6.11Closure methods

Bibliographic detailsStudy type and evidence levelStudy detailsPatient characteristicsIntervention and comparisonsOutcome measures, follow-up and effect sizeComments
Anderson 2004197Systematic review

EL = 1+
Five trials reported wound infection outcomes (Allaire 2000, Cetin 1997, Chelmow 2002, Magann 2002 and Naumann 1995).

Allaire 2000 (USA 1995–97)
Included: 76 women undergoing caesarean section and > 2 cm subcutaneous fat (26 closure gp, 50 non closure gp)
Wounds assessed prior to discharge and @ 7–10 days post-partum

Cetin 1997 (Turkey 1995–97)
Included: 164 women undergoing caesarean section (unclear how many women randomised to each treatment gp)
Wounds assessed during hospital admission

Chelmow 2002 (USA 1995–97)
Included: 327 women undergoing caesarean section ( 162 closure gp, 165 non-closure gp)
Wounds assessed 4–8 weeks post-partum

Magann 2002 (USA 1998–2001)
Included: 590 women eligible – undergoing caesarean section and >2 cm subcutaneous fat – 191 closure gp, 399 non-closure gp
Unclear when wounds were assessed

Naumann 1995 (USA 1991–93)
Included: 245 women undergoing caesarean section and >2 cm subcutaneous fat – 117 closure gp, 128 non-closure gp.
Wounds assessec at discharge and at 7–10 days post-partum
Population: Participants were women undergoing caesarean section

Five RCTs with 1348 female participants
During caesarean section:

Group 1
Sutured closure of subcutaneous fat layer (including Camper’s fascia)

Group 2
Non-closure of subcutaneous fat layer (including Camper’s fascia)
Total RR 1.02, 95% CI 0.69 to 1.50, P = 0.9 and I2 = 0.0%Funding = not stated
Study: Beresford 1993179

Setting: Canada
RCT

EL = 1−
Total no. of participants
n = 94
Intervention group
n = 48
Control group
n = 46
Inclusion: patients having elective abdominal hysterectomies

Exclusion: Other surgical procedure, not low transverse incision, prophylactic antibiotics required
Group 1
Absorbable staples

Group 2
Sutures
Wound infection not defined
Assessed at: 6 & 18 weeks
Assessor: not stated

Wound infection:
Group 1: 5/48 (11%) Group 2 446 (9%)
Centres: Single
Study: Bhatia 2002167

Setting: UK
RCT

EL = 1−
Total no. of participants
n = 31
Group 1
n = 13
Group 2
n = 18
Inclusion: 31 patients having surgery for Dupuytren’s contracture, mean age 61 years

Exclusion: Not stated Baseline comparability: Not stated
Loss to follow-up: none
Group 1
Staples
Group 2
Interrupted sutures
Wound infection defined on 1–10 scale, not clearly defined A

Assessed at: 7 & 14 days

Assessor: not stated

Wound infection:
Group 1: 0/13
Group 2: 1/18
Centres: Single

Years: 2000–2001
Buchweitz 2005156

Study Location: Germany
RCT

EL = 1+
Total no. of participants
n = 104
Group 1
n = 52
Group 2
n = 52
Group 3
n = 23
Inclusion criteria: Laparascopic surgery with at least two 5-mm ports in the lower abdomen and lacking previous abdominal operations

Exclusion criteria: intraoperative enlargement of port sites for intact specimen extraction

Age: Mean 33 ± 6.7 years

Loss to follow up: 8/60
Infection recording period: complications noted during hospital stay and afterwards a questionnaire sent before 3 months postoperative appt
Baseline comparability:no details given.
Intervention:

At least 2 port closures per patient recorded

Group 1 Subcutaneous absorbable suture 4–0 polyglactin 910

Group 2 Transcutaneous absorbable suture 4–0 polyglactin 910

Group 3
Where a third trocar wound was made, adhesive paper tape (Steri- Strip)
Outcomes

Wound infection
Study Duration: Oct 2002 – Sep 2003

Funding: Not stated
1)4/52
2)1/52
3)1/23
Wound dehiscence
1)2/52
2)0/52
3)1/23
Cameron 1987180

Setting: UK
RCT

EL = 1+
Total no. of participants
n = 301
Group 1
n = 143
Group 2
n = 141
Incision: Patients undergoing emergency and elective laparotomy by vertical abdominal incision under the care of three consultants

Exclusions:re-operations through same excision excluded

Type of wound: gastric = 31.5%, biliary = 48.3%, colonic = 38.6%, other = 35.9%

Age: Mean 61 years

Baseline comparability: Excess colonic sugery in Prolene group otherwise well balanced
Group 1
No 1 gauge polydioxanone (PDS) (;69.9% completed)
Group 2
No 1 polypropylene (Prolene) (63.8% completed)

17 patents excluded due to death or re-operation within 14 days, group assignment not reported

Interrupted mass figure-of-eight suture beginning and ending beneath rectus sheath to bury knots, skin closed with clips or nylon, most patients received subcutaneous heparin, bowel preparation and antibiotic prophylaxis according to surgeon’s usual practice. Study duration: Minimum 12 months follow- up, mean 14.7 months
Wound infection defined as: discharge of pus within one month postoperative period

Assessed at: up to 1 month post op by 1 registrar blind to method of closure

Assessor: House officers

Outcomes:
SSI rate:
Group 1 – 12/143 Group 2 – 21/141
P = 0.11

Wound dehiscence: 1/143 Group A 9/141 Group B No significant difference P=>0.05
Centres: Single
Years: dns
Cardosi 2006162

Study location: USA
RCT

EL = 1+
Total no. of participants
n = 225
Group 1
n = 77
Group 2
n = 78
Group 3
n = 67
Inclusion criteria: patients undergoing elective pelvic surgery through a vertical midline incision with presumed subcutaneous thickness of 3 cm or more

Exclusion criteria: patient unable to provide consent,>24 hours preoperative hospitalisation, laparotomy after unsuccessful laparoscopy or vaginal surgery

Age: range 28 to 84 years

Baseline comparability: details given. Age, height, weight, race, risk factors no statistical significance found.

