Use of 2 Types of Air-cell Mattresses for Pressure Ulcer Prevention and Comfort Among Patients With Advanced-stage Cancer Receiving Palliative Care: An Interventional Study

Patients with advanced- or terminal-stage cancer and persons receiving palliative care are at high risk for pressure ulcers (PUs). Purpose: The purpose of this study was to examine the rate of PU development and levels of comfort of a dual-fit, air-cell mattress compared with an alternating, 2-layer overlay air-cell mattress in patients with advanced- or terminal-stage cancer receiving palliative care. Methods: From January 2011 to December 2013, hospitalized patients with advanced- or terminal-stage cancer who were referred to a palliative care team, at least 20 years of age, able to communicate, experiencing pain, and did not have a PU were recruited to participate. Patients who consented were alternately placed on the intervention (dual-fit, air-cell) or control (2-layer air) mattress until hospital discharge or death. Demographic and clinical data, pain scores, performance status, Palliative Performance Scale scores, Braden Scale scores, tissue interface pressure, and comfort were assessed via interview using closed-end questions. If a PU developed, clinical characteristics were assessed using DESIGN-R. Descriptive statistics and the Mann-Whitney U, chi-squared, and Fisher’s exact tests were used to analyze the data. Results: Of the 123 eligible patients, 73 were randomized and 52 completed the study (23 intervention patients, median age 63 [range 27–80] years; and 29 control group patients, median age 61.0 [range 27–82] years). Mattresses were used a median of 17 (range 4–113) days in the intervention group and a median of 32 (range 3–270) days in the control group. The incidence of PUs did not significantly differ between the 2 groups (13% in the intervention and 17.2% in the control group). Interface pressures were significantly higher in the intervention group (27.0 mm Hg vs. 24.3 mm Hg). Comfort scores at rest were significantly better in the intervention than in the control group (sinking into bed [3 vs. 14, respectively]; slipping on bed [o vs. 16, respectively]; and feel pressure of air cell [2 vs. 14, respectively]), as were scores with movement (instability during movement [4 vs. 18, respectively] and feeling of floating of the buttocks [6 vs. 21, respectively]) (P <.05). Conclusion: Dual-fit, air-cell mattresses may help prevent PUs and improve comfort at rest and during activity among patients with end-stage cancer receiving palliative care. Further research regarding mattress selection protocols for this patient population is warranted.

Pressure ulcer (PU) prevention is important for patients with advanced-stage cancer receiving palliative care. A case study by Schank¹ reported 2 patients died 6 weeks and 5 months, respectively, after developing Kennedy Terminal Ulcers (a type of PU), and a prevalence study² noted 55.7% of people who developed PUs died within 6 weeks.² An incidence study³ reported 62.5% of patients receiving hospice care died within 2 weeks of developing PUs. In these studies, PUs developed from 2 weeks to several months before death, inferring that the time for PU prevention care in persons with terminal-stage disease varies widely.

Multiple risk factors influence the development of PUs in persons with terminal-stage disease, including the appearance of symptoms such as pain, dyspnea, and general malaise. When body movement is painful and repositioning is difficult, patients become immobile and are at increased risk for PUs from external factors such as pressure, friction, and shearing and from internal factors such as skin fragility from malnutrition and edema.⁴

Studies have shown a 10% prevalence of PUs (99 of 980) in hospice patients⁵ and a 54.7% prevalence (64 of 117) in terminal-stage nursing home residents.⁶ In other research, the prevalence of PUs was 19% in a palliative care ward,⁷ 22.9% among cancer patients receiving palliative care,⁸ and 18.8% in advanced-stage cancer patients receiving home palliative care.⁹

According to data released at a PU summit,¹⁰ PUs are difficult to prevent at the terminal stage, and as such, according to a survey¹¹ of wound, ostomy, and continence (WOC) nurses, some PUs may be considered unavoidable.

