Sacral Skin Temperature and Pressure Ulcer Development: A Descriptive Study

Existing evidence is inadequate to assume increased skin temperature is a risk factor for the development of pressure ulcers (PUs). PURPOSE: The purpose of this prospective, descriptive study was to examine the relationship between sacral skin temperature and PU development. METHODS: Using convenience sampling methods, patients who were hospitalized in the tertiary intensive care unit (ICU) of the internal medicine department of a university hospital in İzmir, Turkey, between April and December 2015 were eligible to participate if they were >18 years of age, had an expected hospital stay of at least 5 days, a Braden score ≤12, and were admitted without a PU. Demographic and clinical data collected included age, gender, body mass index, diagnosis, mattress type, length of follow-up (days), systolic and diastolic blood pressure, body temperature, hemoglobin level, sacral skin temperatures in the supine and lateral positions, room temperature, PU stage and duration, and Braden score. Temperature was measured the day of hospitalization as a baseline measurement (day 1) and once every day thereafter up to 22 days, until the patient did or did not develop a PU, died, was no longer undergoing position change, or was discharged. Sacral skin temperature was taken immediately after the patient was moved to a lateral position following 120 minutes of supine position (referred to as supine position sacral skin temperature measurement) and after 30 minutes in lateral position (referred to as lateral position sacral skin temperature measurement). Data were collected using paper-and-pencil questionnaires and entered into a software program for analysis. Descriptive statistics, Student’s t test, one-way analysis of variance test, Pearson product-moment correlation analysis, and Spearman’s rank-order correlation analysis were used for data analysis. RESULTS: Of the 37 patients who met the inclusion criteria and were monitored for at least 5 days, 21 (56.8%) developed PUs. No statistically significant difference in supine position sacral skin temperature on day 1 or day 5 was found between patients who did and did not develop a PU (36.90˚ C ± 0.29˚ C and 37.15˚ C ± 0.53˚ C, respectively, on day 1; t = -1.656, P = .112; and 37.37˚ C ± 0.53˚ C and 37.30˚ C ± 0.79˚ C, respectively, on day 5; t = 0.259, P = .798). Day 5 lateral position skin temperatures also did not differ significantly between the 2 groups (37.44˚ C ± 0.44˚ C and 37.31˚ C ± 0.75˚ C, respectively; t = 1.306, P = .621). A statistically significant difference was noted between mean sacral skin temperature in the supine position among patients ages 75 to 90 years compared with patients 38 to 64 years and 65 to 74 years (36.93˚ C ± 0.39˚ C; F = 13.221, P = .000) and with use of a viscoelastic mattress compared with an alternating pressure air mattress and continuous lateral rotation alternating pressure air mattress (37.85˚ C ± 0.54˚ C; F = 14.039, P = .000). No statistically significant differences in sacral skin temperatures were found for any of the of the other variables assessed. CONCLUSION: Sacral skin temperatures were not statistically different between ICU patients who did and did not develop a PU. Additional research may help increase understanding of the relationship between skin temperature and PU development. 

Although many pressure ulcers (PUs) are preventable, they continue to be a major problem in all health care institutions and can cause serious pain and discomfort^1,2 and lead to infections,³ prolonged hospital stays,⁴ and increased mortality and morbidity rates.^5-7 

A PU prevalence study⁶ conducted in Turkey (N = 142) found PU rates of 9.8% in intensive care unit (ICU) patients. Karadağ and Gümüşkaya⁸ reported a prevalence rate of 54.8% among 148 surgical patients in Turkey, Tokgöz and Demir⁹ found an incidence rate of 15% among 46 patients in neurologic units of a Turkish university hospital, and Uzun et al¹⁰ reported an 11.6% PU prevalence in among 344 patients in Turkey. In a descriptive study among 1100 long-term care patients in Ireland, Moore and Cowman¹¹ found a PU prevalence rate of 9%.

