‘Laryngoscope 86 Deemer 1986 COMPARISON OF THE MICROBIOLOGY OF RECURRENT TONSILLITIS BETWEEN CHILDREN AND ADULTS*+ Inzitak BRoox, MD, MSe Bethesda, MD PERRY A. Foore, JR, MD Gainesville, FL ABSTRACT suffering from st Bacteroides melaninogenteus group 0.8 of the 23 tonsils removed from adults {1 2 isolates per patient) (P=0,04). These findings suggest th In children, The differences in the tonillae flora may be due to Ue ee ‘doles and the changes in tonsllarUnsue that occur In thi age grou es per pati Recurrent tonsilitis refractory to penicillin thera- py presents a major medical problem in all age groups. The inability of penicillins to eradicate the infection has also become a great clinical concern. ‘This has been attributed to the emergence of penicil- lin resistant strains among the mouth flora.' Previous work related to the microbiology of re- curent tonsillitis was aimed primarily at the detec: tion of aerobic pathogenic microorganisms such as Group A beta-hemolytic streptococci (GABHS) or Staphylococcus aureus. Although anaerobes were isolated from peritonsillar and retropharyngeal ab- scesses? and actinomyces were frequently recovered from tonsillar tissue,’ the role of anaerobic microor- ganisms in this infection was rarely studied. Furthermore, none of the previous studies com- pared the bacterial findings in the core of tonsils in young children to those of adults. Since tonsillitis may be due to different pathogens in different age groups, we compared in this study the bacterial iso- lates recovered from children to those recovered from adults. All of these patients underwent elective tonsillectomies because of recurrent tonsillitis re- fractory to penicillin therapy. PATIENTS AND METHODS "The tonsils studied were removed from 48 patlnts (25 children and 29 adults) admitted for elective Lonsillectomies, between ‘Tune 1980 and September 1984, Children's ages ranged from 3 Years to 1d years (mean 7 years) and adults’ ages anged from 18 Years to 96 years (mean 25 years) 32 wore male and 16 female. ‘They were selected at random from the total population of pe tients admitted for tonsillectomy for recurtont episodes of tonsil ~ Sri th Departments of Otolaryngology, Uniformed Services Univer tity ofthe Honleh Selecan, Betbead, MD. and th Alachua Genera! a ‘Nor Flra Reglonal Hopital, Gainesville, FL- “Tis opinions and arsrtions contained brea arth private onescf the srliars and reno tobe constr ofl or retloting ia vews ofthe Navy Departnat or the Novel Soviet la [Bator Note This Manueript was aecapted fr publication May 15, ‘Send Reprint Reuss ta tea Brook, MD, Armed Forces Radobioo ay Reveach Insite Bathondn, MD 208149148. 1385 in children (8. per tonsil) The difference between these groups was due to a eygiia were were recovered in ry ratein adults of imilarilnens. More bacter tea por adult, O8 per chil) and Baefevoldes fraps group (0 per adull, 0.2 per chil). Co lated in neven (28%) children ap compared to thee lolation in one (4% adult. Mare .tamase producing organism (BLPO) pe tonal were recovered in adul ‘Forty-three BLPO were detected in 21 (01%) i} as compared to 31 isolate in 16 (645) of the 25 tonsils removed from the elllogy of recurrent consis in adults may difer from that tol many tore courses of antimicrobials given over the years to P lar pharyngitis, Patients suffered from recurrent tonsilitis, de fined as a disease of at least 2 years duration, with atleast three episodes of inflammation of th tonsils within the last year. The ‘duration of the patient's symptoms ranged from 2 Lo 5 years {average 2.8 years), They recived multiple courses of antibiotics, bbutnone received any antimicrobial Uherapy for atleast 1 month prior to surgery Attempts to Isolate GABHS wore made in most of the epi sodes of tonailitis in these patints. Cultures for GABHS were ‘obtained using a sterile eotton swab aystem (Cultarette, Marion Scientific Corporation, Rockford, TL}. In the adult group, GABHS were recovered in th last 2 years ofthe study in 8 of the 23 eases: once in 6 patios and bwice in 2 patients. All ton- sils from the adulls were hypertzophie with evidence of