Effect of Alertness Level and Backrest Elevation on Skin Interface Pressure

Introduction. Critically ill patients may experience reduced mobility and sensation related to various pharmacologic therapies and treatments, making this patient population especially susceptible to pressure ulcers. An alert patient may be better able to reposition in response to discomfort, therefore preventing the development of pressure ulcers. However, little is known about the effect of an individual’s alertness level on skin interface pressures. This study describes the effect of alertness level and backrest elevation on skin interface pressures. Materials and Methods. Fifty healthy participants were recruited from the Virginia Commonwealth University (Richmond, VA) population. Participants simulated each of 2 alertness levels (sedated or alert) while in 3 backrest elevations (30°, 45°, or 60°). Activity level, backrest elevation, and interface pressures were recorded continuously for 30 seconds. Random effects models were used to examine the effects of alertness level and backrest elevation on average and peak pressure. Participants had a mean age of 30 and 82% were female. Results. There was a significant interaction between alertness level and angle as related to average pressure (P < 0.0001) and peak pressure (P < 0.0001). Increases in backrest elevation increased average pressure and peak pressure. Interface pressures were generally greater when participants were simulating the alert state. Conclusion. These findings may indicate that interface pressure is a poor indicator of patient discomfort. Higher body mass index (BMI) was associated with higher average pressure (P < 0.0001), but lower peak pressure (P < 0.0001), suggesting better pressure distribution across the patient’s body area. These findings are similar to previous studies in which low BMI is associated with increased pressure ulcer risk.

Pressure ulcers occur primarily as a result of pressure, or pressure in combination with shear and/or friction, and may be affected by several intrinsic and extrinsic factors.^1-3 Decreased mobility and sensation are significant factors that contribute to the increased pressure experienced by the critically ill.^4-8 Pharmacologic therapies, such as sedatives and analgesics, can affect a patient’s alertness level and may, therefore, impair the patient’s ability to reposition himself or herself in response to discomfort from compressive forces.

  Only a few studies have described the effect of the patient’s ability to reposition and the risk of pressure ulcer development. Exton-Smith and Sherwin⁹ observed in a hospital geriatric unit that 90% of elderly patients who made fewer than 10 movements over a 7-hour period developed a pressure ulcer. However, those who made at least 54 movements did not develop a pressure ulcer. Barbenel et al¹⁰ observed the movements of 40 hospital patients and showed that patients identified as being at risk of developing a pressure ulcer according to the Norton Scale for Assessing Risk of Pressure Ulcers made a reduced number of movements. Spontaneous movements made by patients may be responsible for a temporary increase in skin blood flow, thereby reducing the risk of pressure ulceration.¹¹

  Patients receiving mechanical ventilation (MV) may be particularly affected by low alertness levels due to administration of sedatives or neuromuscular blockers.^12-14 Additionally, to prevent ventilator-associated pneumonia, guidelines suggest that MV patients be positioned at backrest elevations > 30°. Increased backrest elevation has been shown to increase interface pressure¹⁵ and shear.¹⁶ It is important to note that, while shear plays an important role in the development of pressure ulcers, it is not possible to directly measure shear of the skin as this would depend on being able to observe the underlying fascia. However, it can be deduced that the greater the movement of a patient, the greater the potential for shear. Studies have shown that when shear strains are involved, the pressure required to occlude capillaries decreased.¹⁷ Therefore, an increase in either shear or pressure will increase the pressure ulcer risk; however, little is known about whether a patient’s alertness level can influence the effect of backrest elevation on interface pressure.

  The specific aim of this study was to describe the effect of alertness level and backrest elevation on skin interface pressures. Using adult volunteers in a laboratory setting, skin interface pressure data were collected during simulated alertness levels reflective of those found in the critically ill (ie, alert or sedated) at varying degrees of backrest elevation.

The study was conducted in the Clinical Learning Center of the Virginia Commonwealth University (Richmond, VA) and approved by the Institutional Review Board. Because the study involved collecting data of 2 different alertness conditions with the same participant, the use of hospitalized patients would have required frequent monitoring to identify a period of alertness or sedataion. Additionally, equipment setup following the presence of an alert or sedated episode could be intrusive and may have required a considerable amount of time, during which the patient’s status could change. Because of the difficulties associated with obtaining real-time data from critically ill patients, the use of volunteers with simulated behaviors was deemed a reasonable substitute. A sample of 50 healthy participants was obtained from volunteers 18 years of age or older. Exclusion criteria included sacral skin disorders, neuromuscular disorders (eg, cerebral palsy or Parkinson’s disease), inability to move (eg, stroke), or inability to speak English. Self-reported demographic information was collected from the study participants (ie, age, weight, height, and gender).

