Non-invasive liver disease assessment to identify portal hypertension: A systematic review supporting the AASLD Practice Guideline.

INTRODUCTION: Portal hypertension is a serious complication of cirrhosis, which leads to life-threatening complications. Hepatic venous pressure gradient (HVPG), a surrogate of portal pressure, is the reference standard test to assess the severity of portal hypertension. However, since HVPG is limited by its invasiveness and by its availability, non-invasive liver disease assessments (NILDAs) to assess portal pressure, especially clinically significant portal hypertension (CSPH), are needed. METHODS: We conducted a systematic review of Ovid MEDLINE(R) and Epub Ahead of Print, In-Process & Other Non-Indexed Citations, and Daily, Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, and Scopus from each database's inception to April 22 nd , 2022. We included only studies in English that examined ≥50 patients in single liver disease etiologies which compared non-invasive tests (blood, and/or imaging) to HVPG for predicting clinically significant portal hypertension (CSPH; defined as HVPG ≥10 mm Hg) in patients with chronic liver disease (this therefore limited the number of studies that could be included). Outcomes reported included measures of diagnostic test accuracy. Additionally, a narrative review of studies not eligible for the systematic review is also provided. RESULTS: Nine studies with 2,492 patients met the inclusion criteria. There was substantial heterogeneity with regard to liver disease studied and cutoff values used to detect CSPH. Blood based tests, including aspartate to platelet ratio index (APRI) (56% sensitivity and 68% specificity) and fibrosis-4 (FIB-4) (54% sensitivity and 73% specificity) had low accuracy measures. Imaging based tests (transient elastography (TE) and shear wave elastography detection of liver stiffness (LSM)) had better accuracy, but also had substantial variation; at 15 kPa, TE sensitivity was 90%-96% and specificity was 48%-50% while at 25 kPa, its sensitivity and specificity were 57%-85% and 82%-93%, respectively. The narrative review suggested that imaging based tests are the best available NILDA to detect CSPH, CSPH is highly unlikely to be present at an LSM ≤15 kPa and likely to be present at an LSM ≥25 kPa. CONCLUSION: While imaging-based NILDA appeared to have higher accuracy than blood-based tests to detect CSPH, only 9 studies fit the a priori established inclusion criteria for the SR. In addition, there was substantial study heterogeneity and variation in cutoffs for LSM to detect CSPH, limiting the ability to establish definitive cutoffs to detect CSPH.

Imaging-based non-invasive liver disease assessment for staging liver fibrosis in chronic liver disease: A systematic review supporting the AASLD Practice Guideline.

BACKGROUND AND AIMS: Transient elastography (TE), shear-wave elastography (SWE), and/or magnetic resonance elastography (MRE), each providing liver stiffness measurement (LSM), are the most studied imaging-based noninvasive liver disease assessment (NILDA) techniques. To support the American Association for the Study of Liver Diseases guidelines on NILDA, we summarized the evidence on the accuracy of these LSM methods to stage liver fibrosis (F). APPROACH AND RESULTS: A comprehensive search for studies assessing LSM by TE, SWE, or MRE for the identification of significant fibrosis (F2-4), advanced fibrosis (F3-4), or cirrhosis (F4), utilizing histopathology as standard of reference by liver disease etiology in adults or children from inception to April 2022 was performed. We excluded studies with <50 patients with a single disease entity and mixed liver disease etiologies (with the exception of HCV/HIV co-infection). Out of 9447 studies, 240 with 61,193 patients were included in this systematic review. In adults, sensitivities for the identification of F2-4 ranged from 51% to 95%, for F3-4 from 70% to 100%, and for F4 from 60% to 100% across all techniques/diseases, whereas specificities ranged from 36% to 100%, 74% to 100%, and 67% to 99%, respectively. The largest body of evidence available was for TE; MRE appeared to be the most accurate method. Imaging-based NILDA outperformed blood-based NILDA in most comparisons, particularly for the identification of F3-4/F4. In the pediatric population, imaging-based NILDA is likely as accurate as in adults. CONCLUSION: LSM from TE, SWE, and MRE show acceptable to outstanding accuracy for the detection of liver fibrosis across various liver disease etiologies. Accuracy increased from F2-4 to F3-4 and was the highest for F4. Further research is needed to better standardize the use of imaging-based NILDA, particularly in pediatric liver diseases.

