A Comparative, Descriptive Study of Systemic Factors and Survival in Elderly Patients With Sacral Pressure Ulcers

Sacral pressure ulcers (PUs) are a serious complication in frail elderly patients. Thin tissue in the sacral area, low body mass index, and anatomical location contribute to the development of sacral PUs. A comparative, descriptive study was conducted to identify patient systemic factors associated with sacral PUs and to compare survival time in patients with and without PU.

  All consecutive patients with PUs (n = 77) and without sacral PUs (n = 53) admitted to the skilled nursing department of a geriatric hospital in Jerusalem, Israel between July 1, 2008 and December 31, 2011 were eligible to participate. Charts of previously admitted patients were abstracted and patients were prospectively followed until discharge, death, or the end of the study. Patient demographics, comorbidities, nutritional status, physical and cognitive function (measured using the Reisberg’s Functional Assessment Staging Tool [FAST], Stages of Dementia of Alzheimer Scale, and the Glasgow Coma Scale), PU status, number of courses of antibiotic treatment during admission, length of hospitalization, and mortality were compared between patients admitted with and without a sacral PU using descriptive and univariate statistics. Logistic regression models were used to estimate the odds ratio (OR) and 95% confidence intervals (CI) for sacral PU versus without PU by study covariate. The association between sacral PU and survival time was assessed using Kaplan-Meier models. Patients with a sacral PU were significantly older (average age 81.60 ±10.78 versus 77.06±11.19 years old, P = 0.02) and had a higher prevalence of dementia (70% versus 30%, P = 0.007), Parkinson’s disease (92.3% versus 7.7%, P = 0.03), and anemia (67.7% versus 32.3%, P = 0.06) than patients admitted without a PU. Patients with a sacral PU also had a lower body mass index (23.1 versus 25.4, P = 0.04), and lower hemoglobin (10.54 versus 11.11, P = 0.03), albumin (26.2 versus 29.7, P = 0.002), and total protein levels (61.3 versus 65.7, P = 0.04). In addition, antibiotic treatment was significantly higher in the patients with PU (50.6% versus 28.3%, P = 0.01). Patients with a sacral PU also had significantly lower physical and cognitive functioning scores and their median survival time was 70 days compared to 401 days in the non-PU group (P <0.001). These findings are generally consistent with the literature regarding risk factors for PU development and confirm the need for preventive measures. In addition, clinicians need to address the overall goal of patient care and patient quality of life when considering PU management interventions in this patient population.

Pressure ulcers (PU) are a serious complication in bedbound patients. The National Pressure Ulcer Long-Term Care Study¹ in the United States reported 60% to 70% of PUs occur in the pelvic girdle (sacrum, greater trochanter, and ischium) and 25% to 35% occur in the lower legs, heel, ankle, and lateral area of the foot. These percentages are confirmed in additional epidemiologic studies conducted in nursing homes in the US by Smith.² A prevalence study³ of 1,173 patients in 55 wards in Ostergoland County, Sweden found 31% of PUs were situated in the sacrum and 23% in the heels. A national study⁴ of 17,560 patients performed in acute care facilities from 34 states in the US found a sacral PU prevalence of 26% and a 31% incidence.

  The anatomical location of the sacral area at the end of the spinal column with the coccyx in proximity of the anus exposes the skin to fecal and urine secretion.⁵ In a comparative study,⁶ sacral tissue thickness was found to be approximately 1–3 cm, which is thinner than other places of the body when measured using B-mode ultrasound, and renders the skin of sacral area more vulnerable to ulcer formation. In a randomized, controlled study, Ek et al⁷ reported patients with low body mass index (BMI) and low food intake developed significantly more PUs in comparison to patients who received extra nutritional support. In experimental study, Bergstrand et al⁸ reported a positive relationship between low BMI and tissue thickness under high compression in the sacral tissue of elderly patients. When sacral skin tissue thickness is compressed, skin thickness decreases by 60%, and the thinner tissue becomes a contributing factor for development of sacral PU.⁸ In comparative studies,^9,10 blood perfusion and interface pressure were found to play important roles, especially in people of advanced age and with disabilities; occlusion of the sacral skin blood flow leads to tissue ischemia and oxygen deprivation that make the skin prone to PU development.

  The sacral area is influenced by changes in the position of the patient. Lifting the head of the bed during feeding and elevating the legs in order to relieve pressure on the heels direct the pressure and shearing forces toward the sacral area. Similar conditions prevail when the patient is in a sitting position or slipping in a chair.¹¹

  The purpose of this comparative, descriptive study was to identify systemic factors and outcomes in patients admitted to a skilled nursing department with and without a sacral PU.