Of seven pre-op diagnoses, two significant differences – uterine ca higher in control gp and benign ovarian neoplasm higher in suture gp.
Interventions

Gp 1 = closure of skin only

Gp 2 = approximation of Camper’s fascia with a running 00 polyglactin suture

Gp 3 = closed suction drainage of the subcutaneous space exiting through a separate stab incision

All skin closure with staples

All participants given antibiotics
Outcomes
Infection recording period: postoperatively, observations daily during hospitalisation the at outpatients at 2 and 6 weeks

Cellulitis defined as erythema and tenderness requiring antibiotic treatment.
Gp 1 = 4/77
Gp 2 = 3/78
Gp 3 = 1/67

Wound disruption defined as an open (spontaneous or iatrogenic) postoperative wound

Gp 1 = 9/77
Gp 2 = 678
Gp 3 = 10/67
Study duration: July 2000 to June 2002

Funding: not stated
Study: Carlson 1995168

Setting: USA
RCT

EL = 1+
Total no. of participants
n = 225
Group 1
n = 112
Group 2
n = 113
Inclusion: Patients undergoing laparotomy through vertical midline incision, elective and emergency procedures (clean and clean- contaminated wounds),
Excluded: patients with less than 2 years life expectancy, peritonitis and pre-existing ventral hernia. Type of wound: as above

Age: dns

Baseline comparability: Procedure status (elective v emergency) wound class and presence or absence of colon procedure not different between groups, no other baseline data reported
Group 1
0-looped nylom (Ethilon)
82% completed

Group 2
0-looped polyglyconate (Maxon)
72% completed

Running mass closure by Senior or Chief residents incorporating both layers of rectus sheath, technique for skin closure surgeon choice, prophylactic antibiotics for clean- contaminated cases,

Study duration: 2 years
Wound infection not defined

Assessed at: 2 and 6 weeks, 6 and 12 months, and 2 years,

Assessor:

Outcomes: Early postoperative wound complications (including infection and dehiscence), dehiscence

SSI rate:
Group 1: 4/112
Group 2: 2/113
No statistically significant difference

Wound dehiscence: Group 1: 3/112
Group 2: 0/113
P=>0.05.
Centres: Single

Years: dns
Chughtai 2000186

Canada
RCT

EL = 1−
Total no. of participants
n = 162
Group 1
n = 81
Group 2
n = 81
Inclusion: patients undergoing coronary artery bypass grafting (elective or emergency)

Exclusion: Not stated

Baseline comparability: yes except clip group had more women (32.1% vs 17.3%, P = 0.03) and more patients with diabetes (29.6% vs 16.0%, P = 0.04)
Loss to follow-up: Not stated
Group 1: Sternal and leg incisions closed with clips

Group 2: Sternal and leg incisions closed with subcuticular sutures
Wound infection Sternal wound infection (redness plus discharge or wound required re-opening or patient required antibiotics):

Assessed at: not stated

Assessor: not stated In hospital:

Sternal wound infection
Group 1: 2/81 (2.5%) Group 2: 1 (1.2%), P = 0.55

Mediastinitis:
Group 1: 1/81 (1.2%)
Group 2:1/81 (1.2%)
P = 0.99

Leg wound infection
(assessed 3–6 weeks post discharge by same surgeon):

Group 1: 1/81 (1.2%) Group 2: 2/81 (2.5%)
P = 0.58

At follow-up (final denominators not stated):
Sternal wound infection:
Group 1: 6/81 (7.4%) Group 2: 1 (1.2%)
P = 0.05.

Sternal wound dehiscence: 4 (4.9%) vs 2 (2.5%), P = 0.39

Mediastinitis:
Group 1:2/81 (2.5%) Group 2 0/81
P = 0.15
Leg wound dehiscence: 7 (8.6%) in A vs 8 (9.9%) in B, P = 0.77

Leg wound infection:
Group 1: 9/81 (11.1%) Group 2: 9/81 (11.1%)
P = 0.98
Dowson 2006169

Setting: UK
RCT

EL = 1−
Total no. of participants
n = 168
Group 1
n = 58
Group 2
n = 61
Inclusion: All patients undergoing laparoscopic general surgical procedures
Exclusion: Diabetes, corticosteroid use, keloid scarring, cyanoacrylate allergy, previous radiation treatment, surgery in ast month causing scar 1 cm from intended incision site, >5 cm wound without deep layer sutures, pregnant women or trying to conceive, emergency surgery, ASA score 4/5, mentally incapacitated patients, unable to give consent Baseline comparability: similar for age, sex, procedure, no of incisions, incisional length
Sample size = 168
Power calculation – yes
Loss to follow up at 4 weeks – 35
Gp 1
n-butyl-cyanoacrylate tissue adhesive applied to laparoscopy wound avoiding subcutaneous layers, held for 20–30 s. Wound not routinely dressed

Gp 2
3/0 interrupted monofilament nylon suture (Ethilon) to close laparoscopy wound and wound routinely dressed
Infection definition: 3 or more of erythema, oedema, tenderness, inflammation, drainage/discharg, malodour.

Wound assessment at 4–6 weeks and 3 months postoperatively.

Unclear if assessment performed blind.

