Pract Otol(Kyoto) 96:11;983～987, 2003 983

Antibiotics Therapy for Acute Bacterial Tonsillitis

Shinya Takano and Hideki Kurihara Tokyo Women's Medical University Daini Hospital

We reviewed the treatment of 96 patients with acute bacterial tonsillitis.

Using multivariate analysis, we examined whether oral or intravenous administration of antibiotics

(cephems, penicillin, other ƒÀ-lactams, tetracycline, clindamycin, new quinolones, and macrorides) had an influence on the treatment period.

We found that the best class of oral antibiotics for acute bacterial tonsillitis was cefcapene pivoxil

(CFPN-PI) and the best intravenous agent was sulbactum/cefoperazone (SBT/CPZ).

Key words:retrospective study, acute bacterial tonsillitis, antibiotics, oral administration, intravenous injection

Introduction tis such as infectious mononucleosis were excluded by

Acute bacterial tonsillitis is one form of upper respira- blood tests.

tory tract infection, in which bacteria invade the palatine We selected the administration of antibiotics at ran-

tonsil and cause fever, sore throat, pus, and pain on swal- dom.

lowing. The day when symptoms resolved and the WBC and

In Europe and The United States, the drug of first CRP were normalized was defined as a "the day of cure"

choice for acute bacterial tonsillitis is penicillin. How- and the treatment period defined as the interval between

ever, there have been no reports on the first line treat- the first examination and this day. Using multivariate

ment for acute tonsillitis in Japan, and only new analysis, we examined whether oral administration

antibiotics have been examined for effectiveness against (levofloxacin (LVFX), clarithromycin (CAM), minocy-

acute tonsillitis. cline (MIND), cefcapene pivoxil (CFPN-PI), ampicillin

Penicillin is contraindicated in patients with infectious (ABPC), faropenem (FRPM), and no medication) or

mononucleosis because it causes an exacerbation of their intravenous administration (cefepirome (CPR), clinda-

rash1). Therefore, it is important to discriminate acute mycin (CLDM), minocycline (MINO), sulbactum/cefop-

tonsillitis from infectious mononucleosis, and it is neces- erazone (SBT/CPZ), and no injection) and/or the bacteria

sary to administrate the appropriate antibiotics. isolated (K. pneumoniae, S. aureus, H. influenzae, S. Because it is important to determine the antibiotics of pneumoniae, normal flora, and negative culture) influ- first choice for acute infectious diseases in Japan, we ret- enced the duration of the treatment period.

rospectively examined the clinical use of antibiotics for Analysis was performed with Stat Partner for Win-

acute tonsillitis and reviewed the literature. dows Ver.3 (NEC Software Company).

Subjects and Methods Results

The subjects were 96 patients treated at our hospital 1. Bacteriologicalexamination

in 2001. There were 51 females (26.5•}10.6 years old: The distribution of bacteria detected in tonsillar pus is

mean•}SD) and 45 males (33.9•}10.6 years). Patient age shown in Figure 2.S. pneumoniae was detected in 25%

distribution is shown in Figure 1. Causes of viral tonsilli- of the patients, while H. influenzae and S. aureus were 984 Shinya Takano and Hideki Kurihara pract Otol (Kyoto) 96:11

Fig. 1 Age distribution of the subjects.

(r=4.263). 3. Relationship between the treatment period and oral antibiotics The relationship between the treatment period and oral antibiotics administered is shown in Figure 3. CFPN-PI and FRPM shortened the treatment period, while CAM and MINO actually extended the treatment period. 4. Relationship between treatment the period and bacteria isolated The relationship between the treatment period and the bacteria isolated is shown in Figure 4. The treatment

Fig. 2 Bacteria detected in all patients period was longer when S. aureus was isolated. 5. Relationship between the treatment period and intravenous antibiotics each detected in 8% of the patients. In addition, normal The relationship between the treatment period and flora was detected in 54% and no bacteria were isolated the type of antibiotics given intravenously is shown in in 3%. Resistant bacteria were not detected. Figure 5. SBT/CPZ shortened the treatment period. In 2. Multivariate analysis contrast MIND extended the treatment period. The treatments used and the combinations of oral and The average treatment period is stratified according to intravenous antibiotics varied between patients, so we the kind of antibiotics and the bacteria isolated in Table 2. investigated which combination provided the best ther- CFPN-PI were effective after oral administration and apy for acute bacterial tonsillitis by multivariate analysis. SBT/CPZ were effective after intravenous injection The results are shown in Table 1. The multiple correla- when the bacteria were S. pneumoniae and H. influenzae, tion coefficient was r=0.5239, and the most important which were detected commonly. determinant of the treatment period was oral administra- tion (r=4.877), followed by the type of bacteria isolated Pract Otol(Kyoto) 96:11 Acute Tonsillitis 985

Fig. 3 Relationship between the treatment period and oral administration

Fig. 4 Relationship between the treatment period and the bacteria isolated

Fig. 5 Relationship between the treatment period and intravenous administration 986 Shinya Takano and Hideki Kurihara Pract Otol (Kyoto) 96:11

Table 1 Results of multivariate analysis.

