Journal of Medicine and Biomedical Research, Vol. 7, No. 1 & 2, December 2008, pp. 37-44
Bacterial isolates of Tonsillitis and Pharyngitis in a Paediatric Casualty Setting
1 Wilson E. Sadoh, 2 Ayebo E. Sadoh , 3 Adegbenro O. Oladipo, 4Olusola O. Okunola
1 Department of Child Health, Lecturer/ Consultant
Code Number: jm08006
A prospective study was carried out to determine the pattern of bacterial isolates and their antibiotic sensitivity amongst children with tonsillopharyngitis. Consecutive children presenting with sorethroat, difficulty with swallowing, fever and/ or evidence of inflamed pharynx and/ or tonsils at the paediatric casualty of the University of Benin Teaching Hospital, Benin City, between February and October 2006 were recruited for the study. The patient's biodata were obtained and socioeconomic status was determined. Throat swabs were taken for microbiologic analysis. Seventy three throat swabs were analysed. Bacteria were isolated from 39 patients. Of which 19 (48.72%) were β haemolytic Streptococcus (BHS). others were S. aureus five (12.83%), seven (17.95%) were Klebsiella mirabilis and three (7.69%) each of Pseudomonas aeroginosa and Proteus mirabilis. BHS and S. aureus showed 100% sensitivity to cefuroxine, azithromycin, ceftazidine and genticin. All the isolates had little or no sensitivity to ampicillin and cotrimoxazole. BHS is a significant cause of pharyngitis and tonsillitis in our environment and therefore poses a potential danger of rheumatic fever and rheumatic heart disease, a non-suppurative sequalae of BHS. Ampicillin and cotrimoxazole two affordable and commonly available drugs are ineffective in tonsillitis and pharygitis.
KEYWORDS: Tonsillitis, pharyngitis, streptococcus
Pharyngitis and tonsillitis are two common upper respiratory tract infections(URTI) causing ill health in children presenting to primary care physicians.1,2 Although most cases of pharyngitis/ tonsillitis are viral in origin, bacterial causes are important because of the non suppurative sequalae like rheumatic fever and rheumatic heart disease in group Aβ hemolytic Streptococcus (GABHS) infection. Rheumatic fever (RF) and its cardiac complication of rheumatic heart disease (RHD) remain major health problems in developing countries4,5 .
There have been reported outbreak of streptococcal infections amongst 6 schoolchildren in the United states. The introduction of an invasive strain into a strongly associated with the outbreaks 6. Such outbreak contributed to increased morbidity including rheumatic fever and RHD.
In Jos, Nigeria, streptococcus pneumoniae accounted for over one third of bacterial isolates causing URTI in under five children.7 The streptococcal throat carriage rate was 9.78% amongst primary school pupils in Benin, Nigeria.8 This suggests significant probability of rheumatic fever and rheumatic heart disease among children in this community.
Early treatment of GABHS URTI has been shown to prevent rheumatic fever. 9 It is therefore important to evaluate the pattern of bacterial URTI, determine the frequency of GABHS involvement and their current susceptibility to various antibiotics therapy. This will facilitate formulation of empirical antibiotic therapy especially in clinical setting with poor microbiological facilities. A prospective study was therefore carried out to determine the pattern of bacterial isolates causing tonsillitis/ pharyngitis and their antibiotic sensitivity at the paediatric casualty of the University of Benin Teaching Hospital, Benin City, Nigeria.
PATIENTS AND METHODS
Consecutive children presenting with pharyngitis or tonsillitis or both at the paediatric casualty of the University of Benin Teaching Hospital, Benin City, between February and October 2006 were recruited for the study. Inclusion criteria: patients with a history of sorethroat, difficulty swallowing, fever (Including a recorded temperature of > 38.0° at presentation) and/ or evidence of inflamed tonsils/ pharynx (presence of erythema and exudates on the pharynx and/ or tonsils on examination of the throat and anterior cervical adenitis). Cervical adenitis was adjudged present if the anterior cervical lymph nodes were tender and palpable. All patients were seen by the attending paediatrician (Who was a senior registrar or a registrar. They were trained by one of the authors on how to make the diagnosis and obtain throat swab). Patients who were receiving or had received antibiotics within two weeks of presentation were excluded.
The patients’ biodata of age and sex were obtained and socioeconomic status (SES) was determined by the method described by Olusanya et al.10 The clinical features in each case were documented. Patients who had hyperaemia, exudates or ulcers on the pharyngeal wall (excluding the tonsils) were said to have pharyngitis, while if these findings were restricted to the tonsils, a diagnosis of tonsillitis was made. When the features were found in both areas, the term tonsillopharyngitis was used.
