The expected outcomes of this study include:
-
Determination of the overall incidence of post-COVID-19 condition in the study cohort.
-
Identification of patients who continue to experience symptoms beyond 2–3 months from the acute COVID-19 disease.
-
Identification of cluster of symptoms that identify different clinical phenotypes of post-COVID-19 condition.
-
Identification of predictors of post-COVID-19 condition, such as demographic, clinical, and virological factors.
-
Evaluation of long-term organ damage associated with post COVID-19 condition, specifically in the lungs, heart, central nervous system, and peripheral nervous system.
-
Characterisation of the virological and immunological phenotype of post-COVID-19 condition, which can provide insight into the mechanisms underlying this condition.
Table of Contents
Study design
This is an observational retrospective and prospective, multicentre, cohort study. The study will last for 3 years, from February 2022 to February 2025.
Study population
The study will enrol adult patients (> 18 years old) diagnosed with acute COVID-19 disease between 01/03/2020 and 01/02/2025 at the participating centres. The COVID-19 patients will be divided into two groups: Group 1 (hospitalised) and Group 2 (outpatients).
Inclusion criteria for the study require a confirmed diagnosis of SARS-CoV-2 infection based on a positive SARS-CoV-2 RNA test on nasopharyngeal swab or other respiratory samples. Mild COVID-19 disease is defined as cases without hospital admission, while moderate/severe disease includes cases requiring hospitalisation for COVID-19 or hospitalisation for other medical issues with a positive SARS-CoV-2 RNA sample. Patients must provide informed consent for the study to be eligible for inclusion.
Exclusion criteria include death during hospitalisation and patient refusal to participate.
The study will also enrol a control group of hospitalised patients for respiratory infections other than COVID-19 disease during the same study period, with enrolment occurring at hospital discharge and follow-up appointments scheduled at the post COVID clinic with the same timing as COVID-19 patients. These patients will be enrolled if they are adults hospitalised for respiratory infections requiring any type of oxygen therapy or Intensive Care admission (Fig. 1).
Depicting the study population and timeline
Study centres
The coordinator centre is the Clinic of Infectious Diseases, San Paolo Hospital, ASST Santi Paolo and Carlo, Department of Health Sciences, University of Milan, Italy. Other participating centres are Vilnius University Hospital, Santaros Klinikos, Vilnius, Lithuania; University Hospital Heinrich Heine of Dusseldorf, Dusseldorf, Germany; Policlinico Tor Vergata, Università degli Studi di Roma Tor Vergata, Rome, Italy; Regional Hospital Dr. Juan Graham Casasús, Villahermosa, Tabasco, Mexico; Centro Hospitalar de Lisboa Ocidental, Lisbon, Portugal; Pomeranian Medical University, Szczecin, Poland and Federal University of Minas Gerais, Minas Gerais, Brazil.
Patient assessment and follow-up protocol
The patient assessment and follow-up protocol for acute COVID-19 disease involves follow-up visits at 2–3, 6–9, and 12–15 months after clinical recovery. At each visit, routine blood exams and medical visits will be conducted, along with several assessments via questionnaires, including the short version of the post COVID-19 WHO Case Report Form (WHO CRF) for collection of symptoms and newly emerged complaints after COVID-19, the Fatigue Numerical Rating Scale, the Medical Research Council (MRC) dyspnoea scale, the 5-level EQ-5D test (EQ5D-5L), the 12-item WHO Disability Assessment Schedule (WHODAS), the Montreal Cognitive Assessment (MOCA), the Hospital Anxiety and Depression Scale (HADS-A/D), and a screening tool named PCL-5 for Post-Traumatic Stress Disorder (PTSD). If patients are unable to attend the post-COVID-19 clinic, the questionnaires will be able to be completed through a phone interview. All questionnaires will be standardized and available in multiple languages (Fig. 1).
Further instrumental diagnostic evaluations such as diffusing capacity of lung carbon monoxide (DLCO), spirometry, lung computed tomography (CT) scans, pneumological evaluation, echocardiography, cardiological evaluation, neurological evaluation, rehabilitation medicine specialist examination, and psychological evaluation will be conducted as necessary.
Pulmonary symptoms are common after the acute phase, thus patients undergo the MRC dyspnoea scale and oxygen saturation measurement at each post-COVID visit. Further evaluations such as chest CT scans, diffusing capacity of lung carbon monoxide (DLCO), spirometry, and a complete pneumological evaluation will be performed if oxygen saturation is < 95% or MRC dyspnoea scale grade is 2–5.
Patients who recovered from acute COVID-19 disease could present a higher risk of cardiac diseases, including heart failure, myocardial infarction, and arrhythmia in the mid- and long-term [18]. During the scheduled post-COVID visits, patients undergo a complete medical examination, including heart rate and blood pressure measurement, and fill in the MRC dyspnoea scale. Further evaluations such as an electrocardiogram, a cardiologic visit, and an echocardiography will be performed as necessary.
