Long-Term Outcomes of Full-Thickness Pressure Ulcers: Healing and Mortality
D espite advances in preventive interventions such as pressure-reduction surfaces, risk assessment scales, and improved knowledge of prevention techniques, pressure ulcers continue to occur in healthcare facilities. Pressure ulcer prevalence and incidence studies show that overall rates in the last decade have remained essentially the same, averaging between 8% and 15% from 1989 to 1999.1,2 Wound care specialists do not agree on whether the majority of pressure ulcers are preventable.3 The concept of skin failure in relation to pressure ulcer development has been suggested.4 The Centers for Medicare and Medicaid Services, whose directives are the foundation for the majority of state healthcare regulations, include the determination of avoidable versus unavoidable pressure ulcers in their surveyor guidelines.5 Nevertheless, the occurrence of a pressure ulcer continues to stigmatize caregivers - whether medical/nursing professionals, healthcare facilities, or patient family members - with an aura of negligence. This can result in guilt on the part of the caregiver for "letting" the ulcer occur or the pursuit of legal action by aggrieved family members.
Less publicized in the debate surrounding pressure ulcer risk assessment, prevention, and treatment are follow-up studies on outcomes of patients with pressure ulcers. Berlowitz6 demonstrated in a study of 19,981 nursing home residents that while patients with pressure ulcers are more likely to die, the increased risk is largely related to the frailty and high disease burden of the resident and is not a direct result of the ulcer. Thomas,7 in a study of 286 hospital patients, determined that 59.5% of residents who developed a pressure ulcer died within 1 year of developing the ulcer. It appeared that the development of new pressure ulcers was a marker of coexisting illnesses, impaired nutrition, and functional status and not an independent risk factor for increased mortality. Berlowitz,8 studying a group of 301 nursing home admissions, also discovered that the presence of a pressure ulcer on admission, the development of a new ulcer, and failure of the ulcer to heal were all associated with a two- to threefold increase in the risk of dying during a 6-week period following admission. The results of this study also suggested that the pressure ulcer itself did not cause the observed increased mortality. Other research has shown that 1) death occurs in acute hospitalizations in 67% of patients who develop pressure ulcers,9 2) 55.7% of nursing home residents who die with a pressure ulcer do so within 6 weeks of the onset of the pressure ulcer,10 and 3) nursing home residents with pressure ulcers experience a 6-month mortality rate of 77.3%.11
Data on pressure ulcer healing and mortality outcomes are difficult to acquire in many healthcare settings. A comprehensive patient chart often lacks information due to the common practice of transferring patients to multiple facilities based on the level of care. Acquisition of a complete chart from several facilities is, therefore, a time-consuming and expensive undertaking for any researcher. The Veterans Affairs Health Care System is unique among healthcare systems and since 1999, virtually all patient data for any admission at the facility where this study was conducted are available through the Computerized Patient Record System (CPRS). This allows broad, efficient access to a variety of data from history, physical assessments, and discharge summaries to laboratory tests and progress notes.
To determine if outcomes were consistent with published literature, quality-assurance data on facility-acquired pressure ulcers were obtained and analyzed. The facility, a tertiary referral center in the North Central Texas area, comprises 173 acute care (medical/surgical), 45 intensive care (medical, coronary, surgical, and thoracic), and 120 long-term care (geriatric rehabilitation, interval placement, and hospice) beds. Yearly average admissions to this center total 8,134 for acute care, 956 for intensive care, and 632 for long-term care. For all facilities, the average pressure ulcer prevalence rates (Stage I to Stage IV) were 8.4% upon admission and 6.2% for nosocomial only. Specifically, the total prevalence in acute care was 4.4% (3.1% nosocomial) and 27% (25.6% nosocomial) in intensive care. In long-term care, the prevalence rate was 8.6% (4.9% nosocomial). Ulcer severity distribution was as follows: 40.4% were Stage II, 14.4% were Stage III, 29.9% were Stage IV, and 7.3% could not be staged because the ulcer was covered with eschar.