In an investigative study¹² of 237 patients with terminal-stage cancer, among whom the primary diagnosis was malignant neoplasm, 47.0% developed PUs, an incidence of 17.0%. This is higher than the estimated rate of 1.6% in the general hospital population and 1.16% according to a national survey in Japan,¹³ as well as the 1.6% rate reported in bedbound older adults.¹⁴ The fact that it is difficult to change body position effectively because of the emergence of symptoms or other problems in advanced- and terminal-stage cancer patients increases the probability that PUs will develop.¹⁵

Regardless of disease or condition, PU prevention is important. A Cochrane review¹⁶ found that eliminating or reducing pressure from external forces by changing body position or using a support surface is a necessary part of preventive care¹⁶; this complies with Japanese PU prevention guidelines,¹³ which also recommend the use of assessment tools and support from nurses certified in PU care.¹⁷ Repositioning and small shifts in body position also have been investigated in a PU prevention intervention study.¹⁸ In a controlled trial¹⁹ conducted among 82 elderly Japanese patients, a low percentage (3.4%) of patients using an alternating, double-layer, air-cell overlay developed PUs, suggesting that air mattresses may help redistribute pressure and, as such, could play a role in PU prevention. In the terminal stage of cancer, perpetually changing symptoms actively affect mattress comfort, but guidelines²⁰ recommend air mattresses for the prevention of PUs in patients who have difficulty changing position on their own, such as persons who are terminally ill.

Air mattress comfort, another consideration in support surface selection, is a factor of resilience, ventilation, and ease of movement when rolling over. A quasi-experimental, interrupted, time-series design study²¹ in which changes in heart rate were an indication of comfort in bedbound older patients using air mattresses showed comfort was not compromised by mattress function. A randomized controlled trial²² comparing 2 widely used alternating air pressure mattresses showed one outperformed the other in terms of comfort. A randomized evaluation²³ of alternating pressure overlays versus alternating pressure replacement mattresses showed little difference in comfort.

The increased instability of the body due to “sinking” when air mattresses decompress can affect comfort. A descriptive study²⁴ (N = 14 patients) suggested using foam rather than air mattresses (as recommended in PU prevention guidelines for terminal-stage cancer patients) because air mattresses allow patients to sink into the bed, after which, PUs can develop. An experimental study²⁵ reported mattresses that prevent sinking of the buttocks have been developed, but no assessments of their use in patients with terminal-stage cancer have been conducted. Because comfort preferences based on immersion and envelopment affect air mattress use,¹⁷ an air-dispersion mattress that is suited to an individual’s preferences and the care environment should be selected.²⁰

The aim of care in the terminal stage of disease is to maintain or improve the patient’s quality of life (QOL). The development of pressure-redistribution mattresses should include consideration of how patients spend their time during rest and activity, in addition to PU prevention. Air mattresses for terminal-stage patients should resolve problems related to PU development and comfort preferences; however, air mattresses that provide both PU prevention and comfort for terminal-stage patients are not commonly used in Japan. 

Alternating dual-fit, air-cell mattresses are used to prevent the development of PUs after cardiovascular surgery. These mattresses have an accompanying pump that adjusts air-cell firmness, and they have been shown in a randomized controlled study²⁶ to provide support and stability based on the patient’s actions and to improve stability in bed movements by adapting to the body. However, their use in patients with terminal-stage advanced cancer has not been evaluated. Therefore, the objective of the present study was to examine PU development and levels of comfort of a dual-fit, air-cell mattress compared with an alternating, 2-layer overlay, air-cell mattress in patients with advanced-stage cancer.

Study design. This intervention study was conducted from January 2011 to December 2013. Participants were followed from the time they began using an air mattress until hospital discharge or death.

