Assessing the utility of natural language processing for detecting postoperative complications from free medical text.

BACKGROUND: Postoperative complication rates are often assessed through administrative data, although this method has proven to be imprecise. Recently, new developments in natural language processing have shown promise in detecting specific phenotypes from free medical text. Using the clinical challenge of extracting four specific and frequently undercoded postoperative complications (pneumonia, urinary tract infection, sepsis, and septic shock), it was hypothesized that natural language processing would capture postoperative complications on a par with human-level curation from electronic health record free medical text. METHODS: Electronic health record data were extracted for surgical cases (across 11 surgical sub-specialties) from 18 hospitals in the Capital and Zealand regions of Denmark that were performed between May 2016 and November 2021. The data set was split into training/validation/test sets (30.0%/48.0%/22.0%). Model performance was compared with administrative data and manual extraction of the test data set. RESULTS: Data were obtained for 17 486 surgical cases. Natural language processing achieved a receiver operating characteristic area under the curve of 0.989 for urinary tract infection, 0.993 for pneumonia, 0.992 for sepsis, and 0.998 for septic shock, whereas administrative data achieved a receiver operating characteristic area under the curve of 0.595 for urinary tract infection, 0.624 for pneumonia, 0.571 for sepsis, and 0.625 for septic shock. CONCLUSION: The natural language processing approach was able to capture complications with acceptable performance, which was superior to administrative data. In addition, the model performance approached that of manual curation and thereby offers a potential pathway for complete real-time coverage of postoperative complications across surgical procedures based on natural language processing assessment of electronic health record free medical text.

Acute primary CMV infection complicated by pneumonitis and ITP in young immunocompetent woman in a regional Queensland Hospital.

We present the first published case of simultaneous pneumonitis and immune thrombocytopenic purpura secondary to primary cytomegalovirus (CMV) infection in an immunocompetent patient. Treatment with oral valganciclovir for 2 weeks successfully led to complete clinical recovery. CMV is traditionally associated with infection in immunocompromised patients and neonates; however, evidence of severe CMV infections in immunocompetent hosts is emerging. It is important to highlight the broad range of clinical presentations of CMV infections to prevent diagnostic delay and associated morbidity and expense.

Alveolar-arterial oxygen gradient: An early marker to predict the severity of community-acquired pneumonia in children.

To study the clinical significance of alveolar-arterial oxygen gradient (PA-aO2) for children with community-acquired pneumonia (CAP). A prospective study was carried out from January 2020 to June 2023, Overall, 100 patients were included in the study, 35 had severe pneumonia, whereas, 65 had non-severe pneumonia. Clinical and laboratory data were retrospectively collected at the time of hospital admission and during hospitalization. Patients were divided into severe and non-severe groups. PA-aO2 was significantly higher among children with severe pneumonia, as determined by WHO, PRESS (P < .001). PA-aO2 was significantly higher among children with mechanical ventilation, shock, sepsis, and mortality. Receiver operating characteristic curve (ROC) analysis for PA-aO2 showed that the area under the curve was 0.76 (P value < .05), with a sensitivity of 84.3% and a specificity of 67.9%. Our study suggests that PA-aO2 level has a predictive value for detecting community-acquired pneumonia severity in children.

Construction and external validation of a scoring prediction model for mortality risk within 30 days of community-acquired pneumonia in children admitted to the pediatric intensive care unit: A multicenter retrospective case-control study.

