Assessing Skin Blood Flow and Interface Pressure in Patients With Spinal Cord Injury Provided an Alternating Pressure Overlay: A Cross-sectional Study

The effects of an alternating pressure (AP) overlay on the skin are not fully understood. Purpose: This study was conducted among persons with spinal cord injury (SCI) to examine skin blood flow (SBF) and interface pressure (IP) during and after AP overlay use. Methods: In this cross-sectional, repeated measures study, persons eligible for participation were clinic outpatients from a large metropolitan area in the midwest United States who were 18 to 65 years old with a SCI with a neurologic level of injury at T10 or above for more than 1 year and used a wheelchair for primary mobility. Persons with a current pressure injury, diabetes mellitus, and/or hypertension or other vascular or pulmonary diseases were excluded. Data regarding age, gender, body mass index (BMI), duration of SCI, and American Spinal Injury Association Impairment Scale scores were collected. The experimental study involved 3 protocols: the AP protocol (participants lay supine for 40 minutes on an operating room [OR] pad with a low-profile AP that used a 10-minute inflation-deflation cycle); the post-AP protocol (participants lay on the 2-inch foam OR pad for 40 minutes), with 30 minutes of rest in between; and the control protocol, comprised of 40-minutes of laying supine on the OR pad. Each participant served as his/her own control. Outcome variables included 1) peak IP (the highest value among adjoining sensors located at the highest pressure point); 2) averaged IP (the averaged value of the sensors), calculated from pressure mapping system data from the sacrum and left heel; and SBF, measured using a laser Doppler flowmetry system. Descriptive analyses were performed for all variables to determine need for parametric or nonparametric analyses. The mean value of peak IP, averaged IP among inflation and deflation cycles of AP, and post-AP and control protocols were compared using repeated measures analysis of variance (ANOVA). Mean SBF among inflation and deflation cycles of AP and post-AP and control protocols were compared using the nonparametric Friedman test, and Wilcoxon signed rank tests were used to compare the SBF responses during the post-loading period. If the results of repeated measures ANOVA or Friedman tests were statistically significant,  paired t tests and Wilcoxon signed rank tests were used for pairwise comparison with Bonferroni correction at alpha level 0.0125, respectively. Results: Among the 15 participants (11 men, 4 women; age 41.77 ± 14.49 [range 20–62] years; BMI 26.81 ± 4.13 [range 22–37]; injury duration 17 ± 14.62 [range 1–48] years; mostly (11) African American),  peak IP decreased during the AP deflation at sacrum (51.47 ± 30.18 mm Hg vs. 114.13 ± 60.97 mm Hg; P = .002) and heel (26.79 ± 12.91 mm Hg vs. 53.05 ± 18.22 mm Hg; P = 0 .001), and SBF increased at the heel (27.92 ± 32.15 vs. 10.43 ± 11.16 au; P = .006) but was not significant at the sacrum (15.54 ± 15.33 au vs. 11.96 ± 10.26 au, P = .023). Peak IP decreased during post-AP at the sacrum (104.62 ± 58.17 mm Hg; P = .002) but not at the heel (47.69 ± 16.21 mm Hg; P = .097). SBF increased during post-AP at the sacrum (15.78 ± 15.82 au; P = .012) but not at the heel (16.31 ± 29.18 au, P = .427). Conclusion: An AP overlay redistributed IP and increased SBF at the sacrum and heel during use, and its effect 40 minutes after removal was observed only at the sacrum. Studies, including evaluating the lasting effect of AP on weight-bearing tissue protection at different anatomical locations, are needed.

A multicenter cohort study¹ and a retrospective chart review study² have shown pressure injuries are prevalent (11% to 39%) secondary complications in people with spinal cord injury (SCI). A cross-sectional study³ and 2 review articles⁴ showed that this population is at high risk for developing pressure injuries due to lack of sensation, immobility, incontinence, malnutrition⁵ and changes in tissue properties, such as muscle atrophy and accumulation of intramuscular fat. Therefore, pressure injury  preventative strategies are crucial for this population.

