Otology & Neurotology 35:371Y376 Ó 2014, Otology & Neurotology, Inc.

Clinical Outcome Parameters for Necrotizing Otitis Externa *Ayzegu¨l Verim, *BarNz Naibo?lu, *Çi?dem Tepe Karaca, *Lu¨tfu¨ yeneldir, †Semra Ku¨lek0i, and *Ça?atay Oysu *Department of Otorhinolaryngology/Head and Neck Surgery, Haydarpaza Numune Educational and Research Hospital, Istanbul; and ÞDepartment of Otorhinolaryngology/Head and Neck Surgery, Ku¨tahya Educational and Research Hospital, Ku¨tahya, Turkey

Objective: To investigate the duration of time elapsed between the onset of symptoms for necrotizing external otitis (NEO) and admission to hospital that may play a role in patient outcome. Study Design: Retrospective case review. Setting: Tertiary referral center. Patients: Fourteen consecutive male patients with NEO with no improvement from the previous course of antibiotherapy and with findings of osteomyelitis on temporal bone CT, MRI, and positive detection of Tc-99m methylene diphosphonate on temporal bone, admitted as inpatients between 2008 and 2012. Intervention(s): Medical treatment of NEO and surgical debridement. Main Outcome Measure(s): Patients were divided into 2 groups according to median time elapsed between onset of symptoms and hospitalization (G30 d or 930 d). HbA1c, fasting blood sugar, erythrocyte sedimentation rate, C-reactive protein, pain intensity, radiologic grade, improvement since diagnosis, and total time to cure were compared according to the groups. The

relationships between the laboratory data were analyzed to determine the parameters associated with time to recovery. Results: Otalgia was significantly worse in patients who were admitted to hospital greater than 30 days after symptom onset (Mann-Whitney U test, p G 0.002). Blood glucose increased related to delayed admission time ( p G 0.001). CRP results were independently elevated from the admission time ( p G 0.112). There was a statistically significant difference between groups according to ESR levels and recovery time (Mann-Whitney U test, p G 0.004 and p G 0.01). There was a positive correlation between HbA1c levels and recovery time in Group 1 and between ESR levels and recovery time in Group 2 (r = 0.872, p = 0.044; r = 0.630, p = 0.039). Conclusion: Clinical, laboratory, and outcome data worsen later than 30 days in NEO. Key Words: DiabetesVNecrotizing external otitisVOsteomyelitisVSkull baseVTc-99m bone scintigraphyV Temporal bone. Otol Neurotol 35:371Y376, 2014.

Necrotizing external otitis (NEO) was first described in 1959 as a progressive osteomyelitis of the temporal bone caused by Pseudomonas aeruginosa in an elderly diabetic (1). Also referred to as malignant otitis externa, the disease is a serious infection of the external auditory canal, which may spread to the skull base, resulting in a more life-threatening complication: central skull base osteomyelitis. NEO presents with deep seated otalgia refractory to analgesics, most commonly affects elderly patients, 90% to 100% of subjects are diabetic (2Y4). Although uncommon, children with IgG and IgA deficiency, acute monocytic leukemia, or other immunocompromised conditions may also be affected.

Pseudomonas aeruginosa, not normally present in the external acoustic canal flora, is the pathogen usually responsible for the disease. This organism commonly causes benign otitis externa after water irrigation or trauma to the skin of the external auditory canal (EAC). In cases of impaired phagocytosis, or endarteritis due to immunodeficiency or diabetes mellitus, the infection presents a more devastating progress (5). The infection initiates as an inflammation in the dermal and epidermal tissue at the junction of the cartilaginous and osseous EAC and spreads through the Santorini fissures, medially to the tympanomastoid fissure, petrous apex, and skull base bone; inferiorly to the stylomastoid foramen, infratemporal fossa, and neck; posteriorly to the mastoid bone and sigmoid sinus; and anteriorly to the parotid gland and surrounding tissues. Treatment is based on effective management of diabetes, biopsy, and culture guided long-term antibiotherapy. Oral or intravenous ciprofloxacin with antibiotic eardrops

Address correspondence and reprint requests to Ayzegu¨l Verim, M.D., Department of Otorhinolaryngology/Head and Neck Surgery, Haydarpaza Numune Educational and Research Hospital, SelimiyeMh. 34668 ¨ sku¨dar, Istanbul, Turkey; E-mail: aysegulverim@ hotmail.com U The authors disclose no conflicts of interest.

