Necrotizing pneumonia (NP), sometimes called cavitary pneumonia, cavitary necrosis or pulmonary gangrene, involves the loss of normal pulmonary parenchymal architecture with the presence of thin-walled cavitary lesions.1,2 NP was first described in adults in a context of comorbidity and coinfection with anaerobic bacteria1 but is a rare complication of community-acquired pneumonia (CAP) in children and its incidence is poorly documented.3,4
The aim of our study was to describe the epidemiology of children admitted to the pediatric Robert-Debré Hospital with NP between 2006 and 2011, their clinical, biological, microbiological and radiological features, as well as their management and outcome.
PATIENTS AND METHODS
Population and Definitions
Children <18 years hospitalized in the Robert-Debré teaching Hospital between May 2006 and April 2011 with community-acquired NP were retrospectively included. The study was approved by the Robert-Debré Hospital Ethics Committee. Patients with nosocomial NP or cavitary pneumonia due to Mycobacterium tuberculosis were excluded. Diagnosis of NP was made by the combination of clinical signs of lung infection and radiological signs of NP determined by the existence of pneumonia (consolidation visible in the form of dense, homogeneous, systematized and nonretractile opacity) with multiple cavity defined as a follow on radiographs and computed tomography (CT) scans: gas-filled space, seen as a lucency or low attenuation area, within pulmonary consolidation, usually produced by the expulsion or drainage of a necrotic part of the lesion via the bronchial tree and sometimes containing a fluid level.2 Cases of lung abscess, identified on radiograph by a single cavity with an air-fluid level and on CT scan by a thick-walled cavity containing fluid and enhanced after injection of contrast agent were excluded. NP were microbiologically documented if at least 1 of the 4 following conditions was satisfied: (1) ≥1 positive blood culture, (2) positive pleural fluid sample with the presence of pus, (3) protected specimen brush with bacteria count >103 CFU/mL, and (4) bacteria count >104 CFU/mL on bronchoalveolar lavage fluid.
Data for patients attending our Pediatric Emergency Department were collected prospectively using the Urqual v5 software (McKesson Corp., Paris, France). The files of children hospitalized for CAP were identified in the French hospitals medico-administrative database (Programme de médicalisation des systèmes d'information; PMSI) and reexamined to identify patients presenting NP on admission or during the course of their stay. Data were collected retrospectively from the patient records. Chest radiographs and CT scans were reexamined by a pediatric radiologist.
Epidemiological and Clinical Data
Between May 2006 and April 2011, 4859 consultations for pneumonia were recorded in our Pediatric Emergency Department: 635 (13%) patients required hospitalization >24 hours and 41 (0.8%) patients presented NP (Fig. 1). The rate of CAP complicated by NP doubled between 2006 to 2009 and 2009 to 2011 from 4.5% to 9%.
The median age was 14 months (1 month–16 years) with a female-to-male ratio of 0.86. None of the children had a known underlying immune deficiency. The children had a history of fever or respiratory signs before hospitalization for a median of 4 days (range: 1–10 days). Twenty-five children (61%) sought medical visits before the day of hospitalization and 18 (44%) received antibiotic within 7 days of hospitalization. At admission, the median of body temperature was 39°C (37.5°C–41°C), 25 (61%) patients presented a decrease in vesicular breath sounds, 14 (34%) the presence of crackles and 13 (32%) had an oxygen saturation ≤92%.
Biological and Radiological Data
Values for C-reactive protein, white blood cells and neutrophils on admission were 141 mg/L (range: 5–434), 20,000/mm3 (range: 1700–44,300) and 13,401/mm3 (range: 714–33,740), respectively.
The chest radiograph revealed necrosis on admission in only 27% of cases. The median time for the appearance of necrosis from the date of hospitalization was 4 days and 20% were diagnosed after 7 days. A chest CT scan was performed in children with persistent fever, despite antibiotic treatment or with suspicion of cavitation on the chest radiograph for 23/41 patients (56%) and led to the diagnosis of 12 cases of necrosis not detected on the chest radiograph. A total of 26 patients presented pleural empyema (63%), including 19 children on admission. Eight patients presented a pneumothorax including 6 hydropneumothorax, whereas 4 patients presented pneumatocele.
All patients had blood cultures, 15 had a pleural fluid examination and 3 and 1 had a bronchoalveolar lavage and a protected specimen brush, respectively. Results of microbiological examinations for each patient with documented NP are presented in the supplementary data (see Table, Supplemental Digital Content 1, http://links.lww.com/INF/B589). A bacterium was identified in 21/41 cases (51%): 13 Staphylococcus aureus, 7 Streptococcus pneumoniae and 1 Fusobacterium nucleatum in a 14-year-old patient with sickle cell anemia (Fig. 1). All the S. pneumoniae strains were isolated in children aged between 6 months and 3 years. The S. aureus strains were isolated in 6 children <6 months, 4 children aged between 6 months and 3 years and 3 children >3 years. All strains encoded genes of Panton-Valentine leukocidin and 1/13 was methicillin-resistant but not belonging to the European ST80 clone. Of the 5 strains of S. pneumoniae, 3 had reduced sensitivity to penicillin and serotyping revealed 1 serotype 7F, 2 serotypes 3 and 2 serotypes 19A in 5 children full vaccinated with a 7- valent pneumococcal conjugate vaccine.
