Effectiveness of Triclosan-Coated Sutures Compared with Uncoated Sutures in Preventing Surgical Site Infection after Abdominal Wall Closure in Open/Laparoscopic Colorectal Surgery

Abstract

BACKGROUND: 

Previous randomized trials have assessed the effectiveness of triclosan-coated sutures in fascia closure after midline laparotomy in preventing surgical site infections (SSIs); however, available evidence remain inconclusive. We aimed to evaluate the effectiveness of triclosan-coated sutures in abdominal fascia closure to prevent postoperative SSIs.

STUDY DESIGN: 

This study was a multicenter prospective trial conducted within 24 Japanese secondary and tertiary care centers and a propensity score (PS)–matched analysis. Patients 20 years of age or older who underwent elective surgery for colorectal cancer (CRC) were included. Between July 2016 and July 2019, 2,207 patients were prospectively enrolled into the triclosan-coated sutures or uncoated sutures groups. The per-protocol population comprised 2,195 patients. The PS matching was performed for 1,579 patients: 926 patients in the coated group and 653 patients in the uncoated group. The abdominal fascia after midline laparotomy was closed with triclosan-coated or uncoated sutures depending on group. The primary endpoint was the incidence of an SSI. Secondary endpoints were length of hospital stay and surgical complication rates.

RESULTS: 

The recorded SSI rates were 4.2% in the triclosan-coated group and 6.74% in the uncoated suture group (p = 0.028). There were no serious adverse events in the groups. The final logistic regression model showed that several variables affected the occurrence of SSI. Our meta-analysis included six phase-III trials, and our study evaluated 4,797 patients. The results show a significant superiority of triclosan-coated sutures over uncoated suture material.

CONCLUSION: 

Triclosan-coated sutures reduce the incidence of SSI after elective CRC surgery.

In Brief

Triclosan-coated sutures reduce the incidence of surgical site infection (SSI) after elective colorectal cancer surgery. The laparoscopic approach comprised 91.0% (n = 1,998 of 2,195). Our study, including meta-analysis evaluating a total of 4,797 patients, has shown the clinical effectiveness of coated sutures for SSI for the first time.

CONTINUING MEDICAL EDUCATION CREDIT INFORMATION

Accreditation: The American College of Surgeons is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.

AMA PRA Category 1 Credits^TM: The American College of Surgeons designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit^TM. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

Of the AMA PRA Category 1 Credits^TM listed above, a maximum of 1 credits meet the requirement for Self-Assessment.

Postoperative surgical site infections (SSIs) are one of the most common complications after open abdominal surgery. Based on the Centers for Disease Control and Prevention healthcare-associated infection prevalence survey, and the incidence of SSI associated with inpatient surgeries, comprehensive advances have been made in the reduction of SSI in clinical practice according to the guidelines¹; however, SSI still relates to mortality (3%) and causes associated deaths.²

The frequency of SSIs in operations involving midline laparotomy is between 12% and 16%³; however, this frequency varies depending on the definition and patient population.³ Additionally, the incidence and treatment costs may exceed approximately $3.3 billion annually and are associated with almost 1 million additional inpatient-days annually.^4,5 SSIs not only require prolonged treatment but also increase the risk of death. SSIs have been attributed to 17% of cases of postoperative mortality.^5-7

Risk factors relating to SSI have been evaluated, and well-known factors include age, patient history, and comorbidities, such as obesity, hypertension, diabetes, steroid intake, immunosuppression, smoking, and consumption of alcohol.^8-10 Recently, advances have been made to prevent SSIs, including antibiotic prophylaxis, mechanical bowel preparation, hair removal at the operation site, skin disinfection, hand decontamination, and the use of sterile gloves and gowns.¹ The most recent approach is the use of triclosan-coated material as sutures to close the abdominal fascia. In comparison with other materials and procedures which increase the incidence of the infection during surgery, this is a novel attempt to reduce the incidence of SSIs.

Triclosan interferes with microbial lipid synthesis, resulting in attenuation of bacterial growth and colonization of the suture material, and this has been validated in both in vivo and in vitro studies.^11,12 There are products available on the market for use during operations, such as triclosan-coated polydioxanone antimicrobial sutures (PDS Plus; Ethicon, Johnson & Johnson, Livingston, Scotland, UK). Six randomized trials have assessed the effectiveness of triclosan-coated sutures for fascia closure after midline laparotomy^3,13-21; however, the evidence remains largely inconclusive because of different surgical procedures and patient groups. Regarding surgical procedures, laparoscopic surgery has been a preferred approach for minimally invasive surgery and contributes to the lower incidence of SSI, even in cases involving colorectal resection.²²

The effectiveness and clinical impact of triclosan-coated sutures remains unclear. Therefore, we investigated whether a clinically relevant reduction in the incidence of SSIs can be achieved with triclosan-coated sutures compared with uncoated sutures. Our aim was to show reliable data on the effectiveness of triclosan-coated sutures for abdominal fascia closure in relation to SSIs, compared with that of uncoated sutures.

