- Email: [email protected]
Malignant external otitis: Insights into pathogenesis, clinical manifestations, diagnosis, and therapy
REVIEW ARTICLE
Malignant External Otitis: Insights into Pathogenesis, Clinical Manifestations, Diagnosis, and Therapy JENNIFER RUBIN, M.D., VICTOR L. Yu, M.D. Pittsburgh, ~ennsyhnia
Malignant external otitis is an infection of the external ear canal, mastoid, and base of the skull caused by Pseudomonas aeruginosa. The condition occurs primarily in elderly patients with diabetes mellitus. Current theories on pathsgenesis and anatomic correlations are reviewed. Severe, unrelenting otalgia and persistent otorrhea are the symptomatic hallmarks of the disease, whereas an elevated erythrocyte sedimentation rate is the only distinctive laboratory abnormality. Iatrogenic causes such as administration of broad-spectrum antibiotics and aural irrigation may play a predisposing role in high-risk populations. The disease can result in cranial polyneuropathies (with facial nerve [VII] paralysis being the most common) and death. The mainstay of treatment is administration of antipseudomonal antibiotics for four to eight weeks. Recurrence is common, and mortality remains at about 20 percent despite antibiotic therapy. Given the increasing longevity of diabetic patients, the frequency of this disease is increasing. Internists, family practitioners, and ambulatory care physicians must now be cognizant of the presenting symptoms, while infectious disease specialists and otolaryngologists need to be appraised of strides in diagnosis and therapy. The role of surgery should be minimized. Use of new diagnostic radiologic modalities and new antipseudomonal antibiotics discussed in this review should lead to improved outcome.
M
alignant external otitis is an often intractable infection caused by Pseudomonas aeruginosa, which occurs primarily in elderly diabetic subjects. The process frequently progresses to a skull base osteomyelitis with resultant cranial neuropathies. Despite long-term systemic antibiotic therapy, the disease is often recurrent and may be associated with significant mortality, Malignant external otitis was first described by Meltzer and Kelemen [I] in 1959 and the name was coined in Chandler’s [2] landmark 1968 article. However, an updated review is indicated since diagnostic, therapeutic, and prognostic aspects have changed. Specifically, new radiographic modalities and new antibiotic regimens have led to improvements in diagnosis and therapy. Theories regarding the pathogenesis of this disease, often conflicting, need to be critiqued based on current clinical and basic science concepts. The limited value of previously reported series of malignant external otitis is now more apparent. These limitations include a relatively small number of cases from any one institution, the retrospective nature of the studies, absence of defined time intervals in reporting patient follow-up, and omission of important clinical information when reporting individual cases (for example, type and duration of antibiotic therapy). Older reviews have presented conflicting views and inconsistencies on the basis for the association of diabetes mellitus and malignant external otitis, the precise etiologic role of P. aeruginosa versus other organisms, the prognostic significance of cranial nerve palsies, the exact role of radiography in the diagnosis and follow-up of patients, and the necessity for strict diabetic control. The incidence of malignant external otitis appears to be increasing, with 72 percent of cases reported after 1980; thus, the family practitioner, internist, and infectious disease specialist, as well as the otolaryngologist, should be familiar with the course and management of this complex disease. Finally, we present some new diagnostic and therapeutic recommendations for the clinician and the basis for these recommendations.
METHODS A computer-assisted search was performed primarily of the English language literature (one French, one Japanese, and one Portuguese article were also reviewed) on MEDLINE using the terms “malignant/ necrotizing external otitis.” Early sources were also identified from references in the more recent papers.
TERMINOLOGY The name of this disease is a source of continuing debate. Chandler [2] coined the term “malignant external otitis” in his initial description because of the high mortality of this syndrome. Evans and Richards September 1988 The American Journal of Medicine
Volume 85
391
MALIGNANT
EXTERNAL
OTITIS
/ RUBlN
AND YU
j
::
Figure 1. Anatomic sites of infection in malignant external otitis: the infection starts in the external ear canal, crosses the osseous-cartilaginous junction, and invades the temporal bone and adjoining structures. Cranial nerve involvement is directly related to its proximity to the path of infection. The cranial nerves affected, in order of frequency, are the facial nerve (VII) as it exits the stylomastoid foramen; the glossopharyngeal (IX), the vagus (X), and the accessory nerves (Xl) as they exit via the jugular foramen; and the hypoglossal nerve (XII) as it passes along the hypoglossal canal. Other sites of potential involvement include the temporomandibular joint, the meninges, sinuses, and parotid gland.
[3] introduced the term “necrotizing” in 1973, offering it as a more accurate description of an infectious process. Cohn [4] then added “progressive,” whereas Doroghazi et al [S] substituted “invasive” for “necrotizing.” It obviously would be less confusing for both the medical librarian and the physician if one name was consistently used. The objections raised to the term “malignant” are well taken. This is a bacterial infection and not a neoplastic process. In addition, despite its presumed beginnings as an infection of the auditory canal, the invasion of the organism into the temporal bone and skull base renders both “external” and “otitis” less accurate. In our opinion, none of the current nomenclature is ideal. In recognition of Chandler’s [2] contribution and because the majority of descriptions in the literature still employ the term malignant external otitis, we will retain this name throughout this article. Perhaps, as the understanding of the pathogenesis improves, a more accurate yet descriptive name can be formulated. And, more optimistically, as we learn to better control the course of this infection, terms such as “malignant,” “invasive,” “necrotizing,” and, “progressive” might be rendered obsolete.
BACTERIOLOGY The causative agent is virtually 392
September
1988
The American
always P. aeruginJournal
of Medicine
Volume
osa. It is unclear why there is such an overwhelming predilection for P. aeruginosa in this infection. This bacterium is an established commensal of the gastrointestinal tract and is rarely found in the external ear canal of healthy persons. Since the natural reservoir for this organism is water, accumulation of water and moisture is a likely inciting event. P. aeruginosa is also the most common etiologic agent of chronic external otitis. Although other organisms are often isolated concomitantly with Pseudomonas, they are generally habitants of normal skin such as diphtheroids, Staphylococcus aureus, S. epidermidis, and Candida albicans, among others. Isolated reports of other organisms causing malignant external otitis include two cases of Aspergillus fumigatus [6,7]; one case of S. aureus [S]; and one case of Proteus mirabilis [9]. One of these hosts was immunocompromised [6] and another was a chronically ill infant [9]. Only two of 260 (0.8 percent) of the cases of malignant external otitis in immunocompetent adults were not caused by P. aeruginosa. Therefore, stringent diagnostic criteria including bony destruction on computed tomographic scan and isolation of the organism from bone or mastoid should be met in any patient suspected of having malignant external otitis from microorganisms other than P. aeruginosa. 85
MALIGNANT
ANATOMY Spread of the pseudomonal infection in malignant external otitis follows a characteristic route that is reflected by the pattern of cranial neuropathies. The process is initiated in the external auditory canal and crosses the cartilaginous-osseous junction into the temporal bone. The infection then passes through the fissures of Santorini in the cartilaginous external meatus to the mastoid. The facial nerve becomes involved as it exits via the stylomastoid foramen (Figure 1). The infection can then spread along the base of the skull and may affect the glossopharyngeal, vagal, and spinal accessory nerves as they pass through the jugular foramen (the “jugular foramen syndrome”), as well as the hypoglossal nerve at the hypoglossal canal (Figure 1). Less often, the abducent and trigeminal nerves are involved at the petrous apex. There is one case in the literature of optic nerve involvement at the orbital apex in malignant external otitis [lo]. Because of its proximity to the initial site of infection, the facial nerve is almost always the first (and often only) cranial nerve affected, followed by the nerves exiting through the more distal jugular foramen and the hypoglossal canal. The trochlear and olfactory nerves are not known to be affected by this disease.
PATHOGENESIS The precise pathogenesis of malignant external otitis is unknown. Since the majority of cases occur in elderly diabetic patients, a theory that encompasses aging, diabetes mellitus, and an increased predilection for this bacterial infection would be ideal, Chronic hyperglycemia has been found in both an animal model and in in vitro studies to exert a depressive effect on host resistance, specifically polymorphonuclear phagocytic function [11,12]. In our experience, the occurrence of malignant external otitis does not necessarily coincide with periods of hyperglycemia. In fact, there are only two reports of diabetic patients with both malignant external otitis and diabetic ketoacidosis [13,14]. Furthermore, both of these cases occurred in adolescents with type I diabetes mellitus, whereas most cases of malignant external otitis arise in adultonset or type II diabetic patients. The increased serum glucose levels frequently noted at the time of presentation of malignant external otitis may reflect the infection exacerbating the diabetes. We believe hyperglycemia, per se, probably is not a factor in the pathogenesis of malignant external otitis. Aging and diabetes mellitus are both associated with altered immune function. The specific defects, however, are unclear. No generalized abnormality in cellular or humoral immunity appears to be present in patients with diabetes mellitus [X]. Investigations of phagocytic cell function in diabetic patients have provided variable results. The presence of a significant primary phagocytic dysfunction is difficult to reconcile with the adequate resistance mounted by diabetic patients against most bacterial organisms. In fact, there is surprisingly little evidence to suggest that diabetes mellitus is primarily associated with a general increased susceptibility to infections [16]. Diabetic microangiopathy is well described and was originally offered as a possible explanation for the predisposition of elderly diabetic patients to malignant external otitis [2]. Histopathologic studies of the capil-
EXTERNAL
OTITIS
/ RUBIN
AND YU
laries in the skin and subcutaneous tissue overlying the temporal bone have confirmed a thickening of periodic acid-Schiff-positive material in the subendothelial basement membrane [5]. Small vessel disease resulting in cutaneous hypoperfusion, diminished local host resistance, and, hence, increased susceptibility to infection is an attractive hypothesis for most patients with malignant external otitis. Moreover, blood vessel abnormalities are associated with diabetes and aging. There is a small but significant group of non-diabetic persons in whom malignant external otitis develops. The majority of pediatric cases of malignant external otitis do not have diabetes mellitus [17]. These children, however, do have evidence of compromised host resistance as manifest by malignancy and chemotherapy [14,18], malnutrition and anemia [9,19], neutrophil chemotactic disorders [20], and Stevens-Johnson syndrome with drug-induced leukopenia [21,22]. Therefore, the pathogenesis of malignant external otitis in the non-diabetic patient is probably related to immune dysfunction or a combination of impairments. This is in contrast to the predisposing vascular abnormalities in the patient with diabetes. Colonization of the external auditory canal by P. aeruginosa has not been shown to occur more frequently in diabetic versus non-diabetic patients [23]. However, we suspect that broad-spectrum oral antimicrobial chemotherapy so often used empirically for suspected ear infection may be the factor predisposing the external ear to P. aeruginosa colonization. For example, ampicillin, cephalexin, cefaclor, and trimethoprim-sulfamethoxazole have no activity against P. aeruginosa. Finally, an unrecognized but possible cause of malignant external otitis may be iatrogenic. Seven of nine patients with malignant external otitis recently seen by us had a history of ear irrigation within two weeks prior to the onset of symptoms. If nonsterile tap water (which may contain P. aeruginosa) is forcefully introduced into the ear of an elderly diabetic patient, we hypothesize that P. aeruginosa plus host factors produce an environment conducive to local invasion of the microbe followed by subsequent spread into bone and cartilage. Interestingly, a review of the literature disclosed other cases that developed following irrigation of the external ear [2,5,24-281.
