Incidence of pressure injury in patients with moderate and severe traumatic brain injury: a systematic review Authors: Ranya Nascimento Ribeiro, Daniel Vieira de Oliveira, Wellingson S Paiva, Regina Marcia C Sousa, Rita de Cassia A Vieira Published: 2024 ABSTRACT AbstractBackgroundPatients with moderate and severe traumatic brain injury (TBI) admitted to the intensive care unit (ICU) may develop pressure injury (PI) due to haemodynamic instability caused by the disease, lack of mobility in bed, as well as intense and prolonged compression in prominent bone areas.ObjectiveThe objective of this review is to assess the incidence and identify risk factor for the development of PI in patients with moderate and severe TBI admitted to the ICU.MethodSearches were conducted in the PubMed, CINAHL, Scopus, Embase, Web of Science, Google Scholar, Trove and Open Grey databases, including all records found up to May 2023. Patients with moderate and severe TBI admitted to the ICU were included in this review.ResultsA total of 368 studies on PI and TBI were identified in the databases. Two authors assessed study bias and extracted data, wit. h a third reviewer as arbitrator. Six studies met the inclusion and exclusion criteria in the review. The incidence of PI varied between 6.5% and 20% among the included studies. Only two studies applied the Braden Scale, which identified stage II lesions (52.6% and 51.5%), located in the sacral region (78.9% and 54.6%). The risk factors identified in the studies for the development of PI were mechanical ventilation, TBI severity, vasoactive drugs, age, fever, use of enteral nutrition, haemoglobin levels and time to perform tracheostomy.ConclusionThe incidence of PI in moderate and severe patients was similar to that found in ICU patients. There were significant differences across the various studies in the ways in which data were collected and reported.Trial registration numberThe protocol has been deposited in the PROSPERO repository (CRD42023428817). Strengths and limitations of this study Although there is a substantial body of literature on intensive care unit patients with moderate and severe traumatic brain injury, only a few studies address the incidence of pressure injuries. Although there is a substantial body of literature on intensive care unit patients with moderate and severe traumatic brain injury, only a few studies address the incidence of pressure injuries. Among the six studies reviewed, merely two employed comparable methodologies for evaluating pressure injuries, which restricted the effectiveness of meta-analysis, sensitivity analysis and subgroup analysis. Among the six studies reviewed, merely two employed comparable methodologies for evaluating pressure injuries, which restricted the effectiveness of meta-analysis, sensitivity analysis and subgroup analysis. The major challenges in our analyses stemmed from insufficient details regarding the development, location and stage of pressure injuries. The major challenges in our analyses stemmed from insufficient details regarding the development, location and stage of pressure injuries. There is an absence of standardised methods for assessing pressure injuries. There is an absence of standardised methods for assessing pressure injuries. The studies included in our systematic review were notably heterogeneous. The studies included in our systematic review were notably heterogeneous. Introduction Pressure injury (PI) refers to localised damage to the skin or underlying tissues, predominantly observed in hospitalised patients due to pressure or friction on bony prominences.1 The development of PI is associated with extended hospitalisation periods, increased hospital costs, clinical complications and adverse effects on the patient’s physical or mental health.2 PIs are classified by severity into several grades: Grade I involve non-blanchable erythema; Grade II includes partial-thickness skin loss; Grade III encompasses full-thickness skin loss reaching the subcutaneous fat layer; and Grade IV represents complete tissue loss, exposing muscles, tendons or bones.3 4 Patients with severe traumatic brain injury (TBI) often exhibit decreased levels of consciousness and mobility, predisposing them to prolonged periods of immobility. This significantly heightens their risk of developing PI. This condition significantly increases the risk of developing PI.135 6 A systematic review spanning 10 articles on adults in intensive care units (ICU) revealed a cumulative incidence of PI ranging from 3% to 39%. Most studies employed visual inspection to identify PI, with only three studies specifying the injury’s grade.7 Patients with TBI in ICUs face complications from both primary and secondary injuries following trauma. Secondary injuries may lead to the development of PIs during hospitalisation.468 Factors such as impaired mobility, prolonged compression on bony prominences, compromised perfusion and oxygenation, injury severity, extended duration of mechanical ventilation, poor nutritional status, use of vasoactive drugs, extended ICU stays and limited physical mobility significantly contribute to PI risk.2 