Infection @ 4–6 weeks
Gp 1 8/58
Gp2 6/61
P = 0.7

Dehiscence @ 4–6 weeks
Gp 1) 2/58
Gp 2) 4/61
Centre: Single

Years:2000–2002
Eldrup 1981187

Denmark
RCT

EL = 1−
Total no. of participants
n =
Group 1
n = 69
Group 2
n = 68
Inclusion: 137 patients undergoing elective abdominal and breast surgery and having a skin incision between 15 and 35 cm long

Exclusion: Not stated

Baseline comparability: Not stated
Loss to follow-up: Not stated
Group 1: Staples Group 2: SutureWound infection not defined

Assessed at: not stated

Assessor: not stated Wound infection:
Group 1: 3/69
Group 2: 7/68
NS
Ford 2005152

Study location: USA
RCT

EL = 1−
Total no. of participants
n = 151
Group 1
n = 91
Group 2
n = 44
Inclusion criteria: paediatric patients scheduled for clean or clean/contaminated surgery

Exclusion criteria: contaminated wound sites, retention sutures, inappropriate age, malnutrition, debilitation, conditions impairing wound healing eg AIDS, incision sites prone to stretching, ophthalmic or cardiovascular or neurologic surgical sites, more than one procedure, prior participation, triclosan allergy

Age: mean 9.8 years (1–18 years)

Infection recording period:
postoperatively, observations at 1–2 d, 14 d ± 2 d, 80 d ± 5 d.
Baseline comparability: details given. Age, height, weight, race, risk factors no statistical significance found.
Intervention

Gp 1 – coated polyglactin 910 suture with triclosan

Gp 2 – coated polyglactin 910 suture without triclosan

IV Antibiotics given to 65% Gp 1 and 82% Gp 2
Outcomes

Wound infection – no definition given at day 14

Gp 1 = 2/91
Gp 2 = 0/44
Study duration: not stated

Funding: Ethicon
Study: Gennari 2004183

Setting: Italy
RCT

EL = 1+
Total no. of participants
n = 133
Group 1
n = 69
Group 2
n = 64
Inclusion: patients having breast surgery for breast cancer or benign mammary lesions

Exclusion: Insulin-dependent diabetes mellitus, blood-clotting disorders, personal or family history of keloid or hypertrophic scar formation, allergy to materials used.

Baseline comparability: similar
Loss to follow-up: 16.5% RCT
Group 1: 2-octylcyanoacrylate glue (dermabond)

Group 2: Suture
Wound infection not defined

Assessed daily 5–10 days, 6& 12 months post op

Assessor: plastic surgeon

Wound infection or dehiscence: None
Centres: Years:

Funding: Not stated
Gislason 1995
Norway
RCT

EL = 1+
Total no. of participants
n = 599
Group 1
n = 203
Group 2
n = 199
Group 3
n = 197
Inclusion: adults having major operations for gastrointestinal conditions

Exclusion: Urological, gynaecological and minor general surgical operations; laparotomy in last 3 months

Baseline comparability: yes Loss to follow-up: 17.5% Length of follow up: 1 year
Group 1
Continuous mass polyglyconate (Maxon) double suture with loop

Group 2
Continuous mass polyglactin 910 (Vicryl)

Group 3
Interrupted polyglactin 910 (layered for transverse and mass for midline incisions)
Wound infection -
inflammation or discharge or both, confirmed by culture, or abscess [wound requires opening
Assessed at: not stated

Assessor: not stated

Wound infections]: Group 1: 40/199 (20%)
Group 2 and 3 : 44/384 (11.4%) P = 0.015

Burst abdomen (wound dehiscence): 8/197 (4%) in A vs 3 (2%) in B vs 3 (2%) in C, NS
Greene 1999159

USA
RCT

EL = 1−
Total no. of participants
n = 20
Group 1
n = 15
Group 2
n = 5
Inclusion: adult patients requiring bilateral blepharoplasty for functional or aesthetic indications.

No specific exclusion criteria are described.

split-body design study – a blepharoplasty model with identical skin sites on the same patient and each patient acting as his or her own control.

No patients lost to follow-up.
Group 1
One upper eye lid with octyl-2-cyanoacrylate (Dermabond®, Ethicon Inc, a Johnson & Johnson company, Somerville, New Jersey, USA)

Group 2
Other eyelid closed with 6.0 suture (10 fast-absorbing gut or 10 polypropylene, Prolene, Ethicon Inc, a Johnson & Johnson company, Somerville, New Jersey, USA).

Blepharoplasties were closed on the tissue adhesive side by using Castroviejo forceps to approximate the skin edges in 15 patients and by using 3–4 sutures as handles to facilitate apposition and eversion of edges in 5 patients.
Wound infection defined as: not defined

Assessed at 1,2 &4 weeks post op

Assessor: physician

Dehiscence, infection, patient satisfaction, surgeon satisfaction and cosmetic appearance at 1, 2, and 4 weeks.
Study: Grgić 2002192

Setting: Croatia
RCT

EL = 1−
Total no. of participants
n =
Group 1
n = 25
Group 2
n =
Incision: consecutive patients with head & neck tumours requiring surgical resection.

Exclusion: not stated

Type of wound: head & neck

Age: not stated

Baseline comparability: no details
Group 1
Skin Staples (Proximate Plus MD35 Wide)
Group 2
Interrupted Sutures (3–0 monofilament nylon-Ethicon)

Staples & sutures removed at 7 days post op.

Study duration: not stated.
Wound Infection not defined

Assessed at: not stated

Assessor: not stated

SSI: none

Wound dehiscence: none
Centres: Single

Years: no details
Study: Gys 1989195

Setting: Belgium
RCT

EL = 1+
Total no. of participants
n = 167
Group 1
n = 65
Group 2
n = 67
Population: Consecutive patients undergoing elective or emergency laparotomy
Type of wound: Colorectal = 33.3%, pancreaticobiliary = 30.3%, gastric = 21.2%, vascular = 4.5%, other = 10.6%

Baseline comparability: States balanced groups but slightly higher proportion of patients with risk factors in PA group (malignancy, obesity, shock and dirty wounds), only reported for evaluable population
Group 1
Polyglyconate (Maxon) absorbable synthetic monofilament
Group 2
Polyamide (Ethilon) a non-absorbable synthetic monofilament

Numbers per group only reported for patients evaluated (79% of randomised sample) Continuous layered closure technique of closing peritoneum with No 0, and musculo-aponeurotic layer with No 1, skin closed with 2/0 vertical mattress sutures except if contaminated, when skin closed partially to enable secondary healing
Wound Infection defined as: purulent discharge confirmed by bacteriological growth

Assessed at: observed daily as inpatient, reviewed at 1,6 & 12 months

Assessor: not stated

SSI rate:

Group 1: 14/67
Group 2: 10/65
No statistically significant difference.

