Physiological measurements of tissue health; implications for clinical practice

Abstract

Pressure mapping alone insufficiently describes tissue health. Comprehensive, quantitative non invasive assessment is crucial. Interface pressures (IPs) and transcutaneous blood gas levels [transcutaneous tissue oxygen (T_cPO₂)] were simultaneously assessed over both ischia and the sacrum to investigate the hypotheses: (i) tissue oxygenation decreases with sustained applied pressure; (ii) tissue oxygen and IP are inversely correlated in loaded soft tissues; (iii) multisite assessments are unnecessary because healthy individuals are symmetrical. Measurements were taken at 5‐minute intervals for 20 minutes in both sitting and supine lying for a cohort of 20 able‐bodied adults. There were no statistically significant changes over time for either variable in 96% of timepoint comparisons. Specifically, no significant differences were seen between 10 and 20 minutes in either position. These findings imply that a 10‐minute assessment can reliably indicate tissue health and that tissue may adapt to applied load over time. No statistically significant correlations between T_cPO₂ and IP were observed. However, the left and right ischia were significantly different for both variables in supine lying (P < 0·001) and for sitting IP (P < 0·010). Thus, even in this healthy cohort, postural symmetry was not observed and should not be assumed for other populations with restricted mobility. If a multisite technique cannot be used, repeated tissue health assessments must use the same anatomic location.

INTRODUCTION

The maintenance of tissue health is a critical component of clinical management for every at‐risk patient group but not often measured in clinical practice. Any individual with restricted mobility is at increased risk of tissue breakdown. Other risk factors include impaired sensation, poor nutrition and muscle atrophy. Examples of high‐risk groups are the spinal cord injured (SCI) and elderly populations (1). Tissue breakdown because of prolonged applied pressure can develop in any region of the body but most frequently over weight‐bearing body surfaces, bony prominences and where soft tissue coverage is reduced. The sacrum and ischial tuberosities are common sites of tissue breakdown. Such breakdown often leads to a chronic non healing wound, negatively impacting on the patients' rehabilitation and quality of life.

Development of serious tissue breakdown at pressure points caused by prolonged sitting affects nearly 50% of the 1·4 million people who rely on wheelchairs for mobility (2). For wheelchair users, it is commonly accepted that effective pressure relief is an important strategy for the prevention of pressure ulcers. Moreover, for persons with SCI, absence of an effective and reliable pressure relief method is thought to be one of the major reasons for the continued high incidence of pressure ulcers (3). The optimal treatment strategy for pressure ulcers is prevention; one meta‐analysis systematically reviewed all relevant literature published from 1980 to 2007 comparing various prevention and treatment of pressure ulcers in those with SCIs (4). The meta‐analysis found numerous studies on the effectiveness of weight shifts in pressure ulcer prevention. However, there has been no definitive research to indicate optimal pressure relief regimens.

Real‐time pressure mapping can provide biofeedback on the efficacy of pressure relief for an at‐risk individual. The effectiveness and overall pressure distribution of wheelchair seating devices such as cushions can also be evaluated. However, applied pressure cannot be measured directly and is thus reported in terms of interface pressure (IP) values. Motor paralysis and subsequent muscle atrophy often lead to an increase in IP between the weight‐bearing body part and support surface such as the wheelchair cushion or mattress. This increase in IP cannot be felt by those with sensory impairment and can also lead to a decrease in repositioning frequency. The absence of sensory feedback can thus also contribute to the risk of tissue breakdown. Pressure mapping can objectively measure this IP and quantify one of the risk factors in tissue breakdown.

In addition to the measurement of IP by pressure mapping, other physiological parameters can be measured to determine intrinsic risk factors. However, there is no current guideline on which physiological parameters should be measured to measure pertinent risk factors most accurately. This motivates this study. Impaired vascularity and oxygenation of the soft tissues are increasingly recognised as risk factors for the impairment of tissue health 5, 6. Maintenance of tissue health requires an adequate supply of nutrients from the blood and tissue oxygenation. Therefore, tissue oxygenation measurement is important for quantitative identification of pressure ulcer risk, especially at areas of increased mechanical loading, such as the sacrum and ischial tuberosities. Without adequate blood flow and oxygenation, toxic metabolites accumulate locally, increasing the rate of cell death. This leads to ulceration and necrosis of skin and underlying tissue and development of pressure ulcers (7). Maintenance of tissue health requires the management of these risk factors, which in turn affect the tissue tolerance for pressure. Attempts have been made to minimise the load on the tissue by means of wheelchair cushions and seating devices that create postural control and pressure distribution. These seating systems were showed to positively influence IP and skin oxygenation 8, 9.