4 8 9 TBI is known to have long-term physical, cognitive and psychological consequences following injury, including the development of PI, as well conditions like hypoxia, reductions in haemoglobin levels, use of vasoactive drugs, hypotension and medical devices due to prolonged hospital stays and injury severity.10 11 Notably, the incidence rate of medical device-related pressure ulcers in ICU patients ranges from 0.69% to 8.33%, with the ear being the most commonly affected area.12 Despite the extensive research on factors associated with PI in patients with TBI, a systematic review estimating the incidence of PI in those with moderate-to-severe TBI admitted to the ICU has not yet been conducted. Such a review is important for evidence-based practice, offering a comprehensive evaluation of existing literature and providing synthesised insights into the incidence, risk factors and outcomes of PIs in this vulnerable population. Understanding the incidence of PI among patients with TBI is critical for multiple reasons: it helps estimate nursing workloads in the ICU, identifies gaps in current research, guides future research priorities and supports the implementation of effective preventive measures. Additionally, this knowledge aids in cost estimation and systematic healthcare planning. This review aims to thoroughly evaluate the incidence of PI in patients with moderate-to-severe TBI admitted to the ICU, identify and analyse clinical and demographic factors contributing to the occurrence of PI and provide a comprehensive overview of these injuries. The findings are intended to inform clinical guidelines, emphasising effective prevention and management strategies to enhance patient care and improve clinical outcomes. Methods This systematic review was reported following the Joanna Briggs Institute (JBI) Manual for Evidence Synthesis and synthesis without Meta-analysis for prevalence and incidence reporting items.13 14 The protocol was registered at the Prospective Register of Systematic Reviews (PROSPERO; Center for Reviews and Dissemination, University of York; and the National Institute for Health Research). Search strategy The initial search was performed simultaneously in all databases, including all records found in these databases until May 2023. The searches were carried out for records indexed in the following databases: National Library of Medicine National Institutes of Health (PubMed), Cumulative Index to Nursing and Allied Health Literature (CINAHL), Scopus, Embase, Web of Science, Google Scholar, Trove and Open Grey, using keywords or descriptors in accordance with the Health Sciences Descriptors (DeCS), Medical Subject Headings (MeSH) and Emtree from Elsevier Life Science (Emtree). No filters regarding the language or date of publication restrictions were used. The strategy involved the PI and TBI descriptors were combined using the Boolean operator ‘AND’, in addition to being customised according to the characteristics of each database. Online supplemental material 1 describes the search strategy carried out in each database including studies published until May 2023. After searching the databases, the records were stored in the EndNote tool (EndNote X9, Thomson Reuters) classified in different folders and duplicates were excluded using the EndNote program (EndNote X9, Thomson Reuters). Selection criteria The criteria for including titles in the review was based on the CoCoPop mnemonic for research questions as described by the JBI methodology for systematic reviews.14 For the research questions we used population (Pop): patients with moderate and severe TBI; condition (Co): PI; context (Co): patients who were admitted to the ICU. The guiding research questions were: What is the incidence of PI in patients with moderate and severe TBI who were admitted to the ICU? What are the associated factors for PI in patients with moderate and severe TBI who are admitted to the ICU? The population criteria for both questions are synonymous. Thus, patients with moderate and severe TBI who presented PI during hospitalisation in an ICU were included in the study, without age range restrictions, were included in the review. The following criteria for excluding records were adopted: studies that did not provide detailed information regarding the incidence of PI in patients with moderate and severe TBI; studies in which the patient’s admission to the ICU was not reported; experimental studies on animals; studies without abstracts or without full articles; and full texts were not available, even after attempting to contact the corresponding authors by email for two consecutive weeks. Theses, dissertations, book chapters, case series, commentaries, editorials, forum summaries, literature reviews, systematic reviews and meta-analyses were also not included in the review. Patients with mild TBI were excluded from the review because they are generally less severe, less life-threatening and are infrequently admitted to the ICU. Information sources and study selection A selection of articles was conducted by two independent reviewers (RNR and DVO). Titles and abstracts were screened using an