Nonsurgical home treatment of middle ear effusion and associated hearing loss in children. Part II: validation study.

Abstract

In this prospective follow-up investigation, we examined the efficacy of a modified Politzer device in the home treatment of persistent middle ear effusion (MEE) and associated hearing loss in children who had previously participated in a similar clinical trial. Our study group was made up of 38 patients who had been either (1) untreated control participants in the previous study whose hearing in one or both ears had not returned to normal within 11 weeks of their initial audiologic pretest ("former control group"; n = 30), or (2) active-treatmentparticipants in the previous study whose hearing sensitivity in at least one ear had not improved to within normal limits after treatment and who elected to undergo a continuation of treatment ("extended-treatment group"; n = 8). Treatment efficacy was determined by comparing differences in pre- and posttreatment air-conduction thresholds and otoscopic findings. Following treatment, the former control group experienced significant improvements in hearing sensitivity at all frequencies; at the posttreatment test, hearing sensitivity was within normal limits in 43 of 60 ears (71.7%), and normal or moderate tympanic membrane mobility was observed in 30 of 34 otoscopically examined ears (88.2%). In the extended-treatment group, hearing sensitivity returned to within normal limits in 9 of 10 impaired ears (90.0%). These findings further substantiate the efficacy of our modified Politzer device in improving middle ear function and hearing sensitivity in children with MEE, and they support the reliability of the findings reported in our previous study. These results also indicate that many patients in whom initial treatment is not successful may benefit from extended treatment.

Introduction

We recently reported the results of a randomized controlled clinical trial of a modified Politzer device for the home treatment of persistent middle ear effusion (MEE) and associated hearing loss. (1) Our handheld, battery-operated device emits controlled air pressure and air flow that can be adjusted in accordance with the degree of hearing loss and the patient's age.

In our previous study, we found that twice-daily home treatment over a period of 7 weeks was highly efficacious in restoring hearing in children between the ages of 4 and 11 years who had MEE. At the posttreatment audiologic test, hearing sensitivity had returned to normal in 65 of 88 treated ears (73.9%) and in at least one ear in 40 of 47 treated patients (85.1%); the corresponding figures for the control group were only 23 of 86 (26.7%) and 15 of 47 (31.9%). We also reported that pneumatic otoscopy revealed normal or moderate tympanic membrane mobility in 61 of 65 successfully treated ears (93.8%). (1)

At the conclusion of our previous study, we offered the controls whose heating had not recovered the opportunity to undergo active treatment. We also offered extended treatment to those patients in the active-treatment group who had not completely improved. In this article, we report the findings of our prospective follow-up study.

Patients and methods

Instrumentation. The design and function of our modified Politzer device are described in the report of our previous study. (1) The device emits a controlled air pressure and volume velocity sufficient to effect improvement without discomfort. The device has two settings. Setting number 1 delivers an air pressure of 5.2 psi at a volume velocity of 1,524 ml/min; setting 2 delivers an air pressure of 2.5 psi at a volume velocity of 1,690 ml/min. The device was very similar to one that we had used in two other studies reported in 19992 and 2000. (3)

Patients. Our study population was made up of patients who had furnished informed consent and who satisfied six study criteria: (1) age 4 to 11 years, (2) at least a 2-month history of MEE and associated hearing loss as documented by a physician, (3) pure-tone air-conduction thresholds of 20 dB HL or more at 3 frequencies between 500 and 4,000 Hz with air-bone gaps of 15 dB or more at these frequencies or pure-tone air-conduction thresholds of 25 dB HL or more at 2 frequencies between 500 and 4,000 Hz with air-bone gaps of 15 dB or more at these frequencies at the final pretest, (4) a tympanometric peak pressure of -100 daPa or less at the final pretest, (5) an otologic diagnosis of MEE at the final pretest, and (6) an absence of enlarged adenoids, acute otitis media, and other ear abnormalities at the final pretest otologic examination.

Former control group. At the completion of our previous study, 32 of the 47 controls had not achieved normal hearing in at least one ear. Of these 32 patients, 30 accepted our offer to undergo active treatment.

Extended-treatment group. Eight patients in the active-treatment group of our previous study who had not achieved recovery of heating in one or both ears accepted our offer to continue treatment in the current study.