Alternating pressure (AP) overlays are considered active (ie, powered) support surfaces. According to a clinical practice guideline,⁶ different active support surfaces have similar efficacy in terms of reducing pressure injury rate and are recommended for people with “higher risk of pressure injury development when frequent manual repositioning is not possible.”⁶ Offloading pressure is difficult for the SCI population during acute hospitalization or while undergoing prolonged surgical procedure in the operating room (OR).

Several studies have evaluated the potential protective mechanisms of AP support surfaces by measuring physiological outcomes rather than pressure injury incidence. In their cross-sectional study, Chai and Bader⁷ examined the performance of a prototype AP mattress by measuring transcutaneous oxygen tensions and interface pressure (IP) on 12 healthy participants during supine laying. The alternating low pressure surface was provided only at the sacral region, and the AP was adjusted based on the internal mattress pressure. The skin at the scapula served as the control site for the same participant. The researchers found that with head-of-bed angle at 0˚ (supine laying flat), the transcutaneous oxygen tension at the sacrum remained similar to the value at the scapula or fluctuated at adequate level for viability, and the IP at the sacrum rarely exceeded 60 mm Hg. However, the oxygen tension decreased in some participants when the head-of-bed angle increased. A subsequent cross-sectional study⁸ from the same group examined the performance of a commercial AP mattress with various combination of AP patterns at the sacrum in 5 young (20 to 26 years old), healthy participants. AP patterns tested (a total of 18 combinations) included 3 pressure amplitudes (100/0, 60/40, and 30/20 mm Hg), 2 cell profiles (1:2 and 1:4), and 3 cycle periods (3, 9, and 15 minutes). This study⁸ found transcutaneous oxygen tension increased during majority of test conditions; however, at 100/0 mm Hg pressure, the oxygen tension level decreased along with increased transcutaneous carbon dioxide tension level in 46% of cases. The mean IP remained lower than 60 mm Hg in most conditions.

In their cross-sectional study, Arias et al⁹ examined the effects of an AP cushion (prototype) on IP and tissue oxygenation (oxygenated hemoglobin and deoxygenated hemoglobin) at both ischial tuberosities (IT) during seating among 10 healthy participants. Outcomes were compared within each participant before and after 1 hour of AP use. The researchers found no significant difference in mean IP, but the peak pressure decreased significantly (P =.010), especially at the left IT. Oxygenated hemoglobin increased after AP application on both sides. A cross-sectional study by Karg et al¹⁰ assessed the effects of an AP overlay on sacral skin blood flow (SBF) of 19 healthy participants (10 between 18 and 40 years of age, 9 ≥60 years of age). Each participant was assessed in the supine position laying on top of an OR pad with and without the AP overlay. The authors found IP decreased during the deflation cycle and was accompanied by a trend toward increase in SBF with the use of AP overlay.

Each of these studies suggested that using an AP protocol, whether via prototypes or commercial products, had some measurable physiological benefits on the weight-bearing skin of healthy participants. However, literature on skin tissue response to AP in a high-risk population is limited. In their cross-sectional study, Jan et al¹¹ examined the effects of an AP protocol on sacral SBF in 14 healthy participants and 14 participants with chronic SCI. Each participant underwent 2 protocols: the AP protocol was 0 mm Hg and 60 mm Hg of pressure with 4, 5-minute cycles, and a control protocol of constant pressure at 30 mm Hg for 20 minutes. Participants laid prone during the experiment, and the pressure was applied at the sacrum by a single-point pressure indenter. The researchers found that SBF increased with the AP protocol as compared with constant pressure in both groups. No significant difference was noted in the SBF response between participants with and without SCI. Because studies on the effects of AP in a high-risk population are limited, investigation of physiological responses (IP and SBF) during the use of AP in the at-risk population was proposed in the current study.

