Kashyap, Rahul

Background There are limited data on unconscious bias in healthcare, but there is consistent evidence that it alters clinical decision-making. COVID-19 exacerbated many pre-existing disparities, and this paper seeks to identify, deconstruct, and propose mitigation strategies for a few of them. Discussion Five of the largest disparities amplified by the pandemic are discussed in this paper. Older people, Black people, uninsured people, rural communities, and people with lower education levels have been disproportionally affected in both morbidity and mortality. Conclusions The disparities discussed above did not occur in a vacuum but are the result of systemic issues. Equity starts with understanding and addressing the root cause, and it can be worked toward with practical and impactful solutions.

Objective: Artificial intelligence and machine learning (AI/ML) are increasingly being applied to big clinical data to tackle research questions that cannot be answered with traditional statistical methods. The field is still in its nascent stages and there is a paucity of guidelines for conducting and reporting AI/ML research in hypertension. The objective was to apply the HUMANE checklist to survey the present landscape of AI/ML in hypertension to inform the development of hypertension-specific guidelines and recommendations. Design and method: The HUMANE checklist was developed by global clinical and AI/ML experts through the Delphi method. It assesses the quality of medical AI/ML articles based on whether they cover subjects expected in any peer-reviewed, clinical or AI/ML research publication. A cooping review was carried out to identify articles presenting original research in AI/ML and hypertension published in 2019–2021. Two independent reviewers applied the checklist to each article and in the case of discordance, the response was adjudicated by an AI/ML expert. Results were analysed to assess compliance with the survey (% of papers satisfying checklist requirements). Results: A total of 63 manuscripts was reviewed. A summary of results is shown in Figure 1. Highest compliance was seen for items relating to general article presentation, with compliance ranging from 68% to 98% (description of statistical analysis methods and background context, respectively). Lowest compliance was seen with checklist items relating to clinical research and AI/ML methods. 44% of reviewed articles described the demographics of their dataset and 48% stated their inclusion/exclusion criteria. Nonetheless, datasets were deemed appropriate for investigative aims in 93% of articles. 30% of manuscripts reported their calibration measures, while 73% stated their performance metrics. Internal validation was carried out in 75% of studies, but external validity was assessed in only 14% of cases. Algorithmic bias was addressed in 11% of papers. Conclusions: Application of AI/ML methods in hypertension research is growing, but the majority of current work has major shortfalls in reporting quality, model validation and algorithmic bias. Our study identifies areas of improvement to enable the full realisation of the potential of AI/ML in hypertension.

Objectives: Peripherally inserted central catheters (PICCs) are increasingly recognized as an alternative to centrally inserted central catheters (CICCs) in critical care, yet the data regarding the safety and feasibility of this choice in septic shock management is growing but still lacking. In this study, we aimed to determine the feasibility, safety, and impact on outcomes of using dedicated vascular access specialist (VAS) teams to insert PICCs versus CICCs on patients admitted to the ICU with septic shock. Design: Retrospective cohort study. Setting: Mayo Clinic Rochester Medical ICU and Mayo Clinic Arizona Multidisciplinary ICU from 2013 to 2016. Patients: All adult patients hospitalized with diagnosis of septic shock excluding patients who declined authorization for review of their medical records, mixed shock states, and readmissions. Interventions: None. Measurement and main results: Comprehensive data regarding septic shock diagnosis and resuscitation were abstracted from electronic medical records. A total of 562 patients with septic shock were included in the study; 215 patients were resuscitated utilizing a PICC and 347 were resuscitated using a CICC. On univariate analysis, the time to central line insertion and time to vasopressor initiation were found to be reduced in those who received PICC at time of ICU admission versus CICC. Other favorable outcomes were also observed in those who received PICC versus CICC including shorter ICU length of stay and lower unadjusted hospital mortality. A multivariable analysis for hospital mortality showed that after adjusting for important covariates, neither the time to central line insertion nor the time to vasopressor initiation was associated with a lower hospital mortality. Conclusions: Across two tertiary referral centers within the same enterprise, use of a dedicated VAS team for insertion of PICCs for initial resuscitation in patients with septic shock was feasible and associated with shorter time to central venous access and initiation of vasopressors; however, adjusted hospital mortality was not different between the two groups.