Setting and patients. All patients consecutively admitted between July 1, 2008 and December 31, 2011 with PUs in the sacral area and patients without any PUs admitted to the Skilled Nursing Department of the Herzog Hospital Jerusalem were eligible for participation. Admission criteria for the Skilled Geriatric Nursing Department included being bedbound with one or more of the following: extensive and deep PU (grade III or IV, according to the European Pressure Ulcer Advisory Panel [EPUAP] staging system), terminal cancer requiring palliative care, renal failure requiring hemodialysis, oxygen dependence, noninvasive ventilation (continuous positive airway pressure [CPAP] use), and tracheostomy.

  Ethics approval for the study was received from the Helsinki Committee of Herzog Hospital. Patient confidentiality was ensured by encoding all of the data according to numbers without the name or identification number of the patient.

  Procedures and variables. A retrospective chart review was conducted and, if the patient was eligible to participate, pertinent patient data were abstracted and patients were followed prospectively until discharge, death, or the end of the study period (December 31, 2011). The attending physician completed a structured questionnaire for every patient upon admission and at the end of follow-up (date of discharge, death, or end of study period on December 31, 2011).

  Data collected from the medical charts included sociodemographic characteristics (age, gender, family status, transfer status, previous location, diseases/comorbidities), use of urinary catheter and tube feeding, tracheostomy, and PU assessment (site, stage, and Pressure Ulcer Scale for Healing¹² [PUSH] scores). Changes in PU status were monitored weekly using the PUSH¹² and National Pressure Ulcer Advisory Panel¹³ (NPUAP) staging system.

  Mental state assessment using the Mini-Mental State Examination (MMSE) was impossible to perform in most cases; therefore, the Reisberg’s Functional Assessment Staging Tool (FAST), Stages of Dementia of Alzheimer Type¹⁴ (SDAT), the Glasgow Coma Scale¹⁵ (GCS), and the Norton Scale¹⁶ for PU risk were employed. The FAST Scale¹⁴ assesses cognition as Stage 1: no cognitive decline; Stage 2: very mild cognitive decline; Stage 3: mild cognitive decline; Stage 4: moderate cognitive decline; Stage 5: moderately severe cognitive decline; Stage 6: severe cognitive decline; and Stage 7: very severe cognitive decline. The GCS¹⁵ measures consciousness (scores range from 3 to 15, with a score of 5–9 indicating vegetative state and >10 indicating alertness. The Norton Scale¹⁶ score ranges from 4 to 24 points and assesses 5 risk-based items: sensory perception, physical, mental condition, incontinence, activities, and mobility. Scores <14 indicate high risk for developing PUs. These instruments are well established diagnostic clinical tools.^17-19 The Norton Scale was administered at patient admission; likewise, the FAST and GCS were assessed upon patient admission.

  New ulcers and Stage I ulcers were not included. Data regarding patient nutritional status, including BMI, weight, cholesterol, and total lymphocyte count also were collected and serum albumin, total protein, and hemoglobin levels were obtained at admission. Complications such as systemic infections — documented by the number of provided antibiotic courses, length, and outcomes of hospitalization — were summarized at the end of the study.

  Data collection and analyses. Data were collected by the attending physician from the medical charts and entered into the data analysis software. Descriptive statistics were used to assess all data for all study participants. The characteristics of the 2 groups were compared using univariate analysis. Logistic regression models were used to estimate the odds ratio (OR) and 95% confidence intervals (CI) for sacral PU versus without PU by study covariates. All variables were entered at the first stage and backward elimination was used to remove nonsignificant factors; criterion for removal from the model was a P value >0.10.

  The association between sacral PUs and survival time was assessed using the Kaplan-Meier model. All data were analyzed with SPSS software (version 17.0; Chicago, IL). All statistical tests were 2-sided; a P value <0.05 was considered significant. Length of stay was calculated and entered as date of death/discharge minus the date of admission plus 1.

During the study period, 130 patients admitted to the Skilled Nursing Department met the inclusion criteria. Of those, 77 (59.2%) were admitted with a diagnosis of sacral PU and 53 (40.8%) were admitted with no PU at all. At admission, the average age of all patients was 79.7 ± 11.3 years (median age 82). The sacral PU group was significantly older (81.6 ± 10.78 versus 77.0 ±11.19, P = 0.02). Of the 130 patients, 83 (64%) were transferred from general hospitals after acute episodes of illness; sacral PU were found to be significantly more prevalent in the patients who were transferred from home and nursing institutions than from general hospitals (71% versus 53%, P <0.05) (see Table 1).