Studies assessing AP during use are scant, and no research on the effect of AP after removal is available to date. Given that ischemic reperfusion injury, as well as prolonged ischemia, is an etiologic pathway for pressure injury formation,^12,13 the authors suspected it may be beneficial to examine the effect of AP on SBF and IP during the reperfusion period (ie, after AP removal). Ischemic conditioning is widely investigated in patients with ischemic heart disease.¹⁴ These patients are likely to undergo coronary artery bypass grafting surgeries, which according to a review¹⁵ can be complicated by perioperative myocardial injury (a type of ischemia-reperfusion injury). The ischemic conditioning strategy could be implemented by exposing the tissue to several cycles of short-term ischemia (eg, 5 minutes) before prolonged ischemia (more than 1 hour). For example, 4, 5-minute cycles of occlusion and reflow of coronary artery before exposure to 90 minutes of occlusion has been shown to reduce myocardial infarct size in a dog model.¹⁶ Similar findings were later replicated in human studies on myocardial infarction¹⁵ and stroke patients.¹⁷ Although previous studies showed promising results in heart and brain surgery patients, few studies have investigated conditioning strategies on ischemic skin. The results of a controlled prospective cohort study by Kraemer et al¹⁸ showed in patients who underwent free-flap surgery that remote intermittent ischemic conditioning increased tissue oxygen saturation and capillary blood flow and reduced the postcapillary venous filling pressure. In the controlled prospective cohort by Shaked et al¹⁹ that investigated the effect of remote post-conditioning on diabetic foot ulcer healing, repeated remote ischemic post-conditioning of the lower limb was shown to help heal the wound faster compared to regular treatment only.  

Because AP provides cyclical short periods of high and low amplitude pressure on the skin, the current authors suspected it could be used to simulate ischemic conditioning scenarios on the skin during weight-bearing. Given the positive findings from previous studies, the authors anticipated the effects of AP on SBF and IP would be measurable after AP removal. The purpose of this study was to investigate physiological responses (SBF and IP) during and after the use of AP in a population of high-risk patients.

Study design. An experimental, cross-sectional study was conducted in a university laboratory setting. A repeated measures design was utilized, with each person participating in both the experimental and control procedures. To prevent any cross-over effect, the order of the 2 procedures was randomized by drawing envelopes containing 2 different orders, and a washout period of 2 hours was provided between the procedures. Each procedure commenced with a 15-minute rest (baseline) period and ended with 20 minutes of rest (post-loading).
 

Participants. People with SCI were recruited for this study from the greater Chicago area using a convenience sampling method. Flyers describing the study were posted on public bulletin boards at different outpatient clinics, and potential participants were encouraged to contact the research team. Inclusion criteria stipulated participants must be 18 to 65 years old, have a SCI with a neurologic level of injury at T10 or above of more than 1 year, and use a wheelchair for primary mobility. People with a pressure injury, diabetes mellitus, and/or hypertension or other vascular or pulmonary diseases were excluded. Independent patient demographic and injury information variables collected included age, gender, body mass index (BMI), duration of SCI, and American Spinal Injury Association (ASIA) Impairment Scale (AIS) scores were used to assess the sensory and motor levels affected by the SCI. For example, Grade A means the impairment is complete (no motor or sensory function exists below the level of injury) and Grade B stipulates impairment is incomplete (ie, sensory function, but not motor function, is preserved below the first normal level above the level of injury).
 

Procedure. The experimental procedure consisted of 2 protocols: the AP protocol followed by the post-AP protocol, with 30 minutes of rest in between. During the AP protocol, participants were asked to lay supine for 40 minutes on an low-profile AP overlay (Dabir Surface Inc) that was placed over a 2-inch foam OR pad (Alimed Inc) (see Figure 1A). A 10-minute inflation-deflation cycle was applied throughout this protocol (4 inflation/deflation cycles). Compared to typical AP overlays and mattresses, the low-profile AP overlay evaluated in this study is less than an inch in height when fully inflated, and the air cells are an inch in width, providing periodic micropressure relief. The AP overlay then was removed for the 30 minutes of rest. During the post-AP protocol, participants lay on the OR pad only for 40 minutes.

