Background

Pressure ulcers (PUs) may develop when persistent pressure on bony prominences obstructs a healthy capillary flow, leading to tissue necrosis [1]. The primary cause of PUs is ischemic changes induced by pressure-related external forces, including shear force [2], and it is a common complication of immobility among the elderly [3]. Moreover, the elderly are at a high risk of developing PU because of immobility [4, 5], poor nutritional status [6, 7], and decreased body weight [5]. Despite recent research developments in the process of wound healing, treatment strategies for pressure ulcers remain inconclusive [8]. PU is a major contributor to an impaired quality of life by increasing mortality and the length of stay, burdening healthcare systems [9, 10].

An important factor in treatment resistance is the undermining formation, commonly observed in deep PUs [11]. Treatment-resistant PUs may be associated with changes in physical wound properties because of aging, such as mobility and deformity [12], and the high incidence of the undermining formation in PUs over the sacrum, coccyx, and greater trochanter [13]. Evaluation of wound deformity and mobility has been shown to improve healing following topical ointment therapy [14]; moreover, an active control of topical ointment application has been demonstrated to be important for moist wound healing in PUs [15].

In Japan, PUs have been treated by a team comprising physicians, pharmacists, and nurses, based on physical properties of the wound and using topical ointment for moist wound healing. Thus, this concept has been shown to be therapeutically effective at an improved medical cost [16]. Further, it has been spreading to pharmacists in Japan. However, there have been no detailed descriptions of interventions by pharmacists for PU treatment. Based on this background, we newly proposed the concept of topical ointment therapy by pharmacists, called “Furuta method” (Appendix), for PU, considering the physical properties and moist environment of wounds, and we promote widespread training sessions according to Furuta method in Japan. The aim of this study was to investigate the effectiveness of PU therapeutic effects treated by trained pharmacist according to “Furuta method” in Japan and to compare the results based on the compliance surveys of this “Furuta method.”

Methods

Setting

The study was based on compliance surveys and retrospective data collection from 37 hospitals and five dispensing pharmacies in Japan. Hospitals and dispensing pharmacies were appealed to participate via the “decunet” mailing list used by pharmacists involved in PU. Pharmacists who participated in this study underwent training sessions according to “Furuta method” in Japan. We set the suitable standards for the PU team, which comprised physicians, pharmacists, nurses, and other medical staff. Data of the National Center for Geriatrics and Gerontology (NCGG) were excluded because Furuta was involved in PU treatment. Finally, 35 hospitals and four pharmacies were included. All researchers collected data in a severely managed the personal information. The study protocol was approved by the ethics committee of NCGG.

Compliance survey

For compliance confirmation of pharmacist intervention guidelines on PU topical ointment therapy called “Furuta methods” (see Appendix), a self-administered 10-item questionnaire including the questions was provided: 1. are you treating PU more than once a week?; 2. do you assess the wound using descriptive ulcer dermatology?; 3. do you assess ointment base properties?; 4. do you assess wound surface moisture and use the Expert Furuta Blend ointment?; 5. do you assess the physical properties of the wound?; 6. do you treat the wound using wound fixation?; 7. do you assess the wound surface and edge?; 8. do you control the moisture of the wound by assessing granulation tissue deformation by an external force?; 9. do you assess external forces applied to the wound and treat by wound fixation?; and 10. do you assess residual tissue, such as the dermis, subcutaneous tissue, and fascia? This questionnaire was completed by responding yes (1 point) or no (0 point). This questionnaire was a single-blind method, which allocated patients to the compliance (≥8 points) and non-compliance (<8 points) groups. This was followed by a retrospective data collection.

Patients

A total of 888 consecutive patients who underwent PU treatment in Japan between August 2010 and July 2014 were enrolled in this study (Fig. 1). Only patients who met the following criteria were included: (a) patients whose diagnosis were assigned as DESIGN-R [17] ≥ d2; (b) patients who received intervention from the PU team for ≥7 days during the observational period. Patient with missing information on the following data were excluded from this study: age, sex, hemoglobin, serum albumin, site of PU, DESIGN-R score, and observation period. Thus, a total of 868 patients were included; the mean patient age was 80.0 ± 11.3 years (standard deviation, Table 1).

Fig. 1
figure 1

Flow chart for study group inclusion. Patients were grouped according to the DESIGN-R category of depth and were divided into two groups based on compliance survey results. Finally, patients were allocated by propensity score matching

Full size image
Table 1 Baseline characteristics of patients with pressure ulcers based on the DESIGN-R category of depth
Full size table

Propensity score matching

To reduce treatment-selection bias and potential confounding variables, we performed rigorous adjustment for marked differences in baseline patient characteristics with propensity score (PS) matching using the following algorithm: 1:1 optimal match with a ±0.03 caliper and no replacement. Possible confounders were selected based on clinical knowledge for their potential association with the outcome of interest. The PS model was estimated using a logistic regression model that adjusted for patient characteristics, such as age, sex, hemoglobin, albumin, DESIGN-R score at baseline, and observation period. To measure covariate balance, an absolute standardized difference above 10 % represented meaningful imbalance [18, 19].

Study variables and statistical analysis

The main outcome was a comparison of healing days, which was calculated based on a previous study [20] that used the following equation:

$$ \mathrm{Treatment}\ \mathrm{period} = \mathrm{DESIGN}\hbox{-} \mathrm{R}\ \mathrm{score}\ \mathrm{at}\ \mathrm{baseline}/\mathrm{Healing}\ \mathrm{rate} $$
$$ \mathrm{Healing}\ \mathrm{rate} = \left(\mathrm{DESIGN}\hbox{-} \mathrm{R}\ \mathrm{score}\ \mathrm{at}\ \mathrm{baseline}\right)\ \hbox{-}\ \left(\mathrm{DESIGN}\hbox{-} \mathrm{R}\ \mathrm{score}\ \mathrm{at}\ \mathrm{endpoint}\right)/\ \left(\mathrm{Intervened}\ \mathrm{period}\right) $$

DESIGN-R was used for monitoring of PU healing progression and is defined by the following components: depth, exudate, size, inflammation/infection, granulation tissue, necrotic tissue, and pocket [17]. The secondary endpoint was rates of unhealed wound.

