Long-short-term memory machine learning of longitudinal clinical data accurately predicts acute kidney injury onset in COVID-19: a two-center study DOI Creative Commons
Justin Y. Lu,

Joanna Zhu,

Jocelyn Zhu

et al.

International Journal of Infectious Diseases, Journal Year: 2022, Volume and Issue: 122, P. 802 - 810

Published: July 22, 2022

ABSTRACT

Objectives

This study used the long-short-term memory (LSTM) artificial intelligence method to model multiple time points of clinical laboratory data, along with demographics and comorbidities, predict hospital-acquired acute kidney injury (AKI) onset in patients COVID-19.

Methods

Montefiore Health System data consisted 1982 AKI 2857 non-AKI (NAKI) hospitalized COVID-19, Stony Brook Hospital validation 308 721 NAKI Demographic, longitudinal (3 days before onset) tests were analyzed. LSTM was fivefold cross-validation (80%/20% for training/validation).

Results

The top predictors glomerular filtration rate, lactate dehydrogenase, alanine aminotransferase, aspartate C-reactive protein. Longitudinal yielded marked improvement prediction accuracy over individual points. inclusion comorbidities further improves accuracy. best an area under curve, accuracy, sensitivity, specificity be 0.965 ± 0.003, 89.57 1.64%, 0.95 0.03, 0.84 0.05, respectively, dataset, 0.86 0.01, 83.66 2.53%, 0.66 0.10, 0.89 dataset.

Conclusion

accurately predicted approach could help heighten awareness complications identify early interventions prevent long-term renal complications.

Language: Английский

Incidence of New-Onset Hypertension Post–COVID-19: Comparison With Influenza DOI
Vincent Zhang, Molly Fisher, Wei Hou

et al.

Hypertension, Journal Year: 2023, Volume and Issue: 80(10), P. 2135 - 2148

Published: Aug. 21, 2023

SARS-CoV-2 may trigger new-onset persistent hypertension. This study investigated the incidence and risk factors associated with hypertension during COVID-19 hospitalization at ≈6-month follow-up compared influenza.This retrospective observational was conducted in a major academic health system New York City. Participants included 45 398 patients (March 2020 to August 2022) 13 864 influenza (January 2018 without history of hypertension.At 6-month follow-up, seen 20.6% hospitalized 10.85% nonhospitalized COVID-19. Persistent among did not vary across pandemic, whereas that decreased from 20% March ≈10% October (R2=0.79, P=0.003) then plateaued thereafter. Hospitalized were 2.23 ([95% CI, 1.48-3.54]; P<0.001) times 1.52 1.22-1.90]; P<0.01) more likely develop than counterparts. common older adults, males, Black, preexisting comorbidities (chronic obstructive pulmonary disease, coronary artery chronic kidney disease), those who treated pressor corticosteroid medications. Mathematical models predicted 79% 86% accuracy. In addition, 21.0% no prior developed hospitalization.Incidence is higher influenza, constituting burden given sheer number Screening at-risk for following illness be warranted.

Language: Английский

Citations

55

Incidence of new-onset in-hospital and persistent diabetes in COVID-19 patients: comparison with influenza DOI Creative Commons
Justin Y. Lu,

Jack Wilson,

Wei Hou

et al.

EBioMedicine, Journal Year: 2023, Volume and Issue: 90, P. 104487 - 104487

Published: Feb. 28, 2023

Language: Английский

Citations

45

Long term outcomes of patients with chronic kidney disease after COVID-19 in an urban population in the Bronx DOI Creative Commons
Jimmy Lu, Justin Y. Lu, Stephen Wang

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 19, 2025

Abstract We investigated the long-term kidney and cardiovascular outcomes of patients with chronic disease (CKD) after COVID-19. Our retrospective cohort consisted 834 CKD COVID-19 6,167 without between 3/11/2020 to 7/1/2023. Multivariate competing risk regression models were used estimate (as adjusted hazard ratios (aHR) 95% confidence intervals (CI)) progression a more advanced stage (Stage 4 or 5) major adverse events (MAKE), (MACE) at 6-, 12-, 24-month follow up. Hospitalized 12 24 months (aHR 1.62 CI[1.24,2.13] 1.76 [1.30, 2.40], respectively), but not non-hospitalized patients, higher compared those Both hospitalized MAKE 12- 24-months 1.73 [1.21, 2.50], 1.77 [1.34, 2.33], 1.31 [1.05, 1.64], MACE increases in CKD. These findings highlight need for close up care therapies that slow this high-risk subgroup.

Language: Английский

Citations

3

Acute kidney injury in critically ill COVID-19 patients in a tertiary hospital: short and long-term kidney and patient outcomes DOI
Juliana Alves Manhães de Andrade, Gisele Meinerz, Raphael Palma

et al.

