- Email: [email protected]
Selection of antibiotics after incision and drainage of peritonsillar abscesses
Selection of antibiotics after incision and drainage of peritonsillar abscesses DAVID A. KIEFF, MD, NEIL BHATTACHARYYA, MD, NOAH S. SIEGEL, MD, and SALAH D. SALMAN, MD, Boston, Massachusetts
Despite the fact that peritonsillar abscess is the most common complication of acute tonsillitis, the treatment of peritonsillar abscess remains controversial. One element of controversy is the choice of antibiotics after drainage of the abscess. In an attempt to assess the effect of antibiotic choice on the treatment of peritonsillar abscess, we conducted a retrospective review of records from patients with peritonsillar abscess treated with incision and drainage. Our review identified 103 patients, comprising two groups: 58 patients treated with broadspectrum intravenous antibiotics and 45 patients treated with intravenous penicillin alone. These patients were hospitalized after incision and drainage, and therefore their clinical courses and responses to therapy could be rigorously assessed. Characterization of illness based on patient age, temperature, and white blood cell count revealed similar severity of illness between the two groups. Comparison of clinical outcomes with respect to hours hospitalized (mean 44.3 ± 6.6 and 38.3 ± 7.1 hours, 95% confidence interval, for broad-spectrum and penicillin groups, respectively) and mean hours febrile (16.9 ± 5.0 and 13.3 ± 4.2 hours, 95% confidence interval) were not statistically significantly different (p = 0.222 and 0.269, respectively) between groups, indicating that broad-spectrum antibiotics failed to show greater efficacy than penicillin in the treatment of these patients. The microbiologic characteristics of these infections, failures of therapy, and complication rates were similar to those reported in the literature. These results suggest that intravenous penicillin remains an excellent choice for therapy in cases of peritonsillar abscess requiring parenteral antibiotics after drainage. (Otolaryngol Head Neck Surg 1999;120:57-61.) From the Department of Otolaryngology–Head and Neck Surgery, Massachusetts Eye and Ear Infirmary (Dr. Kieff, Dr. Siegel, Dr. Salman) and the Joint Center for Otolaryngology (Dr. Bhattacharyya), Harvard Medical School. Reprint requests: Neil Bhattacharyya, MD, Joint Center for Otolaryngology, 333 Longwood Ave., Boston, MA 02115. Copyright © 1999 by the American Academy of Otolaryngology– Head and Neck Surgery Foundation, Inc. 0194-5998/99/$8.00 + 0 23/1/88637
Peritonsillar abscess (PTA) occurs with an incidence of 30 per 100,000 in the United States among patients 5 to 59 years of age, accounting for approximately 45,000 cases per year. Although PTA is the most common complication of tonsillitis, its treatment varies considerably among otolaryngologists.1-5 Some advocate incision and drainage as the initial treatment, others needle aspiration, and still others quinsy tonsillectomy. In addition, the antibiotics chosen to treat PTA vary widely, ranging from penicillin, an agent with activity against oropharyngeal gram-positive aerobes and anaerobes, to ampicillin/sulbactam, an agent with broad gram-positive and gram-negative aerobic and anaerobic activity.6 Despite the expense of expanded-spectrum antibiotics, pressure for their use comes from the pharmaceutical industry, as well as many reports of resistance among the pathogens typically found in PTA. Penicillin resistance has been reported to occur in more than 50% of cases.1 Herzon1 attempted to establish treatment guidelines for PTA. He proposed outpatient needle aspiration in conjunction with oral penicillin therapy on the basis of an analysis of a cohort study of 123 patients, a practitioner survey, and a meta-analysis of prior studies. However, in spite of his extensive work, he found that the selection of antibiotic therapy for PTA remains controversial. To date, no investigator has reported on the effect of antibiotic therapy as an independent variable in treatment outcome. We report a retrospective study comprising 103 patients and comparing penicillin therapy with broad-spectrum antibiotic therapy in cases of PTA first treated with incision and drainage. The relevant microbiologic findings are also addressed. METHODS AND MATERIAL We screened admissions to our hospital from January 1993 to June 1996 for patients with the diagnosis of PTA. This contemporary time period was chosen to keep the results relevant to current antibiotic resistance patterns. Admissions were included in the study if the patient’s age was between 16 and 65 years; pediatric patients were excluded because they are treated according to a different protocol at our institution. Patients with significant comorbid factors such as diabetes, immunosuppression, or HIV infection that might alter their response to infection were also excluded. Medical records were analyzed for the following variables: 57
58
Otolaryngology– Head and Neck Surgery January 1999
KIEFF et al
Table 1. Demographics and variables characterizing patient groups with peritonsillar abscesses Broad-spectrum (n = 58) Variable
Age (yr) Maximum temperature (°F) Maximum WBC count (1000/mm3) % Quinsy tonsillectomy Hospitalization (hr) Febrility (hr) % Recrudescence of abscess
Mean ± 95% CI
28.1 ± 2.4 99.8 ± 0.8 14.9 ± 1.3 6.9 ± 6.7 44.3 ± 6.6 16.9 ± 5.0 3.4 ± 4.8
Penicillin alone (n = 45)
Range
Mean ± 95% CI
16.0–62.0 97.0–102.6 7.9–30.3
31.8 ± 2.5 99.8 ± 0.3 14.3 ± 0.9 0 38.3 ± 7.1 13.3 ± 4.2 6.6 ± 7.5
12.0–136.0 0.0–96.0
Range
18.0–53.0 98.6–102.2 9.0–24.2 12.2–120.0 0.0–48.0
CI, Confidence interval; WBC, white blood cell.
