Life-threatening dyskalemia, defined as an abnormal serum potassium concentration, is common in emergency settings that requires timely recognition and treatment and can be detected via AI-enabled electrocardiography. We conducted a pragmatic, open-label, randomized controlled trial with physician-level randomization to evaluate whether a real-time AI-enabled electrocardiography alert could improve physicians' management of dyskalemia. Over a six-month period in 2022, 70 emergency physicians were randomized (35 intervention, 35 control) and provided care to 14,989 patients (7506 in the intervention group and 7483 in the control group) at an academic medical center and a community hospital (ClinicalTrials.gov NCT05118022). The trial had two primary outcomes: the rate of hyperkalemia-related treatment and hypokalemia-related treatment within three hours. The intervention consisted of a real-time pop-up alert in the electronic health records that categorized patients at risk of moderate-to-severe hyperkalemia (≥6.0 mmol/L) or hypokalemia (≤3.0 mmol/L). Physicians in the control group did not receive alerts. Overall, the rate of hyperkalemia-related treatment was not significantly greater in the intervention group (8.0%) than in the control group (7.7%) (hazard ratio 1.05; 95% CI 0.94-1.17; p = 0.420). Similarly, the rate of hypokalemia-related treatment did not differ significantly (2.1% vs. 2.4%; hazard ratio 0.91; 95% CI 0.74-1.13; p = 0.392). Among patients identified by AI-enabled electrocardiography as having hyperkalemia, however, hyperkalemia-related treatment occurred more frequently in the intervention group (69.1% vs. 41.6%; hazard ratio 2.23; 95% CI 1.44-3.46; p < 0.001). This trial demonstrates that a real-time AI-enabled electrocardiography alert facilitated earlier treatment among patients identified as high risk for hyperkalemia.

OBJECTIVE:To evaluate the prognostic value of N-terminal pro-B-type natriuretic peptide (NT-proBNP) across different heart failure phenotypes in patients with chronic kidney disease (CKD). METHODS:This retrospective cohort study used data from the TriNetX Global Research Network to identify adults with non-dialysis-dependent CKD and concurrent heart failure with reduced ejection fraction (HFrEF) or heart failure with preserved ejection fraction (HFpEF) between January 1, 2002, and December 31, 2020. Propensity score matching was applied to balance baseline characteristics. NT-proBNP and metabolic markers were analyzed, and outcomes were evaluated over a 3-year period, including all-cause mortality, major adverse cardiac events (MACE), and major adverse kidney events. RESULTS:Among 14,758 matched patients (n=7379 per group), those with HFrEF had significantly higher NT-proBNP levels and an elevated risk of all-cause mortality (HR, 1.09; P=.02) and MACE (HR, 1.17; P<.001) compared with patients with HFpEF. Major adverse kidney events incidence did not differ significantly between groups. Elevated NT-proBNP was strongly associated with adverse cardiovascular outcomes in the HFrEF subgroup. Distinct metabolic and nutritional profiles also influenced outcome trajectories. CONCLUSION:Among patients with CKD, those with HFrEF exhibited higher risks of MACE and mortality than those with HFpEF, especially in the presence of elevated NT-proBNP. Although NT-proBNP facilitates cardiovascular risk stratification, its predictive value for renal outcomes in patients with concurrent CKD and heart failure is limited.

