INTRODUCTION
Acute respiratory distress syndrome (ARDS) occurs with pulmonary inflammation resulting in hypoxia; reported mortality rates are as high as 48%.1 Current standard of care for ARDS consists of mechanical ventilation, limiting tidal volume to less than 6 mL/kg of ideal body weight, and a plateau pressure of less than 30 cm H2O.2,3 In patients with moderate to severe ARDS (defined as PaO2/FIO2 [fraction of inspired oxygen] ratio of <150 mm Hg), prone positioning is recommended for over 12 hours per day.4,5 It is postulated that prolonged placement in a prone position improves perfusion of the ventral lungs and increases blood oxygen levels.6
The COVID-19 pandemic has sharply increased the incidence of ARDS, with approximately 28% of COVID-19 patients in the intensive care unit (ICU) being treated with prone positioning.7 A recent meta-analysis of the use of prone positioning in ARDS patients found evidence of decreased mortality; nevertheless, this intervention was also associated with higher rates of endotracheal tube obstruction and hospital-acquired pressure injuries (HAPIs).8 Research suggests that up to 57% of patients who had prone positioning developed pressure injuries.9,10 Unlike the majority of HAPIs, these pressure injuries typically occur on the face, thorax, and trochanter.11,12
Placing a patient in the prone position safely is a labor-intensive task that requires careful planning to prevent dislodgement of lines and tubes along with strategies to prevent HAPIs.7,13 Prone positioning can be achieved using a specialized bed frame that turns the patient automatically or by manual positioning; the latter usually requires more than 1 staff member.13 With increased ICU patients requiring prone positioning due to the pandemic, both modalities may be necessary due to specialty bed shortages.
Little is known about the influence of manual versus specialty bed-assisted prone positioning on the development of HAPIs. Prior to the pandemic, our facility exclusively used a specialty bed capable of placing patients in the prone position (Rotoprone; ArjoHuntleigh Inc, San Antonio, Texas). However, due to a sudden increase in patients admitted with ARDS during the COVID-9 pandemic, we found it necessary to place patients in a prone position manually. The aim of this study was to compare HAPI incidence rates among patients placed in a prone position manually versus positioning using a specialty bed designed to facilitate prone positioning. We also compared mortality rates between these 2 cohorts.
METHODS
We retrospectively reviewed electronic health records (EHRs) of patients receiving care for ARDS including those managed by prone positioning. The study setting was a 355-bed, not-for-profit regional hospital in the Western United States (Stockton, California), specializing in cardiovascular care, women and children's, and 24-hour emergency services. The hospital has 4 adult ICUs with 39 beds, with 24/7 intensivist coverage; the facility also has a neonatal intensive care unit (NICU). Our facility had 18,095 admissions in 2020, of which 1600 were COVID-19 related. This study spanned a period of time in which there were 2 major surges of the COVID-19 pandemic. Initially, patients requiring prone therapy were placed exclusively on a specialty bed designed to facilitate prone positioning. However, due to a rapid and significant increase in demand for these beds during the COVID-19 pandemic, we manually placed many patients in the prone position as specialty beds supplies were exhausted. Standard pressure injury prevention care for patients placed in a prone position at our facility is evidence based and includes application of a soft silicone multilayer foam dressing to the face and all bony prominence areas (eg, shoulders, chest, iliac crest, elbows, tibia).14,15 Thin foam dressings are also placed under all medical devices. These preventive interventions were used for all patients during data collection regardless of the method used for prone positioning. Pressure injury staging was based on National Pressure Injury Advisory Panel (NPIAP) guidelines.16
The target population was adult patients with an International Classification of Diseases, Tenth Revision, Clinical Modification codes for ARDS and COVID-19 admitted to one of our adult ICUs from July 1, 2019, to January 31, 2021, and placed in a prone position for more than 24 hours. All patients were treated per our ARDS protocol similarly except for manual versus bed-assisted prone positioning. Study procedures were reviewed by Dignity Health Institutional Review Board (IRB # CANV DHIRB-2020-568). The committee deemed this study exempt from individual informed consent March 29, 2021.
