Objectives: Upon completion of this article, the reader should be able to (1) explain why a new scale is needed for assessing the risk of pressure ulcers in individuals with spinal cord injury; (2) list the 15 major risk factors for pressure ulcer development in individuals with spinal cord injury; (3) calculate a score and categorize the level of risk for pressure ulcer development for an individual with spinal cord injury; (4) describe the ideal study design to test the accuracy of this scale. Level: Comprehensive
The Association of Academic Physiatrists is accredited by the Accreditation Council for Continuing Medical Education to sponsor continuing medical education for physicians.
The Association of Academic Physiatrists designates this continuing medical education activity for one credit hour in Category 1 of the Physician's Recognition Award of the American Medical Association.
*Disclosure statements have been obtained regarding the authors' relationships with financial supporters of this activity. There is no apparent conflict of interest related to the context of participation of the authors of this article.
Diminished neurologic function places individuals with permanent spinal cord injury (SCI) at significant risk for pressure ulcers throughout their lifetimes. Yet, prevention of this costly problem requires a further understanding of the risk factors specific to this population. Although it is well known that the risk of pressure ulcers is attributable to multiple factors, the sheer number of factors that have been associated with pressure ulcers is over-whelming.1, 2 Moreover, there is limited published evidence that a knowledge of these risk factors can lead to a measurable decrease in the incidence of pressure ulcers.3-6 Many scales and systems of assessing the risk level for pressure ulcers have been designed, generally by simplifying the essential risk factors into additive scales. Most, however, were designed for the geriatric7-9 or critically ill10 patients, and because they were not developed using SCI patients, these scales are inadequate for SCI individuals. Furthermore, because the literature suggests that these existing scales are of questionable use for this population, we developed a new pressure ulcer risk assessment scale-specifically for individuals with spinal cord injury.
The setting was a Spinal Cord Injury Unit in a 240-bed, long-term care Department of Veterans Affairs Medical Center. This was a retrospective study that used only existing data and was, therefore, exempt from review by the Institutional Review Board. A description of the database used for this study has been published.11 This analysis included information on spinal cord-injured individuals (with and without pressure ulcers), who were treated or evaluated at annual examinations during a 6-yr period. Pressure ulcers were documented by complete chart reviews, patient interviews and, usually but not always, by direct observations.
Selection of Patients
From the database, we analyzed a subgroup (n = 219) of the SCI individuals. The four inclusion criteria were as follows: (1) a spinal cord injury (International Classification of Diseases [ICD-9CM]12 codes 806.0-806.9, 952.00-952.9, 192.2, 198.3) with a neurologic deficit attributable to a disease or damage of the spinal cord but not the cortices or brainstem; (2) treatment or evaluation at an annual examination between August 4, 1987, and December 14, 1993, at the Spinal Cord Injury Unit of the Department of Veterans Affairs Medical Center, Castle Point; (3) mobility limitations, attributable primarily to a neurologic deficit; (4) a traumatic or neoplastic etiology with a sudden onset of SCI. Degenerative causes were excluded.
“Mechanism of injury” was recorded as 11 categories plus “other.” Only four patients in this other group had a neoplasm that caused the SCI. These patients were included in the analyses to ensure that the results were suitable for the general SCI population.
A pressure ulcer was defined as a lesion on any skin surface that resulted from pressure, shear, or friction.13 The standard staging system developed by the National Pressure Ulcer Advisory Panel14 was used for pressure ulcer severity as follows: “Stage I, nonblanchable erythema of intact skin, the heralding lesion of skin ulceration; Stage II, partial thickness skin loss involving epidermis and/or dermis. The ulcer is superficial and presents clinically as an abrasion, blister, or shallow crater; Stage III, full thickness skin loss involving damage or necrosis of subcutaneous tissue that may extend down to, but not through, underlying fascia. The ulcer presents clinically as a deep crater with or without undermining of adjacent tissue; Stage IV, full thickness skin loss with extensive destruction, tissue necrosis or damage to muscle, bone, or supporting structures (for example, tendon or joint capsule).”
Comorbidities were defined as “treated and documented conditions that existed before the development of the current pressure ulcer.” Complications of the pressure ulcer or treatment were not recorded as comorbidities. Impaired cognitive function was defined as a mental disorder comorbidity, a mental status change, or a significant personality disorder. Patients classified as comatose, semicomatose, confused, or lethargic also were coded as having an impaired cognitive function. For these conditions, a patient had to have a history and treatment documented in the chart. We did not use the Diagnostic and Statistical Manual of Mental Disorders criteria to define these conditions.
