Medical Device-Related Pressure Injury in an Intensive Care Unit: A Cross-Sectional Study

BACKGROUND: Medical devices can cause pressure injuries. PURPOSE: This study was conducted to determine the prevalence of and factors associated with medical device-related pressure injury (MDRPI) in an intensive care unit (ICU). METHODS: A cross-sectional study was performed among adult patients (at least 18 years of age) admitted to an ICU in a referral hospital in Brazil between December 2019 and February 2020. The skin of patients who consented to participate was assessed for the presence of an MDRPI, and the use of all medical devices was noted. Other independent variables (sociodemographic variables, medical history, pressure injury risk factors, medications, and length of hospitalization) were abstracted from the medical records. Bivariate data analysis included Pearson’s chi-square test or Fisher’s exact test; odds ratio and a confidence interval of 95% also were established. Correlation among independent variables and MDRPI was determined using the ρ Spearman correlation test, and a hierarchical binary logistic regression analysis was performed using statistically significant variables from the bivariate analysis. P < .05 was considered statistically significant. RESULTS: The 125 study participants ranged in age from 15 to 97 years (mean, 63.02 ± 19.2), 76 (60.8%) were men, and 76 (60.8%) were White. Of the 125 participants, 43 (34%) experienced MDRPI; the total number of MDRPIs was 58 (3 patients had 3 injuries, and 7 patients had 2 injuries). Of those 58 MDRPIs, 46 were stage 1, and 12 were stage 2. Polypharmacy (> 4 medications) was a significant risk factor for MDRPI. Use of a nasal catheter, cord for orotracheal tube fixation, oximeter, intra-abdominal pressure equipment, and indwelling urinary catheter was significantly associated with the presence of MDRPI. Renal and respiratory diseases and the presence of infection were positively related to the presence of MDRPI. CONCLUSION: Medical device-related pressure injury was prevalent in this patient population. Most of these injuries were stage 1, which suggests that frequent monitoring and device repositioning (when possible) may help prevent more serious injuries. Additional research involving other hospitals in Brazil is needed to increase the understanding of the prevalence and risk factors of MDRPIs in patients in the ICU.

Hospital-acquired pressure injuries remain an important concern. Pressure injury typically involves damage to the skin or underlying soft tissue, usually over bony prominences, and may be related to medical equipment or another type of device.^1,2 The presence of a pressure injury is a negative indicator of the quality of care, impacting clinical, social, and economic aspects of the lives of patients and health care institutions.³ The consequences of a pressure injury not only affect that person but also health care providers.⁴

Pressure injuries are classified according to the degree of involvement of the skin tissue and mucous membranes or caused by medical devices. The National Pressure Injury Advisory Panel (NPIAP) defines medical device-related pressure injury (MDRPI) as wounds resulting from the use of devices designed and applied for diagnostic or therapeutic purposes. These injuries usually conform to the pattern or shape of the device and are staged similarly to other types of pressure injury.⁴

According to prevalence and analysis research⁵ of more than 7000 individuals in the United States and Canada, MDRPI usually is caused by poor positioning of equipment that is tightly secured to prevent loss of the device or to ensure a secure seal, as well as failure to change the insertion site of the device. That study also found that 1 in 3 of the most serious pressure injuries is related to medical devices and that the face, head, and auricular region are the most commonly affected areas.

The incidence and prevalence of MDRPI vary widely depending on the population (adult or pediatric) and the type and shape of the device. Patients in intensive care units (ICU) are at high risk of MDRPI development because of immobility, sedation (inability to report pressure or discomfort), and the essential use of medical devices for treatment.⁵

Determining MDRPI prevalence and characterizing this type of injury regarding independent variables, such as patient sociodemographic and clinical characteristics, and their effect on MDRPI development may enhance the understanding of how nursing practice can impact prevention, early identification, and treatment. This has important ramifications not only for patients but also because pressure injuries are used as a parameter for evaluating care in hospital settings.³

According to a systematic review,⁶ epidemiological estimates of MDRPI are important for assessing the extent of the problem globally and informing health organizations in developing guidelines for MDRPI prevention and management. To the authors’ knowledge, few publications describe MDRPI in Brazilian health services. The purpose of this study was to evaluate the prevalence of MDRPI in an ICU in Brazil.