Accuracy of blood-based biomarkers for staging liver fibrosis in chronic liver disease: A systematic review supporting the AASLD Practice Guideline.

BACKGROUND AND AIMS: Blood-based biomarkers have been proposed as an alternative to liver biopsy for non-invasive liver disease assessment (NILDA) in chronic liver disease (CLD). Our aims for this systematic review were to evaluate the diagnostic utility of selected blood-based tests either alone, or in combination, for identifying significant fibrosis (F2-4), advanced fibrosis (F3-4) and cirrhosis (F4), as compared to biopsy in CLD. APPROACH AND RESULTS: We included a comprehensive search of databases including Ovid MEDLINE(R), EMBASE, Cochrane Database, and Scopus through to April 2022. Two independent reviewers selected 286 studies with 103,162 patients. The most frequently identified studies included the simple aminotransferase-to-platelet ratio index (APRI) and fibrosis (FIB)-4 markers (with low-to-moderate risk of bias) in hepatitis B virus (HBV) and C virus (HCV), HIV-HCV/HBV co-infection, and nonalcoholic fatty liver disease (NAFLD). Positive (LR+) and negative (LR) likelihood ratios across direct and indirect biomarker tests for HCV and HBV for F2-4, F3-4, or F4 were 1.66-6.25 and 0.23-0.80, 1.89-5.24 and 0.12-0.64, and 1.32-7.15 and 0.15-0.86 respectively; LR+ and LR for NAFLD F2-4, F3-4 and F4 were 2-65-3.37 and 0.37-0.39, 2.25-6.76 and 0.07-0.87, and 3.90 and 0.15 respectively. Overall, proportional odds ratio indicated FIB-4 <1.45 was better than APRI <0.5 for F2-4. FIB-4 >3.25 was also better than APRI >1.5 for F3-4 and F4. There was limited data for combined tests. CONCLUSIONS: Blood-based biomarkers are associated with small-to-moderate change in pre-test probability for diagnosing F2-4, F3-4, and F4 in viral hepatitis, HIV-HCV co-infection, and NAFLD, with limited comparative or combination studies for other CLD.

Integrating large language models in systematic reviews: a framework and case study using ROBINS-I for risk of bias assessment.

Large language models (LLMs) may facilitate and expedite systematic reviews, although the approach to integrate LLMs in the review process is unclear. This study evaluates GPT-4 agreement with human reviewers in assessing the risk of bias using the Risk Of Bias In Non-randomised Studies of Interventions (ROBINS-I) tool and proposes a framework for integrating LLMs into systematic reviews. The case study demonstrated that raw per cent agreement was the highest for the ROBINS-I domain of 'Classification of Intervention'. Kendall agreement coefficient was highest for the domains of 'Participant Selection', 'Missing Data' and 'Measurement of Outcomes', suggesting moderate agreement in these domains. Raw agreement about the overall risk of bias across domains was 61% (Kendall coefficient=0.35). The proposed framework for integrating LLMs into systematic reviews consists of four domains: rationale for LLM use, protocol (task definition, model selection, prompt engineering, data entry methods, human role and success metrics), execution (iterative revisions to the protocol) and reporting. We identify five basic task types relevant to systematic reviews: selection, extraction, judgement, analysis and narration. Considering the agreement level with a human reviewer in the case study, pairing artificial intelligence with an independent human reviewer remains required.

The association of sensitivity and specificity with disease prevalence: analysis of 6909 studies of diagnostic test accuracy.

BACKGROUND: Sensitivity and specificity are characteristics of a diagnostic test and are not expected to change as the prevalence of the target condition changes. We sought to evaluate the association between prevalence and changes in sensitivity and specificity. METHODS: We retrieved data from meta-analyses of diagnostic test accuracy published in the Cochrane Database of Systematic Reviews (2003-2020). We used mixed-effects random-intercept linear regression models to evaluate the association between prevalence and logit-transformed sensitivity and specificity. The model evaluated all meta-analyses as nested within each systematic review. RESULTS: We analyzed 6909 diagnostic test accuracy studies from 552 meta-analyses that were included in 92 systematic reviews. For sensitivity, compared with the lowest quartile of prevalence, the second, third and fourth quartiles were associated with significantly higher odds of identifying a true positive case (odds ratio [OR] 1.17, 95% confidence interval [CI] 1.09-1.26; OR 1.32, 95% CI 1.23-1.41; OR 1.47, 95% CI 1.37-1.58; respectively). For specificity, compared with the lowest quartile of prevalence, the second, third and fourth quartiles were associated with significantly lower odds of identifying a true negative case (OR 0.74, 95% CI 0.69-0.80; OR 0.65, 95% CI 0.60-0.70; OR 0.47, 95% CI 0.44-0.51; respectively). Pooled regression coefficients from bivariate models conducted within each meta-analysis showed that prevalence was positively associated with sensitivity and negatively associated with specificity. Findings were consistent across subgroups. INTERPRETATION: In this large sample of diagnostic studies, higher prevalence was associated with higher estimated sensitivity and lower estimated specificity. Clinicians should consider the implications of disease prevalence and spectrum when interpreting the results from studies of diagnostic test accuracy.

