Background A growing body of observational evidence supports the value of ceftazidime-avibactam (CAZ-AVI) in managing infections caused by carbapenem-resistant Enterobacteriaceae (CRE). Methods We retrospectively analyzed observational data on the use and outcomes of CAZ-AVI therapy for infections caused by KPC-producing K. pneumoniae (KPC-Kp) strains. Multivariate regression analysis was used to identify variables independently associated with 30-day mortality. Results were adjusted for propensity score for receipt of CAZ-AVI combination regimens vs. CAZ-AVI monotherapy. Results The cohort comprised 577 adults with bloodstream infections (BSIs) (n=391) or non-bacteremic infections (nBSIs) involving mainly the urinary tract, lower respiratory tract, intra-abdominal structures. All received treatment with CAZ-AVI alone (n=165) or with one or more other active antimicrobials (n=412). The all-cause mortality rate 30 days after infection onset was 25% (146/577). There was no statistically significant difference in mortality between patients managed with CAZ-AVI alone and those treated with combination regimens (26.1% vs. 25.0%, P=0.79). In multivariate analysis, mortality was positively associated with the presence at infection onset of septic shock (P=0.002), neutropenia (P <0.001), or an INCREMENT score >8 (P=0.01); with LRTI (P=0.04); and with CAZ-AVI dose adjustment for renal function (P=0.01). Mortality was negatively associated with CAZ-AVI administration by prolonged infusion (P=0.006). All associations remained significant after propensity score adjustment. Conclusions CAZ-AVI is an important option for treating serious KPC-Kp infections, even when used alone. Further study is needed to explore the drug’s seemingly more limited efficacy in LRTIs and the potential survival benefits of prolonging CAZ-AVI infusions to 3 hours or more.

Background Few data are reported in the literature about the outcome of patients with severe extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) infections treated with ceftolozane/tazobactam (C/T), in empiric or definitive therapy. Methods A multicenter retrospective study was performed in Italy (June 2016–June 2019). Successful clinical outcome was defined as complete resolution of clinical signs/symptoms related to ESBL-E infection and lack of microbiological evidence of infection. The primary end point was to identify predictors of clinical failure of C/T therapy. Results C/T treatment was documented in 153 patients: pneumonia was the most common diagnosis (n = 46, 30%), followed by 34 cases of complicated urinary tract infections (22.2%). Septic shock was observed in 42 (27.5%) patients. C/T was used as empiric therapy in 46 (30%) patients and as monotherapy in 127 (83%) patients. Favorable clinical outcome was observed in 128 (83.7%) patients; 25 patients were considered to have failed C/T therapy. Overall, 30-day mortality was reported for 15 (9.8%) patients. At multivariate analysis, Charlson comorbidity index >4 (odds ratio [OR], 2.3; 95% confidence interval [CI], 1.9–3.5; P = .02), septic shock (OR, 6.2; 95% CI, 3.8–7.9; P < .001), and continuous renal replacement therapy (OR, 3.1; 95% CI, 1.9–5.3; P = .001) were independently associated with clinical failure, whereas empiric therapy displaying in vitro activity (OR, 0.12; 95% CI, 0.01–0.34; P < .001) and adequate source control of infection (OR, 0.42; 95% CI, 0.14–0.55; P < .001) were associated with clinical success. Conclusions Data show that C/T could be a valid option in empiric and/or targeted therapy in patients with severe infections caused by ESBL-producing Enterobacterales. Clinicians should be aware of the risk of clinical failure with standard-dose C/T therapy in septic patients receiving CRRT.

BackgroundMesothelioma is a lung cancer that kills thousands of people worldwide annually, especially those with exposure to asbestos. Diagnosis of mesothelioma in patients often requires time-consuming imaging techniques and biopsies. Machine learning can provide for a more effective, cheaper, and faster patient diagnosis and feature selection from clinical data in patient records.Methods and findingsWe analyzed a dataset of health records of 324 patients having mesothelioma symptoms from Turkey. The patients had prior asbestos exposure and displayed symptoms consistent with mesothelioma. We compared probabilistic neural network, perceptron-based neural network, random forest, one rule, and decision tree classifiers to predict diagnosis of the patient records. We measured classifiers’ performance through standard confusion matrix scores such as Matthews correlation coefficient (MCC). Random forest outperformed all models tried, obtaining MCC = +0.37 on the complete imbalanced dataset and MCC = +0.64 on the under-sampled balanced dataset. We then employed random forest feature selection to identify the two most relevant dataset traits associated with mesothelioma: lung side and platelet count. These two risk factors resulted so predictive, that decision tree focusing on them achieved the second top accuracy on the complete dataset diagnosis prediction (MCC = +0.28), outperforming all other methods and even decision tree itself applied to all features.ConclusionsOur results show that machine learning can predict diagnoses of patients having mesothelioma symptoms with high accuracy, sensitivity, and specificity, in few minutes. Additionally, random forest can efficiently select the most important features of this clinical dataset (lung side and platelet count) in few seconds. The importance of pleural plaques in lung sides and blood platelets in mesothelioma diagnosis indicates that physicians should focus on these two features when reading records of patients with mesothelioma symptoms. Moreover, doctors can exploit our machinery to predict patient diagnosis when only lung side and platelet data are available.

Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Fig. 1 Timeline of platelet count after treatment of secondary (systemic lupus erythematosus-related) ITP recurrence in a 37-year-old female patient with COVID-19 infection. IVIG, intra-venous immune globulins; IVMP, intra-venous methyl-prednisolone; PDN, prednisone