online software programme (Rayyan, Qatar Computing Research Institute) and applying the eligibility criteria by the reviewers separately. Any discrepancies were resolved by a consensus discussion with a third reviewer (RCAV) who was involved in making a final decision, if necessary. After reading the titles and abstracts, the studies that would be included in the review were defined and were read in full by two independent reviewers (RNR and DVO) and the third author was only be contacted if there was a discrepancy in the results (RCAV). Data extraction and management After reading the articles included in the review in full, the following information was extracted: year of publication, country of publication of the study, objective, type and direction of studies, research location, sample size, age, severity of TBI, scales for classification of PI, incidences of PI, location of PI, average time for the development of PI, statistical test applied, factors associated with the emergence of PI and length of hospital stay. The accuracy of the information extracted was confirmed in a meeting, and a third (RCAV) reviewer was consulted in case of any disagreement. The collected data were categorised and entered into a Microsoft Office Excel 2019 spreadsheet and organised into tables for better interpretation of the categorised variables. At this stage, a manual search for additional studies was also performed in the reference sections of the included studies by the first and second reviewers. To assess the influence of confounding factors on the outcomes of the included studies, a systematic methodology was adopted. Initially, key confounding factors such as comorbidities, sex, length of stay, Glasgow Coma Scale and age were identified and documented for each study. An analysis was conducted to determine whether these factors were adjusted for or controlled within the result analyses. This information is compiled in table 1 and online supplemental table 1, which detail the treatment of these confounding factors as patient characteristics in each study. The JBI Critical Appraisal Checklist for Studies Reporting Prevalence Data14 was used to assess the potential for bias introduction. The implications of these factors on the study outcomes and their overall interpretation are discussed in the Results and Discussion sections. Table 1Distribution of publications included in the review (n=6)TitleYearCountry of publicationStudy designStudy locationAgeSample sizeType of TBIPressure ulcer in patients with severe traumatic brain injury: significant factors and association with neurological outcome162013IndiaProspective cohortAdult patients admitted within 24 hours of severe TBI to the department of neurosurgery.18 and 60 years.89 patientsSevere TBI.Tracheostomy timing in traumatic brain injury: a propensity-matched study172014Canada and the USARetrospective cohort135 American College of Surgeons’Trauma Quality Improvement Programme.Over 16 years.1142 patientsAbbreviated Injury Scale higher than or equal to 3 (moderate and severe).Timing of tracheostomy placement among children with severe traumatic brain injury: a propensity-matched analysis182019USARetrospective cohortNational Trauma Data Bank Research Dataset (2007–2015).Under 15 years.361 patientsSevere TBI Abbreviated Injury Scale ≥3.Incidence and risk factors associated with pressure injury in patients with traumatic brain injury152020BrazilProspective cohortReference hospital for neurotrauma in the city of Manaus.Mean of 33 years (18–85).240 patients110 mild TBI (45.8%), 69 moderate (28.8%) and 61 (25.4%) severe.Identification of adverse events in paediatric severe traumatic brain injury patients to target evidence-based prevention for increased performance improvement and patient safety192020CanadaRetrospective cohortPaediatric intensive care units of the Children’s Hospital at London Health Sciences Centre.Under 18 years.193 patientsSevere TBIGlasgow Coma Scale ≤8 and a head maximumAbbreviated Injury Scale ≥4.Risk factors associated with pressure ulcers in patients with traumatic brain injury admitted to the intensive care unit202022Republic of KoreaRetrospective cohortThe trauma ICU level ITrauma Centre in South Korea.Mean of 62 years (SD 19.5).237 patientsMean GCS was 10.36 (SD=4.52).GCS, Glasgow Coma Scale; ICU, intensive care unit; TBI, traumatic brain injury Distribution of publications included in the review (n=6) GCS, Glasgow Coma Scale; ICU, intensive care unit; TBI, traumatic brain injury Evaluation of the methodological quality and summary measures The methodological quality of the studies included in the review was assessed by three independent reviewers using the JBI Critical Appraisal Checklist for Studies Reporting Prevalence Data.14 This tool consists of nine questions, referred to as domains, aimed at assessing the sample structure of the studies incorporated. All reviewers deliberated on each appraisal item while evaluating the methodological quality of each study design in their review. This exchange occurred before carrying out the appraisal independently.14 For items Q1 to Q4, the methodology assessed studies that