Predicting late anemia in critical illness

Milbrandt, Eric B; Clermont, Gilles; Martinez, J; Kersten, Alex; Rahim, M T; Angus, Derek C.

doi:10.1186/cc4847

Cited by 18 publications

(8 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The day 1 prevalence speaks to the significant systemic derangement that likely has already occurred in CAP patients at the time of presentation, long before repeated blood draws or the dilutional effects of intravenous fluids could explain low hemoglobin levels. The precipitous decline in hemoglobin values that occurred over the first few days of hospitalization is consistent with that seen in the ICU, where hemoglobin values may decline by >0.5 g/dL/day in non-bleeding patients [ 33 , 34 ]. These changes are believed to be due not only to dilutional effects of fluids and frequent blood draws [ 6 ], but also to other sources of blood loss (gastric stress bleeding, surgical procedures), effects of inflammatory cytokines, inadequate red cell production, and excessive red cell destruction [ 35 ].…”

Section: Discussionsupporting

confidence: 60%

The prevalence of anemia and its association with 90-day mortality in hospitalized community-acquired pneumonia

et al. 2010

View full text Add to dashboard Cite

BackgroundThe prevalence of anemia in the intensive care unit is well-described. Less is known, however, of the prevalence of anemia in hospitalized patients with lesser illness severity or without organ dysfunction. Community-acquired pneumonia (CAP) is one of the most frequent reasons for hospitalization in the United States (US), affecting both healthy patients and those with comorbid illness, and is typically not associated with acute blood loss. Our objective was to examine the development and progression of anemia and its association with 90d mortality in 1893 subjects with CAP presenting to the emergency departments of 28 US academic and community hospitals.MethodsWe utilized hemoglobin values obtained for clinical purposes, classifying subjects into categories consisting of no anemia (hemoglobin >13 g/dL), at least borderline (≤ 13 g/dL), at least mild (≤ 12 g/dL), at least moderate (≤ 10 g/dL), and severe (≤ 8 g/dL) anemia. We stratified our results by gender, comorbidity, ICU admission, and development of severe sepsis. We used multivariable logistic regression to determine factors independently associated with the development of moderate to severe anemia and to examine the relationship between anemia and 90d mortality.ResultsA total of 8240 daily hemoglobin values were measured in 1893 subjects. Mean (SD) number of hemoglobin values per patient was 4.4 (4.0). One in three subjects (33.9%) had at least mild anemia at presentation, 3 in 5 (62.1%) were anemic at some point during their hospital stay, and 1 in 2 (54.5%) survivors were discharged from the hospital anemic. Anemia increased with illness severity and was more common in those with comorbid illnesses, female gender, and poor outcomes. Yet, even among men and in those with no comorbidity or only mild illness, anemia during hospitalization was common (~55% of subjects). When anemia was moderate to severe (≤ 10 g/dL), its development was independently associated with increased 90d mortality, even among hospital survivors.ConclusionsAnemia was common in hospitalized CAP and independently associated with 90d mortality when hemoglobin values were 10 g/dL or less. Whether prevention or treatment of CAP-associated anemia would improve clinical outcomes remains to be seen.

show abstract

Section: Discussionsupporting

confidence: 60%

The prevalence of anemia and its association with 90-day mortality in hospitalized community-acquired pneumonia

et al. 2010

View full text Add to dashboard Cite

show abstract

“…A LR model can predict the outcome variable, such as the disease state ( i.e. sick or healthy) [24] , by the new predictor inputs. The LASSO (Least Absolute Shrinkage and Selection Operator) algorithm [25] is a -norm regularized logistic regression, which is extensively used for feature selection.…”

Section: Resultsmentioning

confidence: 99%

Which Biomarkers Reveal Neonatal Sepsis?

