BACKGROUND: Stomal mucocutaneous separation (SMS) is a serious and common short-term and long-term complication of ostomy surgery, but optimal methods to help prevent it have not been established. The authors hypothesized that seromuscular-dermal (SM-D) suturing may be better than all layer-dermal (AL-D) suturing to help prevent SMS. METHODS: This retrospective study evaluated the short-term SMS rate of patients who underwent colostomy or ileostomy surgery between 2015 and 2019. Patient demographics, medical and surgical history variables, as well as SMS outcomes were abstracted. Postoperative SMS severity was categorized by extent as follows: grade A (mild), grade B (moderate), and grade C (severe). RESULTS: In total, 105 patients (AL-D group, 45 patients; SM-D group, 60 patients) were enrolled in the study. SMS occurred in 24 patients (23%). The overall SMS rate was 18% (n = 11) in the SM-D group and 29% (n = 13) in the AL-D group (P = .202). The occurrence of severe (grade C) or moderate and severe SMS (grades B + C) in the SM-D compared with the AL-D group was significantly different (2% vs 16% [P = .011] and 10% vs 24% [P = .047], respectively). A history of steroid use was a risk factor for moderate and severe SMS (P = .016, odds ratio 5.694). Stomal height ≤1 cm was a a risk factor for all cases of SMS (P = .037, odds ratio 2.650). AL-D suture technique and a history of steroid use were independent risk factors for severe and moderate plus severe SMS (P = .021, odds ratio 12.844 and P = .027, odds ratio 4.808, respectively). CONCLUSION: In this study, use of the AL-D suturing technique and a history of steroid use were independent risk factors for the short-term development of moderate or severe SMS. Patients whose stoma was secured using the SM-D technique had a significantly lower rate of moderate or severe SMS.

Stoma creation is commonly performed in patients with colorectal cancer, inflammatory bowel disease, trauma, and other types of bowel obstruction.¹ Stoma creation is a simple undertaking. However, short-term and long-term ostomy complications, including stomal necrosis, retraction, prolapse, stenosis, hernia, and mucocutaneous separation, can occur.^2-4 In a systematic review of randomized controlled trials, Malik et al³ reported high rates of stoma-related complications: 14.3% following loop ileostomy, 26.3% for loop colostomy, and 62.6% following end colostomy surgery. These complications negatively affect satisfactory recovery after surgery.

Stomal mucocutaneous separation (SMS) refers to the separation of the ostomy from the peristomal skin, which is a common short-term complication of stoma creation. It occurs in 12% to 24% of patients early in the postoperative period.^4-6 SMS is associated with parastomal skin irritation, which causes fecal leakage due to problems with stoma appliance adherence. One cause of SMS is excessive traction between the skin and stoma.⁷ During abdominal surgery, closure of the midline wound can increase traction forces at the stomal-cutaneous junction.⁸ Miyo et al⁹ examined risk factors associated with SMS by surveying their loop ileostomy cases in a retrospective study. They found that long skin incisions and low distal stoma limbs were significant risk factors for SMS. However, only a few reports have mentioned an appropriate procedure of stoma creation that helps prevent SMS.¹⁰

Suturing the stomal intestinal edge to the skin is an essential part of stomal creation, creating a tight stomal-cutaneous junction. This is performed using either of the following suturing procedures: suturing all the layers on the intestinal side or suturing only the seromuscularis layer of the intestinal wall. In the former method, all layer-dermal (AL-D) suturing confers tight adherence because substantial amounts of intestinal tissue are sutured.¹¹ In the latter, however, seromuscular-dermal (SM-D) suturing provides proper tissue adaptation because the intestinal seromuscularis is fixed to the dermis. Most surgeons use AL-D suturing (including all layer-transcutaneous suturing) because this is easier to perform.¹² The Japanese Manual of Intestinal Stoma Construction estimated that only 30% of surgeons use SM-D suture.¹³ In a review article, Kwiatt and Kawata⁸ stated that suturing the full thickness of the intestinal wall was important for preventing SMS; however, the current authors frequently encounter patients with SMS despite using the AL-D suture. 

The authors hypothesized that SMS was related to the suturing method used in the stomal-cutaneous junction and that SM-D suturing might be better than AL-D suturing. Thus, the aim of this study was to examine the effects of SM-D and AL-D suturing on the rate of short-term SMS. The authors included only bowel diversions in this study because urinary diversions were predominantly performed by urologists in the authors’ hospital.