BACKGROUND: The use of high-flow nasal cannula (HFNC) and noninvasive ventilation (NIV) for hypoxemic respiratory failure secondary to COVID-19 are recommended by critical-care guidelines; however, apprehension about viral particle aerosolization and patient self-inflicted lung injury may have limited use. We aimed to describe hospital variation in the use and clinical outcomes of HFNC and NIV for the management of COVID-19. METHODS: This was a retrospective observational study of adults hospitalized with COVID-19 who received supplemental oxygen between February 15, 2020, and April 12, 2021, across 102 international and United States hospitals by using the COVID-19 Registry. Associations of HFNC and NIV use with clinical outcomes were evaluated by using multivariable adjusted hierarchical random-effects logistic regression models. Hospital variation was characterized by using intraclass correlation and the median odds ratio. RESULTS: Among 13,454 adults with COVID-19 who received supplemental oxygen, 8,143 (60%) received nasal cannula/face mask only, 2,859 (21%) received HFNC, 878 (7%) received NIV, 1,574 (12%) received both HFNC and NIV, with 3,640 subjects (27%) progressing to invasive ventilation. The hospital of admission contributed to 24% of the risk-adjusted variation in HFNC and 30% of the risk-adjusted variation in NIV. The median odds ratio for hospital variation of HFNC was 2.6 (95% CI 1.4–4.9) and of NIV was 3.1 (95% CI 1.2-8.1). Among 5,311 subjects who received HFNC and/or NIV, 2,772 (52%) did not receive invasive ventilation and survived to hospital discharge. Hospital-level use of HFNC or NIV were not associated with the rates of invasive ventilation or mortality. CONCLUSIONS: Hospital variation in the use of HFNC and NIV for acute respiratory failure secondary to COVID-19 was great but was not associated with intubation or mortality. The wide variation and relatively low use of HFNC/NIV observed within our study signaled that implementation of increased HFNC/NIV use in patients with COVID-19 will require changes to current care delivery practices. (ClinicalTrials.gov registration NCT04323787.)

AbstractObjectivesSuggested therapeutic options for Multisystem Inflammatory Syndrome in Children (MIS‐C) include intravenous immunoglobulins (IVIG) and steroids. Prior studies have shown the benefit of combination therapy with both agents on fever control or the resolution of organ dysfunction. The primary objective of this study was to analyze the impact of IVIG and steroids on hospital and ICU length of stay (LOS) in patients with MIS‐C associated with Coronavirus Disease 2019 (COVID‐19).Study DesignThis was a retrospective study on 356 hospitalized patients with MIS‐C from March 2020 to September 2021 (28 sites in the United States) in the Society of Critical Care Medicine (SCCM) Discovery Viral Infection and Respiratory Illness Universal Study (VIRUS) COVID‐19 Registry. The effect of IVIG and steroids initiated in the first 2 days of admission, alone or in combination, on LOS was analyzed. Adjustment for confounders was made by multivariable mixed regression with a random intercept for the site.ResultsThe median age of the study population was 8.8 (Interquartile range (IQR) 4.0, 13) years. 247/356 (69%) patients required intensive care unit (ICU) admission during hospitalization. Overall hospital mortality was 2% (7/356). Of the total patients, 153 (43%) received IVIG and steroids, 33 (9%) received IVIG only, 43 (12%) received steroids only, and 127 (36%) received neither within 2 days of admission. After adjustment of confounders, only combination therapy showed a significant decrease of ICU LOS by 1.6 days compared to no therapy (exponentiated coefficient 0.71 [95% confidence interval 0.51, 0.97, p = 0.03]). No significant difference was observed in hospital LOS or the secondary outcome variable of the normalization of inflammatory mediators by Day 3.ConclusionsCombination therapy with IVIG and steroids initiated in the first 2 days of admission favorably impacts ICU but not the overall hospital LOS in children with MIS‐C.