Data were presented as mean ± standard deviation. Paired comparisons were performed using paired t-test for categorical variables and Mann–Whitney U test for continuous variables. Kaplan–Meier plots were created for evaluating the study endpoints, and the respective 95 % CI values were calculated. For the Kaplan–Meier plot for PU progression, all patients were included except for healed cases. Moreover, two-sided log rank tests were used for between-group comparisons of PU progression. All analyses were performed using SPSS version 22.0 software (SPSS Inc, Chicago, IL, USA). A two-tailed P value of <0.05 was considered statistically significant.

Results

The mean compliance survey point was 7.3 ± 2.8, and results were collected from 19 (437 patients) and 20 (451 patients) facilities. In the questions 2, the compliance rate based on descriptive ulcer dermatology was the lowest.

Based on the DESIGN-R criteria, 343 patients were diagnosed as d2, 220 as D3, 159 as D4, 5, and 146 patients as DU (Table 1). Individuals who did and did not undergo treatment based on “Furuta method” differed for all measured characteristics.

Table 2 shows the detailed characteristics of patients included in the final analysis: d2 (n = 202); D3 (n = 130); D4, 5 (n = 76); and DU (n = 76). The covariate balance in the matched cohort considerably improved. In Fig. 2, for each DESIGN-R group of patients, the respective duration of healing was significantly shorter in the compliance group than in the non-compliance group: d2 (23.6 ± 36.8 vs. 32.2 ± 16.6 days; P < 0.001); D3 (46.8 ± 245.5 vs. 137.3 ± 52.7 days; P < 0.001); D4, 5 (122.5 ± 225.7 vs. 258.2 ± 92.7 days; P < 0.001); and DU (78.1 ± 298.9 vs. 142.5 ± 79.4 days; P < 0.001). Moreover, wound healing was significantly different among the DESIGN-R groups (P < 0.001, data not shown).

Table 2 Characteristics of patients with pressure ulcers after propensity score matching
Full size table

Figure 3 shows the Kaplan–Meier plot for PU progression, according to the DESIGN-R score. The number of cases with PU progression was significantly lesser in the compliance group (n = 92) than in the non-compliance group (n =157) (15 vs. 54, respectively; P = 0.003; Fig. 3).

Fig. 2
figure 2

Treatment period for pressure ulcers according to DESIGN-R category of depth. The two groups in each DESIGN-R score was compared using Mann–Whitney U test

Full size image
Fig. 3
figure 3

Progression of pressure ulcer. Kaplan–Meier estimates for the progression of pressure ulcer. There were 15 events of progression in the compliance group (n = 92) and 54 events of progression in the non-compliance group (n = 157; P = 0.003). *The two groups were compared by log rank test

Full size image

Discussion

The results of our study demonstrate that “Furuta method,” was effective for PU treatment in terms of early complete recovery and prevention of PU progression. However, there is a need for aggressive pharmacist intervention while using this method. It is important to conduct simultaneous prevention and treatment for PU; in this study, it is noteworthy that the compliance group did not deteriorate. Pharmacists who participated in this study, result of trained workshop according to Furuta methods, more improved assessment of topical ointment therapy considering the external force for PUs were compared with similar factors in the non-compliance group.

In a previous report conducted at NCGG, treatment periods according to National Pressure Ulcer Advisory Panel (NPUAP) stage definitions were 10 ± 6 days for Stage II, 43 ± 17 days for Stage III, and 80 ± 32 days for Stage IV [20]. In our study, the mean duration of healing in the compliance group was 23.6 ± 36.8 days in patients with d2, 46.8 ± 245.5 days in patients with D3, and 122.5 ± 225.7 days in patients with D4 and D5. In a previous study, only NCGG data (based on “Furuta method”), encompassed a wide range of treatment periods that was comparable with the present study. We believe the low compliance observed with “Furuta method” may be influenced by the treatment period. Further, the compliance rate with descriptive ulcer dermatology was the lowest, and this pathological assessment was not routinely performed by general healthcare workers. It is necessary to educate pharmacists and general healthcare workers regarding these techniques. A randomized single-blind controlled trial reported 8-week healing rates for Stage II ulcers of only 50 % in both groups [21]. Moreover, Horn et al. reported treatment failure rates for Stage III and IV ulcers of 90 % at 8 weeks [22]. These reports mainly investigated dressing therapy for PU and did not evaluate wound assessment according to wound physical properties and descriptive ulcer dermatology. It has been proposed that topical ointment therapy according to Furuta methods had a greater efficacy than dressing therapy.

Furthermore, in this study, there were significantly fewer cases of PU progression in the compliance group than in the non-compliance group (15 vs. 54; P = 0.003). Thus, we believe that an appropriate wound assessment according to “Furuta method” has utility in preventative management.

This study has several limitations. First, it is a retrospective, albeit multicenter, study; hence, there is a potential for bias. This was minimized by performing PS matching, which has been previously shown to be useful [23]. Second, this study did not consider the involvement of the nutritionist and physical therapist. Therefore, prospective trials that consider these other factors are necessary in the future.

Conclusion

We introduce a concept of PU topical ointment therapy, called “Furuta method,” which considers the physical properties and moist environment of wounds and pharmacist intervention guidelines and may be aided by pharmacist participation. This method may possibly lead to an increase in PU healing rates.