Partial breast irradiation compared with whole breast irradiation: a systematic review and meta-analysis.

BACKGROUND: Early-stage breast cancer is among the most common cancer diagnoses. Adjuvant radiotherapy is an essential component of breast-conserving therapy, and several options exist for tailoring its extent and duration. This study assesses the comparative effectiveness of partial-breast irradiation (PBI) compared with whole-breast irradiation (WBI). METHODS: A systematic review was completed to identify relevant randomized clinical trials and comparative observational studies. Independent reviewers working in pairs selected studies and extracted data. Randomized trial results were pooled using a random effects model. Prespecified main outcomes were ipsilateral breast recurrence (IBR), cosmesis, and adverse events (AEs). RESULTS: Fourteen randomized clinical trials and 6 comparative observational studies with 17 234 patients evaluated the comparative effectiveness of PBI. PBI was not statistically significantly different from WBI for IBR at 5 years (RR = 1.34, 95% CI = 0.83 to 2.18; high strength of evidence [SOE]) and 10 years (RR = 1.29, 95% CI = 0.87 to 1.91; high SOE). Evidence for cosmetic outcomes was insufficient. Statistically significantly fewer acute AEs were reported with PBI compared with WBI, with no statistically significant difference in late AEs. Data from subgroups according to patient, tumor, and treatment characteristics were insufficient. Intraoperative radiotherapy was associated with higher IBR at 5, 10, and over than 10 years (high SOE) compared with WBI. CONCLUSIONS: Ipsilateral breast recurrence was not statistically significantly different between PBI and WBI. Acute AEs were less frequent with PBI. This evidence supports the effectiveness of PBI among selected patients with early-stage, favorable-risk breast cancer who are similar to those represented in the included studies.

A systematic review supporting the Society for Vascular Surgery guidelines on the management of heritable aortopathies.

BACKGROUND: To support the development of clinical practice guidelines on the management of patients with genetic aortopathies and arteriopathies, a writing committee from the Society for Vascular Surgery has commissioned this systematic review. METHODS: We conducted a systematic review and searched multiple databases for studies addressing six questions identified by the Society for Vascular Surgery guideline committee about evaluating and managing patients with genetic aortopathies and arteriopathies. Studies were selected and appraised by pairs of independent reviewers. RESULTS: We included 12 studies in this systematic review. We did not identify studies about the long-term outcomes of endovascular repair for aortic aneurysm in patients with heritable aortopathy or about new aortic events in pregnant women with a history of aortic dissection (AD) or aneurysm. A small case series demonstrated a 100% survival rate and 100% aortic intervention-free survival at 15 months (range, 7-28 months) after endograft repair for type B AD. A positive genetic diagnosis was discovered in 36% of patients with aortic aneurysms and dissections who had no risk factors for hereditary aortopathies, and these patients had a mortality rate of 11% at a median follow-up duration of 5 months. Black patients had lower 30-day mortality than White patients (5.6% vs 9.0%, respectively), but they had a higher overall aortic reintervention rate at 30 days after AD repair (47% vs 27%, respectively). Aortic reinterventions owing to aneurysmal expansion and endoleak at 30 days were higher in Black patients than White patients. The certainty of evidence was judged to be very low across all the outcomes evaluated in this systematic review. CONCLUSIONS: The available evidence suggests high survival after thoracic endovascular aortic repair for type B AD in young patients with heritable aortopathies, but with limited long-term follow-up. Genetic testing in patients with acute aortic aneurysms and dissections had a high yield. It was positive for most patients with risk factors for hereditary aortopathies and in more than one-third for all other patients, and was associated with new aortic events within 15 years.