Inclusion criteria. Data on nosocomial pressure ulcer occurrence for use in quality-assurance projects were collected over a 5-year period from October 1997 to October 2002 by a registered nurse and Certified Wound, Ostomy and Continence nurse. The number of people in the sample was based on consults to the wound care team and does not include every person who may have developed a pressure ulcer at this center. Based on available prevalence data, it was estimated that this sample captures more than 90% of nosocomial pressure ulcers during the study time period. The National Pressure Ulcer Advisory Panel staging system was used to determine severity of the pressure ulcers. Only full-thickness, Stage III and Stage IV ulcers were included in the data set. Ulcers with eschar were included if they were determined to be full-thickness wounds or progressed to full-thickness ulcers.
Exclusion criteria. Since foot and toe lesions resulting from peripheral vascular disease are the result of a pathologic process separate and distinct from pressure ulcer formation, they were excluded from the data set. Persons with spinal cord injury (SCI) also were excluded because this population has a uniquely high risk, prevalence, and history of pressure ulcers when compared to the non-SCI population, which would skew the outcomes data. Specifically, pressure ulcer prevalence in this population ranges from 23.7% to 46% - significantly higher than the non-CSI- injured population.12-14 Pressure ulcers acquired outside the facility also were excluded because records about the exact date of onset are often ambiguous.
Ulcer acquisition. Analysis of the patient records and interviews with staff members determined whether ulcers were hospital-acquired. Date of onset was established as either the date of the original wound care consult or through clearly documented evidence of onset in the patient record. Location of the wound was defined as the major bony prominence underneath the ulcer (eg, sacrum, coccyx, ischium).
Care setting and diagnoses. To ascertain and compare mortality rates, variables and patient data were grouped based on care setting (intensive care, acute care, and long-term care). The acute care setting comprised 20 (27.0%) of all hospital acquired ulcers, and the intensive and long-term care units comprised 30 (40.5%) and 24 (32.4%), respectively. A maximum of two major diagnoses were recorded for each patient because overall patient health status is not always adequately reflected in an admitting diagnosis. For example, a person may be admitted for a urinary tract infection (UTI) but the reason for the UTI was the presence of an indwelling catheter secondary to the effects of a cerebrovascular accident (CVA). Therefore, CVA was used as the major diagnosis rather than UTI. A mortality endpoint of 180 days from ulcer onset was chosen because this time frame is consistent with the results of other published studies and the time-frame for end-of-life or hospice care.
Data analysis. All data were entered into a computer spreadsheet and descriptive statistics were used to analyze the data and calculate ulcer prevalence and incidence. To compare time-to-death outcomes, subjects were divided into two major groups: those who died within 180 days of ulcer onset (the "deceased group") and those who lived beyond 180 days after ulcer onset (the "living group").
Of the 74 patients (one woman, 73 men) who developed nosocomial pressure ulcers over a 5-year period, 51 (68.9%) died within 180 days of first detection of the ulcer (see Table 1). The 1-year mortality rate was 78.4%; the 2-year rate was 83.8%. Patients acquiring ulcers in the acute care setting had a 180-day mortality rate of 75.0%, compared to 66.7% for both the intensive and long-term care settings. The average number of days from ulcer onset to death for all patients combined was 47.0 days with a standard deviation (SD) of 40.0 days (see Table 2). In the long-term care setting, the average number of days to death was 60.9 days (SD 45.5 days), approximately 20 days longer than in the acute care (average 41.3, SD 34.3 days) or intensive care (average 42.2, SD 36.7 days) setting.
The major diagnoses for all patients who developed pressure ulcers were: a history of cerebrovascular accident (18.4%), diabetes of any type (13.6%), cancer of any type (12.8%), and coronary artery disease (11.2%) (see Table 3). Most patients (63.5%) had more than one major diagnosis adding to their disease burden. Diabetes of any type, congestive heart failure, and ethanol abuse were more prevalent in the living group by a few percentage points. Chronic obstructive pulmonary disease, renal failure of any type, and cancer of any type were more prevalent in the deceased group, also by only a few percentage points. The average age of patients at ulcer onset was higher for the deceased group with the exception of the long-term care setting. In this setting, persons who died were an average of 1.6 years younger than the living group (see Table 4).