This cross-sectional, analytical study was conducted in an ICU of 40 beds in a large hospital that is an important referral center in the state of Minas Gerais, Brazil. Data were collected between December 2020 and February 2021.

Population and sample. To determine the study population size, a sample calculation was used that considered an average occupancy rate of 90% of beds per month and a median length of stay of 5.9 days per patient,⁷ with a margin of error of 5% and a confidence interval (CI) of 95%. A sample of 144 patients was estimated. Inclusion criteria stipulated that participants must be adults (at least 18 years of age) of either sex who were admitted to the ICU and had used a medical device for more than 24 hours.

The researchers explained the objectives of the study to the participants, who provided signed informed consent. If the patient was sedated or confused, the consent form was signed by a family member or guardian.

Data collection. Data were collected once for each patient during the 45-day study period using a semistructured form that captured sociodemographic and clinical data from the printed medical records as well as the records created during skin inspection. Data were collected to spreadsheets by 3 nursing students and a research nurse trained in the assessment of MDRPI.

Study variables. The independent variables collected from the medical records included length of hospital stay, sex, age (dichotomized into younger than 60 years and 60 years or older), race, marital status, comorbidities as listed in the medical record, and polypharmacy (type and number of drugs prescribed) quantified as the use of 4 or more drugs regardless of drug class. Risk factors of pressure injury included immobility, paralysis, skin retractions, xerosis, moist skin, history of alcoholism, history of smoking, presence of infection, clinical evaluation of edema, dehydration, malnutrition, circulatory failure, and altered blood pressure (hypertension or hypotension).

Body mass index (BMI) was calculated by patient-reported weight and height and classified as underweight, normal, overweight, obese, and extremely obese according to criteria by the World Health Organization (WHO).⁸ For study purposes, results were dichotomized as normal when BMI was between 18.5 and 24.9 kg/m² and other values were considered as altered BMI.

Skin was inspected by the researchers to determine the quantity, types, and locations of medical devices (nasogastric tube, nasal catheter, central venous access, tracheal tube, peripheral venous access, electrodes, cords for orotracheal tube fixation, tracheostomy, oximeter, intra-abdominal pressure device, and indwelling urinary catheter) as well as the presence of MDRPI (outcome variable).

Statistical analysis. The data recorded on the form were entered into a Microsoft Excel version 2016 spreadsheet and exported for analysis by the Statistical Package for the Social Sciences version 23.0 (IBM, Inc.). All analyses (prevalence of pressure injuries, presence of pressure injury and clinical characteristics, presence of comorbidities, risk factors, and devices used) involved only patients with pressure injuries, regardless of the number of lesions in each patient.

For bivariate analysis, Pearson’s chi-square test or Fisher’s exact test was used; P < .05 was considered statistically significant. Odds ratio (OR) and a CI of 95% also were established. To determine the correlation among some characteristics and MDRPI, the ρ Spearman correlation test was performed and considered as follows: 0.8 to 1.0, strong correlation; 0.5 to 0.8, moderate correlation; 0.2 to 0.5, weak correlation; and 0.0 to 0.2, insignificant correlation.

A hierarchical binary logistic regression analysis was performed among the variables that showed P < .005 from the bivariate analysis. The Wald chi-square test was used; P < .05 was considered statistically significant.

Ethical considerations. The study complied with the ethical precepts established by the Helsinki Declaration and research ethics regulatory bodies in Brazil, which determine that any data, direct or indirect, related to human beings to be used in research can only be collected with the approval of a Research Ethics Committee. Approval to conduct the study was obtained from the Ethics Committee in Research With Human Beings, located at the Federal University of São João del-Rei and also an Ethics Committee of the study hospital.

Patients and injury characteristics. Of the 144 eligible patients recruited for the study, 19 declined to participate, which left 125 participants (mean age, 63.02 ± 19.2 years; 76 [60.8%] male; 76 [60.8%] white; 68 [54.4%] single; 2.8 [standard deviation (SD), 1.57] risk factors per individual). The average number of medical devices used by a participant was 5.5 (range, 2–10; SD, 1.99). Average length of ICU stay was 101.4 hours (range, 25–792; SD, 130.20).