RATIONALE & OBJECTIVE:Severe hypokalemia requires prompt management and close surveillance. Although artificial intelligence-enabled electrocardiography (AI-ECG) rapidly detects severe hypokalemia, its application for monitoring serum potassium (K+) dynamics during treatment remains unexplored. This study assessed AI-ECG performance in monitoring K+ changes during supplementation. STUDY DESIGN:Multicenter retrospective cohort study. SETTING & PARTICIPANTS:191 adults with severe hypokalemia (Lab-K+ ≤2.5 mmol/L; matched ECG-K+ <3.5 mmol/L) and ≥1 follow-up paired measurement within 24 hours of K+ supplementation at three teaching hospitals between September 2019 and August 2024. TESTS COMPARED:Laboratory-measured K+ (Lab-K+) and K+ estimated by ECG (ECG-K+) overall and stratified by the etiology of hypokalemia (acute K+ shift vs. chronic K+ deficit). OUTCOMES:Primary: agreement between paired ECG-K+ and Lab-K+. Secondary: diagnostic accuracy and K+ trajectories. ANALYTICAL APPROACH:Linear mixed-effects models with patient-level random intercepts; repeated-measures correlation (rmcorr) and Bland-Altman plots; patient-level clustered bootstrapped ROC analysis for diagnostic accuracy. RESULTS:Of 191 patients, 156 (81.7%) had chronic K+ deficits (most commonly gastrointestinal disorders [n=47] or diuretic use [n=35]), and 35 (18.3%) had acute K+ shifts (most commonly thyrotoxic periodic paralysis [n=25]). The chronic K+ deficits group had more comorbidities and use of medications affecting K+. ECG-K+ correlated strongly with Lab-K+ (rmcorr 0.847; 95% CI, 0.81-0.88; p<0.001). The relationship was modified by hypokalemia etiology (interaction p<0.0001) with a lower correlation in patients with chronic K+ deficits. The diagnostic accuracy of ECG-K+ with Lab-K+ ≤3.5 mmol/L was reflected by an AUC of 0.920; 95% CI, 0.863-0.961. It was higher in patients with acute K+ shift. ECG-K+ preceded Lab-K+ results by a mean of 52.5 minutes. Patients with acute K+ shift corrected approximately threefold faster than those with chronic K+ deficit (0.121 vs. 0.039 mmol/L/h). Rebound hyperkalemia was detected by ECG-K+ in two patients before laboratory confirmation. LIMITATIONS:Retrospective design; treatment-protocol heterogeneity; limited inpatient medication granularity and potential selection bias. CONCLUSIONS:AI-ECG enables real-time, within-patient monitoring of serum K+ dynamics during treatment for severe hypokalemia, with superior performance in the setting of acute hypokalemia due to K+ shift. As a non-invasive adjunct, AI-ECG may shorten time to detect changes in K+ and reduce the need for laboratory K+ measurements than exclusive reliance on Lab-K+ measurements. Confirmatory studies are warranted.

BACKGROUND:Hyperkalemia can be detected by point-of-care (POC) blood testing and by artificial intelligence- enabled electrocardiography (ECG). These 2 methods of detecting hyperkalemia have not been compared. OBJECTIVE:To determine the accuracy of POC and ECG potassium measurements for hyperkalemia detection in patients with critical illness. METHODS:This retrospective study involved intensive care patients in an academic medical center from October 2020 to September 2021. Patients who had 12-lead ECG, POC potassium measurement, and central laboratory potassium measurement within 1 hour were included. The POC potassium measurements were obtained from arterial blood gas analysis; ECG potassium measurements were calculated by a previously developed deep learning model. Hyperkalemia was defined as a central laboratory potassium measurement of 5.5 mEq/L or greater. RESULTS:Fifteen patients with hyperkalemia and 252 patients without hyperkalemia were included. The POC and ECG potassium measurements were available about 35 minutes earlier than central laboratory results. Correlation with central laboratory potassium measurement was better for POC testing than for ECG (mean absolute errors of 0.211 mEq/L and 0.684 mEq/L, respectively). For POC potassium measurement, area under the receiver operating characteristic curve (AUC) to detect hyperkalemia was 0.933, sensitivity was 73.3%, and specificity was 98.4%. For ECG potassium measurement, AUC was 0.884, sensitivity was 93.3%, and specificity was 63.5%. CONCLUSIONS:The ECG potassium measurement, with its high sensitivity and coverage rate, may be used initially and followed by POC potassium measurement for rapid detection of life-threatening hyperkalemia.