In this study, we constructed and validated a scoring prediction model to identify children admitted to the pediatric intensive care unit (PICU) with community-acquired pneumonia (CAP) at risk for early death. Children with CAP who were admitted to the PICU were included in the training set and divided into death and survival groups according to whether they died within 30 days of admission. For univariate and multifactorial analyses, demographic characteristics, vital signs at admission, and laboratory test results were collected separately from the 2 groups, and independent risk factors were derived to construct a scoring prediction model. The ability of the scoring model to predict CAP-related death was validated by including children with CAP hospitalized at 3 other centers during the same period in the external validation set. Overall, the training and validation sets included 296 and 170 children, respectively. Univariate and multifactorial analyses revealed that procalcitonin (PCT), lactate dehydrogenase (LDH), activated partial thromboplastin time (APTT), and fibrinogen (Fib) were independent risk factors. The constructed scoring prediction model scored 2 points each for PCT ≥ 0.375 ng/mL, LDH ≥ 490 U/L, and APTT ≥ 31.8 s and 1 point for Fib ≤ 1.78 g/L, with a total model score of 0-7 points. When the score was ≥ 5 points, the sensitivity and specificity of mortality diagnosis in children with CAP were 72.7% and 87.5%, respectively. In the external validation set, the sensitivity, specificity, and accuracy of the scoring model for predicting the risk of CAP-related death were 64.0%, 92.4%, and 88.2%, respectively. Constructing a scoring prediction model is worth promoting and can aid pediatricians in simply and rapidly evaluating the risk of death in children with CAP, particularly those with complex conditions.

Severe Pneumonia Caused by Respiratory Syncytial Virus and Adenovirus in Children from 2 to 24 Months at Children's Hospital 1 in Ho Chi Minh City, Vietnam.

In Vietnam, due to the lack of facilities to detect respiratory viruses from patients' specimens, there are only a few studies on the detection of viral pathogens causing pneumonia in children, especially respiratory syncytial virus (RSV) and adenovirus (Adv). Here, we performed a cross-sectional descriptive prospective study on 138 children patients from 2 to 24 months old diagnosed with severe pneumonia hospitalized at the Respiratory Department of Children's Hospital 1 from November 2021 to August 2022. The number of patients selected in this study was based on the formula n = ([Z(1 - α/2)]2 × P [1 - P])/d2, with α = 0.05, p = 0.5, and d = 9%, and the sampling technique was convenient sampling until the sample size was met. A rapid test was used to detect RSV and Adv from the nasopharyngeal swabs and was conducted immediately after the patient's hospitalization. Laboratory tests were performed, medical history interviews were conducted, and nasotracheal aspirates were collected for multiplex real-time PCR (MPL-rPCR) to detect viral and bacterial pathogens. The results of the rapid test and the MPL-rPCR in the detection of both pathogens were the same at 31.9% (44/138) for RSV and 8.7% (7/138) for Adv, respectively. Using MPL-rPCR, the detection rate was 21% (29/138) for bacterial pathogens, 68.8% (95/138) for bacterial-viral co-infections, and 6.5% (9/138) for viral pathogens. The results showed few distinctive traits between RSV-associated and Adv-associated groups, and the Adv group children were more prone to bacterial infection than those in the RSV group. In addition, the Adv group experienced a longer duration of treatment and a higher frequency of re-hospitalizations compared to the RSV group. A total of 100% of Adv infections were co-infected with bacteria, while 81.82% of RSV co-infected with bacterial pathogens (p = 0.000009). This study might be one of the few conducted in Vietnam aimed at identifying viral pathogens causing severe pneumonia in children.

Management and documentation of pneumonia - a comparison of patients consulting primary care and emergency care.

OBJECTIVE: To compare management and documentation of vital signs, symptoms and infection severity in pneumonia patients seeking primary care and emergency care without referral. DESIGN: Medical record review of vital signs, examination findings and severity of pneumonia. SETTING: Primary and emergency care. SUBJECTS: Two hundred and forty patients diagnosed with pneumonia. MAIN OUTCOME MEASURES: Vital signs, examination findings and severity of pneumonia. Assessments of pneumonia severity according to the reviewers, the traffic light score and CRB-65. RESULTS: Respiratory rate, blood pressure, heart rate and oxygen saturation were less often documented in primary care (p < .001). Chest X-ray was performed in 5% of primary care patients vs. 88% of emergency care patients (p < .01). Primary care patients had longer symptom duration, higher oxygen saturation and lower respiratory rate. In total, the reviewers assessed 63% of all pneumonias as mild and 9% as severe. The traffic light scoring model identified 11 patients (9%) in primary care and 53 patients (44%) in emergency care at high risk of severe infection. CONCLUSIONS: Vital signs were documented less often in primary care than in emergency care. Patients in primary care appear to have a less severe pneumonia, indicating attendance to the correct care level. The traffic light scoring model identified more patients at risk of severe infection than CRB-65, where the parameters were documented to a limited extent.