Of the 125 participants, 43 (34%) experienced MDRPI; the total number of MDRPIs was 58 (3 patients with 3 injuries, and 7 patients with 2 injuries). Of those MDRPIs, 46 were stage 1 and 12 were stage 2.

Factors associated with MDRPI

Patient and clinical factors. Of the 43 patients who had an MDRPI, 28 (65.11%) were male (P = .564; OR = 1.322; CI, 0.615-2.843), 24 (55.81%) were White (P = .408; OR = 0.729; CI, 0.344-1.544), 24 (55.81%) were single (P = .852; OR = 0.917; CI, 0.426-01.925), 22 (51.16%) had no altered BMI (P = .851; OR = 0.905; CI, 0.432-1.894), 31 (72.1%) had polypharmacy (P = .000; OR = 5.262; CI, 2.341-11.830), 39 (90.7%) had at least 1 risk factor for developing pressure injury (P = .180; OR = 0.244; CI, 0.043-1.389), 22 (51.6%) were 60 years of age or older (P = .572; OR = 1.282; CI, 0.611-2.689), and 41 (95.4%) had at least 1 comorbidity (P = 1.000; OR = 1.285; CI, 0.206-7.996) (Table 1). Only polypharmacy showed a statistically significant association with the presence of MDRPI (P = .011; OR = 0.322; CI, 0.137-0.756).

A Spearman analysis regarding the number of injuries and patient age (ρ = –0.010; P = .911), patient age and number of risk factors (ρ = 0.123; P = .170), patient age and length of hospitalization (ρ = –0.138; P = .125), number of injuries and risk factors (ρ = 0.238; P = .008), number of injuries and number of devices (ρ = 0.176; P = .049), and length of hospitalization and number of injuries (ρ = 1.97; P = .028) showed insignificant or weak correlation.

Devices. Among the 43 patients who used a nasogastric tube, 3 experienced injuries (P = .039; OR = 3.050; CI, 2.366-3.932). Seven (7) injuries were noted among 24 patients who used a nasal catheter (P = .000; OR = 3.278; CI, 2.496-4.304), 1 injury was noted among 48 patients with central venous access (P = .344; OR = 2.952; CI, 2.309-3.770), and 5 injuries were noted among 30 patients with a tracheal tube (P = .018; OR = 10.658; CI, 1.203-94.410). In addition, there were 3 injuries among 102 patients with peripheral venous access (P = .117; OR = 6.075; CI, 0.612-60.270), 1 injury with a cord for orotracheal tube fixation (P = .013; OR = 3.103; CI, 2.396-4.017), and 2 injuries among 8 patients with a tracheostomy, (P = .117; OR = 3.000; CI, 2.337-13.852). Eighteen (18) injuries were noted in 114 patients with an oximeter (P = .000; OR = 4.280; CI, 3.037-6.032), 2 injuries in 26 patients using intra-abdominal pressure equipment (P = .117; OR = 3.000; CI, 2.337-3.852), 3 injuries in 64 patients with external pressure equipment (P = .039; OR = 3.050; and CI, 2.366-3.932), and 9 injuries in 39 patients with an indwelling bladder catheter (P = .000; OR = 3.412; CI, 2.572-4.526) (Table 2). Thus, the use of nasal catheter, cord for orotracheal tube, external pressure equipment, and indwelling urinary catheter were significantly associated with MDRPI in relation to those who used such devices and had no lesions.

Comorbidities and risk factors. Of the 43 patients with MDRPI, 6 (14.0%) had cancer (P = .473; OR = 0.669; CI: 0.241–1.857), 10 (23.3%) had diabetes mellitus (P = .031; OR = 0.387; CI: 0.169–0.889), 8 (18.6%) had kidney disease (P = .008; OR = 6.019; CI: 1.506–24.052), 14 (32.6%) had heart disease (P = .281; OR = 1.716; CI: 0.752–3.916), 1 (2.3%) had infectious disease (P = 1.000; OR = 1.929; CI: 0.118–31.612), 10 (23.3%) had respiratory disease (P = .008; OR = 4.667; CI: 1.480–14.714), 6 (14.0%) had a stroke (P = .554; OR = 1.500; CI: 0.485–4.642), and 25 (58.1%) had hypertension (P = .848; OR = 0.848; CI: 0.419–1.884) (Table 3).