CLINICAL FEATURES In an attempt to give an overview of the clinical aspects of malignant external otitis, data are summarized from the five largest series [2,5,29-401 (Table I). We caution that interpretation of this table should be tempered by the following limitations: nonuniform criteria for diagnosis, nonstandardized data collection, and missing information, For example, the presenting signs and symptoms are noted in only 31 percent (49 of 158) of the cases of these series and specific times of follow-up are even more rarely specified. The range of ages in the five studies is 11 to 91 years old, with the majority of patients over the age of 60. Ninety to 100 percent of the patients displayed some degree of glucose intolerance. These findings are consistent with the majority of reports describing malignant external otitis as a disease primarily of the old.er diabetic’patient. The specific diabetic status was only available for two of the series profiled [2,29,30,39,40] for a total of 69 patients. Of these individuals, 40 were
Seotember
1988
The American
Journal
of Medicine
Volume
85
393
MALIGNANT
EXTERNAL
OTITIS
/ RUBIN
AND YU
insulin-dependent, 25 were taking an oral hypoglycemic agent, 19 had their disease controlled with diet alone, and nine were diagnosed at presentation with malignant external otitis by a glucose tolerance test (chemical diabetes). As discussed previously, the development of malignant external otitis does not seem to be related to the severity of the diabetes or control of the hyperglycemia. In our experience, the disease developed in persons with only a positive glucose tolerance test as well as those with insulin-dependent diabetes of many years’ duration. It should be emphasized that the usual signs of infection are notably absent. For example, fever and tachycardia are distinctly unusual. The patients are usually non-toxic alert. Otalgia is the most common presenting symptom in the literature (noted 75 to 100 percent of the time). Although otalgia occurs in patients with external otitis and otitis media, the degree of pain is extraordinarily severe and unrelenting. The very intensity should suggest the diagnosis. One of our patients has described the excruciating pain as “starting from the top of my head and piercing through my skull down to my spine” and another as “a drilling pain that drives me to my knees.” Holder et al [lo] reported a case of suicide brought on by two years of unrelenting pain and infection. The pain tends to be nocturnal, frequently awakening the patient from sleep [30]. Most of our patients require analgesics before they can sleep. Headache (usually temporal or occipital) may be present and is remarkable too for its severity. We have also been impressed with the presence of classic temporomandibular joint pain with tenderness at the joint area and pain upon opening the mouth. Since the temporomandibular joint impinges onto the external ear canal (Figure l), it is surprising that this symptom has not received greater emphasis in previous clinical descriptions. In some patients, the pain is so severe that food intake is voluntarily limited. Tenderness and swelling about the periauricular area are common, especially if the patient has had symptoms for many months prior to seeing a physician. Purulent otorrhea is usually present at some point in the course of the illness (50 to 81 percent). The degree of drainage is variable, ranging from copious, foul-smelling, greenish exudate to minimal accumulation of moisture. Late in the course of the disease, the otorrhea appears to decrease. When parenteral antipseudomonal antibiotics are given, resolution of the drainage usually occurs within days. Granulation tissue is usually present in the external canal; histologic examination generally shows nonspecific inflammatory changes. Edema with erythema of the ear canal is common; however, the tympanic membrane is usually intact. Cranial nerve dysfunction can appear as early as one week after the onset of symptoms, although it usually occurs more than two months into the illness [35]. The incidence of facial nerve involvement in these series ranges from 24 to 43 percent, while other cranial neuropathies were found 14 to 35 percent of the time. It is noteworthy that the incidence of cranial neuropathies associated with malignant external otitis has not diminished since Chandler’s [2] first reported series in 1963. This finding implies a consistent delay in diagnosis. All too often these patients are only recognized 394
September
1988
The American
Journal
of Medicine
Volume
85
MALIGNANT EXTERNAL OTITIS / RUBIN AND YU
as having malignant external otitis after weeks to months of treatment for simple external otitis. The prognostic significance of cranial nerve (specifically facial) palsies has been debated. Although the mortality is greater in those with cranial neuropathies, it is probably not the nerve palsy that directly contributes to the patient’s demise but rather is a reflection of the advanced disease. The cranial nerve paralysis may not necessarily be permanent. For example, Chandler [31] reported that of those patients with facial nerve paralysis who survived, 22 percent experienced a complete recovery and 9 percent a partial recovery, whereas 17 percent sustained $ermanent paresis [31]. Recovery of function, however, is unpredictable on a case-by-case basis. There have been isolated case reports of other complications, including brain abscess [41-431, sphenoidal sinusitis [27,44-471, mycotic aneurysm [47], parotitis [41], lateral sinus thrombosis [3], and sigmoid sinus thrombosis [30]. These sequelae are noteworthy both in their seriousness and extreme rarity. All of the patients in the literature with meningitis, dural sinus thrombosis, or intracranial abscess died of their disease.
infectious process [56]. A mere 10 percent increase in osteogenic activity is sufficient for detection by bone scanning; this is the reason for its superiority to conventional radiography in demonstrating disease activity [56]. Although this modality is highly sensitive (reported to be 100 percent) [55], the specificity is low. The tracer collects at sites of osteogenic activity in response to both osteolytic and osteoblastic lesions. Thus, a “positive” bone scan may be seen in any inflammatory disease of the head and neck including carcinoma, temporomandibular joint disorders, fractures, and infections. In one study, radiologists who were not informed of the patients’ diagnoses were unable to differentiate scans of patients with malignant external otitis from those with severe external otitis [57]. Although Ostfeld et al [49] claimed that bone scan results correlated well with the clinical course of the disease, this finding has not been confirmed. We conclude that bone scanning with technetium 99 may be helpful in confirming the clinical diagnosis of malignant external otitis, but it may not differentiate this disease from simple external otitis or carcinoma (the leading alternate diagnoses), and is not particularly useful in monitoring the course of disease.
Numerous radiologic techniques have been used in the assessment of patients with malignant external otitis. These include plain films, polytomography, technetium bone scan, gallium scan, computed tomography, and magnetic resonance imaging. Transmission Radiography Most of the radiographic experience in the literature is based upon plain films and polytomography. The combination of skull films, mastoid series, and temporal bone tomograms may demonstrate soft tissue within the external auditory canal as well as increased density of the mastoid air cells with or without bone destruction. These studies are inadequate, however, in demonstrating infratemporal or intracranial extension of disease [38,48,49]. Some authors have suggested that the findings with transmission radiography correlate well with the clinical stage of malignant external otitis [36,48]; specifically, “early” and “late” disease may be differentiated on the basis of the presence or absence of bone destruction. We agree that bony erosion on plain films or tomograms certainly confirms that disease is advanced. However, since 30 to 50 percent demineralization of cancellous bone occurs before a lytic lesion can be visualized radiographically [50], we suggest that these techniques lack adequate sensitivity. In addition, any disease that produces a destructive lesion of the temporal bone (malignancy) can mimic the conventional radiographic findings in malignant external otitis [48]. Because of the limited sensitivity and specificity, plain films and tomographic scans of the skull and mastoid appear to be of minimal usefulness in the evaluation of this disease. Bone Scanning (Technetium 99) Bone scanning with technetium 99 is frequently used in confirming the clinical diagnosis of malignant external otitis [33,49,51-581. The radionuclide tracer accumulates at sites of osteoblastic activity presumably in response to bone destruction caused by the
Gallium Scanning Gallium citrate is incorporated into granulocytes and localizes in areas of inflammation in soft tissue and bone [59-611. The intense inflammatory activity in malignant external otitis in contrast to the minimal accumulation of gallium normally seen in the head region is what makes it useful in this disease [55]. The sensitivity of gallium 67 scanning in the diagnosis of malignant external otitis has ranged from 89 to 100 percent [33,53,56]. Like bone scanning, gallium scanning has the disadvantages of low specificity and imprecision in anatomic localization of disease. Its usefulness in clinical follow-up is unclear. Some investigators have suggested that findings from serial gallium scans can be helpful in predicting clinical resolution [33,51,54,55]. On the other hand, Gherini et al [52] have described two patients who had recurrence of disease despite a normal gallium study. We conclude that controlled studies correlating results of serial gallium scans and clinical course are necessary in order to determine its use in monitoring therapy for malignant external otitis. Computed Tomography (CT) Computed tomography is the current modality of choice for defining the anatomic extent of disease in malignant external otitis [33,38,62]. The ability of the CT scan to delineate fat and muscle planes makes it the ideal modality for viewing the subtemporal area [38]. The sensitivity of CT in diagnosing and following the course of the disease has not yet been established, although it has been reported to be as low as 30 percent at presentation in one report [56]. The majority of studies have found that clinical remission does not appear to coincide with normalization of the scan [34,38,40,52,56]. On the other hand, Gold et al [33] concluded that serial scans could aid in evaluating the effectiveness of therapy given the ability of the CT scan to demonstrate the resolution of soft tissue disease. Serial scans could provide objective criteria for terminating antibiotic therapy. Although this conclusion seems reasonable, confirmation in a prospective
SeWember
1988
The American
Journal
of Medicine
Volume
85
395
MALIGNANT
EXTERNAL
OTITIS
/ RUBIN
AND YU
study would be useful. Meaningful data gleaned from a CT scan might also include abnormalities of the external auditory canal, presence of fluid in the middle ear or mastoid, subtemporal involvement of cranial nerve VII, and spread of the infection across the midline [63]. Recurrence or progression of disease can be documented by sequential studies when a baseline study is obtained and results are correlated with clinical data [33,38,62]. A limitation of the CT scan is its inability to differentiate malignant external otitis from carcinoma. Two cases have been reported in the literature [29,64] in which malignant external otitis apparently coincided with or pre-dated squamous cell carcinoma. We recommend obtaining a scan at the time of initial presentation and repeating the study if the patient fails to show a response to antibiotic therapy, or if symptoms develop following completion of treatment. Magnetic Resonance Imaging Magnetic resonance imaging (MRI), like CT, is an anatomic imaging modality. Experience with this technique in malignant external otitis is limited. A single report comparing CT and MRI in defining the anatomic extent of disease concluded that the latter was superior in delineating soft-tissue involvement; two of four patients had evidence of contralateral disease by MRI that was not apparent on CT [52]. Both CT and MRI findings remained abnormal in these patients at three months and five months, respectively, after clinical resolution of disease. Comparison of the two modalities in more patients is needed before the relative value of each may be assessed.