provided comprehensive details on the selection of patients with moderate and severe TBI admitted to the ICU. These included considerations of adequate sample size and extensive descriptions of research participants. Items Q5 to Q7 evaluated whether the studies identified potential biases, employed validated methods and used reliable instruments to assess the incidence of PI. Items Q8 and Q9 focused on the accuracy of the statistical analyses, the completeness of the results and the identification of any incomplete or missing data. Each item was categorised as ‘yes’, ‘no’ or ‘unclear’ based on the information provided in each study.14 The incidence of PI was assessed in all patients with moderate and severe TBI admitted to the ICU, serving as the primary outcome. Data were synthesised descriptively. We evaluated heterogeneity within studies based on methodological and outcome characteristics. Methodologically, heterogeneity was assessed by examining differences in sample sizes and the scales used for assessing PI. This involved a thorough comparison of study methodologies, including inclusion and exclusion criteria and approaches to data collection, to identify variations that could contribute to heterogeneity. For outcome characteristics, we analysed variability in reported incidence rates of PI, as well as differences in definitions or classifications of outcomes across studies. Data synthesis was conducted through narrative analysis, supported by tables to facilitate comparisons among studies. We reviewed the inclusion criteria of each study to ensure compliance with our predefined selection standards. The methodological quality was evaluated using established criteria to ascertain the reliability and validity of the findings. Variations in study methodologies and results were examined to identify potential sources of variability. This comprehensive approach allowed us to effectively summarise and compare data, providing a clear understanding of the incidence and risk factors associated with PI in patients with moderate and severe TBI. A meta-analysis was deemed inappropriate due to the high heterogeneity observed among the included studies. Moreover, only two studies provided data on the incidence of PI using specific instruments designed for this purpose. This lack of uniformity in data reporting and measurement tools precluded the combination of results in a meta-analysis. Additionally, the studies varied significantly in their methods of evaluating PI, including whether PI was assessed at ICU admission and the age ranges included, thereby introducing substantial heterogeneity. Data analysis Descriptive analyses of the incidence rates were initially conducted, followed by a pooled analysis.14 The incidence rate of PI was calculated by dividing the number of PI cases by the total number of patients with moderate and severe TBI admitted to the ICU. All descriptive analyses were conducted using Jamovi (The Jamovi Project, 2022, V.2.3), available at https://www.jamovi.org. Patient and public involvement statement None. Results A total of 368 studies on PI and TBI were found in the databases, of which 59 were excluded due to duplications, leaving 309 studies for analysis of titles and abstracts. In the second stage of the study, 295 studies were suppressed considering the exclusion criteria, leaving 14 studies to be read in full. Of these 14 studies, only six were included in the review1520 (online supplemental material 2). Figure 1 shows the study selection flowchart. Flow diagram of the study selection process in the systematic review. *Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all daabases/registers). **If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools. PI, pressure injury; TBI, traumatic brain injury. Table 1 describes the characteristics of the six articles conducted on 2.262 patients (ranging from 8916 to 114217) included in the review.1520 Of the six articles included, only two used the Braden Scale to assess PI,15 20 one assessed PI based on another less well-known scale,15 but with the same purpose as the Braden Scale and the other three articles did not describe the specific instrument to assess PI.1719 Incidence of PI Although the studies employed varied methods for identifying PI in patients with moderate and severe TBI admitted to the ICU, five studies151720 provided data on the number of injuries acquired during the ICU stay. One study16 was excluded from the analysis due to insufficient detail regarding the incidence of PI in the ICU, which hindered accurate inclusion in the meta-analysis, sensitivity analysis and subgroup analysis. The results indicated that the proportion of patients with moderate and severe TBI who developed PI was 17% (95% CI 6% to 28%). The data in online supplemental table 1 show that the PI incidence varied between 6.5% and 20% among the included studies. In paediatric studies, the mean PI incidence was recorded at 7.85%18 and 18.7%.19 For studies involving adult patients with TBI, the incidences were 6.5%17 and 15.3%,16 while studies including adults and/or older patients reported incidences of 13.9%20 and 20%.15 Although