Bedside Technologies to Enhance the Early Detection of Pressure Injuries: A Systematic Review : Journal of Wound Ostomy & Continence Nursing

Secondary Logo

Journal Logo

Wound Care

Bedside Technologies to Enhance the Early Detection of Pressure Injuries

A Systematic Review

Scafide, Katherine N.; Narayan, Mary Curry; Arundel, Linda

Author Information
Journal of Wound, Ostomy and Continence Nursing: March/April 2020 - Volume 47 - Issue 2 - p 128-136
doi: 10.1097/WON.0000000000000626

Abstract

INTRODUCTION

Hospital-acquired pressure injuries (HAPIs) are a dangerous and costly patient safety concern affecting more than 2.5 million patients annually in acute care settings in the United States.1 In addition to the physical, psychological, and financial impact on a patient's quality of life, the care related to HAPIs presents a major economic burden on the healthcare system.2,3 In the United States, the estimated cost of treating pressure injuries (PIs) ranges from $9.1 billion to $11.6 billion.1 A 2007 ruling by the Centers for Medicare & Medicaid Services (CMS) classified stage 3 and stage 4 HAPIs as “never events” and in 2008, stopped reimbursing hospitals for the costs related to their care.4 Per the CMS, a single stage 3 or stage 4 HAPI, as a secondary diagnosis, adds $43,180 to each hospital stay.5 Thus, early detection and prevention of PIs is critical for improving patient outcomes and reducing the economic burden on patients and the healthcare system.6

The current National Pressure Ulcer Advisory Panel (NPUAP) staging system is widely adopted internationally and classifies PIs based on anatomical features and the extent of existing tissue loss.7 Pressure injuries are staged from 1 to 4 with the additional categories of unstageable, deep tissue pressure injury (DTPI), and medical device-related PI.8 The NPUAP defines a stage 1 PI as intact skin with a localized area of nonblanchable erythema.8 Pressure-related blanchable erythema (PrBE) or alterations in sensation, temperature, or firmness often precede stage 1 pressure-related nonblanchable erythema (PrNBE).8 Because melanin is not blanchable, distinguishing PrBE from a stage 1 PI is problematic on individuals with dark skin.

A DTPI is defined as intact or nonintact skin with localized area of persistent nonblanchable deep red, maroon, purple discoloration, or epidermal separation revealing a dark wound bed or blood-filled blister.8 The injury is often preceded by a change in temperature or the presence of pain. The discoloration with stage 1 PI and DTPI may appear differently on darkly pigmented skin. Injury to deep tissue may occur despite overlying intact skin. Black and colleagues9 noted necrosis under intact skin may take 48 hours from time of injury until it can be visually detected. Additionally, 7 to 10 days may be required for a DTPI to deteriorate into a full-thickness stage 3 or stage 4 PI.9

There is increasing evidence in the literature supporting the idea that PIs are the result of DTPI, progressing outward until the damage can be seen visually.10–12 Traditionally, visual skin assessment has been the standard used by researchers and clinicians for identifying early signs of skin damage.10,13 However, visual skin inspection is difficult and unreliable, predisposing individuals, especially those with darker skin tones, to advanced skin damage before it is recognized.10,13 Thus, the ability to have a nonvisual assessment tool to identify DTPI at the time of hospital admission would have an important impact on quality of patient care.

The purpose of this systematic review was to determine whether sufficient research evidence exists to support the use of bedside technologies for early detection of PIs, which is inclusive of pressure-related erythema (PrE) and DTPI. Our review sought to answer the question: What available bedside technologies are effective for enhancing the early detection of PIs? Because blanching cannot be reliably assessed on individuals with dark skin, we included both PrBE and PrNBE in this review.

METHODS

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol was used to guide our systematic review process.14 We derived our Population-Intervention-Comparison-Outcome (PICO) elements directly from the search question, and these elements served as the basis for our inclusion and exclusion criteria. Inclusion criteria were peer-reviewed, English-language publications or conference proceedings. Eligible studies also included primary quantitative studies, systematic reviews, meta-analyses, which used technologies to identify tissue injury—PrBE, PrNBE (stage 1), or DTPIs, which had not yet progressed to epidermal separation or blistering—as an outcome measure. Technologies had to be bedside accessible (ie, ultrasound, thermography, subepidermal moisture (SEM) measurement, alternative light sources, spectrophotometry, laser Doppler, etc). Exclusion criteria eliminated dissertations, case studies/series (eg, sample size < 5), qualitative studies, and studies using animal models, computer simulation, and/or healthy volunteers. We also excluded studies that used remote technologies such as telemedicine, focused on stage 2, 3 or 4 PIs, and did not differentiate between stage 1 or DTPIs. However, studies that failed to differentiate PrBE and PrNBE were not excluded, as both conditions are indicators of early PI damage.

Four databases were searched including Medline, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Web of Science, and Cochrane with the assistance of a health sciences research librarian. We selected search terms, subject headings, and MeSH terms based on our PICO criteria (see Supplemental Digital Content 1, available at: https://links.lww.com/JWOCN/A53, for specific search language/strategy). Studies were abstracted from the earliest year available for each database through March 1, 2018. Using Zotero15 reference management software, the lead author (K.N.S.) imported all publications and removed duplicates (Figure). To decrease the risk of selection bias, 2 researchers (K.N.S. and M.C.N.) then each independently screened the titles and abstracts for eligibility. Full-text versions of the remaining publications were retrieved. Screening results were compared and discrepancies discussed until agreement was reached. A similar process was used to perform full-text reviews of the surviving studies, while enlisting the assistance of a third researcher, a certified Wound Ostomy Continence Nurse (L.A.), to clarify terminology and patient-related issues. An ancestry search of the bibliographies of eligible publications was conducted by K.N.S. and subjected to a similar review process. Data were extracted from each of the eligible studies and organized into Tables 1 to 4 that show abbreviated versions of the data collected. Again, to reduce risk of selection bias, we used a triple independent data extraction process with all 3 researchers (K.N.S., M.C.N., and L.A.) independently collecting the data from each of the eligible studies. Data were then compared and combined for synthesis.