Management and Outcome
The median duration of fever, hospitalization and total antibiotherapy was 7 days (range: 1–25 days), 16 days (range: 7–43 days) and 42 days (range: 31–60 days), respectively, without differences between S. aureus and S. pneumoniae NP. Twenty-six patients (63%) needed oxygenotherapy and 7 (17%) required intensive care unit. Only one 5-year- old patient hospitalized for methicillin-susceptible S. aureus NP with previous influenza-like symptoms, hemoptysis and leukopenia (neutrophil count of 714 mm3) on admission had a very severe outcome marked by acute respiratory distress syndrome. On the 26 patients who presented an empyema, 10 had thoracocenthesis, 3 had chest tube thoracostomy followed by thoracoscopic surgical drainage and 3 had a thoracoscopic surgical drainage at the outset. There were no deaths in our study group.
Over the past few years, an increase in complicated CAP has been reported in the literature, but there have been few reports on the incidence of pediatric NP.4,5 To the best of our knowledge, the present series comprising 41 patients is the largest series of pediatric NP analyzed in Europe. Between 2006 to 2009 and 2009 to 2011, the proportion of NP among CAP doubled from 4.5% to 9%. The increased number of complicated CAP cases during the H1N1 pandemia might explain this trend. However, the retrospective nature of our study has not allowed viral research that could support this hypothesis.
Thirteen patients (32%) presented with S. aureus NP with a median age of 6 months (range: 1 month–16 years). Interestingly, all the S. aureus strains encoded the Panton-Valentine leukocidin (PVL), an exotoxin which forms pores in the cell membrane and leads to lysis of the host’s blood mononuclear cells such as neutrophils, monocytes and macrophages, inducing a significant release of inflammation mediators.6 In 2002, cases of very severe community-acquired PVL-positive S. aureus pneumonia affecting patients hospitalized in intensive care were described in France for the first time.7 PVL-positive S. aureus NP occurred in healthy children and young adults presenting respiratory signs following influenza-like symptoms with signs of serious sepsis and leukopenia and the death rate was above 50%.7,8 In our study, only 1 patient presented these clinical features. The other 12 patients (median age: 5 months) were dramatically less severe without leukopenia, and only 25% were hospitalized in intensive care. Our S. aureus NP seem more similar to the historic form of pediatric S. aureus pneumonia described in the 1950s under the term “bullous staphylococcal pneumonia.”9 Carrillo-Marquez et al10 also recently described infants hospitalized for PVL-positive S. aureus CAP and presenting a favorable outcome, despite cavitary lesions. The pathophysiological mechanism of these NP is probably different from that described by Gillet et al7,8 for which necrosis affects the respiratory epithelium (bronchial surface and alveoli), leading to major alteration of gas exchanges without cavitation in most cases.
S. pneumoniae was isolated in 17% of cases, and the 5 serotypes identified in the present series were not covered by the 7-valent pneumococcal conjugate vaccine. PCV13 including 6 additional serotypes (1, 3, 5, 6A, 7F and 19A) and introduced in June 2010 in France should cover these serotypes.
In the present series, illness was prolonged, with apyrexia obtained after 7 days, but none of the patients underwent drainage of the necrotic lesions or resection of lung tissue. Outcome was favorable in all cases with medical treatment alone, as in the 80 children in the series documented by Sawicki et al.4
The current study has some limitations. The study population was drawn from a single pediatric hospital, it was a retrospective study and 18/41 patients received antibiotics that could bias the results of microbiological tests.
NP is a rare complication of CAP in children, and further researches are needed to accurately define their epidemiology. In addition, comparative studies of patients hospitalized for NP, empyema or CAP might help identify factors predictive of the development of NP.
The authors thank Sheila Carrodus for editorial assistance.
1. Gadkowski LB, Stout JE. Cavitary pulmonary disease. Clin Microbiol Rev. 2008;21:305–333 table of contents.
2. Hansell DM, Bankier AA, MacMahon H, et al. Fleischner Society: glossary of terms for thoracic imaging. Radiology. 2008;246:697–722
3. Hacimustafaoglu M, Celebi S, Sarimehmet H, et al. Necrotizing pneumonia
. Acta Paediatr. 2004;93:1172–1177
4. Sawicki GS, Lu FL, Valim C, et al. Necrotising pneumonia is an increasingly detected complication of pneumonia in children
. Eur Respir J. 2008;31:1285–1291
5. Lee GE, Lorch SA, Sheffler-Collins S, et al. National hospitalization trends for pediatric pneumonia and associated complications. Pediatrics. 2010;126:204–213
6. Labandeira-Rey M, Couzon F, Boisset S, et al. Staphylococcus aureus Panton-Valentine leukocidin
causes necrotizing pneumonia
. Science. 2007;315:1130–1133
7. Gillet Y, Issartel B, Vanhems P, et al. Association between Staphylococcus aureus strains carrying gene for Panton-Valentine leukocidin
and highly lethal necrotising pneumonia in young immunocompetent patients. Lancet. 2002;359:753–759
8. Gillet Y, Vanhems P, Lina G, et al. Factors predicting mortality in necrotizing community-acquired pneumonia caused by Staphylococcus aureus containing Panton-Valentine leukocidin
. Clin Infect Dis. 2007;45:315–321
9. Edwards RL, Gresham GA. Staphylococcal pneumonia in the neonatal period. Lancet. 1960;2:1057–1058
10. Carrillo-Marquez MA, Hulten KG, Hammerman W, et al. Staphylococcus aureus pneumonia in children
in the era of community-acquired methicillin-resistance at Texas Children
’s Hospital. Pediatr Infect Dis J. 2011;30:545–550