Ann oi»t 88 :1979
ANTIBODY ACTIVITY IN OTITIS MEDIA WITH EFFUSION DANIEL M. LEWIS, PhD
JAMES L. SCJffiAM, BS
HERBERT G. BIRCK, MD
DAVID J. LIM, MD COLUMBUS, OHIO
Bacteria were isolated from a high percentage of the effusions from patients with otitis media with effusion (OME, serous otitis media). In an attempt to determine if the isolated bacteria were involved in the disease process, we analyzed the serum and effusion of 25 OME patients for the presence of antibacterial antibodies by the indirect immunofluorescence antibody method. Specific antibody activity was detected in 20 of 25 effusions (80%) and 19 of 22 sera (86%). IgG antibodies were the most frequently found class of antibodies in both sera and effusions, but IgA antibodies were detected more frequently in the effusions than in the sera. Hemophilus influenzae, Streptococcus pneumoniae, and diphtheroids were the most frequently isolated organisms, and antibody activity to all bacterial species isolated was detected. The results support the concept that the isolated bacteria are not contaminants but are actively involved in the disease process.
Although otitis media with effusion (OME, serous otitis media, secretory otitis media) has been the subject of numerous investigations, the pathogenesis of this disease is poorly understood. Middle ear effusions from patients with OME were thought to be bacteriologically sterile.'-" but this concept has been refuted by the demonstration that bacteria can be isolated from a high percentage of the effusions."' However, whether the bacteria isolated from the effusions were involved in the pathogenesis of OME or were merely contaminants has not been clearly established. In an attempt to better define the role of bacteria in OME we have assayed a series of serum and effusion samples from patients with OME for the presence of specific antibodies to the isolated bacteria. The presence of such antibodies in effusions is generally accepted as evidence that the bacteria are involved in the disease process." METHODS AND MATERIALS The effusions were collected from children who were diagnosed as having chronic OME by established criteria" and who were undergoing therapeutic myringotomy with tubal insertion. The patients ranged in age from IJ~ to II years, with a mean age of 5.2 years. Ef-
fusions in which there was gross evidence of blood contamination were excluded from the studies. The effusions were processed for bacteriologic, cytologic and biochemical studies as previously described," and those effusions from which, on primary culture, bacteria were isolated in high numbers (greater than 30,000 colonies/ml) and in pure culture were selected for further study. Antibody to the isolated bacteria was detected by a slight modification of the indirect fluorescent antibody technique (IFA) described by Sloyer et al.": Briefly, the isolated bacteria were cultured overnight in an appropriate broth, and the bacteria were washed twice with normal saline (0.85% NaC!) and once with sterile distilled water. Bacteria were then suspended in a small volume of distilled water, and 0.05 ml of the suspension was placed on a microscope slide and allowed to air dry. If not immediately used, the slides were stored at 4 C until needed. The smears were then exposed to either effusion or serum of the patient from whom the bacteria had been isolated for 45 minutes at 37 C. The slides were then washed three times with saline, after which fluorescein-conjugated antisera to either human IgG, IgA or IgM was placed on the smears and allowed to react for 30 minutes at 37 C. Following an additional three washes, the final wash being distilled water, the slides were allowed to air dry and were examined by fluorescence microscope ( Zeiss). A saline control was included for each bacterial isolate to control for nonspecific fluorescence, and a positive serum control" was included in each experiment.
·Positive serum control was provided by Dr. John Sloyer, Huntsville, AL. From the Otolog-ical Research Laboratories. Department of Otolaryngology. Ohio State University College of Medicine. and the Department of Otolaryngolog-y. The Children's Hospital, Columbus. Ohio. This study was supported In part by grants from NIH-NINCDS (NS08854-08) and The Deafness Research Foundation.