F1
Figure.:
PRISMA14 flow diagram of search strategy.
TABLE 1. - Summary of Studies: Ultrasound
Study Study Design Sample Size/PI Equipmenta/Anatomical Site Findings Quality Ratingb
Sato and colleagues,16 Japan Prospective, comparative 30 long-term care residents
31 stage 1 PIs developed
Braden: Median 13 (PI healed), median 10 (PI advanced in stage)
7.5 MHz, Echo Camera SSD-500
Multiple sites
At least 59% of PIs had superficial hypoechoic legions (unclear exact number); 26% had deep hypoechoic legions
PIs that advanced in stage (n=11) more likely to have deep hypoechoic legions (P = .012).
No interrater analysis
III.B
Quintavalle and colleagues,17 USA Prospective, descriptive 119 long-term care residents
Braden: all ≤18
Unknown incidence of PrE
1139 US images
20 MHz, EPISCAN I-200
Heel, sacrum, ischial regions
60% of images with superficial hypoechoic lesion had visible erythema
20% of images with deep hypoechoic legion had visible erythema
<1% normal images had erythema
Interrater reliability image interpret: 97%
III.B
Helvig and Nichols,12 USA Longitudinal, descriptive 99 hospital inpatients
83% white
Braden: all 10-17
6 developed stage 1 PIs
4 developed sDTI (per supplemental material)
“High-frequency,” EPISCAN
Heels
Hypoechoic legions preceded 3 stage 1 and one sDTI
All other stage 1 PIs and sDTIs had prior borderline scans (not normal/clearly abnormal)
No interrater analysis
III.C
Porter-Armstrong and colleagues,18 Ireland/UK Longitudinal, descriptive 50 hospital inpatients
Braden: 3 at risk (score 15-18)
127 cases of PrE
3 stage 1 PIs
1492 US images
20 MHz, EPISCAN I-200
Heels, coccyx, and sacrum
All 3 stage 1 PIs had normal scans
Heels: 86% subjects with hypoechoic legion; 32% with subcutaneous damage (unclear if also had concurrent erythema)
Interrater agreement: κ = 0.80
III.C
Aliano and colleagues,19 USA Cross-sectional, descriptive 20 hospital inpatients
8 stage 1
8 sDTI
12 MHz, unknown equipment
Sacrum
88% and 100% of stage 1 and sDTI, respectively, had hypoechoic legions
63% and 100% of stage 1 and sDTI, respectively, had unclear structural layers
No interrater analysis
III.C
Abbreviations: PI, pressure injury; PrE, pressure-related erythema; sDTI, suspected deep tissue injury; US, ultrasound.
aEcho Camera SSD-500 (Aloka, Tokyo, Japan), EPISCAN (Longport, Glenn Mills, Pennsylvania).
bJHNEBP Rating Scale20 based on the systematic review's variables of interest.

During the data collection process, we each independently analyzed the quality of the eligible studies using the Johns Hopkins Nursing Evidence-Based Practice (JHNEBP) Rating Scale.20 This grading scale assesses the strength and quality of research studies. The strength of a study is based on the study's design: experimental studies are graded as I, quasi-experimental studies are graded as II, and nonexperimental studies and qualitative studies are graded as III. Quality is measured against indicators of the rigor of the study methodology, such as sampling design, sample size, instrument reliability/validity, data collection plan, analysis of the findings, and other strengths and limitations of the studies. After assessing the study against the criteria, a letter grade is assigned: A for high quality, B for good quality, and C for low quality or major flaws. The researchers discussed the grades they assigned to each of the studies until consensus was reached.

RESULTS

Search Results

Of the 1050 articles identified across the database and ancestry searches, 850 were screened by title and abstract resulting in 128 full-text articles retrieved for review. Ultimately, 18 studies met eligibility for inclusion in our final synthesis (Figure). These studies examined technologies that were accessible for the bedside use; the most common were ultrasound (n = 5, Table 1), thermography (n = 7, Table 2), and SEM measurement (n = 5, Table 3). In addition, 1 study each examined the effectiveness of spectroscopy, laser Doppler, and spectrophotometry on early detection of PrBE, PrNBE, and DTPI (Table 4). Two of the studies used more than 1 technology.16,23 Nearly half of the studies examined the effectiveness of the technology specifically in the assessment of stage 1 PIs.12,16,18,19,27,29–31 Another 8 studies did not distinguish between PrBE and PrNBE.13,17,22–27 Deep tissue pressure injury was directly assessed in only 4 studies,5,12,19,25 while several others attempted to predict deterioration of PrE as a possible indicator of suspected underlying DTPI.16,21,24,26

