Portal de Periódicos da CAPES

HOME MONITORING OF THE MIDDLE EAR SYSTEM WITH SPECTRAL GRADIENT ACOUSTIC REFLECTOMETRY: DISTINGUISHING ACUTE OTITIS MEDIA FROM UPPER RESPIRATORY INFECTION

2000; Lippincott Williams & Wilkins; Volume: 19; Issue: 4 Linguagem: Inglês

10.1097/00006454-200004000-00023

1532-0987

Elizabeth D. Barnett, Howard Cabral, Jerome O. Klein,

Phonocardiography and Auscultation Techniques

Acoustic reflectometry has been used for more than a decade as an adjunct to pneumatic otoscopy and tympanometry for diagnosis of middle ear effusion (MEE).1 The latest models of the instrument use spectral gradient analysis to determine the probability of MEE by measuring the response of the tympanic membrane to a tone directed into the ear canal. The machine analyzes data from the incident and reflected sound waves and presents the data as a spectral gradient angle (professional model) or a number from 1 to 5 (consumer model).2 The instrument is generally well-tolerated by children and easy to use with children as young as 6 months of age. The availability of a consumer version of the instrument makes possible daily monitoring of the status of the middle ear system in the home, with little inconvenience to families. We studied home monitoring of the middle ear system using spectral gradient acoustic reflectometry in homes where pediatrician or nurse practitioner parents were able to perform pneumatic otoscopy. We hypothesized that in this setting, changes in the middle ear associated with upper respiratory signs occurring in the presence and absence of acute otitis media (AOM) as well as diurnal variation in readings might be identified. The goal of the study was to obtain information about ease of use of the instrument, to study spectral gradient angle (SGA) readings in the presence and absence of upper respiratory illness signs and during periods of good health as well as periods of time surrounding episodes of AOM and to assess diurnal variation of readings. Methods. Children were enrolled from families in which a parent was able to perform pneumatic otoscopy and was willing to take daily readings with the reflectometer. Parents were not required to be validated otoscopists. Healthy children 6 months to 6 years of age were eligible. Parents were asked to take at least daily, and twice daily when possible, readings with the spectral gradient acoustic reflectometer (EarCheck Pro, Becton Dickinson, Franklin Lakes, NJ) and to record the readings and the times at which they were taken. If the child remained well parents were asked to examine the child's ears weekly with pneumatic otoscopy and record the results. If the child was ill (fever, earache or signs of AOM, or upper respiratory infection) parents were asked to perform daily examinations with pneumatic otoscopy until the child was asymptomatic. Children continued to receive usual medical care, including assessment and treatment by their primary care doctor for ear infections or other illnesses. AOM was defined as one or more signs of acute ear disease (ear pain, acute hearing loss or ear drainage) or systemic illness plus presence of MEE by pneumatic otoscopy. For purposes of the analysis children were defined as being in an otitis interval from 3 days before until 2 weeks after diagnosis of an episode of AOM and in an otitis-free interval at all other times. Instrument readings were recorded as SGAs. For the commercial instrument readings are grouped into five levels (Level 1, SGA >95; Level 2, 70 to 95; Level 3, 60 to 69; Level 4, 49 to 59; and Level 5, SGA <49). Levels are based on the likelihood of presence of MEE, with higher levels and lower SGAs associated with greater likelihood of MEE. Mean values, standard errors and P values were calculated based on models that made statistical adjustment for within child correlation, given that children were assessed multiple times throughout the study.3 The study was approved by the Human Studies Committee at Boston Medical Center. Results. Seventeen children were monitored by 7 parents from February 9, 1998, through May 31, 1998. One parent was unable to obtain any valid readings for 1 child; results were available for 16 children. Children ranged in age from 9 months to 6 years (mean, 39.5 months; median, 34 months). Valid readings were obtained from 16 children on 80% of attempts. Thirteen children had no episodes of AOM during the study period. For these children 95% of readings (1499 of 1572) were in Levels 1 or 2 (SGA ≥ 70), and 2% of readings (37 of 1572) were in Levels 4 