The safety of intravenous peripheral administration of 3% hypertonic saline: A systematic review and meta-analysis.

BACKGROUND: Three percent hypertonic saline (3% HTS) is used to treat several critical conditions such as severe and symptomatic hyponatremia and increased intracranial pressure. It has been traditionally administered through a central venous catheter (CVC). The avoidance of peripheral intravenous infusion of 3% HTS stems theoretically from the concern about the ability of the peripheral veins to tolerate hyperosmolar infusions. The aim of this systematic review and meta-analysis is to assess the rate of complications associated with the infusion of 3% HTS using peripheral intravenous access. METHODS: We conducted a systematic review and meta-analysis to assess the rate of complications related to the peripheral infusion of 3% HTS. We searched several databases for available studies that met the criteria until February 24th, 2022. We included ten studies conducted across three countries examining the incidence of infiltration, phlebitis, venous thrombosis, erythema, and edema. The overall event rate was calculated and transformed using the Freeman-Tukey arcsine method and pooled using the DerSimonian and Laird random-effects model. I2 was used to evaluate heterogeneity. Selected items from Newcastle-Ottawa Scale2 were used to assess the risk of bias in each included study. RESULTS: A total of 1200 patients were reported to have received peripheral infusion of 3% HTS. The analysis showed that peripherally administered 3% HTS has a low rate of complications. The overall incidence of each of the complications was as follows: infiltration 3.3%, (95% C.I. = 1.8-5.1%), phlebitis 6.2% (95% C.I. = 1.1-14.3%), erythema 2.3% (95% C.I. = 0.3-5.4%), edema 1.8% (95% C.I. = 0.0-6.2%), and venous thrombosis 1% (95% C.I. = 0.0-4.8%). There was one incident of venous thrombosis preceded by infiltration resulting from peripheral infusion of 3% HTS. CONCLUSIONS: Peripheral administration of 3% HTS is considered a safe and possibly preferred option as it carries a low risk of complications and is a less invasive procedure compared to CVC.

Lack of patient-centered evaluation of outcomes in intermittent claudication literature.

BACKGROUND: Peripheral arterial disease, characterized as arterial atherosclerotic disease, can lead to insufficient flow in the lower extremities and ischemia, with the most common clinical manifestation being intermittent claudication (IC). In 2022, the Society for Vascular Surgery (SVS) developed appropriate use criteria for the management of IC that used this systematic review as a source of evidence. The objective of this study is to synthesize the findings of the systematic review and identify evidence gaps. METHODS: A comprehensive search of literature databases including MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Scopus was conducted from January 1, 2000, to November 30, 2022. Noncomparative and comparative observational studies and randomized controlled trials were included. Included studies evaluated exercise therapy, endovascular or open revascularization for the treatment of IC. Outcomes of interest (freedom from major adverse limb event, health-related quality of life, and walking distance) were compared in various subgroups (age, sex, diabetes, smoking status, anatomical location of disease, and optimal medical therapy). RESULTS: Twenty-six studies reported the outcomes of interest for the evidence map. The general conclusions of the studies that reported freedom from major adverse limb events were that reintervention rates for endovascular therapy at ≥2 years were >20%, major amputation rates were often not reported, and, after endovascular therapy, the 1-month mortality was low (<2%). Quality of life and walking distance data were sparse, limited to only endovascular intervention, and insufficient to make any strong conclusions. CONCLUSIONS: IC in patients with peripheral arterial disease poses a significant socioeconomic and health care burden. Major, consequential gaps exist in the IC literature with respect to the assessment of patient reported outcome measures, standardized measures of walking distance and the comparative effectiveness of initial exercise therapy vs invasive intervention. The evidence gaps identified by the Society for Vascular Surgery appropriate use criteria on IC systematic review serve as a guide for future research efforts to optimize care for this patient population.

Diagnostic accuracy of the physical examination in emergency department patients with acute vertigo or dizziness: A systematic review and meta-analysis for GRACE-3.

