INTRODUCTION

Creation of a fecal ostomy is an integral part of the surgical management of neoplastic lesions, inflammatory bowel diseases (IBDs), congenital malformations, or trauma of the intestinal tract.1 Approximately 1 million new stomas are created annually, with the largest portion of stomas being created in patients between 50 and 70 years of age.2 In the United States, approximately 100,000 to 150,000 intestinal stomas (fecal diversions) are created annually.3,4 The most common causes of fecal diversions include colorectal cancer (44.6%-74.1%) and IBDs (12.1%-28.2%).3,5-8 Other indications for fecal diversions are complicated diverticulitis (12.7%) and fecal incontinence (6.7%).5 Fecal diversions are occasionally created for the management of radiation proctitis, traumatic injury, infectious colitis, chronic wounds, anastomotic leak, and neurogenic bowel dysfunction following spinal cord injury. Ostomy creation can be permanent or temporary, as well as elective or part of an urgent/emergent surgical procedure.

Stoma complications affect a significant proportion of persons living with a fecal ostomy.9-12 Stoma complications include mucocutaneous separation, stomal retraction, stenosis, necrosis, prolapse, fistula, trauma, and peristomal hernia. Peristomal complications include peristomal moisture-associated skin damage, allergic contact dermatitis, mechanical injury (both pressure and adhesive-related), fungal infection/candidiasis, varices, folliculitis, pyoderma gangrenosum, hyperplasia, and suture granulomas.13

While these complications may occur at any point following ostomy creation, evidence suggests that the majority of complications develop within the first 5 years following stoma surgery.4 Reported incidence rates of stomal and peristomal complications vary from 2.9% to 81%.1,4,11,14-17 Complications may prolong hospital stay, increase the risk of hospital readmissions, and escalate the number of visits to outpatient clinics, resulting in substantial costs related to ostomy care; moreover, these complications negatively affect health-related quality of life (HRQOL).18-20

Due to the growing number of ostomies created on an annual basis, the prevalence of stoma-related complications is expected to increase.21-24 Determining factors that contribute to the development of stoma complications could help identify patients at a greater risk for complications and possibly guide interventions to decrease risk. Multiple risk factors have been examined including body mass index (BMI), tobacco or alcohol abuse, underlying disease leading to ostomy surgery and treatments associated with the disease (chemotherapy, radiation therapy, chronic steroid use), presence of comorbid conditions such as diabetes mellitus, coronary heart disease, hypertension, chronic obstructive pulmonary disease, and surgery-related factors (type of stoma such as on the small or the large intestine, loop, terminal, surgeon experience), position of the stoma on the abdomen and in relation to the rectus sheath, preoperative marking of the stoma site, and the setting in which the ostomy was created (urgent/emergent vs planned).6,25-35 However, studies vary and no consensus around which modifiable factors are most amenable to preventive interventions exists.31,33,34,36-38 Therefore, the purpose of our scoping review was to identify patient-related risk factors contributing to an increased risk of stoma and peristomal complications in patients with fecal ostomies.

METHODS

Two reviewers (P.K. and P.Z.) searched the literature independently using the following online bibliographic databases: PubMed, MEDLINE, Cochrane, Google Scholar, and EMBASE for studies published between January 1980 and December 2018. The search string used to search the databases is depicted in Table 1. We included the following types of articles/elements: systematic reviews with or without meta-analysis of pooled analysis, randomized controlled trials (RCTs) including reviews from the Cochrane Library, nonrandomized cohort studies, observational studies, and other types of review articles. We included elements published in the Dutch, English, French, German, and Spanish languages. The search strategy is detailed in the Figure. Studies were originally screened in abstract format seeking additional inclusion criteria for study participants 18 years and older, original research or review that primarily focused on fecal stomal and parastomal complications, and original research/reviews analyzing risk factors for the development of stomal and parastomal complications in persons with fecal ostomies. Additional articles were identified through the examination of reference lists (ancestry search). We excluded gray literature sources (abstracts), clinical audits that did not provide detailed quantitative measurements, editorials, letters to the editor, and individual case reports.