As a result of the remarkable increase in epidemiological studies in recent years, there is a better understanding of the important role risk factors play in the development of PUs. PU risk can be determined using a risk assessment scale combined with a structured approach, comprehensive skin assessment, and clinical decision-making.^4,12,13 Existing risk assessment scales employ environmental and physiological risk factors such as sensory perception, exposure to moisture, activity and mobility levels, nutritional status, exposure to friction and shear, level of consciousness, age, weight, respiratory status, and hemodynamic status.^14-16 Comprehensive skin assessment determines skin integrity and monitors for the presence of inflammation characterized by erythema, pain, and increases in temperature in at-risk sites.^17,18 Because determining the degree of inflammation is difficult and subjective, more concrete and objective criteria are required.^19,20 

A quasiexperimental study²¹ and a descriptive study²² have shown increased skin temperature is assumed to be a risk factor for PU development and to be a potential harbinger of damage that already occurred. A quasiexperimental study¹⁸ reported that when the skin comes in contact with the support surface, heat accumulates between the support surface and the skin by convection, which increases skin temperature. Observational,²⁰ quasiexperimental,²³ descriptive,^22,24 and cohort²⁵ studies demonstrated that when the skin is exposed to exogenous heat such as that which accumulates between the support surface and the skin, the skin temperature increases and vasodilatation occurs; subsequently, oxygen consumption, carbon dioxide production, and metabolic waste product excretion increases in that area. Skin temperature also can increase as a normal response to PU-related inflammation, especially in areas where tissue ischemia begins^18,20-25; these studies have shown monitoring and evaluating skin temperature increase can be helpful in predicting PU development. However, results of descriptive, comparative, prospective cohort, observational, and quasiexperimental studies^19-22,24-26 regarding the effect of skin temperature on the PU development are inconsistent. The purpose of this prospective, descriptive study was to examine the relationship between skin temperature and PU development in ICU patients. 

Study design and sample. This descriptive study was conducted using a convenience sample of patients who were hospitalized in the tertiary ICU of the internal medicine department of a university hospital in İzmir, Turkey, between April and December 2015. All patients admitted to this unit at the start of the study and who met the study criteria were eligible to participate. The inclusion criteria for the study stipulated study participants must be at least 18 years old, have an expected hospital stay of at least 5 days, have no PU on admission to the unit, and have a Braden score ≤12, which indicates high risk.

Nursing care. The care nurses provided to protect and maintain skin integrity complied with the instructions stated in the hospital quality certificates that were developed based on the 2009 National Pressure Ulcer Advisory Panel (NPUAP) guideline.⁴

Instruments. 

Demographic and clinical characteristics form. Data collected included patient age (categorized as 38 to 64 years, 65 to 74 years, and 75 to 90 years), gender, body mass index (normal weight: 18.5-24.9 kg/m², overweight: 25-29.9 kg/m², obese: ≥30 kg/m²), diagnosis, systolic and diastolic blood pressure, body temperature, hemoglobin level, sacral skin temperatures in the supine and lateral positions, PU stage, time when the PU was first noted after admission, and Braden Risk Assessment Scale score, as well as room temperature, mattress type (alternating pressure air mattress, continuous lateral rotation alternating pressure air mattress, viscoelastic mattress), and length of follow-up (days). Blood pressure, body temperature, and sacral skin temperatures were measured once daily; patient serum hemoglobin levels were extracted daily from the medical records by the researcher. If a PU developed, it was assessed and recorded as Stage 1 through Stage 4, as defined in the NPUAP guideline.⁴ 

Braden Risk Assessment Scale. The scale has 6 subscales: sensory perception, moisture, activity, mobility, nutrition (rated from 1 to 4), and friction and shear (rated from 1 to 3). The total score ranges from 6 to 23; lower scores indicate higher risk (patients with scores of 15 to 18 points are at mild risk, 13 to 14 at moderate risk, 10 to 12 at high risk, and ≤9 points at very high risk).^15,27,28 Risk assessment commenced within the first 24 hours of admission to the ICU and was repeated daily. 

Sacral skin temperature measurement. Participant sacral skin temperatures were measured using the PlusMed Infrared Temporal Artery Thermometer (model pM 1-802; Istanbul, Turkey). The thermometer was supplied by the manufacturer and calibrated twice by an authorized technical service separate and independent from the manufacturer, once before the study began and again during the study (measurements ranged from 36˚ C to 39˚ C˚ ± 0.2˚ C). In addition, linearity and accuracy were verified on 30 healthy persons by using a mercury thermometer as a reference.²⁴ The mean differences between the 2 thermometer types was 0.1˚ C ± 0.05˚ C (range 0.0˚ C – 0.2˚ C); the Pearson correlation coefficient was 0.962, P = .000.