LABORATORY PARAMETERS Hematologic and chemical laboratory parameters are generally unaffected by this condition. Surprisingly, even leukocytosis and the presence of immature forms in the differential count (“left shift”) are uncommon, reflecting the insidious but localized aspects of this infection. Erythrocyte Sedimentation Rate We have found the erythrocyte sedimentation rate (ESR) to be valuable in establishing the diagnosis and in monitoring treatment, yet it has not been widely employed in the diagnosis or follow-up of patients with malignant external otitis. Although it is a nonspecific indicator of disease, its sensitivity in malignant external otitis suggests its use in confirming the diagnosis in patients with pseudomonal otitis refractory to therapy. Even more importantly, it can serve as an objective standard for the status of infection in an individual patient. In our experience, the ESR was elevated in 25 of 25 (100 percent) of the cases of malignant external otitis (mean = 87 mm/hour). Moreover, the value was greater than 100 mm/hour in many patients, placing malignant external otitis in that group of diseases that can result in an ESR greater than 100 mm/hour. The differential diagnosis of an extremely elevated ESR includes malignancy, collagen vascular disease, and inflammatory bowel disease in addition to infections [65]. We note that these findings also support previous contentions that infections of bone may have particular propensity to produce markedly accelerated erythrocyte sedimentation [66].
396
September
1988
The American
Journal
of Medicine
Volume
Since carcinoma of the head and neck and simple external otitis (the two primary alternate diagnoses when considering malignant external otitis) rarely produce ESR elevations to the degree seen in malignant external otitis, this simple noninvasive test might be particularly useful in selected patients in whom malignant external otitis is suspected. In addition to supporting the diagnosis at presentation, the ESR is also a good indicator of disease activity [40]. The ESR will usually respond within two weeks of effective therapy, but it may be months before normal values are attained, especially in patients with prolonged infection prior to initiation of antibiotic therapy. We suggest that a baseline ESR value be obtained in all patients with a suspected diagnosis of malignant external otitis as an index of inflammatory disease; subsequent values can be used to monitor the response to therapy. Re-elevation of the ESR can be the first objective indication of recurrent disease. Glucose Tolerance Approximately 10 percent of the cases of malignant external otitis in the literature occur in non-diabetic patients, with the majority developing in the elderly diabetic host. Therefore, we suggest that glucose tolerance be evaluated in any patient with suspected malignant external otitis and no history of diabetes. If random and fasting serum glucose values are normal, a complete (usually five-hour) glucose tolerance test should be considered to rule out chemical diabetes.
THERAPY Meticulous control of the serum glucose level in diabetic patients is often recommended for those with malignant external otitis. Progression of the infection has never been correlated with severity of hyperglycemia; nevertheless, we agree it is prudent to maintain euglycemia in the diabetic patient with malignant external otitis. The role of surgery in the treatment of malignant external otitis has changed since Chandler’s [2] initial report in 1968 in which all 13 of his patients underwent at least one major surgical procedure in attempting to control the infection. Nine to 100 percent of the patients in these series underwent some form of surgery (usually mastoidectomy, facial nerve decompression, polypectomy, or a deep tissue debridement). Facial nerve decompression is often performed, but in our experience, this procedure has not contributed to the recovery of function. Subtotal temporal bone resection for the drainage of deep-seated cranial abscesses has been advocated in selected patients whose condition is unresponsive to aggressive medical therapy and surgical debridement [67]. We now believe that the role of surgery in malignant external otitis should generally be limited to local debridement and the excision of accessible foci of infection such as polypectomy or sequestrum removal. We believe topical antibiotics, although commonly used in external otitis, have no role in the treatment of malignant external otitis; local instillation of antibacterial agents is insufficient for this invasive pseudomonal infection and only hinders culture isolation of the bacterium from ear drainage. The hyperbaric oxygen chamber has been used with varying success in anecdotal reports [32,53,68,69].
85
MALIGNANT
However, based on the known aerobicity of Pseudomonas and the small number of cases treated, this modality must still be considered experimental. With the advent of semi-synthetic penicillins, specifically carbenicillin in 1969, systemic antipseudomonal antibiotics have emerged as the mainstay of therapy. This transition from primarily surgical to medical management resulted in a decreased mortality; Chandler [29] quoted a mortality rate decrease from 46 percent to 23 percent. However, it is important to note that the death rate has remained essentially unchanged in 20 years despite the introduction of the antipseudomonal penicillins and aminoglycosides. Currently, accepted therapy is the long-term administration of an aminoglycoside (e.g., tobramytin) with a beta-lactam agent (e.g., piperacillin or azlocillin). The optimal duration of treatment has not been established, but most authors recommend at least four to eight weeks of intravenous therapy. Third-generation cephalosporins, notably moxalactam and cefsulodin, have been reported to be successful in well-described anecdotal reports [34,39,68,70]. However, it appears that recurrence occurs as commonly with these newer agents as with previous antipseudomonal beta-lactam therapy. Other antipseudomonal agents that may be considered in selected situations in which a penicillin or aminoglycoside would be hazardous include aztreonam, ceftazidime, and imipenem. A recent study at our institution has demonstrated efficacy in the treatment of 11 patients with malignant external otitis using an oral regimen of ciprofloxacin, a new quinolone agent, and rifampin. At a mean followup of 18 months, there have been no deaths from disease [40]. We anticipate this regimen will be a significant advance in this difficult-to-treat infection.
PROGNOSIS Mental status deterioration is a poor prognostic sign, generally reflecting intracranial complications from the disease. Increasing mortality is associated with cranial nerve involvement, with the highest mortality associated with cranial polyneuropathies. Recurrence is a well-recognized aspect of this disease (9 to 27 percent) and is usually first manifest by the insidious onset of otalgia followed by headache and temporomandibular joint pain. Surprisingly, otorrhea may be minimal or not reappear at all, reflecting the dominance of skull base osteomyelitis in causation of the clinical syndrome. The first objective sign is usually the re-elevation of the ESR. The issue of when to consider an infection as having been cured has practical implications for patient follow-up. Our experience and a review of the literature indicate that infection can recur as long as four to 12 months after the termination of antibiotic therapy. Outcome and follow-up are inconsistently reported in the cases described in the literature. We suggest, therefore, that a period of one year elapse after the discontinuation of antibiotics before a patient’s infection is reported as cured.
SUMMARY Malignant external otitis is a Pseudomonas infection that begins in the external auditory meatus and not infrequently spreads to involve the temporal bone,
EXTERNAL
OTITIS
/ RUBIN
AND YU
zygomatic root, and the base of the skull. Small blood vessel abnormalities, associated with both aging and diabetes, probably contribute to the tendency for infection by disrupting nutrient supply to cutaneous tissues. In addition, iatrogenic factors such as the empiric institution of broad-spectrum antibiotics and aural irrigation may play a predisposing role. The diagnosis of malignant external otitis is supported by unrelenting otalgia frequently accompanied by purulent otorrhea in elderly diabetic patients. The ESR is usually markedly elevated and radiographic studies, especially CT scan, may show evidence of bone destruction. The proclivity towards cranial neuropathies, most commonly facial nerve paralysis, and a 20 percent mortality rate necessitates early diagnosis and prompt institution of appropriate therapy. Currently, the mainstay of therapy is long-term (four to eight weeks) intravenous administration of combination antipseudomonal antibiotics in conjunction with local debridement of infected tissue. The quinolone antibiotics, particularly ciprofloxacin, will likely prove to be of major usefulness.
ACKNOWLEDGMENT We thank Drs. Ralph Caprosa, Donald Kamerer, Eugene Meyers, and J.R. Chandler for their critical review,‘Dr. Nicholas Cauna for assistance in the design of Figure 1, Mr. Ron Filer for the artwork for Figure 1, and Ms. Shirley Brinker for manuscript preparation.