Figure. PRISMA Flow Diagram: Patient-related risk factors associated with stomal and peristomal complications following fecal ostomy surgery.

TABLE 1. Scoping Review Search Strategy

SEARCH STRATEGIES

Two reviewers searched the literature (P.Z., P.K.) and screened the abstract titles and study type. A second pair of reviewers (P.K., M.Z.) read selected elements in full and selected elements meeting inclusion and exclusion criteria. Any disagreements were resolved by discussion. Four investigators (P.Z., P.K., J.P.Z., and J.M.-W.) extracted relevant data from each element including authors, publication date, study design, number of participants, outcomes, overall complications rates, complication rates related to stoma type and duration of follow-up, conclusions, and study heterogeneity. Specific data related to potential risk factors for each stoma or parastomal complication and the rates of any complication in groups based on sex, age, BMI, and the underlying cause of ostomy were also extracted, when available. Discrepancies involving lack of concordance in extracted data values were resolved by discussion with the other investigators (A.D., M.M.).

Extracted data were then divided into 3 categories and grouped into relevant tables. Table 2 summarizes details of included elements (authors, number of participants, number of studies included in systematic reviews, reported complication rates, and study or review design). Table 3 summarizes complication rates for ostomy types and data regarding analysis of patient-related risk factors. The conclusions drawn from the selected studies/reviews in all 3 groups were ranked using the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) approach by 2 reviewers (M.Z., M.M.).26,27 Two independent reviewers (J.P.Z. and A.D.) verified the ratings, and any disagreements were reconciled by discussion. The rating for each study included in the review is depicted in Table 2.

TABLE 2. Summary of All Studies Included in the Scoping Review Assorted by Date of Publication

TABLE 3. Rates and Risk Factors for Ostomy-Related Complications From Original Investigative Reports Included in the Scoping Review

We selected the GRADE instrument to evaluate the quality of underlying evidence and generate strength of recommendations from the studies included in our review.26,27 In addition to examining study design, the GRADE approach enabled ranking of evidence based on risk of bias, imprecision, inconsistency, indirectness of results, and publication bias. Ultimately, the quality of evidence for each outcome was divided into 4 categories ranked from high to very low.39,40

FINDINGS

The initial search produced 1266 potentially relevant articles and another 30 identified via ancestry search. The flowchart describing the process through which eligible studies were selected is presented in detail in the Figure. After removing the duplicate results from the different databases, 679 records remained. Three hundred seventy-two records were then removed after initial abstract title and study type were reviewed. Abstracts of 307 elements were screened for those meeting inclusion criteria, resulting in 142 possibly relevant studies for which full-text articles were reviewed. Seventy-four elements were excluded mostly due to lack of ability to retrieve the data for fecal ostomies from cohorts that included both fecal stomas and urostomies or those that reported findings from pediatric and adult patients. Articles were also removed if full text were not available (at all or in inclusion languages) or had issues with data reporting. This resulted in 68 studies included in our review.

Descriptive features of elements included in our scoping reviews are summarized in Table 1. They include 4 systematic reviews with meta-analysis of pooled findings (one was a Cochrane review), 7 narrative reviews, 2 prospective randomized studies, 13 prospective observational or cohort studies, 3 cross-sectional surveys, and 39 retrospective observational or case-control studies.

More than 60% of elements included in the review were of retrospective design, and their quality as per GRADE ranking was assessed as low or very low. These studies were mostly observational in nature and exposed the risk of bias related to detecting and properly reporting stoma and peristomal complications without any intentional randomization according to stoma type or risk factors. A limited number of prospective or cross-sectional surveys are included in our scoping review; these elements were retained because the risk for bias was reduced by providing regular assessment for the presence of complications by physicians or nurses during planned follow-up visits or randomizing the patient to specific subgroups. We assert that the comparatively small number of systematic reviews with meta-analysis illustrates the generally low quality of data in this area of care.