Brazilian Journal of Nephrology, Journal Year: 2025, Volume and Issue: 47(1)

Published: Jan. 10, 2025

Abstract Introduction: Acute kidney injury (AKI) in the setting of COVID-19 is associated with worse clinical and renal outcomes, limited long-term data. Aim: To evaluate critically ill patients AKI that required nephrologist consultation (NC-AKI) a tertiary hospital. Methods: Prospective single-center cohort adult NC-AKI from May 1st, 2020, to April 30th, 2021. Kidney replacement therapy (KRT), recovery function, death at 90-day 1-year follow-up were evaluated. Results: 360 included, 60.6% male, median age was 66.0 (57.0–72.0) years, 38.1% had diabetes, 68.6% hypertension. stages 1, 2, 3 detected 3.6%, 5.6%, 90.8% patients, respectively. KRT indicated 90% patients. At follow-up, 88.1% died 10.0% recovered function. Female gender (p = 0.047), older stage 0.005), requirement < 0.0001), mechanical ventilation superimposed bacterial infection 0.0001) significantly within 90 days. 1 year, mortality 89.3%. Amongst surviving 72% although lower eGFR compared baseline (85.5 ± 23.6 vs. 65.9 24.8 mL/min, p 0.003). Conclusion: Critically presented high frequency requirement, mortality. Surviving rates one-year baseline.

Language: Английский

Citations

2

Clinical predictors of acute cardiac injury and normalization of troponin after hospital discharge from COVID-19 DOI Creative Commons

Joyce Q. Lu,

Justin Y. Lu, Weihao Wang

et al.

EBioMedicine, Journal Year: 2022, Volume and Issue: 76, P. 103821 - 103821

Published: Feb. 1, 2022

Although acute cardiac injury (ACI) is a known COVID-19 complication, whether ACI acquired during recovers unknown. This study investigated the incidence of persistent and identified clinical predictors recovery in hospitalized patients with 2.5 months post-discharge.

Language: Английский

Citations

60

Long-term outcomes of COVID-19 survivors with hospital AKI: association with time to recovery from AKI DOI Open Access
Justin Y. Lu,

Montek Singh Boparai,

Caroline Shi

et al.

Nephrology Dialysis Transplantation, Journal Year: 2023, Volume and Issue: 38(10), P. 2160 - 2169

Published: Jan. 24, 2023

Although coronavirus disease 2019 (COVID-19) patients who develop in-hospital acute kidney injury (AKI) have worse short-term outcomes, their long-term outcomes not been fully characterized. We investigated 90-day and 1-year after hospital AKI grouped by time to recovery from AKI.This study consisted of 3296 COVID-19 with stratified early (<48 hours), delayed (2-7 days) prolonged (>7-90 days). Demographics, comorbidities laboratory values were obtained at admission up the follow-up. The incidence major adverse cardiovascular events (MACE) (MAKE), rehospitalization, recurrent new-onset chronic (CKD) 90-days discharge.The was 28.6%. Of AKI, 58.0% experienced recovery, 14.8% 27.1% recovery. Patients a longer had higher prevalence CKD (P < .05) more likely need invasive mechanical ventilation .001) die .001). Many developed MAKE, within 90 days, these incidences in group .05). MACE peaked 20-40 days postdischarge, whereas MAKE 80-90 postdischarge. Logistic regression models predicted 82.4 ± 1.6% 79.6 2.3% accuracy, respectively.COVID-19 survivors are high risk for especially those times history CKD. These may require follow-up cardiac complications.

Language: Английский

Citations

38

Long-term outcomes of hospitalized patients with SARS-CoV-2/COVID-19 with and without neurological involvement: 3-year follow-up assessment DOI Creative Commons

Anna Eligulashvili,

Moshe Gordon,

Jimmy S. Lee

et al.

PLoS Medicine, Journal Year: 2024, Volume and Issue: 21(4), P. e1004263 - e1004263