Wound dehiscence: 2/67 PA, 1/65 PG
Centres: Single

Years: May 1986 to January 1987

Study duration: 1
Harvey 1986160

Ireland
RCT

EL = 1−
Total no. of participants
n = 20
Group 1
n = 10 pts 20 infections
Group 2
n = 10 pts 20 infections
Inclusion: 20 patients with bilateral varicose veins

Exclusion: Not stated

Baseline comparability: each patient own control Loss to follow-up: Not stated
Length of follow up: 1 month
Group 1: 1 limb closed with staples

Group 2: 1 limb closed with sutures
Wound infection defined as: redness, necrosis & infection score(0=absent, 1=present)

Assessed at: 2–5 days post op & at out patient follow up, time not stated

Assessor: not stated

Infection: 1/20 in each group
Jallali 2004170

Setting: UK
RCT

EL = 1+
Total no. of participants
n = 25
Group 1
n = 13, 51 wounds
Group 2
n = 12, 48 wounds
Inclusion: consecutive patients requiring laparoscopic cholycystectomy.

Exclusion:

Baseline comparability: Baseline characteristics were comparable. Patients in the tissue adhesive group were younger than suture group
Group 1 Subcuticular 3/0 polydiaxonone (Vicryl) (n = 13–51 wounds)
Group 2 2-octylcyanoacrylate (Dermabond)

All patients received 2/0 polybutylate coated polyester(Ethibond) to close the linea alba under the epigastric & umbilical ports.
Wound Infection not defined
Assessed at: 6–8 weeks post op. Assessor: plastic surgeon blind to closure method

SSI: none
Centres: Single

Years: October 2002 – March 2004
Johnson 1997161

USA
RCT

EL = 1−
Total no. of participants
n = 242
Group 1
n =
Group 2
n =
Inclusion: patients with sternal and saphenous vein harvest wounds for clean incisions for coronary artery bypass operations had half of each wound (upper or lower) closed by each method;

mean age 64.7 years

Exclusion: Emergency operations

Baseline comparability: similar
Loss to follow-up: 17 patients (9 died, 2 not closed according to protocol, 6 did not return for follow up)

Length of follow up: 3–4 weeks
Group 1: Staples

Group 2: Intradermal sutures
Wound infection defined – purulent drainage, antibiotic therapy or debridement

Assessed at: discharge, 1 week post discharge & 3 weeks post op.

Assessor: 1 of 3 doctors blind to closure technique.):

Leg wounds:
Group 1: 46/(8.9%) Group 2: 48 (9.3%)
P = 0.99
Sternal wounds:
Group 1: 12 (2.5%)
Group 2: 2 (0.4%)

P stated to be 0.061 in table and 0.128 in abstract
Keng 1989193

UK
RCT

EL = 1−
Total no. of participants
n = 43 pts, 46 incisions
Group 1
n =
Group 2
n =
Inclusion: patients requiring 46 groin incisions (inguinal hernia, femoral hernia, sapheno ligations, testicular operations and lymph-node biopsies).

Exclusion criteria were not stated.

Seven-month follow-up. Three patients lost to follow-up: one from the suture group and two from the tissue adhesive group.
Group 1
Skin incisions were closed with either Histoacryl tissue adhesive

Group 2
dexon subcuticular suture.

In the three bilateral operations the left side was closed with Histoacryl and the right with dexon.
Dexon suture on straight needle using anchoring knot both ends or opposing the wound with forceps and Histoacryl
Wound infection defined as: pus &/or open wound Assessed at 7 days & 1 month post op

Assessor: not stated
Krukowski 1987171

Setting: UK
RCT

EL = 1+
Total no. of participants
n = 757
Group 1
n = 374
Group 2
n = 383
Inclusion: All patients undergoing elective or emergency laparotomy through a midline vertical incision under the care of two consultants

Exclusion: patients undergoing incisional hernia repair excluded.

Type of wound: as above Age: 49% < 60 years Baseline comparability: balanced for age, sex, type of operation and degree of contamination
Group 1
Polydioxanone (PDX) 4-metric 76.2% completed

Group 2
Polypropylene (PPL) 4-metric 77.0% completed

Abdominal wall sutured with tension-free continuous mass closure technique using 50 mm reverse cutting needle, knots buried deep to linea alba. Skin closure with interrupted monofilament polyamide mattress sutures removed day 10,
Wound Infection Wound infection defined as discharge of pus from wound or growth of a pathogenic organism from serous or sanguineous discharge, or any reported discharge from the wound, assessed at 4–6 weeks after discharge from hospital.

Assessed at 4–6 weeks post discharge

Assessor: not stated

SSI rate:
Group 1: 13/374 Group 2: 27/383 P < 0.05

Wound dehiscence: 1 in each group
Centres: Presumed single

Study duration: 1 year
Leaper 1985a 157

Setting: UK
RCT

EL = 1−
Total no. of participants
n = 95 patients with 111 wounds
Group 1
n = 54
Group 2
n = 57
Inclusion: Patients undergoing elective inguinal surgery (hernia or sapheno-femoral ligation) under care of one surgeon Type of wound: hernia = 48.6%, veins = 51.4%
Age: Mean 47 years Baseline comparability: no imbalances reported
Group 1
Subcuticular 2/0 polypropylene Prolene
88.8% completed
Polydioxanone (PDX)

Group 2
85.9% completed Suture ends loosely tied over wound after insertion, same skin preparation used throughout,.

subcutaneous fat layer closed with 3/0 continuous chromic catgut, each wound covered with Op-site, wounds left undisturbed until 7th day, then prolene removed but PDX clipped flush with skin Study duration: 6 months
Wound infection defined as: persistent superficial cellulitis or induration lasting beyond 7th post-op day, meaningful if antibiotics prescribed, spontaneous discharge of pus or wound required to be opened
Assessed: 7 days, 6 weeks& 6 months post op
Assessor: not stated, included GP & patient recall.

