Pharmacotherapy最新文献

Objective: To use machine learning methods to identify factors associated with corticosteroid (CS) discontinuation 1 year after adult heart transplantation (HT).

Design: Retrospective, observational, cohort study.

Data source: This study used data from the United Network for Organ Sharing (UNOS) database.

Patients: We included adults (age ≥ 18 years) who underwent their first HT between January 2000 and December 2023 in the UNOS database with follow-up through December 2024, who were discharged on a CS.

Measurements: We divided the cohort into those with or without CS at 1-year post-transplant follow-up. We used the eXtreme Gradient Boosting (XGBoost) algorithm to build a model predicting CS discontinuation at 1 year. Relevant recipient, donor, and transplant variables were included to train the model, with Shapley Additive Explanations (SHAP) used to identify and interpret the most important and predictive features.

Results: We identified 72,730 HT recipients; 13,017 (17.9%) had CS discontinued within the first year. Compared with the CS cessation group, those who continued CS were more likely to have had a lower BMI, lower ischemic etiology of cardiomyopathy, lower intra-aortic balloon pump (IABP) and left ventricular assist device (LVAD) use before HT, better renal function, and sustained longer donor ischemic time. Model performance was strong, with an area under the curve of 0.854 (95% confidence interval: 0.848-0.861). Lower average transplant center volume (number of transplants per year), shorter donor ischemic time, and LVAD use at transplant predicted CS discontinuation.

Conclusions: In a large national database, utilizing novel ML modeling techniques, we identified annual transplant center volume, donor ischemia time, and LVAD use at the time of HT as the best predictors of CS discontinuation 1 year after HT.

Introduction: Vancomycin is a critical antibiotic for treating methicillin-resistant Staphylococcus aureus and other gram-positive bacterial infections, but achieving and maintaining therapeutic trough concentrations is challenging.

Objectives: We hypothesized that a deep learning model could accurately predict vancomycin trough concentrations 2 days in advance in critically ill patients and provide recommendations for optimal dosing adjustments to achieve target drug concentrations.

Methods: We trained and validated the model using electronic health record (EHR) data from adults admitted to the University of California San Diego Health system intensive care units (ICUs) from January 1, 2016, to June 30, 2024. Features included patient demographics, comorbidities, vital signs, laboratory measurements, medications, and vancomycin dosing information. The model architecture combined Long Short-Term Memory and Multi-Head Attention layers, supplemented with skip connections to incorporate past dosage information at the final layer of the deep learning model. Model performance was evaluated using mean absolute error (MAE) and root mean square error (RMSE) metrics.

Results: A total of 2205 encounters met the eligibility criteria. The median age was 57 years, and the median ICU length of stay was 4.9 days. The model achieved an MAE of 3.15 mg/L and an RMSE of 4.17 mg/L, comparable to that of a critical care pharmacist aided by a Bayesian dosing software. Additionally, deviations from patient-specific model-based dose recommendations were generally associated with nontherapeutic vancomycin levels.

Conclusion: This study demonstrates the potential to leverage deep learning to individualize and support vancomycin therapeutic drug monitoring in critically ill patients.

Post-resuscitation cardiac arrest care begins at the time of hospital admission and focuses on preventing the sequelae of ischemia-reperfusion injury, including secondary brain damage and post-cardiac arrest syndrome. Clinical practice guidelines on post-resuscitation cardiac arrest care from international medical organizations outline evidence-based practices but sometimes lack pragmatic details needed by bedside clinicians to implement change. Lack of information has required providers to extrapolate from other patient populations with acute brain injuries or utilize lower quality data from patient registries or observational studies. We organized a group of content experts to address selected pharmacotherapeutic controversies encountered during temperature control after out-of-hospital cardiac arrest in adult patients. Data on pre-hospital interventions, in-hospital cardiac arrest, and most non-pharmacologic treatments were excluded. A Delphi process was completed using three rounds of voting to reach consensus on pharmacotherapeutic controversies to review. The original list of 11 topics was narrowed iteratively and the final list included: (i) sedation and analgesia, (ii) seizures and myoclonus, (iii) shivering, and (iv) early-onset pneumonia prevention. Writing groups conducted systematic literature searches of MEDLINE using PubMed focusing on contemporary publications in the past 30 years when temperature control was considered standard care. Each section reviews the scope and management of the controversy and provides a conclusion with suggested future research directions. The information in this review is intended to support providers with the implementation of recommendations made in clinical practice guidelines on post-resuscitation cardiac arrest care and should not necessarily supplant them.

