In this study, we aimed to identify differences in evolution of children with non-severe acute lower respiratory tract infection between those with and without radiographically diagnosed pneumonia receiving amoxicillin. The results on clinical failure are presented elsewhere [6].

Evolution follow-up: Patients were evaluated in the hospital twice a day, for up to 5 days and a senior pediatrician personally checked each evaluation. The patients stayed in the observation ward for children without signs of severity with their respective caregiver so that they could be carefully followed-up by the research team, who also assisted the drug administration. Amoxicillin for up to the tenth day of treatment and the telephone number of the surveillance coordinator were supplied to the guardian adult at discharge and a phone call was advised if the patient developed any new complaint. Each child was discharged from hospital when there was no more fever and respiratory discomfort. At the fifth day of treatment, a telephone call was made to the caregiver to enquire for symptoms and interventions. Cure was recorded when the caregiver informed of absence of all the presenting symptoms. A follow-up examination was performed 2 to 4 weeks after recruitment. Every patient received a telephone call on the day before the scheduled outpatient visit to remind them. All clinical data were collected without knowledge about the final CXR assessment.

Radiological studies: For the purpose of this study, CXR were read by two pediatric radiologists who were unaware of the patients’ clinical history and evolution. Each pediatric radiologist gave his assessment independently. In case of disagreement regarding diagnosis of pneumonia, CXR was read by a third pediatric radiologist without knowledge on the previous evaluations. Radiographic findings were recorded in a standardized form according to previously published recommendation [9]. The pediatric radiologist looked for the presence of pulmonary infiltrate, consolidation, pleural effusion, atelectasis, hyperinflation, abscess, peribronchial thickening, pneumatocele and pneumo-thorax. The pulmonary infiltrate was characterized as alveolar, interstitial or alveolar-interstitial. Pneumonia was radiographically diagnosed if there was agreement on presence of pulmonary infiltrate or pleural effusion in 2 independent assessments.

Data management and analysis: Epi-Info (version 6.04) was used for data entry, Statistical Package for the Social Sciences (SPSS 9.0) and STATA (version 9.0) were used for statistical analysis. The presence of a symptom or sign at least once during the two daily evaluations was considered as presence of that finding on that day of evolution. Differences in proportions were assessed by using the Pearson chi square or Fisher’s exact tests and continuous variable by the Student t or Mann-Whitney U tests, as appropriate. Bivariate and multivariate analyses, using logistic regression models, were performed to identify association between radiographically diagnosed pneumonia on admission and findings during evolution. The 95% confidence intervals were calculated. Sensitivity, specificity and predictive values were calculated. The positive and negative likelihood ratio (LR) were estimated. The study protocol was approved by the Ethics Committee of the Federal University of Bahia.

Overall, 630 patients were evaluated, out of whom 239 were excluded [antibiotic use in the previous 48 hours (n=95), signs of severe disease (n=92), refusal to give informed consent (n=18), underlying chronic illness (n=17), hospitalization in the previous 7 days (n=10) or amoxicillin allergy (n=7)]. Moreover, 6 families refused to continue in the study, 3 children had poor quality on CXR at the evaluation of the first 2 pediatric radiologists and there was no agreement between 2 of 3 participant radiologists in the reading of CXR of 10 patients. Therefore, the study group included 372 patients.

The median (Inter quartile range, IQR) age was 26 (13-40) months. Seventy five (20.2%) children were younger than 1 year. There were 201 (54.0%) males and 14 (3.8%) malnourished patients. The median (IQR) duration of disease was 6 (4-10) days and the most frequent complaints were cough (98.7%), fever (94.1%), difficulty breathing (68.8%), vomiting (47.6%) and wheezing (29.8%). Rales (70.7%), crackles (62.4%), tachypnea (50.8%) and fever (36.3%) were the most common findings. None presented with cyanosis, abdominal distension or stridor. Only 5 (1.3%) patients presented with chest retraction, but nobody showed lower chest indrawing.

Agreement was found between the first 2 pediatric radiologists in 284 (74.3%) cases. By adding the evaluation from the third pediatric radiologist, agreement was found in 372 (97.4%) cases. The final radiographic diagnosis was pneumonia (n = 192; 51.6%), normal CXR (n = 152; 40.9%) and other diagnosis (n = 28; 7.5%) which comprised peribronchial thickening (2.9%), atelectasis (2.4%), hyperinflation (1.6%), peribronchial thickening plus atelectasis (0.3%) or plus hyperinflation (0.3%). Abscess, pneumatocele and pneumothorax were not described. Table I shows the comparison of baseline characteristics according to the radiographic diagnosis subgroup.