  Alertness level. This study compared 2 alertness levels: alert and sedated. Patients in the intensive care unit receiving MV may be sedated to tolerate having an endotracheal tube; however, this level of sedation may result in general decreased sensation and decreased muscle tone. Due to ethical considerations, study participants were not actually sedated but were asked to simulate a sedated state. The sedated state was described to volunteers as relaxing their body completely (ie, going limp) and not moving in response to a change in backrest elevation. Thus, healthy volunteers simulated the way a sedated patient might slide down toward the bed’s bending points when the head of the bed is elevated. It is important to note, however, that the simulated sedated state cannot reflect the effects of decreased muscle tone. When simulating an alert state, volunteers were asked to respond to feelings of discomfort as they normally would, repositioning themselves if they felt the need.

  An actigraphy watch (Basic Motionlogger, Ambulatory Monitoring Inc, Ardsley, NY) was placed on the nondominant wrist of each participant to measure and verify participant alertness level. Wrist actigraphy has been shown to be associated with level of sedation and subjective sedation-agitation scale scores, such as the Richmond Agitation Sedation Scale and the COMFORT Scale for Pain Assessment,^18,19 and was used to simulate alertness levels as previously described.²⁰ The data stored in the actigraphy watch were downloaded at the end of each data collection session and synchronized with interface pressure data post hoc.

  Backrest elevation. Backrest elevations of 30°, 45°, and 60° were used to simulate the usual levels of backrest elevation found in the critical care setting. Backrest elevations were chosen based on the recommended elevation for a MV patient (30°), a typical elevation for a patient in upright positioning therapy (60°), and a transitional elevation angle between the 2 settings. An inclinometer was used to confirm the backrest elevation angle. The inclinometer is a custom-designed device that consists of 3 accelerometers (Model ADXL203, Analog Devices, Norwood, MA), each of which were attached to the 3 steel pivoting sections of the bed (ie, backrest, hip, and knee). Each accelerometer features a measurement range of ± 90° and is capable of sensing a minimal change of position of 0.01°. The analog voltages of the 3 sensors were sampled through a data acquisition device (NI-USB 6009, National Instruments, Austin, TX) and collected using a software program (LabVIEW 8.01, National Instruments, Austin, TX). Data from the inclinometer could be visualized on the laptop screen using the software program and was used to verify the backrest elevation angle during the data collection process.

  Skin interface pressure. Interface pressures were measured using a pressure-sensing mat (XSensor pressure mapping systems, XSensor Corp, Calgary, CA). The mat consists of a 48 x 144 grid of capacitive sensors calibrated to a pressure range of 0 to 200 mm Hg with an accuracy of ± 10%. The grid has a total sensing area of 24 in x 72 in and a total thickness of 0.04 in. The mat was secured into place on a standard hospital bed with a standard hospital sheet placed over it. Data were collected at a sampling frequency of 2 frames per second.

  Procedures. The pressure sensing mat and inclinometer were set up on a standard hospital bed and the actigraphy watch was placed on the participant’s nondominant wrist. A calibration procedure was performed before data collection began to ensure each participant was in the same body position in relation to the hospital bed’s bending points. During the calibration procedure, the backrest elevation was raised to 45° and the participants were asked to reposition themselves in such a way that the natural bending points of their bodies aligned with the bed’s bending points. Then, the bed was returned to a flat position.

  Before the bed was raised and data collection began, volunteers were asked to simulate a sedated state by remaining completely relaxed and avoiding any body adjustment to increasing backrest elevation. Backrest elevation was then increased to a randomly selected angle, either 30°, 45°, or 60°, and the participant was asked to simulate a sedated state for 30 seconds. While simulating a sedated state, the participant was asked to refrain from repositioning his or her body in response to discomfort.

  The participant was then asked to simulate an alert state for 30 seconds. During the alert state, the participant was told they could reposition their body until a comfortable position was reached. If the participant felt he or she was already in a comfortable position, then no repositioning was necessary. After a 30 second rest period, the procedure was repeated until data for all 3 study angles were collected. Throughout the entire procedure, a study researcher visually verified participant alertness level and backrest elevation using the program to read data from the inclinometer.