age, use of oral antibiotics before admission, in-hospital antibiotic therapy, maximum temperature, and maximum white blood cell count during hospitalization, duration of hospitalization, recrudescence (recollection of fluid in the abscess cavity after drainage) of the abscess within 1 week, quinsy tonsillectomies, and gram-stain and culture results from the abscess fluid. The hours the patient was febrile were taken from time of emergency department evaluation until the time of the last recorded fever. Patients with PTA at our institution are uniformly treated with incision and drainage in the emergency department by an otolaryngologist, followed by admission for intravenous antibiotic therapy and hydration until discharge. Hours of hospitalization were recorded from time of incision and drainage to time of discharge. Discharge criteria include demonstration of adequate oral intake (>750 ml in the preceding 8 hours) and the absence of fever. Recurrences within 1 week were classified as recrudescences of the primary abscesses, and these, as well as readmissions to the hospital, were noted. On the basis of intravenous antibiotic therapy, the patients were stratified into two groups after incision and drainage: those who received intravenous penicillin therapy and those who received broader spectrum intravenous therapy. We then statistically analyzed the results for each group and compared them using a spreadsheet-based statistical package. RESULTS
Review of records identified 103 patients with PTA who qualified for the study. Of these, 58 patients received broad-spectrum intravenous antibiotics and 45 received intravenous penicillin alone. Summary data for the two groups are represented in Table 1. All patients underwent incision and drainage in the emergency department. The procedure was performed after topical and infiltration anesthesia, needle localization of the abscess, and mucosal incision and blunt opening of the abscess cavity. Intravenous antibiotic therapy and hydration were instituted on admission for all patients reviewed. The group of 58 patients treated with broad-spectrum antibiotics included patients who received single agents such as clindamycin (43 patients) or ampi-
cillin/sulbactam (9 patients), as well as patients who received combination therapy such as clindamycin plus a second-generation cephalosporin or ampicillin/sulbactam (2 patients), in addition to 2 patients who received penicillin and metronidazole. Two additional patients received combination first- and second-generation cephalosporins. In this group, age ranged from 16 to 62 years, with a mean of 28.2 years. Six patients’ abscesses recurred: Four were treated with quinsy tonsillectomy (6.9%), and two patients (3.4%) underwent repeat incision and drainage. Specifically, deep neck space infection requiring external drainage developed in none of the patients, and no patient had systemic sepsis, Clostridium difficile colitis, or delayed postdrainage hemorrhage. There were no adverse reactions to the administered antibiotics. In this group, one patient’s intravenous antibiotic regimen was changed (1.7%), from clindamycin to ampicillin/sulbactam after 2 days of worsening fever and trismus while the patient received the former drug. The clinical outcome variables results: hours of hospitalization and hours febrile for these patients are also listed in Table 1. Microbiologic data were available for 48 of the 58 patients (82.8%). There was no growth on culture in nine patients (18.8% of those cultured); the remaining culture results are displayed in Table 2. No adjustments of antibiotics were made on the basis of culture results; patients were either responding to therapy or were switched to alternate antibiotics before culture data were available. All patients were discharged home with oral antibiotics consisting of amoxicillin/clavulanate in 13 cases, clindamycin in 39 cases, penicillin in 3 cases, and other antibiotics in 3 cases. No patients returned to the hospital for near-term recurrence (recrudescence) of their abscess. The other group of 45 patients was treated with intravenous penicillin. The average age in this group was 31.8 years (range 18 to 53 years). Three patients (6.7%) suffered recrudescence of their abscesses and had another drainage procedure; none underwent quinsy
Otolaryngology– Head and Neck Surgery Volume 120 Number 1
KIEFF et al
59
Table 2. Microbiologic findings in PTA patients Broad-spectrum group Isolate
No.
Penicillin group % of Total cultured
α-Hemolytic Streptococcus Anaerobic gram-negative rods Bacteroides spp. β-Hemolytic non–group A Streptococcus Coagulase-negative Staphylococcus spp. Fusobacterium Group C Streptococcus Haemophilus influenzae Pseudomonas aeruginosa Staphylococcus aureus Streptococcus pyogenes
13 1 4 5 1 6 3 3 1 5 12
27.1 2.1 8.3 10.4 2.1 12.5 6.3 6.3 2.1 10.4 25.0
More than one organism No growth Not cultured
15 9 10
31.3 18.8 17.2
tonsillectomy. No complications occurred during the hospital course in this group, and there were no adverse drug reactions. Three patients’ antibiotic therapies were altered during hospitalization this group. The changes were from penicillin to clindamycin in two cases, and from penicillin to ampicillin/sulbactam in one case. These changes were necessitated by lack of response to penicillin alone, manifested by worsening pain, fever, and trismus. The clinical outcome variables results for these patients are also listed in Table 1. Microbiologic data were available for 36 of the 45 patients (80.0%). There was no growth on culture in two patients (5.6% of those cultured); the remaining culture results are displayed in Table 2. Again, no adjustments of antibiotics were made on the basis of culture results. All patients were discharged home with oral antibiotics consisting of penicillin in 37 cases, amoxicillin/clavulanate in 2 cases, clindamycin in 2 cases, and amoxicillin in 4 cases. No patient returned to the hospital for recrudescence of an abscess. Comparison of the data for the two groups revealed similar parameters characterizing the groups and the severity of their presenting PTA. The mean ages of the two groups were very similar (28.2 vs. 31.8 years). No statistically significant difference was demonstrated between groups for the maximum temperatures (p = 0.735, two-tailed Student’s t test), and highest white blood cell counts (p = 0.434), suggesting that the two groups had similar severities of disease. The outcome variables for the two groups were compared with the use of a two-tailed Student t distribution. This demonstrated no statistical difference between groups with respect to
Isolate
No.