Pneumonia patients attending primary care have less affected vital signs than those attending emergency care.Vital signs were less documented in primary care than in emergency care.Patients with pneumonia seem to attend the correct level of care when they have the possibility to choose without a referral.CRB-65 was not possible to count in most primary care patients due to lack of documentation.

Assessment of sarcopenic obesity as a predictor of pneumonia in patients with stable schizophrenia-A prospective study.

OBJECTIVE: The European Society for Clinical Nutrition and Metabolism (ESPEN) and the European Association for the Study of Obesity (EASO) recently released the first international consensus on the diagnostic criteria for Sarcopenic obesity (SO). The present study aimed to explore the ability of SO to predict the risk of pneumonia in patients with stable schizophrenia. METHODS: This was a prospective study involving hospitalized patients with schizophrenia aged ≥50 years from two mental health centers in western China. Baseline patient data were collected from September 1 to September 30, 2020. Follow-up data on pneumonia were collected from October 2020 to October 2022. The diagnosis of SO was based on the ESPEN/EASO criteria. Patients were assessed for handgrip strength (HGS), skeletal muscle mass/weight (SMM/W), and fat mass percentage (FM%). Logistic regression analysis was used to explore the effect of SO on the risk of pneumonia in patients with stable schizophrenia. RESULTS: A total of 320 patients with stable schizophrenia were included. Of these, 74 (23.13%) were diagnosed with SO, while 117 (36.56%) developed pneumonia. Compared with patients in the non-low HGS, non-low HGS + non-low SMM/W (or non-low HGS + low SMM/W or low HGS + non-low SMM/W) and non-SO groups, the proportions of patients with pneumonia in the low HGS (42.3% vs. 25.9%, p = 0.004), low HGS + low SMM/W (45.3% vs. 33.3%, p = 0.048), and SO (47.3% vs. 33.3%, p = 0.029) groups, respectively, were higher. However, there was no difference in the proportion of patients with pneumonia in the low SMM/W group and the obese group compared with the non-low SMM/W and non-obese groups. Further logistic regression analysis after adjustment for potential influencing factors showed that compared with the non-low HGS group, patients in the low HGS group had a higher risk of pneumonia (OR = 1.892, 95%CI: 1.096-3.264). CONCLUSION: SO defined according to the ESPEN/EASO criteria was not found to be significantly associated with the development of pneumonia in patients with stable schizophrenia. Further verification of these results is needed with larger sample sizes and the establishment of a cutoff value for this population.

Immune checkpoint inhibitor-related pneumonitis: research advances in prediction and management.

The advent of immune-checkpoint inhibitors (ICIs) has revolutionized the treatment of malignant solid tumors in the last decade, producing lasting benefits in a subset of patients. However, unattended excessive immune responses may lead to immune-related adverse events (irAEs). IrAEs can manifest in different organs within the body, with pulmonary toxicity commonly referred to as immune checkpoint inhibitor-related pneumonitis (CIP). The CIP incidence remains high and is anticipated to rise further as the therapeutic indications for ICIs expand to encompass a wider range of malignancies. The diagnosis and treatment of CIP is difficult due to the large individual differences in its pathogenesis and severity, and severe CIP often leads to a poor prognosis for patients. This review summarizes the current state of clinical research on the incidence, risk factors, predictive biomarkers, diagnosis, and treatment for CIP, and we address future directions for the prevention and accurate prediction of CIP.

Diagnostic and Prognostic Value of Peripheral Neutrophil CD64 Index in Elderly Patients with Community-Acquired Pneumonia.

