Murugan, Raghavan

Background: Paroxysmal atrial fibrillation (AF) is a common arrhythmogenic complication in the intensive care unit (ICU) associated with significant costs and morbidity. AF detection models in the literature have utilized entropic analyses of telemetry to diagnose AF, but they have not been utilized to forecast its onset. Hypothesis: Deep learning techniques applied to high-frequency telemetry data can be utilized to forecast AF with clinically significant lead time. Methods: Telemetry data were collected from 72 ICU monitored beds in a tertiary care center from 4/2018-12/2023. Coefficient of sample entropy (COSEn), a metric of heart rate variability complexity, was calculated on successive 1-minute windows and utilized to identify areas of AF initiation as well as control windows without AF. ECG lead II (256Hz) and beat-to-beat (B2B) measurements were analyzed by the MOMENT time-series foundational model to generate representation embeddings. A binary classification multilayer perceptron (MLP) was trained on these embeddings to forecast AF onset 5m in advance. All results were reported on separate test sets. Group-based trajectory modelling (GBTM) was performed on risk score trajectories from the 20m before AF onset or absence. Results: Over 230,000 hours of B2B data from 2263 patients was analyzed using COSEn to identify 287 transitions to AF from 136 patients. The same number of AF-absent windows were randomly selected. COSEn alone as a forecaster produced an area under the receiver operating characteristic curve (AUC) of 0.65 (95% CI: 0.61-0.69). The MOMENT + MLP model (MMM) achieved an AUC of 0.81 (0.78-0.83) (Fig. 1). COSEn and MMM were subsequently applied to the two hours preceding AF onset or absence. For COSEn, the mean risk trajectory dipped in the baseline period whereas the MMM risk trajectory sustained a high level through the baseline, highlighting its capacity to forecast AF (Fig. 2). For AF patients, GBMT yielded two types of trajectories: one with elevated risk throughout, and another with a rising trend 5 minutes prior to AF onset (Fig. 3). Conclusions: Deep learning techniques can forecast the onset of AF with up to 5 minutes of lead-time and GBTM allowed for the identification of different pre-AF risk trajectory phenotypes.The ability to forecast AF onset has potential to clinically prevent poor outcomes in the ICU, though further refinements to augment the amount of lead time are warranted.

Objective: Acute kidney injury requiring renal replacement therapy in severe vasodilatory shock is associated with an unfavorable prognosis. Angiotensin II treatment may help these patients by potentially restoring renal function without decreasing intrarenal oxygenation. We analyzed the impact of angiotensin II on the outcomes of acute kidney injury requiring renal replacement therapy. Design: Post hoc analysis of the Angiotensin II for the Treatment of High-Output Shock 3 trial. Setting: ICUs. Patients: Patients with acute kidney injury treated with renal replacement therapy at initiation of angiotensin II or placebo ( n = 45 and n = 60, respectively). Interventions: IV angiotensin II or placebo. Measurements and Main Results: Primary end point: survival through day 28; secondary outcomes included renal recovery through day 7 and increase in mean arterial pressure from baseline of ≥ 10 mm Hg or increase to ≥ 75 mm Hg at hour 3. Survival rates through day 28 were 53% (95% CI, 38%–67%) and 30% (95% CI, 19%–41%) in patients treated with angiotensin II and placebo ( p = 0.012), respectively. By day 7, 38% (95% CI, 25%–54%) of angiotensin II patients discontinued RRT versus 15% (95% CI, 8%–27%) placebo ( p = 0.007). Mean arterial pressure response was achieved in 53% (95% CI, 38%–68%) and 22% (95% CI, 12%–34%) of patients treated with angiotensin II and placebo ( p = 0.001), respectively. Conclusions: In patients with acute kidney injury requiring renal replacement therapy at study drug initiation, 28-day survival and mean arterial pressure response were higher, and rate of renal replacement therapy liberation was greater in the angiotensin II group versus the placebo group. These findings suggest that patients with vasodilatory shock and acute kidney injury requiring renal replacement therapy may preferentially benefit from angiotensin II.

Objective: To evaluate the quality of available evidence of drug class combinations and their association with the development of acute kidney injury (AKI). Data Sources: A search of MEDLINE and Embase databases was completed using the following terms: “risk factor AND (acute kidney injury or acute kidney failure) AND (drug or medication).” Study Selection and Data Extraction: Inclusion criteria were the following: English language, full-text availability, and at least 1 drug-combination. Each citation was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) criteria. The literature was evaluated using the quality of evidence component of GRADE. No standardized definition of AKI was applied throughout.. Data Synthesis: Out of 2139 total citations, 151 were assessed for full-text review, with 121 citations (6%) meeting inclusion criteria, producing76 unique drug class combinations. Overall, 56 combinations (73.7%) were considered very low quality; 12 (15.8%) were considered low quality. There were 8 (10.5%) of moderate quality, and no combination was considered high quality. 58 (76%) combinations that had a single citation,with a mean of 1.6 citations per drug class combination. The combination of nonsteroidal anti-inflammatory drugs (NSAIDs) and diuretics was reported in 10 citations, the largest number of citations. Conclusions: Our study demonstrates a lack of well-designed studies addressing drug class combination–associated AKI. The combination of NSAIDs and diuretics with or without additional renin-angiotensin aldosterone agents had the strongest level of evidence. Despite limitations, the information included in this review may result in additional scrutiny about combining certain individual nephrotoxic drugs.