α-hemolytic Streptococcus 15 Anaerobic gram-negative rods 1 Bacteroides spp. 4 β-Hemolytic non–group A Streptococcus 4 Clostridium spp. 1 Coagulase-negative Staphylococcus spp. 1 Fusobacterium 4 Group C Streptococcus 2 H. influenzae 2 Moraxella catarrhalis 1 Nocardia spp. 1 P. aeruginosa 1 S. aureus 2 S. pyogenes 14 More than one organism 13 No growth 2 Not cultured 9
% of Total cultured
41.7 2.8 11.1 11.1 2.8 2.8 11.1 5.6 5.6 2.8 2.8 2.8 5.6 38.9 36.1 5.6 20.0
duration of hospitalization (p = 0.222) and hours to defervescence (p = 0.269). The recrudescence rate (Table 1) was slightly less for patients treated with broad-spectrum agents, but this was not statistically significant. DISCUSSION
The incidence of PTA in the United States has been reported to be 30 per 100,000 person-years, accounting for approximately 45,000 cases and an estimated cost of greater than $32 million annually.1 Treatment of PTA remains varied among practitioners, and it is controversial despite several studies on the subject.2,7 Incision and drainage of PTA was the most common form of management in the United States before 1960 because of the hypothetical risk of producing rheumatic fever or glomerulonephritis from septicemia after abscess tonsillectomy. Volk and Brandow’s report in 1960 paved the way for a change in management of PTA to an “abscess,” or “quinsy” tonsillectomy.8 Although quinsy tonsillectomy became the favored practice on the basis of the impression that it led to the quickest recovery, subsequent studies suggested that antibiotic therapy with abscess aspiration or incision and drainage were equally effective as quinsy tonsillectomy and antibiotic therapy.5 A corollary of these studies was that needle aspiration of PTA yielded equivalent results to incision and drainage. Resurrecting this controversy, a recent retrospective study by Wolf et al.3 refuted the conclusion that aspiration of PTA is as efficacious as incision and drainage. The impetus for the use of broader-spectrum antibiotics in PTA stems from two arguments. First, there have been several reports of increased bacterial resis-
60
KIEFF et al
tance in pharyngotonsillar infections, leading to concern that broader-spectrum antibiotics are needed for effective treatment of the disease.4 Whereas group A βhemolytic streptococci have been classically associated with PTA and are universally penicillin-sensitive, penicillin-resistant aerobes and anaerobes, as well as gramnegative organisms, have been detected in PTA cultures. Brook et al.9 noted that 52% of 25 PTA specimens grew β-lactamase–producing organisms. The abscess is often polymicrobial, and the organisms include Staphylococcus aureus and Bacteroides fragilis, as well as Bacteroides melaninogenicus, a common oral anaerobe. Such organisms are able to survive penicillin therapy and can theoretically protect penicillin-sensitive organisms by releasing penicillinase enzyme into the abscess microenvironment. Therefore Brook et al. believed that antibiotic therapy against PTA should be both penicillinase-stable and effective against anaerobes. A national survey indicated that 52% of otolaryngologists use broad-spectrum agents in first-line treatment of PTA.1 Second, the efficacy of penicillin therapy has also been questioned because of the finding that it does not penetrate well into the abscessed tonsil, whereas cephalosporins are preferentially concentrated in abscessed tonsillar tissue.10 Despite these two theoretical concerns regarding penicillin therapy for PTA, no outcome data demonstrate an advantage of broad-spectrum antibiotic therapy. Indeed, our analysis of 103 patients treated for PTA failed to demonstrate a statistically significant advantage of broad-spectrum antibiotic therapy over penicillin. The patient populations in the broad-spectrum and penicillin groups were not significantly different in their age, maximum temperature, or white blood cell profiles; nor were they significantly different in outcome with respect to hours hospitalized, hours febrile, need for quinsy tonsillectomy, or recurrence rate. The microbiologic data were similar between groups and are consistent with those reported by others: Streptococcus pyogenes and α-hemolytic Streptococcus species predominated.2,9,11 We do not routinely culture for anaerobes because of the cost and the fact that rarely have we needed to adjust antibiotic therapy for anaerobes that were not sensitive to empiric antibiotics. In this series, very few patients had their antibiotics changed during therapy. In the patients whose antibiotics were changed (four total), the choice of the new antibiotic was not based on specific culture results but broadened empirically, suggesting that clinical response is more significant than microbiologic data in the selection of antibiotic therapy and that routine culture of the aspirate may be of relatively limited value. Because no reports have compared the efficacy of incision and drainage alone vs.