TABLE 2. - Summary of Studies: Thermography
Study Study Design Sample Size/PI Equipmenta/Anatomical Site Findings Quality Ratingb
Newman and Davis,21 UK Prospective, descriptive 91 hospital inpatients
7 (8%) at risk for PI (<14 Norton score)
21 had PIs on admission
6 developed PI
Infrared, prototype
Sacrum and ischial regions
3 subjects had PrE with elevated skin temperature
11 subjects had no PrE and “diffuse warm patch” (ie, 1°C/cm); 5 developed PI within 10 d
Temperature differences not provided
No interrater analysis
III.C
Sprigle and colleagues,22 USA Cross-sectional, comparative 65 acute rehab patients
51% light pigment
80 cases of PrE (35% PrNBE)
Liquid crystal contact monitors, DermaTherm
Multiple sites
Significant mean difference (0.82°F) in skin temperature between PrE and control skin (P < .0015)
Mean PrE was warmer than control skin (90.5°F PrE vs 89.5°F control)
23% of PrE cases were cooler than control skin
No interrater analysis
III.B
Sprigle and colleagues,23 USA Prospective, comparative 76 acute rehab patients
61% white
95 areas of PrE (44% PrNBE)
Liquid crystal contact monitors, DermaTherm
Multiple sites
No significant difference in temperature between PrBE and PrNBE relative to control sites (P = .43)
No significant difference in temperature between sDTI (15%) and superficial PrE (“appeared limited to epidermis/dermis”) relative to control sites (P = .67)
No interrater analysis
III.B
Sato and colleagues,16 Japan Prospective, comparative 30 long-term care residents
31 stage 1 PIs developed
Braden: Median 13 (PI healed), median 10 (PI advanced in stage)
Infrared, Thermo Tracer
Heel and sacrum
Median temperature difference between PI and control skin: 0.2°C healed group, −0.1°C advanced group
PIs that advanced in stage (n = 11) had significantly lower temperatures relative to control skin than healed group (P = .049)
No interrater analysis
III.B
Farid and colleagues,24 USA Retrospective, comparative 85 hospital inpatients with PrE
Braden: Mean 10.5 (SD 2.5)
Infrared, Flir i7
Multiple sites
On average, PrE was cooler than surrounding skin for 65% cases (−1.2°C, SD 1.0°) and warmer for 35% (1.2°C, SD 0.7°)
PIs that advanced in stage (n = 30) were more likely to be cooler than surrounding skin (P < .001)
Cooler PrE 32 times greater odds of advancing in stage (P = .001)
No interrater analysis (retrospective design)
III.A
Cox and colleagues25 USA Prospective, predictive 67 nursing home residents
All with PrE or sDTI (specific frequencies not reported)
94% white
Braden: Mean 14.1 (SD 2.4)
Infrared, Flir i
Multiple sites
61% PrE/sDTI warmer and 39% cooler than surrounding skin (mean difference 3.99°C)
PrE temperature was not related to whether the injury advanced in stage after 2 wk (P = .93)
When controlling for other factors, odds of PrE/sDTI advancing in stage after 7 d was 6.3 times greater for cooler injuries (P = .08)
Interrater reliability performed but not reported
III.B
Mayrovitz and colleagues,5 USA Prospective, comparative 100 hospital inpatients
Braden: Mean 16.5 (SD 4.0)
16 developed DTI
Infrared, Scout SCA100
Sacral region
Temperature difference of −1.5°C (affected area − control) was threshold for PI risk
Only 2 of 16 DTIs met threshold. No significant difference in temperature between affected area and control skin for subjects with DTI (P value not reported)
III.B
Abbreviations: PI, pressure injury; PrBE, pressure-related blanchable erythema; PrE, pressure-related erythema; PrNBE, pressure-related nonblanchable erythema; SD, standard deviation; sDTI, suspected deep tissue injury.
aThermo Tracer TH5108ME (NEC San-ei Instruments, Tokyo, Japan), DermaTherm Perfusion Monitors (Sharn Inc, Tampa, Florida), Flir i/i7 (Flir Systems, Boston, Massachusetts), and Scout SCA100 (Wound Vision LLC, Indianapolis, Indiana).
bJHNEBP Rating Scale20 based on the systematic review's variables of interest.

TABLE 3. - Summary of Studies: Subepidermal Moisture Measurement
Study Study Design Sample Size/PI Equipmenta/Anatomical Site Findingsb Quality Ratingc
Bates-Jensen and colleagues,26 USA Longitudinal, predictive 35 nursing home residents
80% white
Braden: Mean = 16.5 (SD 3.6)
25 are totally dependent
1366 PrEs repeated observations
Unknown incidence of PrE
NOVA Petite dermal phase meter
Multiple sites
Mean SEM was higher for PrE (192 DPU, SD 188) compared to normal skin (91 DPU, SD 122) across all anatomical sites (P < .001)
100 DPU change in SEM predicted 26% of PrE 1 wk later on sacrum/buttocks
SEM increased with advancing PI stage over time (P < .05)
SEM interrater reliability: r = 0.63
III.A
Bates-Jensen and colleagues,27 USA Longitudinal, predictive 31 nursing home residents
72% white
Braden: Mean = 16.7 (SD 3.5)
31% totally dependent
318 PrBE and 28 stage 1 repeated observations
Unknown incidence of PrBE/stage 1
NOVA Petite dermal phase meter
Multiple sites
Mean SEM was higher for stage 1 (264 DPU, SD 208) compared to PrBE (185 DPU, SD 138) or normal skin (104 DPU, SD 115) across anatomical sites (P < .0003)
100 DPU change in SEM predicted 99% of concurrent PrE across anatomical sites
100 DPU change in SEM predicted 32% of PrE 1 wk later on sacrum
SEM interrater reliability: r = 0.63
III.B
Bates-Jensen and colleagues,28 USA Longitudinal, predictive 66 nursing home residents
83% white
Braden: Mean = 16.3 (SD 3.6) light skin; mean = 17.5 (SD 3.5) dark skin
335 PrE repeated observations on light skin, 35 dark skin
Unknown incidence of PrE
NOVA Petite dermal phase meter
Multiple sites
Mean SEM was higher for PrE (150.4 DPU, SD = 128.2) compared to normal skin (83.5 DPU, SD = 100.6) across anatomical sites (P < .001)
50 DPU threshold significantly predicted PrE in dark skin (OR 5.3, 95% CI 1.8-15.1)
300 DPU threshold significantly predicted PrE in light skin (OR 2.1, 95% CI 1.1-4.2)
100 DPU change in SEM predicted 15% on light skin and 88% on dark skin of PrE 1 wk later on sacrum/buttocks
No reliability statistics provided
III.B
Guihan and colleagues,13 USA Longitudinal, descriptive-comparative 32 subjects
All nonambulatory with spinal cord injury
56% white
No Braden reported
47% had at least 1 existing PI
66 cases PrE
Moisture Meter D
Multiple sites
Mean SEM was slightly higher for PrE (42 DPU, SD = 10) than normal skin (41 DPU, SD = 10) across all anatomical sites
SEM interrater reliability: r = 0.86
III.B
Bates-Jensen and colleagues,29 USA Longitudinal, predictive 417 nursing home residents
37% white
Braden: Mean = 15.6 (SD 3.2)
33% developed stage 1 PI; unknown incidence of PrBE
564 PrBE and 149 stage 1 PI repeated observations
Moisture Meter D
Sacrum and ischial regions
Mean SEM was higher for stage 1 PI (39.3 TDC) compared to normal (37.1 TDC) and erythema (38.4 TDC) at sacrum (P < .0001)
SEM predicted concurrent stage 1 PI on sacrum of light and medium skin subjects (RRR 1.03, 95% CI 1.0-1.1)
43 TDC threshold significantly predicted stage 1 PI on sacrum (AUC = 0.57; RRR: 2.0, 95% CI: 11.4-3.0) with 37.4% sensitivity and 77.7% specificity
SEM not a significant predictor of stage 1 PI at 1 wk
Interrater reliability: r = 0.92
III.A
Abbreviations: AUC, area under curve; CI, confidence interval; DPU, dermal phase units; OR, odds ratio; PI, pressure injury; PrBE, pressure-related blanchable erythema; PrE, pressure-related erythema; RRR, relative risk ratio; SD, standard deviation; SEM, subepidermal moisture; TDC, tissue dielectric constant.
aMoisture Meter D (Delfin Technologies, Ltd, Kuopio, Finland); NOVA Petite (NOVA Technology Corp, Portsmouth, New Hampshire).
bSEM values are relative to the equipment used.
cJHNEBP Rating Scale20 based on the systematic review's variables of interest.