Complication Rates

The reported rates of stoma and peristomal complications are presented in Table 3. A majority of articles reported stoma and peristomal complication rates around 40%, though the range of rates varied widely from 12% to 82%.9,41,42 The most common early ostomy-related complication was peristomal complications, whereas parastomal hernia was the most common late complication. Multiple studies focused exclusively on these 2 types of complications.27,43-50

Peristomal complications were reported in 6% to 74% of patients.6,51-53 Peristomal skin complications were reported in 0% to 74% of patients with ileostomies and from 3% to 20% of patients with colostomies.6,14,26,51,54,55 The literature we reviewed also suggested that peristomal complications tend to be underrecognized and underreported.

We also extracted data related to long-term complications after ostomy surgery. The rates of parastomal hernia were reported between 1% and 40%, with a median of 10%.8,34,43-45,56-58 We found sparse research related to stoma or peristomal complications occurring 5 years or more after ostomy surgery. Nevertheless, findings from the literature suggest that the rate may be as high as 58% for long-term complications associated with colostomies5 and as high as 76% for those living with an ileostomy.6

Our scoping review also identified potential systemic complications that were more frequently observed in patients after ostomy surgery such as the need for postoperative mechanical ventilation (13.7%), the need for blood transfusion after surgery (8.9%), sepsis (7.9%), and pneumonia (7.1%).35,59 Shellito60 found the risk of sepsis after surgery for intestinal stoma in his narrative literature review study was between 1% and 15%, whereas readmission rates following ostomy creation for dehydration were found in retrospective studies on 201 patients by Paquette and colleagues61 and on 603 patients by Messaris and colleagues62 to range from 17% to 43%. In addition, Messaris and colleagues62 reported the 60-day readmission rate was 16.9% (n = 102) for those with dehydration, the most common systemic complications in their retrospective study comprising of 603 patients with loop ileostomies (43.1%).62

Age as a Risk Factor for Stoma and Peristomal Complications

Age emerged as a potential risk factor for the development of stoma and peristomal complications. Age was reported as an independent risk factor for ostomy-related complications in 12 original studies and was identified as a potential risk factor in systematic reviews.44,61,63-65 Most of the studies reported that the risk of ostomy complications increases with patients 65 to 68 years and older. Moreover, studies reported that advanced age was associated with a higher risk of systemic complications described previously. Nevertheless, findings from some studies indicated that only certain complications are associated with aging whereas others occur more frequently in younger adults. For example, Sung and colleagues66 reviewed medical records in a single-center study and reported that stoma retraction is more common in patients older than 65 years than in younger groups ([chi]² = 5.4, df = 1, P = .021) but the risk of stoma hyperplasia was higher in younger patients ([chi]² = 12.2, df = 1, P = .001).

A minority of researchers found that younger age may increase the overall risk for stoma and peristomal complications. For example, in a prospective observational study of 3970 patients, Cottam and colleagues33 reported the risk of stoma complications decreases with age. They found that patients aged 26 to 40 years had a significantly higher rate of stoma and peristomal complications when compared to participants aged 71 to 85 years (38.5% vs 31.1%; P = .009).33 Similarly, Jayarajah and colleagues67 retrospectively studied medical records of 192 patients and reported a higher risk of complications in patients younger than 60 years old, though this difference was not statistically significant (P = .063). Specifically, participants younger than 60 years developed more parastomal hernias (3% vs 12%; P = .06) and hyperplasia of the peristomal skin (38% vs 23%; P = .06). Several factors may have influenced these findings: (1) more than 70% of participants were younger than 60 years, and younger patients were more likely to undergo ileostomy surgery (50% vs 14%; P = .02).