Otolaryngology– Head and Neck Surgery January 1999
incision and drainage of PTA with antibiotic therapy, and because several of the patients who responded to penicillin in our study grew resistant organisms on culture, it is possible that incision and drainage without antibiotic therapy is adequate treatment of PTA. On theoretical grounds, we advocate antibiotic therapy, and this remains the standard of care.1 Regardless, these data suggest that penicillin therapy is as efficacious as broader-spectrum therapy in the treatment of drained PTA. Given the retrospective nature of this study, several sources of bias may exist. A bias may exist toward treating patients in whom PTA develops while they are being treated with oral antibiotics, especially penicillin, with broader-spectrum antibiotics. In our groupings however, the percentages in both groups of patients who were treated with oral antibiotics before PTA developed were similar. The culture analysis showed similar types of organisms and positive culture rates, regardless of prior antibiotic therapy, indicating the minimal effect of antibiotic therapy on the microbiology of a developing PTA. Interestingly, more cultures were sterile in the broad-spectrum group. This finding should have favored the broad-spectrum group because they theoretically had fewer active organisms at the time of drainage, yet no clinical advantage for broad-spectrum agents was shown. Moreover, the selection bias of sicker patients’ receiving broader-spectrum antibiotics is adjusted for by the similar severities of illness between the two groups, based on maximum temperature and maximum white blood cell count. It is important to note that the admission of patients with PTAs for intravenous antibiotic therapy is becoming less common.1 During the period of this study, all PTAs were treated with incision and drainage, followed by admission for observation, intravenous antibiotics, and fluids. This permitted close follow-up of these patients with respect to temperature, time to adequate oral intake, and need for further procedures. Nevertheless, we expect that the conclusions from our results can be applied to those patients who undergo aspiration and intravenous therapy and, possibly, to those patients undergoing aspiration and outpatient oral antibiotic therapy. It could be argued that broad-spectrum antibiotics should be used when oral therapy is chosen after PTA drainage because the high blood levels of intravenous penicillin may have overwhelmed possible penicillin resistance in this study; however, this argument remains speculative. In fact, our results showed that a greater number of patients (six) in the broad-spectrum group required quinsy tonsillectomy for failure of incision and drainage than did patients in the penicillin group. At our institution, the choice between quinsy tonsillectomy and repeat incision and drainage for PTAs that
Otolaryngology– Head and Neck Surgery Volume 120 Number 1
fail initial drainage and antibiotic therapy is usually based on physician preference. A bias may have existed in treating recrudescences in the broad-spectrum group with quinsy tonsillectomy as opposed to repeat incision and drainage out of concern for continued failure of nonoperative therapy. However, the difference in rates of quinsy tonsillectomies between the two groups was not of statistical significance. In fact, the difference in the rate of quinsy tonsillectomy is also offset by the difference in the rate of recrudescence of the abscess (the penicillin group had a higher rate of recrudescence) so that, taken together, the rate of quinsy tonsillectomy and repeat drainage are about equal between groups. CONCLUSIONS
A cost-effective approach should be adopted for the treatment of PTA. Despite the fact that PTA is a polymicrobial infection that may involve penicillin-resistant organisms, intravenous penicillin is as clinically effective as broad-spectrum antibiotics in cases of PTA, as demonstrated by this study of 103 patients. Because penicillin remains one of the least expensive antibiotic therapies and is well tolerated, we recommend that it be used as the first-line treatment of PTA after incision and drainage, especially when intravenous hydration is required. Achieving effective abscess drainage is a key element in therapy. Cultures of abscess fluid have little
KIEFF et al
61
bearing on the clinical management of PTA, and the routine use of such cultures should be reconsidered. Further prospective studies addressing the comparative clinical effectiveness of various oral antibiotics will help encourage safe and cost-effective therapy. REFERENCES 1. Herzon FS. Peritonsillar abscess: incidence, current management practices, and proposal for treatment guidelines. Laryngoscope 1995;105(Suppl):1-17. 2. Savolainen S, Jousinies-Somer HR, Makitie AA, et al. Peritonsillar abscess: clinical and microbiologic aspects and treatment regimens. Arch Otolaryngol Head Neck Surg 1993;119:521-4. 3. Wolf M, Even-Chen I, Kronenberg J. Peritonsillar Abscess: repeated needle aspiration versus incision and drainage. Ann Otol Rhinol Laryngol 1994;103:554-7. 4. Poole MD. Penicillin may no longer be the drug of choice for Streptococcal pharyngitis. ENT J 1993;4:330-2. 5. Parker GS, Tami TA. Management of peritonsillar abscess in the 90s: an update. Am J Otolaryngol 1992;13:284-8. 6. Fairbanks DN. Antimicrobial therapy in otolarygology–head and neck surgery. 7th ed. Alexandria, Va.: American Academy of Otolaryngology–Head and Neck Surgery; 1993, p 9-11. 7. Maharaj M, Rajah V, Hemsley S. Management of peritonsillar abscess. J Laryngol Otol 1991;105:743-5. 8. Volk BM, Brandow EC Jr. Bilateral tonsillectomy for peritonsillar abscess. Trans Am Laryngol Rhinol Otol Soc 1960;63:776-85. 9. Brook I, Frazier E, Thompson DH. Aerobic and anaerobic microbiology of peritonsillar abscess. Laryngoscope 1991;101:289-92. 10. Maisel RH. Peritonsillar abscess: tonsil antibiotic levels in patients treated by acute abscess surgery. Laryngoscope 1982;92:80-7. 11. Jokinen K, Sipila P, Jokipii L, et al. Peritonsillar abscess: bacteriological evaluation. Clin Otolaryngol 1985;10:27-30.
Otoplasty and Reconstruction of Auricular Defects
An International Course on Otoplasty and Reconstruction of Auricular Defects will be held March 21-22, 1999, at the ENT Department, Medical University of Lübeck, Lübeck, Germany. For further information, contact PD Dr Dr R. Siegert/M. Haase, Department of Otorhinolaryngology and Plastic Head and Neck Surgery, Medical University of Lübeck, Ratzeburger Allee 160, D-23538 Lübeck, Germany; phone, 49-451-5003189; fax, 49-451-500-4192; e-mail, [email protected].