three studies1719 did not primarily focus on evaluating PI incidence as the main outcome in patients with TBI, they were retained in the review because they provided data on PI incidence in patients with moderate and severe TBI. The incidence of PI following moderate and severe TBI was reported in all six studies (figure 2). Three of these studies noted that PIs occurred after ICU admission,15 19 20 with the incidence of PI at ICU admission ranging from 13.9% to 20%.15 19 20 Notably, only three studies explicitly listed in their exclusion criteria patients who had experienced PI prior to ICU admission.15 19 20 Among these, the study by Stewart et al19 highlighted adverse events occurring during paediatric ICU stays. In contrast, the Brazilian study15 reported a notably higher incidence of PI compared with others: 55% within 5 days and 82.2% within 10 days of ICU admission. Additionally, another study20 indicated that the average onset time of PI was 14.21 days (SD=10.33) during the ICU stay, with 54.6% (18/33) of patients developing PI after 15 days. Summary of pressure injury incidence. PI, pressure injury. Risk factor of PI Online supplemental table 1 presents the risk factors of PI assessed in the six included studies. Due to the significant variability in the risk factors evaluated across the studies, conducting a pooled analysis was not feasible.1520 The significant risk factors of PI were age (OR 1.03, 95% CI 1.00 to 1.005, p<0.01), moderate TBI (OR 7.03, 95% CI 1.86 to 26.47, p<0.01), severe TBI (OR 10.91, 95% CI 2.70 to 44.08, p<0.01), injury severity score ≥25 (OR 0.06, 95% CI 0.007 to 0.459, p=0.007), norepinephrine use (OR 1.04, 95% CI 1.01 to 1.07, p<0.01), infusion of vasopressors (OR 3.94, 95% CI 1.065 to 14.543, p=0.040), use mechanical ventilation (OR 11.54, 95% CI 1.534 to 86.847, p=0.018), presence of 7 days or more of fever (OR 5.61, 95% CI 1.169 to 26.965, p=0.031), enteral feed (OR 5.65, 95% CI 1.6 to 19.9, p=0.027), enteral feeding for more than 7 days (OR 6.87, 95% CI 1.178 to 40.082, p=0.032), mean change in haemoglobin at 2 weeks (OR −2.07, 95% CI −3.5 to −0.7, p=0.046), early tracheostomy (OR 0.43, 95% CI 0.26 to 0.71 p=0.001), early tracheostomy rate ratio 0.25; 95% CI (0.08 to 0.80) and Braden Scale Score (OR 0.56, 95% CI 0.343 to 0.926, p=0.024).151720 Only one study did not assess the risk factors of PI after moderate and severe TBI.19 It is important to highlight that the studies by Alali et al 2014,17 McLaughlin et al 2019,18 and Charyk Stewart et al19 did not have the main objective to evaluate the incidence and risk factors for the development of PI after moderate and severe TBI. Interestingly, the presence of comorbidity was not associated with PI in two15 20 studies, while four studies did not evaluate this association with PI because they had children18 19 as a subject, did not include the variable in the study16 and PI was a secondary outcome of the study.17 Quality assessment Among six included studies,1520 two studies15 20 were assessed as fully meeting all criteria of the JBI Critical Appraisal Checklist for Studies Reporting Prevalence Data,14 as shown in online supplemental table 2. Regarding the domain of sample size (question 3) one study16 received an uncertain judgement due to it is the inclusion of 89 patients without adequate description of the sample size or the statistical power of the study.16 For the domain concerning the validity of the method used for identifying the condition (question 6), two studies17 18 received negative judgements and one study19 was marked as uncertain. Similarly, negative evaluations (n=2)17 18 and uncertainty (n=1)19 were noted in question 7, primarily because these studies1719 either did not use a validated scale to assess PI or failed to provide precise information about the scale used. It is important to note that the primary focus of these studies1719 was not to evaluate the incidence and risk factors for PI, which may explain some of the methodological limitations observed. Discussion This systematic review found that the pooled analyses PI incidence was 17% (95% CI 6% to 28%), with a variation of 95% CI 6.5% to 20.0%. This finding contrasts with systematic reviews of other ICU populations, which report a broader range of PI incidence rates. Specifically, pooled data from these reviews show cumulative PI incidence rates of 95% CI ranging from 10.0% to 25.9%,7 and for medical device-related PI, between 0.9% and 41.2%.12 The observed variability across studies can be attributed to differences in population characteristics, the nature of the included studies, the severity of illness among patients, types of medical devices used and the tools employed for assessing PI.712 1520 These differences underscore the importance of using standardised assessment tools and methodologies to better compare and understand the risk factors contributing to PI across diverse ICU populations. Furthermore, studies that employed validated instruments to assess PI reported incidences ranging between 13.9% and 20%.14 18 20 In comparison, a systematic review with meta-analysis of 22 studies, including adult patients in the