392
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
393
ANTIBODY ACTIVITY IN OME
TABLE 1. BACTERIAL SPECIES ISOLATED FROM MIDDLE EAR EFFUSIONS ---~------
No. of Isolates
Bacterial Species
12
Hemophilus influenzae Streptococcus pneumoniae Diphtheroids Staphylococcus aureus Staphylococcus epidennidis Neisseria species Pseudomonas species
s 5 2 1 1 1 27
Total
RESULTS
From 92 effusions collected between January and June of 1977, 27 samples were selected for study by the IFA procedure. The bacterial species isolated are shown in Table 1. The two Staphylococcus aureus isolates were found to react nonspecifically with the fluorescein-labeled goat antisera used in this study. Hence, class-specific antibodies to these organisms could not be detected, and these two samples were therefore excluded from further study. Effusions were classified as either serous or mucoid at the time of surgery, and of the 25 effusions used in this study 13 were considered serous and 12 mucoid. Cytologic analysis of the effusions showed that neutrophils were the dominant cell type in all 25 samples, and therefore these samples could be classified as neutrophilic. The 25 effusions were obtained from 22 patients, i.e., 3 patients had bilateral effusions. When the effusion samples and their paired serum samples were assayed by the IFA procedure, it was found that 20 of the 25 effusions (80%) and 19 of the
22 serum samples (86%) showed antibody activity. The results are summarized in Table 2. As can be seen by examining this table, IgG antibodies were the most frequently found class of antibody in both the effusions and the sera. When effusions were compared with their matched serum samples, it was found that there was a good correlation between the effusion and serum antibodies of the IgG or IgM classes (Table 3) in that there was only one case in which IgG antibodies were found in the effusion only and no cases in which IgM antibodies were found only in the effusion. IgA antibodies were detected more frequently in the effusions than in the sera. There were three effusions in which IgA antibodies were present only in the effusion, suggesting that local IgA production might have occurred. However, no attempt was made to determine if these were secretory IgA antibodies or not. There were only two patients in whom no antibody activity could be detected in either the serum or effusion. The results were next analyzed with respect to the species of bacteria isolated. Results for the three most frequently isolated organisms are shown in Table 4. Hemophilus influenzae was the only organism for which IgM antibodies were found. The incidence of antibodies to S. pneumoniae and diphtheroids was similar except that serum IgA antibodies to diphtheroids were not seen. Although not shown in the table, one isolate each of Staphylococcus epidermidis, Pseudomonas, and Neisseria were also examined. IgG antibodies to the S. epidermidis were found in the serum but not the effusion, while the Pseudomonas and Neisseria isolates showed reactions
TABLE 2. IMMUNOGLOBULIN CLASS OF ANTIBACTERIAL ANTIBODIES IN EFFUSIONS AND SERA OF PATIENTS WITH OTITIS MEDIA WITH EFFUSION Fluid Type (n) Effusion (25) Serum (22)
Antibody Activity"
20"" (14) 19 (12)
IgG 17 (11) 18 (11)
Immunoglobulin Class IgA IgM 9 (7) 7 (7)
7 (7)
7 (7)
n - Number of samples tested. "Antibody activity detected In one or more of the immunoglobulin classes tested. ""Number of positive samples. () - Number of samples from whic-h H. ill!lllclIoae or S. pneumoniae were isolated.
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
394
LEWIS ET AL
TABLE 3. CORRELATION OF ANTIBODY ACTIVITY BETWEEN EFFUSION AND SERUM SAMPLES Antibody Activity in
Antibody Activity
19 1 3 2
Effusion and serum Effusion only Serum only Neither serum nor effusion
(16) (1) (3) (2)
IgG
16 1 5 3
Immunoglobulin Class IgA
(14) (1) (5) (2)
7 3 1 14
(6) (3) (1 ) ( 12)
IgM
7 (5) o (0) 2 (2) 16 (15)
Number of patients is shown in parentheses. Three patients had bilateral effusions.
for IgG antibodies in both serum and effusion. No other antibodies were detected for these three isolates. Finally, the data were analyzed with respect to the ages of the patients. It was found that there were two patients in our study under the age of two and that these two patients were the only subjects who showed no antibody activity in either their effusions or sera. Both of the patients had potential pathogens isolated from their effusions, one H. influenzae and the other an S. pneumoniae, which should have provided sufficient antigenic stimulation for antibody production. The remaining patients ranged from 3 to 11 years of age, and no age-associated pattern of response could be discerned. However, the number of samples studied was too small really to permit adequate analysis of age-associated responses. DISCUSSION
The results of this study show that a significant proportion of the culturepositive effusions from children with OME possess an antibody to the bacteria isolated from them. This finding can be considered as direct support for the idea that the bacteria isolated are involved in the disease process. When
one considers only patients over two years of age, 87% of the effusions were positive for antibody activity. The lack of antibody activity in the two patients who were under two years of age may simply reflect the immunologic immaturity of these patients, as other workers have noted poor antibody responses in this age group," The finding of IgM antibodies to H. influenzae is particularly interesting. Since IgM antibodies are associated with a primary immune response, and in a normal situation do not persist in the serum for a long period of time, this finding suggests that these patients were actively infected by the Hemophilus. Perhaps the most difficult result to explain is the lack of IgM antibodies to the other bacterial species, particularly the S. pneumoniae. Perhaps pneumococcal infections were of a longer duration and the IgM phase of the immune response had passed. That is to say, the persistent pneumococcal infections might preferentially stimulate IgG and IgA antibodies instead of IgM. This question will require further studies on a larger number of patients before we can ascertain the significance of the lack of IgM antibodies to pneumococci in our study.