METHODS

Participants

This study was initially designed as a multicenter prospective trial and propensity-matched analysis. The study was conducted in the surgical departments of 24 Japanese secondary and tertiary care centers. Those eligible for participation were patients 20 years of age or older who underwent elective surgery for colorectal cancer (CRC), with Eastern Cooperative Oncology Group performance status 0 or 1.²³ The exclusion criteria were as follows: dirty or infected surgical sites; impaired mental state; past laparotomy designed for the predetermined surgical site; participation in another trial that would interfere with the intervention or outcome of this trial; uncontrolled diabetes; uncontrolled cardiac failure; uncontrolled cardiac angina; uncontrolled abnormal cardiac rhythm; uncontrollable hypertension; severe emphysema; severe interstitial pneumonia; or severe pulmonary fibrosis.

After being screened for these criteria, patients provided written informed consent and were enrolled in the study. Patient recruitment started on July 22, 2016. The first interim analysis was prespecified and performed in March 2018 when the primary outcome was available for 1,131 patients. The multicenter protocol of this trial was approved by the ethics committee of the Osaka International Cancer Institute, Japan, on July 22, 2016 (Reference number: 1607229068).

Procedures

A study plan including the procedures was approved by all participating institutions. Patients underwent routine preparation according to the established standards of the participating institutions. Antibiotic prophylaxis had to be completed and documented according to the recently updated Japanese guidelines of surgical infection from the Japan Society for Surgical Infection.²⁴

Patients received antibiotic prophylaxis before making the incision. The intervention included closure of the abdominal fascia after midline laparotomy with triclosan-coated polydioxanone (PDS Plus: Ethicon, Johnson & Johnson, Livingston, Scotland, UK)/polyglactin 910 (Vicryl Plus: Ethicon, Johnson & Johnson, Livingston, Scotland, UK) sutures in one group and fascial closure using uncoated sutures in the other. The difference between coated and uncoated sutures depended on the institutions. Fascial closure was achieved with the help of continuous closing technique with the use of loops or nodule suture technique. Subcutaneous drains were allowed. Skin closure was performed using sutures or using surgical skin staples.

Postoperative care and evaluation of the clinical course was observed according to the principles and standards of the participating institutions. The timepoints for the assessment of the primary endpoint were at discharge or on postoperative day 30 (whichever occurred first) and on postoperative day 90. All patients were monitored, and compliance with the protocol was observed.

All patient data were documented on case report forms. Queries about missing data were generated centrally and were subsequently sent to the participating institutions. Serious adverse events were reported to the Data Center of the Osaka University promptly.

Outcomes

The primary endpoint was the incidence of a superficial or deep SSI, according to the Centers for Disease Control and Prevention/National Healthcare Safety Network,^25,26 and Japan Nosocomial Infection Surveillance.²⁷ A superficial incision SSI was considered only if an infection fitted the description, occurred within 30 days after the operative procedure, and involved only the skin and subcutaneous tissue of the incision. The patient must also have had at least 1 of the following: purulent drainage from the superficial incision; organisms isolated from an aseptically obtained culture from the incision; at least 1 sign or symptom of infection, such as pain or tenderness, localized swelling, redness, or heat, and if the superficial incision was deliberately opened by a surgeon, unless the culture was negative; and the diagnosis of superficial incisional SSI established by an infection control team, a surgeon, or attending physician. Deep SSIs must have occurred within 30 days after the operative procedure, involved deep soft tissues, such as the fascial and muscle layers, and presented with at least 1 of the following: a purulent drainage from the deep incision, but not from the organ/space component of the surgical site; a deep incision spontaneously dehisced or deliberately opened by a surgeon, unless the culture was negative; had at least 1 sign or symptom, such as fever, pain, or tenderness; and an abscess or other evidence of infection involving the deep SSI was diagnosed by a surgeon or attending physician.

The outcome assessors analyzed the SSIs on 2 occasions within 30 days postoperatively, during follow-up: one on the day of hospital discharge or postoperative day 30 (whichever occurred first) and the other on day 30 after the operation, in the participating institutions.

Secondary endpoints were length of hospital stay and surgical complication rates.

Statistical analysis

Our sample size was calculated based on an assumed SSI rate of 7.0% in the uncoated suture group and a reduction of 3.0% in the coated suture group. In a fixed design, 1,816 patients were needed to achieve a power of 80% for the chi-square test at a 2-sided significance level of 5%. This design allowed for early termination for efficacy or recalculation of the sample size if the study was continued after the interim analysis. In the protocol, the first interim analysis was planned once the primary outcome was available for 1,131 patients. The primary analysis was based on the protocol for the registered patients to evaluate the clinical expediency of the surgical procedures using uncoated or coated suture material for fascia closure.

An imbalance in crucial covariates related to outcomes could have biased the estimation of the effect of using coated sutures. To deal with selection bias, we performed propensity score (PS) matching. The PS included effective covariates: age; sex; BMI; American Society of Anesthesiologists (ASA) physical status¹⁶; history of hypertension, hyperlipidemia, diabetes, heavy use of steroids, smoking, and drinking; other medical history; white blood cells (WBCs); hemoglobin; platelets; glutamate-oxaloacetate transaminase; glutamic pyruvic transaminase; total bilirubin; creatinine; blood urea nitrogen; albumin; C-reactive protein; surgery time; intraoperative blood loss; presence of blood transfusion; surgical approach; intraoperative complications; surgical wound contamination class; incision length; suture type (braid or monofilament); and presence or absence of drainage. Covariates were chosen for their potential association with the outcome of interest based on clinical knowledge. The PS for each patient was estimated using a logistic regression model, and we used the PS to match patients who were treated surgically with coated sutures with corresponding control patients who were treated surgically without antibacterial sutures (uncoated sutures). We applied the nearest neighbor matching without replacement, with 2-to-1 pair matching to avoid overly limiting the cases enrolled within the study period. A caliper width of 0.2 of the standard deviation of the logit of the PS score was used. We also used a standardized difference to measure covariate balance, whereby an absolute standardized difference above 0.1 represented meaningful imbalance.

