Artificial intelligence (AI) is rapidly gaining interest in medicine, including pathological assessments for personalized medicine. In this issue of Cancer Cell, Wagner et&#xa0;al. demonstrate superior accuracy of transformer-based deep learning in predicting biomarker status in CRC. The work has implications for increased efficiency and accuracy in clinical diagnostics guiding treatment decisions in precision oncology.

Deep learning transforms colorectal cancer biomarker prediction from histopathology images.

Cancer Cell

A strategy of administering a transfusion only when the hemoglobin level falls below 7 or 8 g per deciliter has been widely adopted. However, patients with acute myocardial infarction may benefit from a higher hemoglobin level.

In this phase 3, interventional trial, we randomly assigned patients with myocardial infarction and a hemoglobin level of less than 10 g per deciliter to a restrictive transfusion strategy (hemoglobin cutoff for transfusion, 7 or 8 g per deciliter) or a liberal transfusion strategy (hemoglobin cutoff, &lt;10 g per deciliter). The primary outcome was a composite of myocardial infarction or death at 30 days.

A total of 3504 patients were included in the primary analysis. The mean (&#xb1;SD) number of red-cell units that were transfused was 0.7&#xb1;1.6 in the restrictive-strategy group and 2.5&#xb1;2.3 in the liberal-strategy group. The mean hemoglobin level was 1.3 to 1.6 g per deciliter lower in the restrictive-strategy group than in the liberal-strategy group on days 1 to 3 after randomization. A primary-outcome event occurred in 295 of 1749 patients (16.9%) in the restrictive-strategy group and in 255 of 1755 patients (14.5%) in the liberal-strategy group (risk ratio modeled with multiple imputation for incomplete follow-up, 1.15; 95% confidence interval [CI], 0.99 to 1.34; P&#x2009;=&#x2009;0.07). Death occurred in 9.9% of the patients with the restrictive strategy&#x2009;and in 8.3% of the patients with the liberal strategy (risk ratio, 1.19; 95% CI, 0.96 to 1.47); myocardial infarction occurred in 8.5% and 7.2% of the patients, respectively (risk ratio, 1.19; 95% CI, 0.94 to 1.49).

In patients with acute myocardial infarction and anemia, a liberal transfusion strategy did not significantly reduce the risk of recurrent myocardial infarction or death at 30 days. However,&#xa0;potential&#xa0;harms&#xa0;of a&#xa0;restrictive transfusion strategy&#xa0;cannot be excluded. (Funded by the National Heart, Lung, and Blood Institute and others; MINT ClinicalTrials.gov number, NCT02981407.).

Restrictive or Liberal Transfusion Strategy in Myocardial Infarction and Anemia.

New England Journal of Medicine

The influence of restrictive fluid resuscitation and the early administration of vasopressors on the clinical outcomes in patients with septic shock is not fully understood. The purpose of this study was to evaluate the effects of restrictive fluid management on mortality and organ dysfunction in patients with septic shock.

This study included consecutive patients with septic shock in need of fluid resuscitation. Based on the fluid management provided in the initial resuscitation phase, a comparison was made between a restrictive group and a standard fluid management group. The primary outcome was in-hospital death, while secondary outcomes included organ dysfunction and other adverse events.

A total of 238 patients were included in this study. Restrictive fluid management was administered to 59.2% of patients, while 40.8% received standard fluid management. Restrictive resuscitation was associated with a lower in-hospital mortality rate (24.8% vs. 52.6%), as well as a shorter median ICU stay (8.0 vs. 11.0 days). The restrictive strategy was associated with a significantly lower prevalence of new-onset acute kidney injury (25.5% vs. 51.5%) and a decrease in the incidence of renal replacement therapy (20.6% vs. 40.2%). The standard group had a higher risk of the need for mechanical ventilation and a significantly lower median number of days without a ventilator than the restrictive group. The median duration of vasopressor-free days in the restrictive group was significantly longer than that in the standard group (25.0 vs. 18.0). The administration rate of inotropes in the restrictive group was significantly lower than that in the standard group. A multivariate logistic regression model showed that restrictive fluid management (OR 0.312; 95% CI 0.098-0.994) and vasopressor-free days (OR 0.807; 95% CI 0.765-0.851) protect against in-hospital death, whereas APACHE II scores (OR 1.121; 95% CI 1.018-1.234) were independent risk factors for in-hospital death.

