Study design and participants
The INFRAGEN study is a prospective, observational multicenter study involving the Acute Geriatric Units of four Italian centers: S. Anna University Hospital in Ferrara (Emilia-Romagna region, Italy), Padua University Hospital (Veneto region, Italy), IRCCS Foundation San Gerardo dei Tintori in Monza (Lombardy region, Italy), and Federico II University Hospital (Campania region, Italy).
Participants will be selected based on the following criteria: age ≥ 70 years, non-frailty or mild frailty status before admission (CFS < 6), and confirmed diagnosis of acute infectious diseases at the time of hospital admission or during the hospital stay, according to specific ICD-9 codes, with or without a systemic inflammatory response.
Terminally ill patients with an estimated life expectancy of less than 3 months, those with moderate to severe frailty, and individuals who do not wish to participate in the study or complete follow-up assessments will be excluded.
Written informed consent will be obtained from all participants.
Sample size calculations
Based on previous reports on the topic [10], we estimated that a total sample of at least 340 older adults with acute infections would be sufficient to detect a minimal clinically significant difference in Frailty Index (FI) of at least 0.03 points from the pre-admission status to hospital discharge, with an alpha of 0.05, and a power of 80%.
Study procedures
Data will be collected using the RedCap platform. The study will include three assessment points (Fig. 1): hospital admission (Time 1), discharge (Time 2), and three months after hospital discharge (Time 3).
Graphical summary of the INFRAGEN study protocol
Time 3 assessment will be performed through an outpatient follow-up visit, home visit, or telephone interview, depending on the centers’ resources and the patient’s characteristics and availability.
Each assessment will include face-to-face interviews, administration of validated scales and questionnaires, physical tests, and blood sampling (the latter only at Time 1 and 3, and limited to the subgroup undergoing outpatient clinic visits).
The data collected from each participant are briefly described below.
Sociodemographic factors
Sociodemographic information includes age, sex, pre-admission setting (home, family home, long-term care, other hospital wards), and assistance needs from home.
Comprehensive geriatric assessment
Functional status will be assessed using the validated Basic Activities of Daily Living (BADL) and Instrumental Activities of Daily Living (IADL) scales [14, 15]. The presence of cognitive deficits will be evaluated using the Short Portable Mental Status Questionnaire (SPMSQ) [16].
Chronic medical illness burden will be quantified using the Cumulative Illness Rating Scale (CIRS) [17]. Delirium screening will be performed using the 4AT screening tool [18].
Nutritional status and sarcopenia screening will be evaluated using the Mini Nutritional Assessment Short Form (MNA-SF) and SARC-F [19, 20]. Anthropometric measurements, including body weight, height, and calf and arm circumferences, will be recorded for each participant. Pressure sore risk will be determined using the Exton-Smith Scale (ESS) [21]. Depressive symptoms will be assessed using the Geriatric Depression Scale (GDS) [22], and physical performance through handgrip strength and 4-m walking speed, only for those who are able to walk independently or with walking aids.
Furthermore, we will collect data on the patient’s medication history, including medications used chronically before admission and those prescribed at discharge. Information on vaccination history, alcohol consumption, and smoking status will also be obtained.
Infectious disease
The presence of infectious diseases will be determined through routine clinical, radiological, biochemical, and microbiological analyses and coded according to the following ICD-9 codes: 001–139 (Infectious diseases); 460–466 (Acute respiratory infections); 480–487 (Pneumonia, flu disease); 320–326 (Nervous system infections); 540–543 (Appendicitis); 590, 595.0, 595.4, 599.0, 597 (Urinary system infections); 601, 604 (Prostatitis, orchitis, and epididymitis); 616, 574.0, 571.1 (Gynecological system infections); 572.0, 573.1, 573.2 (Acute hepatitis, liver abscess); 575–576.1 (Cholecystitis, cholangitis).
Each documented infectious disease will be characterized by its etiology (if known), and the prescribed antimicrobial therapy will be recorded, including the specific drug and duration. Any concomitant or subsequent infectious diseases arising during hospitalization will be similarly documented. In addition, the Sequential Organ Failure Assessment (SOFA) score will be calculated for each patient to assess infections’ severity [23].
Frailty status
Frailty will be evaluated using the Primary Care – Frailty Index (PC-FI) [24] and the Clinical Frailty Scale (CFS) [25]. The PC-FI is a reliable and validated tool developed in Italian older adults to assess frailty in the primary care setting. It was chosen considering that both the pre-admission and post-discharge assessment will involve older adults in primary care. In particular, pre-admission frailty status will be derived retrospectively to reflect the status of participants two weeks before hospital admission.