Setting and participants. This study was conducted in an 832-bed university hospital in Japan. At this hospital, PU preventive care is provided in accordance with the guidelines of the Japanese Society of Pressure Ulcers by a PU control team that includes nurses certified in WOC care.¹⁷ All terminal-stage cancer patients referred to a palliative care team for pain were invited to participate and were considered eligible for participation if they were at least 20 years old, had a Japan Coma Scale (JCS)²⁷ score of I, and a Glasgow Coma Scale (GCS)²⁸ score >15. Patients who refused to provide consent or had an existing PU were excluded. All procedures, from participant eligibility and selection to completion of randomization, were performed by the researchers.

Randomization and blinding. The participants were randomized to the intervention group or the control group by one of the researchers using an alternate randomization schedule.

Blinding was difficult because participants could identify the type of air mattress by removing the bedsheet, and nurses could identify the type when setting up the air mattress. Therefore, after consent to participate in the study was obtained, patients were given an explanation of the type of air mattress they would be using. Once enrolled, participants were given a unique identification number to ensure anonymity.

Intervention. The air mattress used in the intervention group was an alternating, exchange-type, dual-fit, air-cell mattress (NEXUS; Cape Co, Ltd, Kanagawa, Japan). The bottom part has a single layer base mat. The structure of the upper part consists of 24 air cells in a bottom supporting layer and an upper layer with 48 separated air cells. The accompanying pump has 2 pressure settings to adjust firmness that were controlled by a nurse. The amount of firmness is related to patient stability and movement in bed, movement while getting into and/or out of bed, and sitting on the bed; the firmness of the air cells is adjusted according to the user’s activity. The height of the mattress is 13 cm, including the base mat and upper layer of air cells.

The air mattress for the control group was an alternating, 2-layer overlay, air-cell mattress (TRICELL; Cape Co, Ltd Kanagawa, Japan), which has a standard mattress base. The upper part of the mattress consists of 24 integrated air cells in the top and bottom layers. This mattress has 1 setting for the accompanying pump to raise the internal pressure for firmness and stability. The height of the cells in the upper part of the mattress is 10 cm.

Procedure. Nurses at the study facility were provided basic knowledge and skills for PU prevention. Nurses on the study ward also received an explanation regarding mattress pressure adjustment using the accompanying pump before the start of the study, and the mattress manual was available at bedside. The researchers checked the mattress pressure adjustment once a day.

At the start of the study, participants were switched from the hospital’s usual thin foam mattress to the assigned air mattress. The air mattress pressure setting was based on the participant’s most recent weight and was adjusted as necessary to achieve an interface pressure in the sacral area of ≤40 mm Hg per manufacturer specifications and the standard value for mattress assessment for PU prevention in Japan.²⁹

Data collection. Participant characteristics were collected at the start of the study from patient records and included age, gender, cancer diagnosis, treatment history (eg, chemotherapy and radiotherapy), presence of pain (Yes/No), pain numeric rating scale (NRS) rating (0 = no pain, 10 = the most intense pain³⁰), performance status (PS) rating (0 = fully active; 4 = cannot carry out any self-care such as standing/sitting, bathing, and toileting; 5 = dead³¹), Palliative Performance Scale³² (PPS; a validated and reliable tool used to assess a patient’s progressive decline in functional performance), activity score rating (100% = normal activity with no evidence of disease, 0% = death), and body mass index (BMI). In addition, interface pressure²⁹ in the sacral region as measured by the researcher using a multipad pressure evaluator,²⁹ Braden Scale score³³ (0 = low risk, 23 = high risk), and K scale³⁴ rating (an assessment used in Japan; 0 = low risk, 6 = high risk) for PU risk were assessed and recorded. Data on the duration of air mattress use also were collected. Patients were assessed once a week by the researchers. The nurses and researchers determined whether PUs had developed. PU variables were assessed using DESIGN-R³⁵ rating, where d0 = no particular skin lesion and no redness; d1 = persistent redness; d2 = lesion extends into dermis; D3 = lesion extends into the subcutaneous tissue; D4 = lesion extends to muscle, tendon, and bone; D5 = lesion extends into the articular or body cavity; and DU = it is impossible to measure the depth for the PU severity assessment.