BACKGROUND: History and physical examination are key features to narrow the differential diagnosis of central versus peripheral causes in patients presenting with acute vertigo. We conducted a systematic review and meta-analysis of the diagnostic test accuracy of physical examination findings. METHODS: This study involved a patient-intervention-control-outcome (PICO) question: (P) adult ED patients with vertigo/dizziness; (I) presence/absence of specific physical examination findings; and (O) central (ischemic stroke, hemorrhage, others) versus peripheral etiology. Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) was assessed. RESULTS: From 6309 titles, 460 articles were retrieved, and 43 met the inclusion criteria: general neurologic examination-five studies, 869 patients, pooled sensitivity 46.8% (95% confidence interval [CI] 32.3%-61.9%, moderate certainty) and specificity 92.8% (95% CI 75.7%-98.1%, low certainty); limb weakness/hemiparesis-four studies, 893 patients, sensitivity 11.4% (95% CI 5.1%-23.6%, high) and specificity 98.5% (95% CI 97.1%-99.2%, high); truncal/gait ataxia-10 studies, 1810 patients (increasing severity of truncal ataxia had an increasing sensitivity for central etiology, sensitivity 69.7% [43.3%-87.9%, low] and specificity 83.7% [95% CI 52.1%-96.0%, low]); dysmetria signs-four studies, 1135 patients, sensitivity 24.6% (95% CI 15.6%-36.5%, high) and specificity 97.8% (94.4%-99.2%, high); head impulse test (HIT)-17 studies, 1366 patients, sensitivity 76.8% (64.4%-85.8%, low) and specificity 89.1% (95% CI 75.8%-95.6%, moderate); spontaneous nystagmus-six studies, 621 patients, sensitivity 52.3% (29.8%-74.0%, moderate) and specificity 42.0% (95% CI 15.5%-74.1%, moderate); nystagmus type-16 studies, 1366 patients (bidirectional, vertical, direction changing, or pure torsional nystagmus are consistent with a central cause of vertigo, sensitivity 50.7% [95% CI 41.1%-60.2%, moderate] and specificity 98.5% [95% CI 91.7%-99.7%, moderate]); test of skew-15 studies, 1150 patients (skew deviation is abnormal and consistent with central etiology, sensitivity was 23.7% [95% CI 15%-35.4%, moderate] and specificity 97.6% [95% CI 96%-98.6%, moderate]); HINTS (head impulse, nystagmus, test of skew)-14 studies, 1781 patients, sensitivity 92.9% (95% CI 79.1%-97.9%, high) and specificity 83.4% (95% CI 69.6%-91.7%, moderate); and HINTS+ (HINTS with hearing component)-five studies, 342 patients, sensitivity 99.0% (95% CI 73.6%-100%, high) and specificity 84.8% (95% CI 70.1%-93.0%, high). CONCLUSIONS: Most neurologic examination findings have low sensitivity and high specificity for a central cause in patients with acute vertigo or dizziness. In acute vestibular syndrome (monophasic, continuous, persistent dizziness), HINTS and HINTS+ have high sensitivity when performed by trained clinicians.

A Systematic Review Supporting the Endocrine Society Clinical Practice Guideline on the Treatment of Hypercalcemia of Malignancy in Adults.

CONTEXT: Hypercalcemia is a common complication of malignancy that is associated with high morbidity and mortality. OBJECTIVE: To support development of the Endocrine Society Clinical Practice Guideline for the treatment of hypercalcemia of malignancy in adults. METHODS: We searched multiple databases for studies that addressed 8 clinical questions prioritized by a guideline panel from the Endocrine Society. Quantitative and qualitative synthesis was performed. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach was used to assess certainty of evidence. RESULTS: We reviewed 1949 citations, from which we included 21 studies. The risk of bias for most of the included studies was moderate. A higher proportion of patients who received bisphosphonate achieved resolution of hypercalcemia when compared to placebo. The incidence rate of adverse events was significantly higher in the bisphosphonate group. Comparing denosumab to bisphosphonate, there was no significant difference in the rate of patients who achieved resolution of hypercalcemia. Two-thirds of patients with refractory/recurrent hypercalcemia of malignancy who received denosumab following bisphosphonate therapy achieved resolution of hypercalcemia. Addition of calcitonin to bisphosphonate therapy did not affect the resolution of hypercalcemia, time to normocalcemia, or hypocalcemia. Only indirect evidence was available to address questions on the management of hypercalcemia in tumors associated with high calcitriol levels, refractory/recurrent hypercalcemia of malignancy following the use of bisphosphonates, and the use of calcimimetics in the treatment of hypercalcemia associated with parathyroid carcinoma. The certainty of the evidence to address all 8 clinical questions was low to very low. CONCLUSION: The evidence summarized in this systematic review addresses the benefits and harms of treatments of hypercalcemia of malignancy. Additional information about patients' values and preferences, and other important decisional and contextual factors is needed to facilitate the development of clinical recommendations.

