Applying a Silver-containing Dressing to the Incision Site and Its Effect on the Development of Surgical Site Infection After Ostomy Closure: A Prospective Randomized Clinical Pilot Study

A surgical site infection (SSI) is defined as a wound site infection that occurs within 30 days of surgery or within 1 year in the presence of implants.¹ SSIs are the most common hospital-acquired infections in surgical patients (38%), and they rank third among all hospital-acquired infections.² The risk of SSI varies depending on the patient, type of surgery, and degree of wound contamination. The degree of contamination of surgical wounds has been classified by the US Centers for Disease Control and Prevention as class I for clean, class II for clean-contaminated, class III for contaminated, and class IV for dirty.¹ After colorectal surgery, the rate of SSIs classified as clean-contaminated and contaminated wounds varies between 3% and 30%.³ Despite a consensus on the precautions to be taken preoperatively and intraoperatively in the SSI prevention guidelines, no postoperative recommendation has been made other than the necessity to cover the incision site with sterile dressing for 24 to 48 hours.² Although there is insufficient evidence concerning the use of topical antibiotics in surgical wound care, it has been stated that local antibiotic applications can reduce SSIs associated with procedures such as vertebral, joint replacement, and cataract surgeries.⁴

The powerful antimicrobial properties of silver have been used since 4000 BCE.² Today, hydrofiber wound dressings containing silver ions are used in the treatment of burn and chronic wounds. Silver ion has also been used in incision dressings after orthopedic, vertebral, genitourinary, and gastrointestinal surgeries, all of which have a high risk of SSI development.⁴ Silver is also effective against bacteria, fungi, and viruses, including methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococci, which are the most common causative agents of nosocomial infections. In contrast to antibiotics, there is a low risk of developing resistance to silver, which has both bactericidal and bacteriostatic effects.² However, it should be kept in mind that when silver is used above a certain concentration, it may cause delays in wound healing due to its toxic effect on fibroblasts.⁵

Studies have reported that moist treatment of wounds results in faster healing and lesser pain compared with the traditional dry wound care approach.⁶ Wound care products that may optimize wound healing by eliminating dead space and maintaining a moist environment while also preventing infections with their antimicrobial properties have been developed. Aquacel Ag Surgical (ConvaTec Inc.), an occlusive wound dressing with a hydrocolloid adhesive (top layer) bound to an outer polyurethane film (making it occlusive) that consists of sodium carboxymethylcellulose fibers, provides full penetration into the skin, creates a moist environment, and prevents the reproduction of microorganisms by releasing 1.2% ionic silver as an antimicrobial agent.³

In colorectal surgery, temporary ostomy is frequently used to ensure the safety of anastomosis. SSIs, with a reported incidence of 0% to 40%, constitute one of the most common complications after ostomy closure.⁷ In the literature, the technique of surgical ostomy closure has been primarily discussed to prevent postoperative SSI development.⁷ To the best of the authors’ knowledge, a postoperative surgical incision dressing method to prevent such infections has not been discussed previously.

In this study, the effects of a silver-containing hydrofiber dressing (Aquacel Ag Surgical) and a standard wound dressing on the development of SSIs at the abdominal incision site after ostomy closure were evaluated in a prospective, randomized manner. The primary goal of the study was to achieve healing without SSI. The secondary goal was the identification of factors influencing the occurrence of SSI.

This study was initiated after obtaining approval from the ethics committee of Hitit University (November 14, 2017, under approval no. 2018-33). The inclusion criteria were age older than 18 years, healthy peristomal skin, at least 5 days of hospitalization, and weekly visits for 30 days of follow-up postoperatively. Exclusion criteria included immunosuppressive disorders, diseases affecting wound healing, current steroid treatment, associated intervention during ostomy closure, and ineligible for follow-up (eg, resident of another city). The study was initiated in April 2018 and ended in May 2019. All participants were informed about the study, and written consents were obtained. For the power analyses for the assessment of the number of patients for each group, 10 patients were required for 80% confidence and 16 were required for 95% confidence. The study aimed to have 32 patients in total (16 patients for each group).