ICU, found that only seven used tools for skin inspection to identify PI, resulting in a 95% CI for cumulative incidence ranging from 9.4% to 27.5%.7 In our review, the Braden Scale was applied in two18 20 of the three studies that used an instrument to assess PI. The Braden Scale is widely used in clinical practice and uses domains to assess and identify the risks of PI.21 The scale assesses sensory perception, humidity, activity, mobility, nutrition, friction and shear. The Braden Scale had moderate predictive validity for PIs in ICU adults with an Area under the Curve (AUC) of 0.78, predictive sensitivity of 0.89 and specificity of 0.28. Using the scale to identify PI risks is important for nurses and other health professionals, making it possible to identify vulnerable points and reinforce constant assessment as a form of prevention in clinical care.22 However, it is uncertain whether risk assessment using the Braden Scale has any impact on PI incidence.23 Stage II and the location of PI in the sacral region were the most frequent in the studies evaluated in our systematic review. Unlike our findings, a systematic review with meta-analysis that evaluated the incidence and prevalence of PI in cardiac patients admitted to the ICU showed that stage I PI was more frequent (pooled proportion 17%; 95% CI 8.3% to 28.3%) than stage II (the pooled proportion 13.1%; 95% CI 5.8% to 22.7%).24 However, considerable heterogeneity was observed in both these analyses,24 with I2=98.5% for stage I and I2=98.5% for stage II PI, indicating high within-study variance. Chaboyer et al7 and Fulbrook et al24 reported that the most common location of PI frequency was the sacral region (50.9%; 26.9–48.0%), constituting data similar to our findings. Because our sample included patients with moderate and severe TBI admitted to the ICU, immobilisation caused by haemodynamic instability and severity of the injury can lead to broken skin, exposing patients to increased risk of infection and worsening the outcome.146 The impairment of functionality in patients with moderate and severe TBI in the acute phase of trauma (after 2 weeks of trauma) can be evidenced in a study by McCrea et al,25 in which 79% of patients with moderate TBI and 94% with severe TBI had moderate-to-severe disability or worse functionality and approximately 80% required assistance in basic aspects of everyday function. This functional incapacity caused by the severity of TBI may contribute to developing PI.18 20 Preventive strategies are essential to reduce the incidence of PI in patients with moderate and severe TBI admitted to the ICU, such as prevention bundles, repositioning and turning of patients, prevention of medical device-related PI, education in prevention, regular risk assessment and an expert nurse available to support and guide practice.26 A systematic review with meta-analysis published by Cochrane in 2020 showed that although repositioning the patient is effective in preventing PI, there is no consensus in the literature regarding the appropriate time to reposition the patient in bed. The repositioning frequency should consider the individual’s activity and pain levels, sleep disruption, ability to independently reposition, cost and nurses’ time.27 Multiple risk factors were associated with the development of PI in patients with moderate and severe TBI in our review. Systematic reviews of patients admitted to the ICU show similar results to those evidenced in our review, in which age (OR 1.06), Braden Scale (OR 1.22), admission to the ICU due to trauma (p<0.001), longer usage time of medical devices (OR=1.11), use of vasoconstrictors (OR 6.07), fever (OR 3.44), ICU stay time (OR 2.06), enteral nutrition (OR 2.12), mechanical ventilation (p<0.01) and neurosurgical ICU (OR 6.221) were risk factors for developing PI.2831 Some authors recommend applying machine learning (ML) algorithms to predict the risk of PI early to minimise the effects caused by the occurrence of PI in hospitals.32 Application of ML models show an excellent pooled AUC of 0.94, sensitivity of 0.79 (95% CI 0.78 to 0.80) and specificity of 0.87 (95% CI 0.88 to 0.87) in predicting PI.32 The combination of biomarkers, socio-demographic and clinical variables, medical history and environmental factors can result in accurate PI prediction models that are easy to apply in clinical practice. However, it is important to highlight that these models must undergo external validation before being implemented at different care levels. Out of the six studies included, only two15 20 received a fully positive evaluation across all domains of the JBI Critical Appraisal Checklist for Prevalence Studies.14 Variability in methodological quality, particularly regarding sample size and reliable methods for identifying PI, significantly impacted the overall synthesis. One study16 received uncertain judgements regarding sample size due to incomplete reporting, which undermined confidence in the reliability of its findings. Moreover, several studies were negatively or uncertainly judged concerning the validity of the methods used to identify PI, either due to the absence of a validated