REFERENCES 1. Meltzer PE, Kelemen G: Pyocutaneous osteomyelitis of the temporal bone, mandible and zygoma. Laryngoscope 1959; 169: 1300-1316. 2. Chandler JR: Malignant external otitis. Laryngoscope 1968; 78: 1257-1294. 3. Evans ITG. Richards SH: Malignant (necrotizing) otitis externa. J Otolaryngol Otol 1973; 87: 13-20. 4. Cohn AM: Proeressive necrotizine otitis. Arch Otolarvnnol 1974; 99: 136-139. 5: Doroghazi RM: Nadol JB, Hyslop NE, Baker AS, Axelrod L: Invasive external otitis. Report of 21 cases and review of the literature. Am J Med 1981; 71: 603614. 6. Petrak RM, Pottage JC, Levin S: Invasive external otitis caused by Aspergilus fumigatus in an immunocompromised patient. J Infect Dis 1985; 151: 196. 7. Cunningham MJ, Yu VL, Turner J, Curtin HD: Necrotizing otitis externa due to Aspergillus in an immunocompetent patrent. Arch Otolaryngol Head Neck Surg 1988; 114: 554-556. 8. Bayardelle P, Jolivet-Granger M, Larochelle D: Staphylococcal malignant external otitis. Can Med Assoc J 1982; 126: 155-156. 9. Coser PL, Stamm AEC, Lobo RC. Pinto JA: Malignant external otitis in infants. Laryngoscope 1980; 90: 312-315. 10. Holder CD, Gurucharri M, Barteis LJ, Colman MF: Malignant external otitiswith optic neuritis. Laryngoscope 1986; 96: 1021-1023. 11. Drachman RH, Root RK, Wood WB: Studies on the effect of experimental nonketotic diabetes mellitus on antibacterial defense. J Exp Med 1966; 124: 227246. 12. Bagade JD, Nielson K, Root J, Bulger R: Host defense in diabetes mellitus: the feckless phagocvte during poor control and ketoacidosis. Diabetes 1970; 19: 364. 13. Merritt WT, Bass JW,Bruhn FW: Malignant external otitis in an adolescent with diabetes. J Pediatr 1980; 96: 872-873. 14. Sobie S, Brodsky L, Stanievich JF: Necrotizing external otibs in children: report of two cases and review of the literature. Laryngoscope 1987; 97: 598-601. 15. Galbraith RM: Immunological aspects of diabetes mellitus. Boca Raton, Florida: CRC Press, 1979. 16. Miller ME: Phagocyticcellfunctions in diabetes mellitus. In: Gupta S, ed. Immunology of clinical and experimental diabetes. New York: Plenum Medical, 1986; 369-383. 17. Rubin J. Yu VL. Stool S: Malignant external otitis in children: a critical review. J Pediatr (in press). 18. Rubenstein JM. Ostfeld E. Ben-Zarav S. Schiby G: Necrotizing external otitis. Pediatrics 1980; 66: 618-620. 19. Joachims HZ: Malignant external otitis in children. Arch Otolaryngol1976; 102: 236-237. 20. lchimura K, Hoshino T, Yano J, Nozue M: Neutrophil disorders in a child wrth necrotizing external otitis. J Otolaryngol 1983; 12: 129-133. 21. Sherman P, Black S, Grossman M: Malignant external otitis due to Pseudomo nas aeruginosa in childhood. Pediatrics 1980; 66: 782-783. 22. Horn KL, Gherini S: Malignant external otitis of childhood. Am J Otol 1981; 2:
September
1988
The American
Journal
of Medicine
Volume
85
397
MALIGNANT
EXTERNAL
OTITIS
/ RUBIN
AND YU
402-404. 23. Salit IE, Miller B, Wigmore W. Smith JA: Bacterial flora of the external canal in diabetics and non-diabetics. Laryngoscope 1982; 92: 672-673. 24. Dawson DA: Malignant external otitis. J Laryngol Otol 1978; 92: 803-810. 25. Dinapoli RP. Thomas JE: Neurologic aspects of malignant external otitis: report of three cases. Mayo Clin Proc 1971; 46: 339-344. 26. John AC, Hopkin NB: An unusual case of necrotizing otitis externa. J Otolaryngol Otol 1978; 92: 259-264. 27. Schwarz GA, Blumenkrantz MJ, Sundmaker WLH: Neurologic complications of malignant external otitis. Neurology 1971; 21: 1077-1084. 28. Wilson DF, Pulec JL, Linthicum FH: Malignant external otitis. Arch Otolaryngol 1971q: 419-422. 29. Chandler JR: Pathogenesis and treatment of facial paralysis due to malignant external otitis. Ann Otol 1972; 81: 648-658. 30. Chandler JR: Malignant external otitis and facial paralysis. Otolaryngol Clin North Am 1974; 7: 375-383. 31. Chandler JR: Malignant external otitis. Further consideration Ann Otol 1977; 86: 417-428. 32. Lucente FE, Parisier SC, Som PM: Complicationsof the treatment of malignant external otitis. Laryngoscope 1983; 93: 279-281. 33. Gold S, Som PM, Lawson W, Lucente FE, Mendelson M, Parisier SC: Radiographic findings in progressive necrotizing “malignant” external otitis. Laryngoscope 1984: 94: 363-366. 34. Mendelson MH, Meyers BR, Hirschman SZ, Shapiro ER, Parisier SC: Treatment of invasive external otitis with cefsulodin. Rev Infect Dis 1984; 6: S698-S704. 35. Corey JP, Levandowski RA, Panwalker AP: Prognostic implications of therapy for necrotizing external otitis. Am J Otol 1985; 6: 353-358. 36. Kim BH: Roentgenographic findings of malignant external otitis. AJR 1971; 112: 366-372. 37. Sando I, Harada T, Saito R, Okano Y, Caparosa RH: Temporal bone histopathology of necrotizing external otitis. Ann Otol 1981; 90: 109-115. 38. Curtin HD, Wolfe P, May M: Malignant external otitis: CTevaluation. Radiology 1982: 145: 383-388. 39. Haverkos HW, Caparosa R, Yu VL, Kamerer D: Moxolactam therapy-its use in chronic suppurative otitis media and malignant external otitis. Arch Otolaryngol 1982: 108: 329-333. 40. Yu VL, Stoehr G, Rubin J, Matador A, Kamerer D: Efficacy of oral ciprofloxacin plus rifampin for therapy of malignant external otitis (abstr 188). In: Proceedings of the 27th Interscience Conference on Antimicrobial Agents and Chemotherapy. New York: American Society of Microbiology, 1987. 41. Caruso VF, Griffiths CM, Bailey BJ: Malignant external otitis. Tex Med 1977: 73: 106-113. 42. Kohut RI, Lindsay JR: Necrotizing(“malignant”) external otitis histopathologic processes. Ann Otol 1979; 88: 714-720. 43. Soliman AE: A rare case of malignant external otitis externa in a non-diabetic patient. J Laryngol Otol 1978; 92: 811-812. 44. Youngs R, Bagley J: Sphenoidal sinusitis secondary to malignant external otitis. J Laryngol Otol 1986; 100: 341-344. 45. Uri N, Kitzes R, Meyer W, Schuchman G: Necrotizing external otitis: the importance of prolonged drug therapy. J Otolaryngol Otol 1984; 98: 1083-1085. 46. Pace-Balzan J: Necrotizing otitis externa. J Laryngol Otol 1977; 91: 735738. 47. Watson RC: The neuropathology of malignant external otitis. Bull Los Angeles Neural Sot 1977; 42: 51-56.
398
September
1988
The American
Journal
of Medicine
Volume
48. Mendez G, Quencer RM, Donovan Post MJ. Stokes NA: Malignant external otitis: a radiographic-clinical correlation. AJR 1979; 132: 957-961. 49. Ostfeld E, Aviel A, Pelet D: Malignant external otitis: the diagnostic value of bone scintigraphy. Laryngoscope 1981; 91: 960-964. 50. Jones AG, Francis MD, Davis MA: Bone scanning: radionuclide reaction mechanisms. Semin Nucl Med 1976; 6: 3-18. 51.Garty I, Rosen G, Holdstein Y: The radionuclide diagnosis, evaluation and follow-up of malignant otitis externa (MOE)-the value of immediate blood pool scanning. J Laryngol Otol 1985; 99: 109-115. 52. Gherini SG. Brackmann DE, Bradley WG: Magnetic resonance imaging and computerized tomography in malignant external otitis. Laryngoscope 1986; 96: 542-548. 53. Mendelson DS, Som PM, Mendelson MH, Parisier SC: Malignant external otitis: the role of computed tomography and radionuclides in evaluation. Radiology 1983; 149: 745-749. 54. Parisier SC, Lucente FE, Hirschman SZ, Som PM, Arnold LM, Ruffman JD: Nuclear scanning in necrotizing progressive “malignant” external otitis. Laryngoscope 1982; 92: 1016-1020. 55. Reiter D, Bilaniuk LT, Zimmerman RA: Diagnostic imaging in malignant otitis externa. Otolaryngol Head Neck Surg 1982; 90: 606-609. 56. Strashun AM, Nejatheim M, Goldsmith SJ: Malignant external otitisearly scintigraph detection. Radiology 1984; 150: 541-545. 57. Noyek AM: Bone scanning in otolaryngology. Laryngoscope 1979; 89(suppl 18): l-29. 58. Levin WJ, Shary JH, Nichols LT, Lucente FE: Bone scanning in severe external otitis. Laryngoscope 1986; 96: 1193-1195. 59. Tsan M, Chen WY, Scheffel V. Wagner HN: Studies on gallium accumulation in inflammatory lesions: gallium uptake by human polymorphonuclear leukocytes. J Nucl Med 1978; 19: 36-43. 60. Menon S, Wagner HN, Tsan MF: Studies on gallium accumulation in inflammatory lesions: uptake by Staphylococcus aureus: concise communication. J Nucl Med 1978; 19: 44-47. 61. Silva J, Harvey WC: Detection of infections with gallium-67 and scintigraphic imaging. J Infect Dis 1974; 130: 125-131. 62. Chakeres DW, Kapila A, LaMasters D: Soft-tissue abnormalities of the external auditory canal: subject review of CT findings. Radiology 1985; 156: 105-109. 63. Rubin J, Curtin HD, Yu VL, Hirsch BE: The utility of serial CT scans in the diagnosis and follow-up of-malignant external otitis. American Academy of Otolaryngology-Head and Neck Surgery, 1988; Washington, D.C. 64. Matucci KF. Setzen M, Galantich P: Necrotizing otitis externa occurring concurrently with epidermoid carcinoma. Laryngoscope 1986; 96: 264-266. 65. Zacharski LR, Kyle RA: Significance of extreme elevation of erythrocyte sedimentation rate. JAMA 1967; 202: 116-l 18. 66. Wyler DJ: Diagnostic implications of markedly elevated erythrocyte sedimentation rate: a reevaluation. South Med J 1977; 70: 1428-1430. 67. Raines JM, Schindler RA: The surgical management of recalcitrant malignant external otitis. Laryngoscope 1980; 90: 606-609. 68. Mader JR, Love JT: Malignant external otitis-cure with adjunctive hyperbaric oxygen therapy. Arch Otolaryngol 1982; 108: 38-40. 69. Lucente FE, Parisier SC, Som PM, Arnold LM: Malignant external otitis: a dangerous misnomer? Otolaryngol Head Neck Surg 1982; 90: 266-269. 70. Meyers BR, Mendelson MH, Parisier SC, Hirschman SZ: Malignant external otitis: comparison of monotherapy vs. combination therapy. Arch Otolaryngol Head Neck Surg 1987; 113: 974-978.