In terms of risk factors of pressure injury, of the 43 patients with pressure injuries, 6 (14.0%) were immobile (P = .091; OR = 0.091; CI, 0.841-11.888), 1 (2.3%) experienced limb paralysis (P = 2.309; OR = 2.952; CI: 2.309-3.776), 3 (7.0%) were restrained (P = .413; OR = 1.975; CI, 0.381-10.231), 8 (18.6%) had xerosis (P = .259; OR = 2.114; CI, 0.733-6.098), 8 (18.6%) had moist skin (P = 1.000; OR = 1.110; CI, 0.425-2.898), 7 (11.6%) had a history of smoking (P = .812; OR = 0.869; CI: 0.325–2.324), 20 (46.5%) had infection (P = .008; OR = 3.043; CI: 1.373-6.746), 22 (51.2%) had non-generalized edema (P = .084; OR = 2.020; CI: 0.952–4.287), 10 (23.3%) were dehydrated (P = .021; OR = 3.838; CI, 1.289-111.433), 24 (55.8%) were malnourished (P = .023; OR = 2.436; CI, 1.145-5.185), 11 (25.6%) had decreased tissue perfusion (P = .053; OR = 0.439; CI, 0.195-0.989), and 18 (41.9%) had altered blood pressure (P = .132; OR = 0.254; CI, 0.254-1.132) (Table 3).

Statistical analysis. Binary regression analysis was performed between the dependent variable (presence of MDRPI) and the various independent variables. Statistically significant independent variables (P < .005) in the bivariate analysis included diabetes mellitus, presence of renal and respiratory comorbidities, infection, dehydration, and malnutrition. Significant factors after adjustments by the regression model included renal comorbidity (P = .016; OR = 5.86; CI, 1.383-5.861), respiratory comorbidity (P = .011; OR = 4.734; CI, 1.427-15.708), and infection (P = .033; OR = 1.07; CI, 2.515-5.879) (Table 4). Table 4 also shows that renal and respiratory comorbidities, as well as the presence of infection, showed a positive relationship (presence of MDRPI), as seen in the OR, with a significant final model (X2Wald [1] = 4.509; P = .034 R2 Nagelkerke = 0.205).

The prevalence of MDRPI in this study was higher than observed in a systematic review⁹ and point prevalence study¹⁰ involving patients in the ICU (research that becomes reliable with statistics ≥10%). The results point to an association between polypharmacy, certain risk factors (infection, dehydration, and malnutrition), comorbidities (kidney and respiratory disease), and the use of specific medical devices (cord for orotracheal tube fixation, nasal catheter, oximeter, and indwelling urinary catheter).

Observational research showed that MDRPIs developed most commonly in intensive care.¹¹ An integrative review¹² of 9 original research articles and 3 case studies identified 11 devices most commonly associated with these injuries, including orotracheal tubes and noninvasive ventilation masks. That review also found that the majority of these lesions (66%) occurred in the cervical area and 40% in the nasal area, with pressure injuries also noted at other locations (around the tracheostomy, frontal region, occipital region, upper limbs, and calcaneus).

An integrative literature review¹³ and a pilot study¹⁴ showed that patients in the ICU usually had a higher risk of developing pressure injury and, consequently, MDRPI, which can be explained by the severity of illness of that population (ie, an unstable clinical picture) that requires more intense mechanical or drug interventions for life maintenance. Still, a case–control study¹⁵ affirmed that devices such as respiratory equipment, urinary catheters, sequential compression devices, and intravenous catheters are indispensable. Health care professionals need to have heightened awareness of the actions that may predispose the patient’s skin to injuries during the use of medical devices. Thus, procedures for systematic pressure injury dressing changes and alternating the insertion site of devices must be considered whenever possible.^13,16

Recommendations for the prevention of MDRPI^13,15 state that health care professionals should be vigilant, reposition devices (when possible) at pressure sites, and keep the skin dry and free of debris. There is also a need for assessment of the patient by a multidisciplinary team and constant reevaluation of the medical devices to help prevent MDRPIs.

