A Systematic Review and Meta-analysis of Randomized Controlled Trials Comparing Interventions for Peristomal Skin Care

Standard skin care procedures for percutaneous endoscopic gastrostomy (PEG) tubes and peristomal skin care for colostomy and ileostomy patients are not always sufficient to prevent peristomal skin problems.   A systematic review and meta-analysis of randomized controlled trials (RCTs) was conducted to compare the effectiveness of standard peristomal skin care to adjunctive techniques or barriers (including glycogel dressings, gelatin- and pectin-based skin barriers, glycerin hydrogel wound dressing, Acacia senegal fiber pockets, hydrocolloid powder crusting, and German chamomile) to manage or treat patients with a stoma. Using systematic literature search techniques, all healthcare databases were searched up through September 2014. No language restrictions were applied. Studies were included if they met criteria for published RCTs or quasi-RCTs that evaluated the outcome of standardized peristomal skin care and other adjunctive techniques or barriers used among patients with a stoma or PEG tube. A meta-analysis was performed to calculate a pooled effect size by using random-effect models for the primary (skin irritation/reaction) and secondary (length of pouch wear time) outcomes. Six RCTs comprising 418 total patients were identified. Four evaluated the outcome of colostomy or ileostomy peristomal skin care, and no significant differences were detected in the incidence of skin problems (RR 0.67; 95% CI: 0.31-1.41). In the two studies that included length of pouch wear time, no significant differences were observed (RR 0.48; 95% CI: 0.03-7.97). No significant differences were seen in the rate of skin irritations of gastrostomy patients (RR 0.56; 95% CI: 0.20-1.59), but the difference in treatment outcomes of peristomal damage in patients with a colostomy was significant (P = 0.01). The limited number of studies, study quality, heterogeneity of variability in peristomal care strategies and techniques, clinical factors, and nonuniform reporting of clinical parameters contributed to the heterogeneity among the trials. Well-designed RCTs are needed to investigate the efficacy of various barriers and techniques for peristomal skin care and to help develop evidence-based standards of caring for the skin of patients with a colostomy, ileostomy, or gastrostomy.

A stoma is an artificial opening created in the abdominal wall that allows effluents, such as stool or urine, to pass from the body. Stomas are created as a result of tumor removal, bowel injuries, defect repair, and inflammatory bowel disease.¹ Based on the site of the artificial opening, stomas result from a colostomy or an ileostomy and include loop, end, and double-barrel types; stomas may be temporary or permanent. In addition, a urostomy may be created for urinary diversion in cases where drainage of urine through the bladder and urethra is not possible. A percutaneous endoscopic gastrostomy (PEG) can be performed to provide enteral feeding to patients with functionally normal gastrointestinal tracts who cannot meet their nutritional requirements because they cannot consume adequate food orally.

Maintaining healthy, intact peristomal skin is essential. However, some narrative reviews^2-5 suggest that despite advances in ostomy care, peristomal skin conditions are common complications among patients. Depending on the type of stoma created, peristomal skin ailment rates of 30% to 60% have been reported in prospective studies.^6,7 A narrative review8 concluded the loss of peristomal skin integrity has been related to chemical injury, mechanical destruction, infectious disease, immunological reactions, and disease-related conditions.⁸ Peristomal skin ailments can range from mild abrasions to full-thickness wounds; stoma-related skin complications make managing the stoma and alleviating discomfort extremely challenging.^4,6,9-11

During typical use of an ostomy pouching system, the peristomal skin constitutes the base plate of the pouching system; keeping this area intact is a critical concern because skin impairment affects the adhesive properties of the appliance (ie, the flange, face plate, or base plate). A stoma accessory may be used in addition to the stoma appliance to prevent appliance leakage and to store excreta.¹² In contrast, gauze dressings are typically used for skin care around PEG tubes. However, standard peristomal skin care is not always adequate for preventing peristomal problems. Numerous approaches are available to prevent or treat peristomal skin complications, including using a different stoma appliance, mixing hydrocolloid powder with a liquid to develop a peristomal skin barrier (also described as a crusting technique), and applying traditional herbal products.^4,11

Two studies compared the effectiveness of stoma appliances and crusting techniques in preventing peristomal skin complications, and several randomized controlled trials (RCTs) have been conducted.^13,14 However, these studies have been inconclusive, most likely because of small sample sizes. The purpose of this systematic literature review and meta-analysis of RCTs was to compare the effectiveness of various standard peristomal skin care techniques — as described by the authors — to other adjunctive approaches or barrier products used by patients with a colostomy, ileostomy, or PEG tube.

