The review considered the randomized clinical trial (RCT), quasi pretest-posttest, and descriptive studies published in English. Participants in the studies were adult, aged over 18 years, considered to be risk to develop PU, have no PU at the onset of the study, and managed at any healthcare setting. The primary outcome measured in the included studies was considered as the incidence of hospital acquired pressure ulcer (HAPU).

Any direct wound intervention, could be topical or commercial wound dressing applied to bony prominence areas. The intervention of RCTs, descriptive, or comparative were included in this review, to compare with two different interventions, or to compare the intervention with the standard care, or to compare the intervention with placebo.

We searched the following electronic database to evaluate the effectiveness use of topical agent of different dressings to prevent pressure ulcers: EBSCO CINHAL Plus (Cumulative Index to Nursing and Allied Health Literature, 2010 to 30 March 2019), the Cochrane Central Register of Controlled Trials (CENTRAL). We selected the English language to retrieve the included studies.

According to Cochrane guidelines for systemic reviews of interventions, firstly we assessed the titles. Then, abstracts’ eligibility was assessed to be included in the review. Next, we reviewed the available full version of studies. The excluded studies that used different interventions rather than topical agent or direct dressings to prevent pressure ulcer. These interventions as; massage, electromagnetic waves, nutritional management, redistributing of pressure sites, wheelchair cushions, improve self-management and self-efficacy of patients, using risk assessment tool, and follow preventive evidence-based practices. Last, we completed a PRISMA flowchart to summarize the selection process (Figure 1).

Two reviewers independently reviewed the following data from the eligible studies: author, title, care setting, inclusion/exclusion criteria, participants’ characteristics, study design, method of randomization, intervention details, type and frequency of dressing, outcome measure, length follow-up, and conclusion as reported by authors. According to Cochrane methods [6] we considered the risk ratio (RR) with confidence interval (CI) of 95% for dichotomous variables. For continuous variables, the mean difference with confidence interval (CI) 95%. Disagreement was resolved through discussion, or with third party judgment if needed.

The GRADE system rates the quality of the evidence, and summary the findings.

Table presents the results of the most important outcomes in the reviewed studies [7] . The GRADE system rates the quality of evidence for each outcome, from a rating of high to very low. The GRADE baseline is high rates for RCTs, and low for non-RCTs. The assessment criteria to consider in baseline rating for downgraded, or upgraded are: risk of bias, inconsistency, indirectness, imprecision, and publication bias (mainly for RCTs) [7] .

The summary table of findings (Table 1) includes: 1) The comparison between the intervention and control with similar intervention. 2) The comparison of different interventions of similar group at the same time limit. 3) The incidence of pressure ulcer. Our search identified 8 studies.

Five studies are RCTs with total of 1846 participants [8] - [13] . Results are summarized in Table 1 “summary of findings”.

The five RCTs’ participants were adults patients (>18 years) who were admitted to hospitals free from any sign of skin lesion. All included participants were free of pressure ulcers before study involvement. The included studies used Braden scale to assess participants’ eligibility, and the acceptable scale range to participate in the study varies to be (13 - 15) [8] [11] [12] [13] , and Waterlow scale in addition to Braden scale [10] .

The setting where the studies were conducted was; two studies in Spain [12] [13] , one study conducted at eight medical-surgical wards in Academic Acute Tertiary hospital in Singapore [8] , one study conducted at orthopedic unit in Iran [10] , and one study conducted at critical care unit at A magnet hospital, but did not mention the country of [11] .

The included studies used randomization of participants’ allocation [8] [10] [11] , but two studies used multicenter, parallel randomization [12] [13] . The blinding method used in these studies were; triple-blinding in two studies [10] [12] , double-blinding in one study [13] , non-blinding in one study [11] , and in one study the blinding method was single for research coordinator [8] .

Aloweni, et al., [8] assess in their study the effectiveness of a prophylactic silicon foam dressing and tropical application of fatty acid in addition to standard preventive measures. The intervention groups received; 1) the standard care (repositioning, positioning devices, diaper change, and barrier cream). 2) In addition to standard care, Silicon Foam dressing, Mepilex Border Sacrum^TM, that applied to the sacral area. Wound nurse specialist changes the dressing every seven days or when seals. 3) fatty acid oil plus standard care received Linovera Oil^Ò which consists of hyperoxygenated fatty acid, that applied onto sacral area three times daily.

