PEDIATRICS Vol. 118 No. 6 December 2006, pp. e1607-e1611 (doi:10.1542/peds.2006-1025)
ARTICLE |
Pharyngitis in Low-Resources Settings: A Pragmatic Clinical Approach to Reduce Unnecessary Antibiotic Use
a Serviço de Pediatria, Hospital Universitário de Brasília, Universidade de Brasília, Brasília, Brazil
b Département des Maladies Infectieuses, Hôpital Universitaire des Enfants Reine Fabiola
c Laboratoire de Génétique des Procaryotes, Institut de Biologie et de Médecine Moléculaire
e Unité d'épidémiologie et d'hygiène hospitalière, Hôpital Brugmann, Université Libre de Bruxelles, Brussels, Belgium
d Serviço de Microbiologia, Hospital Regional da Asa Sul, Brasília, Brazil
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVE. Existing scoring systems for the diagnosis of group A streptococcus pharyngitis are insensitive or inapplicable in low-resources settings. Bacterial cultures and rapid tests can allow for antibiotic prescription abstention in high-income regions. These techniques are not feasible in many low-resources settings, and antibiotics often are prescribed for any pharyngitis episode. However, judicious antibiotics prescription in the community also is of concern in low-income countries. The objective of this study was to develop a clinical decision rule that allows for the reduction of empirical antibiotic therapy for children with pharyngitis in low-resources settings by identifying non–group A streptococcus pharyngitis.
PATIENTS AND METHODS. We prospectively included children with pharyngitis in 3 public hospitals of Brazil during 9 months in 2004. We filled out clinical questionnaires and performed throat swabs. Bilateral 
2 (2-tailed test) and multivariate analysis were used to determine score categories. The outcome measures were sensitivity, specificity, positive likelihood ratio, and posttest probability of non–group A streptococcus infection with the clinical approach as compared with throat culture.
RESULTS. A total of 163 of the 220 children had non–group A streptococcus pharyngitis (negative culture). We established a 3-questions decision rule (age and viral and bacterial signs) with 3 possible answers. The use of this score would prevent 41% to 55% of unnecessary antimicrobial prescriptions. The specificity of the score for non–group A streptococcus pharyngitis was >84%.
CONCLUSION. Such a clinical decision rule could be helpful to reduce significantly unnecessary antibiotic prescriptions for pharyngitis in children in low-resources settings.
Key Words: pharyngitis • practice guidelines • antibiotic use • developing countries
Abbreviations: GAS—group A streptococcus
The global burden of antimicrobial resistance is particularly pressing in developing countries.1 The level of antibiotic resistance is correlated clearly to antibiotic consumption.2–6 In Western Europe and North America, modifications of prescribing practice have been attained with diagnosis and prescribing guidelines.7 These strategies have not been evaluated in developing countries, where health care systems differ greatly from occidental ones, so locally validated interventions are needed.8
Acute pharyngitis is 1 of the most common childhood illnesses. Viruses are responsible for >80% of them, and group A Streptococcus (GAS) accounts for 
15% of these infections.9 In cases of GAS as the cause, antibiotics are prescribed frequently.9 Recent studies underlined the high proportion (20%–50%) of GAS carriers among children, and distinguishing between carriers with viral pharyngitis and real GAS pharyngitis is impossible in general practice.10
In low-income regions, there is no access to microbiologic diagnosis of common GAS infections such as pharyngitis. Presumptive diagnosis relies on clinical signs only, although the positive predictive value of clinical signs and symptoms has been shown to be extremely low.11 The World Health Organization clinical decision rule for streptococcal pharyngitis suggests treating children who are younger than 5 years when pharyngeal exudate plus enlarged, tender cervical nodes are found. The evaluation of this approach was performed in Brazil, Egypt, and Croatia and presented a very low sensitivity (3.6%–8.5%).12 Some clinical scores, developed in medical settings where microbiology is available in the daily practice, have already been published.13–15 Because of the unavailability of microbiologic testing in low-income regions, clinicians often empirically treat all pharyngitis with antibiotics. In these settings, there clearly is a need for modified clinical decision rules with a higher sensitivity and adequate specificity. To our knowledge, only 1 interesting study proposed a clinical decision rule that is designed for the management of streptococcal pharyngitis in low-incomes regions (Egypt).16 However, it has not yet been validated clinically, and we have no data so far that it could be applied in other low-income settings.