Statistical Analysis
Descriptive statistics were computed to describe the age, weight, height, and gender for the sample of participants. Pressure images were analyzed using a numerical computing enviroment (MATLAB [matrix laboratory], MathWorks, Natick, MA) to segment the images into 5 regions of interest that represent the body areas where skin interface pressure effects are most often seen: left scapula, right scapula, sacrum, left heel, and right heel.²¹ Descriptive statistics for skin interface pressures experienced throughout the whole body and each body segment were calculated by backrest elevation. Random effects models were used to examine the effects of alertness level and backrest elevation on pressure. Statistical analysis was performed using software (JMP software, SAS Institute, Cary, NC).

Participants. Fifty participants were recruited for this study (Table 1). Based on actigraphy summed counts, the 2 alertness levels were significantly different from each other regardless of angle (alert, mean 991.7 ± 1124.7; sedated, mean 550.4 ± 1126.0; P = 0.0002).

  Peak pressure. Peak pressure measured across the whole body, sacrum, and heels generally increased with increased backrest elevation; peak pressure measured across the scapula generally decreased as backrest elevation increased (Figure 1). Peak pressure measured over the whole body was typically greater in the alert condition than the sedated condition (Figure 1a). There was a significant interaction between alertness level and angle with respect to peak pressure (P = 0.0369). Thus, the differences in peak pressures between the alertness levels were not the same for all backrest elevation angles. Specifically, whole-body peak pressure differences were greatest when backrest elevation was 45° and least when backrest elevation was 30° (Figure 1a). Sedated subjects had lower peak pressure at 30°, 45°, and 60° than alert subjects, though differences were smaller and nonsignificant at 60° (P = 0.1451).

  Sacrum peak pressure increased as backrest elevation increased (Figure 1b). Peak pressures at 60° were an average of 1.84 mm Hg greater than at 30° (difference between pressure at the 60° elevation and the 30° elevation = 1.84 mmHg, standard error [SE] = 0.44). Sacrum peak pressures increased for the alert condition at backrest elevations of 30° (difference between pressure at alert and sedated states at 30° elevation = 1.84 mm Hg, SE = 0.44) and 45° (difference between pressures at alert and sedated states at 45° elevation = 3.06 mm Hg, SE = 0.44). At backrest elevation of 60°, there was no significant difference between the simulated states.

  Significant differences in scapula peek pressures were observed in both the right and left scapula. Scapula peak pressures decreased as backrest elevation was increased (Figure 1c and d); where peak pressures at 60° were less than at 30° (alert: difference = -13.71 mm Hg, SE = 1.46; sedated: difference = -17.98 mm Hg, SE = 1.22). For the left scapula, alertness level had a significant effect only at backrest elevation of 60°, with alert condition lesser than sedated condition (difference = -1.93, SE = 0.18). For the right scapula, alertness level had a significant difference on peak pressures between the alert condition compared to sedated at backrest elevations of 30° (difference = 1.03 mm Hg, SE = 0.19), 45° (difference = 0.73 mm Hg, SE = 0.19), and 60° (difference = -0.75 mm Hg, SE = 0.19).

  Significant differences in heel peak pressures were only observed in the left heel. Left heel peak pressures decreased for the alert condition at backrest elevations of 30° (difference between alert and sedated states, 30° elevation, at the left heel = -3.26, SE = 0.66), 45° (difference between alert and sedated states, 45° elevation, at the left heel = -10.16, SE = 0.65), and 60° (difference between alert and sedated states, 60° elevation, at the left heel = -12.53, SE = 0.65). Body mass index (BMI) was significantly related to peak pressure (P < 0.0001). Specifically, a higher BMI was associated with a decrease in peak pressure. Age, weight, height, and gender did not have significant effects on peak pressure.

  Average pressure. Average pressures measured across the whole body increased with increased backrest elevation (P < 0.0001). Average pressures were significantly greater for the alert condition. However, differences in average pressure between alert and sedated states were no more than ± 1.0 mm Hg such that, while statistically significant, they are not likely to be clinically relevant.

  There was a significant interaction between sedation state and angle as related to average pressure (P < 0.0001). As BMI increased, the average pressure increased. Sedated subjects had a lower average pressure at 30°, 45°, and 60° than alert subjects, though differences were smaller and nonsignificant at 60°. As the angle increased, so did the average pressure for both alert and sedated subjects. Within subject type (alert vs sedated), average pressure at 45° was significantly greater than that for 30°, and average pressure at 60° was significantly greater than that for 45°. Body mass index was significantly related to average pressure (P < 0.0001).