Despite the fact that peritonsillar abscess is the most common complication of acute tonsillitis, the treatment of peritonsillar abscess remains controversial. One element of controversy is the choice of antibiotics after drainage of the abscess. In an attempt to assess the effect of antibiotic choice on the treatment of peritonsillar abscess, we conducted a retrospective review of records from patients with peritonsillar abscess treated with incision and drainage. Our review identified 103 patients, comprising two groups: 58 patients treated with broadspectrum intravenous antibiotics and 45 patients treated with intravenous penicillin alone. These patients were hospitalized after incision and drainage, and therefore their clinical courses and responses to therapy could be rigorously assessed. Characterization of illness based on patient age, temperature, and white blood cell count revealed similar severity of illness between the two groups. Comparison of clinical outcomes with respect to hours hospitalized (mean 44.3 ± 6.6 and 38.3 ± 7.1 hours, 95% confidence interval, for broad-spectrum and penicillin groups, respectively) and mean hours febrile (16.9 ± 5.0 and 13.3 ± 4.2 hours, 95% confidence interval) were not statistically significantly different (p = 0.222 and 0.269, respectively) between groups, indicating that broad-spectrum antibiotics failed to show greater efficacy than penicillin in the treatment of these patients. The microbiologic characteristics of these infections, failures of therapy, and complication rates were similar to those reported in the literature. These results suggest that intravenous penicillin remains an excellent choice for therapy in cases of peritonsillar abscess requiring parenteral antibiotics after drainage. (Otolaryngol Head Neck Surg 1999;120:57-61.) From the Department of Otolaryngology–Head and Neck Surgery, Massachusetts Eye and Ear Infirmary (Dr. Kieff, Dr. Siegel, Dr. Salman) and the Joint Center for Otolaryngology (Dr. Bhattacharyya), Harvard Medical School. Reprint requests: Neil Bhattacharyya, MD, Joint Center for Otolaryngology, 333 Longwood Ave., Boston, MA 02115. Copyright © 1999 by the American Academy of Otolaryngology– Head and Neck Surgery Foundation, Inc. 0194-5998/99/$8.00 + 0 23/1/88637
Peritonsillar abscess (PTA) occurs with an incidence of 30 per 100,000 in the United States among patients 5 to 59 years of age, accounting for approximately 45,000 cases per year. Although PTA is the most common complication of tonsillitis, its treatment varies considerably among otolaryngologists.1-5 Some advocate incision and drainage as the initial treatment, others needle aspiration, and still others quinsy tonsillectomy. In addition, the antibiotics chosen to treat PTA vary widely, ranging from penicillin, an agent with activity against oropharyngeal gram-positive aerobes and anaerobes, to ampicillin/sulbactam, an agent with broad gram-positive and gram-negative aerobic and anaerobic activity.6 Despite the expense of expanded-spectrum antibiotics, pressure for their use comes from the pharmaceutical industry, as well as many reports of resistance among the pathogens typically found in PTA. Penicillin resistance has been reported to occur in more than 50% of cases.1 Herzon1 attempted to establish treatment guidelines for PTA. He proposed outpatient needle aspiration in conjunction with oral penicillin therapy on the basis of an analysis of a cohort study of 123 patients, a practitioner survey, and a meta-analysis of prior studies. However, in spite of his extensive work, he found that the selection of antibiotic therapy for PTA remains controversial. To date, no investigator has reported on the effect of antibiotic therapy as an independent variable in treatment outcome. We report a retrospective study comprising 103 patients and comparing penicillin therapy with broad-spectrum antibiotic therapy in cases of PTA first treated with incision and drainage. The relevant microbiologic findings are also addressed. METHODS AND MATERIAL We screened admissions to our hospital from January 1993 to June 1996 for patients with the diagnosis of PTA. This contemporary time period was chosen to keep the results relevant to current antibiotic resistance patterns. Admissions were included in the study if the patient’s age was between 16 and 65 years; pediatric patients were excluded because they are treated according to a different protocol at our institution. Patients with significant comorbid factors such as diabetes, immunosuppression, or HIV infection that might alter their response to infection were also excluded. Medical records were analyzed for the following variables: 57
58
Otolaryngology– Head and Neck Surgery January 1999
KIEFF et al
Table 1. Demographics and variables characterizing patient groups with peritonsillar abscesses Broad-spectrum (n = 58) Variable
Age (yr) Maximum temperature (°F) Maximum WBC count (1000/mm3) % Quinsy tonsillectomy Hospitalization (hr) Febrility (hr) % Recrudescence of abscess
Mean ± 95% CI
28.1 ± 2.4 99.8 ± 0.8 14.9 ± 1.3 6.9 ± 6.7 44.3 ± 6.6 16.9 ± 5.0 3.4 ± 4.8
Penicillin alone (n = 45)
Range
Mean ± 95% CI
16.0–62.0 97.0–102.6 7.9–30.3
31.8 ± 2.5 99.8 ± 0.3 14.3 ± 0.9 0 38.3 ± 7.1 13.3 ± 4.2 6.6 ± 7.5
12.0–136.0 0.0–96.0
Range
18.0–53.0 98.6–102.2 9.0–24.2 12.2–120.0 0.0–48.0
CI, Confidence interval; WBC, white blood cell.