The control procedure consisted of only 1 protocol. During this protocol, participants lay on the OR pad only for 40 minutes. The difference between the control protocol and the post-AP protocol was the lack of AP protocol 30 minutes before the control protocol. To ensure consistency throughout the study, participants were asked to lay on their right side during all rest periods, including 15 minutes of baseline, 30 minutes rest between AP protocol and post-AP protocol, and 20 minutes post-loading. Pillows and towels were provided throughout both procedures to maintain patient posture; the biggest positioning difficulty encountered was the external rotation of the participants’ legs. A rolled towel was provided lateral to the leg to ensure the area of skin in contact with the overlay or pad was consistent throughout both procedures.

Study protocols were approved by the University of Illinois at Chicago Institutional Review Board. All participants provided written informed consent before study participation.

Outcome measures.
IP. The 2 study outcome measures were IP and SBF. Whole-body IP was measured using the XSENSOR X3 pressure mapping system (XSENSOR Technology Corp) (see Figure 1B). The IP data from the 3 protocols (AP, post-AP, and control) were collected during the 40 minutes of laying supine at 1 Hz. The x-sensor pressure mapping software was used for IP data acquisition, and data for selected areas and frames (as detailed in data analysis section) were exported to a Microsoft Excel spreadsheet for further analyses. Area and time frame details are described in data analysis section.

SBF. SBF was measured using the laser Doppler flowmetry system (Moor Instruments Inc). Two (2) silicone soft flat probes (VP11SC, see Figure 1C) were taped to the highest pressure points (determined by the pressure map readings while participants were laying supine on top of the OR pad with pressure map in between) at the sacrum and left heel. The silicone soft flat probes are flexible and can deform to stay in total contact to the skin. The SBF data were collected throughout the baseline, 3 protocols, and post-loading periods for both experimental and control procedures. PowerLab (ADInstruments Inc) was used for SBF data acquisition at 20Hz, and spreadsheets of the data were generated via the LabChart software (ADInstruments Inc) for further analyses. One (1) measurement was taken for both outcome measures during both experimental and control procedures for each participant.

Data analysis. Two (2) variables — peak IP (the highest value among the 9 adjoining sensors located at the highest pressure point) and averaged IP (the averaged value of these 9 sensors) — were calculated from data collected at the highest pressure points at the sacrum and left heel. Once the 9 sensors were identified for the highest pressure points, the peak IP and the averaged IP of each frame were exported to an Excel spreadsheet for further analyses.

To examine the effect of AP use, the mean values of peak IP, averaged IP, and SBF during the inflation and deflation cycles of the AP protocol were compared to values from the 40 minutes of the control protocol. To examine the effect of AP after removal, the mean values of peak IP, averaged IP, and SBF during the 40 minutes of post-AP protocol also were compared to the control protocol outcomes. The first and last minute of the 40 minutes laying supine on OR pad and the first and last minute of the 5 minutes inflation/deflation cycle were excluded from the analyses to eliminate the artifact caused by transition between different postures or inflation/deflation cycle of the AP overlay.

To further examine the effect of AP on SBF after removal, 4 SBF variables derived from the SBF data collected during the rest periods of baseline (the 15 minutes before loading) and post-loading were compared between the experimental and the control procedures: baseline SBF, peak reactive SBF, normalized peak reactive SBF, and the total reactive SBF. Figure 2 shows the SBF response recorded from 1 of the participants during the post-loading period of the control protocol at the heel and includes baseline SBF, peak reactive SBF, and total reactive SBF. Baseline SBF was averaged over the 15 minutes of rest at the beginning of experimental and control protocol. A polynomial equation was used to determine the SBF response during the post-loading periods.20 The peak reactive SBF is the highest point on the fit curve. Normalized peak reactive SBF was calculated using the following equation:

           (peak reactive SBF – baseline SBF)/baseline SBF x 100%

Total reactive SBF is the total amount of increase in SBF from baseline (the interval between the fit curve and the baseline value). The formula used to calculate variables from the fit curve was based on previous literature.²⁰ Based on previous literature,^21,22 the amount of increase in SBF during the post-loading period indicated the severity of tissue ischemia during the loading period. The more severe the tissue ischemia during the weight-bearing condition, the larger the SBF increase (normalized peak reactive SBF or total reactive SBF) during the post-loading period.