Disposable paper rulers were used to determine sacral skin area and the thermometer distance from sacral skin surface on patients. When the patient was in the lateral position, the sacral area was determined in line with the vertebral column, 5 cm above the coccyx.²⁴ The thermometer was held vertically 10 cm from the sacrum according to the manufacturer’s suggestion. At each measurement period, consecutive measurements were performed at 1-second intervals until the temperature difference between 2 measurements was 0˚ C, and the results of the measurements then were recorded.

The participants’ sacral skin temperatures were measured within the first 24 hours of admission (day 1) in the ICU and continued consecutively at 24-hour intervals between 5:00 pm and 7:30 pm until the patient developed a PU or died, was discharged, or was no longer having a position change.

Sacral skin temperatures were measured when patients were placed in the lateral position immediately after 120 minutes of being in the supine position (referred to as supine position sacral skin temperature measurement) and immediately following 30 minutes in the lateral positions (referred to as lateral position sacral skin temperature measurement) and recorded. When the sacral skin temperature was measured, the room temperature also was measured and recorded with the infrared thermometer in surface temperature mode. The measurements were taken by the researcher who positioned the patients.

PU protocols. Standard hospital protocol that stipulated patients should be repositioned every 2 hours; provided a support surface; protected from moisture, friction, and shear; and have a bath every day was applied to all participating patients. Patients using an alternating pressure air mattress or viscoelastic mattress were placed in the right lateral, supine, and left lateral positions every 2 hours by the researcher with the help of the support team. When a continuous lateral rotation alternating pressure air mattress was in use, the position change intervals were set to 1 hour because the device rotates at a 1-hour maximum.

Data analysis. Data were entered from paper-and-pencil questionnaires into the SPSS, version 22.0 (IBM Corp, Armonk, NY), for statistical analysis. Descriptive and inferential statistics were used to analyze the data. For the descriptive statistics, mean values and standard deviation, frequencies, and percentages were used. Descriptive characteristics, sacral temperature averages, PU development, and the difference between sacral skin temperatures in terms of gender were analyzed using the Student’s t test. The one-way analysis of variance test was used to assess the differences between sacral skin temperatures in terms of age, body mass index, and mattress type. Patients that developed PUs were followed-up for between 4 and 12 days; patients who did not develop a PU were followed-up until they died, were discharged, or no longer had their position changed for between 5 and 22 days. For most (66.5%), PUs developed within the first 5 days (day 1 to day 5); sacral skin temperatures of all the patients within the first and fifth days were used to examine the effect of sacral skin temperature on PU development. 

The relationship between body temperature and sacral skin temperatures on day 1 and day 5 was assessed using the Pearson product moments correlation and Spearman’s rank correlation coefficient (Spearman’s rho). Results were evaluated at 95% confidence interval; P <.05 was considered significant.

Ethical considerations. Before the study was conducted, approvals were obtained from the Scientific Ethics Committee of Ege University Nursing Faculty (dated 07.03.2014 and numbered 2014-29) and the hospital where the study was to be conducted. Patients or their relatives were given verbal and written information about the study; persons who agreed to participate in the study gave their written informed consent. The participants were told they could withdraw from the study at any time and that their credentials would be kept confidential and would not be shared with any third party. First letters of name and surname and the last 3 digits of the protocol number of each patient were combined to give a code to define and verify the patient. This ensured patient confidentiality.

Demographic and clinical characteristics. Of the 61 patients enrolled after meeting the initial study inclusion criteria, 9 could not undergo position change, 10 died, and 5 were discharged, leaving a sample of 37 patients (19 [51.4%] female), mean age 68.43 ± 11.71 years, mean follow-up time 6.13 ± 4.85 days. Patients were monitored for up to 22 days; 21 (56.8%) developed a PU. No statistically significant difference was found between the participants in terms of PU development and sociodemographic characteristics (all P >.05; see Table 1). All of the patients’ Braden scores were <12 through follow-up. 

Environmental conditions. Average room temperature was 23.23˚ C ± 1.41˚ C (range 20.50˚ C–25.80˚ C) on day 1 and 23.57˚ C ± 0.94˚ C (range 22.30˚ C– 26.00˚ C) on day 5. 

PU stages. Of the 21 patients who developed PUs, 17 (81%) were stage 1 and 4 (19%) were stage 2. PUs occurred in the sacral area in 14 patients (66.5%) within the first 5 days following admission (see Table 2). 