  Sacral PU patients had significantly more underlying medical conditions than the non-PU group. Significant differences were noted between the 2 patient groups with regard to dementia (49 [70%] versus 21 [30%], P = 0.007; OR: 2.6, 95% CI: 1.29-5.47; presence of Parkinson’s disease: 12 [92.3%] versus 1 [7.7%], P = 0.03, OR: 9.5, 95% CI: 1.20-76.2; and anemia on admission: 42 [67.7%] versus 20 [32.3%], P = 0.06, OR:1.98, 95% CI: 0.97-4.04).

  A significantly higher prevalence of cancer was found in the non-PU group: 21 (60%) versus 14 (40%), P = 0.008, OR: 0.34, 95% CI: 0.15-0.76. The prevalence of other medical conditions such as cerebrovascular accidents, diabetes mellitus, ischemic heart disease, peripheral vascular disease, was not significantly different between the 2 groups.

  The mean GCS score was significantly lower (11.39 versus 12.91, P = 0.006), and the mean FAST score was significantly higher in the sacral PU group (6.47 versus 6.19, P = 0.01). The total Norton Scale score was significantly lower in the sacral PU group (8.91 ± 2.0 versus 10.68 ± 3.0, P = 0.0004). Feeding tubes were significantly more common in sacral PU patients (58 [68.25%] versus 27 [31.76%], P = 0.004).

  A higher prevalence of urinary catheters was found in the sacral PU group (51 [64%] versus 29 [36%], P = 0.18) (see Table 1). Among patients in both groups, 54 (41.5%) received 1 or more courses of treatment with antibiotics; patients in the PU group had a significantly higher use of antibiotic treatment episodes (39 [50.6%] versus 15 [28.3%], P = 0.01, OR: 2.6, 95% CI:1.23-5.48), indicating greater infection severity.

  Weight, BMI, hemoglobin levels, serum albumin, and total protein at admission were significantly lower in the sacral PU group (weight 61.6 versus 69, P = 0.03; BMI 23.1 versus 25.4, P = 0.04; hemoglobin 10.54 versus 11.11, P = 0.03; albumin 26.2 versus 29.7, P = 0.002; total protein 61.3 versus 65.7, P = 0.04). The total lymphocyte count and the total cholesterol were similar in the 2 groups (see Table 2).

  Patients with a sacral PU had significantly shorter hospitalization (152 days versus 240 days, P = 0.04) (see Table 1). The median survival time of the patients with sacral PU was significantly shorter than of the non-PU group (70 days versus 401 days, P >0.001, log rank test) (see Figure 1).

  Among patients with a sacral PU at admission, 27 out of 77 (34%) had a Stage IV PU with a PUSH score >10. At the end of the study, the PUSH score of 7 patients had improved (<10).

In this comparative, descriptive study, diseases such as dementia, Parkinson’s, and anemia, as well as low levels of consciousness and severe cognitive impairment, were significantly more prevalent among persons admitted with a sacral PU. Parkinson’s disease and advanced dementia are common neurodegenerative disorders of the elderly.²⁰ The natural course of these diseases includes impaired mobility and increased rigidity and spasticity. Atiyeh and Hayek²¹ described a case report of a young quadriplegic patient with severe spasticity presenting with a large infected ischial PU. Jaul²² reported in a cohort study of patients in a skilled nursing department that severe spasticity is one of the major risk factors for atypically located PUs. In a comparative study,²³ advanced dementia was commonly accompanied by aspiration pneumonia, febrile episodes, difficulties with eating, and results in malnutrition.

  Studies show cognitive impairment is a risk factor for PUs. Based on results from a large database of 75,168 elderly patients living in a community setting in the UK, Margolis et al²⁴ reported various medical conditions as risk factors for PU. Their multivariate analysis showed the presence of Alzheimer’s disease and Parkinson’s disease indicated high relative risk (2.0 and 2.3, respectively) for PU development. A prospective cohort study²⁵ of 323 nursing home residents with advanced dementia, followed for 18 months before death, reported 38.7% developed a PU.

  Anemia of chronic disease may indicate the severity, progression, and instability of the patient’s medical condition that contribute significantly to PU development. Anemia was common in the PU group in this study (see Table 1), confirming the results of Chauhan et al’s²⁶ cross-sectional study in medical and surgical wards in India, where anemia was a significant risk factor for the development of PUs. Anemia of chronic diseases is known to have a negative effect on PU healing due to impaired tissue oxygenation. In their retrospective study, Keast and Fraser²⁷ reported that providing erythropoietin to 4 patients with a grade 4 PU improved their mean hemoglobin level and decreased mean wound surface and depth. Margolis’ cohort study²⁴ showed anemia had a relative risk factor of 1.7 in univariate analysis but not in multivariate analysis. The explanation for this discrepancy is that anemia among the elderly in community settings is less severe and less likely to be due to chronic disease than anemia occurring among patients in a skilled nursing department. The current study supports the conclusion anemia is more prevalent in patients with PU; whether anemia is a risk factor for PU development remains to be determined.