TABLE 4. - Summary of Studies: Reflectance Spectrometry and Laser Doppler
Study Study Design Sample Size/PI Equipmenta/Anatomical Site Findings Quality Ratingb
Sprigle and colleagues23 Prospective, comparative 76 acute rehab patients
61% white
95 areas of PrE (44% PrNBE)
Tissue reflectance spectrometry (TRS) algorithm calculates an erythema index relative to hemoglobin concentration
Spectrophotometer, Monolight 6800
Multiple sites
No relationship between blanchability and whether erythema was present (P > .30).
When pressure applied to PrE (“blanching”), TRS erythema relative to resting state was significantly higher for PrNBE (z score mean: light skin 14.8, SD 11.6; dark skin, 3.77, SD 6) vs PrBE (z score mean: light skin 7.4, SD 6.3; dark skin 0.59, SD 6)
Erythema algorithm sensitivity: 79%. Prior interrater analysis reported (ICC = 0.81)
III.B
Nixon and colleagues,30 UK Longitudinal, comparative 37 hospital inpatients
50 images: 26 PrBE, 8 PrNBE
Skin perfusion is assessed through Doppler flux with image data reported in Perfusion Units
Laser Doppler Perfusion Imager
Sacrum and buttocks
More perfusion observed in PrNBE compared to PrBE (P < .001)
No interrater analysis
III.B
Sterner and colleagues,31 Sweden Longitudinal, comparative 78 hospitalized inpatients
100% white
21 PrE on admission
44 stage 1 on discharge
Erythema index (E-index) is calculated based on proportion of absorbed green light (associated with erythrocytes) relative to red
Narrow-band spectrophotometer, Derma Spectrometer
Sacrum
PrNBE had significantly higher E-index than PrBE (mean difference 1.88-4.74; P < .001)
E-index cut-off 14.5, sensitivity 75% and specificity 41% in distinguishing PrNBE and PrBE
Test-retest reliability E-index: ICC = 0.82-0.96
III.B
Abbreviations: ICC, intraclass correlation coefficient; PI, pressure injury; PrBE, pressure-related blanchable erythema; PrE, pressure-related erythema; PrNBE, pressure-related nonblanchable erythema.
aMonolight 6800 (Imaging and Sensing Technologies, Horseheads, New York), Laser Doppler Perfusion Imager (Moor Instruments Ltd, Devon, UK), and Derma Spectrometer (Cortex Technology, Hadsund, Denmark).
bJHNEBP Rating Scale20 based on the systematic review's variables of interest.

Research Quality

The methodological rigor of the included studies varied greatly based on how they addressed our specific search question. The following JHNEBP grades were assessed: 3 As,24,26,29 11 Bs,5,13,16,17,22,23,25,27,28,30,31 and 4 Cs.12,18,19,21 All of the studies were observational in design (level 3), 14 were prospective/longitudinal, and 4 were cross-sectional. With a few exceptions, earlier research (prior to 2007) tended to be more descriptive and exploratory in nature. Seven studies had small samples sizes (n < 50) of PIs 5,12,16,19,21,30,31 of which 4 had 11 or fewer PIs.12,19,21,30 Only 4 studies conducted a power analysis as a basis for sample size determination.24,25,30,31 Sample ethnic diversity ranged from 3% to 35% African Americans or persons with dark skin across 9 studies, although half of the studies did not report those participant characteristics.5,16–19,21,24,30,31

All but 4 studies16,19,21,23 provided sufficient detail on the theoretical basis and execution of each of the technologies studied. The reliability of instruments' application and associated measurements were formally evaluated in all SEM studies13,26,28–30 but only in 1 thermography study,25 1 ultrasound study,17 and 1 spectrophotometry study.31 The selection of control or comparison skin sites also varied across thermography and SEM studies. Only 3 studies reported blinding the clinician who performed the visual assessment from the results of the technology application.12,18,21 Finally, environmental factors associated with temperature such as exposure to ambient temperature and moisture including incontinence were controlled in 8 of the 12 studies.