Sex

Sex was identified as an independent risk factor for ostomy-related complications in 7 original studies.28,33,44,49,50,66,68 In 5 of the 7 studies, female sex increased the risk for complications. Our review also found that some authors reported that female sex was associated with a higher likelihood of specific types of stoma and peristomal complications and that women with ostomies may suffer more frequently from specific ostomy complications. In a prospective multicenter study conducted in the United Kingdom of 3970 participants with ostomies, female sex was associated with overall higher rates of ostomy complications (38% vs 30%; P < .001).33 The most frequent complication was ostomy retraction (occurring in 40% of all patients), followed by mucocutaneous separation. The same trend was observed in a large retrospective study in Korea (1170 patients), in which rates of stoma retraction (11.6% vs 4.2%; P < .001), flat stoma (12.4% vs 5.7%; P < .001), and parastomal hernia (8% vs 4.2%; P < .005) were higher in female patients. There are several possible explanations for these outcomes. In the single-center retrospective study of 199 patients conducted by Qureshi and colleagues,16 a stoma was more often created in females in an emergent settings whereas males were more likely to undergo planned stoma creation. Moreover, in prospective observational studies by Carlsson and colleagues57 (n = 207) and Cottam's group33 (n = 3970), the researchers postulated that females may be at a higher risk of developing ostomy-related complications due to having lower stoma height (P = .002 and P < .001, respectively) when compared to males. This difference may be related to a higher percentage of females undoing surgery in emergency settings without preoperative stoma siting, along with technical difficulties during surgery, precluding creation of proper stoma height. The authors of this study and other reports in the literature concluded that lower stoma height was also influenced by female preference related to body image.16,29,33

Male sex was identified as a significant risk factor in 2 studies.50,68 One was a retrospective single-center study of 1076 patients and the other of 5019 patients and focused on parastomal hernia/parastomal bulging.

Body Mass Index

Body mass index was associated with an increased likelihood of stoma and peristomal complications in multiple studies.10,31,33,37,44,46,47,49,66 Development of peristomal complications (allergic dermatitis, pyoderma gangrenosum) and stomal complications (parastomal hernia, stoma retraction) occurred more frequently in participants classified as overweight and obese (BMI >=25).10,33,37,44,47 For example, in a retrospective case-control study of 202 patients conducted by Harilingam and colleagues37 and a prospective multicenter observational study of 192 patients by Parmar and colleagues,10 patients with BMI 30 or more had a 3 times higher risk of developing ostomy-related complications compared with individuals with BMI less than 30 (OR = 3.30; 95% CI, 1.61-6.78).

Underlying Indication for Ostomy Surgery

We found 4 studies that reported early stoma complications were significantly more frequent in patients with malignant disease as compared to benign disease.8,17,32,69 In addition, Malik and colleagues17 reported higher complication rates among those with end colostomies due to malignancies in a systematic review of RCTs with a pooled enrollment of 1009 patients. Thalheimer and associates32 retrospectively evaluated 120 patients in a single-center retrospective study and reported a higher rate of complications in patients undergoing adjuvant chemotherapy or combined radio- and chemotherapy than in patients receiving no additional therapy (25.5% vs 9.2%). Oliphant and coworkers69 reviewed the medical records of 222 patients and reported that 3 months after surgery, patients with a loop ileostomy who received adjuvant chemotherapy had significantly poorer ostomy function scores compared to those who underwent surgery alone (4.55 vs 1.53; P = .041). In contrast, they found there were no differences in ostomy function scores at 3 months in patients who underwent colostomy (2.00 vs 2.62; P = .411) or end ileostomy (1.00 vs 2.00; P = .170).69 Distinct from these findings, we retrieved a study that found stoma prolapse occurred less frequently in patients diagnosed with malignant disease versus benign disease (OR = 0.330; 95% CI, 0.106-1.027; P = .047).67

Despite a potentially higher risk of stoma and peristomal complications in patients with underlying malignancies, we found that clinically relevant rates of stoma and peristomal complications are observed in patients with IBD, particularly Crohn's disease, where surgical stoma creation may be a necessary treatment in the course of the disease.14,33,34,36,48,70,71 In a retrospective case-control study of 164 patients conducted by Duchesne and colleagues,36 11 (55%) of 20 patients with IBD were noted to have ostomy complications. Moreover, IBD was a significant predictor of ostomy malfunction (OR = 4.49; 95% CI, 1.16-17.36).36 Similar results were obtained by Pittman and colleagues48 in their cross-sectional, descriptive case-control study of 239 patients in which they observed less serious skin complications and ostomy pouch leakage in patients with colorectal cancer compared to patients with IBD (13.6% vs 30.3%; P = .003) or intestinal diverticulitis (18.2 vs 43.6; P = .0001). In a single-center retrospective cohort study conducted by Takahashi and associates34 of 43 patients with ulcerative colitis and 59 patients with Crohn's disease, stoma-related complications (fistula, retraction, and stenosis) occurred more frequently in patients with Crohn's disease versus those with ulcerative colitis (36.8% vs 17.4%; P < .05).

