Can Artificial Intelligence Speak for Incapacitated Patients at the End of Life?

This Viewpoint examines how artificial intelligence could support surrogate decision-makers while addressing some of the attendant epistemic and moral challenges.

Building upon the foundational science curriculum with physiology-based grand rounds: a multi-institutional program evaluation.

Introduction: Vertically integrating physiology into patient care has the potential to improve clinical reasoning. Clinical Physiology Grand Rounds (CPGR) is a case-based teaching method that brings together students from all years of medical school to focus on linking clinical presentations to core basic science concepts including anatomy, physiology, and pathophysiology. In this study, we describe the implementation of CPGR at two different institutions in the United States and assess student-reported outcomes.Methods: We survey students who participated in CPGR at Columbia University College of Physicians & Surgeons (P&S) and Medical University of South Carolina (MUSC). Subjects were queried across three domains: the benefits of attending, the impact of concept maps, and the impact of the mixed-learner environment.Results: Despite differences in session leadership and the underlying medical school curricula, conference attendees reported similar benefits at the two schools included in this study. Students overwhelmingly (92.9%) reported that remembering clinical presentations was easier when they understood the underlying physiology. They also reported gaining a true understanding of concepts that were previously memorized (87.5%). Both clinical (92.5%) and preclinical students (93.1%) valued the mixed-learner environment as a component of the conference.Discussion: By assuring a mixed-learner environment with near-peer interactions, using concept maps as a teaching tool, and rigorously linking clinical presentation and management to physiological concepts, we found that the key benefits of CPGR were replicable across different institutions, despite several local differences in how CPGR was implemented, led, and conducted.

Variation in COVID-19 Mortality Across 117 US Hospitals in High- and Low-Burden Settings.

Some hospitals have faced a surge of patients with COVID-19, while others have not. We assessed whether COVID-19 burden (number of patients with COVID-19 admitted during April 2020 divided by hospital certified bed count) was associated with mortality in a large sample of US hospitals. Our study population included 14,226 patients with COVID-19 (median age 66 years, 45.2% women) at 117 hospitals, of whom 20.9% had died at 5 weeks of follow-up. At the hospital level, the observed mortality ranged from 0% to 44.4%. After adjustment for age, sex, and comorbidities, the adjusted odds ratio for in-hospital death in the highest quintile of burden was 1.46 (95% CI, 1.07-2.00) compared to all other quintiles. Still, there was large variability in outcomes, even among hospitals with a similar level of COVID-19 burden and after adjusting for age, sex, and comorbidities.

Unusual Radiographic Presentation of Pneumocystis Pneumonia in a Patient with AIDS.

Pneumocystis jirovecii pneumonia (PCP) typically presents as an interstitial and alveolar process with ground glass opacities on chest computed tomography (CT). The absence of ground glass opacities on chest CT is thought to have a high negative predictive value for PCP in individuals with AIDS. Here, we report a case of PCP in a man with AIDS who presented to our hospital with subacute shortness of breath and a nonproductive cough. While his chest CT revealed diffuse nodular rather than ground glass opacities, bronchoscopy with bronchoalveolar lavage and transbronchial biopsies confirmed the diagnosis of PCP and did not identify additional pathogens. PCP was not the expected diagnosis based on chest CT, but it otherwise fit well with the patient's clinical and laboratory presentation. In the era of combination antiretroviral therapy, routine prophylaxis for PCP, and increased use of computed tomography, it may be that PCP will increasingly present with nonclassical chest radiographic patterns. Clinicians should be aware of this presentation when selecting diagnostic and management strategies.

Reduced replication capacity of NL4-3 recombinant viruses encoding reverse transcriptase-integrase sequences from HIV-1 elite controllers.

BACKGROUND: Identifying viral and host determinants of HIV-1 elite control may help inform novel therapeutic and/or vaccination strategies. Previously, we observed decreased replication capacity in controller-derived viruses suggesting that fitness consequences of human leukocyte antigen (HLA) class I-associated escape mutations in Gag may contribute to this phenotype. This study examines whether similar functional defects occur in Pol proteins of elite controllers. METHODS: Recombinant NL4-3 viruses encoding plasma RNA-derived reverse transcriptase-integrase sequences from 58 elite controllers and 50 untreated chronic progressors were constructed, and replication capacity measured in vitro using a green fluorescent protein (GFP) reporter T-cell assay. Sequences were analyzed for drug resistance and HLA-associated viral polymorphisms. RESULTS: Controller-derived viruses displayed significantly lower replication capacity compared with those from progressors (P < 0.0001). Among controllers, the most attenuated viruses were generated from individuals expressing HLA-B*57 or B*51. In viruses from B*57+ progressors (n = 8), a significant inverse correlation was observed between B*57-associated reverse transcriptase-integrase escape mutations and replication capacity (R = -0.89; P = 0.003); a similar trend was observed in B*57+ controller-derived viruses (n = 20, R = -0.36; P = 0.08). CONCLUSIONS: HIV-1 Pol function seemed to be compromised in elite controllers. As observed previously for Gag, HLA-associated immune pressure in Pol may contribute to viral attenuation and subsequent control of viremia.

