scholarly journals Macrolide combination therapy for patients hospitalised with community-acquired pneumonia? An individualised approach supported by machine learning

2019 ◽  
Vol 54 (6) ◽  
pp. 1900824 ◽  
Author(s):  
Rainer König ◽  
Xueqi Cao ◽  
Marcus Oswald ◽  
Christina Forstner ◽  
Gernot Rohde ◽  
...  
Keyword(s):  
Machine Learning ◽  
Combination Therapy ◽  
Decision Tree ◽  
Controlled Trial ◽  
Community Acquired Pneumonia ◽  
Moderate Severity ◽  
Respiratory Secretion ◽  
Hospitalised Patients ◽  
Atypical Pathogens ◽  
Leukocyte Counts

BackgroundThe role of macrolide/β-lactam combination therapy in community-acquired pneumonia (CAP) of moderate severity is a matter of debate. Macrolides expand the coverage to atypical pathogens and attenuate pulmonary inflammation, but have been associated with cardiovascular toxicity and drug interactions. We developed a decision tree based on aetiological and clinical parameters, which are available ex ante to support a personalised decision for or against macrolides for the best clinical outcome of the individual patient.MethodsWe employed machine learning in a cross-validation scheme based on a well-balanced selection of 4898 patients after propensity score matching to data available on admission of 6440 hospitalised patients with moderate severity (non-intensive care unit patients) from the observational, prospective, multinational CAPNETZ study. We aimed to improve the primary outcome of 180-day survival.ResultsWe found a simple decision tree of patient characteristics comprising chronic cardiovascular and chronic respiratory comorbidities as well as leukocyte counts in the respiratory secretion at enrolment. Specifically, we found that patients without cardiovascular or patients with respiratory comorbidities and high leukocyte counts in the respiratory secretion benefit from macrolide treatment. Patients identified to be treated in compliance with our treatment suggestion had a lower mortality of 27% (OR 1.83, 95% CI 1.48–2.27; p<0.001) compared to the observed standard of care.ConclusionStratifying macrolide treatment in patients following a simple treatment rule may lead to considerably reduced mortality in CAP. A future randomised controlled trial confirming our result is necessary before implementing this rule into the clinical routine.

scholarly journals Pathogen- and antibiotic-specific effects of prednisone in community-acquired pneumonia

2016 ◽  
Vol 48 (4) ◽  
pp. 1150-1159 ◽  
Author(s):  
Sebastian A. Wirz ◽  
Claudine A. Blum ◽  
Philipp Schuetz ◽  
Werner C. Albrich ◽  
Christoph Noppen ◽  
...  
Keyword(s):  
Multiple Testing ◽  
Pneumococcal Pneumonia ◽  
Controlled Trial ◽  
Community Acquired Pneumonia ◽  
Double Blind ◽  
Interaction Terms ◽  
Specific Effects ◽  
Hospitalised Patients ◽  
Secondary Endpoints ◽  
Study Population

In a double-blind, randomised, placebo-controlled trial of hospitalised patients with community-acquired pneumonia (CAP), we demonstrated shorter time to clinical stability (TTCS) with adjunct corticosteroid therapy compared with placebo.We did a pre-planned, exploratory analysis of any association between microbiological diagnosis, antibiotic treatment and procalcitonin level and effect of prednisone on TTCS, mortality, and CAP complications (n=726 participants, enrolled between December 2009 and May 2014). Multiplex viral real time PCR was systematically performed in nasopharyngeal swabs beginning November 2011 (n=489). Other investigations and treatments were at the discretion of the physician. Effect modification was tested with inclusion of interaction terms in the statistical models.Reduced TTCS with prednisone was seen in all microbiological, antibiotic, procalcitonin and afebrile patient subgroups. We found evidence for a different prednisone response in patients with pneumococcal pneumonia in whom intravenous antibiotic duration was not shorter (interaction p=0.01) with prednisone, as was observed in the remaining study population. In patients without macrolide treatment, rehospitalisations were not lower with prednisone (interaction p=0.04). After adjustment for multiple testing, these subgroup effects were no longer significant.Prednisone was associated with shorter TTCS independent of CAP aetiology. In pneumococcal pneumonia, prednisone effects on secondary endpoints may be less favourable.

Zinc Depletion Increases Readmission in Older Patients: An Example of Interactions Between Nutrition and Disease

2017 ◽  
Vol 87 (1-2) ◽  
pp. 10-16 ◽  
Author(s):  
Salah Gariballa ◽  
Awad Alessa
Keyword(s):  
Nutritional Status ◽  
Inductively Coupled Plasma ◽  
Older Patients ◽  
Controlled Trial ◽  
Prognostic Indicators ◽  
Hospitalised Patients ◽  
Increased Risk ◽  
Co Morbidity ◽  
The Impact ◽  
Poor Nutrition

Abstract. Background: ill health may lead to poor nutrition and poor nutrition to ill health, so identifying priorities for management still remains a challenge. The aim of this report is to present data on the impact of plasma zinc (Zn) depletion on important health outcomes after adjusting for other poor prognostic indicators in hospitalised patients. Methods: Hospitalised acutely ill older patients who were part of a large randomised controlled trial had their nutritional status assessed using anthropometric, hematological and biochemical data. Plasma Zn concentrations were measured at baseline, 6 weeks and at 6 months using inductively- coupled plasma spectroscopy method. Other clinical outcome measures of health were also measured. Results: A total of 345 patients assessed at baseline, 133 at 6 weeks and 163 at 6 months. At baseline 254 (74%) patients had a plasma Zn concentration below 10.71 μmol/L indicating biochemical depletion. The figures at 6 weeks and 6 months were 86 (65%) and 114 (70%) patients respectively. After adjusting for age, co-morbidity, nutritional status and tissue inflammation measured using CRP, only muscle mass and serum albumin showed significant and independent effects on plasma Zn concentrations. The risk of non-elective readmission in the 6-months follow up period was significantly lower in patients with normal Zn concentrations compared with those diagnosed with Zn depletion (adjusted hazard ratio 0.62 (95% CI: 0.38 to 0.99), p = 0.047. Conclusions: Zn depletion is common and associated with increased risk of readmission in acutely-ill older patients, however, the influence of underlying comorbidity on these results can not excluded.

Sign in / Sign up

Export Citation Format

Share Document