Background:Severe hyperkalemia is a life-threatening emergency requiring prompt management and close surveillance. Although artificial intelligence-enabled electrocardiography (AI-ECG) has been developed to rapidly detect hyperkalemia, its application to monitor potassium (K+) levels remains unassessed. This study aimed to evaluate the effectiveness of AI-ECG for monitoring K+ levels in patients with severe hyperkalemia. Methods:This retrospective study was performed at an emergency department of a single medical center over 2.5 years. Patients with severe hyperkalemia defined as Lab-K+ ≥6.5 mmol/l with matched ECG-K+ ≥5.5 mmol/l were included. ECG-K+ was quantified by ECG12Net analysis of the AI-ECG system. The following paired ECG-K+ and Lab-K+ were measured at least twice, almost simultaneously, during and after K+-lowering therapy in 1 day. Clinical characteristics, pertinent intervention, and laboratory data were analyzed. Results:Seventy-six patients fulfilling the inclusion criteria exhibited initial Lab-K+ 7.4 ± 0.7 and ECG-K+ 6.8 ± 0.5 mmol/l. Most of them had chronic kidney disease (CKD) or were on chronic hemodialysis (HD). The followed Lab-K+ and ECG-K+ measured with a mean time difference of 11.4 ± 5.6 minutes significantly declined in parallel both in patients treated medically (n = 39) and with HD (n = 37). However, there was greater decrement in Lab-K⁺ (mean 7.3 to 4.1) than ECG-K⁺ (mean 6.6 to 5.0) shortly after HD. Three patients with persistent ECG-K+ hyperkalemia despite normalized Lab-K+ exhibited concomitant acute cardiovascular comorbidities. Conclusions:AI-ECG for K+ prediction may help monitor K+ level for severe hyperkalemia and reveal more severe cardiac disorders in the patients with persistent AI-ECG hyperkalemia.

Abstract Background Hyperthyroidism is frequently under-recognized and leads to heart failure and mortality. Timely identification of high-risk patients is a prerequisite to effective antithyroid therapy. Since the heart is very sensitive to hyperthyroidism and its electrical signature can be demonstrated by electrocardiography, we developed an artificial intelligence model to detect hyperthyroidism by electrocardiography and examined its potential for outcome prediction. Methods The deep learning model was trained using a large dataset of 47,245 electrocardiograms from 33,246 patients at an academic medical center. Patients were included if electrocardiograms and measurements of serum thyroid-stimulating hormone were available that had been obtained within a three day period. Serum thyroid-stimulating hormone and free thyroxine were used to define overt and subclinical hyperthyroidism. We tested the model internally using 14,420 patients and externally using two additional test sets comprising 11,498 and 596 patients, respectively. Results The performance of the deep learning model achieves areas under the receiver operating characteristic curves (AUCs) of 0.725–0.761 for hyperthyroidism detection, AUCs of 0.867–0.876 for overt hyperthyroidism, and AUC of 0.631–0.701 for subclinical hyperthyroidism, superior to a traditional features-based machine learning model. Patients identified as hyperthyroidism-positive by the deep learning model have a significantly higher risk (1.97–2.94 fold) of all-cause mortality and new-onset heart failure compared to hyperthyroidism-negative patients. This cardiovascular disease stratification is particularly pronounced in subclinical hyperthyroidism, surpassing that observed in overt hyperthyroidism. Conclusions An innovative algorithm effectively identifies overt and subclinical hyperthyroidism and contributes to cardiovascular risk assessment.

