DISCUSSION:
In the current study, patients diagnosed with CCAP at presentation were older than those diagnosed with CCAP. Sex and comorbidities were not significant risk factors for CCAP. Meanwhile, age below 2 years, male sex, and comorbidities were associated with invasive pneumococcal infections16,27,28. However, age, sex, and comorbidities were not significant risk factors for pneumococcal pneumonia. In a study conducted in Israel, there was no significant difference in age between CAP and CCAP patients. Moreover, age <2 years was not a risk factor for CCAP development29. Immunodeficiencies and suppurative lung diseases may cause CAP. However, CCAPs are generally observed in children who do not have comorbidities and who were previously healthy29. In a 540 patient series in the USA in 2002, children diagnosed with EMP were older (mean age = 6 years, p < 0.001) than those who were followed-up for CAP30. In a study that assessed the 10-year medical records of patients in Canada and France, CCAP was found to account for 51%–57% of all hospital admissions among children aged below 5 years28,31 Similar findings were also observed in a study including 767 patients (mean age = 4.1 years, p = 0.02) in France in 2010,31.
In this study, the presence of cough, chest pain, and fever and pre-admission time of these symptoms and a history of pneumonia were not considered risk factors for CCAP. Respiratory distress (tachypnea and retraction) and hypoxia were significantly associated with CCAP. Severe CAP is defined as the presence of fever; capillary filling time >2 s; dehydration; inability to feed; subcostal, nasal, and substernal retraction; tachypnea (infants, >70 cycles/min and older children, >50 cycles/min); cyanosis; and moaning, according to the British Thoracic Society 2011 guideline4. Based on these parameters, CCAP is more common in hospitalized patients diagnosed with severe pneumonia4,5.
In our study, CCAP patients were commonly underweight. Even in the absence of severe malnutrition, underweight is associated with an increased pneumonia risk32. Wexler ID et al. conducted a study in Israel. Results showed that CCAP patients were often underweight, and this finding is similar to ours. However, the condition is not an independent risk factor for CCAP development9.
In our study, CCAP patients had a high CRP level at time of enrollment. Moreover, both groups had a high ESR and leukocyte and thrombocyte counts during admission. In the study of Huang C-Y et al., elevated CRP level and leukocyte count upon admission were a significant factor for CCAP development10. In a multicenter study that retrospectively analyzed 111 pneumonia patients in 2006, significant differences in terms of ESR and leukocyte and platelet counts were not observed between CAP and CCAP patients9. In these patient groups, anemia and leukocyte <15.000 mm3 were associated with prolonged fever and longer hospitalization duration9.
In our study, no significant growth was observed in the blood cultures of patients hospitalized due to CAP and CCAP. This was attributed to peroral antibiotic use before hospitalization. Regardless of the presence of pulmonary complications, the bacteremia incidence is low in patients followed-up for pneumonia, and a possible pathogen is rarely detected in the blood culture11,33. Neuman MI et al. conducted a study in the USA in 2017. Results showed that a pathogenic microorganism was detected in the blood culture in only 65 (2.5%) of 2568 patients hospitalized due to pneumonia33. In this study, empirical antibiotic use before hospitalization less likely caused growth in the blood culture33. In a retrospective study conducted by Erlichman I et al., which included 144 children hospitalized due to CCAP, empirical antibiotics use before hospitalization decreased the growth rate in the blood culture from 63% to 22%. However, the pathogen proportion detected in pleural cultures was higher34. In a 20-year retrospective study in the USA, the blood culture growth rate decreased from 67% to 30% with empirical antibiotic use before hospitalization in 369 patients who had EMP35. In a study examining patients who were followed-up for PPE and EMP between 2010 and 2017, the pathogen proportion was significantly lower in the blood culture than in pleural cultures33.
Viral and bacterial PCR amplification of samples collected from the respiratory tract and pleura could not be obtained from each patient due to technical problems. RSV was the most common viral agent detected. Within the last 10 years, respiratory and pleural fluid DNA PCR amplification methods, which can yield faster results than conventional culture and are not affected by empirical antibiotic use, are commonly used in pathogen identification9,20,36. The DNA PCR of samples collected from pleural fluid can provide more sensitive results than that of blood culture samples, and the serotyping results can be obtained significantly faster36. However, there are limitations in the differentiation of pathogenic microorganisms due to colonization when nasopharyngeal, oropharyngeal, sputum samples are used37.