Background: Describe the incidence and associated outcomes of gastrointestinal (GI) manifestations of acute coronavirus disease 2019 (COVID-19) and multisystem inflammatory syndrome in hospitalized children (MIS-C). Methods: Retrospective review of the Viral Infection and Respiratory Illness Universal Study registry, a prospective observational, multicenter international cohort study of hospitalized children with acute COVID-19 or MIS-C from March 2020 to November 2020. The primary outcome measure was critical COVID-19 illness. Multivariable models were performed to assess for associations of GI involvement with the primary composite outcome in the entire cohort and a subpopulation of patients with MIS-C. Secondary outcomes included prolonged hospital length of stay defined as being >75th percentile and mortality. Results: Of the 789 patients, GI involvement was present in 500 (63.3%). Critical illness occurred in 392 (49.6%), and 18 (2.3%) died. Those with GI involvement were older (median age of 8 yr), and 18.2% had an underlying GI comorbidity. GI symptoms and liver derangements were more common among patients with MIS-C. In the adjusted multivariable models, acute COVID-19 was no associated with the primary or secondary outcomes. Similarly, despite the preponderance of GI involvement in patients with MIS-C, it was also not associated with the primary or secondary outcomes. Conclusions: GI involvement is common in hospitalized children with acute COVID-19 and MIS-C. GI involvement is not associated with critical illness, hospital length of stay or mortality in acute COVID-19 or MIS-C.

Abstract ObjectivePrevious studies suggested that hypernatremia or hyperosmolarity may have protective effects in lung injury. We hypothesized that hypernatremia and/or hyperosmolarity would prevent ARDS in those at risk of lung injury and associated with favorable outcomes in ARDS patients.SubjectsAll admissions at medical, surgical, and multidisciplinary intensive care units in Mayo Clinic, Rochester from the year of 2009 to 2019.DesignRetrospective cohort study. The occurrence of ARDS was identified using a validated computerized search strategy. The association between serum sodium/osmolarity and the occurrence of ARDS was analyzed using a multivariable logistic regression model. The relationship between serum sodium/osmolarity and outcomes of ARDS was analyzed using linear and logistic regression models.ResultsAmong 50,498 patients, the serum sodium level on admission did not have a significant association with the occurrence of ARDS, with an adjusted odds ratio of 0.95 [95% CI (0.86, 1.05)]. There was no significant association between calculated serum osmolarity and the occurrence of ARDS, with an adjusted odds ratio of 1.03 [95% CI (1.00, 1.07)]. 1560 patients developed ARDS during the ICU stay. Their serum sodium level did not have a significant association with their outcomes [adjusted estimated difference in 28-day ICU-free days, 0.08 days, 95% CI (-0.85, 1.02), adjusted estimated difference in 28-day ventilator-free days, -0.02, 95% CI (-0.62, 0.58), adjusted odds ratio of 28-day mortality 1.02, 95%CI (0.84, 1.24)]. There was no significant association between their calculated serum osmolarity and their outcomes [adjusted estimated difference in 28-day ICU-free days, -0.26, 95% CI, (-0.63, 0.12); adjusted estimated difference in 28-day ventilator-free days, -0.05, 95% CI (-0.28, 0.19); adjusted odds ratio of 28-day mortality, 1.07, 95% CI (0.99, 1.15)].ConclusionsAdmission serum sodium or serum osmolarity were not associated with the occurrence or outcomes of ARDS in ICU.