Search strategy and study selection. Studies were identified based on keyword searches of the following electronic databases: PubMed, EMBASE, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Physiotherapy Evidence Database (PEDro), Cochrane Central Register of Controlled Trials, and the ClinicalTrials.gov registry. The following terms and Boolean operators were used in MeSH and free-text searches: ostomy OR colostomy OR gastrostomy OR ileostomy OR intestinal stoma, wound, manage OR care OR complications. The “related articles” option in PubMed was used to broaden the search. No language or date of publication restrictions were applied. The final search was performed in September 2014. Additional studies were identified by searching the reference sections of relevant papers and contacting known experts in the field.

Selection criteria. Published RCTs or quasi-RCTs that evaluated the outcome of standardized peristomal skin care and other adjunctive techniques or barriers used among patients with a stoma were considered. A review protocol, developed before the study was conducted, stipulated the research was required to describe the following in order to be included: 1) inclusion and exclusion criteria used in selecting patients; 2) patients with benign or malignant disease admitted to the hospital in order to undergo colostomy, ileostomy, or insertion of a PEG tube; 3) techniques of peristomal skin care employed; 4) the definition of adverse events; and 5) evaluation of peristomal skin condition. Studies were excluded if 1) patients only had a urostomy; 2) clinical outcomes had not been clearly stated; or 3) duplicate reporting of patient cohorts had occurred.

Data extraction. Two of the authors independently extracted the details of the RCTs that pertained to the participants, inclusion and exclusion criteria, peristomal skin care techniques used, skin condition parameters, and complications. The independently recorded decisions of the two author/reviewers were compared, and any disagreements were resolved based on the evaluation of a third reviewer.

Methodological quality appraisal. Two authors independently appraised the methodological quality of each study based on the following criteria: 1) adequacy of the randomization, 2) allocation concealment, 3) blinding, 4) duration of follow-up, 5) number of drop-out patients, and 6) the performance of an intention-to-treat (ITT) analysis. The quality of trials was assessed using the “risk of bias” method recommended by the Cochrane Collaboration.¹⁵

Outcome assessments. The outcomes of peristomal care were evaluated based on the incidence of skin reaction and the length of pouch wear time. Skin reactions included erythema, vesicles, erosion, discoloration, infection, and tissue overgrowth. The variables security and adhesive performance, ease of barrier use, and skin comfort also were rated if reported in the study.

Statistical analysis. Statistical analysis was conducted using Review Manager Software version 5.1 (Cochrane Collaboration, Oxford, England). The meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis guidelines.¹⁶ When necessary, standard deviations were estimated based on the reported confidence interval (CI) limits or standard errors or range values.¹⁷ The effect sizes of dichotomous outcomes were reported as risk ratios (RR), and the weighted mean difference (WMD) was reported for continuous outcomes. The precision of an effect size was reported as a 95% CI. A pooled estimate of the RR was calculated using the DerSimonian and Laird random-effect model.¹⁸ This approach provided wide CIs and an appropriate estimate of the average treatment effect in statistically heterogeneous trials, yielding a conservative statistical claim. The data were pooled only in the case of studies that exhibited adequate clinical and methodological similarity. Statistical heterogeneity was assessed using the I² test, where I² quantified the proportion of the total outcome variability that was attributable to variability among the studies. Subgroup analyses also were performed by pooling estimates for similar subsets of patients across trials, where available.

Characteristics of the trials. The flowchart displayed in Figure 1 shows the process by which the trials were screened and selected. The initial search yielded 657 studies, of which 447 were deemed ineligible based on their titles and abstracts. Of the remaining 208 studies, the following reports were excluded from final analysis: 69 were review articles, seven were animal studies, 15 had used dissimilar comparisons, 68 discussed topics that were not the focus of this study, and 43 were not randomized trials. The remaining six eligible trials^13,14,19-22 were included in analysis and described in Table 1.