All data were transferred from Excel spreadsheets to IBM SPSS, version 25.0 for statistical analyses. Descriptive analyses were performed for all variables to determine need for parametric or nonparametric analyses. To compare the mean value of peak IP and averaged IP among inflation and deflation cycles of AP, and post-AP and control protocols, repeated measures analysis of variance (ANOVA) was used. To compare the mean SBF among inflation and deflation cycles of AP, and post-AP and control protocols, the nonparametric Friedman test was employed. If statistical significance was noted using repeated measures ANOVA or the Friedman test, paired t tests and Wilcoxon signed rank tests were used for pair-wise comparison with Bonferroni correction at alpha level 0.0125, respectively. To compare the SBF response during the post-loading period between the experimental and control procedure, Wilcoxon signed rank tests were performed.

Participants. Among the 15 participants (11 men, 4 women; age 41.77 ± 14.49 [range 20–62] years; BMI 26.81 ± 4.13 [range 22–37]; injury duration 17 ± 14.62 [range 1–48] years; mostly [11] African American) who completed both experimental and control protocols, 8 underwent the experimental protocol first, and 7 participants underwent control protocol first (see Table 1). AIS designates A as complete injury and B as sensory incomplete. The purpose was to ensure not all participants had the same order of the 2 procedures to avoid bias caused by any potential learning effect (which is a limitation to repeated measures study design). Descriptive statistics of demographic and injury data were demonstrated in Table 1.

IP and SBF comparisons among AP, post-AP, and control protocols. Descriptive statistics of the peak IP, averaged IP, and SBF during the inflation and deflation cycles of AP, post-AP, and control protocols are presented in Table 2 and Table 3. Because descriptive analyses showed that peak and averaged IP data were normally distributed, repeated measures ANOVA were performed for the IP data and showed significant differences among the protocols. For peak IP, the results were deflation: 51.47 ± 30.18 mm Hg; inflation: 89.27 ± 53.92 mm Hg; post-AP: 104.62 ± 58.17 mm Hg; control: 114.13 ± 60.97 mm Hg (P = .013). For average IP, the results were and deflation: 36.16 ± 18.47 mm Hg; inflation: 57.21±28.54; post-AP: 81.07 ± 48.62 mm Hg; control: 81.50 ± 46.39 mm Hg, (P = .009) at the sacrum. Significant differences were noted among the protocols in peak IP (deflation: 26.79 ± 12.91 mm Hg; inflation: 52.49 ± 21.21; post-AP: 47.69 ± 16.21; control: 53.05 ± 18.22)(P = .001) and averaged IP (deflation: 14.60 ± 8.37 mm Hg; inflation: 37.88 ± 17.27 mm Hg; post-AP: 30.35 ± 17.59 mm Hg; control: 34.83 ± 16.62 mm Hg (P <.001) at the heel. Because descriptive analyses showed SBF data were not normally distributed, Friedman’s tests were used and showed significant differences among the protocols in SBF at both the sacrum (deflation: 15.54 ± 15.33 au; inflation: 12.65 ± 12.45 au; post-AP: 15.78 ± 15.82 au; control: 11.96 ± 10.26 au) (P = .003) and the heel (deflation: 27.92 ± 32.15 au; inflation: 15.10 ± 20.78 au; post-AP: 16.31 ± 29.18 au; control: 10.43 ± 11.16 au,) (P = .022). For the purpose of visualization, Figure 3 provides an example of pressure map measurements during AP, post-AP, and control from 1 participant. The pressure map data recorded during AP visually demonstrated that rows of inflated and deflated air cells. IP values between post-AP and control appear similar for the same participant. An example of SBF readings from the same participant during AP, post-AP, and control at the sacrum and left heel shows the SBF pattern over the supine-laying period (see Figure 4).