Throat swabs were taken by the attending paediatrician from each patient using a sterile swab stick. The swab stick was introduced into a well exposed mouth with the tongue depressed by a wooden spatula where necessary and visible exudates or hyperaemic areas on the pharyngeal or tonsillar walls were swabbed. The swab stick was carefully replace in its sheath and sent for microbiological analysis within 1 hour or kept in Stuart medium before being analysed at the research laboratory of the Child Health department of the same hospital. The microbiological analysis was done by a medical laboratory scientist. Swabs were simultaneously plated on a blood agar, Mac Conkey and Chocolate agar media and incubated at 37°C for 48 hours. Growths were further sub-cultured to identify species. Cultures were considered negative if there was no significant bacterial growth after 48 hours. Growths of Streptococcus pyogenes, Staphylococcus aureus, Klebsiella aerogenes, Peudomonas aerogenosa, and Eshericheae coli on any of the media were identified by standard methods. ß hemolytic streptococcal (BHS) colonies could not be classified further into their Lancefield groups because of non availability of this facility. The sensitivity of the isolates was tested against ten common antibiotics, using Biotec gram positive and negative discs (Batch number C124679).
Data were analysed using SPSS 11.0 version. Simple proportions were represented in percentages. The sensitivity (The number of patients with bacteria isolates as a percentage of patients with the symptom) and specificity (The number of patient without bacterial isolate as a % of patients without the symptom) of each of the clinical symptom/ sign in predicting possible bacteria upper respiratory tract infections were determined. Chi-square was used to test associations and Student's t test to test the difference between means. p < 0.05 was considered statistically significant.
There were 73 patients who met the study criteria and their samples were analysed. Of which 41(56.16%) were males and 32(43.84%) were females, giving a M: F ratio of 1.3:1. The mean age was 3.68 ± 3.68 years (range, 3 month 16 years), 19(26.03%) were infants and 60(82.19%) were aged between 1 and 5 years. Thirty seven 37(50.68%) patients were from high socioeconomic class, 14(19.18%) from middle class and 22(30.14) from low class.
Most of the patients 54 (73.97%) had tonsillitis, 11(15.07%) had tonsillopharyngitis while eight (10.96%) had pharygitis. Fever and inflamed tonsils/pharynx were the most frequent signs or symptoms occurring in 65 (89.04%) and 64 (87.67%) of subjects respectively while cervical adenitis 20 (27.39%) and presence of exudates 19 (26.03%) were the least frequent signs (Table I).
Although all the clinical features showed relatively low sensitivity and specificity, inflamed tonsils (64.81) and cough (64.28%) were the most sensitive while fever (55.38%) and cervical adenitis (55.00%) were the least sensitive. Cough (54.84%) and presence of exudates (50.00%) showed the most specificity while fever (37.50%) and inflamed tonsils (15.80%) had the least specificity. The highest positive predictors of bacterial URTI were fever (87.80%) and inflamed tonsils (85.37%) and the least predictors were history of sorethroat (30.00%) and difficulty with swallowing (30.00%) (Table I).
The most frequently associated disease was malaria 17 (23.29%) and otitis media 4 (5.48%). Bronchial asthma and paediatric AIDS were present in 3 (4.11%) patients each. Others were measles, mumps, rheumatic heart disease and sickle cell anaemia which had occurred in 1 (1.37%) patient each.
Bacteria were isolated from 39 (53.42%) subjects. Nineteen (48.72%) of the isolates were BHS, while five (12.83%) were Staphyloccocus aureus. Klebsiella spp, Proteus spp and Pseudomonas spp were recovered from the throat of seven (17.95%), three (7.69%) and three (7.69%) subjects respectively (Table II). Of the 39 subjects with isolates, 17(43.59%) were males and 22 females giving M: F ratio of 1:1.3. The gender difference was not significant when compared to the study population x 2=1.1146, p = 0.2845. The mean age of those with isolates was 3.38 ± 4.02 years and this was not significantly different from the mean age of the study population, t = 0.3979, p = 0.6914 (CI 1.194 to 1.794). There were more patients 18(46.15%) from high SES than middle seven (17.95%) and low 14 (36.90%). The patients who had isolates in each socioeconomic class did not significantly differ from those without isolates x 2 = 0.9844, p = 0.6133. The 14 patients with Klebsiella mirabilis, Pseudomonas aeroginosa, Proteus mirabilis and Eshericheae coli had a mean age of 1.32 ±0.94 years (range, 3months 4 years). They were significantly younger than the study population, t = 2.374; p = 0.019 and they were from the middle or low SES (Table II). BHS and S. aureus showed 100% sensitivity to cefuroxine, azithromycin, ceftazidime and Genticin but showed little or no sensitivity to ampicillin and cotrimoxazole, Table III.