Procedures. The procedures used in this study were the same as those followed in our previous report, except that we did not include any untreated controls. All patients in this follow-up study received active treatment.

Each patient's parent administered treatment in the morning before breakfast and again in the evening after supper. Each participant had his or her own device. The parent was given alcohol-soaked wipes and instructed to clean the tip of the pediatric probe, which was coupled to the device, before each use. During each treatment, the patient was instructed to be in a sitting position. To deliver therapy, the parent inserted the tip into one nostril while compressing the other nostril with a finger (figure). The child then held a small amount of water in the mouth without swallowing it. The parent then turned on the device, thereby introducing air flow into the nostril at a constant volume velocity. After 1 or 2 seconds of air flow, the parent asked the child to swallow the water. Immediately thereafter, the same therapy was administered to the other nostril. Approximately 10 minutes later, the procedure was repeated in both nostrils. (Note: Our description of the treatment protocol in part I of this article (1) was incorrect. See "Erratum" on page 614 of this issue.)

The planned duration of treatment for the former control group was 7 weeks. The patients in the extended-treatment group (who had already received 7 weeks of treatment) were scheduled to undergo an additional 2 weeks of treatment. Two weeks after the completion of extended treatment, they underwent an audiologic assessment. Those whose hearing was not within normal limits in one or both ears were given another 2 weeks of treatment and assessed again 2 weeks after its completion. This treatment-and-assessment cycle continued until either hearing sensitivity was restored to within normal limits or no further improvement in hearing sensitivity was observed.

Parents were provided with a daily log to foster and track compliance. Most patients were seen in the office for monitoring at least once near the midpoint of the treatment period. Parents were advised to discontinue treatment if the child developed a head cold or infection and to resume treatment only after obtaining clearance from a study otologist or the child's pediatrician.

Audiometry was performed by audiologists certified by the American Speech-Language-Hearing Association and licensed by New York State. Otologic evaluations were performed by board-certified otolaryngologists with at least 15 years of experience. Audiologic evaluations (air- and bone-conduction threshold testing) and otolaryngologic evaluations (otoscopy) had been performed on all patients upon their enrollment in the previous study. Follow-up audiologic and otologic evaluations were performed at the midpoint of treatment for the former controls and biweekly during treatment for the extended-treatment patients. The final audiologic and otologic evaluations were conducted on all patients 4 weeks after the completion of each patient's final treatment. Hearing sensitivity was judged to be within normal limits if the air-conduction thresholds were less than 20 dB HL at any of 3 frequencies between 500 and 4,000 Hz or less than 25 dB HL at either of 2 of these frequencies. Audiologists were blinded to each patient's otologic findings, and otolaryngologists were blinded to each patient's audiometric findings. An alpha level of 0.05 was used for all statistical tests.

Results

Former control group. Of the 30 former controls, 27 underwent 7 weeks of treatment and 3 underwent 9 weeks.

We obtained pre- and posttreatment mean air-conduction thresholds (pretest and posttest, respectively) in both ears in the former controls and compared them with the same values for the active-treatment group in the previous study (table 1). The pretest values for the former controls were obtained 11 weeks following their enrollment in the previous study, and their posttest values were obtained 4 weeks following the completion of treatment in the current study. The mean pretest air-conduction thresholds in the 2 groups were within 4.6 dB at 500 Hz, within 5.0 dB at 1,000 Hz, within 3.0 dB at 2,000 Hz, and within 2.7 dB at 4,000 Hz. In the former control group, mean pretest 4-frequency pure-tone averages were 28.5 and 28.8 dB HL in the right and left ears, respectively; in the previous active-treatment group, the corresponding figures were 29.6 and 32.6 dB HL. Thus, the mean pure-tone averages were symmetrical within 4.1 dB for both ears of both groups.

Analysis of the mean differences in pre- and posttest air-conduction thresholds for both ears in the former control group revealed that treatment resulted in a statistically significant (p < 0.001) improvement in heating in both ears, ranging from 10.3 to 16.3 dB, across the frequency range (table 2).

We conducted between-group comparisons to determine if treatment was more or less effective in the former control group than it had been in the active-treatment group in the previous study. We found no statistically significant differences (table 3). This finding can be interpreted as supporting the reliability of the findings reported in our previous study.