Study results pertaining to the systematic review question were generally clear with relevant findings presented for individual studies (Tables 1–4). However, the number of PIs or number of PIs identified using the technologies was not completely transparent in 3 studies,5,16,18 requiring access to supplemental material for one of them.12 Finally, due in large part to small sample sizes, 4 studies performed only descriptive analyses, preventing inference to a broader population.12,13,17,18

Technology for Early Detection

A description of the technology and synthesis of data for the 4 technologies used for early detection of PI identified through our systematic review include ultrasound, thermography, SEM, and light.

Ultrasound

Since its development in the 1970s, healthcare providers have used ultrasound as a diagnostic tool for a variety of conditions.32 The technology relies on pulsed sound waves emitted from a probe to penetrate the skin's surface and determine its underlying structure based on the echo received from reflected waves. The speed by which the reflected sound returns determines the underlying structure distance and density from the probe.32,33 Using mathematical formulas, a computer converts the waves to a cross-sectional image, also known as a B-mode scan.32 In the ultrasound images, structures with a greater density appear bright indicating higher echogenicity (hyperechoic). By comparison, areas of fluid or macroscopic edema appear dark (hypoechoic). More recently, the development of high-frequency ultrasound (>10 MHz) has improved image resolution but at the expense of penetration depth, making it ideal for the evaluation of more superficial dermal structures.33 Such devices have become more portable to allow for bedside use.

Of the 5 studies identified in our review that examined the effectiveness of ultrasound for PI detection (Table 1), all reported a hypoechoic area was observed in most images taken of PrBE and PrNBE.12,16–19 Additional image findings included unclear structural layers19 and evidence of inflammation and dermal damage.18 The hypoechoic region was more frequently described as superficial in the presence of erythema.16,17 Quintavalle and colleagues17 noted detection of deep hypoechoic regions with no associated erythema, suggesting evidence of early DTPI. Sato and colleagues16 also noted similar findings among stage 1 injuries that deteriorated. Ultrasounds were performed by nurses in 3 of the 5 studies.12,16,17 Only 2 studies performed interrater analysis of the image interpretation with good results (see Table 1).17,18 None of the studies reported findings based on participant race/skin tone, with the only study by Helvig and Nichols12 providing overall sample characteristics.

Thermography

Perfusion of the skin is reflected in its surface temperature. Increased perfusion from injury or inflammation from reperfusion (hyperemic response) may be associated with a localized increase in skin temperature.24,34 Alternatively, nonviable tissue associated with DTPI would lack sufficient perfusion, resulting in cooler overlying skin.25 Such differences in skin temperature can be detected through conduction by liquid crystal contact monitors placed on the skin's surface or, more commonly today, with infrared, noncontact imaging devices that capture thermal radiation.34,35 Selecting an unaffected adjacent skin site for comparison and controlling for ambient temperature acclimation are both important strategies to ensure accurate thermography results.24,25,35

Five of the 7 studies that evaluated thermography noted a difference in temperature as an indicator of PI (Table 2).16,21,22,24,25 Three of the studies noted that areas of PrE were more often warmer than adjacent skin.21,22,25 Pressure injuries that ultimately deteriorated were significantly more likely to be cooler than the surrounding skin, suggesting the possibility of underlying DTPI.16,24,25 But when assessing actual suspected DTPI, no significant difference in temperature was found in 2 studies.5,23 Sprigle and colleagues23 were unable to distinguish PrBE from PrNBE based on temperature difference. None of the studies reported interrater analysis of the technologies' application. One study examined the relationship between erythema and underlying skin tone, noting no significant association.23

Subepidermal Moisture Measurement

Inflammatory processes resulting from pressure injury-associated cell death contribute to an increase in local microvascular fluid.36 The presence of this extracellular edema within the epidermal and dermal layers can be detected through changes in the skin's SEM.29,36 In brief, SEM measurement technology captures the skin's capacitance (ability to hold an electrical charge) by determining its relative dielectric constant (eg, electrical field resistance).36 The greater the water content in the tissue, the higher the dielectric constant; actual reported SEM values are equipment dependent. Changes in SEM are determined based on delta values calculated from several measurement points taken on the skin.37

All 5 studies that examined SEM measurement as an indicator of PI noted significantly higher moisture readings for PrE compared to normal skin (Table 3). Bates-Jensen and colleagues27 found SEM measurement predicted 99% of concurrent PrE. When comparing PrBE to PrNBE, 2 studies found significantly higher SEM associated with PrNBE (stage 1).28,29 In 2 studies, measurement thresholds relative to 2 different SEM devices significantly predicted concurrent PrE or stage 1 PIs (see Table 3 for specific equipment).28,29 The study by Bates-Jensen and colleagues29 was the only one to report sensitivity (37.4%) and specificity (77.7%). The SEM measurement technology also demonstrated good interrater reliability (r = 0.63-0.92) in 4 studies.13,26,27,29 Finally, 4 of the 5 studies examined whether current SEM values could predict the future development of PIs; the measurements predicted the development of between 15% and 88% of PrE and/or stage 1 PI identified on repeated assessment conducted at least 1 week later, with darker skin tone favoring a greater likelihood of detection.26–29

Light Technology (Spectroscopy and Laser Doppler)

Hemoglobin is 1 of 2 major chromophores contributing to skin color, the other being melanin.38 The concentration of hemoglobin is associated with the skin's perfusion, which is greatly affected by the extent and degree of ischemia. Temporary occlusion of microvasculature can result in reactive hyperemia, or a temporary increase in blood flow, which is typically assessed by pressure applied to the skin to produce blanching.23,39 However, blanching is often difficult to observe, especially in individuals with darker skin tones. Blood flow or flux can be evaluated using laser Doppler, which captures the reflected light from both moving red blood cells and surrounding tissue and converts the information into an electrical signal.40 The resulting images can then be analyzed for perfusion based on intensity. Alternatively, reflectance spectrometry (spectral measurements of light reflected off matter) can provide a proximity measure of perfusion by measuring the degree of erythema, which is highly correlated with the concentration of hemoglobin.38 Using a spectrophotometer, white light is applied to the skin's surface. The reflectance of certain spectral bandwidths is measured and, using an algorithm, converted to an erythema index.

