Gourin CG, Hubbell RN. Otorrhea After Insertion of Silver Oxide–Impregnated Silastic Tympanostomy Tubes. Arch Otolaryngol Head Neck Surg. 1999;125(4):446-450. doi:10.1001/archotol.125.4.446
Silver oxide–impregnated tympanostomy tubes have been shown to decrease the incidence of postoperative otorrhea, but without a significant effect in the first postoperative week.
To evaluate prospectively our results with silver oxide–impregnated tympanostomy tubes and to identify factors associated with a higher incidence of early postoperative otorrhea.
Prospective nonrandomized study.
University referral center.
Patients and Other Participants
Six hundred thirty patients with chronic otitis media with effusion or recurrent otitis media.
Silver oxide–impregnated Silastic tympanostomy tubes were inserted in 1254 ears. Subjects with mucoid or purulent effusions or blood at the myringotomy site at surgery were treated with topical antibiotic prophylaxis (sulfacetamide sodium–prednisolone acetate or neomycin sulfate–polymyxin B sulfate–hydrocortisone) for 5 days after tympanostomy tube placement.
Main Outcome Measures
Incidence of otorrhea after tympanostomy tube insertion at 1 week and 1, 3, 6, 9, and 12 months after surgery.
The overall incidence of postoperative otorrhea was 1.9%. The incidence of otorrhea in the first postoperative week was 5.6%; the incidence of otorrhea after the first postoperative week was 1.2% (P<.001). Within the first postoperative week, a significantly greater incidence of otorrhea was noted in patients younger than 3 years (7.8%), in patients with mucoid effusions at surgery (8.6%), and in patients younger than 3 years with mucoid effusions at surgery (15.2%).
Silver oxide–impregnated tympanostomy tubes are associated with a low overall incidence of postoperative otorrhea. A significantly higher incidence of otorrhea is seen during the first postoperative week, compared with the incidence after the first week. Patients with thick middle ear effusions and age younger than 3 years have a significantly greater incidence of early otorrhea after tympanostomy tube placement.
PURULENT OTORRHEA is the most common complication following tympanostomy tube insertion and has been reported to occur in 9% to 34% of cases.1- 8 Postoperative otorrhea may result from persistent infection of the middle ear space or contamination of the middle ear with recurrent infection. An increased incidence of early otorrhea has been described in patients with purulent or mucoid middle ear effusions, blood at the myringotomy site, and age less than 3 years.1,2,4,5,7,9,10 Recent studies suggest that the use of topical antibiotic prophylaxis in the perioperative period may decrease the incidence of early otorrhea, particularly in these high-risk subpopulations.4,6,8- 11 Other studies have failed to show a statistically significant benefit.2,7,10 The use of topical antibiotic prophylaxis to limit otorrhea has failed to eradicate the problem for those subpopulations most at risk, in whom the incidence remains high.10
Silver compounds have been shown to prevent bacterial adherence and colonization in vitro without causing cell toxicity.12- 14 Clinical trials studying the use of silver oxide–coated urinary catheters in vivo have demonstrated marked reductions in catheter-associated urinary tract infections.15- 17 A pilot study by Chole and Hubbell18 reported that silver oxide impregnation of Silastic tympanostomy tubes decreased the incidence of postoperative otorrhea compared with matched controls. However, no significant difference in the incidence of otorrhea was seen in the immediate postoperative period. This prospective study was designed (1) to evaluate our results with silver oxide–impregnated tympanostomy tubes, (2) to identify factors associated with a higher incidence of postoperative otorrhea, and (3) to determine whether the judicious use of topical antibiotic prophylaxis decreases the incidence of early otorrhea.