  Urinary catheters were more common in the sacral PU patients due to urinary incontinence. However, the use of urinary catheters in both groups was not statistically significantly different (63% in PU group versus 36% non-PU group). A longitudinal, repeated-measures study²⁸ showed patients with urinary incontinence associated with dermatitis in long-term acute care had a PU prevalence rate of 35%. A cohort study²⁹ of frail urinary incontinent elderly women with indwelling urinary catheters residing in the community demonstrated a high prevalence of PUs (44%).

  A statistically significant higher prevalence of tube feeding was found among the sacral PU group due to difficulty swallowing. Oral feeding is known to be challenging in patients with low-level consciousness. A large number of studies indicate poor nutrition contributes to the appearance of PU. Ek et al’s⁷ regression analysis showed low BMI; low albumin; and limited mobility, activity, and food intake are predictors for PUs; in this study, low BMI, low total protein and albumin, and anemia were found to have a significantly higher frequency in the sacral PU patients, confirming the relationship between nutritional status and presence of PUs. A prospective, comparative study³⁰ among 286 adult patients undergoing surgical treatment showed low BMI, weight loss, and low serum albumin were found to be risk factors for PU development. A prospective study³¹ of 412 postsurgical and critically ill patients revealed days without nutrition was a critical factor influencing PU development. In multivariate analysis of their prospective study among institutionalized elderly, Bergstrom and Braden³² found lower dietary protein intake was a risk factor for PU development. Margolis et al²⁴ also reported a high relative risk of 3.0 in the multivariate analysis of PU associated with inadequate nutritional intake.

  Inflammation, deficient hydration, and wound severity affect protein synthesis and consequently contribute to low serum albumin and protein as predictors for PU development.³³ The current study showed the severity of sacral PUs had a number of clinical implications. Patients received more courses of antibiotic treatment due to infections caused by the ulcers or other causes. The length of hospitalization was significantly shorter for the sacral PU group (152 days versus 240 days, P = 0.004) due to the shorter survival time. One third of the patients already exhibited Stage IV PUs on admission. The median survival of the sacral PU patients was significantly less than patients without PUs (70 days versus 401 days, P <0.001).

  The high mortality rate in sacral PUs raises ethical and clinical concerns. In patients with extensive and deep sacral wounds and a poor prognosis for survival, any intervention that causes undue suffering or discomfort (eg, sharp debridement and frequent repositioning) should be avoided. The authors’ recommended treatment for palliative care is to protect the wound from complications and improve the quality of life for the patient by preventing pain, discomfort, bad odor, and depression. The family should be informed and consulted concerning these decisions. Furthermore, measures to prevent sacral ulcers, fortify tissue thickness, and prevent PU progression into advanced stages should be investigated. Nutritional aspects of preventing PUs demands further research as well.

The traditional concerns with recall bias in retrospective studies are not relevant in this case because the review relied on charts rather than patient memory. Another potential limitation of the study relates to extrapolation of the data. Patients in a skilled nursing department are not representative of community-dwelling patients with a PU. Frequently, once community patients have a Stage III or Stage IV ulcer, they are referred to inpatient settings.

In patients admitted to a skilled nursing department, systemic factors such as nutritional status, comorbidities, and functional state differed significantly between patients with and without a sacral PU upon admission. Dementia, Parkinson’s disease, and anemia were also significantly more prevalent within the sacral PU group. Low levels of consciousness and severe cognitive impairment were also significantly more prevalent among the PU patients. Sacral PU patients had a significantly lower survival time compared to those without PU. Patients who had developed a PU in the sacral region had a shorter survival time than those without a PU. The general condition of the PU patients, such as nutritional status and comorbidities, was more severe in comparison to those without PU. The severity of sacral PU as expressed in the low median survival, high number of antibiotic treatments, and high number of comorbidities, demands clinical attention and further investigation involving larger studies.

Dr. Jaul is a Senior Lecturer, Hebrew University Jerusalem; and Director, Geriatric Skilled Nursing Department, Herzog Hospital, Jerusalem, Israel. Prof. Menczel is a Professor of Medicine and Chairman, Department of Geriatric, Herzog Hospital, Hebrew University, Jerusalem, Israel. Please address correspondence to: Efraim Jaul, MD, Herzog Hospital, PO Box 3900, Jerusalem 91035 Israel; email: jaul@zahav.net.il.