Continuous parameters were presented as mean and standard deviation or as median and interquartile range. The Student’s t-test, Mann–Whitney U test, and Pearson chi-square test were used to compare data between groups. For primary endpoint analysis aimed at comparing the effect of coated sutures, the adjusted odds ratio (OR) and its 95% CI were estimated using the logistic regression model. Variables were included in the models based on the existing knowledge of risk factors for SSI. Determining whether there was heterogeneity in the effect of coated sutures, we used a logistic regression model stratified by subgroup. Subgroup analyses were previously specified in the statistical analysis plan. We planned and performed 18 subgroup analyses for the primary outcome using the logistic regression model with the following subgroups: age; sex; BMI; ASA physical status; history of hypertension, diabetes, hyperlipidemia, heavy use of steroids, smoking, and drinking; other medical history; WBC; lymphocytes; surgical approach; suture type; presence or absence of drainage; use of coated suture in the dermis; and intraoperative complications. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC). Statistical significance was assessed using a 95% CI. Analysis items with a 2-tailed p < 0.05 were considered statistically significant.

Meta-analysis

A post hoc meta-analysis was performed, including present outcomes to evaluate the expediency of triclosan-coated vs uncoated sutures to reduce SSIs after closure of midline laparotomy. To identify prospective trials that addressed this topic, we searched PubMed and the Cochrane Central Register of Controlled Trials (Central) with the following search terms: “PDS OR Vicryl OR polyglactin OR polydioxanone OR Monocryl”; “antimicrobial OR antibiotic OR antiseptic”; “triclosan OR triclosan coated”; and “randomized controlled trial.” We performed this meta-analysis with the Mantel-Haenszel random-effects model, and used R software (CRAN, R 3·6·2; https://cran.r-project.org/).

RESULTS

Between July 22, 2016, and July 16, 2019, 2,207 patients were prospectively enrolled into the triclosan-coated or uncoated sutures groups (Fig. 1). Of these, 12 were excluded from the analysis: 2 declined to participate; 7 were double registered; 1 had preoperative chemotherapy; 1 had a colon perforation before the operation; and 1 had a skin incision in the same position as a previous surgery. The intention-to-treat population consisted of 2,195 patients assigned as triclosan-coated sutures (coated, 1,399 patients) or uncoated sutures (uncoated, 796 patients) groups. Therefore, the per-protocol population consisted of 2,195 patients (Fig. 1). The groups were imbalanced in terms of patient and procedural characteristics (Table 1).