371

Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

372

A. VERIM ET AL.

and/or third-generation cephalosporins and/or antipseudomonal penicillins are the preferred antibiotic regimen. In cases not responding to antibiotherapy, the infection progresses, leading to central skull base osteomyelitis, involvement of skull foramina with cranial neuropathies (5th, 6th, 7th, 9th, 10th, 11th cranial nerves), meningitis, cerebritis, intracranial empyema, or venous sinus thrombosis (6). In recent years, the mortality rate from NEO has been reduced from 50% to about 10% with the introduction of new generation antibiotics. However, the disease still poses a challenge to otolaryngologists because of its unpredictable prognosis. With this in mind, we reviewed the records of patients with NEO admitted to our clinic from 2008 to 2012 to investigate the time elapsed between the onset of symptoms and admission to hospital that may play a role in patient outcome and the relationship between laboratory, radiologic, and clinical data and clinical outcome. MATERIALS AND METHODS This study is a retrospective chart review of patients hospitalized with NEO in the otorhinolaryngology clinic of a tertiary referral center between 2008 and 2012. Approval was obtained from the institutional local ethics committee with the serial number HNEAH-KAEK 2012/61. Inclusion criteria were as follows: 1) patients with persistent otalgia, otorrhea, and hearing loss with no improvement after the previous course of antibiotherapy, admitted as inpatients with NEO; 2) severe edema of the EAC, exudate with or without granulation tissue; and 3) findings of osteomyelitis on temporal bone CT, MRI, and positive detection of technetium Tc-99m methylene diphosphonate (MDP) on temporal bone. Data were collected from the anamnesis, clinical presentation, physical examination, laboratory, and imaging findings. A correlation analysis of the time elapsed from the onset of symptoms to hospitalization with laboratory, clinical, and outcome data was performed before deciding on parameters to be compared. The positive correlation observed between the time to admission and the study data prompted us to investigate whether median time to admission would be a factor in clinical laboratory and radiologic outcomes (Table 1). Consequently, patients were divided into 2 groups based on the median time from onset of symptoms to hospitalization. The acute NEO group (Group 1) was composed of those patients admitted earlier than the median time to hospitalization (30 d). Those who were admitted later than the median time to hospitalization were included in the chronic NEO group (Group 2).

Parameters 1. The time interval from symptom onset to hospitalization. 2. A visual analogue scale (VAS) was used to evaluate persistent, severe otalgia; 10 points for the worst imaginable pain and 0 points for no pain. 3. Edema, otorrhea, and/or granulation tissue in the EAC and cranial nerve involvement. 4. Erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), HbA1c, and blood glucose levels at admission. 5. Contrast-enhanced CT scan, MRI, and triphasic bone scintigraphy with intravenous injection of Tc-99mMDP, followed by single photon emission computed tomography/ CT (SPECT/CT).

TABLE 1. Correlation of the data with the elapsed time between onset of symptoms and admission to hospital Time elapsed between symptom onset and admission to hospital n = 14 HbA1c Fasting blood sugar ESR at admission CRP at admission Otalgia VAS Improvement start time Improvement time Total cure time

r

p

0.565 0.671 0.675 0.339 0.823 0.587 0.623 0.666

0.035a 0.009b 0.008b 0.236 0.001b 0.027a 0.017a 0.009b

a

p G 0.05. p G 0.01.

b

Temporal bone images were classified into 4 categories according to the grading system defined by Chin et al. (7): Grade I: disease limited to soft tissue and not involving bone; refractory to standard antibiotic therapy for greater than 1 week; Grade II: earliest form of NEO with bone involvement limited to the mastoid; Grade III: NEO extending medially to involve the petrous temporal bone; and Grade IV: NEO extending medially to involve the petrous apex or with cranial nerve involvement or spreading anteriorly to involve the facial bones, posteriorly to involve the occipital bone, or spreading to the contralateral base of the skull.