Impaired replication capacity of acute/early viruses in persons who become HIV controllers.

Human immunodeficiency virus type 1 (HIV-1) controllers maintain viremia at <2,000 RNA copies/ml without antiretroviral therapy. Viruses from controllers with chronic infection were shown to exhibit impaired replication capacities, in part associated with escape mutations from cytotoxic-T-lymphocyte (CTL) responses. In contrast, little is known about viruses during acute/early infection in individuals who subsequently become HIV controllers. Here, we examine the viral replication capacities, HLA types, and virus sequences from 18 HIV-1 controllers identified during primary infection. gag-protease chimeric viruses constructed using the earliest postinfection samples displayed significantly lower replication capacities than isolates from persons who failed to control viremia (P = 0.0003). Protective HLA class I alleles were not enriched in these early HIV controllers, but viral sequencing revealed a significantly higher prevalence of drug resistance mutations associated with impaired viral fitness in controllers than in noncontrollers (6/15 [40.0%] versus 10/80 [12.5%], P = 0.018). Moreover, of two HLA-B57-positive (B57(+)) controllers identified, both harbored, at the earliest time point tested, signature escape mutations within Gag that likewise impair viral replication capacity. Only five controllers did not express "protective" alleles or harbor viruses with drug resistance mutations; intriguingly, two of them displayed typical B57 signature mutations (T242N), suggesting the acquisition of attenuated viruses from B57(+) donors. These data indicate that acute/early stage viruses from persons who become controllers have evidence of reduced replication capacity during the initial stages of infection which is likely associated with transmitted or acquired CTL escape mutations or transmitted drug resistance mutations. These data suggest that viral dynamics during acute infection have a major impact on HIV disease outcome.

Persistent low-level viremia in HIV-1 elite controllers and relationship to immunologic parameters.

BACKGROUND: Human immunodeficiency virus type 1 (HIV-1) elite controllers are able to control virus replication to levels below the limits of detection by commercial assays, but the actual level of viremia in these individuals is not well defined. Here, we quantify plasma HIV-1 RNA in elite controllers and correlate this with specific immunologic parameters. METHODS: Plasma HIV-1 RNA levels were quantified in 90 elite controllers with use of a real time reverse-transcriptase polymerase chain reaction assay with a sensitivity of 0.2 copies/mL. HIV-1-specific immune responses and longitudinal CD4(+) T cell counts were examined. RESULTS: The median plasma HIV-1 RNA level was 2 copies/mL (interquartile range, 0.2-14 copies/mL). A longitudinal analysis of 31 elite controllers demonstrated 2-5-fold fluctuations in viremia in the majority of individuals; 6 had persistent levels below 1 copy/mL. Viremia correlated directly with HIV-1-specific neutralizing antibodies and Western blot reactivity but not with CD8(+) T cell responses. Absolute CD4(+) T cell decrease was more common among individuals with detectable viremia (P = .04). CONCLUSIONS: Low-level viremia is present in the majority of elite controllers and is associated with higher HIV-1-specific antibody responses. Absolute CD4(+) T cell loss is more common among viremic individuals, suggesting that even very low-level viremia has negative consequences over time.

Differential neutralization of human immunodeficiency virus (HIV) replication in autologous CD4 T cells by HIV-specific cytotoxic T lymphocytes.

Defining the antiviral efficacy of CD8 T cells is important for immunogen design, and yet most current assays do not measure the ability of responses to neutralize infectious virus. Here we show that human immunodeficiency virus (HIV)-specific cytotoxic T-lymphocyte (CTL) clones and cell lines derived from infected persons and targeting diverse epitopes differ by over 1,000-fold in their ability to retard infectious virus replication in autologous CD4 T cells during a 7-day period in vitro, despite comparable activity as assessed by gamma interferon (IFN-gamma) enzyme-linked immunospot (ELISPOT) assay. Cell lines derived from peripheral blood mononuclear cells stimulated in vitro with peptides representing targeted Gag epitopes consistently neutralized HIV better than Env-specific lines from the same person, although ineffective inhibition of virus replication is not a universal characteristic of Env-specific responses at the clonal level. Gag-specific cell lines were of higher avidity than Env-specific lines, although avidity did not correlate with the ability of Gag- or Env-specific lines to contain HIV replication. The greatest inhibition was observed with cell lines restricted by the protective HLA alleles B*27 and B*57, but stimulation with targeted Gag epitopes resulted in greater inhibition than did stimulation with targeted Env epitopes even in non-B*27/B*57 subjects. These results assessing functional virus neutralization by HIV-specific CD8 T cells indicate that there are marked epitope- and allele-specific differences in virus neutralization by in vitro-expanded CD8 T cells, a finding not revealed by standard IFN-gamma ELISPOT assay currently in use in vaccine trials, which may be of critical importance in immunogen design and testing of candidate AIDS vaccines.