Effects on IP and SBF during AP use. The results of repeated measures ANOVA and Friedman’s tests showed significant differences in IP and SBF among protocols; hence, pairwise comparisons utilizing paired t tests and Wilcoxon signed rank tests were performed, respectively, to identify which comparisons were significant. Results of the comparisons between inflation cycle of AP and control and between deflation cycle of AP and control are shown in Table 2. Effect size r between .3 and .5 was considered medium effect size, and r >.5 was considered large effect size.²³ Study results showed that the peak and averaged IP at the sacrum were significantly lower during the AP deflation cycle as compared to control (peak IP 51.47 ± 30.18 mm Hg vs. 114.13 ± 60.97 mm Hg; averaged IP 36.16 ± 18.47 mm Hg vs. 81.50 ± 46.39 mm Hg). The peak and averaged IP at the heel were also significantly lower during the AP deflation cycle  as compared to control (peak IP 26.79 ± 12.91 mm Hg vs. 53.05 ± 18.22 mm Hg; averaged IP 14.60 ± 8.37 mm Hg vs. 34.83 ± 16.62 mm Hg) (ie, large effect size for all 4 comparisons except medium effect size for averaged IP at heel). The SBF at the sacrum was also greater during AP deflation cycle as compared to control; however, it did not reach statistical significance (medium effect). No difference was noted in peak, averaged IP, and SBF when comparing AP inflation cycle and control. The sacral peak IP and averaged IP were significantly higher (P <.001, r = 0.60; P <.001, r = 0.62, respectively) during the inflation cycle as compared to deflation, and the heel peak IP and averaged IP were significantly higher (P <.001, r = 0.74; P <.001, r = 0.73 respectively) during inflation cycle as compared to deflation. The SBF at the sacrum and heel both increased significantly during the deflation as compared to the inflation cycle (P = .002, r = 0.56; P <.011, r = 0.47, respectively). Figure 5 illustrates boxplots of the peak IP, averaged IP, and SBF during AP and control protocols.

Effects on IP and SBF after AP removal. The IP and SBF data collected during post-AP protocol as compared to the control protocol, the peak IP, averaged IP and SBF data of post-AP and control protocols along with pairwise comparison results are shown in Table 3. Paired t tests showed the sacrum peak IP was significantly lower in post-AP protocol as compared to control with medium effect size. Wilcoxon signed rank test showed the sacral SBF was significantly greater in post-AP protocol as compared to control with medium effect size. No statistically significant difference in other outcome measures was noted between post AP and control. Figure 6 illustrates boxplots of the peak, averaged IP, and SBF during post-AP and control protocols.

The SBF data collected during the post-loading period of experimental and control procedures, the baseline SBF, peak reactive SBF, normalized peak reactive SBF, and the total reactive SBF data are presented in Table 4 (this table only includes data collected at the sacrum from 10 of the 15 participants and at the heel from 9 of the 15 participants); 4 participants did not demonstrate any increase in SBF at both the sacrum and heel as the pattern shown in Figure 2, 1 participant had no increase in SBF pattern at sacrum, and 2 participants had no such response at heel. Of the 10 participants with an increased SBF pattern at sacrum, the total reactive SBF was significantly larger during the post-loading period of the control procedure as compared to the experimental procedure with medium effect size. No statistically significant difference in SBF was noted during the post-loading period between the experimental and the control procedures at the heel.

Effects on IP and SBF during AP use. The findings of this study suggest AP use affects IP and SBF. Current observations of decreased IP at the sacrum and heel during AP-deflation compared to control confirm the findings of Karg et al.¹⁰ In their study of healthy volunteers, sacral SBF and IP were measured during weight-bearing at the sacrum and compared between 2 protocols (laying supine on an OR pad with and without the use of AP). The authors¹⁰ found IP decreased with AP use. The current study supported that IP at both sacrum and heel decreased during AP deflation in people with chronic SCI.