Community-acquired pneumonia (CAP) is a leading cause of hospitalization and mortality in the elderly. The peripheral blood neutrophil CD64 (nCD64) index is increasingly recognized for its association with poor pneumonia prognosis. A comprehensive investigation involving 128 elderly patients diagnosed with CAP, including 96 with non-severe CAP and 32 with severe CAP, from January 2020 to January 2021 was performed. The nCD64 index, CD4+, CD8+, C-reactive protein (CRP), white blood cell (WBC) count, procalcitonin (PCT), neutrophil (NEUT), and B lymphocyte count were determined using flow cytometry. Our findings reveal that patients with severe CAP exhibited significantly higher levels of nCD64 index, NEUT, WBC, CRP, and PCT. Intriguingly, lower CRP, nCD64 index, CURB-65 score, and PCT were associated with a higher survival rate. Notably, the nCD64 index demonstrated remarkable predictive efficiency for 28-d survival in CAP patients [area under the curve (AUC) = 0.907], surpassing other markers and even showing enhanced predictive power when combined with the CURB-65 score (AUC = 0.905). Furthermore, a negative association was observed between the nCD64 index and both CD4+, CD4+/CD8+ ratios, and B lymphocytes, highlighting its potential role in immune dysregulation. These findings underscore the critical importance of the nCD64 index in the early diagnosis, risk stratification, and prognostic evaluation of infections and immune responses in elderly CAP patients.

Case report of parapneumonic effusion and mesothelial subdiaphragmatic cyst: relationship or coincidence?

Parapneumonic pleural effusions are common in patients with pneumonia. When colonized by pathogenic bacteria or other microorganisms, these effusions can progress to empyema. Additionally, empyema formation may result in extension of the infection into the infradiaphragmatic region, further complicating the clinical scenario. Many subphrenic collections are found to be mesothelial cysts, which are congenital in origin. However, data regarding the potential association between mesothelial diaphragmatic cysts and parapneumonic effusions are limited. We herein describe a toddler with pneumonia complicated by parapneumonic effusion and a lung abscess with a subphrenic collection. After abscess drainage and a full course of antibiotics, imaging revealed clear lung parenchyma with an interval resolution of the effusion and a persistent unchanged subphrenic collection that was confirmed to be mesothelial diaphragmatic cyst. This case highlights the fact that not every subphrenic collection associated with parapneumonic effusion is a communicated collection formed by seeding. Such a collection can instead be an incidental cyst, which is congenital in origin and known as a mesothelial diaphragmatic cyst. A diaphragmatic mesothelial cyst is an uncommon benign congenital cyst that is unrelated to an adjacent parapneumonic effusion. It is usually incidental and can be monitored without invasive intervention.

Study on red blood cell distribution width in children with severe Mycoplasma pneumoniae pneumonia.

Background: This study aimed to investigate the clinical value of the red blood cell distribution width (RDW) in severe Mycoplasma pneumoniae pneumonia (MPP). Methods: A total of 185 children with diagnosed severe MPP were included. The patients' case records and laboratory examination data were analyzed retrospectively. The children were grouped into quartiles based on RDW. Results: Univariate analysis revealed that RDW was significantly correlated with the Pediatric Risk of Mortality (PRISM) III score, Sepsis-Related Organ Failure Assessment score, incidence of invasive intubation and 30-day in-hospital mortality. After adjustment for the severity of illness, multivariate analysis revealed that the PRISM III score and RDW were factors independently associated with 30-day in-hospital mortality. Conclusion: This study revealed that RDW could be correlated with the long-term prognosis and severity of severe MPP.

A clinically actionable and explainable real-time risk assessment framework for stroke-associated pneumonia.