Sensitivity was defined as the percent of patients with a history of a pressure ulcer who had a score equal to or greater than the cutoff point. Specificity was defined as the percent of the patients, in the control group (no pressure ulcers), who had a score below the cutoff point.
The unit of analysis was a person with an SCI; no individual was included in the analysis more than once. SCI individuals, with a history of one or more pressure ulcers, were compared with SCI individuals in the control group. Categoric variables were assessed (e.g., mechanism of injury) with the Pearson's chi-square statistic and ordinal variables (e.g., level of activity) with the Mantel-Haenszel test for linear association. Continuous variables (e.g., age) were analyzed with the unpaired Student's t test and the one-way analysis of variance, if normally distributed, and the Mann-Whitney U test, if not normally distributed. Potential risk factors that were biologically plausible and had P values of less than 0.2 in the univariate analysis were used in multivariate analysis.
For those with pressure ulcers, logistic regression was used to compare those who developed a recurrence with those who did not. Similarly, patients with Stages 3 or 4 pressure ulcers were compared with Stages 1 or 2. In addition, to determine the risk factors that were independently associated with other measures of outcome, we used multiple forward-stepwise linear regression. These other outcomes were as follows: the total number of pressure ulcers, sites, recurrences, and the size, stage, depth, and time from SCI to first pressure ulcer. A new scale was developed using the cumulative evidence from these analyses. For a factor to be included in this scale it had to meet all four of the following criteria: (1) multiple and consistent statistical evidence that this factor was significantly associated with pressure ulcer development or severity in this study; (2) a biologically plausible mechanism for pressure ulcer development; (3) co-evidence in the literature that this factor was associated with the development of pressure ulcers; (4) an improvement in the sensitivity and specificity of the total score caused by the addition of each factor.
These four criteria were also used to set the cutoff points for the continuous variables included in the scale. The scale was developed by adding one factor at a time, followed by an evaluation of the improvement and distribution of the total score. Multivariate models used only cases that had complete data for all variables used in the analysis. No individuals were excluded from this report because of missing data.
Fifteen risk factors met the four inclusion criteria for the new scale (Fig. 1); operational definitions for this scale are provided in Fig. 2. For each of the 219 patients, the weighted values for the scale factors were totaled. Total individual scores ranged from 0 to 19; the highest possible score was 25, and the median value was 7. Patients in the pressure ulcer group had a significantly higher median score than the nonpressure ulcer control group (7 v 5; P < 0.001). Fig. 3 illustrates sensitivity and specificity at various cutoff points. Using a cutoff point of ≥6 provided a sensitivity of 75.6 percent and a specificity of 74.4 percent. Positive predictive value was 92.4%; negative predictive value was 42.7%.
All 49 patients with scores of nine or higher, had a history of one or more pressure ulcers. The proportion of SCI patients with poor outcome increased with increasing scores (Fig. 4). The median number of pressure ulcers, sites, and recurrences also increased with higher scores (Fig. 5). Fig. 6 is a cumulative frequency polygon displaying the incidence of pressure ulcers at each level of the scale.
Characteristics of the Patients
Among the 219 consecutive spinal cord-injured patients who met the inclusion criteria, 176 (80.4 percent) had a history of one or more pressure ulcers. In 43 (19.6 percent) patients, there was no documented history of a pressure ulcer; they were used as a control group. Table 1 provides a comparison of the pressure ulcer group with the controls. These groups were similar in weight, height, body mass index, employment status, and age at the time of injury. When the hospital and nursing home groups were combined, the predominant living situation was a significant factor. All 17 patients whose predominant living situation was classified as a nursing home or hospital had a pressure ulcer. This was significantly higher than in the 202 patients who lived in individual or group homes (100.0 v 78.7 percent; P = 0.034).
Among the 176 patients with a history of one or more pressure ulcers, the median number of ulcers was three (range, 1-24). For two patients, their most severe pressure ulcer was Grade 1; for 40 patients, it was Grade 2; for 61 patients, it was Grade 3; for 73 patients, it was Grade 4. Ischial pressure ulcers developed in 103 patients, followed by 91 sacral, 38 trochanteric, and 37 heel.
To enable comparisons with other SCI populations in which this scale may be applied, the following results are provided. All spinal cord injuries in this study occurred between 1942 and 1993 (median, 1978). For all 219 patients, the age at the time of SCI ranged from 18 to 88, with a mean of 35.6 ± 15 (median, 31). More patients (n = 16) were injured at age 23 than at any other age; 47.9 percent occurred between ages 18 and 30.
