A Retrospective Chart Review of Pediatric Complicated Community-Acquired Pneumonia: An Experience in the Al Qassimi Women and Children Hospital

Background Community-acquired pneumonia (CAP) is one of the most common global health issues. Even though many vaccinations and new diagnostic tools are available, CAP has a higher mortality rate, especially in children less than five years of age. Complicated CAP (CCAP) in a healthy child is a severe disease characterized by a combination of local complications, such as parapneumonic effusion (PPE), empyema (EMP), necrotizing pneumonia (NP), abscess, pneumothorax, and bronchopleural fistula, and systemic complications, such as bacteremia, metastatic infection, multiorgan failure, acute respiratory distress syndrome, disseminated intravascular coagulation, and, rarely, death. This study describes the demographic features, clinical presentation, management, and outcomes of patients diagnosed with CCAP at the Al Qassimi Women’s and Children’s Hospital (AQWCH). Methodology This retrospective chart review aims to collect and explore the data of all previously healthy children admitted with CCAP between the ages of one month and 13 years at AQWCH from January 2018 to December 2020. The primary study outcome measure is to provide clinicians with the diagnostics, evaluation, and management required to treat complicated pneumonia. Results A total of 195 patients were diagnosed with CAP, of whom 30 (15.3%) were diagnosed with CCAP. Of these, 14 (46.6%) patients had NP, eight (26.7%) had PPE, and eight (26.7%) had EMP. The median age of patients was 2.5 years, with 13 (43%) males and 17 (57%) females. The median duration of their stay in the hospital was 16 days. All patients were vaccinated with Hib, PCV13, or PCV7, and 57% of the patients received antibiotics before admission. The most common findings were consolidation and pleural effusion. Blood culture was negative in all cases, and pleural culture was positive only in three cases. A total of 17 (57%) patients underwent video-assisted thoracoscopic surgery (VATS), and post-VATS surgical emphysema was found to be the most common complication. Chest X-rays normalized after three months in 65% of patients. On comparing patients who were admitted to the Pediatric Intensive Care Unit (PICU) before any surgical intervention with those who were not, it was found that patients who required PICU admission were young (median = 2 years; interquartile range (IQR) = 1-4.5; p = 0.044) and had higher respiratory rate (mean = 49 breaths/per minute, standard deviation (SD) = 11; p = 0.000). In addition, they had lower median albumin (median = 2 g/L; IQR = 1.8-2.23; p = 0.004). On comparing patients who required VATS and those who did not require VATS, it was found that the former had a higher median respiratory rate (48 per min; range = 42-54; p = 0.01). A cavity in the chest computed tomography (CT) was found in 86% of patients with VATS (p = 0.017), and they had lower median albumin (median = 2 g/L; IQR = 1.92-2.24; p = 0.012), as well as longer median duration of using oral antibiotics (median = 21 days; IQR = 19-26; p = 0.025). Patients with complicated NP had a higher respiratory rate and higher PICU admission, and more cavity in the chest was found in the CT study. Most NP patients also underwent VATS and had longer median days of using oral antibiotics. One patient developed a bronchopleural fistula, and one patient diagnosed with NP died. Conclusions CCAP is a major cause of hospitalization in children. It is important to suspect CCAP in all CAP patients not responding to treatment after 48-72 hours.


Results
A total of 195 patients were diagnosed with CAP, of whom 30 (15.3%) were diagnosed with CCAP. Of these, 14 (46.6%) patients had NP, eight (26.7%) had PPE, and eight (26.7%) had EMP. The median age of patients was 2.5 years, with 13 (43%) males and 17 (57%) females. The median duration of their stay in the hospital was 16 days. All patients were vaccinated with Hib, PCV13, or PCV7, and 57% of the patients received antibiotics before admission. The most common findings were consolidation and pleural effusion. Blood culture was negative in all cases, and pleural culture was positive only in three cases. A total of 17 (57%) patients underwent video-assisted thoracoscopic surgery (VATS), and post-VATS surgical emphysema was found to be the most common complication. Chest X-rays normalized after three months in 65% of patients.
On comparing patients who were admitted to the Pediatric Intensive Care Unit (PICU) before any surgical intervention with those who were not, it was found that patients who required PICU admission were young (median = 2 years; interquartile range (IQR) = 1-4.5; p = 0.044) and had higher respiratory rate (mean = 49 breaths/per minute, standard deviation (SD) = 11; p = 0.000). In addition, they had lower median albumin (median = 2 g/L; IQR = 1.8-2.23; p = 0.004).
On comparing patients who required VATS and those who did not require VATS, it was found that the former had a higher median respiratory rate (48 per min; range = 42-54; p = 0.01). A cavity in the chest computed tomography (CT) was found in 86% of patients with VATS (p = 0.017), and they had lower median albumin (median = 2 g/L; IQR = 1.92-2.24; p = 0.012), as well as longer median duration of using oral antibiotics (median = 21 days; IQR = 19-26; p = 0.025).
Patients with complicated NP had a higher respiratory rate and higher PICU admission, and more cavity in the chest was found in the CT study. Most NP patients also underwent VATS and had longer median days of using oral antibiotics. One patient developed a bronchopleural fistula, and one patient diagnosed with NP died.