The current medical practice is more responsive rather than proactive, despite the widely recognized value of early disease detection, including improving the quality of care and reducing medical costs. One of the cornerstones of early disease detection is clinically actionable predictions, where predictions are expected to be accurate, stable, real-time and interpretable. As an example, we used stroke-associated pneumonia (SAP), setting up a transformer-encoder-based model that analyzes highly heterogeneous electronic health records in real-time. The model was proven accurate and stable on an independent test set. In addition, it issued at least one warning for 98.6 % of SAP patients, and on average, its alerts were ahead of physician diagnoses by 2.71 days. We applied Integrated Gradient to glean the model's reasoning process. Supplementing the risk scores, the model highlighted critical historical events on patients' trajectories, which were shown to have high clinical relevance.

Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data.

PURPOSE: Although immune checkpoint inhibitors (ICIs) have improved outcomes in certain patients with cancer, they can also cause life-threatening immunotoxicities. Predicting immunotoxicity risks alongside response could provide a personalized risk-benefit profile, inform therapeutic decision making, and improve clinical trial cohort selection. We aimed to build a machine learning (ML) framework using routine electronic health record (EHR) data to predict hepatitis, colitis, pneumonitis, and 1-year overall survival. METHODS: Real-world EHR data of more than 2,200 patients treated with ICI through December 31, 2018, were used to develop predictive models. Using a prediction time point of ICI initiation, a 1-year prediction time window was applied to create binary labels for the four outcomes for each patient. Feature engineering involved aggregating laboratory measurements over appropriate time windows (60-365 days). Patients were randomly partitioned into training (80%) and test (20%) sets. Random forest classifiers were developed using a rigorous model development framework. RESULTS: The patient cohort had a median age of 63 years and was 61.8% male. Patients predominantly had melanoma (37.8%), lung cancer (27.3%), or genitourinary cancer (16.4%). They were treated with PD-1 (60.4%), PD-L1 (9.0%), and CTLA-4 (19.7%) ICIs. Our models demonstrate reasonably strong performance, with AUCs of 0.739, 0.729, 0.755, and 0.752 for the pneumonitis, hepatitis, colitis, and 1-year overall survival models, respectively. Each model relies on an outcome-specific feature set, though some features are shared among models. CONCLUSION: To our knowledge, this is the first ML solution that assesses individual ICI risk-benefit profiles based predominantly on routine structured EHR data. As such, use of our ML solution will not require additional data collection or documentation in the clinic.

Diagnostic Stewardship in Community-Acquired Pneumonia With Syndromic Molecular Testing: A Randomized Clinical Trial.

Importance: Lower respiratory tract (LRT) infections, including community-acquired pneumonia (CAP), are a leading cause of hospital admissions and mortality. Molecular tests have the potential to optimize treatment decisions and management of CAP, but limited evidence exists to support their routine use. Objective: To determine whether the judicious use of a syndromic polymerase chain reaction (PCR)-based panel for rapid testing of CAP in the emergency department (ED) leads to faster, more accurate microbiological test result-based treatment. Design, Setting, and Participants: This parallel-arm, single-blinded, single-center, randomized clinical superiority trial was conducted between September 25, 2020, and June 21, 2022, in the ED of Haukeland University Hospital, a large tertiary care hospital in Bergen, Norway. Adult patients who presented to the ED with suspected CAP were recruited. Participants were randomized 1:1 to either the intervention arm or standard-of-care arm. The primary outcomes were analyzed according to the intention-to-treat principle. Intervention: Patients randomized to the intervention arm received rapid syndromic PCR testing (BioFire FilmArray Pneumonia plus Panel; bioMérieux) of LRT samples and standard of care. Patients randomized to the standard-of-care arm received standard microbiological diagnostics alone. Main Outcomes and Measures: The 2 primary outcomes were the provision of pathogen-directed treatment based on a microbiological test result and the time to provision of pathogen-directed treatment (within 48 hours after randomization). Results: There were 374 patients (221 males [59.1%]; median (IQR) age, 72 [60-79] years) included in the trial, with 187 in each treatment arm. Analysis of primary outcomes showed that 66 patients (35.3%) in the intervention arm and 25 (13.4%) in the standard-of-care arm received pathogen-directed treatment, corresponding to a reduction in absolute risk of 21.9 (95% CI, 13.5-30.3) percentage points and an odds ratio for the intervention arm of 3.53 (95% CI, 2.13-6.02; P < .001). The median (IQR) time to provision of pathogen-directed treatment within 48 hours was 34.5 (31.6-37.3) hours in the intervention arm and 43.8 (42.0-45.6) hours in the standard-of-care arm (mean difference, -9.4 hours; 95% CI, -12.7 to -6.0 hours; P < .001). The corresponding hazard ratio for intervention compared with standard of care was 3.08 (95% CI, 1.95-4.89). Findings remained significant after adjustment for season. Conclusions and Relevance: Results of this randomized clinical trial indicated that routine deployment of PCR testing for LRT pathogens led to faster and more targeted microbial treatment for patients with suspected CAP. Rapid molecular testing could complement or replace selected standard, time-consuming, laboratory-based diagnostics. Trial Registration: ClinicalTrials.gov Identifier: NCT04660084.