age, use of oral antibiotics before admission, in-hospital antibiotic therapy, maximum temperature, and maximum white blood cell count during hospitalization, duration of hospitalization, recrudescence (recollection of fluid in the abscess cavity after drainage) of the abscess within 1 week, quinsy tonsillectomies, and gram-stain and culture results from the abscess fluid. The hours the patient was febrile were taken from time of emergency department evaluation until the time of the last recorded fever. Patients with PTA at our institution are uniformly treated with incision and drainage in the emergency department by an otolaryngologist, followed by admission for intravenous antibiotic therapy and hydration until discharge. Hours of hospitalization were recorded from time of incision and drainage to time of discharge. Discharge criteria include demonstration of adequate oral intake (>750 ml in the preceding 8 hours) and the absence of fever. Recurrences within 1 week were classified as recrudescences of the primary abscesses, and these, as well as readmissions to the hospital, were noted. On the basis of intravenous antibiotic therapy, the patients were stratified into two groups after incision and drainage: those who received intravenous penicillin therapy and those who received broader spectrum intravenous therapy. We then statistically analyzed the results for each group and compared them using a spreadsheet-based statistical package. RESULTS
Review of records identified 103 patients with PTA who qualified for the study. Of these, 58 patients received broad-spectrum intravenous antibiotics and 45 received intravenous penicillin alone. Summary data for the two groups are represented in Table 1. All patients underwent incision and drainage in the emergency department. The procedure was performed after topical and infiltration anesthesia, needle localization of the abscess, and mucosal incision and blunt opening of the abscess cavity. Intravenous antibiotic therapy and hydration were instituted on admission for all patients reviewed. The group of 58 patients treated with broad-spectrum antibiotics included patients who received single agents such as clindamycin (43 patients) or ampi-
cillin/sulbactam (9 patients), as well as patients who received combination therapy such as clindamycin plus a second-generation cephalosporin or ampicillin/sulbactam (2 patients), in addition to 2 patients who received penicillin and metronidazole. Two additional patients received combination first- and second-generation cephalosporins. In this group, age ranged from 16 to 62 years, with a mean of 28.2 years. Six patients’ abscesses recurred: Four were treated with quinsy tonsillectomy (6.9%), and two patients (3.4%) underwent repeat incision and drainage. Specifically, deep neck space infection requiring external drainage developed in none of the patients, and no patient had systemic sepsis, Clostridium difficile colitis, or delayed postdrainage hemorrhage. There were no adverse reactions to the administered antibiotics. In this group, one patient’s intravenous antibiotic regimen was changed (1.7%), from clindamycin to ampicillin/sulbactam after 2 days of worsening fever and trismus while the patient received the former drug. The clinical outcome variables results: hours of hospitalization and hours febrile for these patients are also listed in Table 1. Microbiologic data were available for 48 of the 58 patients (82.8%). There was no growth on culture in nine patients (18.8% of those cultured); the remaining culture results are displayed in Table 2. No adjustments of antibiotics were made on the basis of culture results; patients were either responding to therapy or were switched to alternate antibiotics before culture data were available. All patients were discharged home with oral antibiotics consisting of amoxicillin/clavulanate in 13 cases, clindamycin in 39 cases, penicillin in 3 cases, and other antibiotics in 3 cases. No patients returned to the hospital for near-term recurrence (recrudescence) of their abscess. The other group of 45 patients was treated with intravenous penicillin. The average age in this group was 31.8 years (range 18 to 53 years). Three patients (6.7%) suffered recrudescence of their abscesses and had another drainage procedure; none underwent quinsy
Otolaryngology– Head and Neck Surgery Volume 120 Number 1
KIEFF et al
59
Table 2. Microbiologic findings in PTA patients Broad-spectrum group Isolate
No.
Penicillin group % of Total cultured
α-Hemolytic Streptococcus Anaerobic gram-negative rods Bacteroides spp. β-Hemolytic non–group A Streptococcus Coagulase-negative Staphylococcus spp. Fusobacterium Group C Streptococcus Haemophilus influenzae Pseudomonas aeruginosa Staphylococcus aureus Streptococcus pyogenes
13 1 4 5 1 6 3 3 1 5 12
27.1 2.1 8.3 10.4 2.1 12.5 6.3 6.3 2.1 10.4 25.0
More than one organism No growth Not cultured
15 9 10
31.3 18.8 17.2
tonsillectomy. No complications occurred during the hospital course in this group, and there were no adverse drug reactions. Three patients’ antibiotic therapies were altered during hospitalization this group. The changes were from penicillin to clindamycin in two cases, and from penicillin to ampicillin/sulbactam in one case. These changes were necessitated by lack of response to penicillin alone, manifested by worsening pain, fever, and trismus. The clinical outcome variables results for these patients are also listed in Table 1. Microbiologic data were available for 36 of the 45 patients (80.0%). There was no growth on culture in two patients (5.6% of those cultured); the remaining culture results are displayed in Table 2. Again, no adjustments of antibiotics were made on the basis of culture results. All patients were discharged home with oral antibiotics consisting of penicillin in 37 cases, amoxicillin/clavulanate in 2 cases, clindamycin in 2 cases, and amoxicillin in 4 cases. No patient returned to the hospital for recrudescence of an abscess. Comparison of the data for the two groups revealed similar parameters characterizing the groups and the severity of their presenting PTA. The mean ages of the two groups were very similar (28.2 vs. 31.8 years). No statistically significant difference was demonstrated between groups for the maximum temperatures (p = 0.735, two-tailed Student’s t test), and highest white blood cell counts (p = 0.434), suggesting that the two groups had similar severities of disease. The outcome variables for the two groups were compared with the use of a two-tailed Student t distribution. This demonstrated no statistical difference between groups with respect to
Isolate
No.