Perioperative Management of Vitamin K Antagonists and Direct Oral Anticoagulants: A Systematic Review and Meta-analysis.

BACKGROUND: The management of patients who are receiving chronic oral anticoagulation therapy and require an elective surgery or an invasive procedure is a common clinical scenario. RESEARCH QUESTION: What is the best available evidence to support the development of American College of Chest Physicians guidelines on the perioperative management of patients who are receiving long-term vitamin K agonist (VKA) or direct oral anticoagulant (DOAC) and require elective surgery or procedures? STUDY DESIGN AND METHODS: A literature search including multiple databases from database inception through July 16, 2020, was performed. Meta-analyses were conducted when appropriate. RESULTS: In patients receiving VKA (warfarin) undergoing elective noncardiac surgery, shorter (< 3 days) VKA interruption is associated with an increased risk of major bleeding. In patients who required VKA interruption, heparin bridging (mostly with low-molecular-weight heparin [LMWH]) was associated with a statistically significant increased risk of major bleed, representing a very low certainty of evidence (COE). Compared with DOAC interruption 1 to 4 days before surgery, continuing DOACs may be associated with higher risk of bleeding demonstrated in some, but not all studies. In patients who needed DOAC interruption, bridging with LMWH may be associated with a statistically significant increased risk of bleeding, representing a low COE. INTERPRETATION: The certainty in the evidence supporting the perioperative management of anticoagulants remains limited. No high-quality evidence exists to support the practice of heparin bridging during the interruption of VKA or DOAC therapy for an elective surgery or procedure, or for the practice of interrupting VKA therapy for minor procedures, including cardiac device implantation, or continuation of a DOAC vs short-term interruption of a DOAC in the perioperative period.

A Systematic Review Supporting the Endocrine Society Guidelines: Management of Diabetes and High Risk of Hypoglycemia.

CONTEXT: Interventions targeting hypoglycemia in people with diabetes are important for improving quality of life and reducing morbidity and mortality. OBJECTIVE: To support development of the Endocrine Society Clinical Practice Guideline for management of individuals with diabetes at high risk for hypoglycemia. METHODS: We searched several databases for studies addressing 10 questions provided by a guideline panel from the Endocrine Society. Meta-analysis was conducted when feasible. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology was used to assess certainty of evidence. RESULTS: We included 149 studies reporting on 43 344 patients. Continuous glucose monitoring (CGM) reduced episodes of severe hypoglycemia in patients with type 1 diabetes (T1D) and reduced the proportion of patients with hypoglycemia (blood glucose [BG] levels <54 mg/dL). There were no data on use of real-time CGM with algorithm-driven insulin pumps vs multiple daily injections with BG testing in people with T1D. CGM in outpatients with type 2 diabetes taking insulin and/or sulfonylureas reduced time spent with BG levels under 70 mg/dL. Initiation of CGM in hospitalized patients at high risk for hypoglycemia reduced episodes of hypoglycemia with BG levels lower than 54 mg/dL and time spent under 54 mg/dL. The proportion of patients with hypoglycemia with BG levels lower than 70 mg/dL and lower than 54 mg/dL detected by CGM was significantly higher than point-of-care BG testing. We found no data evaluating continuation of personal CGM in the hospital. Use of an inpatient computerized glycemic management program utilizing electronic health record data was associated with fewer patients with and episodes of hypoglycemia with BG levels lower than 70 mg/dL and fewer patients with severe hypoglycemia compared with standard care. Long-acting basal insulin analogs were associated with less hypoglycemia. Rapid-acting insulin analogs were associated with reduced severe hypoglycemia, though there were more patients with mild to moderate hypoglycemia. Structured diabetes education programs reduced episodes of severe hypoglycemia and time below 54 mg/dL in outpatients taking insulin. Glucagon formulations not requiring reconstitution were associated with longer times to recovery from hypoglycemia, although the proportion of patients who recovered completely from hypoglycemia was not different between the 2 groups. CONCLUSION: This systematic review summarized the best available evidence about several interventions addressing hypoglycemia in people with diabetes. This evidence base will facilitate development of clinical practice guidelines by the Endocrine Society.

Diagnostic accuracy of neuroimaging in emergency department patients with acute vertigo or dizziness: A systematic review and meta-analysis for the guidelines for reasonable and appropriate care in the emergency department.