Table 1. - Baseline Characteristics

Characteristic	Before propensity score matching				After propensity score matching				Unselected groups after matching
	Coated n = 1,399	Uncoated n = 796	SMD	p Value	Coated n = 926	Uncoated n = 653	SMD	p Value	Coated n = 473	Uncoated n = 143	SMD	p Value
Age, y, mean (SD)	68.35 (11.33)	68.92 (11.29)	0.047	0.258	67.99 (11.31)	68.54 (11.56)	0.048	0.347	69.06 (11.34)	70.66 (9.79)	0.152	0.127
Sex, n (%)
Male	795 (56.83)	418 (52.51)	0.087	0.051	516 (55.72)	348 (53.29)	0.049	0.339	279 (58.99)	70 (48.95)	0.203	0.034
Female	604 (43.17)	378 (47.49)			410 (44.28)	305 (46.71)			194 (41.01)	73 (51.05)
BMI, kg/m2, mean (SD)	22.72 (3.67)	22.28 (3.55)	0.125	0.007	22.5 (3.48)	22.46 (3.52)	0.013	0.797	23.14 (4)	21.5 (3.55)	0.434	<0.0001
ASA-PS, n (%)
1	396 (28.31)	335 (42.09)	0.298	<0.0001	295 (31.86)	250 (38.28)	0.136	0.008	101 (21.35)	85 (59.44)		<0.0001
2–4	1,003 (71.69)	461 (57.91)			631 (68.14)	403 (61.72)			372 (78.65)	58 (40.56)	0.828
Medical history, n (%)
HT
Absence	881 (62.97)	508 (63.82)	0.027	0.693	585 (63.17)	411 (62.94)	0.005	0.924	296 (62.58)	97 (67.83)	0.101	0.252
Presence	518 (37.03)	288 (36.18)			341 (36.83)	242 (37.00)			177 (37.42)	46 (32.17)
DM
Absence	1,172 (83.77)	676 (84.92)	0.035	0.478	787 (84.99)	551 (84.38)	0.017	0.740	385 (81.40)	125 (87.41)	0.161	0.095
Presence	227 (16.23)	120 (15.08)			139 (15.01)	102 (15.62)			88 (18.60)	18 (12.59)
HL
Absence	1,187 (84.85)	679 (85.30)	0.018	0.774	793 (85.64)	561 (85.91)	0.008	0.878	394 (83.30)	118 (82.52)	0.027	0.827
Presence	212 (15.15)	117 (14.70)			133 (14.36)	92 (14.09)			79 (16.70)	27 (17.48)
Steroid
Absence	1,396 (99.79)	793 (99.62)	0.030	0.483	924 (99.78)	651 (99.69)	0.017	0.725	472 (99.79)	142 (99.30)	0.074	0.369
Presence	3 (0.21)	3 (0.38)			2 (0.22)	2 (0.31)			1 (0.21)	1 (0.70)
Other
Absence	968 (69.19)	578 (72.61)	0.082	0.091	627 (67.71)	469 (71.82)	0.091	0.081	341 (72.09)	109 (76.22)	0.087	0.329
Presence	431 (30.81)	218 (27.39)			299 (32.29)	184 (28.18)			132 (27.91)	34 (23.78)
Lifestyle habit, n (%)
Smoking
Nonsmoker	1,121 (80.13)	647 (81.28)	0.026	0.512	737 (79.59)	527 (80.70)	0.028	0.585	384 (81.18)	120 (83.92)	0.066	0.458
Smoker	278 (19.87)	149 (18.72)			189 (20.41)	126 (19.30)			89 (18.82)	23 (16.08)
Drinking
Nondrinker	1,128 (80.63)	643 (80.78)	0.003	0.932	742 (80.13)	529 (81.01)	0.022	0.663	386 (81.61)	114 (79.72)	0.055	0.613
Drinker	271 (19.37)	153 (19.22)			184 (19.87)	124 (18.99)			87 (18.39)	29 (20.28)
Laboratory data
WBC, mean (SD)	6.23 (1.79)	6.15 (1.73)	0.050	0.274	6.13 (1.67)	6.13 (1.71)	0.003	0.958	6.43 (1.99)	6.24 (1.8)	0.101	0.301
Neutrophil, mean (SD)	3.97 (1.53)	3.92 (1.47)	0.027	0.516	3.91 (1.44)	3.91 (1.44)	0.002	0.969	4.09 (1.7)	3.99 (1.63)	0.057	0.563
Lymphocyte, mean (SD)	1.58 (0.54)	1.65 (0.57)	0.123	0.005	1.59 (0.55)	1.62 (0.56)	0.048	0.346	1.55 (0.54)	1.79 (0.6)	0.411	<0.0001
Hb, mean (SD)	12.51 (2.04)	12.65 (2.08)	0.068	0.133	12.62 (2.01)	12.66 (2.08)	0.017	0.733	12.3 (2.08)	12.62 (2.07)	0.153	0.110
Plt, mean (SD)	27.11 (14.94)	30.77 (35.15)	0.138	0.001	26.4 (9.49)	25.76 (8.64)	0.024	0.165	28.5 (21.95)	53.69 (77.01)	0.445	<0.0001
GOT, mean (SD)	22.11 (9.23)	22.25 (9.56)	0.009	0.736	22.38 (9.8)	22.3 (9.98)	0.009	0.870	21.58 (7.97)	22.02 (7.33)	0.058	0.550
GPT, mean (SD)	18.78 (12.5)	18.45 (12.51)	0.033	0.557	18.6 (11.68)	18.57 (12.17)	0.002	0.967	19.13 (13.97)	17.9 (14.02)	0.088	0.357
Total Bil, mean (SD)	0.59 (0.27)	0.64 (0.53)	0.094	0.014	0.61 (0.27)	0.63 (0.4)	0.048	0.227	0.57 (0.28)	0.68 (0.92)	0.161	0.023
LDH, mean (SD)	195.96 (107.57)	188.39 (54.79)	0.089	0.068	197.12 (111.01)	188.09 (57.2)	0.102	0.061	193.73 (100.69)	189.74 (42.51)	0.052	0.647
Cr, mean (SD)	0.83 (0.6)	0.88 (2.82)	0.023	0.548	0.82 (0.72)	0.77 (0.21)	0.025	0.081	0.84 (0.27)	1.35 (6.63)	0.108	0.097
BUN, mean (SD)	15.06 (5.58)	14.75 (8.92)	0.048	0.321	14.93 (5.56)	14.79 (9.49)	0.020	0.701	15.31 (5.62)	14.59 (5.66)	0.127	0.185
Alb, mean (SD)	3.99 (1.28)	3.98 (0.53)	0.008	0.873	4 (0.47)	4 (0.51)	0.001	0.957	3.98 (2.1)	3.92 (0.65)	0.035	0.760
CRP, mean (SD)	0.55 (1.53)	0.48 (1.22)	0.055	0.232	0.5 (1.41)	0.48 (1.22)	0.015	0.758	0.67 (1.73)	0.49 (1.19)	0.121	0.249
Surgical factor
Operating time, min, mean (SD)	244.53 (164.16)	248.69 (96.26)	0.025	0.513	248.09 (99.82)	248.04 (97.64)	0.000	0.992	237.56 (245.39)	251.67 (89.93)	0.076	0.501
Blood loss, mL, mean (SD)	56.9 (155.58)	38.56 (101.08)	0.146	0.003	46.62 (118.83)	41.17 (107.98)	0.041	0.352	77.03 (208.34)	26.64 (59.06)	0.329	0.004
Transfusion, n (%)
Absence	1,386 (99.07)	788 (98.99)	0.00809	0.861	918 (99.14)	647 (99.08)	0.006	0.909	468 (98.94)	141 (98.60)	0.033	0.736
Presence	13 (0.93)	8 (1.01)			8 (0.86)	6 (0.92)			5 (1.06)	2 (1.40)
Approach, n (%)
Laparoscopic	1,237 (88.42)	761 (95.60)	0.281	<0.0001	863 (93.20)	624 (95.56)	0.050	0.139	374 (79.07)	137 (95.80)	0.551	<0.0001
Robot-assisted	43 (3.07)	16 (2.01)			33 (3.56)	16 (2.45)			73 (2.75)	3 (2.10)
Open	119 (8.51)	19 (2.39)			30 (3.24)	13 (1.99)			86 (18.18)	3 (2.10)
Class, n (%)
Class I and II	1,397 (99.86)	794 (99.75)	0.025	0.567	924 (99.78)	653 (100.00)	0.048	0.235	473 (100.00)	141 (98.60)	0.705	0.010
Class III	2 (0.14)	2 (0.25)			2 (0.22)	0 (0.00)			0 (0.00)	2 (1.40)
Disinfection, n (%)
Chlorhexidine	278 (19.87)	202 (25.41)	0.128	0.003	211 (22.79)	154 (23.58)	0.019	0.711	67 (14.16)	48 (33.80)	0.478	<0.0001
Popidon iodine	1,121 (80.13)	593 (74.59)			715 (77.21)	499 (76.42)			406 (85.84)	94 (66.20)
Incision length, mm, mean (SD)	5.78 (4.95)	4.57 (2.85)	0.307	<0.0001	4.85 (3.54)	4.64 (3.05)	0.052	0.216	7.59 (6.57)	4.22 (1.57)		<0.0001
Fascia, n (%)
Monofilament	916 (65.48)	717 (90.19)	0.625	<0.0001	757 (81.75)	577 (88.36)	0.166	0.000	159 (33.62)	140 (98.59)	1.887	<0.0001
Braid	483 (34.52)	78 (9.81)			169 (18.25)	76 (11.64)			314 (66.38)	2 (1.41)
Dermis, n (%)
Monofilament	1,372 (98.07)	764 (95.98)	0.282	0.0002	910 (98.27)	635 (97.24)	0.036	0.351	462 (97.67)	129 (90.21)		<0.0001
Braid	21 (1.50)	13 (1.63)			10 (1.08)	10 (1.53)			11 (2.33)	3 (2.20)
Stapler	6 (0.43)	19 (2.39)			6 (0.65)	8 (1.23)			0 (0.00)	11 (7.69)
Drainage, n (%)
Absence	1,389 (99.29)	750 (94.22)	0.279	<0.0001	917 (99.03)	643 (98.47)	0.032	0.315	472 (99.79)	107 (74.83)	0.817	<0.0001
Drainage	10 (0.71)	46 (5.78)			9 (0.97)	10 (1.53)			1 (0.21)	36 (25.17)
Intraoperative complication*, n (%)
Absence	1,379 (98.57)	794 (99.75)	0.130	0.008	920 (99.35)	651 (99.69)	0.037	0.346	459 (97.04)	143 (100.00)		0.037
Presence	20 (1.43)	2 (0.25)			6 (0.65)	2 (0.31)			14 (2.96)	0 (0.00)