For patients who developed a PU, information obtained included the Braden Scale score; presence of urinary incontinence, fecal incontinence, or a urinary catheter; interface pressure measured at the time of entry to the study, ulcer depth, and symptoms (eg, pain and/or redness, the latter assessed by the researchers and a WOC nurse).

Mattress comfort was assessed subjectively after the air mattress had been used. Pain was assessed every 24 hours as previously described. Researchers also interviewed the study participants about comfort using Yes/No questions from previous studies^23,26 that addressed mattress noise, sinking into the bed, slipping on the bed, being affected by motion, sticky feeling, temperature, and feeling pressure from the air cells. Comfort related to activity was assessed by asking patients direct Yes/No questions about their experiences getting out of bed, getting into a sitting position, and walking to the bed and whether movement induced pain, unstable posture when moving, or a feeling of floating in the buttocks.

Outcomes. The primary outcome was development of a PU. The secondary outcome was air mattress comfort.

Data analysis. Survey data and responses to descriptive questions were entered into a spreadsheet. Scores on the Palliative Prognostic Index (PPI)³⁶ were calculated based on PPS scores while noting the presence of edema, dyspnea at rest, and delirium. Participant characteristic numerical data were compared using the Mann-Whitney U test, and categorical data were compared using the chi-squared test and Fisher’s exact test. PU development data were compared using the chi-squared test and Fisher’s exact test, and ulcer depth was compared qualitatively using DESIGN-R.³⁵ Comfort at rest and while moving was evaluated by comparing categorical data using the chi-squared test and Fisher’s exact test. Statistical analysis was conducted using JMP 11.0 software (SAS Institute, Tokyo, Japan). P <.05 indicated a statistically significant difference.

Ethical considerations. This study was approved by the ethical review board of the Kanazawa University Ethics Committee (No. 883). Any participant who developed a PU was promptly reported to his/her physician and certified WOC nurse, and appropriate measures were taken.

Inclusion and follow-up. Of the 123 patients screened, 73 were enrolled in the study; 50 were excluded because they had an existing PU (3), declined to participate (3), were mentally or physically unstable (12), or had decreased consciousness level (11), emergency surgery (3), or worsening symptoms of disease before allocation (18). The 73 participants enrolled in the study were randomized to either the intervention group (36) or the control group (37). Thirteen (13) participants in the intervention group and 8 in the control group were excluded because of ineligibility after follow-up. The sample size was set to have a power of 0.5, in accordance with a previous study.²⁶

At the completion of the study, 23 participants in the intervention group and 29 in the control group were eligible for analysis (see Figure). No significant differences were noted in gender, age, PS, or PPS scores (P <.05) (see Table 1). In the intervention and control groups, the interface pressures were 28.4 mm Hg and 26.0 mm Hg, respectively (P = .01), and the pain scores were 5 and 4, respectively (P = .01). Duration of air mattress use was 17 days and 32 days, respectively, P = .07).

PU development. PUs developed in 3 intervention and 5 control group patients for incidence rates of 13.0% and 17.2%, respectively (P = .49); dermal skin loss (d2) was noted in both groups (see Table 2). Shortness of breath was found to be a risk factor for PUs in both groups. Urinary and fecal incontinence occurred in the control group, but not in the intervention group (see Table 3). The Braden Scale total scores of the patients who developed PUs were 16 and 12, respectively.

Assessment of comfort.
Comfort at rest. Fewer patients in the intervention than in the control group described sinking into the bed (3 vs. 14, respectively), feeling of pressure from the air cells (2 vs. 14, respectively), and a feeling of stickiness (0 vs. 4, respectively) (P <.05) (see Table 4).

Comfort during movement or activity. Fewer persons in the intervention group expressed discomfort compared with the control group in regard to “instability when moving in bed” (4 vs. 18, respectively) and a “feeling of floating of the buttocks” (6 vs. 21, respectively) (P <.05) (see Table 5).