The current study showed that polypharmacy was a significant factor in MDRPI occurrence. The use of vasoactive drugs can reduce blood flow and decrease tissue perfusion, making patients who rely on these medications more susceptible to pressure. In addition, observational research and systematic reviews showed that decreased sensory perception caused by the continuous use and often high doses of pain medication, muscle relaxants, and sedatives can lessen the response to excessive pressure by reducing the sensation of pain and impairing mobility.^13,16,17 However, adequate and individualized pharmacotherapeutic treatment with oral diabetes medications has been statistically associated with better glycemic control and, consequently, a reduction in the onset of the ill effects of diabetes mellitus, such as chronic wounds.¹⁸ Diabetes mellitus was related to the development of MDRPI in the current study. It is important to note that according to a systematic review, the lack of control of diabetes leads to complications in the circulation, renal system, nerves, and neurons, which can complicate the healing process and foretell the appearance of a pressure injury.¹⁹

The MDRPI associations with kidney disease and respiratory disease corroborate findings involving the effects of homeostasis changes, such as impaired gas exchange, circulatory disturbances, edema, hypoxia, malnutrition and dehydration, humidity, and infection³ and the development of pressure injuries.

The majority of participants in the current were classified as elderly. According to recent observational studies,^20,21 pressure injuries are related to the increase in overall life expectancy; geriatric patients and/or persons with an acute or chronic condition that compromises their daily activities are most affected by MDRPIs because the aging process changes the characteristics of the skin, making it more susceptible to this type of injury. However, in this study the variable age lacked significance.

Among medical devices, the cord for orotracheal tube fixation, nasal catheters, oximeters, external pressure equipment, and indwelling urinary catheters were associated with the emergence of MDRPI, reinforcing the findings of other studies.^13,22 The negative impact of these devices was due to the difficult visualization of the underlying tissue as well as the pressure constantly exerted on the skin, lack of repositioning, and possible lack of knowledge in detecting stage 1 MDRPIs.

These findings call attention to the association between the cord for orotracheal tube fixation (an external device) and the emergence of MDRPI. This same association also can be made for the oximeter; although it is considered a low-risk device in terms of skin damage, it was responsible for a surprising number of injuries in this study. Previous research^12,13 demonstrated similar concerns, which brings into consideration the quality of care in terms of the lack of repositioning of the medical device as well as the absence of other protective measures for the skin. This also may be associated with a lack of knowledge of pressure injury staging.

This study underscores the vulnerability of certain patients for a predisposition to MDRPI when in the ICU. A multidisciplinary team should plan and execute care to prevent and minimize MDRPIs.

The study was conducted in a hospital in Brazil that did not have a specific protocol to identify MDRPIs. In addition, 19 patients refused to participate in the study, interfering with the optimal sample size. This limitation also interfered with the collection period, which had to be extended from 30 to 45 days. Another important limitation was the low rotation of patients within the ICU, which led to a draw with a smaller number of patients for the collection.

Further epidemiological studies will be conducted by the researchers to increase understanding about MDRPI in other Brazilian hospitals.

A cross-sectional study of patients in the ICU using medical devices showed a 34% prevalence of MDRPI. The use of a nasal catheter, cord for orotracheal tube fixation, external pressure equipment, and indwelling urinary catheters were significantly associated with MDRPI. Renal and respiratory disease also were significant risk factors for the occurrence of MDRPI. Thus, medical devices should be used with care and caution in patients in the ICU, particularly in those with renal and respiratory disease.

Mr Estevam dos Santos and Ms Valadares Sinicio Abib are nurses, Ms Furtado Bueno is a nurse and master’s student, Dr Cortez is an adjunct professor of nursing, and Dr Moraes is an adjunct professor of nursing, Federal University of São João del Rei, Campus Centro Oeste, Divinópolis, Minas Gerais State, Brazil. Dr de Mendonça Figueirêdo Coelho is a professor, Department of Nursing, Federal University of Ceará, Fortaleza, Ceará State, Brazil. Address all correspondence to: Juliano Teixeira Moraes, RN, CETN, MD, PhD, Federal University of São João del-Rei, Campus Centro Oeste. Divinópolis, Minas Gerais State, Brazil; email: julianotmoraes@ufsj.edu.br.