Comorbid Conditions

Although it is widely known that the presence of significant comorbid conditions may negatively affect the postoperative course, there is little evidence regarding the development of ostomy-related complications and which conditions are most likely to influence the risk for stoma and peristomal complications. In a retrospective study that examined 1216 patients up to 10 years following creation of a fecal ostomy, Nastro and colleagues5 reported 807 complications that developed in 564 patients (46.4%). Comorbid conditions associated with occurrences of stoma and peristomal complications were heart disease (OR = 1.69; 95% CI, 1.18-2.42; P = .004), musculoskeletal disorders (OR = 2.56; 95% CI, 1.61-4.04; P < .001), cigarette smoking (OR = 1.14; 95% CI, 1.01-1.29; P = .032), diabetes mellitus (OR = 1.73; 95% CI, 1.13-2.65; P = .011), and ASA (American Society of Anesthesiologists) physical status III or IV (OR = 4.33; 95% CI, 2.60-7.23; P < .001). An ASA score of more than 3 was also independently associated with the presence of postoperative complications, including stoma and peristomal complications, in a retrospective case-control study of 203 patients conducted by Emmanuel and colleagues72 and a multicenter retrospective study of 85 patients reported by Dumont and coworkers.73 Type 2 diabetes mellitus was found to have an independent association with late peristomal skin complications (P = .045) and ostomy located in a crease (P = .028) in a prospective study of 97 patients conducted by Arumugam and colleagues.31 Diabetes mellitus emerged as the only significant patient-related factor for high-output ileostomy (P = .034) in a retrospective case-control study of 164 patients reported by Takeda and coworkers.74 Nevertheless, it is important to remember that ileostomy is more likely to result in a higher-volume output than colostomies with subsequent dehydration, electrolyte abnormalities, vitamin deficiencies, and malnutrition.75

Preoperative Stoma Site Marking as a Protective Factor

We found multiple and numerous reports of reductions in frequency of stoma and peristomal complications, along with improved HRQOL in patients receiving perioperative stoma site marking and education/counseling from a WOC or ostomy nurse specialist.8,10,11,16,29,57,76-78 While results from individual studies differ, findings from most studies suggest that preoperative stoma site marking by a WOC nurse is protective against complications. For example, in a prospective, nonrandomized case-control study of 60 patients conducted by Karaveli Cakir and Ozbayir78 findings showed that preoperative stoma site marking increased quality of life at the 6-month period following stoma creation. These findings were similar to those reported by Mahjoubi and colleagues,79 Person and colleagues,80 Maydick,81 and McKenna and colleagues,82 who reported that patients receiving preoperative stoma site marking have significantly higher HRQOL scores than unmarked patients. Mahjoubi and colleagues79 in a cross-sectional, case-control study of 348 patients reported that patients with appropriate stoma sites achieved better results on several functional scores measured with the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire EORTC QLQ-C30 and colorectal-specific QLQ-CR38 questionnaires. In the single-center study of 105 patients, Person and colleagues80 reported that improved HRQOL occurred irrespective of permanent versus temporary stoma surgery or underlying cancer diagnosis. McKenna and colleagues82 reported similar findings in a study of 59 patients who underwent ostomy surgery for the management of an underlying malignancy.