Background: The nephrotoxic risks of combining ceftazidime/avibactam (AVI) with vancomycin (VAN) remain underexplored, despite both agents independently being linked to acute kidney injury (AKI). This study assessed the risk of AKI associated with concurrent VAN and ceftazidime/avibactam (VAN-AVI) therapy and developed synthetic data models to enable early prediction of AKI.

Methods: We conducted a retrospective analysis using electronic health record data from hospitalized adults between 2015 and 2022. The incidence of AKI was compared among patients receiving VAN-AVI or VAN in combination with piperacillin/tazobactam (VAN-TPZ) versus VAN monotherapy. AKI was defined as a composite of de novo and recurrent AKI (i.e., patients had a prior diagnosis of AKI within the preceding 6 months and experienced a new AKI event after 7 days of VAN-AVI initiation). To address sample size imbalance, we applied inverse probability of treatment weighting (IPTW) and generated synthetic datasets using Conditional Tabular Generative Adversarial Networks (CTGAN) and Tabular Variational Autoencoders (TVAE). These synthetic datasets were subsequently used to augment machine learning (ML) models aimed at the early prediction of AKI in patients treated with VAN-AVI combination therapy.

Results: Among the 92 patients receiving VAN-AVI combination therapy, only four (4.3%) patients experienced new-onset AKI, and 66 (71.7%) patients had a recurrent AKI. After applying IPTW, VAN-AVI was associated with a higher risk of AKI Hazard Ratio (HR) = 3.47; 95% Confidence Interval (CI): 1.97-6.11, followed by VAN-TPZ (HR = 1.96; 95% CI: 1.37-2.81), compared to VAN alone. Synthetic data analyses conducted over 1000 iterations supported these findings, with mean HRs for VAN-AVI of 3.80 using TVAE and 4.45 using CTGAN. ML models augmented with synthetic data outperformed those using original data alone. For 30-day AKI prediction, F1-scores improved across all models, with the highest performance observed in the augmented XGBoost and logistic regression classifier (F1 = 0.80).

Conclusion: This study introduces a novel approach that integrates IPTW with synthetic data generation to evaluate drug-associated AKI risk in small-sample cohorts. Although our findings demonstrate a lower incidence of de novo AKI in the VAN-AVI group, the use of synthetic data and augmented ML models significantly improved early AKI prediction. These findings support the potential utility of synthetic data frameworks for scalable drug safety evaluations, although further validation is warranted.

Background: Acute ischemic stroke (AIS) accounts for 87% of all strokes and is associated with high morbidity and mortality. Stroke mimics, conditions that present with stroke-like symptoms but are not AIS, represent an important proportion of stroke cases. Tenecteplase, an alteplase variant with higher fibrin affinity and a longer half-life, has recently gained approval for the treatment of AIS. Limited evidence exists on the safety of tenecteplase in stroke mimics.

Objective: The objective of this study was to determine the incidence of intracranial hemorrhage in patients with stroke-like signs and symptoms who received tenecteplase with and without a confirmed diagnosis of AIS.

Methods: This was a single-center, retrospective cohort study conducted at a comprehensive stroke center. Patients who received tenecteplase for stroke-like symptoms from June 2020 to December 2021 were included. Patients were grouped based on AIS diagnosis confirmed by neuroimaging or discharge diagnosis. The primary outcome was the incidence of intracranial hemorrhage, and secondary outcomes included incidence of symptomatic and asymptomatic intracranial hemorrhage, extracranial hemorrhage, hospital length of stay, angioedema development, and discharge disposition.

