The goal of this observational study is to learn about the population of chronically critical ill patients who refer to respiratory rehabilitation units.

The main questions it aims to answer are:

• what are the clinical and pathophysiological characteristics of the population of chronically critical ill patients who refer to respiratory rehabilitation units after hospital admission with the need for prolonged invasive mechanical ventilation via tracheostomy?
• what is the response of this patient population to respiratory rehabilitation treatment in terms of functional recovery, weaning from invasive mechanical ventilation, weaning from tracheostomy tube, mortality and return to home?
• are there any appreciable indices at admission and/or in the first period of hospitalization which may have a prognostic significance both on short-term objectives (weaning from invasive mechanical ventilation, weaning from the tracheostomy tube) and in the medium-long term (survival, state of health)?

Participants will be subjected to a rehabilitation and weaning program consisting of:

• physical therapy,
• speech therapy,
• nutritional assessment and therapy,
• in subjects on invasive mechanical ventilation at admission a program of weaning and, if needed, shift from invasive to Non-Invasive Ventilation will be performed.

Observational, prospective, non-profit study. The first aim is to define the clinical and pathophysiological characteristics of the population of chronic critically ill patients who refer to respiratory rehabilitation units after hospital admission with the need for prolonged invasive mechanical ventilation via tracheostomy.

The second aim is to identify the following aspects: evaluating the response of this particular patient population to respiratory rehabilitation treatment (functional recovery, weaning from invasive mechanical ventilation, decannulation and home discharge), knowing the mortality of this patient population and evaluating the existence of appreciable indices at admission and/or in the first period of hospitalization which may have a prognostic significance both on short-term objectives (weaning from invasive mechanical ventilation, weaning from the tracheostomy tube) and in the medium-long term (survival, state of health). This last aspect could also be useful for the early identification of patients who could benefit from rehabilitation treatment more than others and therefore to optimize the resources at our disposal.

The present study involves the collection of data from chronically critical patients who will be referred to the Rehabilitation Pneumology Unit of the Don Gnocchi Foundation in Florence after hospitalization in intensive care for respiratory failure with the need for prolonged invasive mechanical ventilation.

The prospective analysis will have a total duration of 6 months.

The enrolled patients will follow the standard care path provided by the center and in this context they will carry out the rehabilitation treatment both within the department and in the dedicated respiratory gym.

The variables of interest will be extracted from the medical records at the time of entry (T0) to the department and at the time of discharge (T1).

It is expected to enroll 220 patients under ordinary hospitalization at the unit of Rehabilitative Pneumology of the Don Gnocchi Foundation of Florence.

The default sample size was set based on the recruiting capacity of the center involved in the project.

The data will be retrieved from the medical records of patients belonging to the department.

The folders will be viewed by the main investigator and his collaborators and the data of interest and the reports of the instrumental tests will be extracted.

Starting from the clinical and anamnestic evaluation upon admission to the department, patients who fall within the inclusion criteria will be identified.

The data will be entered anonymously into a computerized database, in which each subject will be associated with an alphanumeric identification code. The correspondence between the patients' name and numerical identity (progressive enrollment number) will be recorded in a separate table (association key) kept by the principal investigator.

The paper data relating to the study will be stored in an archive accessible only to the Principal Investigator, while the electronic data will be stored on servers located in the territory of the European Union on physical archives protected by double key encryption systems.

During the duration of the study, access to the computerized and anonymized database will be limited to the principal investigator, who may extend access to identified collaborators. The anonymized data will be retained for a period of 7 years after the conclusion of the study. Once the retention period indicated above has expired, the data will be made anonymous so that it is no longer possible to trace, directly or indirectly, the identity of the interested party. The anonymized data may be reused for subsequent research and therefore stored by the joint owners indefinitely.

For the analysis of the primary objective, a descriptive statistic will be used through which the clinical and clinical-functional variables recorded at admission (T0) will be presented. In particular, for continuous variables (or similar to such, e.g. age, Barthel score) mean and standard deviation or median and interquartile range will be used, depending on whether the variables have a normal distribution or not, as descriptors respectively of central tendency and dispersion. The normality or otherwise of the distribution of each variable will be verified by evaluating the asymmetry and kurtosis parameters of the distribution itself. As regards dichotomous variables (e.g. presence of infections), the frequency values in the observed sample and the relative percentages will be reported.

For the analysis of the first secondary objective, a comparison will be made between the values recorded at T0 and at discharge (T1) for the clinical and clinical-functional variables that require a reassessment at T1. The comparison will be carried out using a T-test for paired measures or a Wilcoxon Signed-Rank Test in the case of continuous variables, depending on whether or not they have an approximately normal distribution. In the case of categorical/ordinal variables, the Wilcoxon Signed-Rank Test will be used and the McNemar's test in the case of dichotomous variables. For all tests a p-value<0.05 will be considered significant.

For the analysis of the second and third secondary objectives, logistic regression models will be used to evaluate the association between the independent variables (variables recorded at admission and their changes in the first weeks) and the dependent variables (weaning from invasive mechanical ventilation, weaning from tracheostomy tube, survival and health status).

Furthermore, the changes in parameters measured at admission and at discharge will be compared through repeated measures analyses. In particular, for continuous variables, a repeated measures ANOVA test or a Friedman test will be used, depending on whether the variables have a normal distribution or not. For dichotomous variables, a Cochran's Q test will be used. For all tests, a p-value<0.05 will be considered significant.

Inclusion Criteria:

• age >18 years;
• episode of acute or acute-on-chronic respiratory failure with the need for prolonged invasive mechanical ventilation via tracheostomy

Exclusion Criteria:

condition of clinical instability such as to make it necessary to transfer the patient to acute hospital within the first 7 days of hospitalization.

Patients with respiratory failure who have undergone prolonged invasive mechanical ventilation via tracheostomy.