TABLE 4. ANTIBODY ACTIVITY IN MEE WITH RESPECT TO BACTERIAL SPECIES ISOLATED
Bacterial species Hemophilus lnjluenzae Streptococcus pneumoniae Diphtheroids
19A
IgG
Antibody Activity
IgM
E
S
E
S
E
S
E
S
10/12° 4/5 4/5
9/10 3/4 4/5
8/12 3/5 4/5
9/10 2/4 4/5
5/12 2/5 2/5
5/10 2/4 0/5
7/12 0/5 0/5
7/10 0/4 0/5
E - Effusion, S - Serum. 'Number of IFA positive samples over total number examined,
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
ANTIBODY ACTIVITY IN OME
The high percentage of positive reactions to diphtheroids is somewhat surprising. Since diphtheroids are part of the normal flora of the upper respiratory tract, knowledge of the incidence of serum antibodies to diphtheroids in a normal age match population is necessary before we can determine the significance of this finding. However, the presence of IgA antibodies in the effusion but not the serum in two cases (Table 4) suggests that these organisms are infecting the middle ear. It is possible that diphtheroids are opportunistic pathogens which can invade the middle ear when local defense mechanisms are altered or weakened. The pathogenic potential of diphtheroids in OME is still an open question. In studies on children with acute otitis media, Sloyer et a16 , 7 reported finding antibacterial antibodies in middle ear fluid. In patients with pneumococcal otitis media, approximately 26% of the middle ear fluid samples had antibody activity, while in patients from whom H. influenzae were isolated nearly 76% of the samples possessed an antibody activity. In both studies it was found that the antibody activity could be demonstrated in all three immunoglobulin classes but primarily in the IgG and IgA classes in patients infected with H emophilus. The greater frequency with which we detected antibody activity probably reflects the fact that we were examining patients with chronic disease, while their study considered only those
395
patients with acute otitis media. In a subsequent study by the same authors" a significant correlation between the presence of specific antibody activity in the middle ear fluid and clinical recovery from infection was noted. That study demonstrated that an antibody response can help the middle ear space recover from infection and presumably aid in the defense of the middle ear. We do not have information available as yet to determine if the presence of antibacterial antibodies in effusions is associated with a shortening of clinical recovery for patients with OME, but we do plan to address this question in future studies. The finding of specific antibody activity in middle ear effusions of patients with OME and the finding that culturepositive effusions usually have a neutrophilic infiltration'" indicate that bacterial infections do occur in a high percentage of the cases of this disease. Thus, it appears that certain cases of OME may be viewed as mild chronic infections. The question remains as to why patients with OME failed to clear the infection or, conversely, why they did not develop signs and symptoms of purulent otitis media. A number of factors may be involved, ranging from the pathogenic potential of the bacteria to the nature of the immune reaction, and more detailed analysis of the fluids is needed before which factor (s) causing the pathology seen in this disease can be ascertained.