Hypoalbuminemia as predictor of thrombotic events in patients with community-acquired pneumonia.

BACKGROUND: Hypoalbuminemia complicates acute diseases and infections and is associated with a worst prognosis. The aim is to evaluate whether hypoalbuminemia is associated with higher incidence and risk of thrombotic events in community-acquired pneumonia. METHODS: We retrospectively collected data from a prospective study investigating the incidence of thrombotic events in community-acquired pneumonia hospitalized patients from 2011 to 2016 at University-Hospital Policlinico Umberto I. Baseline characteristics and outcomes were collected. Incidence of outcomes were calculated. Kaplan-Meier curves were created, Cox model used to identify predictors for the outcomes, and competing risk analysis performed. RESULTS: From a total of 231 patients, 130 (56.3%) and 101 (43.7%) had or not hypoalbuminemia. Age, proportion of female, BMI, major comorbidities, and severity of pneumonia were similar between two subgroups. A less proportion of patients with hypoalbuminemia received antithrombotic and statin therapy. Median hospital stay was 11 days in both subgroups. Patients with hypoalbuminemia had higher D-dimer and high- sensitivity C-reactive-protein values with an inverse relation between albumin values and these markers. Incidence of thrombotic events was 26 and 11 per 1000 patient-days in patient with and without hypoalbuminemia. At Cox model, hypoalbuminemia was associated with thrombotic events development in univariable (hazard ratio; 2.67, 95% confidence intervals, 1.30-5.40) and multivariable (hazard ratio 3.19; 95% confidence intervals, 1.48-6.89) analysis. CONCLUSIONS: More than a half of patients with community acquired pneumonia had hypoalbuminemia that is associated with a doubled incidence and a three-fold increased risk of thrombotic events. The inverse relation between baseline albumin and D-dimer values confirms this association.

The predictive value of anticholinergic drug exposure and the outcome of pneumonia: a Danish database study.

INTRODUCTION: Older adults are susceptible to anticholinergic effects. Dysphagia and pneumonia are associated with anticholinergic usage, though a definitive causative relationship has not been established. There is no effective way to predict the prognosis of older adults with pneumonia; therefore, this study investigates the predictive value of anticholinergic burden. METHODS: Patients aged 65 years and above admitted for community-acquired pneumonia from 2011 to 2018 in Denmark were included through Danish registries. We calculated anticholinergic drug exposure using the CRIDECO Anticholinergic Load Scale (CALS). The primary outcome was in-hospital mortality, and other outcomes included intensive care unit admission, ventilator usage, length of stay, 30-day/90-day/1-year mortality, institutionalisation, home care utilisation and readmission. RESULTS: 186,735 patients were included in the in-hospital outcome analyses, 165,181 in the readmission analysis, 150,791 in the institutionalisation analysis, and 95,197 and 73,461 patients in the home care analysis at follow-up. Higher CALS score was associated with higher in-hospital mortality, with a mean risk increasing from 9.9% (CALS 0) to 16.4% (CALS >10), though the risk plateaued above a CALS score of 8. A higher CALS score was also associated with greater mortality after discharge, more home health care, more institutionalizations and higher readmission rates. CONCLUSIONS: High anticholinergic burden levels were associated with poor patient outcomes including short-/long-term mortality, dependence and readmission. It may be useful to calculate the CALS score on admission of older patients with pneumonia to predict their prognosis. This also highlights the importance of avoiding the use of drugs with a high anticholinergic burden in older patients.