Patient data. The patients’ demographic data, comorbidities (diabetes mellitus, hypertension, coronary artery disease, cardiac insufficiency, and malignancy), ostomy type, reason for ostomy, and presence of chemotherapy/radiotherapy were recorded. Perioperative fever, arterial blood pressure, and pulse rate were examined 4 times a day. However, blood biochemical analysis and complete blood count were undertaken once preoperatively, once after surgery on the ward, and the day after surgery. If there was a pathologic finding, the variables were examined more frequently as required. The parameters were directly recorded if more than one parameter existed, whereas peak values were recorded for fever, pulse rate, and arterial blood pressure. For blood samples, pathologic values were recorded and tests were applied at least twice a day when required. Postoperative intensive care requirement, postoperative transfusion requirement, mechanical ventilator requirement, and length of stay were also evaluated as factors affecting the development of SSI.

Patients were randomized as “every other patient” in the operating room (to prevent preoperative factors from affecting scheduling and consequently randomization) and divided into two groups based on the type of wound dressing. The surgery team did not know the type of dressing that would be used until the end of the surgery.

There were 37 eligible patients who underwent temporary ostomy closure at Ankara Numune Training and Research Hospital and Hitit University Medical Faculty Hospital. All patients were administered 1 g of ceftriaxone (third-generation cephalosporin) 30 minutes before surgery as prophylaxis. Preoperative skin antisepsis was performed in the same way in both groups by applying 10% povidone iodine solution twice using the wiping and drying technique. Primary skin closure was performed for all patients.

In the control group, the wound area was covered with a standard sterile gauze dressing for 24 to 48 hours, and then wound cleaning was performed with 10% povidone iodine, followed by daily dressing replacement with sterile gauze for 5 days. The patients in the study group were treated with a silver-containing hydrofiber dressing, which was not changed for 5 days following application in the operating room (Figure).

After the fifth day, the wound site was left open in patients without SSI development. Thereafter, the rate of development of SSIs was compared between the dressing groups. SSI development was followed-up based on the US Centers for Disease Control and Prevention definition of SSI by the same senior surgeon. Accordingly, the presence of serous and/or purulent discharge or wound dehiscence at the wound site within 30 days after surgery was considered an SSI. Vital sign measurement and leukocyte count analysis were performed while the patients were in the hospital. No further vital sign measurements and leukocyte count analyses were done on an outpatient basis unless required. 

Statistical analysis. After the completion of the study, all recorded data were uploaded to the statistical program and divided into the study and control groups based on the dressing used. Numerical data were given as percentages. Mean ± standard deviation values were used to define the normally distributed continuous variables, and non-normally distributed data were expressed as median (range) values. Comparisons between the groups with and without SSIs were undertaken using the chi-square (for parametric values) and Mann-Whitney U (for comparing the means) tests. P < .05 was considered statistically significant.

Five of 37 patients required an associated intervention (stump revision, re-anastomosis) during ostomy closure and were excluded. One patient in the control group was lost to follow-up and, therefore, excluded from the study. All remaining patients (n = 31) were followed up for 30 days with weekly visits postoperatively. The mean age of the 31 patients was 59.52 years (range, 19–77 years). Twenty-three patients (74.2%) were male and 8 (25.8%) were female. Sixteen patients (51.6%) had undergone an ileostomy and 15 (48.4%) had undergone a colostomy. Eighteen patients (58.1%) had a single or multiple comorbidities (Table 1). Among the causes of ostomy, rectal cancer ranked first (n = 16; 51.6%) (Table 2). There was no difference between the study and control groups in terms of age, sex, body temperature, leukocyte count, type of ostomy, existence of malignancy and comorbidity, transfusion requirement, and chemoradiotherapy (Table 3).

SSIs developed in 4 of 31 patients (12.9%) (Table 4), and all 4 patients were in the control group. Although there was no difference in age or sex between patients with and without SSIs, mean body temperature was found to be significantly higher in the SSI group (38.0˚C ± 0.82˚C) than in the group without SSIs (36.96˚C ± 0.35˚C) (P = .015). Malignancy; chemotherapy, radiotherapy, or chemoradiotherapy; postoperative intensive care; and ostomy type were not found to have any significant effect on the development of SSIs (Table 5). However, when the dressing methods applied to the patient groups with and without SSIs were compared, SSIs developed at a higher rate in the control group (n = 4; 26.7%) compared with the study group (n = 0); this result was statistically significant (P = .043) (Table 4). The length of hospital stay was significantly higher in patients with SSIs than in patients without (14 ± 1.63 days vs 4.96 ± 0.85days) (P = .000) (Table 5). There were no patients with a single comorbidity who had an SSI. However, the presence of two or more comorbidities significantly affected SSI development (n = 4 [57.1%] vs n = 3 [42.09%]; P = .023) (Table 6).