Study Procedures
Demographic data collected included age, sex, race/ethnicity, and body mass index (BMI). Pertinent clinical data such as method of prone positioning, time spent in the prone position, use of sedation, paralytic agents, and steroids were also collected. We used the Midas Healthcare Analytics Solutions risk adjustment model (https://www.midasplus.com/Overview.htm) that risk adjusts individual patient encounters by assigning probabilities and expected values for mortality, length of stay (LOS), readmissions, and complications to measure severity of illness. Other clinical variables measured were hospital LOS, ICU LOS, and improved oxygenation (defined as a decrease in positive end expiration pressure [PEEP] and FIO2 within 48 hours after prone therapy was initiated). Patient-level data were retrieved from an administrative database (demographic information) and the EHR system (clinical variables, such as prone positioning and development of HAPI wound care notes). All extracted data were documented using an Excel data sheet (Microsoft Corporation, Inc, Redmond, Washington) in a de-identified manner.
Data Analysis
Data analyses were completed using "R" software version 3.6.2 (Microsoft Corporation). Descriptive statistics were calculated for all variables. Frequency tables were obtained for categorical variables, and means and standard deviations were calculated for continuous variables. Univariate analysis was used to identify potential factors associated with an increased likelihood of developing an HAPI. Specifically, [chi]2 were used to analyze associations between dichotomous variables and independent-samples t tests were used to analyze in continuous variables.
RESULTS
We identified records of 231 patients who experienced ARDS during the data collection period; 164 patients (70.9%) were placed in a prone position. Among these, 106 (64.2%) were positioned manually, 54 (32.9%) were positioned using the specialty bed, and 4 (2.4%) were placed using both methods. Patients placed using both methods were removed from analysis, leaving a final cohort of 160 patients. A majority of the patients were male (n = 93; 58.1%), White (n = 62; 38.7%), and non-Hispanic (n = 74; 46.2%) (Table 1). The study group's mean age was 60.96 years (SD = 12.7), and their mean BMI was 32.81 kg/m2 (SD = 7.25). A majority of the patients were diagnosed with COVID-19 (n = 142; 88.7%), and almost three-fourths (n= 117; 73.1%) expired. They spent an average of 14.79 days on the mechanical ventilator and 9.05 days in a prone position. Patients on manual prone positioning had a significantly lower use of neuromuscular blocking agent infusions (P = .0383) and spent significantly more days in a prone position (P = .0143; Table 2).
Patient Outcomes
Slightly more than half (n = 81; 50.63%) of patients developed HAPIs; 35.8% (n = 29) developed deep-tissue pressure injuries (DTPIs), and 27.2% (n = 22) developed more than 1 DTPI. Five patients (6.2%) developed stage I HAPI, 13.6% (n = 11) stage II, and 17.3% (n = 14) developed stage III, stage IV, and unstageable HAPIs (Figure 1). Pressure injuries varied in location based on the type of prone position (Figure 2). Analysis revealed association with the incidence of HAPIs using manual versus specialty bed for positioning (P = .96). Other types of skin injuries, such as mucosal membrane injury and blisters, occurred in both groups; specifically, of those that developed blisters, 75% (n = 18) occurred in patients placed in a prone position via the specialty bed as compared to 57% (n = 12) of patients who were placed in a prone position manually (Figure 3).
Analysis did reveal a significant association between the method of prone therapy and mortality. More patients who were manually placed in the prone position died (n = 85; 80.19%) versus 58.18% (n = 32) of patients positioned using the specialty bed (P = .003). As anticipated, the mortality risk for patients placed in a prone position (mean = 0.59, SD = 0.28) was significantly higher than those positioned via the specialty bed (mean = 0.49, SD = 0.23; P = .0156). Analysis revealed no significant differences between method of placing patients in a prone position and hospital LOS (P = .380), LOS in an ICU (P = .209), and improvement in oxygenation (P = .4842; Tables 2 and 3).
Factors Associated With Development of HAPIs
Univariate testing was used to identify potential factors associated with HAPI occurrences. A diagnosis of COVID-19 was not significantly associated with HAPI development, with 50% of patients with COVID-19 developing HAPI as compared to 44.4% of non-COVID-19 patients (P = .8462). In addition, we did not find any differences in BMI between those patients who had HAPI and those who did not (P = .3573).