Patients. A retrospective study design was used to examine outcomes of consecutive patients who underwent colorectal surgery with stoma creation at the University of Tsukuba Hospital from 2015 to 2019. The clinical data were retrieved in 2020. The suturing method employed was historically changed as follows: AL-D suturing was performed from April 2015 to April of 2017, whereas SM-D suturing was performed from May 2017 to February 2019. The reason for this change was that SMS had occurred in several cases using the AL-D procedure; therefore, the authors attempted to modify the suturing method for better outcome. Both elective and emergency surgeries were included. The cases were intraoperatively supervised by Y. Ohara, T. Enomoto, and Y. Owada to ensure standardization of the surgical technique. These surgeons were educated on the protocol. Patients who underwent surgery by non–protocol-educated surgeons were excluded. This study was approved by the ethics committee of Tsukuba Clinical Research and Development Organization, University of Tsukuba Hospital. Informed consent for surgery was obtained from all patients. 

Method of stoma creation. All patients, including those who underwent emergency surgery, had preoperative stoma site marking. After intestinal maneuver and/or resection, a round skin incision was made, followed by an appropriate aperture creation in the abdominal wall through the rectus muscle. A segment of the intestine was advanced through the abdominal wall and fixed there to keep a stomal height of 2 to 3 cm. The intestine was opened using electrocautery. Three-point suturing (skin, intestinal edge, and intestinal wall) was placed to attach the intestine to the abdominal wall, keeping stomal height (4–6 sutures). Two-point suturing (skin and intestinal edge) was performed to fix the mucocutaneous junction (12–16 sutures). Additional sutures were placed, depending on the stomal size. The differences in suturing procedures of AL-D and SM-D are shown in Figure 1. Briefly, in the AL-D suture, the needle is threaded through the full thickness (mucosa and seromuscularis) of the intestinal edge (Figure 1A). In contrast, in the SM-D suture, the needle is threaded only through the seromuscularis layer of the intestinal edge (Figure 1B). All sutures are buried sutures, that is, they did not go through the epidermis. Monofilament absorbable strings 4-0 (Monodiox; Alfresa Pharma, Osaka, Japan) were used.

Postoperative stomal care and evaluation of SMS. Stomal care was standardized by the surgeons and the Society of Wound, Ostomy, and Continence Nurses. High-quality photographs of the stoma were taken and collected in the record by the nurses at the time of stomal care. The primary outcome of the study was the occurrence and severity of SMS based on the medical records and photos. 

Postoperative SMS severity was categorized into the following 3 grades by Y. Ohara according to SMS proportion: grade A (mild), grade B (moderate), and grade C (severe) (Figure 2). Grade A was defined as SMS less than a quarter of a circle, Grade B as SMS between a quarter to half of a circle, and Grade C as SMS more than half of a circle. SMS was surveyed during the patients’ hospital stay as the focus of the study was on the early complications of stoma creation.

Clinical data collection. All clinical data were collected from the electronic medical records of the University of Tsukuba Hospital. The abstracted variables included patient demographics (age, sex, body mass index), surgical and medical history (reason for surgery, emergency or planned procedure, stoma type, history of chemotherapy, radiation therapy, steroid use, diabetes mellitus, and length of hospital stay), and occurrence and extent of SMS. 

Statistics. Statistical analyses were performed as descriptive statistics using the chi-square test with Fisher’s exact test, and univariate or multivariate analysis by multiple logistic regression analysis using IBM SPSS Statistics Version 25 (IBM Japan, Tokyo). Differences at P < .05 were considered statistically significant. All factors included in the multivariate analysis were selected if the P value in the univariate analysis was P < .3.

Patient characteristics. In total, data from 105 patients were extracted for the study (45 patients in the AL-D group, and 60 patients in the SM-D group). The majority of patients in both groups were aged > 65 years, were male, had a body mass index < 25 kg/m², and had colorectal cancer. Demographic, history, and stoma characteristics (Table 1) did not differ significantly between the AL-D and SM-D groups. The average length of hospital stay was 30 days (SD = 18) and median length of stay was 26 days in the AL-D group; average length of hospital stay was 29 days (SD =17) and median stay was 25 days in the SM-D group.