Adverse events. No patients in either group experienced an overall floating feeling or poor physical condition. No pain was related to the mattresses, and no equipment malfunction or technical problems in the manipulation of the air mattresses occurred. Patients in both groups experienced feelings of pressure on the back from the air cells (2 in the intervention group, 14 in the control group) (see Table 4). Back pain occurred in 1 patient in the control group when a protuberance on the patient’s back contacted the inflating part of an air cell but was immediately improved with a change in body position. 

The risk for PU development increases in terminal-stage cancer patients receiving palliative care.³⁷ According to reviews of the literature,^38,39 PU prevention in palliative care needs to consider both clinical and comfort preferences — that is, PU prevention for patients receiving palliative care needs to be provided with consideration for QOL. Thus, patient QOL needs to be a factor in selecting a pressure-redistribution mattress.

Air mattresses are recommended for elderly people who are bedridden.¹⁴ However, although air mattresses intended to diffuse pressure may be highly effective in preventing PUs, they have low comfort assessments.²³ At the end of life, when patient comfort preferences are given priority, it may be difficult to continue using air mattresses, but discontinuing use may lead to PU development, even though not all end-of-life patients are completely bedridden. Pain in the terminal stage affects the patient’s level of activity; therefore, the comfort of a pressure-redistribution mattress may affect the patient’s preferences and needs, adding to the challenge of selecting an appropriate pressure-redistribution mattress.

The authors of this study chose to focus on the effects of air mattresses on PU prevention and comfort in terminal-stage cancer patients with respect to QOL. The goal for pressure-redistribution mattress use in the terminal stage is to find a mattress that can respond to changes in patient independence/mobility. To the authors’ knowledge, this is the first study to compare air mattresses in terms of PU prevention and comfort in terminal-stage patients receiving palliative care.

PU development. No significant differences were found in PU development between the 2 groups. Participants with shortness of breath, a symptom of exertion, tended to develop PUs, regardless of group; as such, shortness of breath was regarded as a risk factor for PU in this study.⁴⁰ Interface pressures were significantly higher in the intervention than in the control group at entry to the study. The cutoff value for interface pressure in the sacral area using the multipad was <40 mm Hg.²⁹ PU risk was low when positional changes occurred at intervals of 2 hours for the 2-layer air mattress.⁴¹ Therefore, achieving an interface pressure in the sacral area of ≤40 mm Hg was not a problem.

Incontinence. Among patients who developed PUs in the intervention group, urinary incontinence from indwelling urinary catheters was not a factor; however, the control group included patients with fecal and urinary incontinence. Results of a previous randomized controlled trial⁴² showed PU development is affected by the skin’s moisture environment and incontinence-associated dermatitis (IAD). Although differences in the moisture environment may have affected PU development in the present study, skin was assessed by a WOC nurse, and IAD and PUs were investigated and distinguished from one another to provide appropriate evaluation and treatment.

PU depth. PU depth was d2 (ie, nearly all PUs were shallow). Mattress/patient interface pressure was <40 mm Hg, which is thought to be effective in preventing PUs. Thus, the PU prevention effect was equal in both groups and evaluation of prevention was equal among advanced-stage and critical care cancer patients.²⁶

Length of use. The air mattresses in both groups were used over long periods (ie, more than 2 weeks). In a randomized controlled study⁴² that included patients with malignant tumors, air mattresses were used for 14 days. A retrospective study⁴³ showed PU development among palliative care patients extended the period of hospitalization. Thus, a long period of PU prevention should be expected when treating patients in the terminal stage.

Air mattress comfort. Patients in the intervention group experienced less discomfort in bed than the control group. The improvement in air mattress comfort in the intervention group was similar to the results in perioperative patients.²⁶ One factor for improved comfort was believed to be the fit of the small air cells in the intervention group mattress. Advanced cancer patients are individually positioned based on factors such as pain or dyspnea in a way that maintains a constant bed adjustment height and lateral recumbent angle.