BACKGROUND: Patients presenting to the emergency department (ED) with acute vertigo or dizziness represent a diagnostic challenge. Neuroimaging has variable indications and yield. We aimed to conduct a systematic review and meta-analysis of the diagnostic test accuracy of neuroimaging for patients presenting with acute vertigo or dizziness. METHODS: An electronic search was designed following patient-intervention-control-outcome (PICO) question-(P) adult patients with acute vertigo or dizziness presenting to the ED; (I) neuroimaging including computed tomography (CT), CT angiography (CTA), magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), and ultrasound (US); (C) MRI/clinical criterion standard; and (O) central causes (stroke, hemorrhage, tumor, others) versus peripheral causes of symptoms. Articles were assessed in duplicate. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed. Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) was used to assess certainty of evidence in pooled estimates. RESULTS: We included studies that reported diagnostic test accuracy. From 6309 titles, 460 articles were retrieved, and 12 were included: noncontrast CT scan-six studies, 771 patients, pooled sensitivity 28.5% (95% confidence interval [CI] 14.4%-48.5%, moderate certainty) and specificity 98.9% (95% CI 93.4%-99.8%, moderate certainty); MRI-five studies, 943 patients, sensitivity 79.8% (95% CI 71.4%-86.2%, high certainty) and specificity 98.8% (95% CI 96.2%-100%, high certainty); CTA-one study, 153 patients, sensitivity 14.3% (95% CI 1.8%-42.8%) and specificity 97.7% (95% CI 93.8%-99.6%), CT had higher sensitivity than CTA (21.4% and 14.3%) for central etiology; MRA-one study, 24 patients, sensitivity 60.0% (95% CI 26.2%-87.8%) and specificity 92.9% (95% CI 66.1%-99.8%); US-three studies, 258 patients, sensitivity ranged from 30% to 53.6%, specificity from 94.9% to 100%. CONCLUSIONS: Noncontrast CT has very low sensitivity and MRI will miss approximately one in five patients with stroke if imaging is obtained early after symptom onset. The evidence does not support neuroimaging as the only tool for ruling out stroke and other central causes in patients with acute dizziness or vertigo presenting to the ED.

Effect of TAVR Approach and Other Baseline Factors on the Incidence of Acute Kidney Injury: A Systematic Review and Meta-Analysis.

Background: Acute kidney injury (AKI) is a well-known complication following a transcatheter aortic valve replacement (TAVR) and is associated with higher morbidity and mortality. Objective: We aim to compare the risk of developing AKI after transfemoral (TF), transapical (TA), and transaortic (TAo) approaches following TAVR. Methods: We searched Medline and EMBASE databases from January 2009 to January 2021. We included studies that evaluated the risk of AKI based on different TAVR approaches. After extracting each study's data, we calculated the risk ratio and 95% confidence intervals using RevMan software 5.4. Publication bias was assessed by the forest plot. Results: Thirty-six (36) studies, consisting of 70,406 patients undergoing TAVR were included. Thirty-five studies compared TF to TA, and only seven investigations compared TF to TAo. AKI was documented in 4,857 out of 50,395 (9.6%) patients that underwent TF TAVR compared to 3,155 out of 19,721 (16%) patients who underwent TA-TAVR, with a risk ratio of 0.49 (95% CI, 0.36-0.66; p < 0.00001). Likewise, 273 patients developed AKI out of the 1,840 patients (14.8%) that underwent TF-TAVR in contrast to 67 patients out of the 421 patients (15.9%) that underwent TAo-TAVR, with a risk ratio of 0.51 (95% CI, 0.27-0.98; p = 0.04). There was no significant risk when we compared TA to TAo approaches, with a risk ratio of 0.89 (95% CI, 0.29-2.75; p = 0.84). Conclusion: The risk of post-TAVR AKI is significantly lower in patients who underwent TF-TAVR than those who underwent TA-TAVR or TAo-TAVR.

Using model explanations to guide deep learning models towards consistent explanations for EHR data.