^*Complication: anastomotic leakage (6 cases); bleeding (6 cases); urinary tract injury (5 cases); large intestine injury (2 cases); bladder injury (1 case); urethral injury (1 case); colon ischemia (1 case)

Alb, albumin; ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; Cr, creatinine; CRP, c-reactive protein, DM, diabetes mellitus; GOT, Glutamate-Oxaloacetate Transaminase; GPT, Glutamic Pyruvic Transaminase; Hb, hemoglobin count; HL, hyperlipidemia; HT, Hypertension; LDH, low-density lipoprotein; Lymphocyte, lymphocyte count; Neutrophil, neutrophil count; Plt, platelet count; Total Bil, total bilirubin; SMD, standardized mean difference; Steroid, heavy use of steroids; WBC, white blood cell count.

Figure 1.: Study population diagram. This study was initially designed as a multicenter prospective trial and propensity matched analysis.

The PS matching was performed for 1,579 patients (926 patients, coated group; 653 patients, uncoated group). Table 2 shows the full baseline characteristics of the trial participants, including smoking status, comorbidities, surgical approach, and ASA physical status. All procedures were performed by experienced (board-certified) surgeons. The indication for operation was colorectal open/laparoscopic/robotic surgery, excluding emergency surgery; all operations were done for the treatment of malignant diseases of the colon and rectum. Wound status was clean-contaminated in 1,566 of 1,579 (99%) and contaminated in 13 of 1,579 (0.008%).

Table 2. - Full Baseline Characteristics of the Trial Participants After Propensity Matched Analysis

