Management and clinical outcomes for patients with gastrointestinal bleeding who decline transfusion.

BACKGROUND: The national blood shortage and growing patient population who decline blood transfusions have created the need for bloodless medicine initiatives. This case series describes the management of gastrointestinal bleed patients who declined allogeneic blood transfusion. Understanding the effectiveness of bloodless techniques may improve treatment for future patients while avoiding the risks and cost associated with transfusion. STUDY DESIGN AND METHODS: A retrospective chart review identified 30 inpatient encounters admitted between 2016 to 2022 for gastrointestinal hemorrhage who declined transfusion due to religious or personal reasons. Clinical characteristics and patient blood management methods utilized during hospitalization were analyzed. Hemoglobin concentrations and clinical outcomes are reported. RESULTS: The most common therapy was intravenous iron (n = 25, 83.3%), followed by erythropoietin (n = 18, 60.0%). Endoscopy was the most common procedure performed (n = 23, 76.7%), and surgical intervention was less common (n = 4, 13.3%). Pre-procedure hemoglobin was <6 g/dL in 7 patients, and <5 g/dL in 4 patients. The median nadir hemoglobin was 5.6 (IQR 4.5, 7.0) g/dL, which increased post-treatment to 7.3 (IQR 6.2, 8.4) g/dL upon discharge. One patient (3.3%) with a nadir Hb of 3.7 g/dL died during hospitalization from sepsis. Nine other patients with nadir Hb <5 g/dL survived hospitalization. CONCLUSIONS: Gastrointestinal bleed patients can be successfully managed with optimal bloodless medicine techniques. Even patients with a nadir Hb <5-6 g/dL can be stabilized with aggressive anemia treatment and may safely undergo anesthesia and endoscopy or surgery for diagnostic or therapeutic purposes. Methods used for treating bloodless medicine patients may be used to improve clinical care for all patients.

Black-box and surrogate optimization for tuning spiking neural models of striatum plasticity.

The basal ganglia (BG) is a brain structure that has long been proposed to play an essential role in action selection, and theoretical models of spiking neurons have tried to explain how the BG solves this problem. A recently proposed functional and biologically inspired network model of the striatum (an important nucleus of the BG) is based on spike-timing-dependent eligibility (STDE) and captured important experimental features of this nucleus. The model can recognize complex input patterns and consistently choose rewarded actions to respond to such sensory inputs. However, model tuning is challenging due to two main reasons. The first is the expert knowledge required, resulting in tedious and potentially biased trial-and-error procedures. The second is the computational cost of assessing model configurations (approximately 1.78 h per evaluation). This study addresses the model tuning problem through numerical optimization. Considering the cost of assessing solutions, the selected methods stand out due to their low requirements for solution evaluations and compatibility with high-performance computing. They are the SurrogateOpt solver of Matlab and the RBFOpt library, both based on radial basis function approximations, and DIRECT-GL, an enhanced version of the widespread black-box optimizer DIRECT. Besides, a parallel random search serves as a baseline reference of the outcome of opting for sophisticated methods. SurrogateOpt turns out to be the best option for tuning this kind of model. It outperforms, on average, the quality of the configuration found by an expert and works significantly faster and autonomously. RBFOpt and the random search share the second position, but their average results are below the option found by hand. Finally, DIRECT-GL follows this line becoming the worst-performing method.

Methods of Bloodless Care, Clinical Outcomes, and Costs for Adult Patients Who Decline Allogeneic Transfusions.

