A Machine Learning Approach to Predict Weight Change in ART-Experienced People Living With HIV.

INTRODUCTION: The objective of the study was to develop machine learning (ML) models that predict the percentage weight change in each interval of time in antiretroviral therapy-experienced people living with HIV. METHODS: This was an observational study that comprised consecutive people living with HIV attending Modena HIV Metabolic Clinic with at least 2 visits. Data were partitioned in an 80/20 training/test set to generate 10 progressively parsimonious predictive ML models. Weight gain was defined as any weight change >5%, at the next visit. SHapley Additive exPlanations values were used to quantify the positive or negative impact of any single variable included in each model on the predicted weight changes. RESULTS: A total of 3,321 patients generated 18,322 observations. At the last observation, the median age was 50 years and 69% patients were male. Model 1 (the only 1 including body composition assessed with dual-energy x-ray absorptiometry) had an accuracy greater than 90%. This model could predict weight at the next visit with an error of <5%. CONCLUSIONS: ML models with the inclusion of body composition and metabolic and endocrinological variables had an excellent performance. The parsimonious models available in standard clinical evaluation are insufficient to obtain reliable prediction, but are good enough to predict who will not experience weight gain.

Quality of life and intrinsic capacity in patients with post-acute COVID-19 syndrome is in relation to frailty and resilience phenotypes.

The objective of this study was to characterize frailty and resilience in people evaluated for Post-Acute COVID-19 Syndrome (PACS), in relation to quality of life (QoL) and Intrinsic Capacity (IC). This cross-sectional, observational, study included consecutive people previously hospitalized for severe COVID-19 pneumonia attending Modena (Italy) PACS Clinic from July 2020 to April 2021. Four frailty-resilience phenotypes were built: "fit/resilient", "fit/non-resilient", "frail/resilient" and "frail/non-resilient". Frailty and resilience were defined according to frailty phenotype and Connor Davidson resilience scale (CD-RISC-25) respectively. Study outcomes were: QoL assessed by means of Symptoms Short form health survey (SF-36) and health-related quality of life (EQ-5D-5L) and IC by means of a dedicated questionnaire. Their predictors including frailty-resilience phenotypes were explored in logistic regressions. 232 patients were evaluated, median age was 58.0 years. PACS was diagnosed in 173 (74.6%) patients. Scarce resilience was documented in 114 (49.1%) and frailty in 72 (31.0%) individuals. Predictors for SF-36 score < 61.60 were the phenotypes "frail/non-resilient" (OR = 4.69, CI 2.08-10.55), "fit/non-resilient" (OR = 2.79, CI 1.00-7.73). Predictors for EQ-5D-5L < 89.7% were the phenotypes "frail/non-resilient" (OR = 5.93, CI 2.64-13.33) and "frail/resilient" (OR = 5.66, CI 1.93-16.54). Predictors of impaired IC (below the mean score value) were "frail/non-resilient" (OR = 7.39, CI 3.20-17.07), and "fit/non-resilient" (OR = 4.34, CI 2.16-8.71) phenotypes. Resilience and frailty phenotypes may have a different impact on wellness and QoL and may be evaluated in people with PACS to identify vulnerable individuals that require suitable interventions.

Metabolic-Associated Fatty Liver Disease Is Highly Prevalent in the Postacute COVID Syndrome.

BACKGROUND: A proposal has recently been advanced to change the traditional definition of nonalcoholic fatty liver disease to metabolic-associated fatty liver disease (MAFLD), to reflect the cluster of metabolic abnormalities that may be more closely associated with cardiovascular risk. Long coronavirus disease 2019 (COVID-19) is a smoldering inflammatory condition, characterized by several symptom clusters. This study aims to determine the prevalence of MAFLD in patients with postacute COVID syndrome (PACS) and its association with other PACS-cluster phenotypes. METHODS: We included 235 patients observed at a single university outpatient clinic. The diagnosis of PACS was based on ≥1 cluster of symptoms: respiratory, neurocognitive, musculoskeletal, psychological, sensory, and dermatological. The outcome was prevalence of MAFLD detected by transient elastography during the first postdischarge follow-up outpatient visit. The prevalence of MAFLD at the time of hospital admission was calculated retrospectively using the hepatic steatosis index. RESULTS: Of 235 patients, 162 (69%) were men (median age 61). The prevalence of MAFLD was 55.3% at follow-up and 37.3% on admission (P < .001). Insulin resistance (odds ratio [OR] = 1.5; 95% confidence interval [CI], 1.14-1.96), body mass index (OR = 1.14; 95% CI, 1.04-1.24), and the metabolic syndrome (OR = 2.54; 95% CI, 1.13-5.68) were independent predictors of MAFLD. The number of PACS clusters was inversely associated with MAFLD (OR = 0.86; 95% CI, .76-0.97). Thirty-one patients (13.2%) had MAFLD with no other associated PACS clusters. All correlations between MAFLD and other PACS clusters were weak. CONCLUSIONS: Metabolic-associated fatty liver disease was highly prevalent after hospital discharge and may represent a specific PACS-cluster phenotype, with potential long-term metabolic and cardiovascular health implications.