Characteristic	SSI	Absence	Univariate		Multivariate
	n = 83	n = 1,496	Odds ratio (95%CI)	p Value	Odds ratio (95%CI)	p Value
Fascia, n (%)
Uncoated	44 (6.74)	609 (93.26)	Ref		Ref
Coated	39 (4.21)	887 (95.79)	0.61 (0.39-0.95)	0.028*	0.62 (0.40-0.97)	0.037*
Dermis, n (%)
Uncoated	52 (5.02)	984 (94.98)	Ref
Coated	31 (5.71)	512 (94.29)	1.15 (0.73-1.81)	0.560
Age, n (%)
<70 y	37 (4.63)	762 (95.37)	Ref
≥70 y	46 (5.90)	734 (94.10)	1.29 (0.83-2.01)	0.261
Sex, n (%)
Female	30 (4.20)	685 (95.80)	Ref
Male	53 (6.13)	811 (93.87)	1.49 (0.94-2.36)	0.088
BMI, n (%)
<25 kg/m2	55 (4.65)	1,129 (95.35)	Ref
≥25 kg/m2	28 (7.09)	367 (92.91)	1.57 (0.98-2.51)	0.061
ASA-PS, n (%)
1	24 (4.40)	521 (95.60)	Ref
2–4	59 (5.71)	975 (94.29)	1.31 (0.81-2.14)	0.272
Medical history, n (%)
HT
Absence	52 (5.22)	944 (94.78)	Ref
Presence	31 (5.32)	552 (94.68)	1.02 (0.65-1.61)	0.934
DM
Absence	61 (4.56)	1,227 (95.44)	Ref		Ref
Presence	22 (9.13)	219 (90.87)	2.10 (1.27-3.50)	0.004*	1.98 (1.18-3.31)	0.010*
HL
Absence	75 (5.54)	1,279 (94.46)	Ref
Presence	8 (3.56)	217 (96.44)	0.63 (0.30-1.32)	0.221
Steroid
Absence	83 (5.27)	1,492 (94.73)	NA
Presence	0 (0.00)	4 (100.00)
Other
Absence	57 (5.20)	1,039 (94.80)	Ref
Presence	26 (5.38)	457 (94.62)	1.04 (0.64-1.67)	0.880
Lifestyle habit
Smoking, n (%)
Nonsmoker	66 (5.22)	1,198 (94.78)	Ref
Smoker	17 (5.40)	298 (94.60)	1.04 (0.60-1.79)	0.901
Drinking, n (%)
Nondrinker	19 (6.17)	289 (93.83)	Ref
Drinker	64 (5.04)	1,207 (94.96)	1.24 (0.73-2.10)	0.425
Laboratory data
WBC, mean (SD)	6.11 (1.68)	6.6 (1.81)	1.17 (1.04-1.32)	0.010*	1.14 (1.01-1.28)	0.036*
Neutrophil, mean (SD)	3.89 (1.43)	4.2 (1.49)	1.14 (0.99-1.31)	0.059
Lymphocyte, mean (SD)	1.59 (0.55)	1.74 (0.58)	1.56 (1.09-2.22)	0.015*
Hb, mean (SD)	12.64 (2.05)	12.58 (1.94)	0.99 (0.88-1.10)	0.788
Plt, mean (SD)	26.12 (9.19)	26.38 (8.35)	1.00 (0.98-1.03)	0.806
GOT, mean (SD)	22.31 (9.94)	23.02 (8.5)	1.01 (0.99-1.03)	0.522
GPT, mean (SD)	18.6 (11.96)	18.41 (10.39)	1.00 (0.98-1.02)	0.888
Total Bil, mean (SD)	0.62 (0.33)	0.57 (0.26)	0.49 (0.19-1.25)	0.136
LDH, mean (SD)	193.92 (94.63)	183.21 (38.95)	1.00 (0.99-1.00)	0.274
Cr, mean (SD)	0.8 (0.58)	0.8 (0.25)	1.01 (0.69-1.46)	0.971
BUN, mean (SD)	14.88 (7.56)	14.68 (4.83)	1.00 (0.96-1.03)	0.806
Alb, mean (SD)	4 (0.48)	3.94 (0.57)	0.78 (0.50-1.21)	0.270
CRP, mean (SD)	0.48 (1.33)	0.72 (1.38)	1.09 (0.98-1.22)	0.122
Surgical factor
Operating time, mean (SD)
<224 min	31 (4.06)	52 (6.37)	Ref
≥224 min	732 (95.94)	764 (93.63)	1.61 (1.02-2.54)	0.041*	1.52 (0.96-2.41)	0.078
Blood loss, mean (SD)
<10 mL	53 (5.46)	745 (94.54)	Ref
≥10 mL	40 (5.06)	751 (94.94)	0.92 (0.59-1.44)	0.722
Transfusion, n (%)
Absence			Ref
Presence			1.39 (0.18-10.77)	0.752
Approach, n (%)
Laparoscopic or robot-assisted	81 (5.29)	1,449 (94.71)	Ref
Open	2 (4.08)	47 (95.92)	0.76 (0.18-3.19)	0.709
Class, n (%)
Class I and II	83 (5.26)	1,494 (94.74)	NA
Class III	0 (0.00)	2 (100.00)
Disinfection, n (%)
Chlorhexidine	15 (4.11)	350 (95.89)	Ref
Popidon iodine	68 (5.60)	1,146 (94.40)	1.38 (0.78-2.45)	0.265
Incision length, n (%)
<4 mm			Ref
≥4 mm			1.41 (0.86-2.30)	0.175
Fascia, n (%)
Monofilament	1,261 (94.53)	235 (95.92)	Ref
Braid	73 (5.47)	10 (4.08)	0.74 (0.37-1.44)	0.372
Dermis, n (%)
Monofilament	78 (5.05)	1,467 (94.95)	Ref		Ref
Braid	2 (10.00)	18 (90.00)	2.09 (0.48-9.17)	0.329	1.67 (0.37-7.56)	0.504
Stapler	3 (21.43)	11 (78.57)	5.13 (1.40-18.76)	0.014*	3.99 (1.03-15.42)	0.045*
Drainage, n (%)
Absence	82 (5.26)	1,478 (94.74)	Ref
Drainage	1 (5.26)	18 (94.74)	1 (0.13-7.59)	1.000
Intraoperative complication, n (%)
Absence	81 (5.16)	1,490 (94.84)	Ref		Ref
Presence	2 (25.00)	81 (5.16)	6.14 (1.22-30.87)	0.028*	5.17 (0.94-28.58)	0.059