Bacterial culture and rapid tests are not available for common infections in public hospitals of Brazil (for structural and cost reasons). Like many other public institutions in developing countries, all children with clinical pharyngitis are treated empirically with antibiotics. We developed a clinical decision rule that allows for a reduction in empirical antibiotic therapy for children with pharyngitis in Brazil, as an example of a low-resource settings. We directed our clinical analysis for best identifying non-GAS pharyngitis rather than GAS pharyngitis.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Patients
All children (0–15 years) who attended the emergency department of 3 public hospitals of Brazil (Academic hospital of Brasília, Regional Hospital of the south wing of Brasília, and Medical Unit of São Sebastião, with 250, 180, and 40 beds, respectively) with signs and symptoms of acute pharyngitis from February 1 to October 31, 2004, were enrolled prospectively in the study. This study was approved by the ethical board of all participant hospitals. Exclusion criteria were current antimicrobial treatment or refusal to participate in the study. A written informed consent was obtained from the children's parents or guardian. A predesigned questionnaire that comprised general data (date of birth, gender, antimicrobial treatment in the past 6 months, and the prescribed treatment for current episode), demographic information (number of children living under the same roof), and description of infection was used for data collection. Signs and symptoms of pharyngitis were derived from the practice guidelines of the Infectious Diseases Society of America and included the presence of (1) conjunctivitis, coryza, cough, diarrhea, and viral-like exanthema, features suggestive of viral cause and defined as viral signs; or (2) fever >38.5°C, tender cervical node, headache, petechia on the palate, abdominal pain, and sudden onset (<12 hours), features suggestive of GAS as causative agent and defined as bacterial signs.9 The pediatricians filled out the questionnaire, measured the weight and height, and collected the throat secretion samples using sterile cotton-tipped swabs. Physicians were trained to standardize the swabbing technique.
Laboratory Analysis
The swabs were plated on 5% sheep blood agar plates that were incubated at 37°C for 24 to 48 hours. GAS were identified by a ß hemolysis on blood agar, colony morphology, Gram stain, catalase reaction, sensitivity to 0.04-U bacitracin disk, latex agglutination (Slidex, strepto A, Biomérieux, France), and a commercially available method following the manufacturer's recommendations (MicroScan, Dade Behring, São Paulo, Brazil).
Statistical Analysis
We used the data that were collected in this study to develop the clinical decision rule. The association of clinical findings with negative throat culture was assessed with the use of a bilateral 2 (2-tailed test). Continuous variables also were assessed using a t test analysis (Gaussian distribution). The 3 more significant items (age, viral signs, and bacterial signs) were retained for multivariate analysis (using SPSS 11.0; SPSS, Chicago, IL). The weighting of each question was based on the ß value of the regression coefficients and optimized for best results with our data. The outcome measures were sensitivity, specificity, positive likelihood ratio, and posttest probability of non-GAS infection with the clinical approach as compared with throat culture.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
General Data
A total of 220 patients were included during the study period. The mean ± SD age was 6.6 ± 2.8 years, and 56% were male. An antimicrobial treatment had been prescribed in the past 6 months for 80 (36%) patients. Most (90%) of them had received benzathine penicillin or amoxicillin. The mean number of children who lived under the same roof was 2.2 (range: 1–8 children). Among the 220 patients, 163 (74%) had no GAS isolated from their throat, whereas 57 (26%) had a pharyngeal positive culture. A GAS-positive culture was recovered in 0%, 18%, and 33% of children who were 0 to 36 months of age, 37 to 59 months of age, and older than 59 months, respectively (P = .003). All children were treated empirically with antibiotics independent of the bacteriologic result. Most (83%) received intramuscular benzathine penicillin. Other treatments included amoxicillin (9%), macrolides (3%), cephalexin (3%), and amoxicillin with clavulanate (2%).