Neurocognitive and neuropsychiatric sequelae are important to investigate for their great impact on patients’ quality of life [6, 19, 20]. Neurocognitive and neuropsychological performance will be assessed by questionnaires, including the previously mentioned HADS-A/D, PCL-5, and MOCA, which will be performed at each scheduled visit. Patients will be referred to a psychologist or a neurologist as necessary (Table 1).
Questionnaires administration
The fatigue numerical rating scale will be used to assess the severity of fatigue experienced by patients, while the MRC dyspnoea scale will be used to evaluate breathlessness. Quality of life will be assessed using the EQ5D-5L tool, which consists of the EQ-5D descriptive system and the EQ visual analogue scale (EQ-VAS). The HADS will be used to evaluate symptoms of anxiety and depression, while the PCL-5 will be used for first screening for PTSD and monitoring symptoms change during time. The MOCA will be used to assess cognitive function, and the WHODAS 2.0 will be used to measure health and disabilities in clinical practice. All questionnaires with the relative scores and the CRF will be listed in Table 2 and more detailed information attached as supplementary documents (Supplementary Appendix).
Biological samples collection
The study protocol includes collection of biological samples for hospitalised COVID-19 patients during the acute phase. Blood samples and a nasopharyngeal swab will be collected at hospital admission for assessment of SARS-CoV-2 diagnostics, viral variant and viral sequencing, and cellular immunity. Blood samples will also be collected at different time points after enrolment in the post-COVID-19 cohort for the same analysis as in the acute phase. Whole blood samples will be stored, centrifuged, and separated into plasma, and peripheral blood mononuclear cells will be separated for immunological and viral analysis. Samples will be shipped to the University of Siena, Italy, for SARS-CoV-2 neutralising antibodies and for assessment of neutralisation assays, and to Karolinska Institutet, Sweden, for neutralisation assays and for assessment of SARS-CoV-2 diagnostics (WP2 of EuCare Project). The shipment of samples will follow specific regulations and procedures for packaging and international transport. All stored samples will be analysed and destroyed in accordance with current legal and ethical requirements.
Data collection
During the acute phase of COVID-19 disease, data collection involves gathering demographic, clinical, and viro-immunological data. Subsequently, data collection is requested at each follow-up time point. To facilitate this process, a specially designed data management suite has been implemented. It incorporates a security-focused architecture that enables manual data entry into case report forms and secure file transfer for bulk data uploads. This flexible approach accommodates the various data infrastructures employed by the participating centres, whether they collect data from databases or manually. All data is meticulously curated and consolidated within a centralised relational database.
Data collection procedures strictly adhere to the established protocol, the most recent version of the Declaration of Helsinki, the International Conference on Harmonization Good Clinical Practice (ICH-GCP) or ISO EN 14155 guidelines, the regulation 2016/679 GDPR, as well as all relevant national legal and regulatory requirements.
Data analysis
The study data will be presented as absolute numbers and percentages for categorical data, median and Interquartile Range or mean and standard deviation for quantitative variables. The incidence of post-COVID-19 condition will be calculated, and comparison between groups based on exposure factors of interest will be performed using appropriate statistical tests such as Student t-test or Mann Whitney and Chi-square test. Predictors of post COVID-19 condition will be investigated by fitting a univariable and multivariable logistic regression analysis which will be adjusted for possible confounders.
Due to potential losses to follow-up, not all hospitalised COVID-19 patients and patients diagnosed with SARS-CoV-2 infection will attend the post COVID-19 visit. This may result in a non-representative sample, potentially leading to collider bias and distorting the estimation of the effect of an exposure of interest. This bias can be addressed for hospitalised patients by using Inverse Probability Weighting through the creation of a pseudo-sample where individuals are weighted according to the inverse of the probability of being sampled for the post COVID-19 visit. A multivariable logistic regression model will be fitted to estimate the probability of COVID-19 hospitalised individuals to be included in the post COVID-19 cohort based on their characteristics (propensity score, PS), and weights inversely proportional to this probability will be generated (1/PS). Weighted logistic regression analyses will then be used to investigate predictors of post COVID-19 condition. Missing data will be handled using the missing “indicator method,” or alternatively using multiple imputations technique.
The sample size calculation will be determined by the number of expected events in the follow-up period. Based on the currently available data, we estimated an incidence of WHO post COVID-19 condition of 54% (95%CI 45–69%). We will use the comparison between the risk of developing post-COVID-19 condition in participants infected with the Delta versus Omicron viral variant of concern (VoC) for the sample size calculation. According to recent analysis, after controlling for vaccination and other confounders, the risk of hospitalisation comparing Omicron versus Delta variant ranged between 0.55 and 0.88 [21]. Our assumption is that if post COVID-19 condition occurs due to the direct effects of the virus (e.g., through disease severity, level of hyper-inflammation, or other), a similar difference could be seen when comparing the long-term risk of developing post COVID-19 condition by VoC. The cohort of 2,300 participants, accounting for 35% of follow-up, will still guarantee 90% power to detect a reduction in the risk of post-acute sequelae of at least 25% comparing participants infected with Omicron vs. Delta variants.
Data handling and statistical analyses will be performed by the team of statisticians and Artificial Intelligence experts in WP5. Python computing will be used for executing the statistical analysis.