Icodextrin is widely utilized as an osmotic agent in peritoneal dialysis (PD) prescription for clinical patients with inadequate ultrafiltration, but icodextrin induced acute generalized exanthematous pustulosis (AGEP) has not been well recognized. We described a young-aged female with IgA nephropathy under continuous automated peritoneal dialysis who developed skin erythema with exfoliation over the groin at 7th day after first infusion of icodextrin based PD prescription. Her scaling skin lesion with pinhead-sized pustules invaded the bilateral inguinal folds at first, and then extended to general trunk accompanied by pruritus and mild tingly in rapid succession. As expansion of confluent skin lesion, she was admitted on 14th day of icodextrin exposure and PD treatment was instantly ceased. She was afebrile with stable vital sign and physical examination was notable for widespread erythematous papules with pruritus extending over her groins, abdomen and back. Pertinent laboratory examination showed leukocytosis of 18970 cells/μL with neutrophile count of 17642 cells/μL (92.3 %), and c-reactive-protein: 3.39 mg/dl. Skin biopsy reveals multifocal subcorneal abscess with papillary dermal edema, and upper-dermal neutrophilia with perivascular accentuation, consistent with the diagnosis of AGEP. After discontinuation of PD, she underwent temporary high-flux hemodialysis and treatment of steroid and antihistamine, her dermatologic lesion resolved totally four days later and restarted peritoneal dialysis at 17th day without any skin sequalae. This case highlighted the fact that icodextrininduced AGEP should be early recognized after first use of icodextrin to avoid misdiagnosis to other dermatoses and severe complication without appropriate management.

Abstract Background Central diabetes insipidus (CDI) in patients with intracranial germ cell tumors (GCTs) could develop due to suprasellar involvement or postoperative complication after transsphenoidal surgery (TSS). However, it is important to early recognize the pre-existing CDI caused by GCTS with normal sodium (Na+) level before surgery. Case presentation A 25-year-old male presented with progressive fatigue, weakness, polydipsia and loss of libido in the past one year. Laboratory finding was unremarkable without hypernatremia. Hormonal testing indicating anterior pituitary insufficiency. Brain magnetic resonance imaging showed two mass lesions in the sellar/suprasellar and pineal regions with obstructive hydrocephalus. The patient developed polyuria with hypernatremia after 6-hour fasting and TSS with incisional biopsy. Diagnosis of intracranial GCTs with CDI was confirmed by pathology and desmopressin test. Without surgical damage to posterior pituitary or tumor resection which might led to postoperative CDI, pre-existing CDI before surgery could be unrecognized by normal serum Na+ and unrestricted water intake. Conclusions Clinicians should notice the typical manifestations of intracranial GCTs, especially in CDI, to avoid potential complications. Fluid restriction before surgery is a risk factor to develop hypernatremia caused by CDI that was masked by polydipsia before surgery.

A 48-year-old man was admitted with decreased appetite, lethargy, dizziness, and general weakness. He had a history of chronic kidney disease, stage IV of uncertain cause, for years and hyperparathyroidism with hypercalcemia from a parathyroid adenoma diagnosed about 6 months before his current presentation. His serum intact parathyroid hormone and ionized calcium concentrations declined from 440 to 11 pg/ml and from 6.0 to 4.2 mg/dl, respectively, 1 week after the removal of parathyroid adenoma.

BackgroundRare cases of de novo or relapsed kidney diseases associated with vaccination against coronavirus disease 2019 (COVID-19) have been increasingly reported. The aim of this study was to report the incidence, etiologies, and outcomes of acute kidney disease (AKD) following COVID-19 vaccination.MethodsThis retrospective study extracted cases from renal registry of a single medical center from 1 March 2021 to 30 April 2022, prior to the significant surge in cases of the Omicron variant of COVID-19 infection in Taiwan. Adult patients who developed AKD after COVID-19 vaccination were included. We utilized the Naranjo score as a causality assessment tool for adverse vaccination reactions and charts review by peer nephrologists to exclude other causes. The etiologies, characteristics, and outcomes of AKD were examined.ResultsTwenty-seven patients (aged 23 to 80 years) with AKD were identified from 1,897 vaccines (estimated rate of 13.6 per 1000 patient-years within the renal registry). A majority (77.8%) of vaccine received messenger RNA-based regimens. Their median (IQR) Naranjo score was 8 (6-9) points, while 14 of them (51.9%) had a definite probability (Naranjo score ≥ 9). The etiologies of AKD included glomerular disease (n = 16) consisting of seven IgA nephropathy, four anti-neutrophil cytoplasmic antibodies-associated glomerulonephritis (AAN), three membranous glomerulonephritis, two minimal change diseases, and chronic kidney disease (CKD) with acute deterioration (n = 11). Extra-renal manifestations were found in four patients. Over a median (IQR) follow-up period of 42 (36.5–49.5) weeks, six patients progressed to end-stage kidney disease (ESKD).ConclusionBesides glomerulonephritis (GN), the occurrence of AKD following COVID-19 vaccination may be more concerning in high-risk CKD patients receiving multiple doses. Patients with the development of de novo AAN, concurrent extra-renal manifestations, or pre-existing moderate to severe CKD may exhibit poorer kidney prognosis.