The significant increase in SBF during AP deflation at the heel observed in this study was consistent with the findings of Jan et al¹¹; whereas, the findings of trend toward increase in SBF at the sacrum (not reaching statistical significance) during AP deflation was consistent with Karg et al.¹⁰ The discrepancies between the results from Jan et al¹¹ and Karg et al¹⁰ were mainly due to differences in AP protocol implementation and the size of the area receiving AP during the experiment. In Jan et al,¹¹ AP (alternating 0 mm Hg and 60 mm Hg ) and control protocols (constant pressure of 30 mm Hg) were applied using a single pressure point indenter at the sacrum on 28 participants (14 healthy adults and 14 adults with SCI). The authors¹¹ found SBF increased significantly with AP as compared to the control in participants with and without SCI. In Karg et al,¹⁰ participants were supine on an AP overlay on top of an OR pad during AP protocol; the authors noted that along with the decreased IP during the deflation cycle, SBF tended to increase (but not significantly). Both studies^10,11 examined the skin at the sacrum; however, the skin area that underwent AP was far smaller in Jan et al¹¹ (approximately the size of a quarter) as compared with Karg et al.¹⁰ This discrepancy in results also was demonstrated in the current study at the sacrum versus the heel. Pressure map readings indicated the area of skin in contact with the support surface at the heel was relatively smaller than at the sacrum, resulting in findings similar to  Jan et al,¹¹ where a single-point indenter was used.

Current study results regarding effect size showed that the increase in SBF during the deflation cycle was large (r >0.5) and medium (r >0.3) at the heel.²³  Effect size is the “magnitude of the difference” between the 2 comparisons and “independent of sample size,” whereas statistical significance “depends on both sample size and effect size”.²⁴ Current findings suggested that the phenomenon of increased SBF during deflation was less pronounced at the sacrum, and a larger sample size is needed to reach statistical significance for the data collected at the sacrum.²⁴ The difference between sites may be attributed to the fact that the skin adjacent to the placement of the laser Doppler probe at the heel was not in contact with the support surface; therefore, the adjacent skin was not subjected to the deflation cycle. Skin adjacent to the sacrum was exposed to the inflation cycle of AP. The results showed SBF during the inflation cycle was significantly lower than that during deflation cycle, so the effect of inflation cycle (decreased SBF) on the skin adjacent to the sacrum during the deflation cycle needed to be considered.

In addition, because the IP at the sacrum during the deflation cycle was similar to the magnitude of the IP at the heel during control protocol, the authors postulated that a higher IP value at the sacrum also contributed to the less significant increase in perfusion at this site. The slightly diverse findings between the 2 anatomic sites may help explain the variations observed between the 2 aforementioned studies, where different methods were used to induce weight-bearing. Jan et al¹¹ used a single-point indenter about the size of a quarter, leaving the tissue surrounding the area of interest pressure-free, and Karg et al¹⁰ had participants lay supine and applied pressure to both the specific and surrounding tissue in the area of interest.

It is also worth noting that no differences in IP were observed during AP inflation as compared to the control. The inflation cycle did no more harm than simply having a person lay on top of an OR pad because the AP inflation duration was far shorter (<5 minutes). By comparing and integrating the findings from this study with previous research, the authors believe AP protects weight-bearing tissue during the deflation cycle by decreasing the IP and increasing SBF, especially over bony prominences that are prone to pressure injuries.