From July 1, 1995, to February 1, 1997, all patients undergoing myringotomy with tympanostomy tube insertion by the senior author (R.N.H.) at Fletcher Allen Health Care at the University of Vermont, Burlington, were entered into this study. Indications for tympanostomy tube insertion were chronic otitis media with effusion, defined as the presence of a middle ear effusion for 3 months after failure of 2 or more courses of antibiotics; recurrent otitis media, defined as monthly otitis media during a period of 4 months, after failure of prophylactic antibiotics; or both. Patients undergoing concomitant tonsillectomy, adenoidectomy, or both were included. Patients with known immunodeficiency, immotile cilia syndrome, cystic fibrosis, and cleft palate were excluded from analysis. Patients were also excluded if they did not complete at least 3 postoperative visits.
Silver oxide–impregnated Sheehy-type tympanostomy tubes (Activent Antimicrobial Ventilation Tubes; Xomed Surgical Products, Jacksonville, Fla) were inserted either by or under the direct supervision of the senior author (R.N.H.) while patients were under general anesthesia. Intraoperative middle ear findings and the presence of blood at the myringotomy site were recorded for each ear. Patients with mucoid effusions were treated with topical sulfacetamide sodium–prednisolone acetate (Blephamide Ophthalmic; Allergan, Hormigueros, Puerto Rico). Subjects with purulent effusions or blood at the myringotomy site at surgery were treated with topical neomycin sulfate–polymyxin B sulfate–hydrocortisone (Cortisporin Otic Suspension; Burroughs Wellcome Co, Research Triangle Park, NC). Topical antibiotic eardrops were applied to the affected ear intraoperatively and continued for 5 days after tympanostomy tube placement. Placebo topical preparations were not used in unaffected ears.
All subjects were examined postoperatively at 1 week and 1, 3, 6, 9, and 12 months after surgery to assess for discharge and tube patency. Tube patency was evaluated by otoscopic visualization, tympanometry, and audiometry. Blocked tubes were treated by applying hydrogen peroxide if dried blood was noted to be obscuring the lumen, or by removal of crusted debris while viewing under the operating microscope. Parents were instructed to telephone if otorrhea developed between visits. Patients with otorrhea were treated with topical neomycin sulfate–polymyxin B sulfate–hydrocortisone; if no response was seen after 7 days, conventional oral antibiotic therapy was instituted. Cultures were obtained in refractory cases. Persistent otorrhea lasting beyond 60 days associated with hearing loss resulted in tympanostomy tube removal. Statistical analysis was performed with Fisher exact test (GraphPAD Instat; GraphPAD Software, San Diego, Calif). Significance was attributed to a P value less than
Six hundred thirty patients fulfilled the study criteria, in whom a total of 1254 silver oxide–impregnated tympanostomy tubes were inserted. Three hundred sixty-one patients were male (718 ears) and 269 patients were female (534 ears). The right ear was intubated in 628 cases and the left ear in 626 cases. The average patient age was 3.7±2.9 years (mean±SD). Three hundred nine patients (616 ears) were younger than 3 years and 321 patients (638 ears) were 3 years of age or older. Data were recorded independently for each ear. Of 1254 ears followed up, there were 7397 follow-up visits, for a mean of 5.9 visits per ear. Postoperative visits were not recorded after the tubes had extruded, and some patients did not complete all scheduled follow-up visits.
Operative findings at myringotomy are shown in Table 1. The overall incidence of otorrhea in the first postoperative week was 5.6%; the incidence after the first postoperative week was 1.2% (P<.001) (Table 2). The incidence of otorrhea after the first postoperative week did not differ significantly between 1, 3, 6, 9, and 12 months. There was no significant difference in the incidence of otorrhea at any time between sex groups. Within the first postoperative week, a significantly greater incidence of otorrhea was noted in patients with mucoid effusions at surgery (8.6%). Patients younger than 3 years had a significantly greater incidence of early postoperative otorrhea (7.8%), as did patients who were younger than 3 years and had mucoid effusions at surgery (15.2%) (Table 3 and Table 4). No significant difference was noted in the incidence of otorrhea after the first postoperative week between groups based on age or findings at surgery.