^*Statistically significant

Alb, albumin; ASA, American Society of Anesthesiologists; BUN, blood urea nitrogen; Cr, creatinine; CRP, c-reactive protein, DM, diabetes mellitus; GOT, Glutamate-Oxaloacetate Transaminase; GPT, Glutamic Pyruvic Transaminase; Hb, hemoglobin count; HL, hyperlipidemia; HT, Hypertension; LDH, low-density lipoprotein; Lymphocyte, lymphocyte count; Neutrophil, neutrophil count; Plt, platelet count; Total Bil, total bilirubin; SMD, standardized mean difference; SSI, surgical site infection; Steroid, heavy use of steroids; WBC, white blood cell count.

The primary endpoint, SSI within 30 days after index operation, did not differ between the 2 groups (Table 2). The recorded rate of SSI was 39 (4.2%) of 887 patients in the triclosan-coated and 44 (6.7%) of 609 in the uncoated suture groups. The mean duration of surgery was 224 min, with a mean blood loss of 10 mL and a mean length of incision of 4.0 cm. After PS matching, the recorded rate of SSI was 39 of 926 (4.2%) patients in the triclosan-coated and 44 of 653 (6.74%) in the uncoated suture groups (OR 0.61, 95% CI 0.39–0.95; p = 0.028). No participants died during follow-up. Moreover, there were no serious adverse events in the groups.

In the final logistic regression model, several variables affected the occurrence of SSI: suture type (coated vs uncoated, OR 0.63, 95% CI 0.40–0.99; p = 0.046); diabetes (OR 1.81, 95% CI 1.07–3.07; p = 0.026); WBC (OR 1.14, 95% CI 1.01–1.28; p = 0.046); operating time (OR 1.64, 95% CI 1.01–2.67; p = 0.046); and intraoperative complication (OR 6.06, 95% CI 1.03–35.75; p = 0.047; Table 2). The secondary endpoints, length of hospital stay and surgical complication rates, did not statistically differ between the two groups (Supplemental Digital Content 1, https://links.lww.com/XCS/A76).

In the effect of coated sutures, subgroup analyses were performed for the primary outcome using the logistic regression model with the subgroups: age (younger than 70 years); BMI (less than 25); ASA physical status (2 to 4); history of hypertension (presence), hyperlipidemia (absence), and diabetes (absence); other medical history (absence); and nondrinker. These showed statistical significance, evaluated using the 95% CI (Fig. 2). Significant interactions for SSI were not observed among the 17 factors, except for drinking habit, suggesting no significant differences in the efficacy of coated sutures for these subgroups.

Figure 2.: Subgroup analyses were performed for the primary outcome using the logistic regression model with the subgroups: age (younger than 70 y); BMI (<25 kg/m²); American Society of Anesthesiologists-physical status (ASA-PS; 2 to 4); history of hypertension (presence), hyperlipidemia (absence), and diabetes (absence); other medical history (absence); and nondrinker. DM, diabetes mellitus; HL, hyperlipidemia; HT, hypertension; NA, not applicable; SSI, surgical site infection; WBC, white blood cell count.

Our meta-analysis included 6 phase-III trials, and our study evaluated 4,797 patients.^3,17-21 Six trials provided quantitative data about the incidence of SSI, 3 of which compared triclosan-coated polyglactin 910 braided suture material (Vicryl Plus, Ethicon, Johnson & Johnson) vs uncoated polyglactin 910 (Vicryl; Ethicon).^3,18,21 The other 3 trials compared PDS Plus and PDS II.^17,19,20 Three trials were performed in multicenter settings.^17,20,21 As in our study, 4 focused on colorectal surgery only,^3,17,18,21 whereas the remaining trials included a mixed cohort of patients who underwent general and abdominal surgery.^19,22 Five trials applied the definition of the Centers for Disease Control and Prevention,^3,17,19-21 as we did; 1 study did not define SSI.¹⁸ Antibiotic prophylaxis was administered to both the treatment groups.^3,17-21 The funding source was specified in 2 studies^19,20; 3 trials showed a significant reduction of SSI in the triclosan-coated group,^3,18,19 whereas 3 multicenter trials showed no significant differences between treatment groups.^17,20,21 The aggregated results of the trials, including our study, show a significant superiority of triclosan-coated sutures over uncoated sutures (OR 0.71, 95% CI 0.53–0.95; p = 0.0195; Fig. 3). Tests for heterogeneity showed no significance.

Figure 3.: Meta-analysis of the prospective trials compares the efficacy of using triclosan-coated and uncoated sutures for abdominal fascia closure after midline laparotomy in preventing surgical site infection. OR, odds ratio.

DISCUSSION

SSIs commonly occur after open abdominal surgeries. Our study shows that using triclosan-coated PDS/Vicryl Plus sutures for fascia closure during colorectal surgery reduces the rate of SSI. In our study, coated sutures reduced the incidence of SSI from 6.74% to 4.21% after PS matching; the observed reduction was 2.53%. The overall SSI rate was 5.55%, suggesting that SSI is a still a common complication and remains unsolved. BMI, medical history (diabetes and steroid medication use), preoperative data (higher WBC), and longer operating time were risk factors relating to the increase in the incidence of SSI. This prospective study shows that SSI is related to suture material after patient characteristics and surgical procedures matching.

