Prevention of Pressure Ulcers Among People With Spinal Cord Injury: A Systematic Review

OBJECTIVES

To evaluate the literature on the effectiveness of bed and wheelchair positioning and repositioning in the prevention of pressure ulcers (PUs) in both the spinal cord injury (SCI) and non-SCI populations.

DESIGN

Systematic review.

METHODS

PubMed, CINAHL, PsycINFO, and EMBASE were queried with the subject heading terms "pressure sore," "pressure ulcer," "position or turn in bed, wheelchair," "pressure relief," and "pressure release." All study design types that assessed the effectiveness of bed and wheelchair positioning and pressure relief maneuvers in any patient group and in any setting were sought. Three independent reviewers extracted and summarized details of eligible trials using a standardized method. Two independent reviewers assessed the methodological quality of each trial using the American Academy of Neurology guidelines. When reviewers were not able to reach consensus, a third independent reviewer served as tiebreaker.

RESULTS

We identified 2820 publications, of which 49 met inclusion criteria. Of these publications, the subject population was 2834 (923 persons with SCI, 717 persons without SCI, and 1194 healthy control subjects). Among studies examining pressure related to position or repositioning in bed or sitting, procedures for measuring skin pressure and metabolism were highly variable by anatomic location, measurement technique, outcome measure, study site, participant characteristics, and description of position/turning for bed and seated interventions. Numerous factors can influence tissue interface pressures, and no prospective studies had been performed to determine a causal relationship between interface pressure and skin breakdown. Several studies suggest that skin response to pressure differs between subjects with and without SCI. Conflicting results and insufficient evidence for optimal bed and seated positioning and turning and pressure relief maneuvers to prevent PUs in both SCI and non-SCI populations were limiting factors.

CONCLUSIONS

Although there is no clear optimal positioning or turning frequency in bed, the evidence suggests avoiding the 90° lateral position because of high pressures and PU risk over the trochanters. During sitting, pressures are linearly redistributed from the sitting area during recline and tilt; however, reclining carries with it an increased risk of shear forces on this skin. The evidence does not support conclusive guidelines on positioning or repositioning techniques for PU prevention in bed or during sitting. We conclude that PU risk is highly individualized, with the SCI population at a higher risk, which demands flexible PU prevention strategies for bed/seated positioning and pressure relief maneuvers. Education has and will remain our most powerful ally to thwart this pervasive public health problem.

Objective classification of vehicle seat discomfort

The objective of this study was to identify how physiological measures relate to self-reported vehicle seating discomfort. Twelve subjects of varied anthropometric characteristics were enrolled in the study. Subjects sat in two seats over a 2-h period and were evaluated via three physiological measures (near-infrared spectroscopy, electromyography and pressure mapping) yielding six testing sessions. Subjective discomfort surveys were recorded before and after each session for nine regions of the body. Conditional classification discomfort models were developed through dichotomised physiological responses and anthropometry to predict subjective discomfort in specific body locations. Models revealed that subjects taller than 171 cm with reduced blood oxygenation in the biceps femoris or constant, low-level muscle activity in the trapezius tended to report discomfort in the lower extremities or neck, respectively. Subjects weighing less than 58 kg with reduced blood oxygenation in the biceps femoris or unevenly distributed pressure patterns tended to report discomfort in the buttocks. The sensitivities and specificities of cross-validated models ranged between 0.69 and 1.00.