REFERENCES 1. Siirala U: The problem of sterile otitis media. Prac OtorhinolaryngoI19:159-169, 1956 2. Harcourt FL, Brown AK: Hydrotympanum (secretory otitis media). Arch Oto:aryngol 57: 12-21, 1953 3. Liu YS, Lim DJ, Lang RW, et al: Chronic middle ear effusions: Immunochemical and bacteriological investigations. Arch Otolaryngol 101:278-286, 1975
4. Healy BG, Teele DW: The microbiology of chronic middle ear effusions in children. Laryngoscope 87: 1472-1478, 1977 5. Wilson GS, Miles A: Topley and Wilson's Principles of Bacteriology, Virology and Immunity, ed 6. Baltimore, Williams & Wilkins, 1975, p 1273
6. Sloyer JL [r, Howie VM, Ploussard JH, et al: Immune response to acute otitis media in children. I. Serotypes isolated and serum and middle ear fluid antibody in pneumococcal otitis media. Infec Immun 9: 1028-1032, 1974 7. Sloyer JL Ir, Cate CC, Howie VM, et al: The immune response to acute otitis media in children. II. Serum and middle ear fluid antibody in otitis media due to Haemophilus iniluenzae. J Infect Dis 132:685-688, 1975 8. South MA: Lack of immune response to Hemophilus influenzae: Immune paralysis or immaturity? J Pediatr 80:348-350, 1972 9. Sloyer JL Jr, Howie VM, Ploussard JH, et al: Immune response to acute otitis media:
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
396
LEWIS ET AL
Association between middle ear fluid antibody and the clearing of clinical infection. J Clin Microbiol 4:306-308, 1976 10. Lim DJ, Lewis DM, Schram JL: Cel-
lular aspects of otitis media with effusion: A quantitative study, in Lim DJ (ed): Abstracts of the Midwinter Research Meeting, Association for Research in Otolaryngology 33:1978
ACKNOWLEDGMENT - The authors wish to thank Nancy Sally, Katherine Adamson, and Lee Lytle and the staff of the Department of Anesthesia at The Children's Hospital for their invaluable assistance. REPRINTS - David J. Lim, MD, Otological Research Laboratories, 4331 University Hospitals Clinic, 456 Clinic Drive, Columbus, Ohio 43210.
IV INTERNATIONAL SYMPOSIUM ON ACOUSTIC IMPEDANCE MEASUREMENTS The IV International Symposium on Acoustic Impedance Measurements will be held September 26-28, 1979 in Lisbon, Portugal. For further information write: Organizing Secretariat, Clinica Fono-Audiologica, Rua Conde Redondo, 119-3°-Lisboa, Portugal.
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
ANTIBODY ACTIVITY IN OTITIS MEDIA WITH EFFUSION DANIEL M. LEWIS, PhD
JAMES L. SCJffiAM, BS
HERBERT G. BIRCK, MD
DAVID J. LIM, MD COLUMBUS, OHIO
Bacteria were isolated from a high percentage of the effusions from patients with otitis media with effusion (OME, serous otitis media). In an attempt to determine if the isolated bacteria were involved in the disease process, we analyzed the serum and effusion of 25 OME patients for the presence of antibacterial antibodies by the indirect immunofluorescence antibody method. Specific antibody activity was detected in 20 of 25 effusions (80%) and 19 of 22 sera (86%). IgG antibodies were the most frequently found class of antibodies in both sera and effusions, but IgA antibodies were detected more frequently in the effusions than in the sera. Hemophilus influenzae, Streptococcus pneumoniae, and diphtheroids were the most frequently isolated organisms, and antibody activity to all bacterial species isolated was detected. The results support the concept that the isolated bacteria are not contaminants but are actively involved in the disease process.