In 3 studies, researchers attempted to distinguish PrBE from PrNBE (stage 1) based on indicators of perfusion using the technologies described earlier (Table 4).23,30,31 Findings from these studies demonstrated PrNBE had significantly greater perfusion than PrBE as observed by laser Doppler30 and through higher erythema index values.23,31 Studies examining reflectance spectrometry23,31 reported good interrater and intrarater reliability along with sensitivity and specificity for the erythema algorithms used (see Table 4). Nixon and colleagues30 did not report any such analyses for their use of laser Doppler. The erythema index used in the Sprigle and colleagues23 study was able to compensate for melanin concentration, allowing it to be applied to different skin tones.

DISCUSSION

Pressure injuries remain physical, psychological, and financial burdens on patients and an economic drain on the healthcare system. Visual skin assessment is unreliable, especially in individuals with darker skin tones. Our systematic review identified 5 technologies that have been studied for their potential to provide early detection of PIs. These technologies included ultrasound, thermography, SEM measurement, reflectance spectrometry, and laser Doppler flowmetry.

Evidence from our review supports the use of SEM measurement as a potential tool for the early identification of PI, specifically blanchable erythema and nonblanchable erythema, with ultrasound and alternate light devices warranting further research. Based on our assessment using the JHNEBP Rating Scale,20 the methodological rigor was consistently higher across the SEM measurement studies compared to the other technology identified in the review. A body of research regarding SEM measures, which includes multiple, high-quality studies, increases the reliability of findings identified in our review. As such, these devices may help identify early pressure-related skin damage before clinical manifestations, potentially decreasing healthcare costs and benefitting patients of all skin tones.

Studies using ultrasound to detect early PIs found meaningful and consistent results, but the quality of the research to date precludes us from recommending its use. More rigorous studies with larger sample sizes are needed before ultrasound can be recommended as an effective technology. Two additional considerations for ultrasound include it being limited to detecting macroscopic level tissue damage and its training requirements for image interpretation.36

Thermography was the most studied technology with 7 studies identified; however, findings were inconsistent. Furthermore, thermography requires positioning of the patient to allow the temperature of exposed skin sufficient time to stabilize prior to assessment. Such additional effort may cause nurses to be reluctant to use the technology in daily clinical practice. Measuring erythema using reflectance spectrometry was found to be effective at differentiating between PrBE and PrNBE (stage 1). However, only 2 studies supported this technology, and distinguishing between PrBE and PrNBE skin damage may be of limited clinical importance. The results of a single study investigating laser Doppler flowmetry were promising, but require further research.

The 1 technology consistently able to provide early detection of PIs, supported by 5 high-quality studies, is SEM measurement. The body of evidence supports the technology's use as an effective tool for detecting stage 1 or PrBE. Though literature suggests SEM measurements may also assist in the detection of deep tissue injury,41 currently sufficient evidence is not available to support this practice.

For SEM measurements to be clinically useful, the technology must be applied using a simple, noninvasive, durable, light-weight, and easily transportable instrument that requires only minimal training to operate. Nurses must find the tool quick to use and interpret at the bedside with good inter-rater reliability. Several SEM measurement devices are on the market that appear to meet these requirements, with one used in 2 studies recently approved by the Food and Drug Administration and available in the United States.37,41 The purchase prices of many reliable SEM measurement, thermography, spectrometry, and portable ultrasound devices can range between $2000 and $10,000. The initial investment of early detection technology must be weighed against the potential savings in preventive PI care.

According to a systematic review conducted by Gunowa and colleagues,42 patients with dark skin tones are at higher risk for developing severe PIs. Although reasons for this risk disparity have not been rigorously investigated, evidence suggests that current recommended visual and tactile skin assessment practices are inadequate to meet the needs of patients with darker skin tones. Evidence from our review suggests technologies used to perform SEM measurements and spectroscopy may improve the identification of PrE in this particular population.23,28 Nurses should include assessment techniques that allow equitable care and reduce outcome disparities in high-risk populations. Thus, recognizing and adopting technology that can detect early-stage skin injury in patients with dark skin is critical.

The current state of the science related to technologies that can identify early pressure-related skin damage has implications for clinical practice, education, and research. Larger scale studies are needed to evaluate the effectiveness and feasibility of these devices in the clinical setting to facilitate preventive interventions and decrease the incidence and prevalence of PIs. Additionally, more research is needed to understand how location of the PI on the body may affect the application and interpretation of the different technologies. For example, the impact of peripheral vascular disease and greater skin thickness on the heel may alter measurements of perfusion and ultrasound image interpretation, respectively.17 In the meantime, nurses in clinical practice settings and the nurse educators who train them need to recognize the shortcoming of visual and tactile assessments for identifying early skin injury, especially for patients with dark skin tones. Skin assessments should be performed carefully and with consultation from a certified wound nurse.

LIMITATIONS

Several limitations to our systematic review should be noted. First, despite performing a comprehensive review of the literature, our efforts may have missed eligible studies. A systematic search of relevant journals may have provided additional publications. Second, we included both populations with PrBE and PrNBE in our eligibility criteria. Even though the NPUAP does not include a stage for PrBE, we felt its inclusion was necessary to address the important disparity in identifying stage 1 PIs on individuals with dark skin. Data provided in Tables 1 to 4 demonstrate that much of the available research on PI detection does not distinguish between these types of injuries. Finally, the JHNEBP Rating Scale is a widely used tool to assess the quality of evidence to support clinical decision-making.20 However, its application, as with any grading scale, is subjective. We attempted to overcome this challenge by utilizing 3 raters and focusing our assessment on only those elements relevant to the search question. We did not contact the authors of the included studies, which would have helped clarify challenges identified in some of the findings.