Hearing sensitivity. In the current study, recovery of hearing sensitivity to within normal limits following treatment occurred in 43 of the 60 ears (71.7%) in the former control group. Results of the previous study were similar, as active treatment restored hearing in 65 of 88 ears (73.9%). (1)

Otoscopic findings. We performed posttest pneumatic otoscopy in 34 of the 43 ears in the former control group in which hearing had been restored to within normal limits. Of these 34 ears, tympanic membrane mobility was found to be normal in 29 (85.3%), moderate in 1 (2.9%), slight in 3 (8.8%), and absent in 1 (2.9%). Overall, normal or moderate tympanic membrane mobility was observed in 30 of these ears (88.2%). In our previous study, normal or moderate tympanic membrane mobility was observed in 61 of 65 ears (93.8%). (1) This similarity further substantiates the efficacy of our device in improving middle ear function and heating sensitivity in children with MEE, and it further supports the reliability of the findings reported in our previous study.

Extended-treatment group. The duration of extended treatment ranged from 2 to 4 weeks, bringing the total duration of all treatment to 9 to 11 weeks. We compared the pure-tone thresholds obtained following the completion of the initial treatment and the completion of extended treatment.

Following the initial treatment period, 6 of these patients had a unilateral hearing impairment (patients 2, 3, 4, 5, 6, and 7) and 2 patients had a bilateral heating impairment (patients 1 and 8)--a total of 10 impaired ears (table 4). Following extended treatment, normal hearing was restored in 9 of the 10 ears (90.0%). Normal hearing was restored in 5 of the 6 patients with unilateral impairment and in both ears of the 2 patients with bilateral impairment.

When the number of ears that recovered after initial plus extended treatment (9) is added to the number that recovered after the initial treatment (65), normal heating overall was restored in a total of 74 of 88 ears (84.1%).

Follow-up. Upon the completion of their children's participation in the current study, parents were advised to notify the investigators if they or others noticed any change in hearing status. One parent of a former control whose hearing was restored to within normal limits contacted us 1 month later to report a recurrence of the child's hearing impairment. Audiologic and otologic evaluations revealed that this child had experienced a recurrence of MEE. The parent was instructed to resume treatment for 3 weeks and to return for evaluation 2 weeks after the completion of that treatment. Although the treatment was completed, the parent and patient did not return until 4 months later. At that time, the child's hearing was normal in both ears.

In the report of our previous study, we noted that 3 children whose heating had been restored by treatment subsequently experienced recurrent MEE with hearing loss. The hearing in all 3 of those patients returned to within normal limits after treatment had been reinstituted for 2 weeks. These findings suggest that the use of our modified Politzer device may be effective for the treatment of recurrent MEE in patients who have already responded to previous treatment. Larger studies are needed to confirm the efficacy of our device in treating recurrences.

Discussion

A limitation of this investigation was the absence of a control group. This study, however, represents a follow-up of the participants from our previous study whose hearing sensitivity did not recover to within normal limits following treatment, so we did not recruit another control group.

Therefore, the contribution of spontaneous recovery to the improvement observed cannot be entirely ruled out. Recall that the former control group and the extended-treatment group demonstrated close similarity in pretreatment air-conduction thresholds and in several other measures, including the degree of improvement from pre- to posttest, the percentages of ears demonstrating recovery of heating sensitivity to within normal limits at the posttest, and the percentages of ears with normal-to-moderate tympanic membrane mobility. The close similarity between the two groups in (1) pretest hearing sensitivity, (2) the degree of improvement from pre- to posttest, (3) the percentages of ears demonstrating recovery of hearing sensitivity within to normal limits at the posttest, (4) the substantial and significant (p < 0.001) improvement in heating sensitivity from pre- to posttest, and (5) the lack of change in hearing sensitivity in the former control group throughout the 11-week period in our previous study followed by improvement only after initiation of the home treatment (with increasingly greater improvement observed with the progression of treatment as evidenced by the results of monitoring during the treatment) suggests that the major factor accounting for the improvements seen at the posttest was home treatment rather than spontaneous recovery.