Treatment Swabs from the EAC discharge were taken for culture and antibiotic sensitivity testing before starting treatment. Then, all patients received 750 mg of ciprofloxacin administered by slow infusion over 90 minutes every 12 hours. Blood glucose levels and renal function were monitored by the Department of Internal Medicine. Daily aspiration of the EAC with topical application of quinolone eardrops and debridement of granulation tissue and necrosed bone were performed in addition to systemic therapy. All debris collected were sent for histopathologic analysis. On Day 5, patients whose pain intensity and symptoms did not show any improvement underwent hyperbaric oxygen therapy (HBOT) 5 days a week, a total of 20 times, and an additional antibiotic regimen besides ciprofloxacin, according to culture results and advice from the Infectious Disease Department. Ciprofloxacin was combined with ceftazidime 2 g IV every 8 hours or 4 g piperacillin/0.5 g tazobactam IV every 8 hours. With a reduction of 50% in VAS scores, CRP, and ESR levels; resolution of otorrhea; and decrease in EAC edema from severe to mild, combined antibiotic therapy was modified to single antibiotherapy (ciprofloxacin 750 mg administered orally every 24 h). Treatment was stopped when VAS scores had decreased to 0, CRP levels to 0.0 to 0.8 mg/dl, and ESR to 0 to 30 mm/h for women and 0 to 20 mm/h for men. If present, disappearance of dural enhancement, medullary bone involvement, and EAC skin edema on MRI and temporal bone CT were defined as recovery indicators on images. Patients were all followed up to the present time.

Statistical Analysis NCSS (Number Cruncher Statistical System) 2007&PASS 2008 Statistical Software (UT, USA) was used for statistical analysis. Descriptive (mean, standard deviation [SD], frequency, and median) and quantitative statistical methods were used in the

Otology & Neurotology, Vol. 35, No. 2, 2014

Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

NECROTIZING OTITIS EXTERNA OUTCOME PARAMETERS

FIG. 1. Image of the bare bone after debridement of necrotic debris from the EAC.

evaluation of the study data. For quantitative analyses, the MannWhitney U test was used to compare the 2 groups for parameters that did not have a normal distribution. Qualitative data were compared using Fisher’s exact test. Spearman’s correlation coefficient was performed to examine the underlying relationship between recovery time and HbA1c, ESR, and CRP levels, and VAS scores and grade of temporal bone involvement of the groups. Results were evaluated with a confidence interval of 95% and with p G 0.05 indicating a statistically significant difference.

RESULTS Patients History and Clinical Signs Fifteen male patients were identified from the file review of patients hospitalized for NEO between 2008 and 2012. One patient, who was subsequently diagnosed on the fifth histopathologic analysis as NEO associated with EAC carcinoma, was excluded from the study. Fourteen men with an age range of 58 to 79 years (mean age, TABLE 2.

373

69.57 T 7.03 yr) comprised the study group. All patients (100%) had diabetes. Atherosclerotic vascular disease was the only associated comorbidity in 9 (64%) patients. Persistent otalgia that worsened at night was the major symptom in all patients besides exudate (9 patients, 64%) and EAC edema (14 patients, 100%). Granulation tissue on the floor of the EAC at the bony-cartilaginous junction was present in 6 cases (42%). Exposed bony wall was seen in only 1 patient (Fig. 1). In 3 patients (21%), facial nerve paralysis ensued during the course of treatment. Average duration of symptoms in all patients before hospitalization was 33.64 T 11.71 days (mean T SD), with a range of 21 to 60 days (median, 30 d). Mean and median time to hospitalization of the groups are detailed in Table 2. Otalgia severity scores on VAS at time of hospitalization were 8.00 T 0.71and 9.67 T 0.50 for Groups 1 and 2, respectively (Table 2). Otalgia was significantly worse in patients who were admitted to hospital later than 30 days after symptom onset (Mann-Whitney U test, p G 0.002). Cultures and Laboratory Findings Nine (64%) of the patients had positive culture results for P. aeruginosa and one (7%) for S. aureus. Culture results were negative in 4 (29%) of the patients. Blood glucose levels, HbA1c levels, and inflammatory markers (CRP and ESR) were all elevated above reference values. Blood glucose levels were significantly higher in patients admitted later than 30 days (Group 2) after symptom onset (Mann-Whitney U test, p G 0.001). The mean HbA1c ratio of Group 2 was higher than Group 1; however, there was no statistically significant difference between the groups with regard to HbA1c levels. They were elevated independent of the median time to admission. CRP values of patients admitted to hospital later than 30 days seemed to be higher than in patients admitted earlier. However, the difference between these groups was not statistically significant. CRP results were elevated, independent of the time to hospitalization ( p G 0.112). ESR levels were elevated in all patients, and the difference between the groups was statistically significant

Mean time to hospitalization, otalgia severity scores on VAS, laboratory, and outcome data of the groups Group 1

