Interventions for Optimization of Guideline-Directed Medical Therapy: A Systematic Review.

Importance: Implementation of guideline-directed medical therapy (GDMT) in real-world practice remains suboptimal. It is unclear which interventions are most effective at addressing current barriers to GDMT in patients with heart failure with reduced ejection fraction (HFrEF). Objective: To perform a systematic review to identify which types of system-level initiatives are most effective at improving GDMT use among patients with HFrEF. Evidence Review: PubMed, Embase, Cochrane, CINAHL, and Web of Science databases were queried from January 2010 to November 2023 for randomized clinical trials that implemented a quality improvement intervention with GDMT use as a primary or secondary outcome. References from related review articles were also included for screening. Quality of studies and bias assessment were graded based on the Cochrane Risk of Bias tool and Oxford Centre for Evidence-Based Medicine. Findings: Twenty-eight randomized clinical trials were included with an aggregate sample size of 19 840 patients. Studies were broadly categorized as interdisciplinary interventions (n = 15), clinician education (n = 5), electronic health record initiatives (n = 6), or patient education (n = 2). Overall, interdisciplinary titration clinics were associated with significant increases in the proportion of patients on target doses of GDMT with a 10% to 60% and 2% to 53% greater proportion of patients on target doses of ß-blockers and renin-angiotensin-aldosterone system inhibitors, respectively, in intervention groups compared with usual care. Other interventions, such as audits, clinician and patient education, or electronic health record alerts, were also associated with some improvements in GDMT utilization, though these findings were inconsistent across studies. Conclusions and Relevance: This review summarizes interventions aimed at optimization of GDMT in clinical practice. Initiatives that used interdisciplinary teams, largely comprised of nurses and pharmacists, most consistently led to improvements in GDMT. Additional large, randomized studies are necessary to better understand other types of interventions, as well as their long-term efficacy and sustainability.

Serological response to vaccination in post-acute sequelae of COVID.

BACKGROUND: Individuals with post-acute sequelae of COVID (PASC) may have a persistence in immune activation that differentiates them from individuals who have recovered from COVID without clinical sequelae. To investigate how humoral immune activation may vary in this regard, we compared patterns of vaccine-provoked serological response in patients with PASC compared to individuals recovered from prior COVID without PASC. METHODS: We prospectively studied 245 adults clinically diagnosed with PASC and 86 adults successfully recovered from prior COVID. All participants had measures of humoral immunity to SARS-CoV-2 assayed before or after receiving their first-ever administration of COVID vaccination (either single-dose or two-dose regimen), including anti-spike (IgG-S and IgM-S) and anti-nucleocapsid (IgG-N) antibodies as well as IgG-S angiotensin-converting enzyme 2 (ACE2) binding levels. We used unadjusted and multivariable-adjusted regression analyses to examine the association of PASC compared to COVID-recovered status with post-vaccination measures of humoral immunity. RESULTS: Individuals with PASC mounted consistently higher post-vaccination IgG-S antibody levels when compared to COVID-recovered (median log IgG-S 3.98 versus 3.74, P < 0.001), with similar results seen for ACE2 binding levels (median 99.1 versus 98.2, P = 0.044). The post-vaccination IgM-S response in PASC was attenuated but persistently unchanged over time (P = 0.33), compared to in COVID recovery wherein the IgM-S response expectedly decreased over time (P = 0.002). Findings remained consistent when accounting for demographic and clinical variables including indices of index infection severity and comorbidity burden. CONCLUSION: We found evidence of aberrant immune response distinguishing PASC from recovered COVID. This aberrancy is marked by excess IgG-S activation and ACE2 binding along with findings consistent with a delayed or dysfunctional immunoglobulin class switching, all of which is unmasked by vaccine provocation. These results suggest that measures of aberrant immune response may offer promise as tools for diagnosing and distinguishing PASC from non-PASC phenotypes, in addition to serving as potential targets for intervention.

Awareness of SARS-CoV-2 Omicron Variant Infection Among Adults With Recent COVID-19 Seropositivity.