α-hemolytic Streptococcus 15 Anaerobic gram-negative rods 1 Bacteroides spp. 4 β-Hemolytic non–group A Streptococcus 4 Clostridium spp. 1 Coagulase-negative Staphylococcus spp. 1 Fusobacterium 4 Group C Streptococcus 2 H. influenzae 2 Moraxella catarrhalis 1 Nocardia spp. 1 P. aeruginosa 1 S. aureus 2 S. pyogenes 14 More than one organism 13 No growth 2 Not cultured 9
% of Total cultured
41.7 2.8 11.1 11.1 2.8 2.8 11.1 5.6 5.6 2.8 2.8 2.8 5.6 38.9 36.1 5.6 20.0
duration of hospitalization (p = 0.222) and hours to defervescence (p = 0.269). The recrudescence rate (Table 1) was slightly less for patients treated with broad-spectrum agents, but this was not statistically significant. DISCUSSION
The incidence of PTA in the United States has been reported to be 30 per 100,000 person-years, accounting for approximately 45,000 cases and an estimated cost of greater than $32 million annually.1 Treatment of PTA remains varied among practitioners, and it is controversial despite several studies on the subject.2,7 Incision and drainage of PTA was the most common form of management in the United States before 1960 because of the hypothetical risk of producing rheumatic fever or glomerulonephritis from septicemia after abscess tonsillectomy. Volk and Brandow’s report in 1960 paved the way for a change in management of PTA to an “abscess,” or “quinsy” tonsillectomy.8 Although quinsy tonsillectomy became the favored practice on the basis of the impression that it led to the quickest recovery, subsequent studies suggested that antibiotic therapy with abscess aspiration or incision and drainage were equally effective as quinsy tonsillectomy and antibiotic therapy.5 A corollary of these studies was that needle aspiration of PTA yielded equivalent results to incision and drainage. Resurrecting this controversy, a recent retrospective study by Wolf et al.3 refuted the conclusion that aspiration of PTA is as efficacious as incision and drainage. The impetus for the use of broader-spectrum antibiotics in PTA stems from two arguments. First, there have been several reports of increased bacterial resis-
60
KIEFF et al
tance in pharyngotonsillar infections, leading to concern that broader-spectrum antibiotics are needed for effective treatment of the disease.4 Whereas group A βhemolytic streptococci have been classically associated with PTA and are universally penicillin-sensitive, penicillin-resistant aerobes and anaerobes, as well as gramnegative organisms, have been detected in PTA cultures. Brook et al.9 noted that 52% of 25 PTA specimens grew β-lactamase–producing organisms. The abscess is often polymicrobial, and the organisms include Staphylococcus aureus and Bacteroides fragilis, as well as Bacteroides melaninogenicus, a common oral anaerobe. Such organisms are able to survive penicillin therapy and can theoretically protect penicillin-sensitive organisms by releasing penicillinase enzyme into the abscess microenvironment. Therefore Brook et al. believed that antibiotic therapy against PTA should be both penicillinase-stable and effective against anaerobes. A national survey indicated that 52% of otolaryngologists use broad-spectrum agents in first-line treatment of PTA.1 Second, the efficacy of penicillin therapy has also been questioned because of the finding that it does not penetrate well into the abscessed tonsil, whereas cephalosporins are preferentially concentrated in abscessed tonsillar tissue.10 Despite these two theoretical concerns regarding penicillin therapy for PTA, no outcome data demonstrate an advantage of broad-spectrum antibiotic therapy. Indeed, our analysis of 103 patients treated for PTA failed to demonstrate a statistically significant advantage of broad-spectrum antibiotic therapy over penicillin. The patient populations in the broad-spectrum and penicillin groups were not significantly different in their age, maximum temperature, or white blood cell profiles; nor were they significantly different in outcome with respect to hours hospitalized, hours febrile, need for quinsy tonsillectomy, or recurrence rate. The microbiologic data were similar between groups and are consistent with those reported by others: Streptococcus pyogenes and α-hemolytic Streptococcus species predominated.2,9,11 We do not routinely culture for anaerobes because of the cost and the fact that rarely have we needed to adjust antibiotic therapy for anaerobes that were not sensitive to empiric antibiotics. In this series, very few patients had their antibiotics changed during therapy. In the patients whose antibiotics were changed (four total), the choice of the new antibiotic was not based on specific culture results but broadened empirically, suggesting that clinical response is more significant than microbiologic data in the selection of antibiotic therapy and that routine culture of the aspirate may be of relatively limited value. Because no reports have compared the efficacy of incision and drainage alone vs.