Time from SCI to last follow-up averaged 17.2 ± 12.1 (range, 0.2-51.3) yr. The median time was 14.4 yr. Persons with pressure ulcers were not followed significantly longer than persons who served as controls (17.9 v 14.2; P = 0.073).
During the interval from the SCI to the last follow-up, 26 of the 219 (11.9 percent) patients died. The group with pressure ulcers had a significantly higher proportion that died (Table 1). These figures were based on deaths from all causes, not specifically pressure ulcer complications. Of the 26 patients who died, one had a neoplastic diagnosis (hemangioblastoma) at the age of 75 yr; this patient also had two pressure ulcers.
Spinal Cord Injuries
In the 92 patients with a complete SCI, 89.1 percent had a pressure ulcer (Table 2). Among the 127 with an incomplete SCI, a significantly lower proportion of patients had a pressure ulcer (74 percent; P = 0.005).
Overall, 90 percent of the patients (197/219) used wheelchairs for mobility, and 42 percent (92/219) had a complete SCI. Surprisingly, mechanism of injury and “level of spinal cord injury” were not significantly associated with pressure ulcer development. Besides the nine mechanisms of injury groups listed in Table 2, 34 patients were categorized as other. Their mechanisms included water skiing, downhill skiing, struck by a falling tree, struck by a wave, a fall on ice, and injured by machinery. In four patients with a mechanism of injury of other, the SCI was related to a neoplasm (hemangioblastoma, thoracic tumors, and two with astrocytoma).
Statistically, the “level of activity” was the most significant risk factor for pressure ulcer development. The proportion of patients with pressure ulcers was 30 percent for the 20 ambulatory patients, 85.3 percent for the 197 persons using a wheelchair, and 100 percent in the 2 patients confined to bed (P < 0.001). Among the ambulatory patients, there was a similar incidence of pressure ulcers in those who could walk with and those who could walk without assistance; these groups were, therefore, combined. Persons who used power and manual wheelchairs also had similar pressure ulcer results and were combined. For patients who were temporarily hospitalized for the treatment of a pressure ulcer, their living situation, activity, and mobility levels before the hospitalization were used for this analysis.
Several other measures of functional status were associated with pressure ulcers (Table 3). Mobility level was associated with pressure ulcers, independent of level of activity. The proportion with pressure ulcers was 50 percent for the 10 patients with full mobility, 81.6 percent for the 207 with slight to limited mobility, and 100 percent in the 2 patients who were immobile (P = 0.023). Mobility level added predictive information, independent of the activity level.
Most patients (211/219) had minimal bladder control or were constantly moist. Yet the variable “urinary incontinence level” did help discriminate between the pressure ulcer and the control groups (81.5 v 50 percent; P = 0.028). Autonomic dysreflexia was documented for 30 patients. Although in the univariate analysis, the results were statistically borderline, this group did have a higher proportion with pressure ulcers (93.3 v 78.3 percent; P = 0.054).
All 24 patients with pulmonary disease had a pressure ulcer; this is in contrast to 77.9 percent of the patients without pulmonary disease (P = 0.010; Table 4). Diabetes and renal disease were twice as prevalent in the pressure ulcer group, although our sample size was not large enough to show statistical significance. The proportion with a history of pressure ulcers was higher (90.2 v 77.4 percent; P = 0.044) in the 51 patients with either diabetes or serum glucose levels ≥110 mg/dl. Among the 41 patients with renal disease, 36 patients had a pressure ulcer.
Patients with either cardiac disease or an abnormal electrocardiogram were also at increased risk for pressure ulcers (90.9 v 77.7 percent; P = 0.049). In fact, 40 of the 44 patients in this group had a pressure ulcer. Impaired cognitive function was also associated with pressure ulcers (93.1 v 78.4 percent).
Table 5 shows significant physiologic differences between the pressure ulcer and control groups for albumin, white blood cell count, carbon dioxide content, and hematocrit. Low hematocrit was a risk factor for several types of pressure ulcer outcome. Hematocrit was inversely related to lifetime total ulcers, the depth of the ulcer, and the number of different sites.
For 40 patients, hematocrit values were below 36%. In this group, 87.5 percent had pressure ulcers compared with 79.1 percent of the patients with hematocrit ≥36% (P = 0.224). Of the 212 patients with hematocrit data, only 5.7 percent had levels below 30%. All 12 of these patients had a history of pressure ulcers.