85
Malignant External Otitis: Insights into Pathogenesis, Clinical Manifestations, Diagnosis, and Therapy JENNIFER RUBIN, M.D., VICTOR L. Yu, M.D. Pittsburgh, ~ennsyhnia
Malignant external otitis is an infection of the external ear canal, mastoid, and base of the skull caused by Pseudomonas aeruginosa. The condition occurs primarily in elderly patients with diabetes mellitus. Current theories on pathsgenesis and anatomic correlations are reviewed. Severe, unrelenting otalgia and persistent otorrhea are the symptomatic hallmarks of the disease, whereas an elevated erythrocyte sedimentation rate is the only distinctive laboratory abnormality. Iatrogenic causes such as administration of broad-spectrum antibiotics and aural irrigation may play a predisposing role in high-risk populations. The disease can result in cranial polyneuropathies (with facial nerve [VII] paralysis being the most common) and death. The mainstay of treatment is administration of antipseudomonal antibiotics for four to eight weeks. Recurrence is common, and mortality remains at about 20 percent despite antibiotic therapy. Given the increasing longevity of diabetic patients, the frequency of this disease is increasing. Internists, family practitioners, and ambulatory care physicians must now be cognizant of the presenting symptoms, while infectious disease specialists and otolaryngologists need to be appraised of strides in diagnosis and therapy. The role of surgery should be minimized. Use of new diagnostic radiologic modalities and new antipseudomonal antibiotics discussed in this review should lead to improved outcome.
M
alignant external otitis is an often intractable infection caused by Pseudomonas aeruginosa, which occurs primarily in elderly diabetic subjects. The process frequently progresses to a skull base osteomyelitis with resultant cranial neuropathies. Despite long-term systemic antibiotic therapy, the disease is often recurrent and may be associated with significant mortality, Malignant external otitis was first described by Meltzer and Kelemen [I] in 1959 and the name was coined in Chandler’s [2] landmark 1968 article. However, an updated review is indicated since diagnostic, therapeutic, and prognostic aspects have changed. Specifically, new radiographic modalities and new antibiotic regimens have led to improvements in diagnosis and therapy. Theories regarding the pathogenesis of this disease, often conflicting, need to be critiqued based on current clinical and basic science concepts. The limited value of previously reported series of malignant external otitis is now more apparent. These limitations include a relatively small number of cases from any one institution, the retrospective nature of the studies, absence of defined time intervals in reporting patient follow-up, and omission of important clinical information when reporting individual cases (for example, type and duration of antibiotic therapy). Older reviews have presented conflicting views and inconsistencies on the basis for the association of diabetes mellitus and malignant external otitis, the precise etiologic role of P. aeruginosa versus other organisms, the prognostic significance of cranial nerve palsies, the exact role of radiography in the diagnosis and follow-up of patients, and the necessity for strict diabetic control. The incidence of malignant external otitis appears to be increasing, with 72 percent of cases reported after 1980; thus, the family practitioner, internist, and infectious disease specialist, as well as the otolaryngologist, should be familiar with the course and management of this complex disease. Finally, we present some new diagnostic and therapeutic recommendations for the clinician and the basis for these recommendations.
METHODS A computer-assisted search was performed primarily of the English language literature (one French, one Japanese, and one Portuguese article were also reviewed) on MEDLINE using the terms “malignant/ necrotizing external otitis.” Early sources were also identified from references in the more recent papers.
TERMINOLOGY The name of this disease is a source of continuing debate. Chandler [2] coined the term “malignant external otitis” in his initial description because of the high mortality of this syndrome. Evans and Richards September 1988 The American Journal of Medicine
Volume 85
391
MALIGNANT
EXTERNAL
OTITIS
/ RUBlN
AND YU
j
::
Figure 1. Anatomic sites of infection in malignant external otitis: the infection starts in the external ear canal, crosses the osseous-cartilaginous junction, and invades the temporal bone and adjoining structures. Cranial nerve involvement is directly related to its proximity to the path of infection. The cranial nerves affected, in order of frequency, are the facial nerve (VII) as it exits the stylomastoid foramen; the glossopharyngeal (IX), the vagus (X), and the accessory nerves (Xl) as they exit via the jugular foramen; and the hypoglossal nerve (XII) as it passes along the hypoglossal canal. Other sites of potential involvement include the temporomandibular joint, the meninges, sinuses, and parotid gland.
[3] introduced the term “necrotizing” in 1973, offering it as a more accurate description of an infectious process. Cohn [4] then added “progressive,” whereas Doroghazi et al [S] substituted “invasive” for “necrotizing.” It obviously would be less confusing for both the medical librarian and the physician if one name was consistently used. The objections raised to the term “malignant” are well taken. This is a bacterial infection and not a neoplastic process. In addition, despite its presumed beginnings as an infection of the auditory canal, the invasion of the organism into the temporal bone and skull base renders both “external” and “otitis” less accurate. In our opinion, none of the current nomenclature is ideal. In recognition of Chandler’s [2] contribution and because the majority of descriptions in the literature still employ the term malignant external otitis, we will retain this name throughout this article. Perhaps, as the understanding of the pathogenesis improves, a more accurate yet descriptive name can be formulated. And, more optimistically, as we learn to better control the course of this infection, terms such as “malignant,” “invasive,” “necrotizing,” and, “progressive” might be rendered obsolete.
BACTERIOLOGY The causative agent is virtually 392
September
1988
The American
always P. aeruginJournal
of Medicine
Volume
osa. It is unclear why there is such an overwhelming predilection for P. aeruginosa in this infection. This bacterium is an established commensal of the gastrointestinal tract and is rarely found in the external ear canal of healthy persons. Since the natural reservoir for this organism is water, accumulation of water and moisture is a likely inciting event. P. aeruginosa is also the most common etiologic agent of chronic external otitis. Although other organisms are often isolated concomitantly with Pseudomonas, they are generally habitants of normal skin such as diphtheroids, Staphylococcus aureus, S. epidermidis, and Candida albicans, among others. Isolated reports of other organisms causing malignant external otitis include two cases of Aspergillus fumigatus [6,7]; one case of S. aureus [S]; and one case of Proteus mirabilis [9]. One of these hosts was immunocompromised [6] and another was a chronically ill infant [9]. Only two of 260 (0.8 percent) of the cases of malignant external otitis in immunocompetent adults were not caused by P. aeruginosa. Therefore, stringent diagnostic criteria including bony destruction on computed tomographic scan and isolation of the organism from bone or mastoid should be met in any patient suspected of having malignant external otitis from microorganisms other than P. aeruginosa. 85
MALIGNANT
ANATOMY Spread of the pseudomonal infection in malignant external otitis follows a characteristic route that is reflected by the pattern of cranial neuropathies. The process is initiated in the external auditory canal and crosses the cartilaginous-osseous junction into the temporal bone. The infection then passes through the fissures of Santorini in the cartilaginous external meatus to the mastoid. The facial nerve becomes involved as it exits via the stylomastoid foramen (Figure 1). The infection can then spread along the base of the skull and may affect the glossopharyngeal, vagal, and spinal accessory nerves as they pass through the jugular foramen (the “jugular foramen syndrome”), as well as the hypoglossal nerve at the hypoglossal canal (Figure 1). Less often, the abducent and trigeminal nerves are involved at the petrous apex. There is one case in the literature of optic nerve involvement at the orbital apex in malignant external otitis [lo]. Because of its proximity to the initial site of infection, the facial nerve is almost always the first (and often only) cranial nerve affected, followed by the nerves exiting through the more distal jugular foramen and the hypoglossal canal. The trochlear and olfactory nerves are not known to be affected by this disease.
PATHOGENESIS The precise pathogenesis of malignant external otitis is unknown. Since the majority of cases occur in elderly diabetic patients, a theory that encompasses aging, diabetes mellitus, and an increased predilection for this bacterial infection would be ideal, Chronic hyperglycemia has been found in both an animal model and in in vitro studies to exert a depressive effect on host resistance, specifically polymorphonuclear phagocytic function [11,12]. In our experience, the occurrence of malignant external otitis does not necessarily coincide with periods of hyperglycemia. In fact, there are only two reports of diabetic patients with both malignant external otitis and diabetic ketoacidosis [13,14]. Furthermore, both of these cases occurred in adolescents with type I diabetes mellitus, whereas most cases of malignant external otitis arise in adultonset or type II diabetic patients. The increased serum glucose levels frequently noted at the time of presentation of malignant external otitis may reflect the infection exacerbating the diabetes. We believe hyperglycemia, per se, probably is not a factor in the pathogenesis of malignant external otitis. Aging and diabetes mellitus are both associated with altered immune function. The specific defects, however, are unclear. No generalized abnormality in cellular or humoral immunity appears to be present in patients with diabetes mellitus [X]. Investigations of phagocytic cell function in diabetic patients have provided variable results. The presence of a significant primary phagocytic dysfunction is difficult to reconcile with the adequate resistance mounted by diabetic patients against most bacterial organisms. In fact, there is surprisingly little evidence to suggest that diabetes mellitus is primarily associated with a general increased susceptibility to infections [16]. Diabetic microangiopathy is well described and was originally offered as a possible explanation for the predisposition of elderly diabetic patients to malignant external otitis [2]. Histopathologic studies of the capil-
EXTERNAL
OTITIS
/ RUBIN
AND YU
laries in the skin and subcutaneous tissue overlying the temporal bone have confirmed a thickening of periodic acid-Schiff-positive material in the subendothelial basement membrane [5]. Small vessel disease resulting in cutaneous hypoperfusion, diminished local host resistance, and, hence, increased susceptibility to infection is an attractive hypothesis for most patients with malignant external otitis. Moreover, blood vessel abnormalities are associated with diabetes and aging. There is a small but significant group of non-diabetic persons in whom malignant external otitis develops. The majority of pediatric cases of malignant external otitis do not have diabetes mellitus [17]. These children, however, do have evidence of compromised host resistance as manifest by malignancy and chemotherapy [14,18], malnutrition and anemia [9,19], neutrophil chemotactic disorders [20], and Stevens-Johnson syndrome with drug-induced leukopenia [21,22]. Therefore, the pathogenesis of malignant external otitis in the non-diabetic patient is probably related to immune dysfunction or a combination of impairments. This is in contrast to the predisposing vascular abnormalities in the patient with diabetes. Colonization of the external auditory canal by P. aeruginosa has not been shown to occur more frequently in diabetic versus non-diabetic patients [23]. However, we suspect that broad-spectrum oral antimicrobial chemotherapy so often used empirically for suspected ear infection may be the factor predisposing the external ear to P. aeruginosa colonization. For example, ampicillin, cephalexin, cefaclor, and trimethoprim-sulfamethoxazole have no activity against P. aeruginosa. Finally, an unrecognized but possible cause of malignant external otitis may be iatrogenic. Seven of nine patients with malignant external otitis recently seen by us had a history of ear irrigation within two weeks prior to the onset of symptoms. If nonsterile tap water (which may contain P. aeruginosa) is forcefully introduced into the ear of an elderly diabetic patient, we hypothesize that P. aeruginosa plus host factors produce an environment conducive to local invasion of the microbe followed by subsequent spread into bone and cartilage. Interestingly, a review of the literature disclosed other cases that developed following irrigation of the external ear [2,5,24-281.