Patients infected with Escherichia coli had wound evisceration, which primarily closed after 1 week of parenteral antibiotic delivery and local antimicrobial treatment. Cultured microorganisms and their correlation with the study parameters are shown in Table 7. Patients with coronary artery disease had significantly more SSIs (2 of 4 patients, P = .002). S aureus was isolated from 1 patient with volvulus and 3 patients with rectal cancer.

Whereas length of stay among patients who did not have an SSI was 4.96 days, E coli infection led to the longest stay (16 days), followed by Staphylococcus epidermidis for 14 days, and S aureus for 12 days (4 of 31 patients; mean, 14 days). Patients with SSI had longer length of hospital stay (4 of 31 patients; mean, 14 days) than patients without SSI (27 of 31 patients, mean, 4.96 days) (P = .000). E coli infection consequently led to wound dehiscence, deepening to evisceration, which in turn required reoperation for primary wound closure.

Surgical site infections have high mortality and morbidity rates and are seen in approximately 2% to 5% of all hospitalizations.⁴ These infections have been reported to be the most common cause of unexpected rehospitalizations and increased costs due to prolonged wound care, treatment, and hospital stay.⁸ Thus, many studies have been conducted in areas such as antibiotic prophylaxis, preoperative skin preparation, and surgical wound management; moreover, methods to reduce the risk of SSIs have been discussed in detail.^9-14

The validity of surgical wound dressings has been questioned in a number of studies, and it has been concluded that they are unnecessary, time-consuming, and do not reduce the rate of wound infection.^9,15,16 However, historically, to prevent the development of SSIs, guidelines that have been published since 1999 recommend dressing the wound site after surgery.¹ As a best practice, the UK National Institute for Health and Clinical Excellence¹⁷ recommends that all surgical incisions be closed for 3 to 5 days postoperatively, preferably with semi-adhesive, transparent polyurethane dressings.⁹ In the SSI guideline updated by the American College of Surgeons and Surgical Infection Society in 2016, it is stated that the incision site should be closed for up to 48 hours, although the time and shape are not clearly specified.⁴

One of the most important factors affecting postoperative results is the healing of the surgical wound. Early wound healing reduces the risk of infection and consequent complications like dehiscence, eventration, and evisceration. Bacterial contamination negatively affects wound healing. Studies have been conducted to investigate the use of topical antibiotics to prevent bacterial colonization in the wound area, but the results are not adequate or significant in showing that these agents truly reduce the development of SSIs.⁴ On the other hand, studies using wound dressings designed to utilize the bactericidal effect of silver ions in surgical procedures with a high risk of SSI development have reported that these dressings reduced the development of infections, although the number of cases was limited.^2,9,18

More comprehensive research on the surgical wound-healing process and developing technologies has led to the use of advanced dressing materials. Wound dressings that can absorb excess exudate to prevent maceration, remove bacteria from the wound and prevent their proliferation, and provide a barrier against the external environment have been developed. New-generation dressings, defined as total occlusive ionic silver–containing dressings, have been shown to reduce bacterial colonization when used in surgical incisions.⁹ However, there are limited data in the literature to prove that these dressings reduce the development of SSIs.⁴ The current study aimed to add to the relevant data regarding the use of silver-containing total occlusive dressing on ostomy closure wounds.

In a previous study, when the key words “surgical site infection,” “ostomy,” and “ostomy closure” were searched on PubMed, the results included studies indicating that the rate of SSI development increased up to 40% after ostomy closure⁷; however, no study was found to show the effect of wound dressing on the development of SSIs. Therefore, our study is the first in the literature to show that the development of SSIs can be prevented with the use of silver dressings in ostomy closures. We observed no wound infection within 30 days after the operation in the study group. By contrast, wound infections
developed in 26.7% of control group patients. The difference in the incidence of infections between the two methods was statistically significant.