In our study, according to CT scan data, the antibiotic therapy type and duration was modified and more detailed information could be obtained before the surgical intervention, particularly in our patient with NP and BPF. If fever does not subside within 72 h and the clinical condition does not improve or worsens in patients followed-up for CAP, treatment for CCAP should be modified, if necessary, according to the evaluation results21,38. When CXR and thoracic USG are not effective for the diagnosis of complications, such as BPF and pneumatocele, particularly in the differentiation of NP and LA, more detailed information can be obtained via CT scan22,38
In our study, pleural effusion >10 mm during admission is a significant variable for CCAP diagnosis. Similar studies have shown that the presence of pleural effusion at presentation is associated with CCAP diagnosis9-11. In patients with unresolved radiological and clinical findings, YCA, tuberculosis-related lymphadenopathy, and congenital lung malformations may be observed.
In accordance with studies in the literature, our research showed that CCAP patients commonly present with hypoxia upon admission, and oxygen therapy was required during hospitalization. CCAP patients have a longer hospitalization duration than CAP patients due to local and systemic complications and the need for additional interventions (GT, CDF, and VATS) during follow-up39. BPF was observed, with an incidence rate of 13% in children with NP in the USA and 67% in Brazil12,14. Patients followed-up for CCAP, particularly NP and BPF with complications, had a longer hospitalization, oxygen therapy, and fever duration and higher number of PICU and even ECMO follow-ups16,40.
In our study, intravenous SAM and/or peroral macrolide were administered to patients diagnosed with CAP, and treatment was modified when CCAP was detected. The antibiotic duration and type used before admission were not significant risk factors for CCAP development. In previous retrospective studies and a prospective research conducted by Huang CY et al., disease progression was observed regardless of the empirical antibiotic therapy type provided according to the current treatment guidelines10,15.
VATS is associated with better outcomes and shorter hospitalization duration compared with pleural drainage with GT without fibrinolytics20. In 2021, the American Academy of Pediatric Surgery recommended that VATS should be considered when chemical debridement with CDF cannot be achieved20. In a systematic review and meta-analysis published in 2018, similar perioperative complications were observed in cases in which both methods were used. However, patients who received VATS were exposed to less repetitive interventions and had a shorter hospitalization duration41. Based on these data, in recent years, studies about early VATS use, instead of CDF, in children have been published in accordance with those in adults42. When sufficient debridement with GT is not achieved and VATS is performed in the early period before fibrinolytic use, the hospitalization duration is shortened. Further, the complication incidence is lower in patients treated with VATS than in those treated with CDF42. Based on these developments, VATS is preferred instead of CDF when adequate pleural fluid drainage cannot be achieved with thoracentesis and GT implantation for PPE and EMP treatment in our hospital after 2017, which was the time our study was conducted.
The surgical intervention timing can affect treatment success43,44. CDF can be used for early-stage disease management. However, it is not effective for advanced-stage disease, and complication incidence increases with repeated interventions including CDF. Moreover, hospitalization duration is prolonged43. VATS efficacy decreases, particularly after 1 month, and decortication may be required in patients who underwent CDF43-46. In relation to this, the European Association for Cardio-Thoracic Surgery recommends that VATS should be highly considered in all advanced-stage disease patients at presentation46
The current study had several limitations. That is, the research was retrospective in nature. Hence, antipyretic treatment use, content, and frequency, as well as its effect on CCAP development, could not be investigated because the patients’ family members did not regularly follow-up for fever. Moreover, our hospital is a tertiary referral center, and most patients had received prior treatment. However, late referrals could affect outcomes. Although necessary, not all patients could undergo respiratory and pleural DNA PCR amplification due to technical problems.
In conclusion, pleural effusion, respiratory distress (tachypnea and retraction), and hypoxia at presentation were found to be independent risk factors for CAP. Therefore, experience in interventional radiology and pediatric surgery and use of multidisciplinary approaches may be important for the diagnosis and treatment of children with CCAP admitted in tertiary care centers. However, multicenter, prospective and comparative studies should be conducted to examine CDF and VATS efficacy in children with early- and late-stage diseases.