CLINICAL FEATURES In an attempt to give an overview of the clinical aspects of malignant external otitis, data are summarized from the five largest series [2,5,29-401 (Table I). We caution that interpretation of this table should be tempered by the following limitations: nonuniform criteria for diagnosis, nonstandardized data collection, and missing information, For example, the presenting signs and symptoms are noted in only 31 percent (49 of 158) of the cases of these series and specific times of follow-up are even more rarely specified. The range of ages in the five studies is 11 to 91 years old, with the majority of patients over the age of 60. Ninety to 100 percent of the patients displayed some degree of glucose intolerance. These findings are consistent with the majority of reports describing malignant external otitis as a disease primarily of the old.er diabetic’patient. The specific diabetic status was only available for two of the series profiled [2,29,30,39,40] for a total of 69 patients. Of these individuals, 40 were
Seotember
1988
The American
Journal
of Medicine
Volume
85
393
MALIGNANT
EXTERNAL
OTITIS
/ RUBIN
AND YU
insulin-dependent, 25 were taking an oral hypoglycemic agent, 19 had their disease controlled with diet alone, and nine were diagnosed at presentation with malignant external otitis by a glucose tolerance test (chemical diabetes). As discussed previously, the development of malignant external otitis does not seem to be related to the severity of the diabetes or control of the hyperglycemia. In our experience, the disease developed in persons with only a positive glucose tolerance test as well as those with insulin-dependent diabetes of many years’ duration. It should be emphasized that the usual signs of infection are notably absent. For example, fever and tachycardia are distinctly unusual. The patients are usually non-toxic alert. Otalgia is the most common presenting symptom in the literature (noted 75 to 100 percent of the time). Although otalgia occurs in patients with external otitis and otitis media, the degree of pain is extraordinarily severe and unrelenting. The very intensity should suggest the diagnosis. One of our patients has described the excruciating pain as “starting from the top of my head and piercing through my skull down to my spine” and another as “a drilling pain that drives me to my knees.” Holder et al [lo] reported a case of suicide brought on by two years of unrelenting pain and infection. The pain tends to be nocturnal, frequently awakening the patient from sleep [30]. Most of our patients require analgesics before they can sleep. Headache (usually temporal or occipital) may be present and is remarkable too for its severity. We have also been impressed with the presence of classic temporomandibular joint pain with tenderness at the joint area and pain upon opening the mouth. Since the temporomandibular joint impinges onto the external ear canal (Figure l), it is surprising that this symptom has not received greater emphasis in previous clinical descriptions. In some patients, the pain is so severe that food intake is voluntarily limited. Tenderness and swelling about the periauricular area are common, especially if the patient has had symptoms for many months prior to seeing a physician. Purulent otorrhea is usually present at some point in the course of the illness (50 to 81 percent). The degree of drainage is variable, ranging from copious, foul-smelling, greenish exudate to minimal accumulation of moisture. Late in the course of the disease, the otorrhea appears to decrease. When parenteral antipseudomonal antibiotics are given, resolution of the drainage usually occurs within days. Granulation tissue is usually present in the external canal; histologic examination generally shows nonspecific inflammatory changes. Edema with erythema of the ear canal is common; however, the tympanic membrane is usually intact. Cranial nerve dysfunction can appear as early as one week after the onset of symptoms, although it usually occurs more than two months into the illness [35]. The incidence of facial nerve involvement in these series ranges from 24 to 43 percent, while other cranial neuropathies were found 14 to 35 percent of the time. It is noteworthy that the incidence of cranial neuropathies associated with malignant external otitis has not diminished since Chandler’s [2] first reported series in 1963. This finding implies a consistent delay in diagnosis. All too often these patients are only recognized 394
September
1988
The American
Journal
of Medicine
Volume
85
MALIGNANT EXTERNAL OTITIS / RUBIN AND YU
as having malignant external otitis after weeks to months of treatment for simple external otitis. The prognostic significance of cranial nerve (specifically facial) palsies has been debated. Although the mortality is greater in those with cranial neuropathies, it is probably not the nerve palsy that directly contributes to the patient’s demise but rather is a reflection of the advanced disease. The cranial nerve paralysis may not necessarily be permanent. For example, Chandler [31] reported that of those patients with facial nerve paralysis who survived, 22 percent experienced a complete recovery and 9 percent a partial recovery, whereas 17 percent sustained $ermanent paresis [31]. Recovery of function, however, is unpredictable on a case-by-case basis. There have been isolated case reports of other complications, including brain abscess [41-431, sphenoidal sinusitis [27,44-471, mycotic aneurysm [47], parotitis [41], lateral sinus thrombosis [3], and sigmoid sinus thrombosis [30]. These sequelae are noteworthy both in their seriousness and extreme rarity. All of the patients in the literature with meningitis, dural sinus thrombosis, or intracranial abscess died of their disease.
infectious process [56]. A mere 10 percent increase in osteogenic activity is sufficient for detection by bone scanning; this is the reason for its superiority to conventional radiography in demonstrating disease activity [56]. Although this modality is highly sensitive (reported to be 100 percent) [55], the specificity is low. The tracer collects at sites of osteogenic activity in response to both osteolytic and osteoblastic lesions. Thus, a “positive” bone scan may be seen in any inflammatory disease of the head and neck including carcinoma, temporomandibular joint disorders, fractures, and infections. In one study, radiologists who were not informed of the patients’ diagnoses were unable to differentiate scans of patients with malignant external otitis from those with severe external otitis [57]. Although Ostfeld et al [49] claimed that bone scan results correlated well with the clinical course of the disease, this finding has not been confirmed. We conclude that bone scanning with technetium 99 may be helpful in confirming the clinical diagnosis of malignant external otitis, but it may not differentiate this disease from simple external otitis or carcinoma (the leading alternate diagnoses), and is not particularly useful in monitoring the course of disease.
Numerous radiologic techniques have been used in the assessment of patients with malignant external otitis. These include plain films, polytomography, technetium bone scan, gallium scan, computed tomography, and magnetic resonance imaging. Transmission Radiography Most of the radiographic experience in the literature is based upon plain films and polytomography. The combination of skull films, mastoid series, and temporal bone tomograms may demonstrate soft tissue within the external auditory canal as well as increased density of the mastoid air cells with or without bone destruction. These studies are inadequate, however, in demonstrating infratemporal or intracranial extension of disease [38,48,49]. Some authors have suggested that the findings with transmission radiography correlate well with the clinical stage of malignant external otitis [36,48]; specifically, “early” and “late” disease may be differentiated on the basis of the presence or absence of bone destruction. We agree that bony erosion on plain films or tomograms certainly confirms that disease is advanced. However, since 30 to 50 percent demineralization of cancellous bone occurs before a lytic lesion can be visualized radiographically [50], we suggest that these techniques lack adequate sensitivity. In addition, any disease that produces a destructive lesion of the temporal bone (malignancy) can mimic the conventional radiographic findings in malignant external otitis [48]. Because of the limited sensitivity and specificity, plain films and tomographic scans of the skull and mastoid appear to be of minimal usefulness in the evaluation of this disease. Bone Scanning (Technetium 99) Bone scanning with technetium 99 is frequently used in confirming the clinical diagnosis of malignant external otitis [33,49,51-581. The radionuclide tracer accumulates at sites of osteoblastic activity presumably in response to bone destruction caused by the
Gallium Scanning Gallium citrate is incorporated into granulocytes and localizes in areas of inflammation in soft tissue and bone [59-611. The intense inflammatory activity in malignant external otitis in contrast to the minimal accumulation of gallium normally seen in the head region is what makes it useful in this disease [55]. The sensitivity of gallium 67 scanning in the diagnosis of malignant external otitis has ranged from 89 to 100 percent [33,53,56]. Like bone scanning, gallium scanning has the disadvantages of low specificity and imprecision in anatomic localization of disease. Its usefulness in clinical follow-up is unclear. Some investigators have suggested that findings from serial gallium scans can be helpful in predicting clinical resolution [33,51,54,55]. On the other hand, Gherini et al [52] have described two patients who had recurrence of disease despite a normal gallium study. We conclude that controlled studies correlating results of serial gallium scans and clinical course are necessary in order to determine its use in monitoring therapy for malignant external otitis. Computed Tomography (CT) Computed tomography is the current modality of choice for defining the anatomic extent of disease in malignant external otitis [33,38,62]. The ability of the CT scan to delineate fat and muscle planes makes it the ideal modality for viewing the subtemporal area [38]. The sensitivity of CT in diagnosing and following the course of the disease has not yet been established, although it has been reported to be as low as 30 percent at presentation in one report [56]. The majority of studies have found that clinical remission does not appear to coincide with normalization of the scan [34,38,40,52,56]. On the other hand, Gold et al [33] concluded that serial scans could aid in evaluating the effectiveness of therapy given the ability of the CT scan to demonstrate the resolution of soft tissue disease. Serial scans could provide objective criteria for terminating antibiotic therapy. Although this conclusion seems reasonable, confirmation in a prospective
SeWember
1988
The American
Journal
of Medicine
Volume
85
395
MALIGNANT
EXTERNAL
OTITIS
/ RUBIN
AND YU
study would be useful. Meaningful data gleaned from a CT scan might also include abnormalities of the external auditory canal, presence of fluid in the middle ear or mastoid, subtemporal involvement of cranial nerve VII, and spread of the infection across the midline [63]. Recurrence or progression of disease can be documented by sequential studies when a baseline study is obtained and results are correlated with clinical data [33,38,62]. A limitation of the CT scan is its inability to differentiate malignant external otitis from carcinoma. Two cases have been reported in the literature [29,64] in which malignant external otitis apparently coincided with or pre-dated squamous cell carcinoma. We recommend obtaining a scan at the time of initial presentation and repeating the study if the patient fails to show a response to antibiotic therapy, or if symptoms develop following completion of treatment. Magnetic Resonance Imaging Magnetic resonance imaging (MRI), like CT, is an anatomic imaging modality. Experience with this technique in malignant external otitis is limited. A single report comparing CT and MRI in defining the anatomic extent of disease concluded that the latter was superior in delineating soft-tissue involvement; two of four patients had evidence of contralateral disease by MRI that was not apparent on CT [52]. Both CT and MRI findings remained abnormal in these patients at three months and five months, respectively, after clinical resolution of disease. Comparison of the two modalities in more patients is needed before the relative value of each may be assessed.