Importance: Some individuals who were infected by the SARS-CoV-2 Omicron variant may have been completely unaware of their infectious status while the virus was actively transmissible. Objective: To examine awareness of infectious status among individuals during the recent Omicron variant surge in a diverse and populous urban region of Los Angeles County. Design, Setting, and Participants: This cohort study analyzed the records of adult employees and patients of an academic medical center who were enrolled in a longitudinal COVID-19 serological study in Los Angeles County, California. These participants had 2 or more serial anti-nucleocapsid IgG (IgG-N) antibody measurements at least 1 month apart, with the first occurring after the end of a regional Delta variant surge (September 15, 2021) and a subsequent one occurring after the start of a regional Omicron variant surge (December 15, 2021). Adults with evidence of new SARS-CoV-2 infection occurring during the Omicron variant surge period through May 4, 2022, were included in the present study sample. Exposures: Recent Omicron variant infection as evidenced by SARS-CoV-2 seroconversion. Main Outcomes and Measures: Awareness of recent SARS-CoV-2 infection was ascertained from review of self-reported health updates, medical records, and COVID-19 testing data. Results: Of the 210 participants (median [range] age, 51 (23-84) years; 136 women [65%]) with serological evidence of recent Omicron variant infection, 44% (92) demonstrated awareness of any recent Omicron variant infection and 56% (118) reported being unaware of their infectious status. Among those who were unaware, 10% (12 of 118) reported having had any symptoms, which they attributed to a common cold or other non-SARS-CoV-2 infection. In multivariable analyses that accounted for demographic and clinical characteristics, participants who were health care employees of the medical center were more likely than nonemployees to be aware of their recent Omicron variant infection (adjusted odds ratio, 2.46; 95% CI, 1.30-4.65). Conclusions and Relevance: Results of this study suggest that more than half of adults with recent Omicron variant infection were unaware of their infectious status and that awareness was higher among health care employees than nonemployees, yet still low overall. Unawareness may be a highly prevalent factor associated with rapid person-to-person transmission within communities.

Development and Validation of Machine Models Using Natural Language Processing to Classify Substances Involved in Overdose Deaths.

Importance: Overdose is one of the leading causes of death in the US; however, surveillance data lag considerably from medical examiner determination of the death to reporting in national surveillance reports. Objective: To automate the classification of deaths related to substances in medical examiner data using natural language processing (NLP) and machine learning (ML). Design, Setting, and Participants: Diagnostic study comparing different natural language processing and machine learning algorithms to identify substances related to overdose in 10 health jurisdictions in the US from January 1, 2020, to December 31, 2020. Unstructured text from 35 433 medical examiner and coroners' death records was examined. Exposures: Text from each case was manually classified to a substance that was related to the death. Three feature representation methods were used and compared: text frequency-inverse document frequency (TF-IDF), global vectors for word representations (GloVe), and concept unique identifier (CUI) embeddings. Several ML algorithms were trained and best models were selected based on F-scores. The best models were tested on a hold-out test set and results were reported with 95% CIs. Main Outcomes and Measures: Text data from death certificates were classified as any opioid, fentanyl, alcohol, cocaine, methamphetamine, heroin, prescription opioid, and an aggregate of other substances. Diagnostic metrics and 95% CIs were calculated for each combination of feature extraction method and machine learning classifier. Results: Of 35 433 death records analyzed (decedent median age, 58 years [IQR, 41-72 years]; 24 449 [69%] were male), the most common substances related to deaths included any opioid (5739 [16%]), fentanyl (4758 [13%]), alcohol (2866 [8%]), cocaine (2247 [6%]), methamphetamine (1876 [5%]), heroin (1613 [5%]), prescription opioids (1197 [3%]), and any benzodiazepine (1076 [3%]). The CUI embeddings had similar or better diagnostic metrics compared with word embeddings and TF-IDF for all substances except alcohol. ML classifiers had perfect or near perfect performance in classifying deaths related to any opioids, heroin, fentanyl, prescription opioids, methamphetamine, cocaine, and alcohol. Classification of benzodiazepines was suboptimal using all 3 feature extraction methods. Conclusions and Relevance: In this diagnostic study, NLP/ML algorithms demonstrated excellent diagnostic performance at classifying substances related to overdoses. These algorithms should be integrated into workflows to decrease the lag time in reporting overdose surveillance data.

Perioperative and Long-Term Outcomes of Robot-Assisted Partial Nephrectomy: A Systematic Review.

BACKGROUND: Adoption of the robotic surgical platform for small renal cancers has rapidly expanded, but its utility compared to other approaches has not been established. The objective of this review is to assess perioperative and long-term oncologic and functional outcomes of robot-assisted partial nephrectomy (RAPN) compared to laparoscopic partial nephrectomy (LPN) and open partial nephrectomy (OPN). METHODS: A search in PubMed, Embase, and Cochrane (2010-2019) was conducted. Of 3877 articles screened, 7 observational studies were included. RESULTS: RAPN was associated with 24-50 mL less intraoperative blood loss compared to LPN and 39-84 mL less than OPN. RAPN also demonstrated trends of other postoperative benefits, such as shorter length of stay and fewer major complications. Several studies reported better long-term functional kidney outcomes, but these findings were inconsistent. Recurrence and cancer-specific survival (CSS) were similar across groups. While RAPN had a 5-year CSS of 90.1%-97.9%, LPN and OPN had survival rates of 85.9%-86.9% and 88.5-96.3% respectively. CONCLUSIONS: RAPN may be associated with a lower estimated blood loss and comparable long-term outcomes when compared to other surgical approaches. However, additional randomized or propensity matched studies are warranted to fully assess long-term functional kidney and oncologic outcomes.