Tube extrusion occurred in 58 cases (4.6%). The mean±SD time to extrusion was 18.8±14.1 weeks. It occurred at 1 week in 1 ear, 1 month in 5, 3 months in 17, 6 months in 21, 9 months in 9, and 12 months in 5. Obstruction of the tube by dried blood or debris was seen infrequently (at 1 week in 5 ears, 1 month in 2, 3 months in 4, 9 months in 3, and 12 months in 1) and responded in all cases to dissolution or removal of the obstruction. There were no instances of persistent otorrhea beyond 60 days requiring tube removal, and no cases of cholesteatoma formation were observed during the study period.
The observed incidence of postoperative otorrhea was not significantly different from the incidence reported by Chole and Hubbell18 (Table 5). They reported an incidence of otorrhea that was significantly lower than that of matched controls, but not within the first week. We observed an incidence of otorrhea within the first postoperative week that was significantly higher than the late (>1 week) incidence and the overall incidence of postoperative otorrhea in this study. While this study was neither randomized nor controlled, our patient population was identical to the control and experimental population studied by Chole and Hubbell in their initial pilot study of silver oxide–impregnated tympanostomy tubes with respect to patient characteristics, preoperative intervention, operative findings, and follow-up.18 The addition of topical antibiotic prophylaxis to subjects with mucoid or purulent effusions or blood at the myringotomy site in this study did not significantly affect the incidence of early otorrhea compared with the pilot study population, who did not receive antibiotic prophylaxis.18
This study describes our results with the use of silver oxide–impregnated tympanostomy tubes in the treatment of chronic or recurrent otitis media. We observed a low incidence of postoperative otorrhea that was similar to the incidence reported by Chole and Hubbell18 in an identical patient population receiving silver oxide–impregnated tympanostomy tubes. In Chole and Hubbell's study, the incidence of postoperative otorrhea in the study group was significantly lower than the incidence seen in matched controls except in the first postoperative week, where the incidence of otorrhea was identical for both groups.18 We found an incidence of otorrhea in the first postoperative week that was similar to the incidence reported by Chole and Hubbell and significantly higher than the late and overall incidence. These results support Chole and Hubbell's conclusion that silver oxide–impregnated tympanostomy tubes are associated with a low incidence of postoperative otorrhea after the first postoperative week.18
Early postoperative otorrhea may result from external contamination at the time of tympanostomy tube insertion or from persistent preexisting infection of the middle ear space. Efforts to prevent surgical contamination by ear canal disinfection and the use of the "no-touch" surgical technique have not been shown to affect the incidence of early otorrhea.5,19- 21 An increased incidence of early postoperative otorrhea has been described in patients with blood at the myringotomy site, purulent or mucoid middle ear effusions, and young age.1,2,4,7,9,10,22 Blood at the myringotomy site is associated with an incidence of early otorrhea as high as 25.6%.10,22 However, the association between blood at the myringotomy site and early otorrhea has not been shown to be significant in this study and others, perhaps because of small patient numbers.9,10 The status of the middle ear at the time of surgery appears to be the most important predictor of early postoperative otorrhea. An increased incidence of early otorrhea has been reported for ears with purulent effusions (18%-33%)5,7,9,23 and mucoid effusions (18.6%-23%).9,10,23 Although purulent effusions have been reported to have the highest incidence of postoperative otorrhea, these represent the minority of ears that are seen with early otorrhea.2,3,5,9 We observed a greater incidence of early postoperative otorrhea in ears with purulent effusions, but this difference was not significant because of small numbers in this group. The presence of a mucoid effusion in this study was associated with more than one third of all cases of early otorrhea and fully 50% of all cases of early otorrhea in children younger than 3 years. The identification of bacterial DNA by polymerase chain reaction in otherwise culture-negative chronic middle ear effusions indicates that an active bacterial process may be present in a greater number of cases than previously suspected.24 These findings would suggest that antibiotic prophylaxis should be specifically targeted to these higher-risk subpopulations. However, in all studies that have investigated the use of topical prophylaxis, the duration of prophylaxis as well as the agent used has varied.2,4- 11 The use of topical prophylaxis remains an area of debate, and most clinical use is based on anecdotal experience.11