Transcutaneous blood gas systems measure the partial pressure of oxygen and/or carbon dioxide in the skin. Transcutaneous blood gas measurement has been used in studies to determine both pressure ulcer risk and healing 10, 11, 12, 13. Knight et al. (10) focussed on measuring oxygenation of only sacral skin tissue on a group of 14 male healthy volunteers. Their study strongly suggested that a reduction in tissue oxygen may be a critical factor in tissue metabolism, with evidence of marked increases in tissue transcutaneous tissue carbon dioxide and sweat lactate and urea beyond an apparent threshold reduction in transcutaneous tissue oxygen (T_cPO₂) of 60%. A study by Reenalda et al. (14) on 25 non disabled males concluded that shifting posture at least every 8 minutes is advisable for maintenance of tissue viability in wheelchair users. Furthermore, subcutaneous tissue oxygenation increases on average 2·2% with each posture adjustment, indicating the positive effects of posture shift on tissue viability. Another analysis of healthy sitting behaviour was carried out by Linder‐Ganz et al. (15). They concluded that non disabled subjects change their posture approximately every 9 minutes in the sagittal plane and approximately every 6 minutes in the frontal plane, as measured with pressure sensors. Unfortunately, this study focussed only on truncal movements and the external load on the tissue, as quantified by IP. Other physiological parameters like transcutaneous blood gas measurements were not assessed. The recent availability of multisensor transcutaneous blood gas systems gives researchers the potential ability for simultaneous measurement of several at‐risk skin locations. A combined approach to risk assessment for pressure ulcer using both pressure mapping and transcutaneous blood gas systems has the potential to provide a much more comprehensive approach in the objective determination of risks.

The goal of this study was to investigate the tissue response to loading conditions using objective indicators of tissue health status in able‐bodied adults in both sitting and supine lying. Our methodology was developed to understand these physiological parameters better in this population. Previously, risk assessment has focussed on single sites at risk of tissue breakdown, such as the ischial tuberosities or sacrum 16, 17. Although IP mapping can measure the entire seating interface region, T_cPO₂ electrodes can only determine localised regions of tissue oxygenation. Until quite recently, T_cPO₂systems have only been single channel devices. This precluded concurrent assessment at multiple sites of interest without the use of multiple monitors, such as reported by Makhsous et al. (18). Our study is the first to simultaneously assess multiple sites on the body using a multifactorial tissue health assessment protocol using a multichannel T_cPO₂ system and IP monitoring in both sitting and supine lying postures. The specific hypotheses to be investigated were that, in able‐bodied individuals:

• 1tissue blood oxygenation decreases over time with sustained applied external pressure;
• 2tissue responses to load at pressure ulcer risk sites exhibit an inverse correlation between tissue oxygen and IPs, that is, T_cPO₂ decreases as IP increases; and
• 3multisite assessments are unnecessary because able‐bodied individuals have symmetrical sitting and supine lying postures.

MATERIALS AND METHODS

The standardised Skin Care Research Team tissue health assessment protocol developed by our group uses surrogate measures indicating changes in risk status, including IP measurement and non invasive evaluation of transcutaneous blood gas levels (19).

The current cohort study mapped tissue health profiles for a group of 20 healthy able‐bodied adults: 10 men and 10 women. Cohort size was determined based on our experience from tissue health assessments carried out with individuals with SCI to allow detection of statistically valid changes in study variables. Women and minorities were included in the study population to the extent that they are present in the target study population. Average weight for all subjects was 69 kg [standard deviation (SD) ± 17]. Average weight was 78 kg (SD ± 19) for men and 60 kg (SD ± 8) for women. The average age for both men and women was 24 years (SD ± 2). There were no significant differences between demographic variables thus data were pooled for all subjects at each timepoint.