Maydick81 conducted a survey among 230 participants attending a conference of the United Ostomy Associations of America. Of the 140 participants who met inclusion criteria and provided data, the majority (n = 85; 60.7%) had their stoma site marked by a WOC nurse. More than 75% of the participants received preoperative marking, with a significant difference in quality of life for patients whose stoma site was marked by a surgeon (mean difference in quality-of-life scores (M) = 7.71, standard error (SE) = 0.16) or a WOC nurse (M = 8.82, SE = 0.37) versus another professional (M = 4.83, SE = 1.05) (P = .19).

SUMMARY OF EVIDENCE

Our review indicates that approximately 40% of patients with an ostomy will experience 1 or more stoma or peristomal-related complications.29,35,37 Peristomal complications were prevalent and occurred in nearly 80% of all patients experiencing complications. Patients developing these complications had substantially higher costs of postsurgical care and suffered from considerable difficulty and distress and negative effects on HRQOL.20 Most peristomal complications appeared within first month after the surgery.67 The most frequent stoma complications were parastomal hernia, stoma prolapse, and mucocutaneous separation. With the exception of mucocutaneous separation, stoma complications were more likely to occur later than peristomal complications.

We found evidence suggesting a number of patient-related risk factors. They included nonmodifiable factors: age more than 65 years; female sex; concurrent diabetes mellitus; and IBD or cancer as the underlying cause for ostomy creation. Emergency surgery also emerged as an identified independent risk factor associated with ostomy-related complications, but we did not classify this factor as patient related.83,84 Evidence further suggests that a combination of factors present the greater risk of complications.33 For example, higher BMI is associated with a higher incidence of multiple stoma- and peristomal-related complications and in such patients it is of importance to properly mark the stoma site and have a stoma height above the level of the skin.66 Evidence concerning differences in specific complications based on fecal ostomy type was weak. Limited evidence suggests that colostomy, and especially loop colostomy, is associated with an increased risk of prolapse and parastomal hernia.10,11,17 The strongest evidence supports creation of a loop ileostomy versus a loop colostomy because it is associated with a reduced likelihood of stoma-related and systemic complications.38,85,86 Finally, we found evidence suggesting preoperative stoma site marking by a WOC or ostomy nurse specialist acts as a protective factor reducing the likelihood of peristomal complications and impaired HRQOL.

GAPS IN EVIDENCE

Although the presence of comorbid conditions such as diabetes mellitus was found to be associated with an increased likelihood of stoma and peristomal complications, we found no studies that provided the direct evidence regarding association between the presence and grade of comorbid conditions based on a cumulative risk scale such as the Charlson Comorbidity Index and development of the stoma or peristomal complications. We also observed lack of a unified grading system for reporting severity of stoma or peristomal complications.

LIMITATIONS

Although 4 RCTs, Cochrane systematic reviews, and systematic reviews were included in this analysis, most evidence in our review was retrieved from observational studies or uncontrolled case series with a high risk of bias, limited follow-up, and small sample sizes. Use of the GRADE instrument for study bias revealed that the majority were graded as low or very low, supporting our conclusions regarding a paucity of high-quality studies of stoma and peristomal complications among patients living with a fecal ostomy.

CONCLUSION

We completed a scoping review and found research related to 5 factors associated with an increased likelihood of stoma or peristomal complications: (1) age more than 65 years; (2) female sex; (3) BMI more than 25; (4) diabetes mellitus as a comorbid condition; and (5) abdominal malignancy as the underlying reason for ostomy surgery. We found a single protective factor: stoma site marking and care by a WOC or ostomy nurse specialist. Research indicates that up to 93% of all ostomy surgical services incorporate a WOC or ostomy nurse specialist as part of their health care team.87 Follow-up of the patient by a WOC nurse and a physician should include both the immediate postoperative period and over time because ostomy complications that may need surgical intervention develop in the late postoperative period, mostly within the first 6 months after stoma creation. Further research is needed to improve the evidence-based surgical decision-making regarding the type of stoma and its placement as well as prevention and management of stoma and peristomal complications.

ACKNOWLEDGMENT

The authors wish to thank Michal Mik for comments when preparing the manuscript. This research is partially supported by the European Social Fund (ESF) funded by the National Center for Research and Development (grant no. POIR.04.01.01-00-0066/18).

REFERENCES