A 74-year-old female with a medical history of multiple myeloma, which relapsed after autologous stem-cell transplantation in 2016, was treated with daratumumab, lenalidomide, and dexamethasone. The chemotherapy was administered three weeks before hospital admission. The patient presented with fever (38.7 °C tympanic) and fatigue. Blood analysis at hospitalization showed neutropenia (absolute count neutrophils was 497/mm3) and CRP of 342 mg/L. A CCT scan found an opacification of the left lower lobe with the presence of an abscess (44 × 21 mm), Supp. Materials Fig. 1. The patient was started on broad empiric antibiotics (cefepime 2 g TID and metronidazole 500 mg TID), and a bronchoscopy was ordered. Bronchoalveolar fluid (BAL) cultures were negative, but Legionella pneumophila serogroup 1 NAAT and the galactomannan on the same sample (index:1.12) were both positive. Furthermore, LUA was also positive. Antibiotic therapy was narrowed to moxifloxacin (MFX) 400 mg QD. As the diagnosis of probable invasive pulmonary mold diseases was met following the criteria of EORTC/MSG [16], the patient was also treated with voriconazole. A compatible CCT imaging is part of the EORTC/MSG diagnostic criteria of an invasive fungal infection. Pulmonary aspergillosis may present in an initial phase with typical signs as nodular opacities of a halo sign. However, CCT images may not differ between mold or bacterial infections in a later phase as lung cavitations appear [17]. Considering the neutropenia, Filgrastim 0.3 mg injection was given daily until complete recovery of the neutrophils absolute count. After 14 days of treatment, the patients persisted to be febrile (38.1 °C tympanic), CRP decreased to 110 mg/L and neutrophils were in the normal range (absolute count of 5003/mm3). CCT was repeated and an increase in pleura effusion and the dimension of the abscess was observed (60 × 34 mm), supplementary materials Fig. 1. A multidisciplinary discussion was organized to choose the most appropriate treatment option, involving a pneumologist, infectious disease specialist, haematologist, radiologist, and thorax surgeon. The abscess was not ripe enough to be drained, so a conservative approach was considered. A chest tube was placed to evacuate pleural effusion. Furthermore, moxifloxacin was continued for a total of 21 days, and voriconazole for six weeks. After stopping antibiotic treatment and removing the chest tube, the CCT revealed a persistent abscess and consolidation without pleural effusion. Finally, CCT was repeated four months after the diagnosis of the infection, and a complete cure of the abscess was observed (Supp. Materials Fig. 1). Chemotherapy was restarted only after these reassuring findings.
Second caseA 55-year-old female with a medical history of asthma on inhaled steroids and type 2 diabetes on insulin therapy was diagnosed with systemic sarcoidosis three weeks before admission. The disease involved mediastinal, peri-hepatic, and splenic lymph nodes. Treatment with high-dose methylprednisolone (64 mg QD) was initiated. She presented with a dry cough and haemoptysis. Blood analysis found a CRP of 231 mg/L and leucocytosis (14,500/mm3); chest X-ray showed opacification left upper lobe (Fig. 1). Bronchoscopy was performed, and the BAL sample was analyzed. The sample tested positive for Legionella pneumophila serogroup 1 NAAT. Therefore, MFX 400 mg QD was started, and methylprednisolone was reduced to 32 mg QD. Five days after admission, the patient was afebrile and did not require oxygen. She was discharged with a close follow-up, 10 days after discharge. At that time, she reported persistent cough and chest pain, and inflammatory markers were decreased (CRP of 12 mg/L and white blood cell count of 11,400/mm3). A chest X-ray was obtained, which suggested an evolution into cavitary pneumonia (Fig. 1). This suspicion was confirmed by CCT. Therefore, methylprednisolone was decreased to 16 mg QD, and antibiotic treatment was continued for 21 days in total with favourable clinical and radiological evolution (Fig. 1).
Fig. 1Evolution of chest imaging in a patient affected by Legionnaires’ Disease lung abscess. Evolution over time of chest X-ray of the second reported case, from right to left respectively (subpanel A, B, C). The lung lesion is underlined with a black arrow. A (left): left apical consolidation; B (middle): evolution of the consolidation into cavitation; C (right): partially resolution with shrinking of the lesion
Systematic reviewStudy selection and characteristicsThe PRISMA 2020 flow diagram is shown in Fig. 2. Following the screening procedure, 23 case series or case reports, and six guidelines were included. After reading the full text of the guidelines, one additional guideline was found through the references of the guidelines selected by the research strategy [18]. Table 1 lists the included guidelines. Supp. Materials Table 2 depicts the selected manuscripts and the number of the described patients per article.