The six RCTs were published between 2005 and 2012 and featured sample sizes of 39 to 98 patients. Five evaluated outcomes of using standard skin care/protection compared to additional intervention techniques.^{13,14,19,20,22} One²² involved infants with Hirschsprung’s enterolitis or imperforate anuses who were undergoing colostomy and compared the peristomal skin condition after 4 weeks of treatment using a 2.5% zinc ointment versus Acacia senegal fiber pockets (Heather’s Tummy Care, Inc, Seattle, WA, US). Acacia senegal is a small deciduous Acacia tree known by the common names Rfaudraksha, Gum Arabic Tree, and Gum Senegal Tree; people use the dried seeds of this tree as food and use the gum produced by the tree to soothe the mucous membranes of the intestine and treat inflamed skin.²³

Two of the studies enrolled patients with a PEG and evaluated the outcomes of using standard wound dressing (described in the guidelines of the European Society for Clinical Nutrition and Metabolism²⁴) versus a glycerin hydrogel (Sondofix; Medi-Globe GmbH, Schweinfurt, Germany).^19,20 One study investigated the results of using standard peristomal skin care (cleansing the skin area with tepid water and drying, then applying the ostomy appliances, such as the flange and pouch) versus a crusting technique coupled with dusting hydrocolloid powder in patients with colostomy or ileostomy.¹⁴ An RCT cross-over study¹³ was conducted to assess skin conditions after using a grooved base-plate and wafer adhesive-pouch coupling system (ConvaTec, a Bristol-Myers Squibb Company, Princeton, NJ) versus a gelatin- or pectin-based skin barrier (EasiFlex, Alterna/Assura, and Easy-Removal, Coloplast Corporation, Marietta, GA) in patients with colostomy. One trial²¹ investigated the therapeutic results of using hydrocortisone versus German chamomile compresses in patients with colostomy peristomal skin damage. German chamomile, a plant extract that exhibits anti-inflammatory, antibacterial, and bacteriostatic properties, was reported to facilitate granulation tissue formation and the epithelialization of ulcers.^25,26 Patient and stoma baseline characteristics and details regarding peristomal management techniques are summarized in Table 1.

The methodological quality of the studies was fair. One study was assessed as a quasi-RCT because patient age, body mass index, medication usage, and overall health condition were matched between groups before randomization.²¹ Four studies reported randomization methods.^13,14,20,22 None of the trials described the method of allocation concealment (see Table 2). Only one study reported blinding of the outcome assessors.²¹ Three studies used an ITT analysis.^19,21,22 The patient dropout rate at the 1-month follow-up was considered acceptable at <20% in five of the studies.^13,19-22

Incidence of peristomal skin problems. Four studies provided data regarding the incidence of peristomal skin problems.^14,19,20,22 In two studies,^19,20 the peristomal skin was assessed the day after PEG placement and subsequently 1, 2, and 4 weeks postprocedure by using validated wound scoring systems; the peristomal infection score of each patient was calculated by adding the scores measured for local erythema, induration, and exudation, and the final score was sorted into 5 grades (0 = no reaction, 4 = worst reaction).²⁷ At the end of the first and second weeks, a statistically significant reduction of the mean infection scores was seen in patients with glycerin hydrogel wound dressings in Blumenstein et al.²⁰ However, in Aschl et al,¹⁹ 88% of the patients did not show any relevant sign of infection in both the standard wound dressing and glycerol dressing groups. In the study that recruited infants with Hirschsprung’s enterocolitis or imperforate anuses who were undergoing colostomy,²² the severity of peristomal skin inflammation was defined according to the Physician Global Assessment (no reaction, or mild, moderate, or severe reaction) at 1 month after the intervention.²⁸ Infants who were provided the Acacia senegal barrier showed lower and less severe inflammation rates than those who received zinc sulfate ointment management (P = 0.05).²² In Park et al,¹⁴ the atypical peristomal skin ailments reported included discoloration, erosion, and tissue overgrowth, and the investigators measured the occurrence of these three types of skin problems at 1, 2, and 3 months after the intervention.²⁹ No significant difference was noted in discoloration between the groups, whereas the probability of erosion (OR 0.38, 95% CI: 0.16-0.89) and tissue overgrowth (OR 0.09, 95% CI: 0.02-0.55) was lower in the standard peristomal care group than in crusting technique group.¹⁴

To facilitate comparison in the meta-analysis, studies that addressed only the severe skin reaction assessed at 1 month (scores of 2–4 in the studies of Blumenstein et al²⁰ and Aschl et al¹⁹); moderate and severe cases in the study of Hosseinpour et al²²; and patients with at least one atypical skin complication and tissue overgrowth in the report of Park et al¹⁴ were combined. Overall, 38 out of 150 patients (25.3%) in the experimental groups experienced a peristomal skin reaction compared with 55 out of 157 (35%) in the control groups. The current analysis revealed no significant differences in the incidence of peristomal skin reaction between the two treatment groups, the RR was 0.67 (95% CI: 0.31–1.41), and substantial heterogeneity was observed in the incidence of peristomal skin reaction among the trials (see Figure 2). Subgroup analysis also revealed no significant differences between experimental and control groups in both PEG (RR 0.56, 95% CI: 0.20–1.59) and in colostomy/ileostomy (RR 0.48, 95% CI: 0.03–7.97) subgroups.