Patient-related and non-patient factors play a role in the risk of SSI development. Accordingly, certain populations and procedures constitute a higher risk of SSIs. In the prevention of SSIs, the Surgical Infection Prevention Collaborative of the Centers for Medicare and Medicaid Services has focused on colorectal surgery, abdominal hysterectomy, vaginal hysterectomy, cardiac surgery, vascular surgery, and knee and hip arthroplasty.² In these types of surgery in which SSIs are commonly observed, silver dressings have been used for postoperative wound dressing. It has been shown that these dressings reduce the development of SSIs, especially in colorectal (from 30% to 13%), orthopedic (from 23.7% to 7.9%), spinal (from 2.30% to 0%), and cardiac (from 13% to 0%) procedures.² According to a report of the National Healthcare Safety Network (NHSN), higher rates of SSIs are seen in rectal (3.47%–26.67%), colon (3.99%–9.47%), and small bowel (3.44%–6.75%) surgeries.² According to the European Institute of Oncology data, the average incidence of SSIs after elective colorectal cancer surgery is 23%.² The SSI rate in the current study was 12.9% for the whole group, 15.8% (3 of 19) among patients with cancer, and 9% (1 of 11) among patients with no malignancy.

The SSI guideline published by the American College of Surgeons recommends the purse-string suture method in ostomy closures to prevent the development of SSIs.⁴ However, postoperative dressing style was not defined in the recommendation.⁷ In the same study, the participants were followed-up as outpatients. A total of 113 patients were included. Among them, 55 had their wounds closed using conventional primary suturing and 58 by purse-string suturing. Of the 55 patients who had conventional primary suturing, there was a 27% SSI rate, which was similar to our finding (4 of 15 patients; 26.7%). In the purse-string group, the SSI rate was 16% and their dressing choice was not defined. However, in our study group, there was no SSI (0 of 16 patients). Also in our study, the overall SSI was 12.9% (4 of 31 patients). The purse-string approach increases the length of stay, is time-consuming, and requires a second surgery. Based on our literature search, no study has compared primary closure and purse suturing regarding silver-containing wound dressing applications. However, the results we obtained encourage the use of silver-containing dressing to decrease SSI rates.

It is well-known that the frequency of dressing change affects wound healing. While frequent removal and re-application delay healing,¹⁹ the exudate excess also causes peri-wound damage.²⁰ The wound infection itself increases wound exudate, which consequently requires frequent wound dressing change. In the current study, the study group received a one-step occlusive dressing, which had long-acting silver prophylaxis. The occlusive manner prevents detrimental environmental factors in wound healing, and silver serves as a sufficient prophylactic antibacterial agent. The control group had conventional wound care that consisted of daily dressing change and limited antibacterial prophylaxis.

Comorbidities play an important role in the development of SSIs. The presence of immunosuppression, impaired glucose regulation, and cardiac, pulmonary, and vascular pathologies that impair tissue oxygenation increase the risk of SSI development.⁴ Based on our results, having coronary artery disease increases the risk of SSI, as half of the patients had SSI. However, this factor was not significant, most probably due to the small number of patients in the study group. The presence of more than one comorbidity significantly increased SSI development; however, a single comorbidity did not play a role in the development of SSIs.

Badia et al²¹ found that postoperative SSIs increased the length of hospital stay. Similarly, the current study showed that the development of SSIs significantly increased the length of stay.

The limitations of this study include the small patient population. Studies with larger patient populations should be conducted. Another limitation of the study is its heterogeneity in terms of the type of ostomy closure surgical technique and the reason for ostomy, which also requires further studies including large patient populations to assess and support the current results regarding the effects of ileostomy or colostomy closure. As a one-step approach, the occlusive wound dressing with and without silver should be investigated in further studies.

This prospective randomized clinical pilot study was conducted to address the effectiveness of ionic silver–containing occlusive dressings in the prevention of SSI. The data from 32 patients (n = 16 in the control group; n = 16 in the study group) showed that the use of a wound care product containing ionic silver reduced the rate of SSIs after ostomy closure, thus preventing additional surgical procedures related to infection and increased length of hospital stay. There is a need for multicenter randomized clinical studies involving more patients to reach a consensus on the type of wound dressing, specifically dressings with and without silver.

Tezcan Akin, MD¹; Murat Kendirci, MD²; Ali Emre Akgün, MD¹; Erdinç Çetinkaya, MD¹; Sadettin Er, MD¹; Merve Akin, MD¹;
and Ahmet Çinar Yasti, MD³

¹Department of General Surgery, Ankara City Hospital, Ankara, Turkey. ²Hitit University, Medical School, Department of General Surgery, Çorum, Turkey. ³Health Sciences University, Medical School, Department of General Surgery, Ankara, Turkey

Address all correspondence to: Merve Akin, MD, Üniversiteler Mahallesi 1604, Cadde No: 9, Çankaya/ANKARA; email: merveakin.2002@gmail.com.