LABORATORY PARAMETERS Hematologic and chemical laboratory parameters are generally unaffected by this condition. Surprisingly, even leukocytosis and the presence of immature forms in the differential count (“left shift”) are uncommon, reflecting the insidious but localized aspects of this infection. Erythrocyte Sedimentation Rate We have found the erythrocyte sedimentation rate (ESR) to be valuable in establishing the diagnosis and in monitoring treatment, yet it has not been widely employed in the diagnosis or follow-up of patients with malignant external otitis. Although it is a nonspecific indicator of disease, its sensitivity in malignant external otitis suggests its use in confirming the diagnosis in patients with pseudomonal otitis refractory to therapy. Even more importantly, it can serve as an objective standard for the status of infection in an individual patient. In our experience, the ESR was elevated in 25 of 25 (100 percent) of the cases of malignant external otitis (mean = 87 mm/hour). Moreover, the value was greater than 100 mm/hour in many patients, placing malignant external otitis in that group of diseases that can result in an ESR greater than 100 mm/hour. The differential diagnosis of an extremely elevated ESR includes malignancy, collagen vascular disease, and inflammatory bowel disease in addition to infections [65]. We note that these findings also support previous contentions that infections of bone may have particular propensity to produce markedly accelerated erythrocyte sedimentation [66].
396
September
1988
The American
Journal
of Medicine
Volume
Since carcinoma of the head and neck and simple external otitis (the two primary alternate diagnoses when considering malignant external otitis) rarely produce ESR elevations to the degree seen in malignant external otitis, this simple noninvasive test might be particularly useful in selected patients in whom malignant external otitis is suspected. In addition to supporting the diagnosis at presentation, the ESR is also a good indicator of disease activity [40]. The ESR will usually respond within two weeks of effective therapy, but it may be months before normal values are attained, especially in patients with prolonged infection prior to initiation of antibiotic therapy. We suggest that a baseline ESR value be obtained in all patients with a suspected diagnosis of malignant external otitis as an index of inflammatory disease; subsequent values can be used to monitor the response to therapy. Re-elevation of the ESR can be the first objective indication of recurrent disease. Glucose Tolerance Approximately 10 percent of the cases of malignant external otitis in the literature occur in non-diabetic patients, with the majority developing in the elderly diabetic host. Therefore, we suggest that glucose tolerance be evaluated in any patient with suspected malignant external otitis and no history of diabetes. If random and fasting serum glucose values are normal, a complete (usually five-hour) glucose tolerance test should be considered to rule out chemical diabetes.
THERAPY Meticulous control of the serum glucose level in diabetic patients is often recommended for those with malignant external otitis. Progression of the infection has never been correlated with severity of hyperglycemia; nevertheless, we agree it is prudent to maintain euglycemia in the diabetic patient with malignant external otitis. The role of surgery in the treatment of malignant external otitis has changed since Chandler’s [2] initial report in 1968 in which all 13 of his patients underwent at least one major surgical procedure in attempting to control the infection. Nine to 100 percent of the patients in these series underwent some form of surgery (usually mastoidectomy, facial nerve decompression, polypectomy, or a deep tissue debridement). Facial nerve decompression is often performed, but in our experience, this procedure has not contributed to the recovery of function. Subtotal temporal bone resection for the drainage of deep-seated cranial abscesses has been advocated in selected patients whose condition is unresponsive to aggressive medical therapy and surgical debridement [67]. We now believe that the role of surgery in malignant external otitis should generally be limited to local debridement and the excision of accessible foci of infection such as polypectomy or sequestrum removal. We believe topical antibiotics, although commonly used in external otitis, have no role in the treatment of malignant external otitis; local instillation of antibacterial agents is insufficient for this invasive pseudomonal infection and only hinders culture isolation of the bacterium from ear drainage. The hyperbaric oxygen chamber has been used with varying success in anecdotal reports [32,53,68,69].
85
MALIGNANT
However, based on the known aerobicity of Pseudomonas and the small number of cases treated, this modality must still be considered experimental. With the advent of semi-synthetic penicillins, specifically carbenicillin in 1969, systemic antipseudomonal antibiotics have emerged as the mainstay of therapy. This transition from primarily surgical to medical management resulted in a decreased mortality; Chandler [29] quoted a mortality rate decrease from 46 percent to 23 percent. However, it is important to note that the death rate has remained essentially unchanged in 20 years despite the introduction of the antipseudomonal penicillins and aminoglycosides. Currently, accepted therapy is the long-term administration of an aminoglycoside (e.g., tobramytin) with a beta-lactam agent (e.g., piperacillin or azlocillin). The optimal duration of treatment has not been established, but most authors recommend at least four to eight weeks of intravenous therapy. Third-generation cephalosporins, notably moxalactam and cefsulodin, have been reported to be successful in well-described anecdotal reports [34,39,68,70]. However, it appears that recurrence occurs as commonly with these newer agents as with previous antipseudomonal beta-lactam therapy. Other antipseudomonal agents that may be considered in selected situations in which a penicillin or aminoglycoside would be hazardous include aztreonam, ceftazidime, and imipenem. A recent study at our institution has demonstrated efficacy in the treatment of 11 patients with malignant external otitis using an oral regimen of ciprofloxacin, a new quinolone agent, and rifampin. At a mean followup of 18 months, there have been no deaths from disease [40]. We anticipate this regimen will be a significant advance in this difficult-to-treat infection.
PROGNOSIS Mental status deterioration is a poor prognostic sign, generally reflecting intracranial complications from the disease. Increasing mortality is associated with cranial nerve involvement, with the highest mortality associated with cranial polyneuropathies. Recurrence is a well-recognized aspect of this disease (9 to 27 percent) and is usually first manifest by the insidious onset of otalgia followed by headache and temporomandibular joint pain. Surprisingly, otorrhea may be minimal or not reappear at all, reflecting the dominance of skull base osteomyelitis in causation of the clinical syndrome. The first objective sign is usually the re-elevation of the ESR. The issue of when to consider an infection as having been cured has practical implications for patient follow-up. Our experience and a review of the literature indicate that infection can recur as long as four to 12 months after the termination of antibiotic therapy. Outcome and follow-up are inconsistently reported in the cases described in the literature. We suggest, therefore, that a period of one year elapse after the discontinuation of antibiotics before a patient’s infection is reported as cured.
SUMMARY Malignant external otitis is a Pseudomonas infection that begins in the external auditory meatus and not infrequently spreads to involve the temporal bone,
EXTERNAL
OTITIS
/ RUBIN
AND YU
zygomatic root, and the base of the skull. Small blood vessel abnormalities, associated with both aging and diabetes, probably contribute to the tendency for infection by disrupting nutrient supply to cutaneous tissues. In addition, iatrogenic factors such as the empiric institution of broad-spectrum antibiotics and aural irrigation may play a predisposing role. The diagnosis of malignant external otitis is supported by unrelenting otalgia frequently accompanied by purulent otorrhea in elderly diabetic patients. The ESR is usually markedly elevated and radiographic studies, especially CT scan, may show evidence of bone destruction. The proclivity towards cranial neuropathies, most commonly facial nerve paralysis, and a 20 percent mortality rate necessitates early diagnosis and prompt institution of appropriate therapy. Currently, the mainstay of therapy is long-term (four to eight weeks) intravenous administration of combination antipseudomonal antibiotics in conjunction with local debridement of infected tissue. The quinolone antibiotics, particularly ciprofloxacin, will likely prove to be of major usefulness.
ACKNOWLEDGMENT We thank Drs. Ralph Caprosa, Donald Kamerer, Eugene Meyers, and J.R. Chandler for their critical review,‘Dr. Nicholas Cauna for assistance in the design of Figure 1, Mr. Ron Filer for the artwork for Figure 1, and Ms. Shirley Brinker for manuscript preparation.