Restrictive fluid resuscitation and early vasopressor protocol in patients with septic shock are associated with better outcomes, indicating that this regimen is feasible and safe.

Restrictive fluid resuscitation in septic shock patients has lower mortality and organ dysfunction rates than standard therapy.

Shock

Urinary tract infections are the most common bacterial infections worldwide. Infections can range from mild, recurrent (rUTI) to complicated (cUTIs), and are predominantly caused by uropathogenic Escherichia coli (UPEC). Antibiotic therapy is important to tackle infection; however, with the continued emergence of antibiotic resistance there is an urgent need to monitor the use of effective antibiotics through better stewardship measures. Currently, clinical diagnosis of UTIs relies on empiric methods supported by laboratory testing including cellular analysis (of both human and bacterial cells), dipstick analysis and phenotypic culture. Therefore, development of novel, sensitive and specific diagnostics is an important means to rationalise antibiotic therapy in patients. This review discusses the current diagnostic landscape and highlights promising novel diagnostic technologies in development that could aid in treatment and management of antibiotic-resistant UTIs.

Urinary tract infections: a review of the current diagnostics landscape.

Journal of Medical Microbiology

Chronic cough (CC; &#x2265;8 weeks in duration) is a common and burdensome feature of respiratory diseases. The understanding of cough has progressed significantly in recent years, albeit largely in refractory (unexplained) chronic cough (RCC) in the absence of other respiratory conditions. The prevalence of CC in respiratory diseases is poorly described, but estimates have been reported: asthma (8-58%), chronic obstructive pulmonary disease (COPD; 10-74%), bronchiectasis (82-98%), interstitial lung disease (ILD; 50-89%) and sarcoidosis (3-64%). CC in respiratory conditions generally predicts impaired health status and more severe disease. It is associated with increased symptom burden and disease severity in asthma, COPD, bronchiectasis and ILD, higher exacerbation frequency in asthma and bronchiectasis, and increased mortality and lung transplantation in idiopathic pulmonary fibrosis (IPF). Physiologically, heightened cough reflex sensitivity (CRS) has been reported and postulated to be mechanistic in isolated RCC. Cough reflex hypersensitivity (CRH) has also been reported in asthma, COPD, bronchiectasis, ILD and sarcoidosis. Unlike recent advances in isolated RCC, there are limited studies and understanding of central cough neuropathways in other respiratory conditions. Of note, dysfunctional central voluntary cough suppression neuropathways and physiology were observed in isolation in RCC; cough suppression is preserved in COPD. Understanding in the mechanism of RCC cannot be simply extrapolated to other respiratory conditions. The restricted understanding of cough mechanisms in these conditions has limited cough-specific therapeutic options in this context. There is currently an unmet need to expand our understanding of cough in chronic respiratory conditions, both in order to improve the quality of life of patients, and to improve knowledge of cough in general. This review aims to describe the prevalence, impact, pathophysiology and management of CC in asthma, COPD, bronchiectasis, ILD and sarcoidosis.

Cough in chronic lung disease: a state of the art review.

Journal of Thoracic Disease

Chronic kidney disease (CKD) is a common occurrence in patients with diabetes mellitus (DM), occurring in approximately 40% of cases. DM is also an important risk factor for cardiovascular disease (CVD), but CKD is an important mediator of this risk. Multiple CVD outcomes trials have revealed a greater risk for CVD events in patients with diabetes with CKD versus those without. Thus, reducing the risk of CKD in diabetes should result in improved CVD outcomes. To date, of blood pressure (BP) control, glycemic control, and inhibition of the renin-angiotensin system (RASI), glycemic control appears to have the best evidence for preventing CKD development. In established CKD, especially with albuminuria, RASI slows the progression of CKD. More recently, sodium glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide receptor agonists (GLP1RA) have revolutionized the care of patients with diabetes with and without CKD. SGLT2i and GLP1RA have proven to reduce mortality, heart failure (HF) hospitalizations, and worsening CKD in patients with diabetes with and without existing CKD. The future of limiting CVD in diabetes and CKD is promising, and more evidence is forthcoming regarding combinations of evidence-based therapies to further minimize CVD events.

Deep learning transforms colorectal cancer biomarker prediction from histopathology images.

Full text links

For the best experience, use the Read mobile app

Deep learning transforms colorectal cancer biomarker prediction from histopathology images.

Full text links

Trending Papers

For the best experience, use the Read mobile app