Necrotizing pneumonia in a preterm baby

To the editor Necrotizing pneumonia (NP) is a complication of community-acquired pneumonia that may affect healthy children.1 Particularly severe and rapidly evolving NP have been associated with Staphylococcus aureus strains producing Panton-Valentine Leukocidin (PVL), which is a pore forming exotoxin that causes the lysis of leukocytes, leading to the destruction of the surrounding tissue. Presentation of NP during the first months of life is very uncommon. Here we describe a case of NP caused by methicillin-resistant S. aureus (MRSA) producing PVL in a preterm baby. A 2-month-old girl presented to the Pediatric Emergency at the ASST Big Metropolitan Hospital Niguarda, Milan, for a 3-day history of lethargy and decreased oral intake. She was born late preterm at 34 + 1 weeks of gestation from C-section. Her birth weight was 1.270 g. No ventilatory support was necessary. She was discharged home at day 27 weighing 2.710 g. At physical examination she was afebrile, but tachycardia (160 beats/min), tachypnea (70 breaths/min) and mild respiratory distress were present. Oxygen saturation was 89% in room air. On auscultation, decreased breath sounds and fine crackles at the left lung fields were revealed. Laboratory tests were normal, except for leukocytosis (WBC, 45×109/L), neutrophils (61%), increased C-Reactive Protein (12.5 mg/dL; nv <0.5), and hyponatremia (Na+ 122 mmol/L). Chest X-ray revealed a large consolidation and multiple air-filled cysts in the left lung, ipsilateral pleural effusion, and rightward mediastinal shift (Figure 1). She was admitted to the Neonatal Intensive Care Unit (NICU) where supplementation with high-flow oxygen nasal cannula, correction of hyponatremia by intravenous fluids, and parenteral feeding were provided. Nasal swabs for respiratory viruses including SARS-CoV2 tested negative. Empiric broad-spectrum antibiotic therapy with IV vancomycin and cefotaxime was started. On day 2, she was intubated and mechanically ventilated due to rapid deterioration of respiratory conditions. A chest drain was positioned and removed after draining 20 ml of pleural fluid over 24 h. Blood and urine cultures tested negative. Nasal swab, bronchial aspirate and pleural fluid cultures all grew MRSA. Cefotaxime was therefore replaced with IV ceftaroline according to antibiogram. Nasal and pharyngeal swab cultures in the parents were negative. Over the following days, repeated chest X-ray showed enlarged air-filled cavities in the left lung (Figure 2A). Chest CT-scan with contrast revealed two large pneumatoceles (PCs) and well-demarked necrotic areas into the left lower lobe (Figure 2B). After multidisciplinary team debating, the patient was extubated on day 6, and ventilatory support successfully switched to nasal CPAP. On day 7, PVL toxin tested positive on pleural fluid sample. Antibiotic therapy was adjusted, and vancomycin was replaced by clindamycin leading to slow but progressive improvement of clinical conditions, inflammatory biomarkers, and radiological findings. The patient was discharged home after a 4-week course of antibiotic therapy. Regular clinical and radiological (chest X-ray and lung ultrasonography) follow-up showed a full recovery. Our case is interesting for several reasons. The presentation of NP in the first weeks of life is uncommon. To our knowledge, only two cases have been described so far. In a 23-day-old newborn with NP caused by a MRSA producing PVL, pneumonia was preceded by respiratory syncytial virus infection, the clinical onset was more severe, and the conservative treatment was unsuccessful. The patient developed bronchopleural fistula and pneumothorax, and eventually underwent total right pneumectomy.2 In a more recent description, a 3-month-old female with postinfectious multiple bullous lesions in the right lung developed transient heart tamponade that successfully resolved with conservative treatment.3 NP caused by S. aureus strains producing PVL is an emerging problem, and PVL is a significant risk factor for severe S. aureus disease.4 Both methicillin susceptible S. aureus and MRSA strains can carry the genes of this toxin, but the role of methicillin resistance in the severity of NP remains controversial. Imaging has a fundamental role in the diagnosis of complicated pneumonia. Chest X-ray and ultrasounds are the first-choice imaging tools but are less sensitive than CT scan in distinguishing lung lesions in NP. In our case, chest CT scan was performed because worsening of the radiological picture did not allow to differentiate PC from loculated pneumothorax. Despite the high sensitivity in evaluating lesions in NP, chest CT scan should be considered if a diagnostic doubt is present or when there is no clinical improvement after appropriate treatment.1 Rapid progression of lung lesions is typical of NP. PC develops secondary to localized bronchiolar and alveolar necrosis, which allow one-way passage of air into the interstitial space. In our patient, initial chest X-ray revealed multiple small PCs in the left lung, but they suddenly merged into larger air-filled cavities. Progressive distension of PC is a critical phase, which may worsen over hours and therefore needs careful monitoring of clinical conditions. Indeed, compression of the adjacent parenchyma may induce respiratory or cardiovascular compromise, and/or rupture through the pleural space may provoke pneumothorax and bronchopleural fistula. Patients receiving mechanical ventilation have an increased risk for developing such complications, and any therapeutical decision should be carefully thought out.5 Despite negative cultures for S. aureus infection in the parents, both the long-time from hospital discharge to the onset of symptoms and the absence of MRSA contamination in our NICU over the last 3 years make a community-acquired infection likely. To summarize, we present an unusual case of a 2-month-old premature girl with severe NP caused by MRSA producing PVL. Due to the perceived risk of complications in a baby receiving mechanical ventilation, a conservative approach with close follow-up was successfully adopted. In similar cases, we recommend that intervention (i.e., medical vs. surgical) should be primarily based on clinical judgment rather than on simple imaging evaluation. Valeria Rubinacci: Conceptualization; investigation; writing—original draft. Stefano Martinelli: Writing—review and editing. Alice Proto: Formal analysis. Ottavio Vitelli: Methodology. Marco Fossati: Data curation. Fernando M. de Benedictis: Supervision. The authors declare no conflict of interest. The Big Metropolitan Hospital Niguarda, Milan, Italy, does not require an Ethical statement when single cases are described, and a written informed consent was released by patients/parents. Written informed consent was obtained by patient's parents for the images included in this article. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.