Table I  Baseline Characteristics According to Radiographic Diagnosis 

Among 344 patients with pneumonia or normal CXR, 337 (98.0%) were hospitalized. The median (25th-75th percentile) duration of hospitalization was 1 day (1-2, range 1-21 days). Amoxicillin was substituted with another antibiotic in 12 (3.2%) inpatients at the first (n=1), second (n=2), third (n=5), fourth (n=2) or fifth (n=2) day of treatment because of clinical deterioration (n=5), concurrent infection (sinusitis or otitis) diagnosed during follow-up (n=3), persistent vomiting (n=2), persistence of symptoms (n=1) or immediate allergic reaction (n=1). All patients were discharged after improvement. Table II presents the significant differences on evolution in a univariate analysis and Table III depicts the significant differences on admission or evolution in a multivariate analysis between children with radiographic diagnosis of pneumonia and normal CXR on admission. Table IV shows the validation of tachypnea and documented fever on admission and up to the second day of evolution. The positive LR of tachypnea or fever on the second day of evolution was 1.88 (95% CI: 1.27-2.78) and 3.29 (95% CI: 1.49-7.28), respectively. On the contrary, the negative LR of both characteristics was 0.80 (95% CI: 0.70-0.88) and 0.88 (95% CI: 0.81-0.94), respectively.

Table II	Differences in Evolution Between Children with Pneumonia and Normal Chest Radiograph on Admission 

Table III	Multivariate Analysis of Clinical Findings on Admission or Evolution Between Children with 

Pneumonia or Normal Chest Radiograph on Admission  (N=336)*	

Table IV	Validation of Tachypnea and Documented Fever on Admission and Up to Second Day 

of Evolution for the Diagnosis Of Pneumonia

Successful telephonic contact, between the fourth and seventh day of treatment (median=5 days, IQR = 5-5 days), was obtained with 336 (98.2%) families who informed that the child had complete resolution of symptoms (75.2%), improvement with persistence of at least one initial complaint (24.2%), absence of improvement (n = 1; 0.3%) or worsening (n = 1; 0.3%). At this moment, report of complete resolution of symptoms was more frequent among those with normal CXR (80.7% vs. 70.7%; P = 0.04). Ambulatory follow-up evaluation occurred between 11 and 38 days after recruitment (median 21; IQR= 17-23) among 337 (98.0%) patients; mean duration of amoxicillin administration was 10 ± 1 day among 323 users. No clinical difference at this time was found by comparing patients with pneumonia or normal CXR (data not shown).

The frequency of reported cure was significantly higher on the fifth day telephone call survey for those with normal CXR on admission. This suggests that the resolution of the clinical findings of patients with radiographic pneumonia is slower than that among children with other lower respiratory tract diseases. Although this result might have been expected, it raises worry about the length of antibiotic administration to treat pneumonia. The WHO has recommended antibiotic use for 3-5 days to children with non-severe pneumonia [21], and this recommendation comes from studies in which children were included by using WHO diagnostic criteria [22,23]. A recent meta-analysis recognizes that more well-designed randomized clinical trials are needed to support such recommendation [24]; we now add addition evidence. We did not investigate the etiology and this is a limitation. However, this study was carried out in a scenario that is similar to many ERs, all over the world. In this study, the duration of amoxicillin administration was 10 days to be in accordance with the national Brazilian guidelines, for ethical reason [25].

In this investigation, the final radiographic diagnosis was based on the agreement of 2 independent assessments by trained pediatric radiologists using a standard evaluation form previously validated [9]. The use of such strict criteria was meant to control the subjective bias that may occur in CXR reading [4]. In addition to that, the close evaluation on admission and evolution by a senior pediatrician, the low rate of loss to follow-up and the implementation in only one pediatric center contribute to the reliability of our results.

To conclude, this study, to the best of our knowledge, is the first prospective study enrolling and following children with non-severe CAP treated with amoxicillin, where detailed clinical and radiographic aspects on admission and evolution were registered. From our data, persistence of fever or tachypnea up to the second day of amoxicillin treatment is predictive of radiographically diagnosed pneumonia among children with non-severe lower respiratory tract diseases.

Acknowledgments: The authors are thankful to the pediatricians and nurses of the Professor Hosannah de Oliveira Pediatric Center, Federal University of Bahia, Salvador, Brazil. Maria-Regina A Cardoso and Cristiana M Nascimento-Carvalho are investigators of the Brazilian Council for Science and Technology Development (CNPq).

Contributors: M-SHF, ARM, CCS, MCS, MNB, FO, BBB, and the PNEUMOPAC-Efficacy Study Group collected the clinical data and drafted the manuscript; CAAN, SCA and RVB read the chest X-rays, filled the radiographic forms out, took part in the interpretation of the results and criticized the manuscript; M-RA Cardoso took part in the analysis and revised the manuscript; CMNC conceived the research question and designed the study protocol, analyzed the data and revised critically the manuscript. All authors approved the submitted version.

Funding: This research was funded by the the Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB), Salvador, Brazil. Maria-Regina A Cardoso and Cristiana M Nascimento-Carvalho are investigators of the Brazilian Council for Science and Technology Development (CNPq). Neither FAPESB nor CNPq had any role in handling or completing the research. The authors had full control of all primary data.

Competing interests: None stated.

Appendix I

1. Ranganathan SC, Sonnappa S. Pneumonia and other respiratory infections. Pediatr Clin North Am. 2009;56:135-56.

2. World Health Organization. Acute Respiratory Infections in Children: Case Management in Small Hospitals in Developing Countries. A Manual for Doctors and Other Senior Health Workers. World Health Organization: Programme for the Control of Acute Respiratory Infections. Geneva, Switzerland: 1990.