CONCLUSIONS

Currently, there is promising technology for the early detection of pressure-related injury of the skin. Evidence identified in our systematic review supports several options for further development or immediate practice implementation. The technology must be readily available and easily integrated into clinical practice by the bedside nurse. Skill training for the use of the technology should begin with the certified wound nurse who has the expertise in visual skin assessment and can then progress the technology to bedside nursing.

Visual skin inspection is not highly reliable in identifying pressure-related skin injuries, especially in individuals with darker skin tones. As we learn more about the pathology of PI development and causal factors, cost-effective assistive assessment technology may play an integral role in the timely detection of PI and early implementation of optimal PI prevention measures and treatments.

REFERENCES

1. Agency for Healthcare Research and Quality. Preventing pressure ulcers in hospitals: a toolkit for improving quality of care. https://www.ahrq.gov/sites/default/files/publications/files/putoolkit.pdf. Published 2011. Accessed November 7, 2018.
2. Kanno N, Nakamura T, Yamanaka M, Kouda K, Nakamura T, Tajima F. Low-echoic lesions underneath the skin in subjects with spinal-cord injury. Spinal Cord. 2009;47(3):225–229. doi:10.1038/sc.2008.101.
3. Tubaishat A, Papanikolaou P, Anthony D, Habiballah L. Pressure ulcers prevalence in the acute care setting: a systematic review, 2000-2015. Clin Nurs Res. 2018;27(6):643–659. doi:10.1177/1054773817705541.
4. Centers for Medicare & Medicaid Services. CMS improves patient safety for Medicare and Medicaid by addressing never events. https://www.cms.gov/newsroom/fact-sheets/cms-improves-patient-safety-medicare-and-medicaid-addressing-never-events. Published 2008. Accessed December 29, 2018.
5. Mayrovitz HN, Spagna PE, Taylor MC. Sacral skin temperature assessed by thermal imaging: role of patient vascular attributes. J Wound Ostomy Continence Nurs. 2018;45(1):17–21. doi:10.1097/WON.0000000000000392.
6. Rao AD, Preston AM, Strauss R, Stamm R, Zalman DC. Risk factors associated with pressure ulcer formation in critically ill cardiac surgery patients: a systematic review. J Wound Ostomy Continence Nurs. 2016;43(3):242–247. doi:10.1097/WON.0000000000000224.
7. Edsberg LE, Black JM, Goldberg M, McNichol L, Moore L, Sieggreen M. Revised National Pressure Ulcer Advisory Panel pressure injury staging system. J Wound Ostomy Continence Nurs. 2016;43(6):585–597. doi:10.1097/WON.0000000000000281.
8. National Pressure Ulcer Advisory Panel. NPUAP pressure injury stages. http://www.npuap.org/resources/educational-and-clinical-resources/npuap-pressure-injury-stages/. Published 2016. Accessed November 9, 2018.
9. Black JM, Brindle CT, Honaker JS. Differential diagnosis of suspected deep tissue injury. Int Wound J. 2016;13(4):531–539. doi:10.1111/iwj.12471.
10. Gefen A, Gershon S. An observational, prospective cohort pilot study to compare the use of subepidermal moisture measurements versus ultrasound and visual skin assessments for early detection of pressure injury. Ostomy Wound Manag. 2018;64(8):12–27. doi:10.25270/owm.2018.9.1227.
11. Aoi N, Yoshimura K, Kadono T, et al. Ultrasound assessment of deep tissue injury in pressure ulcers: possible prediction of pressure ulcer progression. Plast Reconstr Surg. 2009;124(2):540–550. doi:10.1097/PRS.0b013e3181addb33.
12. Helvig EI, Nichols LW. Use of high-frequency ultrasound to detect heel pressure injury in elders. J Wound Ostomy Continence Nurs. 2012;39(5):500–508. doi:10.1097/WON.0b013e3182652648.
13. Guihan M, Bates-Jenson BM, Chun S, Parachuri R, Chin AS, McCreath H. Assessing the feasibility of subepidermal moisture to predict erythema and stage 1 pressure ulcers in persons with spinal cord injury: a pilot study. J Spinal Cord Med. 2012;35(1):46–52. doi:10.1179/204577211X13209212104141.
14. Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015;349(1):g7647–g7647. doi:10.1136/bmj.g7647.
15. Roy Rosenzweig Center for History and New Media. Zotero version 4.0.29.10. https://www.zotero.org/. Accessed November 9, 2018.
16. Sato M, Sanada H, Konya C, Sugama J, Nakagami G. Prognosis of stage I pressure ulcers and related factors. Int Wound J. 2006;3(4):355–362.
17. Quintavalle PR, Lyder CH, Mertz PJ, Phillips-Jones C, Dyson M. Use of high-resolution, high-frequency diagnostic ultrasound to investigate the pathogenesis of pressure ulcer development. Adv Skin Wound Care. 2006;19(9):498–505.
18. Porter-Armstrong AP, Adams C, Moorhead AS, et al. Do high frequency ultrasound images support clinical skin assessment? ISRN Nurs. 2013:2013:314248.
19. Aliano K, Low C, Stavrides S, Luchs J, Davenport T. The correlation between ultrasound findings and clinical assessment of pressure-related ulcers: is the extent of injury greater than what is predicted? Surg Technol Int. 2014;24:112–116.
20. Poe SS, Costa L. Evidence appraisal: research. In: Dearholt S, Dang D, eds. Johns Hopkins Nursing Evidence-Based Practice: Models and Guidelines. 2nd ed. Indianapolis, IN: Sigma Theta Tau International; 2012.