or 5 (SGA < 60). When runny nose, cough or runny nose plus cough was present, the mean SGA was lower than when children were symptom-free (Table 1). The mean SGA in symptom-free children was 106; with runny nose the mean SGA was 98.7, with cough it was100.5 and with runny nose and cough it was 95.7 (P = 0.0001 for all 3 comparisons).TABLE 1: Variation in mean spectral gradient angle with signs of upper respiratory infection in children with and without acute otitis mediaThere were 4 episodes of AOM in 3 children during the study. All episodes were confirmed by the child's primary care pediatrician. On the day of diagnosis of AOM 33% of readings (4 of 12) were in Levels 1 or 2, and 50% (6 of 12) were in Levels 4 or 5. During the entire study period 215 of 269 readings (80%) were in Levels 1 and 2, and 22 of 269 (8%) were in Levels 4 or 5. During AOM intervals (3 days before until 14 days after the day of diagnosis), 64 of 102 readings (63%) in these children were in Levels 1 and 2, and 14 of 102 (14%) were in Levels 4 or 5. When these 3 children were in AOM-free intervals, 90% of readings (151 of 167) were in Levels 1 or 2 and 5% (8 of 167) were in Levels 4 or 5. Children who had AOM at any time during the study had no significant change in mean SGA with runny nose or cough. (Table 1). To determine diurnal variation in SGA we compared readings taken in the morning (6 to 10 a.m.) with readings taken in the evening (5 to 9 p.m.). For children who never had AOM during the study period there was no difference between readings taken in the morning and at night. For children who had AOM during the study period, mean SGA was significantly lower in the morning (76.1) than in the evening (86.5) during the AOM-free interval; there was no diurnal variation in readings taken during the AOM interval. Discussion. Parents in this study were able to obtain valid readings with EarCheck on 80% of attempts. Proposed uses of the instrument have included intermittent use in children suspected of having AOM, monitoring resolution of MEE in children immediately after an episode of AOM, and monitoring of presence of MEE in children with persistence of effusion.4 The instrument is equivalent to tympanometry for diagnosis of MEE compared with findings at surgery5 and compared with pneumatic otoscopy.4, 6 This study is the first to examine monitoring of healthy children who did not have any AOM. The results indicate that 95% of children who are AOM-free have readings in Levels 1 or 2. Children with diagnosed AOM had readings in Levels 4 and 5; these readings occurred 50% of the time in children diagnosed with AOM on the day of diagnosis, compared with 14% of the time when these children were AOM-free. Only 2% of readings were in Levels 4 and 5 in children without AOM. The results also demonstrated that although children with other signs (runny nose or cough) may have readings that changed from baseline when the child was completely well, the readings were changed minimally and differently from when children had AOM. This device was able to detect subtle differences in the middle ear system that represent either inflammation of the middle ear mucosa or changes in middle ear pressure. This information might be helpful to both parents and physicians in distinguishing upper respiratory infection from AOM. A potential limitation of this study is the small number of children who developed AOM during the study period. This study was designed, however, not to assess the instrument's ability to diagnose AOM, but to collect information daily about the status of the middle ear system in children with and without AOM. Baseline readings were lower in children who had AOM during the study period; this may represent a persistence of inflammation or other changes in the middle ear system that are not clinically apparent but are measurable by SGAR. Variation in readings from morning to evening suggests that there may be subtle changes in the middle ear that occur in the absence of evident MEE. For the individual child greatest consistency would be obtained if readings were taken at the same time every day. This might be especially important for the child with persistence of MEE for whom readings are being taken to make decisions about further interventions. Acknowledgments. We thank the pediatricians, nurse practitioners and their children who participated in this study. Demian Christiansen provided valuable technical assistance. This study was supported in part by a grant from MDI Instruments, Becton Dickinson, Franklin Lakes, NJ.