the pattern of infectious diseases.36 At the present time, although otitis media has a benign course and typically responds well to treatment overall, some cases (both AOM and COM) develop severe and lethal complications with considerable mortality rates ranging from 8% to 18.6%.

Accurate examination with early diagnosis and intervention in cases of COM, before a complication has already developed, is very important in prevention of delayed ICC. Nonetheless, although there have been significant advances in the diagnosis and treatment resources available in medical practice, as well as undeniable socioeconomic improvements in Brazil over the last two decades, we observed a stable occurrence of ICCs over time.

Hearing loss due to OM is a worldwide public health issue. Our data confirm a high prevalence of significant hearing loss and deafness in cases of COM. The inflammatory and infectious events that take place in OM cases are still undervalued in this regard. Additionally, our data confirm that patients with ICCs due to OM have high hospital costs, permanent neurological sequelae in 29%, and a mortality rate of 7.8%, seen in the young and economically active. According to the Brazilian statewide system of data analysis (www.seade.gov.br), there was a total of 6701 admissions to the Brazilian public health system in the state of Sao Paulo related to “diseases of the ear and mastoid process” which cost R$ 2.5 million and resulted in 30 reported deaths in one year. Due to its prevalence and high socioeconomic cost, OM remains a pressing public health issue. Education and training of PCPs is essential for early identification of uncomplicated ear disease, as early and appropriate treatment may prevent the development of more aggressive OM and avoid the ICCs due to OM.

Once the complication has occurred, a multidisciplinary approach is required to maximize recovery. Currently, it is common that when there is a sick patient with a temporal lobe or cerebellar abscess, the possibility of concomitant OM is often overlooked. However, because the primary infection responsible is often not immediately diagnosed, the time of hospitalization and chances of sequelae are increased.

Neurosurgical drainage of the intracranial abscess, without removing the causative cholesteatoma, is directly related to relapse. In these cases, it is mandatory to surgically extirpate the ear infection because it is the focus of continued contamination of the central nervous system. Mastoidectomy should be performed as early as possible and during the same hospitalization. The abscess is frequently able to be drained into the ear.

Life-threatening intracranial complications from otitis media still occur. Complications in cases of acute otitis media usually occur in children or in the elderly. In contrast, complications of chronic otitis media are generally observed in young adulthood, particularly cases of cholesteatoma that begin in childhood and take several years to develop. Patients with intracranial complications remain hospitalized for a long period of time, and these individuals have significant morbidity from permanent auditory and neurological sequelae, as well as a substantial mortality rate. We have shown that the high incidence of OM, the inaccuracies in diagnosis, and the possibility of life threatening sequelae make this an important ongoing problem. As such, the primary treating physician must maintain a high index of suspicion when examining a patient with OM.

Sujana S. Chandrasekhar, MD – shareholder and board member, Scientific Development and Research, Inc. (Intranasal Surfactant Research). Other authors: none.