SMS occurrence and severity. SMS (grades A + B + C) occurred in 24 of 105 patients (23%) and the incidence was lower in the SM-D group (11 patients, 18%) compared with the AL-D group (13 patients, 29%), but the difference was not significant (P = .202) (Table 2). However, when comparing the rate of severe SMS (grade C) with moderate to severe SMS (grades B + C), SM-D suturing dramatically decreased SMS occurrence versus AL-D suturing (2% vs 16%, P = .011 and 10% vs 24%, P = .047, respectively).

Risk factors of SMS. Univariate analysis for risk factors associated with the occurrence of Grade C as well as Grade B + C SMS showed that a history of steroid use was a significant risk factor for Grades B and C SMS (P = .016, odds ratio 5.694, and 95% CI 1.502-21.584) (Table 3). AL-D suturing was a significant risk factor for grade C SMS (16% vs 2%, P = .020, odds ratio 10.870, and 95% confidence interval 0.011-0.778) as well as grade B + C (24% vs 10%, P = .047, odds ratio 2.915, and 95% confidence interval 0.116-1.015). Stomal height ≤ 1 cm was a significant risk factor in all cases of SMS (33% vs 16%, P = .037, odds ratio 2.650, and 95% confidence interval 1.044-6.728). 

Multivariate analysis showed that AL-D was an independent risk factor of SMS in grade C (P = .021, odds ratio 12.844, and 95% confidence interval 1.464-112.703). Steroid use also was an independent risk factor of SMS in grades B + C (P = .027, odds ratio 4.808, and 95% confidence interval 0.052-0.835) (Table 4). 

Other stoma-related complications. Stoma fistulation occurred in 1 patient in the AL-D group, and stoma stenosis occurred in 1 patient in the SM-D group. No prolapse, retraction, or necrosis occurred in any patient.

This retrospective study compared the effect of SM-D and AL-D suturing technique on the risk of SMS, which is a frequent short-term complication associated with stoma creation.¹⁴ Although the overall difference was not statistically significant, patients whose stoma was secured using the AL-D suturing technique were almost 3 times more likely (odds ratio 2.95) to experience severe or moderate SMS than those whose stoma was sutured using SM-D technique. 

Although stoma creation is a common surgical procedure, a standard method has not been defined clearly. Moreover, optimal suturing for the intestinal edge and skin has not been established. Suturing procedures also vary across surgeons, hospitals, and countries. Uchino et al¹⁰ focused on skin suturing (buried [dermal] suturing versus transcutaneous suturing) for stoma creation in patients with ulcerative colitis (n = 378). They demonstrated that the use of buried suturing was associated with lower SMS occurrence. Sier et al¹² focused on intracutaneous versus transcutaneous suturing to avoid skin irritation and fecal leakage. The results of their randomized clinical trial showed that intracutaneous suturing was superior to transcutaneous suturing. However, only a few studies have assessed the optimal procedure of suturing the stomal intestinal edge.¹⁰ 

Two (2) key factors are important for preventing SMS: mechanical strength and wound healing. Mechanical strength is necessary to maintain a robust attachment between the skin and the intestine. The AL-D suture may improve mechanical strength because a substantial amount of intestinal tissue is held in a sutured string. Wound healing is a fundamental factor in postsurgical care that may be inhibited by steroid use.¹⁵ The use of steroids was shown to be a risk factor for SMS in the current study. The authors considered that wound healing in stomal maturation with SM-D suturing may be similar to intestinal anastomosis with the Gambee suture.¹⁶ Hirata et al¹⁷ compared Gambee anastomosis with Albert–Lembert anastomosis using a dog model. The Gambee anastomosis, which connected layer to layer of intestinal wall (similar to the SM-D suture), showed good tissue healing and angiogenesis on histopathological study. On the other hand, Albert–Lembert anastomosis, which threaded through all layers (mucosa and seromuscularis) of the intestinal wall (similar to the AL-D suture), showed mucosal deficiency and disconnection of the intestinal layer. In the current study, SM-D suturing was considered reasonable for better wound healing. The seromuscularis and dermis were adequately attached together, which provided sufficient tissue angiogenesis and repair. In addition, Krasniqi et al¹⁸ demonstrated that Gambee anastomosis showed advantages in mechanical cohesiveness and histological healing in their rat model. Therefore, SM-D suture might provide both wound healing and mechanical strength on the stomal mucocutaneous junction. In contrast, AL-D suturing might cause inadequate layer formation. The mucosa could collapse into the dermis, or the epidermis could collapse into the seromuscularis. The mucosal surface or the epidermis itself cannot adhere to other tissues. This unfavorable disposition of layers possibly hinders mucocutaneous junction wound healing. Thus, the skin may become irritated or local infection may worsen, finally creating SMS.