21. Newman P, Davis NH. Thermography as a predictor of sacral pressure sores. Age Ageing. 1981;10:14–18.
22. Sprigle S, Linden M, McKenna D, Davis K, Riordan B. Clinical skin temperature measurement to predict incipient pressure ulcers. Adv Skin Wound Care. 2001;14(3):133–137.
23. Sprigle S, Linden M, Riordan B. Analysis of localized erythema using clinical indicators and spectroscopy. Ostomy Wound Manage. 2003;49(3):42–52.
24. Farid KJ, Winkelman C, Rizkala A, Jones K. Using temperature of pressure-related intact discolored areas of skin to detect deep tissue injury: an observational, retrospective, correlational study. Ostomy Wound Manage. 2012;58(8):20–31.
25. Cox J, Kaes L, Martinez M, Moles D. A prospective, observational study to assess the use of thermography to predict progression of discolored intact skin to necrosis among patients in skilled nursing facilities. Ostomy Wound Manage. 2016;62(10):14–33.
26. Bates-Jensen BM, McCreath HE, Kono A, Apeles NCR, Alessi C. Subepidermal moisture predicts erythema and stage 1 pressure ulcers in nursing home residents: a pilot study. J Am Geriatr Soc. 2007;55(8):1199–1205.
27. Bates-Jensen BM, McCreath HE, Pongquan V, Apeles NCR. Subepidermal moisture differentiates erythema and stage I pressure ulcers in nursing home residents. Wound Repair. 2008;16(2):189–197. doi:10.1111/j.1524-475X.2008.00359.x.
28. Bates-Jensen BM, McCreath HE, Pongquan V. Subepidermal moisture is associated with early pressure ulcer damage in nursing home residents with dark skin tones: pilot findings. J Wound Ostomy Continence Nurs. 2009;36(3):277–284. doi:10.1097/WON.0b013e3181a19e53.
29. Bates-Jensen BM, McCreath HE, Patlan A. Subepidermal moisture detection of pressure induced tissue damage on the trunk: the pressure ulcer detection study outcomes. Wound Repair Regen. 2017;25(3):502–511. doi:10.1111/wrr.12548.
30. Nixon J, Cranny G, Bond S. Pathology, diagnosis, and classification of pressure ulcers: comparing clinical and imaging techniques. Wound Repair Regen. 2005;13(4):365–372.
31. Sterner E, Fossum B, Berg E, Lindholm C, Stark A. Objective evaluation by reflectance spectrophotometry can be of clinical value for the verification of blanching/non blanching erythema in the sacral area. Int Wound J. 2014;11(4):416–423. doi:10.1111/iwj.12044.
32. Serup J, Keiding J, Fullerton A, Gniadecka M, Gniadecki R. High-frequency ultrasound examination of skin: introduction and guide. In: Serup J, Jemec G, Grove G, eds. Handbook of Non-Invasive Methods and the Skin. 2nd ed. Boca Raton, FL: CRC Press; 2006:473–491.
33. Grubbs S, Ludwig M, McHale E, et al. The effect of high frequency ultrasound on the prevention of pressure ulcers in long-term care patients. Internet J Acad Physician Assist. 2009;7(1):3–3.
34. Ring E. Thermal imaging of skin temperatures. In: Serup J, Jemec G, Grove G, eds. Handbook of Non-Invasive Methods and the Skin. 2nd ed. Boca Raton, FL: CRC Press; 2006:769–786.
35. International Academy of Clinical Thermology. Quality Assurance Guidelines: Standards and Protocols in Clinical Thermographic Imaging. http://www.breastthermography.com/articles/2015%20IACT%20Standards%20and%20Guidelines.pdf. Published January 2019. Accessed November 5, 2019.
36. Gefen A. The Sub-epidermal Moisture Scanner: the principles of pressure injury prevention using novel early detection technology. Wounds Int. 2018;9(3):30–35.
37. Ross G, Gefen A. Assessment of sub-epidermal moisture by direct measurement of tissue biocapacitance. Med Eng Phys. 2019;73:92–99. doi:10.1016/j.medengphy.2019.07.011.
38. Takiwaki H. Measurement of erythema and melanin index. In: Serup J, Jemec G, Grove G, eds. Handbook of Non-Invasive Methods and the Skin. 2nd ed. Boca Raton, FL: CRC Press; 2006:665–672.
39. Lachenbruch C, Tzen Y-T, Brienza D, Karg PE, Lachenbruch PA. Relative contributions of interface pressure, shear stress, and temperature on ischemic-induced, skin-reactive hyperemia in healthy volunteers: a repeated measures laboratory study. Ostomy Wound Manage. 2015;61(2):16–25.
40. Bircher A. Laser Doppler measurement of skin blood flux: variation and validation. In: Serup J, Jemec G, Grove G, eds. Handbook of Non-Invasive Methods and the Skin. 2nd ed. Boca Raton, FL: CRC Press; 2006:691–696.
41. Moore Z, Patton D, Rhodes SL, O'Connor T. Subepidermal moisture (SEM) and bioimpedance: a literature review of a novel method for early detection of pressure-induced tissue damage (pressure ulcers). Int Wound J. 2017;14(2):331–337. doi:10.1111/iwj.12604.
42. Gunowa NO, Hutchinson M, Brooke J, Jackson D. Pressure injuries in people with darker skin tones: a literature review. J Clin Nurs. 2018;27(17/18):3266–3275. doi:10.1111/jocn.14062.
Keywords:

Blanchable erythema; Deep tissue pressure injury; Diagnostic wound technologies; Early wound detection; Nonblanchable erythema; Pressure injury; Pressure ulcer; Stage 1 pressure injury

Supplemental Digital Content

© 2020 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of the Wound, Ostomy and Continence Nurses Society