Group 2

(n = 5)

Mean time to admission (d) Otalgia severity score (VAS) Mean blood glucose at hospitalization (mg/dl) HbA1c levels (%) CRP levels at hospitalization (mg/L) ESR at hospitalization (mm/h) Start of improvement (d) Time to improvement (d) Time to complete recovery (d)

P p G 0.01

(n = 9)

Mean T SD

Median

Mean T SD

Median

22.20 T 1.79 8.00 T 0.71 129.60 T 11.28 7.80 T 1.66 6.88 T 1.61 85.20 T 6.38 6.00 T 1.00 23.80 T 2.39 61.60 T 4.33

21 8.0 135 7.9 6.1 85 6.0 24.0 63.0

40.0 T 9.69 9.67 T 0.50 171.11 T 23.88 9.75 T 1.82 9.80 T 3.46 107.11 T 12.46 27.11 T 5.18 41.78 T 11.84 126.33 T 23.9

36 10.0 170 10.1 9.2 110 28.0 38.0 130.0

0.002a 0.001a 0,083 0,112 0,004a 0.001a 0.01a 0.01a

MannYWhitney U test. a p G 0.01. Otology & Neurotology, Vol. 35, No. 2, 2014

Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

374

A. VERIM ET AL.

(Mann-Whitney U test, p G 0.004). ESR was an inflammatory marker whose elevated levels were related to the duration of NEO (Table 2). Radiologic Findings Tc-99m MDP SPECT/CT indicated the presence of osteomyelitis of the temporal bone in all patients. MRI revealed temporal bone marrow involvement in all and dural enhancement in 2 patients (14%). Temporal bone involvement was assessed according to the radiologic grading system defined by Chin et al. (8). The distribution of grading in images is detailed in Table 3. Two patients with dural enhancement were rated as Grade 3. There was no statistically significant difference between groups according to the radiologic grading system (Fisher’s exact test). Bony involvement was limited to the mastoid bone or extended to the petrous part of the temporal bone in patients with mean duration of symptoms of 40 T 9.69 days (Table 3). Clinical Outcome The duration until pain severity, EAC edema, and exudate started to decrease was 6.00 T 1.00 days in Group 1 and 27.11 T 5.18 days in Group 2 (Table 2). There was a statistically significant difference between groups according to the time to decline of symptoms (Mann-Whitney U test, p G 0.001). Patients admitted earlier than 30 days responded earlier than the patients admitted later than 30 days after symptom onset. Improvement was defined as a reduction of 50% in VAS scores, a significant decrease in CRP and ESR levels, resolution of otorrhea, and decrease in EAC edema from severe to mild. There was a statistically significant difference between groups according to time to improvement (Mann-Whitney U test, p G 0.01). Patients admitted earlier than 30 days improved earlier than the patients admitted later than 30 days after symptom onset (Table 2). Elimination of all symptoms and return of the inflammatory markers to normal limits were accepted as complete recovery. There was a statistically significant difference between groups according to recovery time. Patients whose treatment started within 1 month after symptom onset recovered earlier than those whose treatment started later than 1 month after symptom onset (Table 2). Parameters were compared in terms of time to recovery to analyze the relationship between laboratory tests, radiologic grading, VAS scores, and disease prognosis. Correlation between HbA1c, ESR, CRP, VAS scores, and recovery time in the groups is shown in Table 3. Spearman TABLE 3. Radiologic grade

Total recovery time Group 1 (n = 5) HbA1c at hospitalization Fasting blood sugar at hospitalization ESR at hospitalization CRP at hospitalization VAS at hospitalization Time to improvement

a

Group 2 (n = 9)

0.872 0.154

0.508 0.529

Y0.158 Y0.205 0.344 0.872a

0.630a 0.723a Y0.092 0.861b

ESR indicates erythrocyte sedimentation rate; CRP, C-reactive protein; VAS, visual analogue scale. a p G 0.05. b p G 0.01.