In a recent prospective follow-up study of children with persistent MEE and heating loss who had been treated early in life with tympanostomy tubes, Valtonen et al concluded that parents must be advised of the necessity of lengthy (5 yr) follow-up monitoring, the possible need for repeated surgical placement of tympanostomy tubes, and the possibility that middle ear complications may require other surgical management. (4) Use of our device might serve as an alternative to repeated surgical insertions of ventilation tubes; further research is needed to investigate this possibility.

In conclusion, the findings of our current study further substantiate the efficacy of our modified Politzer device in improving middle ear function and hearing sensitivity in children with MEE, and they support the reliability of the findings of our previous study.

Acknowledgments

The authors thank Christoph Kimmich, president of Brooklyn College, CUNY, who generously provided the suites for parent training, otolaryngology examination, and audiologic testing throughout our research. In addition, we thank otolaryngologists Abraham Sinnreich, MD, George Braun, MD, and Samuel Nodelman, MD, who served on the medical research team, as well as the three research audiologists, Genevieve Davitt, Stella Boshoyer, and Mafia Manevich. We also thank Carol Ann Silverman, PhD, for her valuable contributions.

References

(1.) Arick DS, Silman S. Nonsurgical home treatment of middle ear effusion and associated hearing loss in children. Part I: Clinical trial. Ear Nose Throat J 2005;84:567-8, 570, 572-574, 576, 578.

(2.) Silman S, Arick D. Efficacy of a modified politzer apparatus in management of eustachian tube dysfunction in adults. J Am Acad Audiol 1999;10:496-501.

(3.) Arick DS, Silman S. Treatment of otitis media with effusion based on politzerization with an automated device. Ear Nose Throat J 2000;79:290-2, 294, 296.

(4.) Valtonen H, Tuomilehto H, Qvarnberg Y, Nuutinen J. A 14-year prospective follow-up study of children treated early in life with tympanostomy tubes: Part 1: Clinical outcomes. Arch Otolaryngol Head Neck Surg 2005;131:293-8.

Shlomo Silman, PhD; Daniel S. Arick, MD, FACS; Michele B. Emmer, PhD

Dr. Arick is on the attending staff of the Department of Otolaryngology-Head and Neck Surgery, New York Eye & Ear Infirmary, New York City. Dr. Silman and Dr. Emmer are with the Center for Auditory Research, Brooklyn College, City University of New York, Brooklyn.

Reprint requests: Dr. Shlomo Silman, Center for Auditory Research, Brooklyn College, CUNY, 2900 Bedford Ave., Brooklyn, NY 11210. Phone: [phone]; fax: [phone]; e-mail: [e-mail]

This study was conducted at the Brooklyn (N.Y.) College Center for Auditory Research and at Arisil Instruments of Brooklyn. This research was supported by a grant (No. 5R44DC003613-03) from the Small Business Innovation Research Program, National Institute on Deafness and Other Communication Disorders, National Institutes of Health to Arisil Instruments. Dr. Arick and Dr. Silman were the principal investigators, and Dr. Emmer was senior research audiologist.

Table 1. Mean (standard deviation) pre- and posttest air-conduction
thresholds (dB HL) in the former control group in the current study
and in the active-treatment group in the previous study

Group             Test   Ear   n   500 Hz  1,000 Hz  2,000 Hz  4,000 Hz

Current study:    Pre   Right  30   31.0     32.0      22.2      28.8
Former control                      (8.9)    (9.2)    (10.6)    (11.7)
  group *               Left   30   30.7     32.7      23.0      28.7
                                   (10.1)   (12.7)    (11.6)    (14.7)

Previous study:   Pre   Right  43   33.0     32.1      23.8      29.4
Active-treatment                   (10.9)   (10.1)    (11.0)    (12.1)
                        Left   45   35.3     37.7      26.0      31.4
                                   (11.4)   (10.8)    (12.2)    (11.7)

Current study:    Post  Right  30   18.5     18.2      11.8      15.0
Former control                     (12.3)   (12.8)    (11.3)    (11.6)
  group                 Left   30   18.7     16.3      11.0      14.2
                                    (8.2)    (9.7)     (9.9)    (11.8)

Previous study:   Post  Right  43   18.3     15.5       7.9      12.0
Active-treatment                   (76.0)    (8.4)    (79.0)     (9.8)
  group                 Left   45   19.0     17.3       8.7      14.7
                                   (10.3)   (11.0)     (8.2)     (8.1)

* The former control group was made up of 30 patients who had served as
untreated controls in the previous clinical trial. (1) Upon conclusion
of their participation in that study, they elected to receive active
treatment.

