The presence of pressure injuries has been considered a quality indicator of health care services and efforts has been made to develop guidelines to prevent this issue. Intensive care unit admissions (ICU) are recognized as risk factors to develop pressure injuries due to reduced mobility of critically ill patients and large number of devices and hard technologies that jeopardize preventive measures, such as, decubitus changes. There are not current definitive evidence about the superiority of any support surface to treat or prevent pressure injuries. The objective of the present study is to analyze if the use of viscoelastic support surface in vulnerable critically ill patients decreases the incidence of pressure injury compared with pyramidal foam support surface.

Randomized clinical trial performed in an intensive care unit for adult patients in a philanthropic hospital. Inclusion criteria are patients admitted to ICU with Braden scale ≤ 14. Exclusion criteria are age under 18 years, less than 24 hours of ICU length of stay, contraindication of performing complete institutional preventive measures for support injuries, presence of support injuries at ICU admission or absence of informed consent. Randomization will be made by computerized generated numbers and patients will be allocated in two groups in a ratio of 1:1. All study patients will be cared for according to standard institutional preventive measures. The interventional group will be placed in an ICU bed with viscoelastic support surface and the control group in an ICU bed with pyramidal foam support surface. The main outcome evaluated will be the occurrence of type II pressure injury. Secondary outcomes are the time to the occurrence of pressure injury, length of ICU and hospital stay and 28 days mortality rate. Significance level will be 5%.

A randomized clinical trial with intent-to-treat analysis will be performed involving critical patients at high risk for development of pressure ulcers, that is; those presenting a Braden ≤ 14 scale at ICU admission in a one year period.

According to the Braden scale, six subscales or risk factors are validated: 1 - sensory perception, 2 - humidity, 3 - activity, 4 - mobility, 5 - nutrition and friction, and 6 - shearing. The total score varies from 6 to 23 points and patients are classified as: very high risk (scores less than or equal to nine), high risk (10 to 12 points), moderate risk (13 to 14 points), low risk (15 to 18 points), and without risk (19 to 23 points).

The exclusion criteria is aged less than 18 years, length of stay in the ICU less than 24 hours, contraindication for the performance of the standard PU prevention measures of the institution, presence of PU at ICU admission, and absence of the Informed Consent Term (ICT).

Randomization will be performed using a computerized table and patients will be allocated into two groups. Patients in the intervention group will be cared for according to the standard PU prevention measures of the institution and use a viscoelastic mattress as a bedding surface with the following characteristics: conventional density foam, density 40, and viscoelastic, density 60, polyurethane coated reinforced with 100% polyester mesh. Patients in the control group will be cared for according to standard PU prevention measures of the institution and use a pyramidal foam mattress as a support surface with a density of 33.

The institution's PU prevention measures are: apply body moisturizer after bathing without massaging bony prominences or areas with hyperemia; inspect the skin in the cephalo-caudal direction, especially the areas of bony prominences every 24 hours to observe hyperemias, dryness, heat, hardness, and maceration; rigorously evaluate whether the entire oral or enteral diet was ingested, taking into account the gastric residue and change in decubitus position every 2 hours; evaluate the need for application of dressings to protect bony protrusions from friction; always use the traction to mobilize the patient in the bed.

As this is a non-pharmacological intervention, the blinding of the health team is not possible, however the statisticians responsible for the analyzes will be blind regarding the identity of the treatments. After the randomization, identification data will be collected, date of birth, gender, weight will be measured by the patient lift (Jack 250®), height measured by a tape measure, severity score (SAPS 3) on ICU admission, score of organ dysfunctions (SOFA), diagnosis of ICU admission, and the Braden scale.

During the patient follow-up, the following data will be collected daily until discharge from the ICU; physical examination for the detection and classification of pressure ulcers, maximum vasopressor dose, and the Braden scale. After the first day, the accumulated fluid balance will be collected and in the course of the follow-up the date of discharge from the ICU and from the hospital. The main outcome of the study is considered as the occurrence of a stage 2 PU. Secondary outcomes are time spent in the ICU and hospital. The research was approved by the National Commission for Ethics in Research (CONEP), through the Certificate of Presentation for Ethical Appreciation (CAAE) no. 51644915.0.0000.5231, and the Committee for Ethics in Research Involving Humans of the Evangelical Beneficent Association of Londrina - AEBEL, according to Opinion no. 3/2015, of October 30, 2015. 4.3.1.

The data will be analyzed in the MedCalc program for Windows, version 9.3.2.0 (MedCalc Software, Mariakerke, Belgium), establishing analyzes pertinent to this research. The level of significance adopted is 5% and the confidence interval is 95%.

To demonstrate the difference between the study groups, considering the outcome variable the occurrence of stage 2 PUs, a significance level of 95%, power of 80%, ratio between exposed and unexposed of 1:1, frequency of outcome in the unexposed group of 80%, and frequency of outcome in the exposed of 30%, a sample size of 60 participants was calculated, 30 in each group.

For the comparison of two groups of continuous variables with independent samples, the Student t test will be used for the variables with normal distribution. For cases where distribution is not normal, the Mann - Whitney test will be applied. Kaplan Meier curves will be constructed and the Log-rank test applied to compare the main outcome (PU stage 2) between the two study groups.

Inclusion Criteria:

• patients admitted to ICU with Braden scale ≤ 14

Exclusion Criteria:

• age under 18 years
• less than 24 hours of ICU length of stay
• contraindication for performing complete institutional preventive measures for support injuries
• presence of support injuries at ICU admission
• absence of informed consent