The data reported by Berg and Seidler¹³ were not pooled because the specific number of patients with a skin problem was not provided. However, these authors reported no participants experienced episodes of skin breakdown based on the incidence and the reports of the detected peristomal skin problems, participant withdrawals, and adverse events of the two groups in the study.

Pouch wear time. Two RCTs^13,14 investigated time between ostomy pouch changes: one described the results of the crusting technique coupled with dusting hydrocolloid powder in patients with colostomy or ileostomy,¹⁴ and the other study¹³ evaluated the outcomes after using a grooved base-plate and wafer adhesive-pouch coupling system, and a gelatin- or pectin-based skin barrier in patients with colostomy. No heterogeneity was noted across the studies (I² = 0%; chi-square = 0.06; P = 0.81) nor statistically significant differences in pouch wear times between the two treatment groups (WMD -0.18; 95% CI: -0.49–0.13) (see Figure 3).

Skin damage treatment outcomes. Charousaei et al²¹ evaluated the results of hydrocortisone and German chamomile compress treatment among patients with peristomal skin damage. Lesions healed significantly faster in the chamomile users compared with the hydrocortisone group (mean times to healing 8.89 ± 4.89 and 14.53 ± 7.6 days, P = 0.001). Peristomal skin pain and itching also were relieved sooner in the chamomile group than the hydrocortisone group.

An optimal colostomy or ileostomy peristomal skin management system should effectively prevent skin problems and appliance leakage to avoid the development of complications. However, the comparative effectiveness of the specific components that constitute such systems has not been established. The results of a systematic review and meta-analysis indicated that when adjunctive techniques or barriers were used, no statistically significant benefit in peristomal skin condition was derived compared with using standardized stomal care methods. In contrast, subgroup analysis from two trials also revealed no significant differences in peristomal skin condition surrounding PEG tubes between experimental and control groups. Although each of the studies analyzed reported advantages associated with the adjunctive techniques used, methodological inconsistencies among the trials confounded attempts to compare the overall outcomes of various skin barriers and techniques used across these studies.

Numerous factors affect ostomy-related complications. A retrospective analysis³⁰ of the medical records of 1,170 patients who underwent end colostomy in South Korea found irritant contact dermatitis, occurring in 15.5% of patients, was the most common peristomal complication. Gender and body mass index were associated with irritant contact dermatitis, hyperplasia, peristomal hernias, flat stomas, and retracted stomas.

Herlufsen et al³¹ conducted a cross-sectional study of 202 patients with ostomies; results indicated that patients with an ileostomy exhibited the highest rate of complications (57%), followed by those with urostomy (48%) and colostomy (35%). A prospective study³² investigated the long-term results of ostomy-related complications: the sample comprised 144 patients who underwent an intestinal ostomy as part of an acute abdominal surgery. Results indicated stomal and peristomal complications were prevalent during the first 2 postoperative years and particularly during the first 6 months.

In the studies examined in the current review, all outcomes were evaluated within 3 months of stoma creation. Therefore, additional RCTs must be conducted to confirm the long-term results of using distinct peristomal skin-care strategies.

The skin of the neonate has been identified as a risk factor that influences peristomal skin complications. A clinical practice guideline has shown premature neonates present diminished cohesion between the dermis and epidermis and can be at risk of systemic toxicity.³³ In the study reviewed in the current research that involved infants with Hirschsprung’s enterolitis or imperforate anuses undergoing double barrel colostomy,²² infants that received the Acacia senegal barrier exhibited lower and less severe inflammation rates at 4 weeks after surgery (P = 0.05) than infants in the zinc sulfate ointment group. Moreover, no side effects were reported. However, Acacia senegal fiber may not be readily available in European countries, and additional RCTs are required for evaluating its preventive effects and for the medical community to establish the specific use of commercially available Acacia senegal fiber in stoma care.

Although numerous options are available for managing peristomal skin ailments, few studies have compared these strategies.² The treatments may involve using compresses, topical corticosteroid preparations, or barrier products and preparations.⁴ One of the reviewed studies compared the effect of a German chamomile solution with that of topical steroids on peristomal skin lesions in colostomy patients, and the results indicated lesions healed considerably faster in the chamomile group compared with the hydrocortisone group.²¹ Because corticosteroids are nonspecific, anti-inflammatory agents, using herbal extracts could facilitate avoiding the side effects observed after long-term topical corticosteroid use. However, to apply the chamomile compress, participants in that study had to remove their appliance twice a day, which is inconvenient and can cause skin irritation. Additional randomized studies are required to confirm the results of this study.