Clinical Decision Rule
The 3 questions (age, viral signs, and bacterial signs) of our decision rule had a ß value of the regression coefficients of 2.8, 0.8, and 1.0, respectively. The scoring system is presented in Table 1, and the decision rule to apply according to the score is shown in Table 2.
|
|
The classification of our children in our study according to this clinical approach and the result of the throat culture is shown in Table 3. A cutoff of 8 was chosen as having the optimal risk/benefit ratio. The discriminative ability of the model at the cutoff value
8 was calculated according to sensitivity, specificity, positive likelihood ratio, and posttest probability in Table 4.
|
|
We designed 2 different decision rules depending on the accessibility to microbiologic diagnosis. A clinical score
8 suggests a symptomatic treatment with an 84% specificity for non-GAS pharyngitis. Those children had a likelihood ratio for non-GAS pharyngitis of 2.6, which give a posttest probability for non-GAS pharyngitis of 88%. Nine (16%) of 57 children with a score 8 nevertheless had a GAS-positive culture. The clinical characteristics of these children are shown in Table 5. In medical settings where bacteriologic diagnosis is unavailable, the application of that recommendation would have reduced the antibiotic prescription by 41%.
|
In medical settings where limited bacteriologic diagnosis is available, a score <8 can be refined into 2 distinct situations. In case of a score result from 5 to 7, an antibiotic treatment would be recommended if the culture were positive for GAS. Only 16% of the children of our population would fall into this intermediate category and would require a microbiologic diagnostic confirmation. A score result <5 indicates an antibiotic treatment without bacteriologic confirmation. The application of this recommendation would have reduced the antibiotic prescription by 55%.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The prevalence of pharyngitis associated with GAS-positive culture was similar (26%) to previous reports.9 Therefore, at least 74% of the children were treated unnecessarily with antibiotics. The presence of GAS in the throat sample depended on the age of the children. As shown previously,9 the incidence of GAS pharyngitis was lower in children who were younger than 5 years than in older children. However, a significant proportion of children had a GAS-positive culture in the 37- to 59-month age group. This would lead us to propose a lower age limit (3 years) for considering a GAS cause for pharyngitis as chosen for the Canadian sore throat score and in contrast with the 5-year cutoff chosen by the Infectious Diseases Society of America. In children who were younger than 36 months, no GAS was isolated in our study. This result differs from other studies that showed the presence of GAS in the throat of children who were younger than 3 years.17,18 However, the proportion of GAS infections (or colonization?) in children who are younger than 3 years is lower and is well correlated to age in this young age group. Epidemiologic variations over time and location also have been demonstrated with GAS.19,20 We assume, until validation and with the present epidemiology, that the absence of GAS in children who were younger than 3 years in our data could justify the symptomatic treatment of all of these children.
A 41% to 55% sparing of antibiotic prescriptions by use of our clinical decision rule would represent a significant impact given the high prevalence of pharyngitis in children. As in most cases, the children were given a benzathine penicillin G shot; this would also mean a significant reduction of the pain associated with this treatment and a decrease of the financial cost related to the antibiotic injection.
The posttest probability for non-GAS pharyngitis was high (corresponding to a theoretical positive predictive value of 88.1%), even without taking into account the GAS carriers among these children. The use of that pragmatic approach in our population would have resulted in antibiotic abstention for 9 (16%) children with pharyngitis that was associated with a GAS-positive culture. However, this seems acceptable because, first, 16% is lower than the general pediatric GAS carriage rates (20%–50%).10 Second, the sensitivity of throat culture is limited to a maximum of 90% in study settings (this could be lower in the day-to-day medical practice), and the culture does not distinguish between carriage and infection. Serologic examinations were not performed in this study, and we do not know whether these 9 children were infected by GAS or only carriers. In an additional validation study, the capacity of this scoring system to discriminate between carriage and infection will be investigated in the patients with high score and positive culture. The true specificity of our clinical decision rule could be underestimated in the present study if such patients were carriers.
Rheumatic fever still is highly prevalent in many developing countries (one third of the Brazilian heart surgeries in 2004).21 In pediatric patients, the efficacy of primary prevention of rheumatic fever by antibiotic treatment of GAS pharyngitis has not been demonstrated firmly.22 Some studies showed a positive effect of this prevention, but design flaws and use of atypical populations limit the applicability of these studies to children in developing countries.23 However, a careful attitude has been favored and antimicrobial treatment of GAS pharyngitis probably can be justified in countries with a high prevalence of rheumatic fever. Nevertheless, the hypothetical prevention of rheumatic fever has to be balanced against the development of antimicrobial resistance and its consequences.