Transcutaneous blood oxygen levels and IPs were mapped for all study participants in both supine lying and sitting positions.

Tissue gas measurement

T_cPO₂ was measured using a multichannel Radiometer TCM400 monitor (Radiometer America Inc., Westlake, OH). The oxygen electrodes (Radiometer, model E5280‐8) were calibrated to room air. The temperature control of the system was set at 43°C, to produce maximal local vasodilation. The variation in room temperature over the course of each assessment was maintained at 25 ± 2°C throughout each assessment.

IP monitoring

IP distribution was recorded using the Tekscan CONFORMat® Pressure Measurement System (Tekscan Inc, Boston, MA) which used thin flexible sensors using resistive sensors in a grid‐based array containing more than 1000 sensors. The CONFORMat® sensor operated within the range 1–250 mmHg (accuracy ±10%) at scanning rate of 100 Hz. Real‐time three‐dimensional images of pressure distribution at the seating interface were produced using graphical display software. The system analysis has the capability to determine several variables (20). In this study, mean pressures were determined for the regions of interest.

DATA COLLECTION

Assessment protocol

To monitor tissue health, the tissues were first unloaded for a 20‐minute equilibration period prior to assessment in the posture of interest. Specifically, the research participant lay in a side‐lying position on a standard hospital bed with a standard hospital mattress (Hill‐Rom, Batesville, IN), with hips and knees flexed to 90° to approximate the limb posture in sitting. The bony landmarks of interest were palpated and fixation rings located over the ischial tuberosities and sacrum (Figure 1). The fixation ring consisted of a 20 mm diameter adhesive ring surrounding a central plastic screw mounting. The electrode mounting was approximately 10 mm in diameter and 6 mm high. The central region of the ring was filled with contact fluid and each electrode securely attached. A 20‐minute equilibration period with the subject remaining in the side‐lying position then followed so that local vasodilation was stabilised, with unloaded values of T_cPO₂ in region of 50–90 mmHg, as previously reported by Knight et al. (10).

Figure 1.

Location of transcutaneous tissue oxygen electrodes in supine lying position.

The Tekscan mattress sensor mat was placed on top of the mattress and the research participant carefully rolled into a supine position. The T_cPO₂ electrodes and all surrounding regions were then gently manually palpated to ensure there was no relative movement of the soft tissue over the bony prominences. The supine lying assessment phase then commenced and transcutaneous gas levels were monitored at 5‐minute intervals concurrent with IP measurement for a 20‐minute period, that is, 0 (baseline), 5, 10, 15 and 20 minutes.

A Tekscan sensor mat was then placed over the cushion of the wheelchair seating system. A standard wheelchair with a foam cushion was used. The research participant then carefully transferred to the seated position (Figure 2). Optimal sitting posture is often considered to include maintenance of the lumbar lordotic curve. However, it has been shown that habitual sitting posture is significantly more flexed and may be kyphotic 21, 22. In this study, although subjects sat upright, their habitual sitting postures showed posterior pelvic tilt and a reduction or loss of lordosis, thus sacral loading was exhibited in sitting. The T_cPO₂ electrodes and all surrounding regions were thus again gently manually palpated to ensure there was no relative movement of the soft tissue over the bony prominences. Monitoring of tissue gas levels then commenced (initial baseline, T = 0). Transcutaneous gas levels were continuously monitored concurrent with IP measurement for a further 20‐minute assessment period.

Figure 2.

Location of transcutaneous tissue oxygen electrodes in seated position with posterior pelvic tilt.

ETHICAL CONSIDERATIONS

The Institutional Review Board of the Louis Stokes Cleveland Department of Veterans Affairs Medical Center (LSCDVAMC) approved this study, and written informed consent was obtained from all the subjects.