Resuscitation status variables were also obtained, and patients were divided into four groups: full code, do not resuscitate (DNR) only, withdrawal of care, and hospice (see Table 5). The full code category included patients who expired after full resuscitative efforts were attempted. People were placed in the DNR only category if they died while an active DNR order was in effect and otherwise routine medical care was provided. The withdrawal-of-care group included patients who underwent withdrawal of all routine medical care, save for comfort measures, per family wishes in an acute or intensive care setting. The hospice group comprised patients enrolled in a hospice program. Resuscitation status was unavailable for 13.7% of the deceased group and 53.3% of the living group - the majority of these patients died either at home or in another facility. Of the 74 persons studied, 66 (89%) died within the 5-year period of data collection. In the deceased group, 35% expired with an active DNR in effect, 19.6% died undergoing withdrawal of care, 17.6% had enrolled in a hospice program, and 13.7% died during full code. Of those in the deceased group, 70.7% of the patients and/or their families had some type of acknowledgment, acceptance, or active engagement in end-of-life planning before their deaths. Within the living group, 40.0% died with an active DNR in effect and 6.7% died while in hospice.
Discharge summaries and physician progress notes were used to determine cause of death. Of the 51 patients with an identifiable cause of death, none was determined to be related to the presence of a pressure ulcer (ie, no sepsis or other acute medical problems directly attributable to an open wound). These data lend further support to previously published conclusions that people do not die from the development or presence of pressure ulcers. Rather, people die from the overwhelming disease burden that results in the profound immobility, malnutrition, and decreased tissue perfusion that allow skin failure to occur.
The majority of ulcers for all patients were located in the sacral/coccygeal area (66.2%) and on the heel (16.2%) (see Table 6). Sacral ulcers comprised 68.6% of the deceased group, compared to 60.9% in the living group, while 3.9% of the deceased and 8.7% of the living group had trochanter ulcer. The difference between the deceased and living groups was less than a 3% for the other ulcer locations.
Wound healing outcomes. Among the 51 patients who died within 180 days of ulcer onset, none of the ulcers had healed (see Table 7). Published data reveal a full-thickness pressure ulcer healing time of between 90 and 150 days,15 so lack of healing was not surprising. Among the 15 patients who died between 181 and 1,058 days, only six (40%) of the ulcers were healed by the time of death. Of the eight patients still living after the data collection cut-off date, four (50%) had a healed ulcer. Two wounds in this group failed to heal due to an underlying osteomyelitis that could not be resolved either medically or surgically because of patient morbidities.
The skin is the largest organ system in the body and can experience severe duress during a critical or chronic illness. In many instances, the skin will fail along with other organ systems. This is especially true in elderly and chronically ill patients when the epidermis and dermis undergo significant atrophic changes due to combinations of disease, medications, and aging. Subcutaneous and muscle padding over bony prominences also can diminish from chronic malnutrition and the general catabolic state typical of advanced age.16 Diminished skin strength significantly increases the risk for ultimate skin failure as a person becomes less and less mobile due to the compounding effects of age and disease.
People suffering from an acute critical illness may not have the fluid volume to adequately maintain open capillaries to deliver oxygen and nutrients to muscle cells despite fluid resuscitation, vasopressor support, pressure reduction surfaces, and frequent turning (every 2 hours or sooner). In both acute and chronic illnesses, tissue ischemia can rapidly result from even low pressure despite use of the most advanced pressure reduction surfaces, aggressive turning schedules, and other preventative measures as a result of a critical or end-stage, untreatable disease state.17,18
Current medical technology allows clinicians to extend life-sustaining organ systems far longer than the body's outermost protective covering sometimes can tolerate. Acute and chronic skin failure as a pathologic process is often under-appreciated by the average healthcare provider. In the presence of consistently documented preventive measures, the occurrence of full-thickness skin failure in critically or chronically ill people should be treated not with censure or guilt, but rather as an indicator that the disease burden is simply becoming too much to endure. The occurrence of skin failure in the critically or chronically ill, like any other organ system failure, should be used to establish a dialogue among all involved parties regarding the value of future aggressive medical interventions. Such information could be useful in palliative care decisions. Discussing end-of-life planning appears to be appropriate in these situations.