Another scoring system was constructed for use in children from Cairo (Egypt).16 The use of this score would have allowed for a 37.5% reduction of unnecessary antibiotic therapy with 9% of missed streptococcal cases in the study population of Cairo. When we applied this Egyptian score to our Brazilian children population, we obtained only a 20% reduction of unnecessary antibiotic therapy with 12.5% missed streptococcal cases. This variability could be attributable to epidemiologic differences between Cairo and Brazil. However, a scoring system directed for best sensitivity to non-GAS pharyngitis with an acceptable specificity theoretically could be more efficient in terms of antibiotic reduction.
There are several limitations to our study. First, the number of included children was small. Second, this clinical decision rule has to be validated on a large population of children who attend different medical settings. Third, the acceptability of this scoring system by the general physician and pediatricians has to be evaluated in the daily practice. The validation also should include bacterial serology and viral culture and serologies to improve distinction between carriage and infection.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This clinical decision rule for pharyngitis management was constructed on the basis of the reality of public pediatric practice in Brazil, which could be representative of many emerging countries of the developing world. Clinically based and locally designed guidelines in such setting could be a potent incentive for physicians to modify their prescribing practice.
|
ACKNOWLEDGMENTS |
|---|
Dr Smeesters is supported by a grant of the Belgian Kid's Foundation. The Medcorp society supported us for the bacteriologic diagnosis (MicroScan).
We thank Maria Custodia Machado Ribeiro, Denise Nogueira, Joceleia de Lira, Maria Gloria Andrade, Daniela de Sales M. dos Santos, Marne Rodrigues, Amabel Fernandes Correia, Marina Sancha Queiroga, Jeane Bezerra Rodrigues, Francisca Moura de Oliveira, Maria do Carmo Teixeira, Divina Maria Dias, Georges Casimir, and André Kahn for help in the realization of this study. We thank Philippe Lepage for critical proofreading of the manuscript. We thank all of the professionals of the 3 participating hospitals for kind collaboration. Dr Van Melderen is chercheur qualifié du Fonds National de la Recherche Scientifique.
|
FOOTNOTES |
|---|
Accepted Jul 11, 2006.
Address correspondence to Pierre Robert Smeesters, MD, Département des Maladies Infectieuses, Hôpital Universitaire des Enfants Reine Fabiola, Av JJ Crocq 15, 1020 Bruxelles, Belgium. E-mail: psmeeste{at}ulb.ac.be
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Okeke IN, Laxminarayan R, Bhutta ZA, et al. Antimicrobial resistance in developing countries. Part I: recent trends and current status. Lancet Infect Dis. 2005;5 :481 –493[CrossRef][Web of Science][Medline]
- Goossens H, Ferech M, Vander Stichele R, Elseviers M. Outpatient antibiotic use in Europe and association with resistance: a cross-national database study. Lancet. 2005;365 :579 –587[Web of Science][Medline]
- Okeke IN, Klugman KP, Bhutta ZA, et al. Antimicrobial resistance in developing countries. Part II: strategies for containment. Lancet Infect Dis. 2005;5 :568 –580[CrossRef][Web of Science][Medline]
- Rubin MA, Samore MH. Antimicrobial use and resistance. Curr Infect Dis Rep. 2002;4 :491 –497[Medline]
- Byarugaba DK. A view on antimicrobial resistance in developing countries and responsible risk factors. Int J Antimicrob Agents. 2004;24 :105 –110[CrossRef][Web of Science][Medline]
- Sirinavin S, Dowell SF. Antimicrobial resistance in countries with limited resources: unique challenges and limited alternatives. Semin Pediatr Infect Dis. 2004;15 :94 –98[CrossRef][Medline]
- Mol PG, Wieringa JE, Nannanpanday PV, et al. Improving compliance with hospital antibiotic guidelines: a time-series intervention analysis.