BACKGROUND: Providing bloodless medical care for patients who wish to avoid allogeneic transfusion can be challenging; however, previous studies have demonstrated favorable outcomes when appropriate methods are used. Here, we report one of the largest series of patients receiving bloodless care, along with the methods used to provide such care, and the resulting outcomes. METHODS: In a retrospective cohort study, 1111 adult inpatients (age ≥18 years) at a single institution who declined allogeneic transfusion for religious or personal reasons between June 2012 and June 2016 were included, and the patient blood management methods are described. Patient characteristics, laboratory data, and transfusion rates, as well as clinical outcomes (morbidity, mortality, and length of stay) were compared to all other patients in the hospital who received standard care, including transfusions if needed (n = 137,009). Medical and surgical patients were analyzed as subgroups. The primary outcome was composite morbidity (any morbid event: infectious, thrombotic, ischemic, renal, or respiratory). Secondary outcomes included individual morbid events, in-hospital mortality, length of stay, total hospital charges, and costs. RESULTS: The bloodless cohort had more females and a lower case mix index, but more preadmission comorbidities. Mean nadir hemoglobin during hospitalization was lower in the bloodless (9.7 ± 2.6 g/dL) compared to the standard care (10.1 ± 2.4 g/dL) group (P < .0001). Composite morbidity occurred in 14.4% vs 16.0% (P = .16) of the bloodless and standard care patients, respectively. Length of stay and in-hospital mortality were similar between the bloodless and standard care patients. After Bonferroni adjustment for multiple comparisons, hospital-acquired infection occurred less frequently in the bloodless compared to the standard care cohort (4.3% vs 8.3%) (P < .0001) in the medical patient subgroup, but not in the surgical subgroup. After propensity score adjustment in a multivariable model and adjustment for multiple comparisons, bloodless care was associated with less risk of hospital-acquired infection (OR, 0.56; 95% CI, 0.35-0.83; P = .0074) in the medical subgroup, but not in the surgical subgroup. Median total hospital charges (by 8.5%; P = .0017) and costs (by 8.7%; P = .0001) were lower in the bloodless compared to the standard care cohort, when all patients were included. CONCLUSIONS: Overall, adult patients receiving bloodless care had similar clinical outcomes compared to patients receiving standard care. Medical (but not surgical) bloodless patients may be at less risk for hospital-acquired infection compared to those receiving standard care. Bloodless care is cost-effective and should be considered as high-value practice.

On the Use of a Multimodal Optimizer for Fitting Neuron Models. Application to the Cerebellar Granule Cell.

This article extends a recent methodological workflow for creating realistic and computationally efficient neuron models whilst capturing essential aspects of single-neuron dynamics. We overcome the intrinsic limitations of the extant optimization methods by proposing an alternative optimization component based on multimodal algorithms. This approach can natively explore a diverse population of neuron model configurations. In contrast to methods that focus on a single global optimum, the multimodal method allows directly obtaining a set of promising solutions for a single but complex multi-feature objective function. The final sparse population of candidate solutions has to be analyzed and evaluated according to the biological plausibility and their objective to the target features by the expert. In order to illustrate the value of this approach, we base our proposal on the optimization of cerebellar granule cell (GrC) models that replicate the essential properties of the biological cell. Our results show the emerging variability of plausible sets of values that this type of neuron can adopt underlying complex spiking characteristics. Also, the set of selected cerebellar GrC models captured spiking dynamics closer to the reference model than the single model obtained with off-the-shelf parameter optimization algorithms used in our previous article. The method hereby proposed represents a valuable strategy for adjusting a varied population of realistic and simplified neuron models. It can be applied to other kinds of neuron models and biological contexts.

Assessing predictors of futility in patients receiving massive transfusions.

BACKGROUND: Massive transfusions are associated with a high mortality rate, but there is little evidence indicating when such efforts are futile. The purpose of this study was to identify clinical variables that could be used as futility indicators in massively transfused patients. METHODS: We retrospectively analyzed 138 adult surgical patients at our institution receiving a massive transfusion (2016-2019). Peak lactate and nadir pH within 24 h of massive transfusion initiation, along with other clinical variables, were assessed as predictors of the primary outcome, in-hospital mortality. RESULTS: The overall rate of in-hospital mortality among our patient population was 52.9% (n = 73). Increasing lactate and decreasing pH were associated with greater mortality among massively transfused patients. Mortality rates were ~2-fold higher for patients in the highest lactate category (≥10.0 mmol/L: 25 of 37; 67.6%) compared to the lowest category (0.0-4.9 mmol/L: 17 of 48; 35.4%) (p = .005), and ~2.5-fold higher for patients in the lowest pH category (<7.00: 8 of 9; 88.9%) compared to the highest category (≥7.40: 8 of 23; 34.7%) (p = .016). Increasing age was also associated with higher mortality (≥65 years: 24 of 33; 72.7%) when compared to younger patients (18-64 years: 49 of 105; 46.7%) (p = .010). CONCLUSIONS: Peak lactate ≥10.0 mmol/L, nadir pH <7.00, and age ≥65 years were significantly associated with higher rates of in-hospital mortality among massively transfused patients. Incorporating these clinical parameters into a futility index for massive transfusions will be useful in situations where blood products are scarce and/or mortality may be unavoidable.