Wound infection; wound appearance at 6 weeks and 6 months SSI rate:
Group 1: 9/54
Group 2: 10/57
No significant difference.
Centres: Presumed single

Years: dns
Leaper 1985b 196

Setting: UK
RCT

EL = 1+
Total no. of participants
n = 233
Group 1
n = 97
Group 2
n = 107
Inclusion: Patients with major laparotomy wounds, midline or transverse
Type of wound: gastric = 16.1%, pancreatico-biliary = 45%, small bowel = 4.4%, colon = 25.5%, other = 8.8%

Age: Mean 57.7 years

Baseline comparability:
demographic details only available for study completers
Group 1
No 1 (BPC) polyamide (nylon)
Group 2
Polydioxanone (PDX)

87.6% of randomised sample completed overall

Continuous mass technique to close abdominal wall, midline and transverse incisions but not through scar tissue, skin closed with continuous subcuticular 00 polypropylene suture, wound covered with Op-site dressing, antibiotic prophylaxis if required
Wound infection defined as: purulent discharge, systemic symptoms causing delayed discharge

Assessed at: daily during inpatient period up to 14 days, 6 week & 6 months at out patient
Assessor: not stated

Outcomes:
SSI rate:
Group 1:9/97
(7 minor, 2 major)
Group 2: 18/107
(14 minor,4 major)
Not statistically significant.

Authors suggest trend towards increased infection with synthetic sutures which are absorbed more slowly.
Centres: Two centres

Years: dns
Maartense 2002163

Netherlands
RCT

EL = 1+
Total no. of participants
n = 140
Group 1
n = 48
Group 2
n = 42
Group 3
n = 50
Inclusion:adult patients requiring elective laparoscopic surgery.

Exclusion: Patients who had undergone previous laparotomy or were pregnant. The study was undertaken at two centres, Sixteen-month follow-up

There were no withdrawals, however seven patients treated with paper tape and three with tissue adhesive were converted to the suture group.
Group 1
Octylcyanoacrylate (Dermabond®, Johnson & Johnson, Amersfoot, The Netherlands) tissue adhesive
Group 2
76 mm × 6 mm adhesive paper tape (SteriStrip® Bioplasty/Uroplasty, Geleen, The Netherlands)
Group 3 intracutaneous poliglecaprone (Monocryl®) 4/0, Johnson & Johnson) interrupted sutures
Wound infection defined as: pus requiring surgical or spontaneous drainage

Assessed at 10–14 days & 3 months

Assessor: surgical house officers.

Notes – There were 43 patients and 46 wounds included. The wounds were treated as independent data which strictly is incorrect, however as there were only three patients with more than one wound it will make little difference to the conclusions so the study and data is included.
Centres: 2
Magann 2002164

Australia
RCT

EL = 1+
Total no. of participants n =
Group 1
n =
Group 2
n =
Inclusion: 964 women having Caesarean section with ≥2 cm subcutaneous tissue Exclusion: Emergency operation Baseline comparability: yes Loss to follow-up: 26.4%RCT Length of follow up: 6 weeksAll skin closure with clips Subcutaneous closure with: A: Not closed (n = 205) B: Running suture of polyglycolic suture (n = 191) C: 7-mm closed drainage system (n = 146)Wound infection defined as: induration, erythema, purulent material Assessed at: 14 & 42 days post op Assessor: not stated Wound infection: 14 in A vs 16 in B vs 14 in C, NS
Mullen 1999165

USA
RCT

EL = 1−
Total no. of participants
n = 80
Group 1
n = 20
Group 2
n = 20
Group 3
n = 19
Group 4
n = 18
Inclusion: patients (3 later excluded) undergoing elective coronary artery bypass surgery

Exclusion: Insertion of a drain, intra-aortic balloon pump in index limb, or inability to complete follow up

Baseline comparability: similar Loss to follow-up: 3.75% Length of follow up: 6–8 weeks
Group 1: Staples, close immediately

Group 2: Staples, close after protamine administration

Group 3: Subcuticular sutures, close immediately

Group 4: Subcuticular sutures, close after protamine administration
Minor leg wound infection rate (erythema+ one of: fever 38.5° C, raised WBC >12 × 109/litre: or purulent discharge):

Group 1: 3/20 (15%)
Group 2: 3/20 (15%)
Group 3: 1/19 (5.3%)
Group 4: 0/18

Major leg wound infection rate (infection requiring i.v. antibiotics or surgical therapy, prolonging hospital stay or requiring readmission):
Group 1: 0/20
Group 2: 0/20
Group 3: 0/19
Group 4: 2/18 (10.5%)

Assessed daily during inpatient stay by phone & 6–8 weeks post discharge

Assessor: not stated
Murphy 1995158

Setting: Ireland
RCT

EL = 1−
Total no. of participants
n = 114, 173 wounds
Group 1
n = 41 wounds
Group 2
n = 38 wounds
Group 3
n = 45 wounds
Group 4
n = 49 wounds
Inclusion: Patients undergoing bypass surgery with a groin incision
Type of wound: Aorto-femoral bypass = 25% femorodistal bypass = 75%

Age: Mean 67 years

Baseline comparability: no imbalances noted
Group 1
Subcuticular Maxon

Group 2
Continuous nylon over and over technique

Group 3
Interrupted nylon

Group 4
Clips

Subcutaneous tissues closed with 3/0 Maxon in two layers to leave skin edges to be closed by four interventions

Study duration: 14 days
Wound infection defined as positive cultures.

Assessed at: alternate days up to 14 days post op

Assessor: not stated

Outcomes: Infection confirmed by bacteriological culture, dehiscence, cost of suture material

SSI rate:
Group 1: 1/41
Group 2: 1/38
Group 3 : 2/45
Group 4 : 1/49

No significant difference between groups.