Introduction
Pneumonia in previously healthy children caused by infection outside the hospital is defined as communityacquired pneumonia (CAP) [1]. CAP remains the largest single cause of morbidity and mortality worldwide in children aged between 28 days (i.e., outside the neonatal period) and five years [2]. Complicated CAP (CCAP) in a previously healthy child is a severe disease characterized by a combination of local complications, such as parapneumonic effusion (PPE), empyema (EMP), necrotizing pneumonia (NP), abscess, pneumothorax, and bronchopleural fistula, and systemic complications, such as bacteremia, metastatic infection, multiorgan failure, acute respiratory distress syndrome, disseminated intravascular coagulation, and, rarely, death [3]. Pleural effusion undergoes three stages, namely, the exudative stage (fluid is serous and sterile), the fibropurulent stage after one to two weeks (inflow of white blood cells and bacteria), EMP over two to four weeks, and progressive thick fibrinous peel (pus formation) [4]. Even though the exact pathology of NP is unknown, it is characterized by severe lung tissue destruction and disintegration [3]. Many studies have shown that genetic predisposition, vascular thrombosis, and vascular occlusion are possible mechanisms of NP [5][6][7]. The process begins with consolidation and necrosis and then progresses to cavitation, which may convert into a large cyst. If ruptured, it can cause bronchopulmonary fistula [7,8].
The symptoms of CCAP are fever, cough, tachypnea, distress, chest pain, and abdomen and/or shoulder pain. Decreased air entry, dullness, and fine crackles are some of the findings on chest examination. Streptococcus pneumonia, Staphylococcus aureus, including methicillin-resistant S. aureus, and Streptococcus pyogenes are the causative organisms [9]. Less common causes include Haemophilus influenzae, Mycoplasma pneumoniae, and Pseudomonas aeruginosa. Children with CCAP have an extended hospitalization duration, radiological evaluation, antibiotic treatment, and analgesic or sedative medications and require more invasive interventions such as chest tube insertion, open decortication, and video-assisted thoracoscopic surgery (VATS), with most patients recovering completely. Systemic complications of CCAP include sepsis and septic shock, metastatic infection, multiorgan failure, acute respiratory distress syndrome, disseminated intravascular coagulation, and death [3]. categorical variables. Data for categorical variables were tested using the chi-square test and Fisher's exact test. Continuous variables were tested using the t-test and Mann-Whitney U test. The normality was tested using the Shapiro-Wilk test and visualization of histograms. As continuous variables, such as age, were not normal, the Kruskal-Wallis test was performed. The results are shown as tables or diagrams (bar and line graph). The alpha value p ≤ 0.05 was used to determine statistical significance.

Frequency
Overall study
The blood test results on admission showed that the median white blood cell count was 17.7 per/mL (interquartile range (IQR) = 14-24), levels of C-reactive protein increased in all patients with a median of 233 mg/L (IQR = 90-300), and the median albumin content was 2.2 g/L (IQR = 1.96-2.74). The blood culture was negative in all cases, and the pleural culture was positive only in three (14%) cases. The isolated organisms were Staphylococcus haemolyticus, S. pneumoniae, and S. pyogenes ( Table 2).
Follow-up chest X-rays after three months of treatment were obtained for 17 (57%) patients, and the findings were normal in 11 (65%) patients; however, consolidation was found in five (29%) patients, and one (6%) patient had a cavity. Chest X-rays after six months were obtained for five (17%) patients, and the finding was normal in all patients (100%) ( Table 3). Chest taping − number (%) 4 (13) Follow-up chest X-ray after 3 months − number (%) 17 (57) Normal findings on the follow up chest number -ray after 3 months − number (%) 11 (65) Consolidation on the follow up chest X-ray after 3 months − number (%) 5 (29) Cavity on the follow up chest X-ray after 3 months − number (%) 1 (6) Follow-up chest X-ray after 6 months − number (%) 5 (17) Normal findings on the follow up chest X-ray after 6 months − number (%) 5 (100)  On testing for association between patients who required PICU admissions and who did not require it, along with other variables, a statistical significance was found (p < 0.05) with age, respiratory rate, and albumin content. Patients who were admitted to PICU had a lower median age of two years (IQR 1-4.5, p = 0.044), a higher mean respiratory rate of 49 beats/minute (SD = 11; p = 0.000), and a lower median albumin content of 2 mg/L (IQR = 1.8-2.23; p = 0.004). There were no significant differences in other frequencies (  On testing variables for the association between patients who required VATS and those who did not require VATS, a significant (p < 0.05) association between patients who required VATS and their clinical, laboratory, radiological findings, and management was found. Clinically, there was a significantly higher respiratory rate of 48 breaths/minute (IQR = 42-55; p = 0.01] in patients with VATS. In total, 12 (86%) patients (p = 0.017) who underwent VATS had a cavity in their chest CT scan on admission with no differences in other frequencies. All laboratory tests were equally abnormal in all patients in both groups except for albumin content, which was remarkably lower in patients with VATS (median = 2 mg/L; IQR = 1.