The nucleotide-binding and oligomerization domain, leucine-rich repeats, and pyrin domain-containing protein 3 (NLRP3) inflammasome plays a crucial role in innate immunity and is involved in the pathogenesis of autoinflammatory diseases. Glycolysis regulates NLRP3 inflammasome activation in macrophages. However, how lactic acid fermentation and pyruvate oxidation controlled by the mitochondrial pyruvate carrier (MPC) affect NLRP3 inflammasome activation and autoinflammatory disease remains elusive. We found that the inactivation of MPC with genetic depletion or pharmacological inhibitors, MSDC-0160 or pioglitazone, increased NLRP3 inflammasome activation and IL-1β secretion in macrophages. Glycolytic reprogramming induced by MPC inhibition skewed mitochondrial ATP-associated oxygen consumption into cytosolic lactate production, which enhanced NLRP3 inflammasome activation in response to monosodium urate (MSU) crystals. As pioglitazone is an insulin sens MSDC-itizer used for diabetes, its MPC inhibitory effect in diabetic individuals was investigated. The results showed that MPC inhibition exacerbated MSU-induced peritonitis in diabetic mice and increased the risk of gout in patients with diabetes. Altogether, we found that glycolysis controlled by MPC regulated NLRP3 inflammasome activation and gout development. Accordingly, prescriptions for medications targeting MPC should consider the increased risk of NLRP3-related autoinflammatory diseases.

BackgroundIdentifying candidates responsive to treatment is important in lupus nephritis (LN) at the renal flare (RF) because an effective treatment can lower the risk of progression to end-stage kidney disease. However, machine learning (ML)-based models that address this issue are lacking.MethodsTranscriptomic profiles based on DNA microarray data were extracted from the GSE32591 and GSE112943 datasets. Comprehensive bioinformatics analyses were performed to identify disease-defining genes (DDGs). Peripheral blood samples (GSE81622, GSE99967, and GSE72326) were used to evaluate the effect of DDGs. Single-sample gene set enrichment analysis (ssGSEA) scores of the DDGs were calculated and correlated with specific immunology genes listed in the nCounter panel. GSE60681 and GSE69438 were used to examine the ability of the DDGs to discriminate LN from other renal diseases. K-means clustering was used to obtain the separate gene sets. The clustering results were extended to data derived using the nCounter technique. The least absolute shrinkage and selection operator (LASSO) algorithm was used to identify genes with high predictive value for treatment response after the first RF in each cluster. LASSO models with tenfold validation were built in GSE200306 and assessed by receiver operating characteristic (ROC) analysis with area under curve (AUC). The models were validated by using an independent dataset (GSE113342).ResultsForty-five hub genes specific to LN were identified. Eight optimal disease-defining clusters (DDCs) were identified in this study. Th1 and Th2 cell differentiation pathway was significantly enriched in DDC-6. LCK in DDC-6, whose expression positively correlated with various subsets of T cell infiltrations, was found to be differentially expressed between responders and non-responders and was ranked high in regulatory network analysis. Based on DDC-6, the prediction model had the best performance (AUC: 0.75; 95% confidence interval: 0.44-1 in the testing set) and high precision (0.83), recall (0.71), and F1 score (0.77) in the validation dataset.ConclusionsOur study demonstrates that incorporating knowledge of biological phenotypes into the ML model is feasible for evaluating treatment response after the first RF in LN. This knowledge-based incorporation improves the model's transparency and performance. In addition, LCK may serve as a biomarker for T-cell infiltration and a therapeutic target in LN.