DISCUSSION
Prone positioning for 12 to 16 hours per day is recommended treatment for ARDS patients by the Society of Critical Care Medicine, the American Thoracic Society, and the European Society of Intensive Care Medicine.17,18 Prone positioning arose as a strategy for improving oxygenation and reduce mortality from a report by Guerin and colleagues,10 who found significant reduction in mortality in ARDS patients10 maintained in a supine position versus patients placed in a prone position for more than 12 hours daily. Nevertheless, this strategy is also associated with an increased likelihood of adverse events such as HAPIs. We found that 51.83% of patients placed in a prone position developed an HAPI; this finding is consistent with other recent studies.10,19
The main aim of this study was to evaluate differences in the occurrence of HAPIs based on the method used to place patients in a prone position (manual vs specialty bed); no significant differences were found. We searched the literature and found a single study that addressed this issue. Morata and colleagues11 reported that manual prone positioning was associated with a lower HAPI rate than positioning using the same specialty bed in our study (P = .007); their sample was smaller than our study (N = 37). They also reported a significant cost savings based on specialty bed rental fees. In contrast to these findings, we found no significant difference in hospital LOS or oxygenation levels (the main purpose of prone positioning), based on technique for prone positioning. With supply chain shortages impacting rental beds, we suspect use of manual prone positioning has become increasingly prevalent in the management of patients with ARDS. Nevertheless, we also acknowledge several staff members are needed for this procedure to safely place patients in a prone position, implement HAPI preventive interventions, and prevent other serious complications such as accidental extubation. Cotton and colleagues20 advocated the formation of a "proning team" to alleviate the burden of this maneuver, reduce related complications, and protect the healthcare team from injuries related to this intervention.
We observed that patients placed in a prone position were more likely to experience HAPI of the face. The most common type of pressure injury was DTPI; we also observed medical device-related pressure injuries of the nares and ears. Other authors have explored prevention of these types of pressure injuries using prophylactic silicone foam dressings.13,21
Although our hospital used evidence-based practices for the prevention of HAPIs, we found that more than half of patients (51.83%) developed an HAPI. These critically ill patients are at higher risk for skin breakdown due to a number of factors such as hemodialysis and hypotension requiring vasopressors.22 Furthermore, pathophysiological factors predictive of HAPI development such as impaired tissue oxygenation and perfusion are common in COVID-19 patients.21,23 Similar to other studies, we found that that neither the presence of COVID-19 or BMI was significantly associated with the development of HAPIs.24,25
Analysis revealed a significant association between the method of prone therapy and mortality (80.19% of the patients managed by manual prone positioning vs 58.18% positioned using a specialty bed). The reason for this difference is not clear. All patients managed by manual prone positioning were diagnosed with COVID-19, and it unknown whether this diagnosis influenced the mortality rate.
Given the high rate of HAPIs seen in our sample, we strongly recommend additional research to determine effective pressure injury preventive interventions in this clearly vulnerable population. Possible interventions deserving of additional research include more frequent repositioning and use of pressure redistribution devices such as air fluidized pillows and silicone foam dressings.13,26 Adherence to evidence-based preventive actions can be a challenge to nurses when working under crisis conditions with increased workload during the pandemic surge. Using creative staffing models to support staff at the bedside could mitigate this strain on staffing.26,27
LIMITATIONS
This study has several limitations, with most linked to its retrospective design. Specifically, data collection was restricted to the clinical data available in the EHR. For example, the identification of HAPIs was exclusively based on the wound care nurse notes. We were unable to collect sufficient data and analyze the influence of other factors on HAPI development such as nutritional makers or vasopressor use.
CONCLUSIONS
We collected retrospective data in a group of clinically ill patients with ARDS requiring prone positioning and found no difference in HAPI occurrences based on technique for prone positioning. Additional research is needed to confirm our findings and compare the cost of prone positioning manually versus using a specialty bed. Additional research is also needed to identify effective preventive interventions in this vulnerable population.
REFERENCES