([dagger]) Data from the active-treatment group in the previous study'
are presented for comparison purposes.

Table 2. Mean (95% confidence interval) differences (posttest minus
pretest) in air-conduction thresholds (dB HL) in the 30 former
controls *

Ear        500 Hz         1,000 Hz       2,000 Hz       4,000 Hz

Right       -12.8          -13.8           -10.3          -13.8
        (-17.5, -8.1)  (-18.5, -9.2)   (-14.3, -6.4)  (-19.1, -8.5)

Left        -12.0          -16.3           -12.0          -14.5
        (-15.6, -8.4)  (-21.2, -11.5)   (-170, -70)   (-20.0, -9.0)

* All differences are statistically, significant (p < 0.001) according
to the Student's paired t test. The negative values represent
improvements in hearing sensitivity following treatment.

Table 3. Mean (95% confidence interval) differences
between groups (values for the previous study's
active-treatment group minus values for the current
study's former control group) in the changes (post-test
minus pretest) in air-conduction thresholds (dB
HQ *

Ear         500 Hz       1,000 Hz       2,000 Hz       4,000 Hz

Right        -1.9          -2.8           -5.6           -3.6
          (-7.4, 3.5)   (-8.5, 2.9)   (-11.3, 0.1)   (-10.2, 2.9)
           p = 0.48      p = 0.33      p = 0.053       p = 0.28

Left         -4.3          -2.0           -5.3           -2.3
          (-9.3, 0.6)   (-7.4, 3.5)   (-11.1, 0.4)   (-8.5, 3.9)
           p = 0.09      p = 0.46       p = 0.07       p = 0.47

Statistical significance was determined according to an independent
samples t test (assuming equal variances). The negative values
represent improvements in hearing sensitivity following treatment.

Table 4. Pure-tone thresholds for 500 through 4,000 Hz in 8 patients
following initial treatment and following extended treatment

                                             Frequency (Hz)

                                                Right ear

Pt.   Test                            500   1,000   2,000   4,000

1     After 7 wks of initial Tx *     30     25      15      25
      After extended Tx ([dagger])    10     10      15      10

2     After 7 wks of initial Tx       15      5       5       5
      After extended Tx               10     10       0       5

3     After 7 wks of initial Tx       25     30      25      20
      After extended Tx               20     15      15      10

4     After 7 wks of initial Tx       30     25      30      25
      After extended Tx               10     15      15      10

5     After 7 wks of initial Tx       30     30      10      10
      After extended Tx               25     15      15      15

6     After 7 wks of initial Tx       10     20      10      10
      After extended Tx               15     15       5      10

7     After 7 wks of initial Tx       15     25      20      10
      After extended Tx               10      0      10      10

8     After 7 wks of initial Tx       35     30      20      20
      After extended Tx               20     20       0      15

                                             Frequency (Hz)

                                                Left ear

Pt.   Test                            500   1,000   2,000   4,000

1     After 7 wks of initial Tx *     25     25       5      25
      After extended Tx ([dagger])    10     10       5       0

2     After 7 wks of initial Tx       35     25      15      30
      After extended Tx               20     15       5      15

3     After 7 wks of initial Tx       25     20      10      15
      After extended Tx               15     15      15      10

4     After 7 wks of initial Tx       15     10      10      15
      After extended Tx               10     10      10      10

5     After 7 wks of initial Tx       15     15       5      15
      After extended Tx               25     20       5      15

6     After 7 wks of initial Tx       25     30      20      25
      After extended Tx               15     15      15      10

7     After 7 wks of initial Tx       55     60      45      40
      After extended Tx               35     45      30      15

8     After 7 wks of initial Tx       35     35      20      15
      After extended Tx               15     10      10       0

* Previous study. Posttest was performed 4 weeks following the
conclusion of the 7-week treatment course. Note that these patients
had been treated for 7 weeks in the previous clinical trial, but they
did not reach the criteria, for complete recovery. Their pure-tone
thresholds upon initial enrollment in the previous clinical trial
were poorer than those indicated in this table.

([dagger]) Current study. Posttest was performed 2 weeks following
the conclusion of extended treatment.