scale or insufficient details about the scales applied. The lack of a reliable measurement tool can lead to an underestimation of the number of injuries identified, underscoring the importance of reliable measurement tools for accurately detecting and quantifying injuries.13 14 This variability in study quality revealed potential biases and inconsistencies in the data, emphasising the need for robust and well-documented methodologies.13 14 Furthermore, the primary focus of most studies did not include an assessment of the incidence and risk factors for PI in patients with moderate and severe TBI. These limitations in quality assessments likely impacted the robustness of our findings, as only two studies provided detailed and reliable information regarding the incidence, location and severity of PI. Limitations Our review is the first on PI incidence in moderate and severe TBI. The study has the following limitations: (1) Six studies were included in the systematic review; however, only two employed comparable methodologies for evaluating PI. This limited evidence base may restrict the generalisability of the results, potentially affecting their broader applicability. The power of the review might be insufficient to detect significant effects or variations, which could undermine the robustness of the conclusions. Such variability in definitions and methods curtailed the feasibility of performing meta-analyses, sensitivity analyses and subgroup analyses; (2) although all studies in our review reported that patients were admitted to the ICU, it was not consistently possible to determine from all included studies whether PIs developed during the ICU stay or if patients were already admitted with these injuries. The lack of detailed information regarding the development, location and stage of PI posed significant challenges in our analyses; (3) there was a notable lack of standardisation in the measurement methods used to assess PIs; (4) despite numerous publications concerning ICU patients with moderate and severe TBI, only a few studies have reported on the incidence of PI; (5) it is important to highlight the heterogeneity of the studies included in our systematic review. Additionally, it was not feasible to aggregate the risk factors contributing to the development time of PI. To ensure the accuracy of the data presented, we contacted two authors from the original studies to clarify uncertainties regarding the incidence of PI and the severity of TBI reported in their research. We recommend the establishment of a standardised method for identifying, collecting and reporting the incidence of PI in patients with TBI in the ICU. The limited of the evidence base restricts our capacity to formulate extensive conclusions and accentuates the necessity for additional research. It is necessary that such research incorporate standardised assessment tools alongside more comprehensive data sets to both expand and refine the applicability and representativeness of the results. The findings of this review, offering substantial insights into the incidence of PI among patients with moderate and severe TBI, must be approached with circumspection due to methodological inconsistencies and definitional variability. This constrained evidence base thus not only limits the scope of our conclusions but also underscores the critical need for further empirical studies. These studies should aim to standardise methodologies and broaden data sources to enhance the reliability and broader relevance of the outcomes. The incidence of PI in patients with moderate and severe TBI is frequently under-reported by the care team, who often fail to document the development or absence of PI throughout the entire hospitalisation period. Despite the workload challenges faced by nursing staff in most services, it is imperative that the entire care team meticulously records the various stages of PI, their locations, the onset times and the number of injuries observed in the different sectors where patients with TBI are admitted. Such comprehensive documentation of PIs will not only foster a more thorough understanding of the issue but also facilitate the implementation of specific measures to enhance care quality, prevent and mitigate PIs, ensure better distribution of professionals in units with greater workload demands and reduce hospital costs. The engagement of the entire care team is essential to address this challenge effectively and promote the safety and well-being of patients with TBI. Conclusion Significant heterogeneity was observed among the included studies, preventing the execution of a meta-analysis. The PI incidence was found to be 17% in patients with moderate and severe TBI. Notably, only two studies used the Braden Scale as a validated assessment measure for PI. Stage II PI, most commonly located in the sacral region, were frequently reported across the studies. Identified risk factors for the development of PI include mechanical ventilation, the severity of the TBI, the use of vasoactive drugs, age, fever, enteral nutrition, haemoglobin levels and the timing of tracheostomy procedures. supplementary material