Association of implicit intensity values incorporated into work RVUs with objective measures.

BACKGROUND: The "intensity" of a surgical procedure is supposed to be incorporated into work RVUs to allow higher compensation rates for more complex procedures. However, updates to work RVUs are subjective and it is unclear if these intensity values correlate to objective measures of a procedure's complexity. METHODS: Centers for Medicare and Medicaid Services (CMS) data were used to calculate intraservice intensity values for CPT codes in 2017 ("CMS intensity values"). Twenty-six objective measures- spanning patient, case, and risk characteristics - were generated using the 2017 participant use file from NSQIP. CMS intensity values were compared to objective measures using scatterplots and correlations. RESULTS: Among 473 CPT codes, CMS intensity values ranged from 0.0031 to 0.142 work RVUs/minute. CMS intensity values were positively associated with 3 objective measures, negatively associated with 5 measures, and not associated with the remaining 18 measures. CONCLUSIONS: Despite intensity values - and therefore compensation rates - varying over 40-fold in the wRVU scale, there was generally no association between their magnitude and objective measures of surgical intensity.

Surgeon work captured by the National Surgical Quality Improvement Program across specialties.

BACKGROUND: The National Surgical Quality Improvement Program (NSQIP) database is increasingly used for surgical research. However, it is unclear how well this database represents the breadth of work performed by different specialties. METHODS: Using the 2017 NSQIP participant use file and the 2017 Medicare Physician/Supplier Procedure Summary file, we evaluated (1) what proportion of surgical work is captured by NSQIP, (2) what procedures and disciplines are undersampled, and (3) the overall concordance between the NSQIP sample and a national sample. RESULTS: The NSQIP database reported at least one case for 4,463 out of the 5,272 Current Procedures Terminology codes in the Medicare file, potentially capturing 97.8% of surgical work across all 10 specialties. However, this proportion decreased to 72.1% when only procedures with at least 100 cases in NSQIP were considered. Limiting our analysis to only those procedures with 100 cases had markedly different effects by specialty. In part, this was owing to undersampling of minor procedures, which are more common in disciplines such as otolaryngology and urology. The overall association between the size of the NSQIP sample and the Medicare sample was 0.08. CONCLUSION: Although NSQIP has the potential to capture a diverse surgical caseload, some specialties and procedures are undersampled, limiting the ability for NSQIP to generate valid benchmarks. There was little correlation between the sample sizes in NSQIP and a national sample. Increasing sampling of underrepresented procedures and developing weights to scale NSQIP to a national sample would strengthen the program's ability to inform health outcomes research and provide valid comparisons across procedures and specialties.

Inhibition of glutamine synthetase in the central nucleus of the amygdala induces anhedonic behavior and recurrent seizures in a rat model of mesial temporal lobe epilepsy.

The prevalence of depression and suicide is increased in patients with mesial temporal lobe epilepsy (MTLE); however, the underlying mechanism remains unknown. Anhedonia, a core symptom of depression that is predictive of suicide, is common in patients with MTLE. Glutamine synthetase, an astrocytic enzyme that metabolizes glutamate and ammonia to glutamine, is reduced in the amygdala in patients with epilepsy and depression and in suicide victims. Here, we sought to develop a novel model of anhedonia in MTLE by testing the hypothesis that deficiency in glutamine synthetase in the central nucleus of the amygdala (CeA) leads to epilepsy and comorbid anhedonia. Nineteen male Sprague-Dawley rats were implanted with an osmotic pump infusing either the glutamine synthetase inhibitor methionine sulfoximine [MSO (n=12)] or phosphate buffered saline [PBS (n=7)] into the right CeA. Seizure activity was monitored by video-intracranial electroencephalogram (EEG) recordings for 21days after the onset of MSO infusion. Sucrose preference, a measure of anhedonia, was assessed after 21days. Methionine sulfoximine-infused rats exhibited recurrent seizures during the monitoring period and showed decreased sucrose preference over days when compared with PBS-infused rats (p<0.01). Water consumption did not differ between the PBS-treated group and the MSO-treated group. Neurons were lost in the CeA, but not the medial amygdala, lateral amygdala, basolateral amygdala, or the hilus of the dentate gyrus, in the MSO-treated rats. The results suggest that decreased glutamine synthetase activity in the CeA is a possible common cause of anhedonia and seizures in TLE. We propose that the MSO CeA model can be used for mechanistic studies that will lead to the development and testing of novel drugs to prevent seizures, depression, and suicide in patients with TLE.