Context: Abnormal serum calcium concentrations affect the heart and may alter the electrocardiogram (ECG), but the detection of hypocalcemia and hypercalcemia (collectively dyscalcemia) relies on blood laboratory tests requiring turnaround time.Objective: The study aimed to develop a bloodless artificial intelligence (AI)-enabled (ECG) method to rapidly detect dyscalcemia and analyze its possible utility for outcome prediction. Methods: This study collected 86,731 development, 15,611 tuning, 11,105 internal validation, and 8401 external validation ECGs from electronic medical records with at least 1 ECG associated with an albumin-adjusted calcium (aCa) value within 4 h. The main outcomes were to assess the accuracy of AI-ECG to predict aCa and follow up these patients for all-cause mortality, new-onset acute myocardial infraction (AMI), and new-onset heart failure (HF) to validate the ability of AI-ECG-aCa for previvor identification.Results: ECG-aCa had mean absolute errors (MAE) of 0.78/0.98 mg/dL and achieved an area under receiver operating characteristic curves (AUCs) 0.9219/0.8447 and 0.8948/0.7723 to detect severe hypercalcemia and hypocalcemia in the internal/external validation sets, respectively. Although < 20 % variance of ECG-aCa could be explained by traditional ECG features, the ECG-aCa was found to be associated with more complications. Patients with ECG-hypercalcemia but initially normal aCa were found to have a higher risk of subsequent all -cause mortality [hazard ratio (HR): 2.05, 95 % conference interval (CI): 1.55-2.70], new-onset AMI (HR: 2.88, 95 % CI: 1.72-4.83), and new-onset HF (HR: 2.02, 95 % CI: 1.38-2.97) in the internal validation set, which were also seen in external validation.Conclusion: The AI-ECG-aCa may help detecting severe dyscalcemia for early diagnosis and ECG-hypercalcemia also has prognostic value for clinical outcomes (all-cause mortality and new-onset AMI and HF).

Mycophenolate (mycophenolate mofetil [MMF]; mycophenolate sodium [MPS]) and tacrolimus (FK-506) are commonly and concomitantly used to prevent rejection in organ transplant. Mycophenolate-induced hepatotoxicity causing the reduced FK-506 metabolism with nephrotoxicity may be less appreciated, leading to inappropriate management. We describe a new living donor kidney recipient receiving pretransplant and post-transplant immunosuppressants including oral mycophenolate (MMF 1 g daily) and tacrolimus (FK-506 4-8 mg daily) who developed progressive liver dysfunction (up to 10-fold increase) despite the reduced FK-506 dosage (6 mg daily). A thorough investigation including infection, inflammation, and autoimmune hepatitis were unremarkable. With a withdrawal of MMF, his liver function improved, but persistently higher trough serum FK-506 level (12-15 ng/mL) and increased serum creatinine were notable. Moreover, the reintroduction of MPS with the reduced FK-506 dosage (4 mg daily) worsened liver function along with FK-506 nephrotoxicity (serum creatinine from 1.4-2.4 mg/dL). The replacement of MPS with mammalian target of rapamycin inhibitor not only resolved liver injury but also normalized serum FK-506 level and kidney function. Mycophenolate should be kept in mind as a cause of drug-induced hepatotoxicity that can reduce tacrolimus metabolism, leading to FK-506 nephrotoxicity and acute kidney injury in organ transplant.