Regarding secondary endpoints, length of hospital stay and surgical complication rates were similar in both the groups, which appear clinically relevant. An association between suture material and severe complications and high mortality was not found. Therefore, our study showed significant differences in the SSI rate without any influence on these complications.

This study was completed within the expected scale—more than 2,000 patients were enrolled within 36 months, with the inclusion of 20 participating institutions.²⁸ Our meta-analysis of preexisting randomized controlled trials (RCTs) assessing coated vs uncoated sutures for abdominal fascia closure showed heterogeneous results. Three single-center trials showed the superiority of coated sutures over noncoated sutures, whereas a multicenter trial showed that coated sutures had no advantage over noncoated sutures.¹⁷ The potential sources of bias distorting these results are small sample size, single-center setting, clinical heterogeneity, and various definitions of SSI. Our prospective study had the largest sample size focusing on elective CRC surgery (here, SSI has the worst rate in gastroenterological surgery). The rates of postoperative complications were 13.71% (127 of 926) in the coated sutures and 15.31% (100 of 653) in the uncoated suture groups (p = 0.373). Reflecting the recent surgical situation for elective CRC surgery, the laparoscopic approach comprised 91.0% (1,998 of 2,195) and robotic surgery comprised 2.69% (59 of 2,195) of the surgeries performed. Our findings show the clinical effectiveness of coated sutures in preventing SSI after laparoscopic/robotic elective colorectal surgery; this shows a significant beneficial effect of triclosan-coated sutures in the prevention of SSI after elective colorectal surgery.

We also show that efficacy leads to effectiveness in real-life clinical situations. Several in vitro and in vivo experiments showed reduced adherence of microorganisms to the local surface of coated suture material.²⁹ We proved the clinical relevance of this effect in elective colorectal surgery. Consequently, this finding delivers 2 messages to surgeons and the industry: first, the results of our trial underpin the importance of large and high-level clinical trials for valid assessment of surgical techniques, materials, and strategies in real-life situations; second, although surgical innovation partly relies on the development of new materials, to show clinical significance with clear proof of effectiveness is similar and possibly better than a RCT approach.

One trial had high internal validity because of standardization of surgical and perioperative care, adequate sample size, and masked and monitored outcome assessment because of the heterogeneous clinical care and complex assessment of SSI.²⁰ In our study, all operations were aimed at the treatment of malignant diseases of the colon and rectum and were performed by board-certified surgeons (excluding cases of emergency surgeries). Additionally, the assessment of SSI was completed and documented according to the recently updated Japanese guidelines of surgical infection from the Japanese Society for Surgical Infection.²⁴ In the context of the costs of treatment and overall healthcare, the effort to reduce the incidence of the SSI with the application of the triclosan-coated sutures and the clinical effectiveness can be appreciated because the high rate of SSI is one of the most important cost-driving factors in surgery.^3,6 One key limitation of this study was that it was not a RCT. However, real-world data should ideally provide answers to general queries presented here, such as whether coated or uncoated sutures are better for controlling SSI.

CONCLUSIONS

Our prospective study, which focused on CRC surgeries, showed the significant effect of triclosan-coated sutures in preventing SSI. Our results are similar to those of previously published RCTs.

APPENDIX

Members of the Clinical Study Group of Osaka University, Colorectal Cancer Treatment Group (CSGOCG): Masatoshi Kitakaze, MD, Department of Gastroenterological Surgery, Osaka University Graduate School of Medicine, Osaka, Japan; Yusuke Takahashi, MD, PhD, Department of Surgery, National Hospital Organization Osaka National Hospital, Osaka, Japan, and Department of Gastroenterological Surgery, Osaka International Cancer Institute, Osaka, Japan; Singo Noura, MD, PhD, Department of Surgery, Toyonaka Municipal Hospital, Osaka, Japan; Hidekazu Takahashi, MD, PhD, FACS, Department of Gastroenterological Surgery, Osaka University Graduate School of Medicine, Osaka, Japan; Chu Matsuda, MD, PhD, Department of Gastroenterological Surgery, Osaka University Graduate School of Medicine, Osaka, Japan, and Osaka International Cancer Institute, Osaka, Japan; Yuko Ohno, PhD, StemRIM Institute of Regeneration-Inducting Medicine, Division of Health Science, Graduate School of medicine, Osaka University, Osaka, Japan; Hirofumi Yamamoto, MD, PhD, Department of Gastroenterological Surgery, Osaka University Graduate School of Medicine, Osaka, Japan, and Department of Molecular Pathology, Division of Health Sciences, Graduate School of Medicine, Osaka University, Osaka, Japan; Tsunekazu Mizushima, MD, PhD, Department of Surgery, Osaka Police Hospital, Osaka, Japan.

Author Contributions

Study conception and design: Miyoshi, Fujino

Acquisition of data: Miyoshi

Analysis and interpretation of data: Miyoshi, Fujino, Kitakaze, Fuji, Ohno

Drafting of manuscript: Miyoshi

Critical revision: Miyoshi

Acknowledgment:

We thank the staff of all participating centers of the CSGOCG trial group for their outstanding engagement and support of the trial. Data collected for the study, including individual participant data and a data dictionary defining each field in the set, will be made available on reasonable request.