J Antimicrob Chemother. 2005;55
:550
–557
[Abstract/Free Full Text] - WHO. WHO Global Strategy for Containment of Antibiotic Resistance. Geneva, Switzerland: WHO; 2001:99
- Bisno AL, Gerber MA, Gwaltney JM Jr, Kaplan EL, Schwartz RH. Practice guidelines for the diagnosis and management of group A streptococcal pharyngitis. Infectious Diseases Society of America. Clin Infect Dis. 2002:35 :113 –125[CrossRef][Web of Science][Medline]
- Martin JM, Green M, Barbadora KA, Wald ER. Group A streptococci among school-aged children: clinical characteristics and the carrier state.
Pediatrics. 2004;114
:1212
–1219
[Abstract/Free Full Text] - Steinhoff MC, Abd el Khalek MK, Khallaf N, et al. Effectiveness of clinical guidelines for the presumptive treatment of streptococcal pharyngitis in Egyptian children. Lancet. 1997;350 :918 –921[CrossRef][Web of Science][Medline]
- Rimoin AW, Hamza HS, Vince A, et al. Evaluation of the WHO clinical decision rule for streptococcal pharyngitis.
Arch Dis Child. 2005;90
:1066
–1070
[Abstract/Free Full Text] - Attia MW, Zaoutis T, Klein JD, Meier FA. Performance of a predictive model for streptococcal pharyngitis in children.
Arch Pediatr Adolesc Med. 2001;155
:687
–691
[Abstract/Free Full Text] - McIsaac WJ, White D, Tannenbaum D, Low DE. A clinical score to reduce unnecessary antibiotic use in patients with sore throat. CMAJ. 1998;158 :75 –83[Abstract]
- Breese BB. A simple scorecard for the tentative diagnosis of streptococcal pharyngitis.
Am J Dis Child. 1977;131
:514
–517
[Abstract/Free Full Text] - Steinhoff MC, Walker CF, Rimoin AW, Hamza HS. A clinical decision rule for management of streptococcal pharyngitis in low-resource settings. Acta Paediatr. 2005;94 :1038 –1042[CrossRef][Web of Science][Medline]
- Amir J, Shechter Y, Eilam N, Varsano I. Group A beta-hemolytic streptococcal pharyngitis in children younger than 5 years. Isr J Med Sci. 1994;30 :619 –622[Web of Science][Medline]
- Levin RM, Grossman M, Jordan C, Ticknor W, Barnett P, Pascoe D. Group A streptococcal infection in children younger than three years of age. Pediatr Infect Dis J. 1988;7 :581 –587[Web of Science][Medline]
- Kaplan EL. Recent epidemiology of group A streptococcal infections in North America and abroad: an overview. Pediatrics. 1996;97(pt 2) :945 –948
- Carapetis JR, Currie BJ, Kaplan EL. Epidemiology and prevention of group A streptococcal infections: acute respiratory tract infections, skin infections, and their sequelae at the close of the twentieth century. Clin Infect Dis. 1999;28 :205 –210[Web of Science][Medline]
- Brazilian Ministry of Health, Sistema Único de Saúde. Ministério da Saúde. Available at: www.datasus.gov.br. Accessed September 22, 2006.
- Carapetis JR, McDonald M, Wilson NJ. Acute rheumatic fever. Lancet. 2005;366 :155 –168[CrossRef][Web of Science][Medline]
- Robertson KA, Volmink JA, Mayosi BM. Antibiotics for the primary prevention of acute rheumatic fever: a meta-analysis. BMC Cardiovasc Disord. 2005;5 :11[CrossRef][Medline]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Karthikeyan and B. M. Mayosi Is Primary Prevention of Rheumatic Fever the Missing Link in the Control of Rheumatic Heart Disease in Africa? Circulation, August 25, 2009; 120(8): 709 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. C. Maltezou, V. Tsagris, A. Antoniadou, L. Galani, C. Douros, I. Katsarolis, A. Maragos, V. Raftopoulos, P. Biskini, K. Kanellakopoulou, et al. Evaluation of a rapid antigen detection test in the diagnosis of streptococcal pharyngitis in children and its impact on antibiotic prescription J. Antimicrob. Chemother., December 1, 2008; 62(6): 1407 - 1412. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