During movement or activity, the intervention group experienced less discomfort than did the control group with regard to a feeling of floating of the buttocks and reported less instability when sitting. The participants in the present study had a median PPS of 40%. In a retrospective study of 77 hospice residents,⁴⁴ the PPS level was 10% to 70%, and 56% of all patients had a PPS of 40% or less. When the PPS is 40%, activity may increase the risk of PU development because of external forces and friction. PPS can assess activity in more detail and has been shown in a prospective cohort pilot study⁴⁵ to correlate with the Braden Scale in terms of PU risk.⁴⁶

The study had several limitations, including a small sample size and the limited external validity of the results. First, blinding of the air mattresses could not be achieved, and other variables that may have affected the study outcome (eg, nutritional status) were not assessed.

Second, the results of this study were limited to advanced-stage cancer patients who reported having pain. A prospective sequential case series⁷found pain in the terminal stage is a predictor of the development of PUs. Moreover, it is possible the effects of pressure from alternating inflation and deflation of the air mattress may elicit a stronger pain response in end-of-life patients.

Third, patients had difficulty with the air mattress controls. With no automatic control, nurse assistance was needed to adjust the pressure. The median PPS among the participants was 40%, and a nurse’s assistance was needed to operate the bed. Although these findings can be adapted to patients in a similar physical condition, terminal-stage patients have fluctuating physical conditions and activity levels, so the present results may have differed depending on the patients’ activity levels and degree of assistance required.

Fourth, patients in both groups experienced a feeling of pressure on the back, affecting comfort. Air mattress inflation and deflation provide pressure reduction and decompression, but in patients with cancer-related back pain, the tumor and other sites of inflammation would be physically stimulated when the air cells are inflated, which may increase pain. This may have caused back pain in both the intervention and control groups. The decompression and pressure reduction functions of the dual-fit, air-cell mattress may have created untoward effects. Consequently, it will be necessary to investigate air cell inflation and deflation and whether pain was induced where the body was in contact with the mattress surface. Methods to minimize pain and assessments of the PU prevention effect also will be necessary.

Fifth, only the control group included patients with incontinence. No IAD assessment tool was used in this study. Had the development of PUs and the identification of IAD been clarified, it is possible that the PU incidence rate would have been different.

Lastly, the human resources required to adjust body position at night were limited in the present study, possibly allowing for inconsistency in appropriate decompression and positioning.

This study compared the PU development rate and comfort of a dual-fit, air-cell mattress and a 2-layer, air-cell overlay mattress in patients with terminal-, advanced-staged cancer receiving palliative care. No statistically significant differences in PU incidence were observed between the 2 groups, but comfort at rest and with movement was significantly better in the intervention than in the control group (P <.05). These findings may provide basic data for the exploration of choices and conditions for PU prevention support surfaces in end-of-life patients. Additional prospective studies, including comparing the use of these types of mattresses with other pressure-redistribution mattresses in patients with terminal stage cancer, are needed. 

The authors sincerely thank everyone who participated in this study.

This trial was registered at UMIN Clinical Trials (UMIN000035568).

Ms. Marutani is a registered nurse, Department of Clinical Nursing, Graduate Course of Nursing Science, Division of Health Sciences, Graduate School of Medical Sciences, Kanazawa University, and Kanazawa University Hospital, Ishikawa, Japan. Dr. Okuwa is a registered nurse, Faculty of Health Sciences, Institute of Medical, Pharmaceutical and Health Sciences; and Dr. Sugama is a registered nurse, Institute for Frontier Science Initiative, Kanazawa University. 

Please address correspondence to: Mayumi Okuwa, RN, PhD, Faculty of Health Sciences, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University, 5-11-80, Kodatsuno, Kanazawa-shi, Ishikawa 920-0942 Japan; email: ookuwa@mhs.mp.kanazawa-u.ac.jp.