DATA ANALYSIS

Each subject was measured repeatedly during the assessment period, thus the longitudinal data were correlated over time. It was not appropriate to fit a conventional linear regression model that assumes independent identical distributed random errors for the data (23). Random coefficient models are a type of regression model that is particularly suitable for analysis of data where repeated measurements have been obtained from a group or sample drawn from a larger population. These models have a nested covariance structure that accounts for random variations within groups. In this study, two sources of variation can be recognised: (i) random variation among subjects due to, for instance, physiological variation and (ii) random variation within individuals caused by the process of actually taking the measurement. Linear random‐effects mixed models were therefore applied to the study data (24) allowing introduction of two separate sources of variation: random variation among individuals and random variation within individuals. Correlation studies were performed to examine relationships between the selected tissue health parameters at the 0·05 significance level, with appropriate corrections for multiple testing.

RESULTS

The subjects were monitored at 5‐minute intervals over a 20‐minute assessment period for each posture of interest. Thus data were collected for IP and T_cPO₂ at five timepoints in two different postures, resulting in 10 data sets for each subject. The outcomes of interest were changes in tissue health variables, specifically IP and T_cPO₂, over time following the initial response to application of external load (obtained at T = 0, baseline). There were no outliers for any variable of interest and no statistically significant differences over time for 96% of comparisons between timepoints for all three body areas (sacral, left and right ischial) measured (Table 1). Measurements with statistically significant differences are summarised in the following paragraphs.

Table 1.

Variation in tissue health parameters over time in lying and in sitting

Location	Time (minutes)	Transcutaneous tissue oxygen (mmHg)			Interface pressure (mmHg)
		Mean	SD	Significant differences	Mean	SD	Significant differences
In lying
Sacrum	0	31	24	NS	34·53	9·69	NS
	5	34	22	NS	34·17	8·90	NS
	10	35	21	NS	34·96	8·96	NS
	15	37	19	* Versus 0	34·31	9·12	** Versus 0
	20	35	20	NS	35·67	11·43	* Versus 5
							*** Versus 0
Left ischial	0	44	12	NS	41·93	10·01	NS
	5	44	12	NS	43·71	9·81	NS
	10	47	14	NS	45·77	12·80	NS
	15	45	16	NS	47·23	14·15	NS
	20	45	15	NS	48·45	13·46	NS
Right ischial	0	52	23	NS	48·98	14·53	NS
	5	51	20	NS	48·52	12·10	NS
	10	55	23	NS	49·12	13·23	NS
	15	51	14	NS	49·76	12·45	NS
	20	47	14	NS	50·01	14·10	NS
In sitting
Sacrum	0	53	13	NS	39·98	12·09	NS
	5	54	14	NS	43·52	17·71	NS
	10	53	13	NS	43·89	16·37	NS
	15	52	13	NS	43·50	17·57	NS
	20	53	15	* Versus 0&10	44·75	16·65	NS
				** Versus 5
Left ischial	0	52	12	NS	52·07	12·35	NS
	5	52	13	NS	54·70	19·18	NS
	10	52	13	NS	54·96	18·94	NS
	15	50	11	NS	57·90	23·04	NS
	20	49	11	NS	56·81	21·82	NS
Right ischial	0	50	19	NS	51·76	16·37	NS
	5	52	18	NS	51·65	15·69	NS
	10	52	18	NS	52·03	14·87	NS
	15	51	16	NS	52·43	16·20	NS
	20	52	17	NS	53·04	13·38	NS

NS, non significance; SD, standard deviation.

^* P < 0 · 05,

^** P < 0 · 01,

^*** P < 0 · 001.

In supine lying, there were no statistically significant differences over time for right ischial T_cPO₂ or IP or sacral IP at any timepoint. The only exception in T_cPO₂ for supine lying was a statistically significant increase in T_cPO₂ over time in the sacral region between baseline loading (T = 0) and 15 minutes (P < 0·050). For the left ischial area IP in supine lying, there were statistically different increases in IP over time observed between baseline loading (T = 0) and both 15‐minute (P < 0·010) and 20‐minute (P < 0·001) intervals. In addition, there was a statistically significant increase in left ischial IP in supine lying between 5‐ and 20‐minute intervals (P < 0·050).

In sitting, there were no statistically significant differences over time for IP at any timepoint. There were also no statistically significant differences over time at any timepoint for right ischial or sacral IP T_cPO_2. Left ischial T_cPO₂ was reduced at 20 minutes resulting in statistically significant decreases relative to both baseline (P < 0·050) and 5‐minute intervals (P < 0·010).