REFERENCES 1. Meltzer PE, Kelemen G: Pyocutaneous osteomyelitis of the temporal bone, mandible and zygoma. Laryngoscope 1959; 169: 1300-1316. 2. Chandler JR: Malignant external otitis. Laryngoscope 1968; 78: 1257-1294. 3. Evans ITG. Richards SH: Malignant (necrotizing) otitis externa. J Otolaryngol Otol 1973; 87: 13-20. 4. Cohn AM: Proeressive necrotizine otitis. Arch Otolarvnnol 1974; 99: 136-139. 5: Doroghazi RM: Nadol JB, Hyslop NE, Baker AS, Axelrod L: Invasive external otitis. Report of 21 cases and review of the literature. Am J Med 1981; 71: 603614. 6. Petrak RM, Pottage JC, Levin S: Invasive external otitis caused by Aspergilus fumigatus in an immunocompromised patient. J Infect Dis 1985; 151: 196. 7. Cunningham MJ, Yu VL, Turner J, Curtin HD: Necrotizing otitis externa due to Aspergillus in an immunocompetent patrent. Arch Otolaryngol Head Neck Surg 1988; 114: 554-556. 8. Bayardelle P, Jolivet-Granger M, Larochelle D: Staphylococcal malignant external otitis. Can Med Assoc J 1982; 126: 155-156. 9. Coser PL, Stamm AEC, Lobo RC. Pinto JA: Malignant external otitis in infants. Laryngoscope 1980; 90: 312-315. 10. Holder CD, Gurucharri M, Barteis LJ, Colman MF: Malignant external otitiswith optic neuritis. Laryngoscope 1986; 96: 1021-1023. 11. Drachman RH, Root RK, Wood WB: Studies on the effect of experimental nonketotic diabetes mellitus on antibacterial defense. J Exp Med 1966; 124: 227246. 12. Bagade JD, Nielson K, Root J, Bulger R: Host defense in diabetes mellitus: the feckless phagocvte during poor control and ketoacidosis. Diabetes 1970; 19: 364. 13. Merritt WT, Bass JW,Bruhn FW: Malignant external otitis in an adolescent with diabetes. J Pediatr 1980; 96: 872-873. 14. Sobie S, Brodsky L, Stanievich JF: Necrotizing external otibs in children: report of two cases and review of the literature. Laryngoscope 1987; 97: 598-601. 15. Galbraith RM: Immunological aspects of diabetes mellitus. Boca Raton, Florida: CRC Press, 1979. 16. Miller ME: Phagocyticcellfunctions in diabetes mellitus. In: Gupta S, ed. Immunology of clinical and experimental diabetes. New York: Plenum Medical, 1986; 369-383. 17. Rubin J. Yu VL. Stool S: Malignant external otitis in children: a critical review. J Pediatr (in press). 18. Rubenstein JM. Ostfeld E. Ben-Zarav S. Schiby G: Necrotizing external otitis. Pediatrics 1980; 66: 618-620. 19. Joachims HZ: Malignant external otitis in children. Arch Otolaryngol1976; 102: 236-237. 20. lchimura K, Hoshino T, Yano J, Nozue M: Neutrophil disorders in a child wrth necrotizing external otitis. J Otolaryngol 1983; 12: 129-133. 21. Sherman P, Black S, Grossman M: Malignant external otitis due to Pseudomo nas aeruginosa in childhood. Pediatrics 1980; 66: 782-783. 22. Horn KL, Gherini S: Malignant external otitis of childhood. Am J Otol 1981; 2:
September
1988
The American
Journal
of Medicine
Volume
85
397
MALIGNANT
EXTERNAL
OTITIS
/ RUBIN
AND YU
402-404. 23. Salit IE, Miller B, Wigmore W. Smith JA: Bacterial flora of the external canal in diabetics and non-diabetics. Laryngoscope 1982; 92: 672-673. 24. Dawson DA: Malignant external otitis. J Laryngol Otol 1978; 92: 803-810. 25. Dinapoli RP. Thomas JE: Neurologic aspects of malignant external otitis: report of three cases. Mayo Clin Proc 1971; 46: 339-344. 26. John AC, Hopkin NB: An unusual case of necrotizing otitis externa. J Otolaryngol Otol 1978; 92: 259-264. 27. Schwarz GA, Blumenkrantz MJ, Sundmaker WLH: Neurologic complications of malignant external otitis. Neurology 1971; 21: 1077-1084. 28. Wilson DF, Pulec JL, Linthicum FH: Malignant external otitis. Arch Otolaryngol 1971q: 419-422. 29. Chandler JR: Pathogenesis and treatment of facial paralysis due to malignant external otitis. Ann Otol 1972; 81: 648-658. 30. Chandler JR: Malignant external otitis and facial paralysis. Otolaryngol Clin North Am 1974; 7: 375-383. 31. Chandler JR: Malignant external otitis. Further consideration Ann Otol 1977; 86: 417-428. 32. Lucente FE, Parisier SC, Som PM: Complicationsof the treatment of malignant external otitis. Laryngoscope 1983; 93: 279-281. 33. Gold S, Som PM, Lawson W, Lucente FE, Mendelson M, Parisier SC: Radiographic findings in progressive necrotizing “malignant” external otitis. Laryngoscope 1984: 94: 363-366. 34. Mendelson MH, Meyers BR, Hirschman SZ, Shapiro ER, Parisier SC: Treatment of invasive external otitis with cefsulodin. Rev Infect Dis 1984; 6: S698-S704. 35. Corey JP, Levandowski RA, Panwalker AP: Prognostic implications of therapy for necrotizing external otitis. Am J Otol 1985; 6: 353-358. 36. Kim BH: Roentgenographic findings of malignant external otitis. AJR 1971; 112: 366-372. 37. Sando I, Harada T, Saito R, Okano Y, Caparosa RH: Temporal bone histopathology of necrotizing external otitis. Ann Otol 1981; 90: 109-115. 38. Curtin HD, Wolfe P, May M: Malignant external otitis: CTevaluation. Radiology 1982: 145: 383-388. 39. Haverkos HW, Caparosa R, Yu VL, Kamerer D: Moxolactam therapy-its use in chronic suppurative otitis media and malignant external otitis. Arch Otolaryngol 1982: 108: 329-333. 40. Yu VL, Stoehr G, Rubin J, Matador A, Kamerer D: Efficacy of oral ciprofloxacin plus rifampin for therapy of malignant external otitis (abstr 188). In: Proceedings of the 27th Interscience Conference on Antimicrobial Agents and Chemotherapy. New York: American Society of Microbiology, 1987. 41. Caruso VF, Griffiths CM, Bailey BJ: Malignant external otitis. Tex Med 1977: 73: 106-113. 42. Kohut RI, Lindsay JR: Necrotizing(“malignant”) external otitis histopathologic processes. Ann Otol 1979; 88: 714-720. 43. Soliman AE: A rare case of malignant external otitis externa in a non-diabetic patient. J Laryngol Otol 1978; 92: 811-812. 44. Youngs R, Bagley J: Sphenoidal sinusitis secondary to malignant external otitis. J Laryngol Otol 1986; 100: 341-344. 45. Uri N, Kitzes R, Meyer W, Schuchman G: Necrotizing external otitis: the importance of prolonged drug therapy. J Otolaryngol Otol 1984; 98: 1083-1085. 46. Pace-Balzan J: Necrotizing otitis externa. J Laryngol Otol 1977; 91: 735738. 47. Watson RC: The neuropathology of malignant external otitis. Bull Los Angeles Neural Sot 1977; 42: 51-56.
398
September
1988
The American
Journal
of Medicine
Volume
48. Mendez G, Quencer RM, Donovan Post MJ. Stokes NA: Malignant external otitis: a radiographic-clinical correlation. AJR 1979; 132: 957-961. 49. Ostfeld E, Aviel A, Pelet D: Malignant external otitis: the diagnostic value of bone scintigraphy. Laryngoscope 1981; 91: 960-964. 50. Jones AG, Francis MD, Davis MA: Bone scanning: radionuclide reaction mechanisms. Semin Nucl Med 1976; 6: 3-18. 51.Garty I, Rosen G, Holdstein Y: The radionuclide diagnosis, evaluation and follow-up of malignant otitis externa (MOE)-the value of immediate blood pool scanning. J Laryngol Otol 1985; 99: 109-115. 52. Gherini SG. Brackmann DE, Bradley WG: Magnetic resonance imaging and computerized tomography in malignant external otitis. Laryngoscope 1986; 96: 542-548. 53. Mendelson DS, Som PM, Mendelson MH, Parisier SC: Malignant external otitis: the role of computed tomography and radionuclides in evaluation. Radiology 1983; 149: 745-749. 54. Parisier SC, Lucente FE, Hirschman SZ, Som PM, Arnold LM, Ruffman JD: Nuclear scanning in necrotizing progressive “malignant” external otitis. Laryngoscope 1982; 92: 1016-1020. 55. Reiter D, Bilaniuk LT, Zimmerman RA: Diagnostic imaging in malignant otitis externa. Otolaryngol Head Neck Surg 1982; 90: 606-609. 56. Strashun AM, Nejatheim M, Goldsmith SJ: Malignant external otitisearly scintigraph detection. Radiology 1984; 150: 541-545. 57. Noyek AM: Bone scanning in otolaryngology. Laryngoscope 1979; 89(suppl 18): l-29. 58. Levin WJ, Shary JH, Nichols LT, Lucente FE: Bone scanning in severe external otitis. Laryngoscope 1986; 96: 1193-1195. 59. Tsan M, Chen WY, Scheffel V. Wagner HN: Studies on gallium accumulation in inflammatory lesions: gallium uptake by human polymorphonuclear leukocytes. J Nucl Med 1978; 19: 36-43. 60. Menon S, Wagner HN, Tsan MF: Studies on gallium accumulation in inflammatory lesions: uptake by Staphylococcus aureus: concise communication. J Nucl Med 1978; 19: 44-47. 61. Silva J, Harvey WC: Detection of infections with gallium-67 and scintigraphic imaging. J Infect Dis 1974; 130: 125-131. 62. Chakeres DW, Kapila A, LaMasters D: Soft-tissue abnormalities of the external auditory canal: subject review of CT findings. Radiology 1985; 156: 105-109. 63. Rubin J, Curtin HD, Yu VL, Hirsch BE: The utility of serial CT scans in the diagnosis and follow-up of-malignant external otitis. American Academy of Otolaryngology-Head and Neck Surgery, 1988; Washington, D.C. 64. Matucci KF. Setzen M, Galantich P: Necrotizing otitis externa occurring concurrently with epidermoid carcinoma. Laryngoscope 1986; 96: 264-266. 65. Zacharski LR, Kyle RA: Significance of extreme elevation of erythrocyte sedimentation rate. JAMA 1967; 202: 116-l 18. 66. Wyler DJ: Diagnostic implications of markedly elevated erythrocyte sedimentation rate: a reevaluation. South Med J 1977; 70: 1428-1430. 67. Raines JM, Schindler RA: The surgical management of recalcitrant malignant external otitis. Laryngoscope 1980; 90: 606-609. 68. Mader JR, Love JT: Malignant external otitis-cure with adjunctive hyperbaric oxygen therapy. Arch Otolaryngol 1982; 108: 38-40. 69. Lucente FE, Parisier SC, Som PM, Arnold LM: Malignant external otitis: a dangerous misnomer? Otolaryngol Head Neck Surg 1982; 90: 266-269. 70. Meyers BR, Mendelson MH, Parisier SC, Hirschman SZ: Malignant external otitis: comparison of monotherapy vs. combination therapy. Arch Otolaryngol Head Neck Surg 1987; 113: 974-978.
85