scarring, fand most had "cheeay.”foulsmeling material in the crypts From the children, GABHS were recovered inthe last 2 years ofthe study in 21 ofthe 25 eases. In 2 cases GABHS were recov fered once, in eases they wore recovered twice, and In 15 m han three times. All tonsils from children were hypertrophic ‘but they had less scarring than tonsils from adults and only a few were fout-amelling. Immediately after excision, one ofthe tonsils was placed in a sterile Petr dish and held by freops. One sid of the tonsil was feauterized with a heated scalpel and an incision was made through that area, cutting the tonsil in half. he core ofthe ton sil was ewabbed with a stelle cotton-tipped applicator that was {introduced into an anaerobie transport media [Port-A-Cul Swab System, BBL, Cockeysville, MD]. This system of transportation ‘of specimen was chosen because ofthe data that shows tho of ‘acy ofthis mode of transportation in preserving aerobic and an Serobic bacteria qualitaively and quantitivey for at least 48 hours." Quality control studios done prior to the study domo strated the survival of organisms up to 48 hours. The time be ‘ween removal af the tonsils and inoculation of the spoeiman was hover in excess of 24 hour, Specimens were inoculated on 59% hoop blood, chocolate, and MaeContey's ager plates fr aerobic land facultative organisms, The plates wore incubated at 37° C ‘erobicaly (MacConkey's) or undor 8% COs and examined at 24 find 48 houre, Por anaerobes, the material was plated on prore- ‘eed media: vitemin Ki enriched brucellablood-agar plato, an- teroble blood ager plate containing kanamycin and vancomycin, fanaerobie blood plate containing phenylethyl alcohol, and en riched thioglyeniate broth (containing heroin and vitaimin Ki) "The anaorobie plates and thigglycolate broth were incubated in GesPakejers (BBL, Cockeysville, MD) and examined at 48 and 96 hours. Anaerabos were Identified by tackniques previously de. Seribed "Aerobie bacteria were identified using conventional Inethode.* Using the ebromogenie cephalosporin enalog 81/12 rmethodology,* beta lactamase activity was determined on all r+1386 BROOK AND FOOTD: MICROBIOLOGY OF RECURRENT TONSILLITIS TABLET. Organisms Isolated in 48 Bxcised Tonsis from 5 Children and “Adults with Recurrent Tons of fol Siren ‘Rorobie and Taeultative Grann-positive cocei ‘S pneumoniae 2 7 Alphachem. streptococe! 20 18 Gamma hemolytic streptococci 15 4 Group Abeta hem. streptococci 7 1 Group B beta hem, streptococci 2 Group C beta hem. streptococci 2 1 Scaureus nan 10 Gor S. epidermidis 4 2 Gram-nogative cove! ‘Branhamella catarrhalis 13( 188) Gram-positive baci ‘Lactobacillus 2p 5 Diphtheroia sp 7 3 Gram-negative baci 1H. influenzae type B HE, parainfluenzac E corrodens E.coll ‘Total Tora Anaerobic ‘Gram-positive cocci Peptococcus sp 13 “4 Peptostreptococeus sp 5 1 Gram-negative coeet Ve paroula 10 8 Gram-positive bal 'B. adoleseentis, 2 3 Bubacterium sp 5 4 Lactobacillus sp 3 6 Peacnes 1 3 ‘Actinomyces sp fl 4 Gram-negative bal Fusobacterium sp Bacteroides 2p B. melaninogenicus grp Boras B, ruminicolla sap, brevis B. fragilis group B. corradens ‘Total {ganioms. Statistical analysis was done using the chisquare and Mann-Whitney teats RESULTS Four hundred forty-six bacterial strains were iso- lated from the tonsils: 188 aerobes and 258 anaerobes (Table I). Mixed aerobic and anaerobic flora were present in all the tonsils, yielding an average of 9.3, isolates (3.9 aerobes and 5.4 anaerobes) per speci- men, No consistent pattern of combinations of the different aerobie and anaerobic bacteria in the ton- sils was noted. ‘The aerobic organisms most frequently isolated were alpha: and gamma-hemolytic streptococci, Branhamella catarrhalis, Staphylococcus aureus, beta-hemolytic streptococci (Groups A, B, and C), and Hemophilus sp. The predominant anaerobie or- ganisms were Bacteroides sp (including Bacteroides melaninogenicus and Bacteroides fragilis groups), Fusobacterium sp, anaerobic gram-positive cocci, and Veillonella parvula. ‘There were differences in the recovery rate of or- ganisms and the species recovered in adults: 10.2 per