The number of studies included was small; three investigated regular colostomy or ileostomy care, two studied skin care surrounding PEG tubes, and one evaluated management of peristomal skin problems. Also, variability in peristomal care strategies and techniques, clinical factors, and nonuniform reporting of clinical parameters contributed to the heterogeneity among the trials reviewed; population characteristics and the types of stoma differed across the studies and clinician experience and patient characteristics can affect clinical outcomes. Also, the peristomal skin condition was evaluated for distinct durations in the studies, rendering the current assessment vulnerable to measurement bias.

Several trials were small and one study recruited only 16 patients in the control group.¹³ None of the studies included in this analysis reported adequate allocation concealment. The assessment personnel were blinded to the outcomes in only one study.²² Half of the investigators analyzed their data according to the per-protocol principle, which may have biased their evaluations regarding the effect of peristomal care. Furthermore, even using comprehensive search strategy, the possibility of publication bias exists. Studies with null results are generally less likely to be published; therefore, they are more likely to be missed in a database search.

The strengths of this review include a comprehensive search for relevant studies, the systematic and unambiguous application of eligibility criteria, the cautious consideration of study quality, and rigorous analysis.

A meta-analysis indicated using advanced techniques or barriers is unlikely to markedly improve peristomal skin condition compared with the routine use of more standard stoma care practices. Although such interventions may improve the state of the peristomal skin, no statistically significant difference was detected in the incidence of peristomal wound reactions and the length of pouch wear time in patients treated with standard or advanced stomal care. Similarly, no significant difference between groups was found in the incidence of peristomal wound reactions in patients with PEG insertion. However, lesions healed significantly faster in the chamomile users compared with the hydrocortisone group in patients with peristomal skin damage. Repeat studies using advanced techniques or barriers are essential to strengthen the statistical power and confirm the effects. Well-designed RCTs are required to investigate the efficacy of varying barriers and techniques for peristomal skin care, enabling clinicians to improve patient quality of life.

Ka-Wai Tam and Tsai-Wei Huang designed the study. Ka-Wai Tam, Tsai-Wei Huang, Jun-Hung Lai, and Hung-Chou Chen extracted the data. Ka-Wai Tam, Tsai-Wei Huang, Jun-Hung Lai, Hung-Chou Chen, and Wen-Hsuan Hou analyzed and interpreted the data. Ka-Wai Tam and Tsai-Wei Huang wrote the first draft. All authors contributed to subsequent versions of the manuscript and approved the final article. Tsai-Wei Huang is the corresponding author.

Dr. Tam is a researcher, Center for Evidence-Based Medicine; Assistant Professor, Department of Surgery, School of Medicine; PhD candidate, Graduate Institute of Clinical Medicine, College of Medicine; staff physician, Division of General Surgery, Department of Surgery, Taipei Medical University, Taipei, Taiwan; and Director, Center for Evidence-Based Health Care, Taipei Medical University—Shuang Ho Hospital, New Taipei City, Taiwan. Dr. Lai is an instructor, Department of Nursing, College of Medicine and Nursing, HungKuang University, Taichung City, Taiwan; and staff physician, Department of Internal Medicine, Erlin Branch of Changhua Christian Hospital, Changhua City, Taiwan. Dr. HC Chen is a staff physician, Department of Physical Medicine and Rehabilitation, Taipei Medical University—Shuang Ho Hospital. Dr. Hou is a researcher, Center for Evidence-Based Medicine, Taipei Medical University; staff physician, Department of Physical Medicine and Rehabilitation, Taipei Medical University; and Assistant Professor, School of Gerontology Healthcare Management, College of Nursing, Taipei Medical University. Dr. Ko is a staff physician, Department of Internal Medicine, Kuang Tien General Hospital; and Assistant Professor, Department of Nutrition, Master Program of Biomedical Nutrition, College of Medicine and Nursing, HungKuang University. Dr. SL Chen and Dr. Huang are Associate Professors, Department of Nursing, College of Medicine and Nursing, HungKuang University.

Please address correspondence to: Tsai-Wei Huang, RN, PhD, Department of Nursing, College of Medicine and Nursing, HungKuang University, Taiwan, 1018, Sec. 6, Taiwan Boulevard, Shalu District, Taichung City 43302, Taiwan; email: m8706001@gmail.com.