Kalowes, et al., [11] assess the effectiveness of 5-layered soft silicon foam dressing versus a control group receiving the standard care. The intervention group received Mepilex Border Sacrum foam dressing^TM over sacral areas, the dressing had changed every three days or when dislodge. The control group received the standard care (Total Care SpO2RT 2 Therapy Bed [Hill-Rom, Inc], repositioning, and incontinence skin care).

Lupinaez-Perez, et al., [12] assess the inferiority of application of hyperoxygenated fatty acid (HOFA) to an intervention group with application of an olive oil composition to the control group, in addition to the usual care for both groups. The HOFA applied topically and included Equisetum Arvense, Hypericum Perforatum and perfume. The olive oil procedure consisted in applying a magistral formula, in liquid spray form, containing 97% extravirginolive oil and 3% Hypericum Perforatum and Peppermint. The olives used for this product is the “Picual”, both products have similar appearance.

Hekmatpou, et al., [10] assess the effectiveness of Aloe-Vera oil vs placebo in preventing PUs for immobilized patients. Te researcher applied the Aloe-Vera from dark glass container and rubbed onto skin twice daily over the pressure points along 10 days duration. The control group received the placebo (water and starch gel) that was similar to the Aloe-Vera consistency, and at same protocol.

Verdú & Soldevilla, [13] compared the efficacy of (IPARZINE-4A-SKR) to prevent PUs over a placebo. The product applied to an intervention group every 12 hours, with gentle massage until it was completely absorbed over high risk areas of PUs. The placebo applied to the control group at same steps.

Our search found three non-randomized clinical trials (Table 2). Total number of participants were 433 patients. One of these studies was quasi-experimental study assess the effectiveness of using intravenous bags as compared with a commercially available heel suspension foam boot especially designed to offload the foot (Heelift; DM Systems Inc, Evanston, Illinois) for patients at high risk of pressure ulcers [14] . The results showed significant effect of using the pressure-relief suspension boot. Thus, non of patients used the pressure-relief suspension boot experienced any signs or symptoms of PUs, whereas 6 out of 15 patients who used the intravenous bag developed PUs signs and symptoms. This study had some methodological limitations that may affect its quality. The sample size is small (n = 30). The participants recruited to this study were patients admitted for knee and hip surgery, this will affect the generalization of the results. Also, the patients in intravenous bag had high response dose of PU incidence rate; that 6 out of 15 patients experienced PUs, which is higher than the result of the International PU Prevalence Survey 2006-2015 which is 9.3% [2] .

The second study of a prospective, nonrandomized, qusi-experimental observational study was conducted to compare the effectiveness of implementation of prophylactic silicon adhesive hydrocellular sacral foam dressing [15] . This study was conducted in three ICU wards, patients are at risk of PUs, the sample size of this study is (n = 243). The results indicate the obvious decreasing in the incidence rate from 7.6 to 3.4 per 1000 population. But there are many consideration in methodology such as; the non-randomization sampling, repositioning of the patients was not monitored, and units increased training programs of preventive

measures during the study period, which may decreased the incidence rate of PUs.

The third study [9] aimed to assess the effectiveness of Hydrocolloid dressing (HD) in comparison with Transparent polyurethane film (PF). A total of 160 patients were recruited in this study, 80 patients were in each group. Participants were observed to evaluate the effectiveness of treatment by the incidence of PUs, reasons to change the dressing, number of dressing per area, and total number of dressing during the study period. The results showed that in HD group the first dressing was after two days for 13 (16.3%) patients, then after seven days for 21 (26.3%) patients. In PF group the first change after five, ten, and twelve days for 23 (28.7%), 34 (42.5%), and 11 (13.8%) respectively. The incidence rate was decreased significantly in PF group was 8.7%, and in HD group was 15%. Despite the significant effect of PF to decrease the incidence of PUs, but there are some differences between groups, it would be better to have control group.

The most effective framework to evaluate a systematic review is GRADE (Grading quality of evidence and strength of recommendations). GRADE is used to rate the certainty of evidence for a treatment efficacy from high to very low. The GRADE system takes in two types of studies: randomized controlled trials (RCTs) and non-randomized trials [7] . The five criteria to evaluate the RCTs and could downgrade one or two level are (risk of bias, in directedness, inconsistency, imprecision, and publication bias). Table 3 illustrates the details of each study evaluation criteria.