Discussion
The purpose of this study is to describe the demographic features, clinical presentation, management, and outcome of patients diagnosed with CCAP at AQWCH. The study group comprises children admitted to AQWCH diagnosed with CCAP between January 01, 2018, and December 31, 2020. The diagnosis of CCAP was made based on the clinical, radiological, and hematological findings. CCAP in previously healthy children was found to be associated with younger age, less than two years, a long duration of fever before admission, asymmetric chest pain, high inflammatory markers, low WBC, iron-deficiency anemia, and pretreatment with analgesic medications [10][11][12][13][14]. Similarly, our study showed that the median age was 2.5 years and the median duration of stay in the hospital was 16 days, with a longer duration of fever and high inflammatory markers.
In our study, the incidence of CCAP was found to be 15.3%. Of these, NP was 46.6%, PPE was 26.7%, and EMP was 26.7%. A higher incidence of complicated pneumonia compared with 3% of the British Association pediatric pneumonia audit [15] was found as a result of the delay in diagnosing and transferring complicated cases from other health facilities. A retrospective, observational study conducted at Boston Children's Hospital over 15 years showed an increased incidence of NP [16].
In this study, all patients had been vaccinated with PCV13 or PCV7 and Hib; however, blood culture was negative in all cases and pleural culture was positive only in three (14%) cases, which may have been due to the antibiotic use before admission or technical issues (methods of blood collection and quantities of blood collected). The retrospective study for EMP conducted in Canada from eight pediatric hospitals showed that 56/88 (63.6%) patients had positive pleural cultures [17], and a study conducted in North America showed that empirical antibiotic use decreases the positive cultures from 60% to 30% [18]. In another study conducted in six children's hospitals in the United States for patients hospitalized with CAP, a positive blood culture was obtained (2-5%) [19]. One of the detailed studies carried out before the introduction of PCV13 in the United States, supported by European studies after the introduction of PCV13, had shown that S. pneumoniae was the most common cause of CCAP. However, immunization with PCV13 was linked to decreased invasive pneumococcal disease and increased incidence of other organisms, especially S. pyogenes [20][21][22][23].
Imaging studies are important tools in CCAP. The chest X-ray helps in diagnosing diseases in patients, but there are chances of missed diagnosis in some patients, or the diagnosis may not differentiate between abscess formation and other thoracic malformations or consolidation [3]. Chest ultrasound is more sensitive than chest X-ray for detecting small effusions, differentiating and diagnosing septation, consolidation, lung abscess, and EMP [24,25]. CT does not provide a better diagnosis from chest ultrasound in CCAP in terms of planning its management and predicting outcomes. The CT should be used for complex cases to guide intervention and unclear diagnosis, or if there is no improvement in disease management [26], which is observed in this study.
Interestingly, in this study, PICU admission was high at 70% and low at 2% in the British Association pediatric pneumonia audit [15], and most patients were admitted preoperatively (67%). This was followed as there are no high-dependency care units in our facility. On analyzing the data for different types of CCAP, it was found that the duration of hospital stay for NP was shorter compared with other studies that ranged between 13.5 days and 27 days. Most patients underwent early surgical intervention (VATS) [27,28].
The surgical treatment for EMP is still disputable. VATS allows pleural debridement and drains pus from the pleural cavity under direct vision. VATS needs a specifically trained surgeon, which is unavailable in most centers. In both retrospective and prospective uncontrolled studies, VATS had a better result than chest tube drainage alone in resolving EMP [29]. In our study, four patients required a chest tube, and two underwent VATS later during the hospitalization. The complications reported with VATS are bronchopleural fistula, lung injury, and death [30]. In this study, most complications post-VATS were surgical emphysema (65%), pneumothorax (24%), and bronchopulmonary fistula (8%). Unfortunately, one (16.7%) patient of the total CCAP patients with NP died despite the required management and care taken in the PICU.

Strengths, limitations, and generalizability
This is a retrospective chart review study; many positive and significant statistical analysis results were found, which is the main strength of the study. Due to the small sample size, some variables could not be analyzed in this study. The data abstractors failed to develop the study proposal and protocol to minimize biases. In other words, they failed to understand the study objectives and research questions. The study was performed in a tertiary pediatric hospital where cases were referred from other hospitals, mostly with complications. Hence, this population did not match the general population. The incidence of CCAP for the whole UAE cannot be determined from this data.
One of the main limitations of this study was the lack of references and articles related to this subject, especially in the UAE, for comparing our study with the published research papers. Another major limitation was the non-availability of preadmission units for patients, which directly affected the progress of the treatment of CCAP.
The Responsible Conduct of Research (RCR) is considered protected health information that will not be revealed to anybody. However, the data may be used for future prospective studies, official monitoring, supervision, or if the law requires it.