3. Puumalainen T, Quiambao B, Abucejo-Ladesma E, Lupisan S, Heiskanen-Kosma T, Ruutu P, et al. Clinical case review: a method to improve identification of true clinical and radiographic pneumonia in children meeting the World Health Organization defition for pneumonia. BMC Infect Dis. 2008;8:95.

4. Wilkins TR, Wilkins RL. Clinical and radiographic evidence of pneumonia. Radiol Technol. 2005;77:106-10.

5. Mahabee-Gittens EM, Grupp-Phelan J, Brody AS, Donnelly LF, Bracey SE, Duma EM, et al. Identifying children with pneumonia in the emergency department. Clin Pediatr. 2005;44:427-35.

6. Fontoura M-S H, Araújo-Neto CA, Andrade SC, Brim RV, Matutino AR, Silva CC, et al. Clinical failure among children with non-severe community-acquired pneumonia treated with amoxicillin. Expert Opin Pharmacother. 2010;11:1451-8.

7. El-Radhi AS, Barry W. Thermometry in paediatric practice. Arch Dis Child. 2006;91:351-6.

8. World Health Organization. Integrated management of childhood illness chart booklet. (WC 503.2). Geneva: WHO, 2008. Available at: http:// whqlibdoc.who.int/publications/2008/9789241597289_eng.pdf. Accessed on January 15; 2009.

9. Cherian T, Mulholland EK, Carlin JB, Ostensen H, Amin R, Campo M, et al. Standardized interpretation of paediatric chest radiographs for the diagnosis of pneumonia in epidemiological studies. Bull World Health Organ. 2005;83:353-9.

10. Sazawal S, Black RE, Pneumonia Case Management Trials Group. Effect of pneumonia case management on mortality in neonates, infants, and preschool children: a meta-analysis of community-based trials. Lancet Infect Dis. 2003;3:547-56.

11. Shah S, Bachur R, Kim D, Neuman MI. Lack of predictive value of tachypnea in the diagnosis of pneumonia in children. Pediatr Infect Dis J. 2010;29:406-9.

12. Hazir T, Qazi SA, Bin Nisar Y, Maqbool S, Asghar R, Iqbal I, et al. Comparison of standard versus double dose of amoxicillin in the treatment of non-severe pneumonia in children aged 2-59 months: a multi-centre, double blind, randomised controlled trial in Pakistan. Arch Dis Child. 2007;92:291-7.

13. Harari M, Shann F, Spooner V, Meisner S, Carney M, de Campo J. Clinical signs of pneumonia in children. Lancet. 1991;338:928-30.

14. Mulholland EK, Simoes EA, Costales MO, McGrath EJ, Manalac EM, Gove S. Standardized diagnosis of pneumonia in developing countries. Pediatr Infect Dis J. 1992;11:77-81.

15. Singhi S, Dhawan A, Kataria S, Walia BN. Validity of clinical signs for the identification of pneumonia in children. Ann Trop Paediatr. 1994;14:53-8.

16. Redd SC, Vreuls R, Metsing M, Mohobane PH, Patrick E, Moteetee M. Clinical signs of pneumonia in children attending a hospital outpatient department in Lesotho. Bull World Health Organ. 1994;72:113-8.

17. Falade AG, Tschäppeler H, Greenwood BM, Mulholland EK. Use of simple clinical signs to predict pneumonia in young Gambian children: the influence of malnutrition. Bull World Health Organ. 1995;73:299-304.

18. Graham SM, English M, Hazir T, Enarson P, Duke T. Challenges to improving case management of childhood pneumonia at health facilities in resource-limited settings. Bull World Health Organ. 2008;86:349-55.

19. Zorc JJ, Hall CB. Bronchiolitis: recent evidence on diagnosis and management. Pediatrics. 2010;125:342-9.

20. Cardoso MR, Nascimento-Carvalho CM, Ferrero F, Alves FM, Cousens SN. Adding fever to World Health Organization criteria for diagnosing pneumonia enhances the ability to identify pneumonia cases among wheezing children. Arch Dis Child. 2011;96:58-61.

21. Grant GB, Campbell H, Dowell SF, Graham SM, Klugman KP, Mulholland EK, et al. Recommendations for treatment of childhood non-severe pneumonia. Lancet Infect Dis. 2009;9:185-96.

22. Pakistan Multicentre Amoxycillin Short Course Therapy (MASCOT) pneumonia study group. Clinical efficacy of 3 days versus 5 days of oral amoxicillin for treatment of childhood pneumonia: a multicentre double-blind trial. Lancet. 2002;360:835-41.

23. Agarwal G, Awasthi S, Kabra SK, Kaul A, Singhi S, Walter SD, et al. Three day versus five day treatment with amoxicillin for non-severe pneumonia in young children: a multicentre randomised controlled trial. BMJ. 2004;328:791-7.

24. Haider BA, Saeed MA, Bhutta ZA. Short-course versus long-course antibiotic therapy for non-severe community-acquired pneumonia in children aged 2 months to 59 months. Cochrane Database Syst Rev. 2008;(2):CD005976.