Otolaryngology– Head and Neck Surgery January 1999
incision and drainage of PTA with antibiotic therapy, and because several of the patients who responded to penicillin in our study grew resistant organisms on culture, it is possible that incision and drainage without antibiotic therapy is adequate treatment of PTA. On theoretical grounds, we advocate antibiotic therapy, and this remains the standard of care.1 Regardless, these data suggest that penicillin therapy is as efficacious as broader-spectrum therapy in the treatment of drained PTA. Given the retrospective nature of this study, several sources of bias may exist. A bias may exist toward treating patients in whom PTA develops while they are being treated with oral antibiotics, especially penicillin, with broader-spectrum antibiotics. In our groupings however, the percentages in both groups of patients who were treated with oral antibiotics before PTA developed were similar. The culture analysis showed similar types of organisms and positive culture rates, regardless of prior antibiotic therapy, indicating the minimal effect of antibiotic therapy on the microbiology of a developing PTA. Interestingly, more cultures were sterile in the broad-spectrum group. This finding should have favored the broad-spectrum group because they theoretically had fewer active organisms at the time of drainage, yet no clinical advantage for broad-spectrum agents was shown. Moreover, the selection bias of sicker patients’ receiving broader-spectrum antibiotics is adjusted for by the similar severities of illness between the two groups, based on maximum temperature and maximum white blood cell count. It is important to note that the admission of patients with PTAs for intravenous antibiotic therapy is becoming less common.1 During the period of this study, all PTAs were treated with incision and drainage, followed by admission for observation, intravenous antibiotics, and fluids. This permitted close follow-up of these patients with respect to temperature, time to adequate oral intake, and need for further procedures. Nevertheless, we expect that the conclusions from our results can be applied to those patients who undergo aspiration and intravenous therapy and, possibly, to those patients undergoing aspiration and outpatient oral antibiotic therapy. It could be argued that broad-spectrum antibiotics should be used when oral therapy is chosen after PTA drainage because the high blood levels of intravenous penicillin may have overwhelmed possible penicillin resistance in this study; however, this argument remains speculative. In fact, our results showed that a greater number of patients (six) in the broad-spectrum group required quinsy tonsillectomy for failure of incision and drainage than did patients in the penicillin group. At our institution, the choice between quinsy tonsillectomy and repeat incision and drainage for PTAs that
Otolaryngology– Head and Neck Surgery Volume 120 Number 1
fail initial drainage and antibiotic therapy is usually based on physician preference. A bias may have existed in treating recrudescences in the broad-spectrum group with quinsy tonsillectomy as opposed to repeat incision and drainage out of concern for continued failure of nonoperative therapy. However, the difference in rates of quinsy tonsillectomies between the two groups was not of statistical significance. In fact, the difference in the rate of quinsy tonsillectomy is also offset by the difference in the rate of recrudescence of the abscess (the penicillin group had a higher rate of recrudescence) so that, taken together, the rate of quinsy tonsillectomy and repeat drainage are about equal between groups. CONCLUSIONS
A cost-effective approach should be adopted for the treatment of PTA. Despite the fact that PTA is a polymicrobial infection that may involve penicillin-resistant organisms, intravenous penicillin is as clinically effective as broad-spectrum antibiotics in cases of PTA, as demonstrated by this study of 103 patients. Because penicillin remains one of the least expensive antibiotic therapies and is well tolerated, we recommend that it be used as the first-line treatment of PTA after incision and drainage, especially when intravenous hydration is required. Achieving effective abscess drainage is a key element in therapy. Cultures of abscess fluid have little
KIEFF et al
61
bearing on the clinical management of PTA, and the routine use of such cultures should be reconsidered. Further prospective studies addressing the comparative clinical effectiveness of various oral antibiotics will help encourage safe and cost-effective therapy. REFERENCES 1. Herzon FS. Peritonsillar abscess: incidence, current management practices, and proposal for treatment guidelines. Laryngoscope 1995;105(Suppl):1-17. 2. Savolainen S, Jousinies-Somer HR, Makitie AA, et al. Peritonsillar abscess: clinical and microbiologic aspects and treatment regimens. Arch Otolaryngol Head Neck Surg 1993;119:521-4. 3. Wolf M, Even-Chen I, Kronenberg J. Peritonsillar Abscess: repeated needle aspiration versus incision and drainage. Ann Otol Rhinol Laryngol 1994;103:554-7. 4. Poole MD. Penicillin may no longer be the drug of choice for Streptococcal pharyngitis. ENT J 1993;4:330-2. 5. Parker GS, Tami TA. Management of peritonsillar abscess in the 90s: an update. Am J Otolaryngol 1992;13:284-8. 6. Fairbanks DN. Antimicrobial therapy in otolarygology–head and neck surgery. 7th ed. Alexandria, Va.: American Academy of Otolaryngology–Head and Neck Surgery; 1993, p 9-11. 7. Maharaj M, Rajah V, Hemsley S. Management of peritonsillar abscess. J Laryngol Otol 1991;105:743-5. 8. Volk BM, Brandow EC Jr. Bilateral tonsillectomy for peritonsillar abscess. Trans Am Laryngol Rhinol Otol Soc 1960;63:776-85. 9. Brook I, Frazier E, Thompson DH. Aerobic and anaerobic microbiology of peritonsillar abscess. Laryngoscope 1991;101:289-92. 10. Maisel RH. Peritonsillar abscess: tonsil antibiotic levels in patients treated by acute abscess surgery. Laryngoscope 1982;92:80-7. 11. Jokinen K, Sipila P, Jokipii L, et al. Peritonsillar abscess: bacteriological evaluation. Clin Otolaryngol 1985;10:27-30.
Otoplasty and Reconstruction of Auricular Defects
An International Course on Otoplasty and Reconstruction of Auricular Defects will be held March 21-22, 1999, at the ENT Department, Medical University of Lübeck, Lübeck, Germany. For further information, contact PD Dr Dr R. Siegert/M. Haase, Department of Otorhinolaryngology and Plastic Head and Neck Surgery, Medical University of Lübeck, Ratzeburger Allee 160, D-23538 Lübeck, Germany; phone, 49-451-5003189; fax, 49-451-500-4192; e-mail, [email protected].