Sacral skin temperature. No statistically significant difference was found in terms of sacral skin temperatures measured in the supine position on day 1 or in the supine and lateral positions on day 5 between patients who did or did not develop a PU (P >.05) (see Table 3 and Figure).

Mean patient body temperature on day 1 was 36.66˚ C ± 0.50˚ C and 36.76˚ C ± 0.64˚ C on day 5. A positive statistically significant correlation was noted between body temperature and sacral skin temperature in the supine position on day 1 and day 5 (37.01˚ C ± 0.42˚ C, r = 0.459, P = .004; and 37.33˚ C ± 0.68˚ C, r = 0.495, P = 0.007, respectively; and in the lateral positions (37.25 ± 0.52°C, r = 0.591, P = 0.000, and 37.37 ± 0.63°C, P = .381, P = .045, respectively) (see Table 4). 

Mean patient sacral skin temperature according to the patients’ age group was 37.35˚ C ± 0.63˚ C for persons 38 to 64 years of age, 37.38˚ C ± 0.70˚ C for persons 65 to 74 years of age, and 36.93˚ C ± 0.39˚ C for persons 75 to 90 years of age. A statistically significant difference was noted between average sacral skin temperature and advanced age (F = 13.221; P = .000).

According to mattress type, the mean patient sacral skin temperature was 37.10˚ C ± 0.60˚ C for alternating pressure air mattresses, 37.28˚ C ± 0.55˚ C for continuous lateral rotation alternating pressure air mattresses, and 37.85˚ C ± 0.54˚ C for viscoelastic mattresses. A statistically significant difference was noted between average sacral skin temperature and viscoelastic mattress use (F = 14.039, P = .000) (see Table 5).

PUs occurred in the sacral area in the majority of patients within the first 5 days following admission to the ICU. The prospective, descriptive, cohort study by Sae-Sia et al²⁴ found sacral skin temperature could rise during the 24- to 96-hour period before a PU developed. However, in the current study, no statistically significant difference was seen between patients with and without PUs in terms of sacral skin temperatures measured on day 1 and day 5. On the other hand, the difference between the mean temperatures measured in the supine position on day 1 and on day 5 in patients who developed PUs was 3 times greater than in the patients who did not develop PU, and the difference between the sacral skin temperature measured in the lateral and supine position on day 5 was 6 times higher. Although this increase was not statistically significant, it may be clinically significant. Convective heat accumulation between the skin and the support surface may have caused this increase in temperature. Another explanation is that local tissue hypoxia, ischemia, tissue inflammation, and skin temperature may increase because blood flow to the tissues may be impeded after the tissue between the skin and the support surface is exposed to pressure for a long time, according to descriptive, descriptive comparative, retrospective and correlational, and/or quasiexperimental studies.^22,24,29-31 

The results of several studies investigating the relationship between PU and sacral skin temperature are contradictory. A descriptive study by Andersen and Karlsmark¹⁹ that examined sacral skin temperature using 4 noninvasive methods among 11 participants found changes in actual skin temperatures were not associated with PU development. A descriptive study by Knox²² conducted among 26 nursing home residents found sacral skin surface temperature measured using a disposable thermistor temperature probe over the sacrum and trochanters was not significantly related to PU development. On the other hand, a prospective, observational study conducted by Yoshimura et al²⁰ among 29 patients undergoing elective surgery at a general hospital in Japan showed change in skin temperature was significantly related to the subsequent development of a PU. An observational study by Rapp et al³² among 20 nursing facility residents who wore skin temperature monitors for 5 days found a significant difference between the residents who did and did not develop PUs. Conflicting results could be due to the differences in sample size, methodology, and temperature measurement device used in the studies.

Body temperature. In the current study, comparing patient body temperature and sacral skin temperature measured in the supine and lateral positions revealed a weak, positive, significant relationship between body temperature and sacral skin temperature (see Table 4). Research has shown the body’s core temperature is generally transferred to the periphery by blood circulation and creates the skin temperature measured from the skin surface. Skin temperature is affected not only by the core temperature, but also by environmental factors such as clothing thickness, air temperature, the structure of chair or mattress surfaces used, and physiological factors such as the amount of subcutaneous adipose tissue.^22,33 However, only one study²² investigated the relationship between sacral skin temperature and body temperature. In contrast to current findings, the descriptive study by Knox²² reported a weak, negative, but statistically significant relationship between body temperature and sacral skin temperature. Therefore, future studies of PU development should include body temperature measurements in addition to sacral skin temperature. 