Table 2 Average treatment period (relationship between the bacteria isolated and the antibiotics used)

Discussion been many reports on the clinical efficacy of newer anti-

Penicillin is the drug of first choice for acute tonsillitis biotics in Japan2•`5, but there is no report describing in Europe and The United States. Since 1980, there have guidelines for the use of antibiotics for acute infectious Pract Otol (Kyoto) 96:11 Acute Tonsillitis 987

diseases in the otolaryngology field. We considered it The detection rate of S. pneumoniae was highest, fol- necessary to develop treatment guidelines for acute lowed by H. influenzae,and S. aureus. infectious diseases in Japan from the aspect of medical The treatment period was dependent on the antibiotics treatment economy. used for oral administration, the bacteria detected, and

For antibiotic therapy, it is important to continuously the antibiotics given intravenously. maintain drug concentrations equal to or higher than the Cefcapene pivoxil (CFPN-PI) is recommended as the

MIC for the bacteria in the lesion. However, this is actu- first choice for oral therapy for acute bacterial tonsillitis,

ally difficult to achieve, and administration every 8-10 while sulbactum/cefoperazone (SBT/CPZ) is the first

hours is desirable in patients with normal immunity6). choice for intravenous therapy.

Penicillin achieves high tissue levels in the palatine

tonsil7). In particular, the tissue levels achieved in the References

1) Takano S, Aramaki H, Yoda K, et al.:Infectious mononucleo- palatine tonsil by amoxicillin/clavulanic acid are excellent sis and bacterial tonsillitis-differencial diagnosis from clinical after oral administration7), while intravenous ampicillin7) symptom-. J of Jpn Soc for Infect Disin Otol 17:70～72, is also excellent in this respect. 1999. In the case of oral cephems, tissue levels in the 2) Kawasaki Y, Sakamoto Y, Ogawa S, et aI.:Clinical studies on

palatine tonsil achieved by cefalexin and cefaclor are T-2588 in the field of otorhinolaryngology. The JPn J Antibiot

good7), while intravenous therapy achieves excellent tis- 39:2710～2715, 1986.

sue levels using cefazolin, cefoperazon, or cefoxitin6). 3) Fujita T, Fukui T, Okui S, et al.:In vitro and in vivo clinical

Especially, cefalexin and cefaclor show higher tissue lev- evaluation of BMY-28100. Chemotherapy 37:222～233, 1989.

4) Obana M, Matsuoka Y and Irimajiri S:Clinical study on S-1108 els than the serum concentration7). in respiratory tract infections. Chemotherapy 41:776～780, The tissue uptake in the palatine tonsil of ƒÀ-lactams is 1993. low after oral administration and intravenous injection7), 5) Majima Y, Shinogi J, Yamagiwa M, et al.:Clinical effects of cef- but a sufficient effect can be expected when considering podoxime proxetiil in otorhinolaryngological infection. Prog

MIC values for the relevant bacteria7). Med 14:1002～1005, 1994.

Suzuki et al.8) reported that there were no changes in 6) Eagle H, Fleischman R, Musselman A, et al.:The bacterial

action of penicillin in vivo;the participation of the host, and the the bacteria detected from patients with acute tonsillitis

slow recovery of the surviving Organisms. Ann Intern Med over in the past ten years, and that S. pyogenes, S. aureus 331:544～571, 1950. S. pneumoniae, and H. influenzae were commonly in 7) Suzuki K:The usage and dosage of anti-bacterial agents from detected in acute tonsillitis. An increase in the detection the tissue concentration-from the point of the systemic

rate of PRSP was reported, but there was no change in administration-. J of JPn Soc for Infect Dis in Otol 6:136～

the MRSA detection rate8). 142, 1988.

These results suggest that CFPN-PI is the drug of 8) Suzuki K and Baba S:Study of isolated bacteria from tonsilli-

tis. J of JPn Soc for Stomato-pharyngol 11:231～237, 1999. first choice for oral administration in patients with acute

bacterial tonsillitis, while SBT/CPZ is the first choice for

intravenous therapy. Received February 20, 2003 Accepted April 23, 2003 Mailing address:Shinya Takano, Department of Otorhi- Summary nolaryngology, Yokohama Rosai Hospital, 3211 Kozukue- cho Kouhoku-ku Yokohama, Kanagawa, 222-0036 We reviewed the treatment of 96 patients with acute

bacterial tonsillitis.