correlation analyses to determine the relationship between data (HbA1c, ESR, CRP, and VAS scores) and recovery time revealed the following: a) There was a positive correlation between HbA1c levels (mean, 7.80 T 1.66%) and recovery time in Group 1 (r = 0.872, p = 0.044). High levels of HbA1c were indicators of late recovery in patients with symptom onset earlier than 1 month. b) There was a positive correlation between ESR levels (mean, 107.11 T 12.46 mm/h) and recovery time in Group 2 (r = 0.630, p = 0.039). High ESR levels were indicators of late recovery in patients with symptom onset later than 1 month. c) There was a positive correlation between CRP levels (mean, 9.80 T 3.46 mg/L) and recovery time in Group 2 (r = 0.723, p = 0.028). High CRP levels were indicator of late recovery in patients with symptom onset later than 1 month. As for pain severity, there was no correlation between VAS scores and recovery time. The severity of otalgia was not an indicator of recovery time (Table 4). Radiologic grading was compared in terms of recovery time. No statistically significant difference was found between radiologic Grades 2 and 3 according to recovery. Nevertheless, late recovery in patients with Grade 3 was clearly higher than that in patients with Grade 2. This is quite likely because of the small sample size of the cohort. We inferred that, with a larger sample size, the difference between grades would be statistically significant, and Grade 3 NEO would be linked to late recovery. Follow-up Patients were all informed to return to the ENT clinic in the case of any symptom recurrence and have been on follow-up for a mean of 26.57 T 9.46 months. Except for 1 patient who was actually a case of squamous cell carcinoma associated with NEO, no mortality was observed in our cohort during this period.

Radiologic grading for the 2 groups

Group 1 (n = 5)

Group 2 (n = 9)

n (%)

n (%)

p

3 (60.0%) 2 (40.0%)

3 (33.3%) 6 (66.7%)

0.580

Grade 2 Grade 3

TABLE 4. Spearman correlation coefficients (r) for intragroup parameters with recovery time

Fisher’s exact test.

DISCUSSION Even in recent years, it is still very difficult to make an accurate diagnosis of NEO because its physical findings simulate external otitis. Patients are almost always misdiagnosed as having a simple EAC skin inflammation

Otology & Neurotology, Vol. 35, No. 2, 2014

Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

NECROTIZING OTITIS EXTERNA OUTCOME PARAMETERS and treated with a single antibiotic until they return with a more serious clinical presentation. The elapsed time before definitive diagnosis leads to spread of the infection through the epidermal layer down to the temporal bone. Thereby, the inflammation changes its pattern from cellulitis to acute osteomyelitis, and if not diagnosed and treated in time, the disease progresses to a subacute form within 1 month and results in avascularization of the bone. This process gives rise to treatment failure leading to mastoid, petrous bone, petrous apex, and skull base involvement. Our reason to start the present study was to examine whether the duration of time between onset of symptoms and admission to hospital plays a role in laboratory, radiologic, and clinical outcomes in patients with NEO. The correlation analysis demonstrated that, apart from CRP levels, HbA1c, fasting blood sugar, ESR, otalgia VAS levels, improvement start time, improvement time, and total cure time were significantly associated with time to admission to hospital (Table 1). Considering the results derived from the present study, we aimed to investigate the critical time when the disease progressed to a more chronic form, in other words, to a worse outcome. Keeping in mind our hypothesis that this would be the median time to admission (30 d), we categorized the patients into 2 groups according to median time elapsed before hospitalization to verify the accuracy of our assertion. Our prevalence of diabetes (100%) was in agreement with some of the studies (2Y7,9). Blood glucose levels were significantly higher in patients with duration of symptoms greater than 30 days, and high HbA1c levels were associated with late recovery in patients with duration of symptoms less than 30 days. However, blood glucose levels were not correlated to recovery time in either of the groups. Both patient groups had poor glycemic control as in the series reported by Chin et al. (7). This may be explained by the fact that elderly diabetics, not insulin balanced, probably have a serious vasculopathy leading to a worse outcome (8). Disease involvement was graded using the system defined by Chin et al. based on temporal bone CT scanning and Tc-99mMDP SPECT/CT (7). In contrast to Chin et al. and Soudry et al., involvement of temporal bone, including Grades 2 and 3, was not correlated with disease progression in our cohort (7,10). The absence of a link between involvement and recovery time may be related to NEO not extending beyond the petrous apex and to the small size of the study. Nine (64%) patients had positive culture results for P. aeruginosa and one (7%) for S. aureus. In 4 patients (29%), pathogens were not identified. These results were in agreement with previous P. aeruginosa prevalence varying from 51% to 26.7% (8,10Y12). In our study, 750 mg of ciprofloxacin administered intravenously every 12 hours was the first-line antibiotic regimen. Otalgia of patients who were admitted to hospital earlier than 30 days started to resolve within 1 week. They went on to receive only ciprofloxacin and recovered with a shorter treatment course (mean, 61.60 T 4.33 d); however, patients who were