Age. In the current study, mean sacral skin temperature in patients in the 75 to 90 year age group was lower than in the other 2 age groups (P <.001), and the mean sacral skin temperature in the patients in the 35 to 64 year group was similar to patients in the 65 to 74 year age group (P >.05) (see Table 5). According to an experimental study by Bergstrand et al,³⁴ skin elasticity over muscle and bone tissue decreases with age, and tissue resistance to pressure decreases due to the inadequate sweat gland function, which suggests that advanced age may be an important factor in PU development. Bergstrand et al³⁴ assigned 115 participants to 3 groups (a healthy group <65 years of age, a healthy group >65 years of age, and an unhealthy group >65 years of age) and found no significant difference in sacral skin temperatures between healthy and unhealthy volunteers age ≥65 years, but a significant difference was noted between healthy volunteers age ≤65 years and the other groups. In a quasiexperimental study conducted by Baldwin²¹ among 38 hospitalized patients, sacral skin temperatures were lower in patients >60 than in patients <60 years of age. On the other hand, a descriptive study by Howell³⁵ that examined sacral skin temperatures according to age and sites where PUs were likely to develop concluded sacral skin temperature did not correlate significantly with age. The results of the current study are consistent with Bergstrand et al,³⁴ Baldwin,²¹ and Howell³⁵ who found that when patient age increases, the sacral skin temperature decreases, but more studies should be conducted to investigate the relationship between age and sacral skin temperatures.

Mattresses. In the current study, mean sacral skin temperature of patients using viscoelastic mattresses was higher than in patients provided 2 other types of mattresses (P <.01) (see Table 5). This can be explained by the fact that heat that accumulates between the surface of the mattress and the skin cannot be transferred by convection because the surface area of the skin in the viscoelastic mattress covers a large area and the viscoelastic mattress has a dense and porous foam structure. A literature review by Lachenbruch³⁶ similarly found the heat absorption capacities of viscoelastic mattresses were higher but heat transport capabilities were lower than other types of mattresses. A quantitative analysis by Rothenberger et al,³⁷ conducted among 25 healthy volunteers, investigated the effects of standard sponge mattresses, air mattresses, and viscoelastic mattresses on blood flow in tissues exposed to pressure; viscoelastic mattresses were found to reduce blood flow compared with air mattresses.³⁷ The current study findings are consistent with the results from Rothenberger et al³⁷ but may contradict the results of descriptive and quasiexperimental studies indicating that viscoelastic mattresses significantly reduce PU incidence.^38,39 In the current study, only a few (3) patients used the viscoelastic mattress, which may have affected the results. Thus, to obtain results that can be generalized to larger populations, future studies with larger samples and different mattress types should be performed. 

This study has several limitations. First, the sample was drawn from one university hospital, which limits generalization. Also, the small sample size, a weakness of the study, may have affected the results. 

The aim of the study was to examine the relationship between sacral skin temperature and PU development among 37 patients. The authors found no statistically significant difference in supine position sacral skin temperature on day 5 between patients who did and did not develop a PU (37.37˚ C ± 0.53˚ C and 37.30˚ C ± 0.79˚ C, respectively; t = 0.259, P = .798 day 5). Day 5 lateral position skin temperatures also did not differ between the 2 groups (37.44˚ C ± 0.44˚ C and 37.31˚ C ± 0.75˚ C, respectively; t = 1.306, P = .621). Thus, study data found no statistically significant relationship between sacral skin temperature and PU development and as such, assessing skin surface temperature measurement was not considered a viable tool for predicting the development of PU in high-risk intensive care patients. Future studies with larger samples and different mattress types are needed. 

Ms. Yilmaz is a research assistant, Dokuz Eylül University, Nursing Faculty, Fundamentals of Nursing Department, Inciralti, İzmir, Turkey. Dr. Günes is a professor, Ege University, Nursing Faculty, Fundamentals of Nursing Department, İzmir, Turkey.

Please address correspondence to: İlkin Yilmaz, PhD(c), RN, Dokuz Eylül University, Nursing Faculty, Fundamentals of Nursing Department, 35340, Inciralti, İzmir, Turkey; email: ilkinyilmaz85@gmail.com.