Fig. 2Study flowchart. Edition PRISMA 2020 flow diagram
Table 1 Considered guidelines and their recommendationsResult of synthesis (case reports)Table 2 illustrates the descriptive characteristics of the 29 patients found by the systematic review. The median age was 48 years and 65% were male. The most reported comorbidity was hematologic malignancy (n:8 61%), and the most prevalent immunosuppressant treatment was corticosteroids (n:17 85%). The preferred diagnostic method was bacterial culture, mostly performed on BAL fluid or lung tissue biopsies, with a sensitivity of 86% (4 false negatives). This was followed by LUA testing, which was performed for 25 patients and had a sensitivity of 52% (12 false negatives, only patients with LD with non-Lps1 pathogenic agent). Lps1 was the most commonly isolated species of Legionella, affecting 52% of the described patients. A delay of the beginning of an effective therapy superior to five days or more occurred in 56% of the patients. After diagnosis, bi-therapy with two or more effective antimicrobials was started in 52% of the patients. Supp. Materials Table 3 illustrates the antimicrobial regimen adaption after diagnosis of LD abscess. Fluoroquinolones were the preferred antibiotic class, with 76% of the patients being treated with one antimicrobial from this class. A co-infection diagnosed through isolation of a co-pathogen occurred in 11 cases, with four patients growing an anaerobic bacterium. Particularly, two times bothFusobacterium nucleatum and Prevotella species [7, 9, 19, 20]. In case of polymicrobial flora, isolates were cultured in lung biopsy (two patients), protected bronchial brush specimen (one patient), empyema sample (two patients), BAL fluid (five patients), and finally one patient had a sputum sample growing Enterococcus faecalis and a positive LUA. An absence of anaerobic coverage was observed in 33% of the patients. The median duration of therapy was 42 days. Frequently the consolidation evolved into empyema, and 59% of the patients needed chest tube drainage. Thoracic surgery was performed in 22% of the patients: two lobectomies, one wedge resection, and two abscess resections. One operation was not in detail described. The reported reasons for surgery were clinical deteriorations (increased pain and inflammation markers, increase in abscess volume, and appearance of empyema) and in two cases unclear diagnostic prompted surgical management [9, 21]. All-cause mortality was 21%.
Table 2 Characteristics of cavity Legionnaires’ disease patients found by systematic reviewResult of synthesis (guidelines)Four guidelines provided suggestions for LD treatment [12,13,14, 20]. Three reports proposed monotherapy [12, 13, 22], while one suggested combination treatment [14]. Furthermore, fluoroquinolones and macrolides were mentioned as the first antibiotic choice in four guidelines [12,13,14, 20]. Finally, two guidelines advocated for initial intravenous therapy before possible oral relay [13, 22].
Four guidelines discussed therapy for lung abscesses, and all four reports suggested the potential pathogenic role of anaerobes [12, 13, 18, 22]. One article proposed a length of treatment of six to eight weeks [12], while another suggested treatment until radiological resolution [22].
Certainty of the evidence reported in the guidelinesRecommendations for LD antimicrobial therapy are based on observational reports [21, 23,24,25]. In a prospective study conducted in Spain during an outbreak, there was no difference in clinical outcome or length of hospitalisation between patients treated with fluoroquinolones and macrolides (292 patients). However, in severe patients, treatment with levofloxacin (LVX) reduced complications and length of stay [23]. Another prospective study observed 45 severe LD patients and found no improvement when using a combination of LVX and rifampicin (RIF) compared to LVX monotherapy [24]. Finally, a study on 25 patients found increased mortality in those treated with monotherapy compared to the combination of clarithromycin and RIF [25].
Anaerobes are one of the main pathogens involved in lung abscesses. A study in Taiwan found anaerobes in 34% of the specimens obtained from 90 patients [11]. Guidelines for managing lung abscesses rely on RCTs [24, 26,27,28]. RCTs comparing the management of pulmonary abscess with antimicrobials without anaerobic coverage resulted in lower clinical cure rates than therapy with an anaerobic spectrum [26, 27]. Additionally, an RCT comparing MFX and ampicillin-sulbactam in 139 patients for the treatment of aspiration pneumonia and pulmonary abscess reported similar clinical outcomes. The median duration of MFX therapy for lung cavitation was 30 days, and the adverse events were comparable between the two treatment arms [28].
Comments (0)