Most of the patients in this study were under five years of age. This is in keeping with previous reports.7,11 The fact that these young children are still in the process of developing immunity to pathogens in their environment may explain this finding. That majority of the subjects were from high SES may reflect the likelihood that people from upper social class are more likely to seek medical attention for an upper respiratory tract infection.
Although sorethroat and difficulty in swallowing are major symptoms in tonsillitis/ pharyngitis,12,13 less than a third of the patients in this study presented with these complaints. This is because children seldom complain of sorethroat or difficulty in swallowing.14 Since almost a third of the studied patients were infants, complaint of sorethroat or difficulty with swallowing would obviously be less.
The use of individual clinical features as predictors of possible bacteria URTIs were poor as shown by their low sensitivity and specificity. This is consistent with other studies that showed the unreliability of using individual sign to predict possible bacteria tonsillitis/ pharyngitis.13,14
Almost half of the bacteria isolates were BHS, the predominance of this organism in this study is consistent with reports of other studies12,15. Although the Lancefield grouping could not be done to determine the proportion of the BHS that was group A (The group that is associated with RF and RHD), the high numbers of BHS underscores the continued potential danger posed by possible non-suppurative sequelae of RF and RHD in our community. The unusual presence of gram negative coliform organisms (Klebsiella, Proteus and Pseudomonas) in this study is noted. Since they are not traditional upper respiratory pathogens, it is possible that they could have been transferred from the anal region to the throat following poor personal hygiene. The fact that the patients with coliform organisms were significantly younger and were from low and middle SES lends credence to such possibility.
Our study showed that BHS and S. aureus had 100% sensitivity to cefuroxine, azithromycin and ceftazidime, three relatively expensive drugs, a finding corroborated by previous studies.16-18 It is worthy of note that amoxicillin showed more activity against the pathogens isolated compared to ampicillin in this study, despite the fact that both drugs are aminopenincillins and are suppose to have similar antibacterial activities.19 Although the reason for this difference is not clear, we suspect it may not be unrelated to possible abuse of the cheaper ampicillin with resultant resistance. All pathogens showed little or no sensitivity to ampicillin and co-trimoxazole, two affordable drugs, whose normal antibacterial spectrum includes streptococcus pneumonia.19 This finding may also be due to abuse of these relatively cheap drugs.
A ten day course of penincillin V in 3 or 4 divided doses is the standard recommendation for treatment of streptococcal pharyngitis. 20 The need to change this first line drug is being advocated by some workers due to increasing resistance of streptococcus to penicillin, because of difficulty with complying with the standard regimen and the non availability of paediatric preparations.21-23 The magnitude of streptococcal resistance to penincillin V could not be evaluated in this study since sensitivity to it could not be tested due to non-availability of the disc. We recommend cefuroxine and azithromycin as alternative drugs where the oral route is desired and ceftazidime as a choice drug for intravenous or intramuscular routes for streptococcal and staphylococcal infections. Where financial constraint is a major issue, amoxicillin an affordable drug with comparable sensitivity with the standard penicillin regimen,24 can be given both orally or intravenously. The coliform organisms showed good sensitivity to genticin, an affordable drug. Genticin would be particularly useful where the parenteral route of drug administration is indicated. However its suitability as a first line drug in infections by these pathogens is limited by the fact that it can only be administered parenterally.
LIMITATION OF STUDY
Since penicillin V is still the standard recommendation for treating streptococcal URTI, our inability to test bacterial isolates for sensitivity to it is noted as a limitation to this study. We therefore suggest future study to determine the sensitivity of upper respiratory pathogens to penicillin V in the study locale.
We thank most sincerely the paediatric residents of the department of Child Health, University of Benin Teaching Hospital, who assisted us in obtaining relevant information and throat swabs from the subjects.
© Copyright 2008 - CMS UNIBEN JMBR
The following images related to this document are available:
Photo images[jm08006t2.jpg] [jm08006t3.jpg] [jm08006t1.jpg]