The overall risk of SMS was not lower with SM-D suturing, but the risk of severe SMS was decreased significantly. These data suggest that even if SMS occurred in cases in which SM-D suturing was used, the SMS was not extended. The authors believed that substantial wound healing obtained by the SM-D suture might resist enlargement of the mucocutaneous gap. Moreover, the authors assumed that the nonsutured mucosal layer of intestine might cover the SM-D junction, protecting it from fecal contact that could worsen SMS. The authors hypothesized that in AL-D sutured stoma, once a small SMS occurred, it would easily destroy adjacent stitches due to insufficient tissue healing between the intestine and skin. Feces would invade into the mucocutaneous gap, and SMS would be extensive. Severe SMS is difficult to treat and requires long periods until good granulation covers the mucocutaneous gap.⁸ Patients with severe SMS undergo intensive postoperative stomal rehabilitation, including skin care for irritation and adjustment for fecal leakage. Thus, prevention of severe SMS by using optimal suturing techniques can profoundly affect patient outcomes. 

Steroid use also was found to be a risk factor for SMS. Barr⁷ reported that SMS occurs more commonly in patients who are immunocompromised and that corticosteroid therapy results in superficial infection and poor healing in these patients. As Miyo et al⁹ demonstrated, low stomal height is also a risk factor for SMS. Keeping stomal height > 1 cm is necessary to direct the effluent into the pouch, which is an important factor for preventing peristomal skin problems.⁵  

This study has several limitations. First, this was a single center retrospective study. Time and selection bias should be considered. A randomized study should be performed in the future. Additional studies are also recommended; for example, studies regarding patients’ ability to learn self-care and economic studies comparing the 2 techniques. Second, this study analyzed various types of patients. Although the authors demonstrated several factors associated with SMS occurrence, SMS could also be influenced by an individual patient’s physical condition (eg, malnutrition), which could not be fully assessed in this study. The authors could not demonstrate the differences of length of hospital stay between the 2 groups, because patients underwent various type of clinical care (eg, chemotherapy or supportive care for cancer) after stoma creation during the same hospital stay. Third, SMS might be affected by each surgeon’s skill. In an attempt to avoid this bias, we only analyzed patient data from 3 colorectal surgery specialists who supervised all cases to minimize surgeon-related variables. Fourth, the authors focused on SMS as an early complication; therefore, the observation of SMS occurrence was performed during the hospital stay. This patient follow-up time might be too short for detecting late manifestation of SMS.

This retrospective study was conducted to examine the effect of SM-D and AL-D suturing technique on the rate of short-term SMS among 105 patients who underwent ileostomy or colostomy surgery. AL-D suturing, low stomal height (≤ 1 cm), and history of steroid use were shown to be risk factors for SMS. These results suggest that SM-D sutures should be used in stomal creation to help prevent SMS. Although SM-D suturing is complicated and difficult, the skill can be mastered to help prevent this serious complication

Drs. Ohara, Enomoto, Owada, Kitaguchi, Hisakura, Akashi, Ogawa, Takahashi, and Shimomura are surgeons and Dr. Oda is a professor, Department of Gastrointestinal and Hepato-Biliary-Pancreatic Surgery, Faculty of Medicine, University of Tsukuba, Tsukuba-shi, Ibaraki, Japan. Address all correspondence to: Yusuke Ohara, MD, Department of Gastrointestinal and Hepato-Biliary-Pancreatic Surgery, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8575, Japan; tel: +81-298-53-3221; fax: +81-298-53-3222; email: y.ohara@md.tsukuba.ac.jp
All authors substantially contributed to the manuscript. YOH, TE, YOW, and DK performed surgeries. KH, YA, KO, OS, and KT decided and approved treatments. YOH and TO were major contributors in writing the manuscript. All authors read and approved the final manuscript.
The datasets generated and/or analyzed during the study are publicly available from the corresponding author on reasonable request.