Impact of body weight on hemoglobin increments in adult red blood cell transfusion.

BACKGROUND: Though weight is a major consideration when transfusing blood in pediatric patients, it is generally not considered when dosing transfusions in adults. We hypothesized that the change in hemoglobin (Hb) concentration is inversely proportional to body weight when transfusing red blood cells (RBC) in adults. METHODS: A total of 13,620 adult surgical patients at our institution were assessed in this retrospective cohort study (2009-2016). Patients were stratified based on total body weight (kg): 40-59.9 (16.6%), 60-79.9 (40.4%), 80-99.9 (28.8%), 100-119.9 (11.3%), and 120-139.9 (2.9%). The primary outcome was the change in Hb per RBC unit transfused. Subgroup analyses were performed after stratification by sex (male/female) and the total number of RBC units received (1/2/≥3 units). Multivariable models were used to assess the association between weight and change in Hb. RESULTS: As patients' body weight increased, there was a decrease in the mean change in Hb per RBC unit transfused (40-59.9 kg: 0.85 g/dL, 60-79.9 kg: 0.73 g/dL, 80-99.9 kg: 0.66 g/dL, 100-119.9 kg: 0.60 g/dL, 120-139.9 kg: 0.55 g/dL; p < .0001). This corresponded with a 35% difference in the change in Hb between the lowest and highest weight categories on univariate analysis. Similar trends were seen after subgroup stratification. On multivariable analysis, for every 20 kg increase in patient weight, there was a ~6.5% decrease in the change in Hb per RBC unit transfused (p < .0001). CONCLUSIONS: Patient body weight differentially impacts the change in Hb after RBC transfusion. These findings justify incorporating body weight into the clinical decision-making process when transfusing blood in adult surgical patients.

Balancing the Blood Component Transfusion Ratio for High- and Ultra High-Dose Cell Salvage Cases.

OBJECTIVE: To assess the ratio of non-red blood cell to red blood cell components required to avoid coagulopathy when transfusing large amounts of salvaged blood using laboratory test-guided therapy. DESIGN: Retrospective cohort study. SETTING: Single-center, academic hospital. PARTICIPANTS: Thoracoabdominal and abdominal open aortic surgery patients. MEASUREMENT AND MAIN RESULTS: Thirty-eight patients in whom at least 1,000 mL of salvaged red blood cells were transfused were identified and divided into the following 2 cohorts: 1,000-to-2,000 mL of salvaged red blood cells (high dose) (n = 20) and >2,000 mL of salvaged red blood cells (ultra-high dose) (n = 18). Compared with the high-dose cohort, the ultra high-dose cohort received ∼4 times more salvaged red blood cells (1,240 ± 279 mL v 5,550 ± 3,801 mL). With transfusion therapy guided by intraoperative coagulation tests and thromboelastography, the adjusted ratio of non-red blood cell to red blood cell components (plasma + platelets + cryoprecipitate:allogeneic + salvaged red blood cells) was 0.59 ± 0.66 in the high-dose and 0.93 ± 0.27 in the ultra high-dose cohorts. Multiple coagulation parameters were normal and similar between cohorts at the end of surgery, as determined by the mean, median, and 95% confidence intervals. CONCLUSIONS: When transfusing large volumes of salvaged blood, it is important to balance the ratio between non-red blood cell and red blood cell components. Through a laboratory test-guided approach, coagulopathy was not detected when transfusing blood in ratios of approximately 1:2 for patients receiving 1,000-to-2,000 mL of salvaged blood and 1:1 for patients receiving >2,000 mL of salvaged blood.