Although otitis media with effusion (OME, serous otitis media, secretory otitis media) has been the subject of numerous investigations, the pathogenesis of this disease is poorly understood. Middle ear effusions from patients with OME were thought to be bacteriologically sterile.'-" but this concept has been refuted by the demonstration that bacteria can be isolated from a high percentage of the effusions."' However, whether the bacteria isolated from the effusions were involved in the pathogenesis of OME or were merely contaminants has not been clearly established. In an attempt to better define the role of bacteria in OME we have assayed a series of serum and effusion samples from patients with OME for the presence of specific antibodies to the isolated bacteria. The presence of such antibodies in effusions is generally accepted as evidence that the bacteria are involved in the disease process." METHODS AND MATERIALS The effusions were collected from children who were diagnosed as having chronic OME by established criteria" and who were undergoing therapeutic myringotomy with tubal insertion. The patients ranged in age from IJ~ to II years, with a mean age of 5.2 years. Ef-
fusions in which there was gross evidence of blood contamination were excluded from the studies. The effusions were processed for bacteriologic, cytologic and biochemical studies as previously described," and those effusions from which, on primary culture, bacteria were isolated in high numbers (greater than 30,000 colonies/ml) and in pure culture were selected for further study. Antibody to the isolated bacteria was detected by a slight modification of the indirect fluorescent antibody technique (IFA) described by Sloyer et al.": Briefly, the isolated bacteria were cultured overnight in an appropriate broth, and the bacteria were washed twice with normal saline (0.85% NaC!) and once with sterile distilled water. Bacteria were then suspended in a small volume of distilled water, and 0.05 ml of the suspension was placed on a microscope slide and allowed to air dry. If not immediately used, the slides were stored at 4 C until needed. The smears were then exposed to either effusion or serum of the patient from whom the bacteria had been isolated for 45 minutes at 37 C. The slides were then washed three times with saline, after which fluorescein-conjugated antisera to either human IgG, IgA or IgM was placed on the smears and allowed to react for 30 minutes at 37 C. Following an additional three washes, the final wash being distilled water, the slides were allowed to air dry and were examined by fluorescence microscope ( Zeiss). A saline control was included for each bacterial isolate to control for nonspecific fluorescence, and a positive serum control" was included in each experiment.
·Positive serum control was provided by Dr. John Sloyer, Huntsville, AL. From the Otolog-ical Research Laboratories. Department of Otolaryngology. Ohio State University College of Medicine. and the Department of Otolaryngolog-y. The Children's Hospital, Columbus. Ohio. This study was supported In part by grants from NIH-NINCDS (NS08854-08) and The Deafness Research Foundation.
392
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
393
ANTIBODY ACTIVITY IN OME
TABLE 1. BACTERIAL SPECIES ISOLATED FROM MIDDLE EAR EFFUSIONS ---~------
No. of Isolates
Bacterial Species
12
Hemophilus influenzae Streptococcus pneumoniae Diphtheroids Staphylococcus aureus Staphylococcus epidennidis Neisseria species Pseudomonas species
s 5 2 1 1 1 27
Total
RESULTS
From 92 effusions collected between January and June of 1977, 27 samples were selected for study by the IFA procedure. The bacterial species isolated are shown in Table 1. The two Staphylococcus aureus isolates were found to react nonspecifically with the fluorescein-labeled goat antisera used in this study. Hence, class-specific antibodies to these organisms could not be detected, and these two samples were therefore excluded from further study. Effusions were classified as either serous or mucoid at the time of surgery, and of the 25 effusions used in this study 13 were considered serous and 12 mucoid. Cytologic analysis of the effusions showed that neutrophils were the dominant cell type in all 25 samples, and therefore these samples could be classified as neutrophilic. The 25 effusions were obtained from 22 patients, i.e., 3 patients had bilateral effusions. When the effusion samples and their paired serum samples were assayed by the IFA procedure, it was found that 20 of the 25 effusions (80%) and 19 of the
22 serum samples (86%) showed antibody activity. The results are summarized in Table 2. As can be seen by examining this table, IgG antibodies were the most frequently found class of antibody in both the effusions and the sera. When effusions were compared with their matched serum samples, it was found that there was a good correlation between the effusion and serum antibodies of the IgG or IgM classes (Table 3) in that there was only one case in which IgG antibodies were found in the effusion only and no cases in which IgM antibodies were found only in the effusion. IgA antibodies were detected more frequently in the effusions than in the sera. There were three effusions in which IgA antibodies were present only in the effusion, suggesting that local IgA production might have occurred. However, no attempt was made to determine if these were secretory IgA antibodies or not. There were only two patients in whom no antibody activity could be detected in either the serum or effusion. The results were next analyzed with respect to the species of bacteria isolated. Results for the three most frequently isolated organisms are shown in Table 4. Hemophilus influenzae was the only organism for which IgM antibodies were found. The incidence of antibodies to S. pneumoniae and diphtheroids was similar except that serum IgA antibodies to diphtheroids were not seen. Although not shown in the table, one isolate each of Staphylococcus epidermidis, Pseudomonas, and Neisseria were also examined. IgG antibodies to the S. epidermidis were found in the serum but not the effusion, while the Pseudomonas and Neisseria isolates showed reactions
TABLE 2. IMMUNOGLOBULIN CLASS OF ANTIBACTERIAL ANTIBODIES IN EFFUSIONS AND SERA OF PATIENTS WITH OTITIS MEDIA WITH EFFUSION Fluid Type (n) Effusion (25) Serum (22)
Antibody Activity"
20"" (14) 19 (12)
IgG 17 (11) 18 (11)
Immunoglobulin Class IgA IgM 9 (7) 7 (7)
7 (7)
7 (7)
n - Number of samples tested. "Antibody activity detected In one or more of the immunoglobulin classes tested. ""Number of positive samples. () - Number of samples from whic-h H. ill!lllclIoae or S. pneumoniae were isolated.