Resilience and Frailty in People Living With HIV During the COVID Era: Two Complementary Constructs Associated With Health-Related Quality of Life.

BACKGROUND: Resilience is defined as an individual's positive adaptation to stressors. The COVID-19 pandemic represents a generalized stressor which may affect differently people living with HIV (PLWH). The objective of this study was to characterize resilience in PLWH with particular regarding the identification of frailty-resilience phenotypes, which may differently affect health-related quality of life (HR-QoL). METHODS: This was an observational study of PLWH attending Modena HIV Metabolic Clinic. Frailty was assessed in 2019, before the onset of the COVID-19 pandemic by using 37-Item frailty index ranging from 0 to 1. The frailty index score was categorized as fit (<0.25) or frail (>0.25). In January 2021, PLWH were offered to complete a set of electronic questionnaires including the CD-RISC-25 for resilience and EQ-5D5L and SF-36 for HR-QoL. Resilience was defined as CD-RISC-25 score >75.7 (ranging from 0 to 100). RESULTS: Of 800 PLWH reached by mail, 575 (72%) completed the questionnaires. The median age and HIV duration were 54.5 and 24.3 years, respectively. Impaired resilience was associated with loneliness [odds ratio (OR = 2.39; 1.20 to 4.76, P < 0.001)]. Predictors for EQ-5D5L <89.7% were the phenotypes "frail/nonresilient" [OR = 5.21, 95% confidence interval (CI): 2.62 to 10.33] and "fit/nonresilient" (OR = 5.48, 95% CI: 2.8 to 10.74). Predictors for SF-36 <64.40 were the phenotypes "frail/nonresilient" (OR = 7.43, 95% CI: 2.57 to 21.22) and "fit/nonresilient" (OR = 6.27, 95% CI: 2.17 to 18.16). Both models were corrected for age, sex, HIV duration, and nadir CD4. CONCLUSIONS: Resilience characterizes the well-being of PLWH during the COVID-19 crisis. This construct is complementary to frailty in the identification of clinical phenotypes with different impacts on HR-QoL.

Contribution of integrase inhibitor use, body mass index, physical activity and caloric intake to weight gain in people living with HIV.

Background: Integrase inhibitor (INSTI) use has been associated with greater weight gain (WG) among people living with HIV (PLWH), but it is unclear how this effect compares in magnitude to traditional risk factors for WG. We assessed the population attributable fractions (PAFs) of modifiable lifestyle factors and INSTI regimens in PLWH who experienced a ≥5% WG over follow-up.Methods: In an observational cohort study from 2007 to 2019 at Modena HIV Metabolic Clinic, Italy, ART-experienced but INSTI-naive PLWH were grouped as INSTI-switchers vs non-INSTI. Groups were matched for sex, age, baseline BMI and follow-up duration. Significant WG was defined as an increase of ≥5% from 1st visit weight over follow-up. PAFs and 95% CIs were estimated to quantify the proportion of the outcome that could be avoided if the risk factors were not present.Results: 118 PLWH switched to INSTI and 163 remained on current ART. Of 281 PLWH (74.3% males), mean follow-up was 4.2 years, age 50.3 years, median time since HIV diagnosis 17.8 years, CD4 cell count 630 cells/µL at baseline. PAF for weight gain was the greatest for high BMI (45%, 95% CI: 27-59, p < 0.001), followed by high CD4/CD8 ratio (41%, 21-57, p < 0.001) and lower physical activity (32%, 95% CI 5-52, p = 0.03). PAF was not significant for daily caloric intake (-1%, -9-13, p = 0.45), smoking cessation during follow-up (5%, 0-12, p = 0.10), INSTI switch (11%, -19-36; p = 0.34).Conclusions: WG in PLWH on ART is mostly influenced by pre-existing weight and low physical activity, rather than switch to INSTI.