tonsil (3.8 aerobes and 6.4 anaerobes) as com- pared to children, 8.4 per tonsil (4.0 aerobes and 4.4 anaerobes). The difference between these groups of patients was therefore due to a higher recovery rate of anaerobes in adults: a higher recovery rate of B. melaninogenicus group in adults (1.6 per adult. as compared to 0.8 per child) (P<0.05) and B. fragilis group (0.4 per adult and 0.2 per child) (P>0.05). A difference in the recovery rate of GABHS was also noted. These organisms were isolated in 7 of the 25 children (0.28 per patient) and only in 1 of the 23 adults (0.04 per patient) (P=0.028). Conversely, al- though not statistically significant (P>0.05), Group B beta-hemolytic streptococci were recovered more often in adults, 5 of 28 patients (0.22 per patient), as compared to 2 of 25 children (0.08 per patient). Bete-lactamase production was detectable in 74 isolates recovered from 87 (77%) tonsils from 21 (01%) adults and 16 (64%) children. ‘These included all isolates of S. aureus and B. fragilis group, 25 of the 58 (48%) B. melaninogenicus group, 8 of 7 (43%) Bacteroides oralis, 5 of 14 (86%) Hemophilus sp, and 5 of 29 (17%) of B. catarrhalis. There were differ- ences in the recovery of beta-lactamase-producing organisms (BLPO) among the children as compared to adults. Forty-three BLPO (1.9 per patient) were recovered in adults as compared to 29 (1.2 per pa- tiont) in children (P=0.04). DISCUSSION ‘This study confirms our previous observations of the polymicrobial, aerobic, and anaerobic flora recov- ered from the cores of recurrently inflamed tonsils.”** We wore able to detect BLPO in over three fourths of our patients, a finding that confirms previous data reported in children with recurrent tonsillitis.'* The ability of BLPO to protect penicillin susceptible organisms has been shown in vitro‘ and in vivo.” These BLPO may degrade penicillin in the area of the infection, protecting not only themselves bbut other penicillin-sensitive pathogens.” Therefore, penicillin therapy directed against a susceptible pathogen might be rendered ineffective by the pres- ence of a penicillinase-producing organism, We were able to detect differences in the bacterial flora in children as compared to adults with recur- rent tonsillar pharyngitis. ‘The higher number of or- ganisms per tonsil in adults was attributed to the in- creased recovery rate of B. melaningenicus and B. fragilis groups. The higher number of beta-lactamaseBROOK AND FOOTE: MICROBIOLOGY OF RECURRENT TONSILLITIS 1387 producing isolates of Bacteroides sp found in adults ‘accounts also for the higher recovery rate of BLPO in older patients, Another important finding is the rareness of recovery of GABHS and the higher iso- lation rate of Group B streptococci in the older age group. The latter finding confirms a previously re- ported study."* ‘These findings may have important clinical impli- cations on patient management. As less GABHS and more Bacteroides sp are found in older patients, a shift in the flora might have occurred in adults, making the Bacteroides sp of organisms more clini- cally important, B. melaninogenicus is part of the normal oral and vaginal flora, It has recently emerged as the predominant Bacteroides sp isolated from an- aerobic pleuropulmonary and upper respiratory tract infections."* B. melaninogenicus is also com- monly isolated from many other other sites of anaer- obic infection, such as pelvic inflammatory disease” and cutaneous abscesses," and has a major role in periodontal disease.” ‘The pathogenicity of B. melaninogenicus group has also been demonstrated in animal models.” B. melaninogenicus was shown to be a necessary com- ponent of many infections." Polymicrobial infec- tions are more pathogenic for experimental animals than are those involving single organisms.” Since B. molaninogenicus is always present in the tonsils with other bacteria, it is possible that the tonsillar infection is polymicrobial, where the Bacteroides are only a part of the infectious flora. ‘An important clinical consideration is the higher number of BLPO recovered in adults, which may further complicate the