  Additional findings. Body mass index was found to be significantly related to both average pressure and peak pressure and, for this reason, its effect on pressure within the different conditions was explored further. Observations were divided by BMI, with a BMI ≤ 25 considered normal and a BMI > 25 considered overweight. Participants who were overweight had significantly lower peak pressures than participants with normal weight (Figure 2a). For participants who were overweight, there was no difference in peak pressure with respect to simulating an alert or sedated condition. Participants with normal weight, however, had significant differences between alert and sedated simulated conditions (P = 0.0031 for peak pressure, P = 0.0032 for average pressure), and this difference increased especially as backrest elevation increased.

Patient repositioning can play a crucial role in the prevention of pressure ulcers.²² Typically, nurses reposition a patient in an effort to minimize interface pressures. However, few studies^9-11 have demonstrated the role of a patient’s ability to reposition himself in lowering interface pressures.

  Effect of alertness level. Volunteers experienced greater peak and average whole-body pressures after they had repositioned themselves. This contradicts the suggested hypothesis that self-positioning is initiated to reduce pressure, and may demonstrate that interface pressure alone is a poor indicator of patient discomfort. Instead, it appears participants felt more comfortable placing a greater load on specific areas of their body and that the exact location of these areas may be patient-specific. While the results of this study do not account for the effect of decreased muscle tone resulting from actual sedation, it is reasonable to assume that a lack of muscle tone might cause more of a patient’s weight to load tissues that are not accustomed to such a load. In short, the magnitude of the compressive force may not be as important as the location of that compressive force. However, due to the anatomical differences inherent in each patient, finding this optimal location may be exceedingly difficult.

  Effect of backrest elevation. As supported by the literature, interface pressures increased as backrest elevation increased.^15,16 Peak pressures occurring in the sacrum increased with backrest elevation, while peak pressures in both scapulae decreased. Pressures may be offloaded from the scapula and distributed to the sacrum as the patient is elevated closer to a sitting position. In other words, pressure on the whole body increases with backrest elevation. The results of this study, therefore, support the recommendation that the elevation of the bed be kept at ≤ 30° for patients considered at risk of developing pressure ulcers, with the understanding that a decreased backrest elevation can lead to increased risk of ventilator-associated pneumonia. It is also important to note that pressure measured at certain body areas decreased (ie, the scapulae), while pressure observed at other areas increased (ie, the sacrum and heels). It is therefore important to measure pressure at different parts of the body to understand the effects of various clinical interventions on interface pressure.

  Recently, Behrendt et al²³ used a continuous bedside pressure mapping system to provide real-time feedback of optimal body position to reduce pressure ulcers. They found significantly fewer pressure ulcers in patients using the mapping systems. Because the immediate effects of alertness level and backrest elevation on skin interface pressures are unknown to the bedside care provider unless continuous pressure mapping is available, these systems may be beneficial to monitor and reduce the risk of pressure ulcers in the near future.

  Finally, it should also be noted that, although shear was not directly measured in this study, it may be possible to use relative movement as a corollary measurement of shear. Further investigation is necessary to determine the role of shear in a clinical setting.

  Effect of body mass index. Higher BMI led to higher average pressure, but lower peak pressure. Participants with higher BMI may have had better pressure distribution across their body area. This finding is similar to previous studies, in which low BMI is associated with increased pressure ulcer risk.^24-26

  Volunteers experienced greater interface pressures across their whole body when simulating an alert state. This differs from the suggested hypothesis that an alert patient would be able to reposition themselves in such a way as to reduce pressure. It may be that the participants in this study felt more comfortable offloading pressure on certain, more sensitive areas of their body at the compromise of placing greater compressive forces on areas better suited for the higher load. This may suggest that the intensity of interface pressure is not a suitable marker for determining whether or not a patient is comfortable. Instead, frequency of repositioning to offload compressive forces on tissues may be a more important factor.

Affiliations: Virginia Commonwealth University, Richmond, VA; and Air Force Institute of Technology, Dayton, OH

Correspondence:
Anathea Pepperl, PhD
aapepperl@vcu.edu

Disclosure: The authors disclose this work was supported by funding from the National Institutes of Health (NIH R01 NR010381, MJ Grap, PI). Christine Schubert Kabban was an employee of the federal government when this work was conducted and prepared for publication.