Abstract Background Preoperative risk scores facilitate patient selection, but postoperative risk scores may offer valuable information for predicting outcomes. We hypothesized that the postoperative Sequential Organ Failure Assessment (SOFA) score would predict mortality after left ventricular assist device (LVAD) implantation. Methods We retrospectively reviewed data from 294 continuous‐flow LVAD implantations performed at Mayo Clinic Rochester during 2007 to 2015. We calculated the EuroSCORE, HeartMate‐II Risk Score, and RV Failure Risk Score from preoperative data and the APACHE III and Post Cardiac Surgery (POCAS) risk scores from postoperative data. Daily, maximum, and mean SOFA scores were calculated for the first 5 postoperative days. The area under receiver‐operator characteristic curves (AUC) was calculated to compare the scoring systems' ability to predict 30‐day, 90‐day, and 1‐year mortality. Results For the entire cohort, mortality was 5% at 30 days, 10% at 90 days, and 19% at 1 year. The Day 1 SOFA score had better discrimination for 30‐day mortality (AUC 0.77) than the preoperative risk scores or the APACHE III and POCAS postoperative scores. The maximum SOFA score had the best discrimination for 30‐day mortality (AUC 0.86), and the mean SOFA score had the best discrimination for 90‐day mortality (AUC 0.82) and 1‐year mortality (AUC 0.76). Conclusions We observed that postoperative mean and maximum SOFA scores in LVAD recipients predict short‐term and intermediate‐term mortality better than preoperative risk scores do. However, because preoperative and postoperative risk scores each contribute unique information, they are best used in concert to predict outcomes after LVAD implantation.

AbstractIntroductionCoronavirus disease 2019 (COVID‐19) is associated with high rates of morbidity and mortality. Primary hypothyroidism is a common comorbid condition, but little is known about its association with COVID‐19 severity and outcomes. This study aims to identify the frequency of hypothyroidism in hospitalized patients with COVID‐19 as well as describe the differences in outcomes between patients with and without pre‐existing hypothyroidism using an observational, multinational registry.MethodsIn an observational cohort study we enrolled patients 18 years or older, with laboratory‐confirmed severe acute respiratory syndrome coronavirus‐2 infection between March 2020 and February 2021. The primary outcomes were (1) the disease severity defined as per the World Health Organization Scale for Clinical Improvement, which is an ordinal outcome corresponding with the highest severity level recorded during a patient's index COVID‐19 hospitalization, (2) in‐hospital mortality and (3) hospital‐free days. Secondary outcomes were the rate of intensive care unit (ICU) admission and ICU mortality.ResultsAmong the 20,366 adult patients included in the study, pre‐existing hypothyroidism was identified in 1616 (7.9%). The median age for the Hypothyroidism group was 70 (interquartile range: 59–80) years, and 65% were female and 67% were White. The most common comorbidities were hypertension (68%), diabetes (42%), dyslipidemia (37%) and obesity (28%). After adjusting for age, body mass index, sex, admission date in the quarter year since March 2020, race, smoking history and other comorbid conditions (coronary artery disease, hypertension, diabetes and dyslipidemia), pre‐existing hypothyroidism was not associated with higher odds of severe disease using the World Health Organization disease severity index (odds ratio [OR]: 1.02; 95% confidence interval [CI]: 0.92, 1.13; p = .69), in‐hospital mortality (OR: 1.03; 95% CI: 0.92, 1.15; p = .58) or differences in hospital‐free days (estimated difference 0.01 days; 95% CI: −0.45, 0.47; p = .97). Pre‐existing hypothyroidism was not associated with ICU admission or ICU mortality in unadjusted as well as in adjusted analysis.ConclusionsIn an international registry, hypothyroidism was identified in around 1 of every 12 adult hospitalized patients with COVID‐19. Pre‐existing hypothyroidism in hospitalized patients with COVID‐19 was not associated with higher disease severity or increased risk of mortality or ICU admissions. However, more research on the possible effects of COVID‐19 on the thyroid gland and its function is needed in the future.