The relationship between T_cPO₂ and IP was then analysed to determine the interaction of these tissue health variables. No statistically significant correlations between T_cPO₂ and IP for either sitting or supine lying at any assessment timepoint were observed.

Data for all timepoints for right and left ischial measurements were pooled to determine postural symmetry (Table 2). Equal left and right ischial region IPs imply that the pelvis is balanced and symmetrical, whilst unequal pressures imply pelvic asymmetry. Statistically significant differences between the left and right ischia were observed for T_cPO₂ in supine lying (P < 0·001) and for IP in both supine lying (P < 0·001) and sitting (P < 0·010).

Table 2.

Differences between left and right ischial tissue health variables

Position	Variable	Left ischial		Right ischial		Significant differences
		Mean	SD	Mean	SD
In lying	TcPO2	39·2	12·2	45·7	15·4	P < 0·001
	IP	44·6	13·1	46·8	9·5	P < 0·001
In sitting	TcPO2	48·9	11·6	47·6	17·2	NS
	IP	55·1	20·9	52·1	16·2	P < 0·01

IP, interface pressure; SD, standard deviation; T_cPO₂, transcutaneous tissue oxygen.

DISCUSSION

Tissue breakdown may originate at the surface, that is, preliminary skin breakdown, or in the deep tissues at the bony interface 25, 26. There is a significant amount of research being carried out to determine the underlying mechanisms leading to deep tissue injury and necrosis 27, 28. However, currently there are no clinically applicable techniques other than magnetic resonance imaging (MRI) for assessment of deep tissue health status 29, 30. However, MRI is not be routinely available in many clinical rehabilitation settings and is unlikely to be cost effective as a preventative assessment technique. In addition to the usual contraindications to the use of MRI, many individuals with restricted mobility cannot tolerate a closed MRI due to difficulties with transferring or claustrophobia. IP measurement and tissue blood gas assessment are clinically practical techniques, as used in seating clinics 31, 32, which can provide quantitative measurements of surface and subsurface tissue health conditions. This approach provides indirect indicators of tissue health and whilst by no means a complete picture, the assessment of multiple objective variables facilitates more comprehensive assessment than subjective risk factor scales and has been shown to improve clinical outcomes (33). Pressure mapping systems provide information on IP distribution, including colour‐coded pressure maps and numerical pressure values. On the other hand, tissue blood gas levels provide an indication of regional vascularity and response to applied load. However, this physiological parameter is rarely measured in clinical practice and there is little known about its correlation with IP measurement.

In this study, it was found that overall there was no statistical difference between measuring T_cPO₂ or IP values at 5 and 20 minutes timepoints for supine lying or sitting in the majority of cases. This finding appears to refute the first study hypothesis. It is of interest to note that the tissue response in the supine lying posture appears to be less stable than in sitting as indicated by more statistically significant differences over the 20‐minute assessment period.

The able‐bodied subject group evaluated in this study did not show a decrease in oxygenation with increased pressures. Bader and Gant (34) showed that tissue response to applied pressure also varies in individuals with SCI. More specifically; they reported that 22–92 mmHg applied pressure was required to produce a 50% decrease in T_cPO₂ relative to the unloaded value at the sacrum. The findings of this study imply that healthy individuals may either have a widely varying threshold for critical applied pressure required to decrease T_cPO₂ or be able to adapt to sustained applied pressure. Healthy individuals with normal vascularity may have an adaptive blood flow to compensate for increased pressures, due in part to the viscoelastic behaviour of muscular bulk soft tissues 35, 36. Thus findings of this study do not support the second study hypothesis.

Tissue health variables were found to exhibit statistically significant side‐to‐side differences in this study. This indicates that IPs were not distributed symmetrically implying that pelvic posture was also asymmetric, Thus indicating that the third study hypothesis is also invalid. Specifically, able‐bodied individuals cannot be assumed to have symmetric posture. An asymmetric posture leads to asymmetric blood flow and IP conditions. If able‐bodied individuals are asymmetric, it is even more unlikely that individuals with muscle atrophy, motor deficits and disabilities such as SCI or stroke are symmetric.