Frailty assessment will be repeated 48 h before hospital discharge and at a 3-month follow-up.
Genetic and epigenetic analyses
At Time 1 and Time 3 (only for a subsample of participants), blood samples will be collected and processed for plasma and blood cell stocking for DNA extraction.
DNA extraction from whole blood will be performed by Automated Genomic DNA Purification EZ1 XL machine (QIAGEN).
Global DNA methylation assessment will be performed by “highly quantitative pyrosequencing” technique as genome-wide DNA methylation levels and as gene promoter associated CpG islands utilizing selected age-related methylation marker loci and at LINE-1, Alu, Sat-alpha repetitive elements (as a surrogate for global genome methylation) as previously described [26,27,28]. Results will be also confirmed by internal Illumina Infinium 450 k beadchip array.
Leukocyte Telomere Length (LTL) will be determined using quantitative real-time PCR (QuantStudio™ 3 System; Applied Biosystems) according to the method of Cawthon [29], the 36B4 will be used as a single copy gene reference (SCR), and the ratio Telomere/36B4 (T/S) will be estimated as indicator of LTL [2 − ∆Ct (△Ct = Ct TL − Ct 36B4)].
Biochemical analyses
We will monitor the plasma levels of circulating markers associated with key biological processes related to frailty, such as (1) pro-inflammatory state (including but not limited to TNF-a, IL6, CRP, TRAIL/OPG, CXCL10, MCP-1); (2) metabolic/endocrine imbalance (including but not limited to Cystatin-C, IGF-1); (3) functional brain modifications (including but not limited to BDNF, CSF); (4) oxidative stress (including but not limited to PON-1, Homocysteine).
Analysis will be performed using Luminex Technology, a platform allowing the simultaneous detection and quantification of multiple secreted proteins, including cytokines, chemokines, and growth factors. Customized panels will be generated with targets of interest.
Ox-stress markers will be analyzed using fluorimetric/ELISA assays.
Selected inflammatory markers will be monitored by ProQuantum High-Sensitivity immunoassays (TermoFisher, Waltham, MA USA) to achieve a highly sensitive assay to detect lower protein levels.
Statistical analysis
The baseline characteristics of the study participants will be described using count and proportion for the categorical variables and mean and standard deviation, or median and interquartile ranges, for the quantitative variables, according to their distribution. These characteristics will be compared between participants by type of infection and the presence of systemic inflammatory response through the ANOVA, Student t-test, Kruskal Wallis test, or Chi-squared test, as appropriate.
The intra-individual change in FI between the pre-admission status and the patient’s condition at hospital discharge will be evaluated through the Student t-test for paired samples and then with linear mixed models. The latter analysis will be set with a random intercept and slope and adjusted for potential confounders in the association between infectious diseases and frailty changes, e.g., sociodemographic information, study center, comorbidities, length of hospital stay, and in-hospital clinical events (unrelated to the infection). The mean adjusted change in FI between pre-admission status and hospital discharge will be expressed as ß coefficient and 95% confidence interval (95% CI). Moreover, we will test whether the presence of systemic inflammation will modify the average intra-hospital change in FI. As secondary analyses, we will investigate which infection-related characteristics will be associated with higher in-hospital mortality (through the Kaplan-Mayer method and Cox regression analysis), longer hospital stay (with linear regressions), and discharge setting (with multinomial logistic regression).
The impact of infectious diseases (distinguished by number, site and severity) on frailty changes will be analyzed with linear mixed models, considering the beta coefficients (95% CI) for the interaction infection*time. Additional analyses will be performed to test whether sociodemographic, functional, clinical, and genetic/epigenetic factors interact with infectious diseases to influence the course of frailty over time.
The patterns of global DNA methylation will be assessed in duplicate for each sample and expressed in percentage as the mean obtained by the two evaluations and considered valuable with a discrepancy < 2%. Methylation percentages will be stratified into quartiles, and the combined middle two quartiles will be used as the reference category to identify potential correlations at the top or bottom of the distribution with the clinical phenotypes. As a measure of the relative Telomere length, the ratio of the telomere repeat copy number to the number of single-copy genes (T/S ratio) will be determined by quantitative PCR using the single-copy gene 36B4 for reference and a standard curve. Quality controls and assay validation tests will be assessed by official commercially recognized standards (Qiagen, LifeTechnology). The association of global DNA methylation pattern and telomere length with the frailty trajectories will be evaluated through a linear mixed model, adjusted for potential confounders. Appropriate subgroup and interaction analyses will be performed to explore the possible modifying effects by sex, the presence of the most prevalent clinical conditions, and infection-related factors (i.e. number, site, and severity of infections) on the studied associations.