antimicrobial therapy given to these patients. The higher recovery rate of BLPO in adults is not surprising, since these patients have received over the years many more courses of anti- microbials, which might have created a shift in the oral flora toward selection of BLPO such as Bac- teroides sp. The observed changes in the tonsillar flora may also be due to the increased immunity against GABHS in adults, which may decrease the infectivity of these organisms in this age group. An- other important factor is the accumulation of post- inflammatory changes in adults, which include sear- ring and microabscesses. ‘These changes make it harder for antimicrobials to penetrate and achieve therapeutic levels inside the tonsils. Our findings suggest that recurrent tonsillitis in adults may be different from that illness in younger individuals both in microbiology and appearance of the tonsils. The management of this recurrent infec- tion may, therefore, be different in these age groups. Howover, individual variability exists in each age group and antimicrobial therapy should be adjusted to the individual patient, ‘The isolation of these potentially pathogenic or- ganisms as well as BLPO from tonsils in patients with recurrent tonsillar pharyngitis raises the question whether the treatment of recurrent tonsillitis with penicillin is adequate in all instances, and whether therapy should be directed at the eradication of these organisms. There are several clinical trials in children that support the administration of antimi- crobial agents that are effective against both aero- bic and anaerobic BLPO in case of recurrent tonsil tis.' In the most recent of these studies, clindamycin therapy was found to be superior to erythromyein or penicillin. Another drug that may be effective in the therapy of this condition is the combination of a beta-lactamase inhibitor, clavulanic acid, and amox- icillin. Further studies are howover needed to evalu- ate the efficiency of such therapies in different age groups. ACKNOWLEDGEMENT ‘The authors gratefully acknowledge J. E. Perry and J. C. Gillmore for technical assistance, W. E. Jackson for statistical analysis, and Ms. Sharon Prince for secretarial assistance, REFERENCES 1, Brook, [The Role of Beta-Lactamnase Procizeing Bacteria In the Persistence of Streptococcal Tonsllar Infection. eu. Inf Dis, 601-60", 1984 2, Farockl, M.A. Bacteriology and Histology of Tonsilar Porenehyma. in ‘Tonsillectomized Specimens. Eye Bar Nose Throat Mom, 46:301-312, 1967 3, Brook, L: Abscssses of the Head and Neck. In: Anaerobic Infeetion in Childhood, f. Brook (Ed). G. V. Hall, Boston, pp. 176-191, 1983. 4. Gruner, O. P. Nz Actinomyces in Tonsillar "issue. A Hise torial Study of Tonsilicetomy Material. Acta Pathol Microbiol Scand, 16:280-244, 1960, 5. Mena, E, Thompson, FS, Armfield, A. Y.,e¢ al: Evalua- Lion of Port‘A-Gul Transport System for Protection of Anaerobic Bacteria, J. Cin, Microbial, 82885, 1978 6, Holdeman, LB. and Moore, W. B. C: Anaerobe Labora tory’ Manual 4th Ed. Vieginia Polytechnic Institute, Blacks burg, 1977 7. Lennelte, EH, Ballows, A., Havsler, W. eta: Manual of Clinial Microbiology, 3rd Bd. Amerlean Society for Mierabiolo- ay. Washington, DC, 1980 8, O'Callagham, C. H., Mort, A., Kitby, SM. etal: Novel Method for Detection of Beta-Lactamase By Using a Chromato genie Cephalosporin Substrate. Antimicrob: Agents Chemother, 12285-2885, 1972, 9. Brook, 1, Yocum, P. and Shah, K-: Surface Vs, Core‘Ton sillar Acrobie ‘and Anaorabie Plora in Recurrent Tonsilitis. JAMA, 244:1696-1698, 1980. 10, Brool, 1, Yocum, P, and Friedman, B. Ms Acrobie and Anacrobie Flore Rocovored From Tonsilar Tissue of Children with ‘Recurrent Tonslliis, Ann, Otol Ahinal Laryngol 5o:269-269, 1981, 11, Simon, H. M, and Sukalr, W. Staphylococeal Antagonism to Poncilin Group Therapy of Hemolytie Streptococcal Pharyn geal Infection: Bifect of Oxaclln. Pediatrics, 31:463-469, 1983. 12, Brook, 1, Pazzaglia, C., Coolbeugh, J. C.,et al: In Vivo, Protection of Group A Beta Hemolytic Streptococci by Bote Lactamase Producing Bacteroides Spaces. J. Antimierob. Che- ‘mother, 12:698-608, 198,