These findings imply that care should be taken to ensure that repeated postural assessments are carried out using the same location for unilateral measurements, that is, all measurements for a specific variable should be taken on the right or left side of the patient's body. In addition, the apparent adaptive response observed in this able‐bodied subject group implies that caution should be used when extrapolating the effect of seating interventions between different subject populations. For example, the effects on tissue health of wheelchair cushions to optimise comfort and prevent pressure ulcers, developed using able‐bodied test subjects cannot be assumed to be directly replicated in disabled populations.

Pressure mapping alone is not sufficient to indicate tissue health (37). Inclusion of other non invasive measurement techniques, such as T_cPO₂, is crucial for a more comprehensive, quantitative assessment of pressure ulcer risk. Currently, IP mapping systems are widely used in seating clinics. Multichannel tissue gas monitoring systems have been commercially available for around 5 years and are most commonly used in intensive care settings, where equipment cost is less of a factor. At first review it may seem that these systems can improve efficacy in a research setting but have little relevance in the clinic. In the rehabilitation field, preventative management and specifically seating and wheelchair assessments are generally highly cost conscious. The purchase cost of a tissue gas monitor may be considered too great. However, this initial outlay may be outweighed by the longer term cost–benefit of improved clinical effectiveness in assessing pressure ulcer risk.

Overall there was found to be no statistically significant differences between measuring T_cPO₂ and IP at 10 or 20 minutes following application of loads due to sitting or supine lying. The findings of this small study of able‐bodied individuals imply that in this population, tissue health variables tend to have stabilised by 10 minutes under applied load. This further implies that a 10‐minute assessment can provide a reliable indicator of tissue health.

The absence of a direct correlation between IPs and T_cPO₂ for the cohort of able‐bodied subjects studied implies that loaded tissue may adapt over time in this population. The absence of an adaptive response may be a factor in increased pressure ulcer risk. Further research into a potential adaptive response to loading in healthy compared with at‐risk individuals may provide insight regarding physiological mechanisms impacting the risks of pressure ulcer formation. This finding may also indicate that caution should be used in directly extrapolating tissue health data obtained from a non disabled population to disabled populations known to be at risk, such as individuals with SCI.

Postural symmetry cannot be assumed. If a multisite technique cannot be applied, then it is critically important that repeated tissue health assessments are carried out at the same anatomic location. For example, when assessing ischial region T_cPO₂ it is recommended that an initial selection of left or right ischium should be made and that the same site then be assessed at all subsequent assessments to determine changes in tissue health over time. It is further recommended that postural assessment should be an integral component of evaluating pressure ulcer risk. To be clinically valuable the assessment should be non invasive, rapid, quantifiable and documented. Techniques such as digital photography and or goniometry may be useful.

The data from this preliminary study contribute to knowledge regarding normal healthy tissue responses to clinically relevant loading postures and may have implications for assessments in at‐risk patient populations.

Study limitations

This study used a small cohort of young able‐bodied adults at low risk for pressure ulcer development. The lack of a direct comparison with an at‐risk population, such as individual with SCI, is a potential study limitation. However, numerous studies and clinical experience have shown that such individuals are at increased risk. This study may be considered to highlight the limitations of extrapolating between populations.

It is also possible that the relative locations of the T_cPO₂ electrodes changed as the study participants transferred from supine lying to sitting. The careful postural transfer process includes manual palpation of the T_cPO₂ electrode and surrounding soft tissue after the new posture is attained. This is carried out to check for any relative movement of the soft tissue over the bony prominence. In this study, it was found that there was no relative movement of the soft tissue that moved the T_cPO₂ electrodes away from the bony prominences.

It is also noted that statistical significance does not always indicate clinical significance. For example, some of the differences in between left and right IPs do not appear great (Table 2). It is acknowledged that all statistical testing runs the risk of type I (false positive) errors. However, as noted in the section Materials and Methods, our statistical methodology included appropriate adjustments for repeated testing.

ACKNOWLEDGEMENTS

This was an unfunded study. All study personnel contributed to this article.