The selection of topical antibiotic prophylaxis in this population was based on the spectrum of activity of the agent as well as the type of middle ear effusion. Sulfacetamide is bacteriostatic against Haemophilus influenzae, Moraxella catarrhalis, and most pneumococci, the most common pathogens isolated in acute otitis media and chronic otitis media with effusion.10 However, the organisms most commonly associated with chronic suppurative otitis media, acute otitis externa, and infected surgical wounds are Staphylococcus aureus or Pseudomonas aeruginosa, neither of which is covered by the spectrum of activity of sulfacetamide.25 Neomycin sulfate–polymyxin B sulfate in combination is bactericidal against Pseudomonas and Staphylococcus species as well as most gram-negative organisms that are associated with chronic suppurative otitis media. Notable exceptions to its activity are Bacteroides fragilis, streptococci, and pneumococci. We chose sulfacetamide eardrops in cases of thick mucoid effusions at surgery on the basis of its spectrum in the setting of chronic otitis media with effusion and on the basis of previous work suggesting that it is precisely in the setting of thick mucoid middle ear effusions that prophylactic use of sulfacetamide drops may show benefit.10 In the setting of a suppurative effusion or blood at the myringotomy site, with a greater risk of surgical wound infection, the antibacterial spectrum of neomycin sulfate–polymyxin B sulfate is a more appropriate choice. The use of neomycin sulfate–polymyxin B sulfate for up to 3 days postoperatively has not been shown to result in a reduction in otorrhea in the setting of a thick middle ear effusion.7,23 However, a statistically significant benefit has been shown with a longer 5-day course in the subgroup of patients having thick middle ear fluid at surgery.9 The significantly greater incidence of early postoperative otorrhea in our patients with mucoid effusions, despite prophylaxis with sulfacetamide, suggests that management of mucoid effusions with sulfacetamide drops does not appreciably reduce the incidence of postoperative otorrhea. Otorrhea in these cases may be a result of infection with Staphylococcus, Pseudomonas, or other organisms that may be resident or introduced during myringotomy. Culture results of middle ear contents would be required to test this hypothesis.
Chole and Hubbell18 reported that the incorporation of silver oxide into tympanostomy tubes appears to decrease the long-term incidence of otorrhea. The known property that silver oxide confers to a tympanostomy tube is the ability to prevent adherence and colonization of selected bacteria rather than affecting an already established infection. Silver ions inhibit the growth of Staphylococcus, Streptococcus, Escherichia coli, and Pseudomonas.13,16,17 Persistent infection present at the time of tympanostomy tube insertion would thus not be affected in the immediate postoperative period by the presence of silver oxide in the tympanostomy tube. It is likely that the higher incidence of early postoperative otorrhea seen in children younger than 3 years and in patients with thick effusions results from factors that increase the risk of otorrhea in these patients irrespective of the type of tube inserted.
While silver oxide–impregnated tympanostomy tubes appear to reduce the overall incidence of postoperative otorrhea, their use does not significantly affect the incidence of otorrhea during the first postoperative week. In high-risk subpopulations previously identified as having an increased incidence of early postoperative otorrhea based on effusion type and age, the incidence of early otorrhea remained significantly elevated. The prophylactic use of topical antibiotic eardrops was neither randomized nor blinded in this study. However, despite prophylaxis in selected high-risk cases, the incidence of otorrhea in the first postoperative week remained high. Future studies appear warranted to determine the effectiveness of topical antibiotic prophylaxis on early otorrhea, and to elucidate the underlying pathophysiological mechanism in refractory cases to set guidelines for management.
Accepted for publication October 27, 1998.
Presented at the 13th annual meeting of the American Society of Pediatric Otolaryngology, Palm Beach, Fla, May 14, 1998.
Reprints: Richard N. Hubbell, MD, Division of Otolaryngology–Head and Neck Surgery, Department of Surgery, University Health Center Campus, 1 S Prospect St, Burlington, VT 05401.