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
394
LEWIS ET AL
TABLE 3. CORRELATION OF ANTIBODY ACTIVITY BETWEEN EFFUSION AND SERUM SAMPLES Antibody Activity in
Antibody Activity
19 1 3 2
Effusion and serum Effusion only Serum only Neither serum nor effusion
(16) (1) (3) (2)
IgG
16 1 5 3
Immunoglobulin Class IgA
(14) (1) (5) (2)
7 3 1 14
(6) (3) (1 ) ( 12)
IgM
7 (5) o (0) 2 (2) 16 (15)
Number of patients is shown in parentheses. Three patients had bilateral effusions.
for IgG antibodies in both serum and effusion. No other antibodies were detected for these three isolates. Finally, the data were analyzed with respect to the ages of the patients. It was found that there were two patients in our study under the age of two and that these two patients were the only subjects who showed no antibody activity in either their effusions or sera. Both of the patients had potential pathogens isolated from their effusions, one H. influenzae and the other an S. pneumoniae, which should have provided sufficient antigenic stimulation for antibody production. The remaining patients ranged from 3 to 11 years of age, and no age-associated pattern of response could be discerned. However, the number of samples studied was too small really to permit adequate analysis of age-associated responses. DISCUSSION
The results of this study show that a significant proportion of the culturepositive effusions from children with OME possess an antibody to the bacteria isolated from them. This finding can be considered as direct support for the idea that the bacteria isolated are involved in the disease process. When
one considers only patients over two years of age, 87% of the effusions were positive for antibody activity. The lack of antibody activity in the two patients who were under two years of age may simply reflect the immunologic immaturity of these patients, as other workers have noted poor antibody responses in this age group," The finding of IgM antibodies to H. influenzae is particularly interesting. Since IgM antibodies are associated with a primary immune response, and in a normal situation do not persist in the serum for a long period of time, this finding suggests that these patients were actively infected by the Hemophilus. Perhaps the most difficult result to explain is the lack of IgM antibodies to the other bacterial species, particularly the S. pneumoniae. Perhaps pneumococcal infections were of a longer duration and the IgM phase of the immune response had passed. That is to say, the persistent pneumococcal infections might preferentially stimulate IgG and IgA antibodies instead of IgM. This question will require further studies on a larger number of patients before we can ascertain the significance of the lack of IgM antibodies to pneumococci in our study.
TABLE 4. ANTIBODY ACTIVITY IN MEE WITH RESPECT TO BACTERIAL SPECIES ISOLATED
Bacterial species Hemophilus lnjluenzae Streptococcus pneumoniae Diphtheroids
19A
IgG
Antibody Activity
IgM
E
S
E
S
E
S
E
S
10/12° 4/5 4/5
9/10 3/4 4/5
8/12 3/5 4/5
9/10 2/4 4/5
5/12 2/5 2/5
5/10 2/4 0/5
7/12 0/5 0/5
7/10 0/4 0/5
E - Effusion, S - Serum. 'Number of IFA positive samples over total number examined,
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
ANTIBODY ACTIVITY IN OME
The high percentage of positive reactions to diphtheroids is somewhat surprising. Since diphtheroids are part of the normal flora of the upper respiratory tract, knowledge of the incidence of serum antibodies to diphtheroids in a normal age match population is necessary before we can determine the significance of this finding. However, the presence of IgA antibodies in the effusion but not the serum in two cases (Table 4) suggests that these organisms are infecting the middle ear. It is possible that diphtheroids are opportunistic pathogens which can invade the middle ear when local defense mechanisms are altered or weakened. The pathogenic potential of diphtheroids in OME is still an open question. In studies on children with acute otitis media, Sloyer et a16 , 7 reported finding antibacterial antibodies in middle ear fluid. In patients with pneumococcal otitis media, approximately 26% of the middle ear fluid samples had antibody activity, while in patients from whom H. influenzae were isolated nearly 76% of the samples possessed an antibody activity. In both studies it was found that the antibody activity could be demonstrated in all three immunoglobulin classes but primarily in the IgG and IgA classes in patients infected with H emophilus. The greater frequency with which we detected antibody activity probably reflects the fact that we were examining patients with chronic disease, while their study considered only those