375

admitted later than 30 days from the start of symptoms did not respond in this period, received dual antibiotic therapy, and displayed late recovery (mean, 126.33 T 23.9 d). It appears that treatment start earlier than 30 days after onset of symptoms is more efficient even with a single IV ciprofloxacin. However, patients whose treatment started later than 30 days displayed later responses, even to dual antibiotherapy. Inflammatory markers were elevated in both groups, and ESR was significantly higher in patients with symptom duration longer than 30 days (mean, 107.11 T 12.46 mm/h). CRP values showed no significant difference between groups. However, increased levels of ESR and CRP were linked to late recovery and were predictors of cure delay in patients admitted later than 30 days after onset of symptoms (r = 0.872, p = 0.044). A significant decrease in ESR levels after 6 months of treatment compared with pretreatment values has been reported to be a good indicator of improvement of the disease (12Y14). As with ESR values, CRP returning to reference intervals is a useful test in the evaluation of complete disease resolution (7,15). In our series, we used CRP and ESR levels as inflammatory markers to monitor disease resolution and to predict the progression of disease. These tests have been proposed by authors as markers for diagnosis and resolution of disease superior to CT, MRI, and Tc-99mMDP (13). CT is sensitive to demineralization of bone greater than 30% but is a poor diagnostic tool for measuring treatment response because of the persistence of bone changes despite resolution of disease (16). MRI is better for showing dural enhancement and bone marrow involvement but is not adequate for displaying bone erosion. Nuclear imaging with Tc-99mMDP is the principal diagnostic imaging modality for NEO. However, this method is not only specific for osteomyelitis; the radiotracer concentrates in areas with osteoblastic activity as in trauma and malignancy. If there is any doubt, tissue biopsy is necessary to rule out malignancy. Indeed, the only patient who died of disease in our study was a case of NEO coinciding with temporal bone squamous cell carcinoma. Tc-99m was unable to make a definitive diagnosis between NEO and squamous cell carcinoma. The definitive diagnosis was made after several biopsies. Nine patients whose pain intensity and symptoms did not show any improvement within 1 week underwent HBOT combined with antibiotic regimen, 5 days a week, a total of 20 times. We infer that the low mortality rate in our cohort was the result of this combined therapy, and we endorse the suggestions of other authors recommending HBOT in all cases of NEO, provided there is no contraindication to hyperoxic treatment (17Y20). The present research has some limitations regarding its retrospective design and low power because of the restricted number of patients. The absence of a link between high radiologic grading and late recovery may be attributable to the small size of the study. A greater sample size would probably strengthen the power and the link between radiologic grading and late recovery. However, Otology & Neurotology, Vol. 35, No. 2, 2014

Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

376

A. VERIM ET AL.

excluding CRP values, data were all significantly related to time to hospitalization. Consequently, we do not expect to see different results in a larger sample size. As for the only patient who died of disease, this was a case of NEO synchronous with temporal bone squamous cell carcinoma in mastoid cells. Although several biopsies were taken, the neoplasm was only diagnosed on the fifth sampling. 18F-labeled fluorodeoxyglucose positron emission tomography may be more sensitive than Tc-99m and Ga-67 bone scans because it reflects intralesional metabolic activity better than these techniques. Since the first identification of NEO in 1959, many parameters such as the presence of cranial nerve paralysis, patient’s age, associated comorbidities, degree of temporal bone involvement on CT or MRI, and inflammatory markers (CRP, ESR), have been examined to try to define the prognostic indicators of the disease course. However, no study has been carried out to determine the role of duration of symptoms on NEO outcome. With this article, we hope to provide clinicians a new perspective on the outcome parameters of the disease. With our results as basis, we suggest that the time from onset of symptoms to admission is an important parameter linked to HbA1c, fasting blood sugar, ESR levels, degree of otalgia, improvement, and total recovery time. In our opinion, clinicians should consider whether duration of symptoms has exceeded 30 days before deciding on treatment modality and disease outcome. CONCLUSION The results of this study have shown that a maximum of 30 days from symptom onset to hospitalization for NEO is critical in terms of disease control. Delayed hospitalization exceeding 1 month has a negative influence on disease duration because of the transformation of osteomyelitis from the acute to chronic form. Otalgia severity, blood glucose levels, CRP concentrations, ESR, and HbA1c levels all increase significantly in patients hospitalized later than 1 month after symptom onset. High levels of ESR (9100 mm/h) and CRP are linked to late recovery and are found to be predictors of cure delay in patients admitted later than 1 month after symptom onset. However, pain severity is not an indicator of recovery time. Nuclear imaging Tc-99m MDP and Ga-67 citrate should be supported by multiple biopsies and FDG-PET imaging to rule out neoplasms.