13 wounds showed local signs of infection but had negative cultures
Centres: Presumed single

Years: dns
Murphy 2004191

Setting: Ireland
RCT

EL = 1−
Total no. of participants
n = 60
Group 1
n = 31
Group 2
n = 29
Inclusion: patients undergoing routine clean orthopaedic procedures

Exclusion: Not stated

Baseline comparability: Not stated Loss to follow-up: Not stated

Length of follow up: 13 days
Group 1: Clips
Group 2: Sutures
Wound infection (erythema plus discharge, wound required opening or antibiotics):
Group 1: 1/31 (3.2%) Group 2: 1/29 (3.4%) NS

Dehiscence: 1 (3.2%) in A vs 1 (3.4%) in B, NS Assessed at 13 days, further assessment not stated Assessor: surgeon
Nasir 2001188

Iraq
RCT
EL = 1+
Total no. of participants
n = 100
Group 1
n = 50
Group 2
n = 50
Inclusion: patients having laparotomy through midline incision

Exclusion: not stated

Baseline comparability: Unclear – patients having emergency operations were more likely to be in group 2 (54% vs 46% in 1, not clear how significant this is; not adjusted for in analysis) Loss to follow-up: none
Group 1: Continuous double loop closure

Group 2: Continuous mass closure (n = 50) Both arms used polypropylene
Wound infection not defined
Assessed at: 5–7, 21 & 42 days post op
Assessor: not stated

Infection:
Group 1: 6/50 (12%) Group 2: 9/50 (18%)

Wound dehiscence: None in A vs 4 (8%) in B
Niggebrugge 1999172

Denmark
RCT

EL = 1+
Total no. of participants
n = 390
Group 1
n = 204
Group 2
n = 186
Inclusion: Patients undergoing midline-laparotomy wound closure

Exclusion: Age 15 or less; laparotomy in previous 3 months
Baseline comparability: similar except 23% emergency in Group 1 vs 33% in Group 2 P = 0.02; accounted for by separate analyses for elective and emergency
Loss to follow-up: none
Group 1: Continuous Running Suture (CRS)

Group 2: Continuous double-loop closure (CDLC)
Wound infection (purulent discharge or positive cultures):

Group 1: 13/204 (6.4%)
Group 2: 17/184 (9.1%)
NS

Wound dehiscence: 4 (2.0%) in A vs 7 (3.8%) in B, NS
Ong 2002153

Singapore
RCT

EL = 1+
Total no. of participants
n = 59
Group 1
n = 26
Group 2
n = 33
Inclusion: children having herniotomies (mean age 4.5 years)

Exclusion: Neonates or children allergic to glue

Baseline comparability: similar
Length of follow up: 2–3 weeks; originally planned to review at 3 months but only 9/59 patients returned Blinding of outcome assessors: yes
Group 1: 2-octylcyanoacrylate glue
Group 2: Sutures
Wound infection defined: not stated

Assessed at 14–21 days & 3 months

Assessor: independent nurse observer.

Wound dehiscence: none in either group

Wound infection: none in either group
Funding: Ethicon
Study: Orr 2003249

Setting: USA
RCT

EL = 1+
Total no. of participants
n = 203
Group 1
n = 104
Group 2
n = 97
Inclusion: High risk factors for poor wound outcome with abdominal incisions, age at least 18 years, evidence of compromised wound healing due to at least 1 of following: age > 70 years, obesity, cancer, diabetes, COPD, chronic steroid use, altered nutritional status, ascites, renal insufficiency, jaundice, prior radiation to surgical site, prior transverse incision crossing study incision

Type of wound: abdominal malignancy = 70%, others = 30%

Age: Mean 55 years; range 21–87 years
Group 1:
No 1 long-term absorbable multifilament suture poly(L-lactide/glycolide) (PLG)

Group 2:
Permanent monofilament (No 1) polypropylene (n = 97)

Overall 99% completed

Continuous fascial closure for both with strict per-operative management protocol
Wound infection: not defined
Assessed at: It is unclear when infections were measured, but complications were recorded in the immediate postoperative period as well as at scheduled visits (4–6 weeks, 6 months).
Assessor: not stated

SSI rates:
Group 1: 8/104
Group 2: 6/97
P = 0.75)

Dehiscence: 4/104 PLG 10/97 No1ppp No significant difference( P = 0.09).
Centres: 9

Years: May 1999 to June 2000
Orr 1990173

USA
RCT

EL = 1+
Total no. of participants
n = 402
Group 1
n = 201
Group 2
n = 201
Inclusion: low risk women having abdominal gynaecological procedures

Exclusion: High risk (e.g. active infection plus cancer plus previous radiation); vaginal operation

Baseline comparability: similar Loss to follow-up: None RCT Length of follow up: 6 months Blinding of outcome assessors: not stated Power calculation: not stated
Group 1: Continuous closure of fascia

Group 2: Interrupted closure of fascia
Wound infection not defined
Assessed at: daily as inpatient, 4 & 24 weeks post discharge

Assessor: Doctor

Infection or seroma: Group 1: 9/201 (4.5%) Group 2: 4/201 (2.0%)
Osther 1995

Setting: Denmark
RCT

EL = 1+
Total no. of participants
n = 204
Group 1
n = 100
Group 2
n = 104
Inclusion: People with suspected impaired wound healing undergoing elective or emergency laparotomy, Impaired wound healing included one or more of following criteria: age > 70 years, COPD for at least 10 years, intra-abdominal malignancy or diffuse peritonitis,

Excluded: incisions through previous scar

Type of wound: dns
Age: Median 75 years (Dexon) and 77 years (Maxon) Baseline comparability: balanced for age, sex, type of incision and operation, risk factors, presence of malignancy or corticosteroid use
Group 1
Early absorbable multifilament polyglycolic acid sutures (Dexon 0/0) for fascial closure
(84% completed 3 months, 70% 12 months)

Group 2
Late absorbable monofilament polyglyconate sutures (Maxon)
(74% completed 3 months, 64.4% completed 12 months)

Fascial closure following abdominal surgery, sutures placed 1.5 cm from fascial margins with 1 cm between each stitch as a simple interrupted suture, subcutaneous tissues not closed, skin closed with interrupted nylon 3/0.
Wound infection defined as purulent discharge leading to surgical drainage

Assessed at: the time of measurement was not stated, but is described as ‘early complications’

Assessor: not stated

SSI incidence:
Group 1:16%
Group 2: 7%
( P = 0.04).
Centres: Presumed single

Years: March 1990 to December 1991
Ozturan 2001175

Turkey
RCT

EL = 1+
Total no. of participants
n = 101
Group 1
n = 34
Group 2
n = 67
Inclusion: patients requiring rhinoplasty or septorhinoplasty were entered to the study.