It has been shown that identical deep learning (DL) architectures will produce distinct explanations when trained with different hyperparameters that are orthogonal to the task (e.g. random seed, training set order). In domains such as healthcare and finance, where transparency and explainability is paramount, this can be a significant barrier to DL adoption. In this study we present a further analysis of explanation (in)consistency on 6 tabular datasets/tasks, with a focus on Electronic Health Records data. We propose a novel deep learning ensemble architecture that trains its sub-models to produce consistent explanations, improving explanation consistency by as much as 315% (e.g. from 0.02433 to 0.1011 on MIMIC-IV), and on average by 124% (e.g. from 0.12282 to 0.4450 on the BCW dataset). We evaluate the effectiveness of our proposed technique and discuss the implications our results have for both industrial applications of DL and explainability as well as future methodological work.

Perioperative Management of Antiplatelet Therapy: A Systematic Review and Meta-analysis.

Objective: To summarize the available evidence about the perioperative management of patients who are receiving long-term antiplatelet therapy and require elective surgery/procedures. Methods: This systematic review supports the development of the American College of Chest Physicians guideline on the perioperative management of antiplatelet therapy. A literature search of MEDLINE, EMBASE, Scopus and Cochrane databases was conducted from each database's inception to July 16, 2020. Meta-analyses were conducted when possible. Results: In patients receiving long-term antiplatelet therapy and undergoing elective noncardiac surgery, the available evidence did not show a significant difference in major bleeding between a shorter vs longer antiplatelet interruption, with low certainty of evidence (COE). Compared with patients who received placebo perioperatively, aspirin continuation was associated with increased risk of major bleeding (relative risk [RR], 1.31; 95% CI, 1.15-1.50; high COE) and lower risk of major thromboembolism (RR, 0.74; 95% CI, 0.58-0.94; moderate COE). During antiplatelet interruption, bridging with low-molecular-weight heparin was associated with increased risk of major bleeding compared with no bridging (RR, 1.86; 95% CI, 1.24-2.79; very low COE). Continuation of antiplatelets during minor dental and ophthalmologic procedures was not associated with a statistically significant difference in the risk of major bleeding (very low COE). Conclusion: This systematic review summarizes the current evidence about the perioperative management of antiplatelet therapy and highlights the urgent need for further research, particularly with the increasing prevalence of patients taking 1 or more antiplatelet agents.

A Systematic Review Supporting the Endocrine Society Clinical Practice Guideline for the Management of Hyperglycemia in Adults Hospitalized for Noncritical Illness or Undergoing Elective Surgical Procedures.

CONTEXT: Individuals with diabetes or newly recognized hyperglycemia account for over 30% of noncritically ill hospitalized patients. Management of hyperglycemia in these patients is challenging. OBJECTIVE: To support development of the Endocrine Society Clinical Practice Guideline for management of hyperglycemia in adults hospitalized for noncritical illness or undergoing elective surgical procedures. METHODS: We searched several databases for studies addressing 10 questions provided by a guideline panel from the Endocrine Society. Meta-analysis was conducted when feasible. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology was used to assess certainty of evidence. RESULTS: We included 94 studies reporting on 135 553 patients. Compared with capillary blood glucose, continuous glucose monitoring increased the number of patients identified with hypoglycemia and decreased mean daily blood glucose (BG) (very low certainty). Data on continuation of insulin pump therapy in hospitalized adults were sparse. In hospitalized patients receiving glucocorticoids, combination neutral protamine hagedorn (NPH) and basal-bolus insulin was associated with lower mean BG compared to basal-bolus insulin alone (very low certainty). Data on NPH insulin vs basal-bolus insulin in hospitalized adults receiving enteral nutrition were inconclusive. Inpatient diabetes education was associated with lower HbA1c at 3 and 6 months after discharge (moderate certainty) and reduced hospital readmissions (very low certainty). Preoperative HbA1c level < 7% was associated with shorter length of stay, lower postoperative BG and a lower number of neurological complications and infections, but a higher number of reoperations (very low certainty). Treatment with glucagon-like peptide-1 agonists or dipeptidyl peptidase-4 inhibitors in hospitalized patients with type 2 diabetes and mild hyperglycemia was associated with lower frequency of hypoglycemic events than insulin therapy (low certainty). Caloric oral fluids before surgery in adults with diabetes undergoing surgical procedures did not affect outcomes (very low certainty). Counting carbohydrates for prandial insulin dosing did not affect outcomes (very low certainty). Compared with scheduled insulin (basal-bolus or basal insulin + correctional insulin), correctional insulin was associated with higher mean daily BG and fewer hypoglycemic events (low certainty). CONCLUSION: The certainty of evidence supporting many hyperglycemia management decisions is low, emphasizing importance of shared decision-making and consideration of other decisional factors.