REFERENCES 1. Meltzer PE, Kelemen G. Pyocutaneous osteomyelitis of the temporal bone, mandible, and zygoma. Laryngoscope 1959;169: 1300Y16. 2. Berenholz L, Katzenell U, Harell M. Evolving resistant pseudomonas to ciprofloxacin in malignant otitis externa. Laryngoscope 2002;112:1619Y22. 3. Djalilian HR, Shamloo B, Thakkar KH, et al. Treatment of culturenegative skull base osteomyelitis. Otol Neurotol 2006;27:250Y5. 4. Peleg U, Perez R, Raveh D, Berelowitz D, Cohen D. Stratification for malignant external otitis. Otolaryngol Head Neck Surg 2007;137:301Y5. 5. Franco-Vidal V, Blanchet H, Bebear C, et al. Necrotizing external otitis: a report of 46 cases. Otol Neurotol 2007;28:771Y3. 6. Slattery WH, Brackmann DE. Skull base osteomyelitis: malignant otitis externa. Otolaryngol Clin North Am 1996;29:795Y806. 7. Chin R, Roche P, Sigston E, Valance N. Malignant otitis externa: an Australian case series. Surgeon 2012;10:273Y7. 8. Chen CN, Chen YS, Yeh TH, Hsu CJ, Tseng FY. Outcomes of malignant external otitis: survival vs mortality. Acta Otolaryngol 2010;130:89Y94. 9. Joshua BZ, Sulkes J, Raveh E, Bishara J, Nageris BI. Predicting outcome of malignant external otitis. Otol Neurotol 2008;29: 339Y43. 10. Soudry E, Joshua BZ, Sulkes J, Nageris BI. Characteristics and prognosis of malignant external otitis with facial paralysis. Arch Otolaryngol Head Neck Surg 2007;133:1002Y4. 11. Chen YA, Chan KC, Chen CK, Wu CM. Differential diagnosis and treatments of necrotizing otitis externa: a report of 19 cases. Auris Nasus Larynx 2011;38:666Y70. 12. Carfrae MJ, Kesser BW. Malignant otitis externa. Otolaryngol Clin North Am 2008;41:537Y49. 13. Rubin J, Yu VL. Malignant external otitis: insights into pathogenesis, clinical manifestations, diagnosis and therapy. Am J Med 1988;85:391Y8. 14. Omran AA, El Garem HF, Al Alem RK. Recurrent malignant otitis externa: management and outcome. Eur Arch Otorhinolaryngol 2012;269:807Y11. 15. Pepys MB, Hirschfield GM. C-reactive protein: a critical update. J Clin Invest 2003;111:1805Y12. 16. Grandis JR, Curtin HD, Yu VL. Necrotizing (malignant) external otitis: prospective comparison of CT and MR imaging in diagnosis and follow-up. Radiology 1995;196:499Y504. 17. Goldstein LJ, Gallagher KA, Bauer SM, et al. Endothelial progenitor cell release into circulation is triggered by hyperoxiainduced increases in bone marrow nitric oxide. Stem Cells 2006;24:2309Y18. 18. Sheikh AY, Gibson JJ, Rollins MD, et al. Effect of hyperoxia on vascular endothelial growth factor levels in a wound model. Arch Surg 2000;135:1293Y7. 19. Park MK, Myers RA, Marzella L. Oxygen tensions and infections: modulation of microbial growth, activity of antimicrobial agents, and immunologic responses. Clin Infect Dis 1992;14:720Y40. 20. Davis JC, Gates GA, Lerner C, et al. Adjuvant hyperbaric oxygen in malignant external otitis. Arch Otolaryngol Head Neck Surg 1992;118:89Y93.

Otology & Neurotology, Vol. 35, No. 2, 2014

Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.

Clinical outcome parameters for necrotizing otitis externa.

To investigate the duration of time elapsed between the onset of symptoms for necrotizing external otitis (NEO) and admission to hospital that may pla...
315KB Sizes 0 Downloads 0 Views