Clinical application of bronchoalveolar lavage fluid metagenomics next-generation sequencing in cancer patients with severe pneumonia.

OBJECTIVE: Metagenomic next-generation sequencing (mNGS), as an emerging technique for pathogen detection, has been widely used in clinic. However, reports on the application of mNGS in cancer patients with severe pneumonia remain limited. This study aims to evaluate the diagnostic performance of bronchoalveolar lavage fluid (BALF) mNGS in cancer patients complicated with severe pneumonia. METHODS: A total of 62 cancer patients with severe pneumonia simultaneously received culture and mNGS of BALF were enrolled in this study. We systematically analyzed the diagnostic significance of BALF mNGS. Subsequently, optimization of anti-infective therapy based on the distribution of pathogens obtained from BALF mNGS was also assessed. RESULTS: For bacteria and fungi, the positive detection rate of mNGS was significantly higher than culture method (91.94% versus 51.61%, P < 0.001), especially for poly-microbial infections (70.97% versus 12.90%, P < 0.001). Compared with the culture method, mNGS exhibited a diagnostic sensitivity of 100% and a specificity of 16.67%, with the positive predictive value (PPV) and negative predictive value (NPV) being 56.14% and 100%, respectively. The agreement rate between these two methods was 59.68%, whereas kappa consensus analysis indicated a poor concordance (kappa = 0.171). After receipt of BALF mNGS results, anti-infective treatment strategies in 39 out of 62 cases (62.90%) were optimized. Moreover, anti-tumor therapy was a high-risk factor for mixed infections (87.18% versus 65.22%, P = 0.04). CONCLUSIONS: The present study showed that cancer patients with severe pneumonia, especially those received anti-tumor therapy, were more likely to have poly-microbial infections. BALF mNGS can provide a rapid and comprehensive pathogen distribution of pulmonary infection, making it a promising technique in clinical practice, especially for optimizing therapeutic strategies for cancer patients.

Compliance with guideline recommendations in the management of patients with hospitalized community-acquired pneumonia.

Compliance with guidelines in the empirical treatment for community-acquired pneumonia (CAP) is very important to increase treatment success and reduce mortality. This study aimed to determine compliance with guideline recommendations for CAP and define the role of the clinical pharmacist (CP). Patients diagnosed with CAP were evaluated retrospectively between January 2018 and January 2020 and prospectively between February 2020 and February 2021. Compliance with guidelines was evaluated according to the local, national (Turkish Thoracic Society), and international (American Thoracic Society and Infectious Disease Society of America, European Society of Clinical Microbiology and Infectious Diseases) guidelines. A total of 751 patients (423 in the retrospective and 328 in the prospective period) were included. It was determined that the 30-day mortality and length of stay were higher in patients who were not treated according to the guidelines. The compliance for empirical treatments was 16.3%-59.1% and 7.8%-30.1% in retrospective and prospective periods, respectively. During the prospective period, a total of 603 recommendations were made by CP, and 578 (95.9%) were accepted and implemented. In the prospective period, treatment duration was shortened, inappropriate fluoroquinolone use was decreased, the switch to oral treatment was increased, and the number of potential drug-drug interactions was decreased (p < 0.001). Compliance with guidelines is essential to be improved to reduce mortality, shorten the length of stay, determine the appropriate antimicrobial duration, and reduce the use of fluoroquinolones and broad-spectrum antibiotics unless necessary. CP intervention contributes to the rational selection of antimicrobials, limiting drug-drug interactions, avoiding toxicities, and compliance with guidelines.