Results: Of 250 patients, 174 were diagnosed with AIS and 76 with stroke mimics. Patients with AIS were older (69 years [SD: 13.8 years] vs. 62.3 years [SD: 16.1 years]; p = 0.0008), had higher hypertension prevalence (132 [75.9%] vs. 44 [57.9%]; p = 0.0042), and lower previous stroke prevalence (9 [5.2%] vs. 11 [14.5%]; p = 0.0126) compared with stroke-mimic patients, respectively. Patients with AIS had shorter median tenecteplase door-to-needle times (38.5 min [IQR: 30 min, 53 min] vs. 46.5 min [IQR: 34.5 min, 64.5 min]; p = 0.0154) compared with stroke mimics. Intracranial hemorrhage occurred in 11.5% of patients with AIS and none in stroke mimics (p = 0.0021). Symptomatic hemorrhage rates were similar between groups. Patients with AIS had longer median hospital stays (3 days [IQR: 2 days, 6 days] vs. 2 days [IQR: 2 days, 4 days]; p = 0.0018). Discharge dispositions and cases of angioedema were similar between groups.

Conclusion: Tenecteplase appears safe for stroke mimics, with no disproportionate harm compared to patients with AIS. This study further supports the safety of tenecteplase in stroke mimics, aiding rapid treatment decisions in stroke-like presentations.

Background: Critically ill patients may be overexposed to valproate because altered protein binding leads to a disproportionate free valproate fraction. The Fraser equation was derived and internally validated to estimate critically ill patients' free valproate concentrations, but it requires external validation.

Methods: Adult intensive care unit (ICU) patients at two academic centers with concurrently measured free and total valproate concentrations were included. Free valproate concentrations were estimated using the Fraser equation which includes total valproate, albumin and BUN concentration, and whether the patient received propofol or aspirin. The primary outcome was Fraser equation performance, assessed using Bland-Altman methods. Comparative performance against estimates from the Doré equation (an alternative predictive equation), therapeutic concordance using a target free valproate range of 5-15 mg/L, and equation improvements were explored.

Results: Overall, 315 patients and 556 free-total valproate concentration pairs were included. The mean (±SD) age was 58 (±17) years and 90 (29%) patients were on valproate prior to hospital admission. The Fraser equation estimated free valproate concentrations were correlated with measured concentrations (r = 0.728) with a negative bias (mean bias -2.77 mg/L, 95% LOA -18.9, 13.4). The Fraser equation increasingly underestimated measured concentrations as measured concentrations increased. 71% of Fraser equation estimates were therapeutically concordant (e.g., estimate and measured both within reference range) compared to 61.9% of Doré estimates (p = 0.001). Fraser equation modifications led to minor performance improvements but did not overcome worsening underestimation with higher measured free valproate concentrations.

Conclusion: The Fraser equation was moderately accurate and corresponded with an appropriate interpretation for 71% of free valproate concentrations. Worse underestimation with higher measured free valproate concentrations suggests direct measurement of free valproate concentrations is warranted. While the Fraser equation could complement therapeutic decision making at centers where direct free valproate measurements are unavailable or slow to return, its accuracy is currently insufficient to replace direct measurement.

Background and objective: Opioids and benzodiazepines are the most widely utilized sedative-analgesic agents in pediatric intensive care units (PICUs). However, prolonged exposure to these drugs may lead to drug tolerance and dependence, whereas abrupt discontinuation or rapid dose reduction can precipitate withdrawal symptoms, which increase the risk of complications such as medical device dislodgement, exacerbated patient discomfort, and prolonged mechanical ventilation dependency and hospital stays. Studies have revealed that up to 94% of children in PICUs develop iatrogenic withdrawal syndrome (IWS), but there is currently no gold standard for a definitive diagnosis of IWS. This study aims to explore the correlation between salivary cortisol levels, sedation-analgesia status, and the occurrence of IWS in critically ill children.

Methods: A total of 118 critically ill children who were exposed to sedative and/or analgesic drugs for a continuous period of ≥ 5 days at Yuying Children's Hospital, Affiliated Second Hospital of Wenzhou Medical University, from October 2022 to September 2024, were screened, and 106 patients were included in the study. Based on the Sofia observation withdrawal symptoms scale (SOS) score, 42 patients were divided into the IWS group (SOS ≥ 4) and 64 patients into the non-IWS group (SOS < 4). The usage of sedative and analgesic drugs was compared between the two groups, as well as the salivary cortisol concentrations at 24, 48, and 72 h after drug withdrawal. The influencing factors of IWS occurrence and the predictive value of related indicators were analyzed.