The mean albumin in the pressure ulcer group was significantly lower compared with controls (3.73 v 3.97 gm/dl; P = 0.025). Although <3.4 gm/dl was selected for the scale, optimal albumin, based on the 95 percent confidence intervals, for SCI patients is probably >3.8 gm/dl. Of the 199 patients with serum albumin data, only 3.5 percent had an albumin level below 3 gm/dl; all seven of these patients had a pressure ulcer. Low albumin was significantly associated with lifetime total pressure ulcers, number of different sites, and recurrences. We also found that albumin was inversely related to the stage of the worst pressure ulcer. There were 62 patients with either an albumin of <3.4 gm/dl or a total protein of <6.4 g/dl. Of these patients, a higher proportion had a history of pressure ulcers (91.9 v 75.8 percent; P = 0.007).
Patients with pressure ulcers were twice as likely to be current cigarette smokers and, on average, had a longer smoking history (Table 6). In 88 patients who never smoked, 73.9 percent had a pressure ulcer. This increased to 77.6 percent in the 58 former smokers and 90.1 percent in the 71 current smokers (P = 0.011). For the 176 patients with a pressure ulcer, the recurrence rates of the first pressure ulcer were 26.9 percent for never, 37.8 percent for former, and 42.2 percent for current smokers. Although alcohol use was highly correlated to cigarette smoking, the consistency of the smoking evidence indicated that cigarette smoking was a more direct risk factor than drinking.
The median time from SCI to the first pressure ulcer was 2 yr (716 days). This interval ranged from 1 day to 46 yr. The group with an early ulcer (within the first 2 yr after their SCI) were significantly older at the time of the SCI, compared with those with a late ulcer (38.3 v 33.2 yr; P = 0.026).
A Stage 3 or a Stage 4 ulcer developed in the 134 patients (61.2 percent). Within this group, we noted a higher prevalence of pulmonary disease, a history of smoking, diabetes, complete SCI, and autonomic dysreflexia (compared with the patients without Stages 3 or 4 ulcers). Of the 176 patients with pressure ulcers, six factors were significantly different for the Stages 3 and 4 ulcer patients compared with Stages 1 and 2 ulcer patients. These severe ulcers were associated with cigarette smoking, pulmonary disease, diabetes, low hematocrit, cardiac, and renal disease.
Combining evidence from this analysis and the extensive pressure ulcer literature enabled us to create a new pressure ulcer risk assessment tool, specifically designed for SCI individuals. It is unclear from the current literature, which factors are the independent, optimal predictors of pressure ulcers for SCI patients. Therefore, this new scale is an important tool for preventing pressure ulcers in this population.
Advantages of the new scale include a selection of risk factors and cutoff points based on scientific evidence from SCI patients. Reliability has been improved by replacing subjective assessments with objective data.15, 16 An additional advantage is that, in this scale, patients with no risk factors are given a score of zero, and those at higher risk have higher scores.
A major strength of this study was the length of time patients had been at risk for pressure ulcers. This long follow-up period proved essential for developing an accurate and stable scale. Many patients develop pressure ulcers soon after their SCI; the median time to pressure ulcer was 2 yr. Therefore, including all patients, despite the length of follow-up, was an important part of the study design. The control group's average follow-up of 14 yr shows that these patients have had sufficient opportunity (time at risk) to develop a pressure ulcer.
Among the pressure ulcer risk factors, critical interactions were detected between these pairs: diabetes and smoking, pulmonary disease and malnutrition, abnormal electrocardiograms and malnutrition. Smokers may have different thresholds for the serum values reported here. Although the connection between serum albumin and pressure ulcers has been studied by many, well designed studies are needed to learn how this information can benefit the patient. These interactions, thresholds, and the questions raised by The National Pressure Ulcer Advisory Panel14 require further exploration.
Prevention of pressure ulcers, before extensive injury has occurred, is of primary importance for SCI individuals. Based on data for 219 SCI individuals and risk factors identified from the literature, this scoring system will identify SCI patients at high risk for developing pressure ulcers. Although preliminary evidence suggests that the incidence of pressure ulcers can be reduced in the high-risk group among SCI patients,17 long-term studies are necessary.