Effects on IP and SBF after AP removal. The secondary finding from this study was that some effects of AP on IP and SBF were observed after AP removal at the sacrum but not at the heel (reduced peak IP and increased SBF during post-AP protocol). Another finding was that total reactive SBF decreased with the experimental procedure, suggesting decreased severity of tissue ischemia during weight-bearing in 10 participants. This may be related to differences in the tissue areas surrounding the region of interest. With the relatively larger area of pressure loading at the sacrum, the authors postulated that the region of interest at the sacrum as well as the surrounding area underwent cyclical changes in IP and perfusion during AP. As a result, the sacrum may experience a more profound ischemic conditioning-like scenario (relatively higher IP and larger area affected) as compared to the heel, underscoring the lasting effect. As described previously, ischemia conditioning is a strategy that is used to protect ischemic tissue by exposing the tissue to cyclical short-duration ischemia (approximately 5 minutes) before or after the prolonged ischemia (more than 1 hour). In 2 controlled prospective cohort studies on ischemic skin (ie, free-flap surgery and diabetic wounds in wounds in persons with diabetes), the cyclical short-duration of ischemia was provided by tourniquet¹⁸ or blood pressure cuff¹⁹ on the limb to completely occlude the SBF. Comparing the nature of weight-bearing at the sacrum and the heel, the large area of sacral tissue experiencing cyclical changes in IP during AP is closer to the ischemic conditioning scenario implemented in previous studies using a tourniquet or blood pressure cuff (ie, relatively larger area of skin experiencing a cyclical amount of high pressure). The other explanation may be related to the differences in tissue characteristics between the 2 anatomical sites. The increased sacral SBF during supine laying on an OR pad after AP protocol suggested weight-bearing did not create complete occlusion at the sacrum, thus allowing a relatively elevated perfusion level shortly after 4 cycles of AP. The posterior aspect of the heel has a thicker epidermis and larger capillary diameter in the papillae as compared to sacrum,²⁵ and the high level of metabolic demand makes this area of tissue vulnerable to ischemia.²⁶ Although the IP observed in the current study was relatively lower at the heel compared to the sacrum, the authors suspected that the amount of tissue ischemia at the heel caused by 40 minutes of supine laying on an OR pad was greater than the ischemia at the sacrum, masking the lasting effect of AP. Based on limited data, the authors cannot definitively conclude whether the mechanism of this lasting effect is true ischemia conditioning. However, the study design limits the ability to evaluate the lasting effect of AP on weight-bearing tissue protection at different anatomical locations. Additional studies, including evaluating the lasting effect of AP on weight-bearing tissue protection, are needed.

This study included 15 individuals with chronic SCI with strict inclusion criteria; therefore, the findings cannot be generalized to the SCI population at large that may have other comorbidities or are in the acute phase of SCI. This study also included individuals with both AIS grades A and B. Further research is needed to study the influence of completeness of injury on the efficacy of an AP overlay. An OR pad was utilized as control protocol in comparison with the AP overlay in individuals with chronic SCI, so results could not be extended to nonsurgical settings where thicker mattresses are primarily used. Because this cross-sectional study was conducted at a university laboratory, long-term follow-up on AP overlay use was not examined.

A cross-sectional study among 15 persons with chronic SCI found IP decreased during the deflation cycle of AP at the sacrum and heel. The SBF increased during the deflation cycle of AP at the heel; however, this phenomenon was less profound at the sacrum. After AP removal, IP decreased and SBF increased at the sacrum but not at the heel when laying on the OR pad. As such, an AP overlay may be effective in redistributing IP and increasing SBF at the heel and sacrum during use and also effective in redistributing IP and increasing SBF at the sacrum after AP removal. Further studies in the OR and various hospital settings are warranted to investigate whether the effects observed in the study affect direct clinical outcomes such as pressure injury incidence.

The authors thank the study participants.

Dr. Tzen is an assistant professor, Department of Applied Clinical Research; an assistant professor, Department of Orthopedic Surgery; and an assistant professor, Department of Physical Medicine & Rehabilitation, University of Texas Southwestern Medical Center, Dallas, Texas. Ms. Purohit is a student researcher, Department of Physical Therapy; and a graduate student, Rehabilitation Sciences Program; and Ms. Mei is a student, Rehabilitation Sciences Program (Honors College), University of Illinois at Chicago, Chicago, Illinois. Dr. Tan is an assistant professor, Department of Physical Medicine & Rehabilitation, UT Southwestern Medical Center; and a Director of Clinical Quality, Spinal Cord Injury Center, VA North Texas Health Care System, Dallas, Texas. Please address correspondence to: Yi-Ting Tzen, UT Southwestern Medical Center, 5323 Harry Hines Boulevard, MC 8878, Dallas, TX 75390-8878; email: Yi-Ting.Tzen@UTSouthwestern.edu.

This study was supported by an educational grant from Dabir Surfaces, Chicago, Illinois.