395
patients with acute otitis media. In a subsequent study by the same authors" a significant correlation between the presence of specific antibody activity in the middle ear fluid and clinical recovery from infection was noted. That study demonstrated that an antibody response can help the middle ear space recover from infection and presumably aid in the defense of the middle ear. We do not have information available as yet to determine if the presence of antibacterial antibodies in effusions is associated with a shortening of clinical recovery for patients with OME, but we do plan to address this question in future studies. The finding of specific antibody activity in middle ear effusions of patients with OME and the finding that culturepositive effusions usually have a neutrophilic infiltration'" indicate that bacterial infections do occur in a high percentage of the cases of this disease. Thus, it appears that certain cases of OME may be viewed as mild chronic infections. The question remains as to why patients with OME failed to clear the infection or, conversely, why they did not develop signs and symptoms of purulent otitis media. A number of factors may be involved, ranging from the pathogenic potential of the bacteria to the nature of the immune reaction, and more detailed analysis of the fluids is needed before which factor (s) causing the pathology seen in this disease can be ascertained.
REFERENCES 1. Siirala U: The problem of sterile otitis media. Prac OtorhinolaryngoI19:159-169, 1956 2. Harcourt FL, Brown AK: Hydrotympanum (secretory otitis media). Arch Oto:aryngol 57: 12-21, 1953 3. Liu YS, Lim DJ, Lang RW, et al: Chronic middle ear effusions: Immunochemical and bacteriological investigations. Arch Otolaryngol 101:278-286, 1975
4. Healy BG, Teele DW: The microbiology of chronic middle ear effusions in children. Laryngoscope 87: 1472-1478, 1977 5. Wilson GS, Miles A: Topley and Wilson's Principles of Bacteriology, Virology and Immunity, ed 6. Baltimore, Williams & Wilkins, 1975, p 1273
6. Sloyer JL [r, Howie VM, Ploussard JH, et al: Immune response to acute otitis media in children. I. Serotypes isolated and serum and middle ear fluid antibody in pneumococcal otitis media. Infec Immun 9: 1028-1032, 1974 7. Sloyer JL Ir, Cate CC, Howie VM, et al: The immune response to acute otitis media in children. II. Serum and middle ear fluid antibody in otitis media due to Haemophilus iniluenzae. J Infect Dis 132:685-688, 1975 8. South MA: Lack of immune response to Hemophilus influenzae: Immune paralysis or immaturity? J Pediatr 80:348-350, 1972 9. Sloyer JL Jr, Howie VM, Ploussard JH, et al: Immune response to acute otitis media:
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
396
LEWIS ET AL
Association between middle ear fluid antibody and the clearing of clinical infection. J Clin Microbiol 4:306-308, 1976 10. Lim DJ, Lewis DM, Schram JL: Cel-
lular aspects of otitis media with effusion: A quantitative study, in Lim DJ (ed): Abstracts of the Midwinter Research Meeting, Association for Research in Otolaryngology 33:1978
ACKNOWLEDGMENT - The authors wish to thank Nancy Sally, Katherine Adamson, and Lee Lytle and the staff of the Department of Anesthesia at The Children's Hospital for their invaluable assistance. REPRINTS - David J. Lim, MD, Otological Research Laboratories, 4331 University Hospitals Clinic, 456 Clinic Drive, Columbus, Ohio 43210.
IV INTERNATIONAL SYMPOSIUM ON ACOUSTIC IMPEDANCE MEASUREMENTS The IV International Symposium on Acoustic Impedance Measurements will be held September 26-28, 1979 in Lisbon, Portugal. For further information write: Organizing Secretariat, Clinica Fono-Audiologica, Rua Conde Redondo, 119-3°-Lisboa, Portugal.
Downloaded from aor.sagepub.com at Purdue University Libraries on July 9, 2015