Taylor18 warned, “The promotion and use of assessment tools that are untested or that have poor validity or reliability can jeopardize the delivery of cost-effective pressure ulcer prevention by encouraging intervention decisions based on chance rather than on sound clinical data.” Therefore, further research must test the validity, reliability, and appropriate time to use new scoring systems. Until such research is completed, the 15 factors in this new scale should be considered indicators for SCI patients at high risk of pressure ulcer development and not causative factors.
We believe that this tool is easy to use and should prove cost-effective. Application of this scale and future studies to refine it will promote the efforts of those working to achieve the goal of the National Pressure Ulcer Advisory Panel, “To assist health care professionals in reducing the incidence of pressure ulcers by 50 percent by the year 2000.”
We thank Dr. Vivian Beyda, the Eastern Paralyzed Veterans Association, Dr. Richard Salcido, Dr. Jane A. Petro, Dr. Paul Visintainer, Mr. Edward May, Dr. Martin W. Ferguson-Pell, the Disability Prevention Program, New York State Department of Health, and the Centers for Disease Control and Prevention.
1. Byrne DW, Salzberg CA: Major risk factors for pressure ulcers in the spinal cord disabled: a literature review. Paraplegia,
in press, 1996.
2. Gosnell DJ: Pressure sore risk assessment: Part II-analysis of risk factors. Decubitus
3. Hodge J, Mounter J, Gardner G, Rowley G: Clinical trial of the Norton Scale in acute care settings. Aust J Adv Nursing
4. Hunter SM, Cathcart-Silberberg T, Langemo DK, et al.: Pressure ulcer prevalence and incidence in a rehabilitation hospital. Rehabil Nurs
5. Langemo DK, Olson B, Hunter S, Hanson D, Burd C, Cathcart-Silberberg T: Incidence and prediction of pressure ulcers in five patient care settings. Decubitus
6. Fletcher DJ, Taddonio RF, Byrne DW, et al.: Incidence of acute care complications in vertebral column fracture patients with and without spinal cord injury. Spine
7. Norton D, McLaren R, Exton-Smith AN: An Investigation of Geriatric Nursing Problems in Hospital
. London: The National Corporation for the Care of Old People, 1962. Reprinted by Churchill Livingston, Edinburgh, 1975:193-238.
8. Andersen KE, Jensen O, Kvorning SA, Bach E: Prevention of pressure sores by identifying patients at risk. BMJ
9. Towey AP, Erland SM: Validity and reliability of an assessment tool for pressure ulcer risk. Decubitus
10. Batson S, Adam S, Hall G, Quirke S: The development of a pressure area scoring system for critically ill patients: a pilot study. Intensive Crit Care Nurs
11. Salzberg CA, Harmatz A, Byrne DW, D'Ariano G, Petro JA: Development of a computerized data base to evaluate pressure ulcers. Decubitus
12. Department of Health, and Human Services: The International Classification of Diseases,
9th rev., clinical modification: ICD-9-CM. Vol. 1. Diseases tabular list. 2nd ed. Washington, DC: Government Printing Office. (DHHS publication no. (PHS) 80-1260), 1980.
13. Parish LC, Witkowski JA, Crissey JT: The Decubitus Ulcer
. New York: Masson Publishing, Chicago: Year Book Medical Publishers, 1983.
14. The National Pressure Ulcer Advisory Panel. Pressure ulcers prevalence, cost and risk assessment: consensus development conference statement. Decubitus
15. Oot-Giromini BA: Pressure ulcer prevalence, incidence and associated risk factors in the community. Decubitus
16. Peiper B, Adams W, Mikols C, Mance B: Visceral protein nutritional assessment of patients placed on a high or low air-loss bed. J Enterostom Ther
17. Rothery FA: Preliminary evaluation of a pressure clinic in a new spinal injuries unit. Paraplegia
18. Taylor KJ: Assessment tools for the identification of patients at risk for the development of pressure sores: a review. J Enterostom Ther
How to Obtain CME Category 1 Credits
To obtain CME Category 1 credit, this educational activity must be completed and postmarked by December 31, 1997. Participants may read the articles and take the exams issue by issue or wait to study several issues together. After reading the three CME Articles in this issue, participants may complete the Self-Assessment Exam by answering the questions on the CME Answering Sheet and the Evaluation pages, which appear later in this section. Send the completed forms to: CME Department, Association of Academic Physiatrists, 5987 E. 71st Street, Suite 112, Indianapolis, IN 46220. Documentation can be received at the AAP National Office at any time throughout the year, and accurate records will be maintained for each participant. CME certificates are issued only once a year in January for the total number of credits earned during the prior year.