These findings are consistent with previously published data that establish a link between the development of nosocomial pressure ulcers and impending death. The majority of pressure ulcers in this data set occurred in people with a heavy disease burden who were approaching the end of life. Therefore, the development of a full-thickness pressure ulcer (more accurately termed skin failure) in the critically and chronically ill appears to be a comorbid pathologic process rather than a simple "failure to turn" as is so often proposed. The data also reveal that ultimate resolution of skin failure in the presence of such a disease burden is difficult at best.
Collection and analysis of pressure ulcer healing and mortality outcomes is a missing component of overall pressure ulcer quality-assurance data. Pressure ulcer prevalence and incidence data are relatively easy to acquire, and reporting of such data is now well standardized. Collection of long-term information about people with pressure ulcers can be a time-consuming and expensive undertaking, and a framework for consistent data analysis and presentation is not yet established. Nevertheless, collection of such data is vital. Great emphasis is currently placed on the elimination and remediation of an apparent end-stage, pathologic condition that may not be preventable or treatable in all cases. More research is required into the relationship between pressure ulcer development and ultimate healing and mortality. A framework for consistent reporting of long-term pressure ulcer outcomes also must be established. Healthcare providers require clear guidelines to determine when the development of a pressure ulcer is an avoidable occurrence or simply another failing organ system. - OWM
1. Panel on the Prediction and Prevention of Pressure Ulcers in Adults. Pressure Ulcer in Adults. Clinical Practice Guideline Number 3: Prediction and Prevention. Rockville, Md: US Department of Health and Human Services. Public Health Service. Agency for Health Care Policy and Research; May 1992. AHCPR Publication 92-0047.
2. Amlung S, Miller W, Bosley L. The 1999 national pressure ulcer prevalence survey: a benchmarking approach. Advances in Skin and Wound Care. 2001;14(6):297-301.
3. Brandeis G, Berlowitz M, Katz P. Are pressure ulcers preventable? A survey of experts. Advances in Skin and Wound Care. 2001;14(5):244-248.
4. Witkowski J, Parish, L. The decubitus ulcer: skin failure and destructive behavior. Int J Dermatol. 2000;39(12):894-896.
5. Appendix P: Investigative Protocol--Pressure Sore/Ulcer; Publication Number 7, State Operations Manual. Baltimore, Md.; Centers for Medicare and Medicaid Services; July 2002;38-40,93-96.
6. Berlowitz D, Brandeis G, Anderson J, et al. Effect of pressure ulcers on the survival of long-term care residents. J Gerontol A Biol Sci Med. 1997;52(2):M106-M110.
7. Thomas D, Goode P, Tarquine P, et al. Hospital acquired pressure ulcers and risk of death. J Am Geriatr Soc. 1996;44(12):1435-1440.
8. Berlowitz D, Wilking S. The short-term outcome of pressure sores. J Am Geriatr Soc. 1990;38(7):748-752.
9. Allman R, Laprade C, Noel L, et al. Pressure sores among hospitalized patients. Ann Intern Med. 1986;105(3):337-342.
10. Kennedy K. The prevalence of pressure ulcers in an intermediate care facility. Decubitus. 1989;2(2):44-45.
11. Michocki R, Lamy P. The problem of pressure sores in a nursing home population: statistical data. J Am Geriatr Soc. 1976;24(7):323-328.
12. Chen D, Apple DF Jr, Hudson LM, et al. Medical complications during acute rehabilitation following spinal cord injury - current experience of the model systems. Arch Phys Med Rehabil. 1999;80(11):1397-1401.
13. Walter JS, Sacks J, Othman R, et al. A database of self-reported secondary medical problems among VA spinal cord injury patients: its role in clinical care and management. J Rehabil Res Dev. 2002;39(1):53-61.
14. Krause JS. Skin sores after spinal cord injury: relationship to life adjustment. Spinal Cord. 1998;36(1):51-56.
15. Brown G. Reporting outcomes for stage IV pressure ulcer healing-a proposal. Advances in Skin and Wound Care. 2000;13(6):277-283.
16. Bryant R. Acute and Chronic Wounds: Nursing Management. St. Louis, Mo.: Mosby Year Book;1992:8-10.
17. Peerless J, Davies A, Klein D, et al. Skin complications in the intensive care unit. Clin Chest Med. 1999;20(2):453-467.
18. Kosiak M. Etiology and pathology of ischemic ulcers. Arch Phys Med Rehabil. 1959;40:62-69.