Results: The cumulative doses of sedative and analgesic drugs administered in the IWS group were statistically significantly higher than those in the non-IWS group (p < 0.05). The median salivary cortisol concentration in the IWS group was significantly higher at 24 h after drug discontinuation compared with the non-IWS group. Logistic results of binary regression analysis showed that a higher level of salivary cortisol at 24 h after discontinuation was an independent predictor for the occurrence of IWS (p < 0.05). Receiver operating characteristic (ROC) curve analysis indicated an area under the curve of 0.857 (95% confidence interval (CI): 0.782-0.932) with sensitivity and specificity of 88.1% and 84.4%, respectively.

Conclusions: A statistically significant correlation was observed between salivary cortisol levels, the cumulative dosage of sedative-analgesic drugs, and the occurrence of IWS. Higher levels of salivary cortisol are a predictor of IWS occurrence. Detecting salivary cortisol levels has clinical application for early diagnosis of IWS and judgment of withdrawal severity in critically ill children.

Hepatorenal syndrome-acute kidney injury (HRS-AKI) is a life-threatening complication of cirrhosis, marked by profound splanchnic vasodilation leading to renal hypoperfusion. Guidelines throughout the years have refined the definitions, classifications, and diagnostic criteria of HRS-AKI to promote earlier recognition and intervention. Vasoconstrictor therapy remains the cornerstone of the management of HRS-AKI. Among the available agents, terlipressin is the most well-studied and has been endorsed by multiple international guidelines as first-line therapy for HRS-AKI. Although approved by the United States Food and Drug Administration for use in HRS-AKI with an intermittent intravenous bolus dosing strategy, recent consensus guidance recommends continuous infusion as the preferred route of administration. This narrative review summarizes current evidence from clinical trials and comparative studies evaluating terlipressin continuous infusion versus intermittent bolus dosing strategies, with a focus on efficacy, safety, and operational considerations. Clinical trials comparing the two strategies have shown that the continuous infusion strategy provides comparable, and possibly superior, efficacy in HRS reversal compared to intermittent bolus dosing, while also demonstrating a lower incidence of adverse effects. Operational feasibility is supported by physical and chemical stability in various diluents over 24-h infusion periods; however, implementation challenges remain, including limited line compatibility data and line access issues in patients with cirrhosis. In conclusion, the terlipressin continuous infusion strategy appears to be an efficacious, safe, and operationally plausible alternative to intermittent bolus dosing and may be considered for routine use in clinical practice.

Background: Cytomegalovirus (CMV) is a major opportunistic infection in solid organ transplant (SOT) recipients. Although valganciclovir is first-line for CMV treatment, its use may be limited by resistance and myelotoxicity. Letermovir (LET) and maribavir (MBV) are increasingly used as alternative therapies, though real-world outcomes remain incompletely defined. This study evaluated real-world outcomes of LET and MBV in SOT recipients for whom standard CMV therapy was inadequate or not feasible. Viral clearance, breakthrough infection, refractory infection, and resistance emergence were assessed.

Methods: We conducted a retrospective cohort analysis of adult SOT patients receiving LET (2018-2025) and/or MBV (2019-2025) at two transplant centers. Primary outcomes included CMV breakthrough (≥ 500 IU/mL) during LET prophylaxis and CMV clearance (< 200 IU/mL) at the end of MBV treatment. Secondary outcomes included resistant, refractory, and recurrent CMV infection within 3 months post-therapy.

Results: Among 47 LET courses (41 patients), breakthrough CMV occurred in 14/32 (44%) secondary prophylaxis (SP) courses (median viral load [VL]: 1037 IU/mL, interquartile range [IQR] 673-3427) but in none of the 15 primary prophylaxis (PP) courses. No resistance to LET was detected during or after prophylaxis. MBV was administered in 21 courses (17 patients); MBV-refractory/resistant (R/R) infection occurred in 8 courses (38%), and 11 (52%) courses achieved CMV clearance. Within 3 months post-MBV, 14 (67%) courses had CMV VL ≥ 1000 IU/mL, and two (10%) courses experienced recurrence. MBV resistance emerged in 6/17 (35%) patients during or after treatment.

Conclusions: LET was effective for PP but associated with frequent breakthroughs in SP patients. MBV was limited by incomplete viral suppression and emergent resistance. These findings highlight the variable effectiveness of LET and MBV in clinical practice and support the need for further research to define optimal antiviral use in SP and refractory CMV treatment.