Exclusion: Patients who had a history of peripheral vascular disease, diabetes mellitus, clotting disorder, keloid or hypertrophy, or allergy to cyanoacrylate or formaldehyde.

There were no withdrawals.
Group 1:
Butylcyanoacrylate (LiquiBand. MedLogic Global Ltd, Plymouth, Devon, UK) tissue adhesive

Group 2
6.0 polypropylene sutures for columellar skin closure after the majority of the tension had been taken up using 5.0 chromic catgut.
Wound infection defined as: not stated

Assessed: weekly for 1st 4 weeks post op, then at 8 & 12 weeks.

Assessor: surgeons

Dehiscence and infection at one week.

Notes – The authors were contacted to clarify the numbers in each group randomised by coin toss and received confirmation that the numbers were correct. We also received clarification that the standard deviations were presented after the means in the results section of the paper.
Paral 2007176

Study location: Czech Republic
RCT

EL = 1+
Total no. of participants
n = 415

Group 1A
n = 101
Group 1B
n = 109
Group 2A
n = 109
Group 2B
n = 105
Population: Participants were patients undergoing elective abdominal surgery and were split into two groups according to wound contamination – clean (n = 201) and clean-contaminated (n = 214)

Age: 18–88 years

Exclusion criteria: long term anticoagulative therapy, haematological disease or haemocoagulation disorders

Infection recording period: days 3, 7, 10 and 30 postoperatively Baseline comparability: details given. Age, BMI, subcutaneous fat thickness and risk factors all similar except sex (Gp A1and A2 considerably fewer females)and nicotine usage.
Intervention:
  1. Sutured closure of subcutaneous fat
  2. No closure of subcutaneous fat layer
4 Groups:
1A.

Clean wound with sutured subcutaneous fat

1B.

Clean wound with no closure of subcutaneous fat (n = 100)

2A.

Clean-contaminated wound with sutured subcutaneous fat (n = 109)

2B.

Clean-contaminated wound with no closure of subcutaneous fat (n = 105)

Outcomes

Wound infection (Superficial and Deep) at Day 30

1A – 3%
1B – 2%
2A – 6.4%
2B – 5.8%

Wound Dehiscence (Partial – no total dehiscence observed) at Day 30

1A – 2%
1B – 2%
2A – 4.6%
2B – 1.9%
Study duration: not stated

Funding: Ministery of Defence and University of Defense of the Czech Republic
Ranaboldo 1992184

UK
RCT

EL = 1−
Total no. of participants
n =
Group 1
n = 22
Group 2
n = 26
Inclusion: 48 patients with midline abdominal wounds, mean age 65 years

Exclusion: Not stated

Power calculation: no – underpowered

Baseline comparability: yes Loss to follow-up: Not stated
Group 1: Staples

Group 2: Sutures
Wound infection not defined
Assessed at: 1 month Assessor: not stated

Wound infection with discharge of pus:
Group 1: 1/22
Group 2: 1/26
Sadick 1994185

Australia
RCT

EL = 1−
Total no. of participants
n = 100
Group 1
n = 50
Group 2
n = 50
Inclusion: Patients having excision of benign pigmented lesions of upper back

Exclusion: not stated

Baseline comparability: not stated
Loss to follow-up: not stated
Group 1: Conventional bi-layered technique

Group 2: Buried vertical mattress technique (n = 50)
Infection:
Group 1: 3/50 (6%) Group 2: 2/50 (4%) iNS
Sebesta 2003194

USA
RCT

EL = 1−
Total no. of participants
n = 59
Group 1
n = 30, 118 incisions
Group 2
n = 29, 110 incisions
Inclusion: 59 patients in whom 228 laparoscopic trocar scars were closed

Exclusion: Wounds in both groups that did not closely approximate received interrupted subcutaneous sutures

Baseline comparability: Not stated
Loss to follow-up: Not stated
Group 1: 2-octylcyanoacrylate glue (dermabond)

Group 2: Sutures
Wound infection defined as: not stated

Assessed: 14 days post op.

Assessor: not stated Wound infection:

Group 1: 1/30
Group 2: 0/29
Funding: Ethicon
Shamiyeh 2001166

Austria
RCT

EL = 1+
Total no. of participants
n = 79
Group 1
n = 26
Group 2
n = 28
Group 3
n = 25
Inclusion: adult patients requiring varicose vein surgery on the leg.

Exclusion: Patients with a history of chronic venous insufficiency with dermatosclerosis, previous phlebectomies, or allergy to plaster or octylcyanoacrylate.

Nine-month follow-up randomised, parallel group study.
Two patients were lost to follow-up from the suture group due to failure to attend and could not be traced by mail or phone.
Mullerian phlebectomy creating average wound length of five mm. Used 5 minutes wound compression followed by skin closure with

Group 1: octylcyanoacrylate tissue adhesive

Group 2: 5.0 monofilament suture or tape.

Group 3: A small plaster was placed over each wound.
Wound infection defined as: not stated

Assessed: 10 days & 2 weeks post op.

Assessor: not stated

Wound dehiscence, infection, at 10 days

Notes – The authors were approached to clarify the reason that two patients were lost to follow up and also to clarify how many patients in each group had dehiscence and infection at 10 days. We received the clarification required to enable us to use the data in this review.
Sinha 2001182

UK.
RCT

EL = 1+
Total no. of participants
n = 44
Group 1
n = 20
Group 2
n = 24
Inclusion: adult patients requiring hand or wrist surgery (carpal tunnel syndrome, trigger finger, De Quervain’s tenosynovitis, ganglions of wrist and hand, and cysts of fingers).

Exclusion: Patients requiring surgery for Dupuytren’s contracture, repeat surgery or had a history of skin allergy, keloid formation, diabetes, or corticosteroid use.
Skin approximating with skin hooks and applying

Group 1: n-butyl 2-cyanoacrylate adhesive (Indermil)

Group 2: suturing with 4.0 monofilament.