et al. 2013

View full text Add to dashboard Cite

We address the identification of optimal biomarkers for the rapid diagnosis of neonatal sepsis. We employ both canonical correlation analysis (CCA) and sparse support vector machine (SSVM) classifiers to select the best subset of biomarkers from a large hematological data set collected from infants with suspected sepsis from Yale-New Haven Hospital's Neonatal Intensive Care Unit (NICU). CCA is used to select sets of biomarkers of increasing size that are most highly correlated with infection. The effectiveness of these biomarkers is then validated by constructing a sparse support vector machine diagnostic classifier. We find that the following set of five biomarkers capture the essential diagnostic information (in order of importance): Bands, Platelets, neutrophil CD64, White Blood Cells, and Segs. Further, the diagnostic performance of the optimal set of biomarkers is significantly higher than that of isolated individual biomarkers. These results suggest an enhanced sepsis scoring system for neonatal sepsis that includes these five biomarkers. We demonstrate the robustness of our analysis by comparing CCA with the Forward Selection method and SSVM with LASSO Logistic Regression.

show abstract

“…There were also studies that aimed to reduce unnecessary laboratory tests to streamline the process and reduce the burden on patients [51,52]. Predicted clinical events also included acute traumatic coagulopathy [53], delirium [54], advanced anemia [55], and fluid resuscitation therapy [56].…”

Section: Other Predictions and Evaluationsmentioning

confidence: 99%

State of the Art of Machine Learning–Enabled Clinical Decision Support in Intensive Care Units: Literature Review

Hong¹,

Liu²,

Gao³

et al. 2022

JMIR Med Inform

View full text Add to dashboard Cite

Background Modern clinical care in intensive care units is full of rich data, and machine learning has great potential to support clinical decision-making. The development of intelligent machine learning–based clinical decision support systems is facing great opportunities and challenges. Clinical decision support systems may directly help clinicians accurately diagnose, predict outcomes, identify risk events, or decide treatments at the point of care. Objective We aimed to review the research and application of machine learning–enabled clinical decision support studies in intensive care units to help clinicians, researchers, developers, and policy makers better understand the advantages and limitations of machine learning–supported diagnosis, outcome prediction, risk event identification, and intensive care unit point-of-care recommendations. Methods We searched papers published in the PubMed database between January 1980 and October 2020. We defined selection criteria to identify papers that focused on machine learning–enabled clinical decision support studies in intensive care units and reviewed the following aspects: research topics, study cohorts, machine learning models, analysis variables, and evaluation metrics. Results A total of 643 papers were collected, and using our selection criteria, 97 studies were found. Studies were categorized into 4 topics—monitoring, detection, and diagnosis (13/97, 13.4%), early identification of clinical events (32/97, 33.0%), outcome prediction and prognosis assessment (46/97, 47.6%), and treatment decision (6/97, 6.2%). Of the 97 papers, 82 (84.5%) studies used data from adult patients, 9 (9.3%) studies used data from pediatric patients, and 6 (6.2%) studies used data from neonates. We found that 65 (67.0%) studies used data from a single center, and 32 (33.0%) studies used a multicenter data set; 88 (90.7%) studies used supervised learning, 3 (3.1%) studies used unsupervised learning, and 6 (6.2%) studies used reinforcement learning. Clinical variable categories, starting with the most frequently used, were demographic (n=74), laboratory values (n=59), vital signs (n=55), scores (n=48), ventilation parameters (n=43), comorbidities (n=27), medications (n=18), outcome (n=14), fluid balance (n=13), nonmedicine therapy (n=10), symptoms (n=7), and medical history (n=4). The most frequently adopted evaluation metrics for clinical data modeling studies included area under the receiver operating characteristic curve (n=61), sensitivity (n=51), specificity (n=41), accuracy (n=29), and positive predictive value (n=23). Conclusions Early identification of clinical and outcome prediction and prognosis assessment contributed to approximately 80% of studies included in this review. Using new algorithms to solve intensive care unit clinical problems by developing reinforcement learning, active learning, and time-series analysis methods for clinical decision support will be greater development prospects in the future.

show abstract

Predicting late anemia in critical illness

Cited by 18 publications

References 10 publications

The prevalence of anemia and its association with 90-day mortality in hospitalized community-acquired pneumonia

The prevalence of anemia and its association with 90-day mortality in hospitalized community-acquired pneumonia

Which Biomarkers Reveal Neonatal Sepsis?

State of the Art of Machine Learning–Enabled Clinical Decision Support in Intensive Care Units: Literature Review

Contact Info

Product

Resources

About