Published: April 4, 2024

Background Acute neurological manifestation is a common complication of acute Coronavirus Disease 2019 (COVID-19) disease. This retrospective cohort study investigated the 3-year outcomes patients with and without significant manifestations during initial COVID-19 hospitalization. Methods findings Patients hospitalized for Severe Respiratory Syndrome 2 (SARS-CoV-2) infection between 03/01/2020 4/16/2020 in Montefiore Health System Bronx, an epicenter early pandemic, were included. Follow-up data was captured up to 01/23/2023 (3 years post-COVID-19). consisted 414 1,199 propensity-matched (for age severity score) manifestations. Neurological involvement phase included stroke, new or recrudescent seizures, anatomic brain lesions, presence altered mentation evidence impaired cognition arousal, neuro-COVID-19 complex (headache, anosmia, ageusia, chemesthesis, vertigo, presyncope, paresthesias, cranial nerve abnormalities, ataxia, dysautonomia, skeletal muscle injury normal orientation arousal signs). There no group differences female sex composition (44.93% versus 48.21%, p = 0.249), ICU IMV status, white, not Hispanic (6.52% 7.84%, 0.380), (33.57% 38.20%, 0.093), except black non-Hispanic (42.51% 36.03%, 0.019). Primary mortality, heart attack, major adverse cardiovascular events (MACE), reinfection, hospital readmission post-discharge. Secondary neuroimaging (hemorrhage, active prior mass effect, microhemorrhages, white matter changes, microvascular disease (MVD), volume loss). More discharged rehabilitation (10.39% 3.34%, < 0.001) skilled nursing facilities (35.75% 25.35%, fewer home (50.24% 66.64%, than matched controls. Incidence any reason (65.70% 60.72%, 0.036), stroke (6.28% 2.34%, 0.001), MACE (20.53% 16.51%, 0.032) higher Per Kaplan–Meier univariate survival curve analysis, such more likely die post-discharge compared controls (hazard ratio: 2.346, (95% confidence interval (CI) [1.586, 3.470]; 0.001)). Across both cohorts, causes death (13.79% neurological, 15.38% control), sepsis (8.63%, 17.58%), influenza pneumonia (13.79%, 9.89%), (10.34%, 7.69%), respiratory distress syndrome (ARDS) 6.59%). Factors associated mortality after leaving involved (odds ratio (OR): 1.802 CI [1.237, 2.608]; 0.002)), discharge disposition (OR: 1.508 [1.276, 1.775]; 0.001)), congestive failure 2.281 [1.429, 3.593]; score 1.177 [1.062, 1.304]; older 1.027 [1.010, 1.044]; 0.002)). radiological findings, that showed significantly age-adjusted loss ( 0.045) The study’s patient limited infected first wave when hospitals overburdened, vaccines yet available, treatments limited. Patient profiles might differ interrogating subsequent waves. Conclusions had worse long-term These raise awareness need closer monitoring timely interventions manifestations, as their course involving enhanced morbidity mortality.

Language: Английский

Citations

11

SARS-CoV-2 infection increases long-term multiple sclerosis disease activity and all-cause mortality in an underserved inner-city population DOI
Roham Hadidchi, Stephen Wang,

David Rezko

et al.

Multiple Sclerosis and Related Disorders, Journal Year: 2024, Volume and Issue: 86, P. 105613 - 105613

Published: April 8, 2024

Language: Английский

Citations

11

Long-term outcomes of patients with a pre-existing neurological condition after SARS-CoV-2 infection DOI
Roham Hadidchi,

Yousef Al‐Ani,

Solbie Choi

et al.

Journal of the Neurological Sciences, Journal Year: 2025, Volume and Issue: unknown, P. 123477 - 123477

Published: April 1, 2025

Language: Английский

Citations

1

Outcomes of Hospitalized Patients With COVID-19 With Acute Kidney Injury and Acute Cardiac Injury DOI Creative Commons
Justin Y. Lu,

Alexandra Buczek,

Roman Fleysher

et al.

Frontiers in Cardiovascular Medicine, Journal Year: 2022, Volume and Issue: 8

Published: Feb. 15, 2022

This study investigated the incidence, disease course, risk factors, and mortality in COVID-19 patients who developed both acute kidney injury (AKI) cardiac (ACI), compared to those with AKI only, ACI no (NI).This retrospective consisted of hospitalized at Montefiore Health System Bronx, New York between March 11, 2020 January 29, 2021. Demographics, comorbidities, vitals, laboratory tests were collected during hospitalization. Predictive models used predict AKI, ACI, AKI-ACI onset. Longitudinal analyzed time-lock discharge alive or death.Of 5,896 patients, 44, 19, 9, 28% had NI, AKI-ACI, respectively. Most presented very early (within a day two) hospitalization contrast (p < 0.05). Patients combined significantly older, more often men higher levels cardiac, kidney, liver, inflammatory, immunological markers NI groups. The adjusted hospital-mortality odds ratios 17.1 [95% CI = 13.6-21.7, p 0.001], 7.2 5.4-9.6, 4.7 3.7-6.1, 0.001] for respectively, relative NI. A predictive model onset using top predictors yielded 97% accuracy. data predicted up 5 days prior outcome, an area-under-the-curve, ranging from 0.68 0.89.COVID-19 markedly worse outcomes only Common variables accurately AKI-ACI. ability identify could lead earlier intervention improvement clinical outcomes.

Language: Английский

Citations

30