All cases have local anaesthetic infiltration with or without general anaesthesia.
Wound infection: Assessed: 7–10 days, 6,12 weeks Assessor: Tissue viability nurse blind to closure method

Dehiscence, infection, pain on movement, cosmetic appearance at 10 days, 2 weeks and 6 weeks.
Stenvik 2006189

Study location: Norway
RCT

EL = 1+
Total no. of participants
n = 239
Group 1
n = 119
Group 2
n = 120
Inclusion criteria: patients undergoing CABG saphenectomy

Exclusion criteria: unclear.
Gp 1 (n = 119) were further randomised to ic only Gp
1A(n = 59)
or
in and sc closure Gp 1B (n = 60).

Gp B were matched controls – not reported here

Age: not stated

Baseline comparability: details given. Diabetes, BMI sex, operation time, unstable angina all similar, no statistical significance found.
Intervention

Gp 1A – intracutaneous closure only using Ethicon Monocryl 3–0 absorbable suture (n = 59)

Gp 1B – intracutaneous continuous closure using Ethicon Monocryl 3–0 absorbable suture and subcutaneous continuous closure using Ethicon Vicryl 3–0 absorbable suture

Antibiotic treatment given
Outcomes

Definition SSI: symptoms of rubor, calor and dolor requiring wound drainage or antibiotic treatment

Infection recording period: unclear – local patients seen in outpatients and others contacted by phone, time not specified

Gp A1 – 2/59
Gp A2 – 4/60
Study duration: during 2003

Funding: not stated
Toriumi 1998154

USA
RCT

EL = 1−
Total no. of participants
n = 111
Group 1
n =
Group 2
n =
Inclusion: patients of 1 year of age and over requiring elective surgery for benign skin lesions predominantly in face and neck.

Exclusion: Patients were excluded if there was a history of significant trauma, peripheral vascular disease, diabetes mellitus, and blood clotting disorder, keloid or hypertrophy, known allergy to cyanoacrylate or formaldehyde.

11 patients were lost to follow-up: reason for withdrawal not stated by group.
Incisions with and without subcutaneous sutures and then randomised for closure with

Group 1: octyl-2-cyanoacrylate suture

Group 2: 5.0 or 6.0 nylon suture.
Wound infection defined as: not stated

Assessed: Dehiscence and infection were noted at 5–7 days.

Assessor: not stated
Van den Ende 2004155

Netherlands
RCT

EL = 1+
Total no. of participants
n =
Group 1
n = 50
Group 2
n = 50
Inclusion: 100 children having herniotomies or orchidopexy

Exclusion: Not stated

Baseline comparability: similar
Loss to follow-up: None stated
Group 1:
N-butylcyanoacrylate glue; mean age 2.5 years
Group 2:
Sutures mean age 3.0 years
Wound infection defined as: not stated
Assessed: 10 days & 6 weeks post op. Assessor: not stated

Wound dehiscence: Group 1: 13/50 (26%) Group 2 ; 0/50 (P < 0.001)

Wound infection:

Group 1: 4/50
Group 2 ; 2/50 (NS)
Funding: Not stated
Velmahos 2002177

USA
RCT

EL = 1+
Total no. of participants
n = 48
Group 1
n = 26
Group 2
n = 22
Inclusion: 48 patients having operations for colon injuries at a trauma centre

Exclusion: Aged under 18 years, pregnant, operated on >6 hours after admission or damage control procedures with abdomen left open Baseline comparability: yes Loss to follow-up: 3 (6%) re-operated within 2 days and left with open abdomen
Group 1:
Primary skin closure

Group 2:
Delayed skin closure
Wound infection (pus, positive wound cultures and, if wound closed, need for opening for drainage):

Group 1: 17/26 (65%) Group 2: 8/22 (36%)
P = 0.04

Wound dehiscence (separation of fascia): Group 1: 8/26 (31%) Group 2: 3/22 (14%)
NS
Wolterbeek 2002178

Netherlands
RCT

EL = 1+
Total no. of participants
n = 170
Group 1
n = 86

Group 2
n = 84
Inclusion: 170 patients with femoro-popliteal or femoro-tibial bypass

Exclusion: Baseline comparability: similar

Loss to follow-up: 10 (6%)
Group 1: Staples but 2 had 2nd procedure and 1 died within 2 weeks

Group 2: Sutures but 5 had 2nd procedure and 2 died within 2 weeks
Superficial infections (erythema + serous leak):

Group 1: 2/86 (2%)
Group 2: 6/84 (8%)
NS
Deep infections:
Group 1:1/86 (1%)
Group 2:1/84 (1%)
NS
Yigit 2005190

Study location: Turkey
RCT

EL = 1+
Total no. of participants
n = 152
Group 1
n = 74
Group 2
n = 78
Population: Participants were male patients with chronic pilonidal sinus undergoing excision and primary closure

Age: median 22 years (range 15–64)

Inclusion criteria: male patients with chronic pilonidal sinus undergoing excision and primary closure

Exclusion criteria: patients with active infection

Baseline comparability: details given. Age, mean lateral distance from midline similar, no statistical significance found. Length of operation significantly higher in Gp 1.
Intervention

Gp 1 – intracutaneous closure only using Ethicon Monocryl 3–0 absorbable suture

Gp 2 – intracutaneous continuous closure using Ethicon Monocryl 3–0 absorbable suture and subcutaneous continuous closure using Ethicon Vicryl 3–0 absorbable suture

Antibiotic treatment given
Outcomes

Wound infection – no definition given
Infection recording period: early complication @ 10 days, late complication during subsequent 6 weeks

Early
Gp 1
1/74 deep infection (1.35%)
5/74 superficial infection (6.7%)

Gp2
1/78 deep infection (1.3%)
2/78 superficial infection (2.6%)

Late
Gp 1
None observed

Gp 2
1/78 wound dehiscence (1.3%) on day 20 P.o.
Study duration:
May 2003 to May 2004

Funding: not stated

From: Evidence tables

Cover of Surgical Site Infection
Surgical Site Infection: Prevention and Treatment of Surgical Site Infection.
NICE Clinical Guidelines, No. 74.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2008 Oct.
Copyright © 2008, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.