Tissue Adhesive Versus Suture Wound Repair at

1 Year: Randomized Clinical Trial Correlating

Early, 3-Month, and 1-Year Cosmetic Outcome

James Quinn, MD*

George Wells, PhDII

Terri Sutcliffe, BScN‡

Mario Jarmuske, MD§

Jennifer Maw, MD#

Ian Stiell, MD, MSc‡II

Peter Johns, MD‡

Study objective: To compare the 1-year cosmetic outcome ofwounds treated with octylcyanoacrylate tissue adhesive andmonofilament sutures and to correlate the early, 3-month, and1-year cosmetic outcomes.

Methods: We prospectively randomized 136 cases of traumaticlaceration to repair with octylcyanoacrylate tissue adhesive or5-0 or smaller monofilament suture. A wound score was assignedby a research nurse, and validated by a second nurse blinded tothe treatment, at 5 to 10 days after injury (early), 3 months, and1 year. Standardized photographs were taken at 3 months and 1year and shown to a cosmetic surgeon blinded to the method ofclosure, who rated the wounds on a validated cosmesis scale.

Results: We were able to examine 77 lacerations at 1 year forfollow-up. No differences were found in the demographic orclinical characteristics between groups. Likewise, at 1 year nodifference was found in the optimal wound scores (73% versus68%, P=.60) or in visual analog scale cosmesis scores (69 versus69 mm, P=.95) for octylcyanoacrylate and sutures, respectively.Agreement was poor between early and 3-month wound scores(κ=.34; 95% confidence interval [CI], .10 to .58) but a strongassociation existed between 3-month and 1-year wound scores(κ=.71; 95% CI, .52 to .90). We noted a moderate correlationbetween 3-month and 1-year results on the visual analog cosmesisscale (intraclass correlation, .48; 95% CI, .30 to .63).

Conclusion: One year after wound repair, no difference is notedin the cosmetic outcomes of traumatic lacerations treated withoctylcyanoacrylate tissue adhesive and sutures. The assessmentof wounds 3 months after injury and wound repair provides agood measure of long-term cosmetic outcome.

[Quinn J, Wells G, Sutcliffe T, Jarmuske M, Maw J, Stiell I,Johns P: Tissue adhesive versus suture wound repair at 1 year:Randomized clinical trial correlating early, 3-month, and 1-yearcosmetic outcome. Ann Emerg Med December 1998;32:645-649.]

From the Section of EmergencyMedicine, University of Michigan,*

Ann Arbor, MI; and the Divisions ofEmergency Medicine‡ and PlasticSurgery§ and the Departments ofClinical EpidemiologyIIandOtolaryngology,# University ofOttawa, Ottawa, Canada.

Received for publication March 31, 1998. Revisionsreceived August 17, 1998. Accepted for publication August 24, 1998.

Presented at the Society for Academic Emergency MedicineAnnual Meeting, Chicago, IL,May 1998.

Supported by a research gift fromClosure Medical Corp.

Address for reprints: Dr JamesQuinn, UCSF Medical Center, Herbst Emergency Services, Box0208, 505 Parnassus Avenue, San Francisco, CA 94143-0208; 415-502-4534, fax 415-665-3999, E-mail quinnj@medicine.ucsf.edu.

Copyright © 1998 by the AmericanCollege of Emergency Physicians.

0196-0644/98/$5.00 + 047/1/94302

O R I G I N A L C O N T R I B U T I O N

6 4 5D E C E M B E R 1 9 9 8 3 2 : 6 A N N A L S O F E M E R G E N C Y M E D I C I N E

W O U N D R E P A I RQuinn et al

6 4 6 A N N A L S O F E M E R G E N C Y M E D I C I N E 3 2 : 6 D E C E M B E R 1 9 9 8

I N T R O D U C T I O N

More than 12 million traumatic wounds are treated eachyear in the United States.1 These injuries have traditionallybeen closed with sutures. Until recently, any new method ofwound closure was usually reported as a case series. How-ever, clinicians have begun to objectively compare wound-closure techniques in clinical trials, and no closure methodhas been subjected to as much scrutiny as tissue adhesives. Todate, 5 randomized trials investigating the repair of trau-matic lacerations have been reported; 2 such studies havebeen focused in the closure of incisions. Three of these trialsdealt with the older and weaker butylcyanoacrylates, focus-ing on small linear lacerations,2-4 and another 3 reportedon the new stronger, medical-grade octylcyanoacrylateformulation (Dermabond; Closure Medical Corp, Raleigh,NC) that recently gained approval by the US Food andDrug Administration.5-7

The primary outcome in these trials has been cosmesis—appropriately so, because this is the most important out-come for patients.8,9 To date, all trials have reported cos-metic outcomes up to 3 months and have demonstratedno differences in the cosmetic outcomes of tissue adhesive–and suture-repaired wounds. Although never objectivelyproved, the opinion has been voiced by some experts thatcosmesis 1 year after wound repair should be reported.10

In this study we sought to compare, 1 year after repair,traumatic lacerations treated with octylcyanoacrylate andthose treated with sutures. The secondary purpose of thestudy was to correlate early (5- to 10-day), 3-month, and1-year cosmetic outcomes to help determine an appropri-ate time at which to measure cosmetic outcomes.

M A T E R I A L S A N D M E T H O D S

Adult patients who had sustained traumatic lacerationsrequiring suture on the face, torso, or extremities—regardless of length or need for deep sutures—were eligiblefor study enrollment. Exclusion criteria included preg-nancy, soiled or heavily contaminated wounds requiringdebridement, animal or human bite or scratch wounds,puncture wounds, stellate crush wounds, wounds cross-ing mucocutaneous junctions, scalp lacerations; patientswith aknown history of keloid formation; ear lacerations;wounds on the hands or feet; any wounds crossing joints;and extremity lacerations—in patients with a history ofperipheral vascular disease, diabetes, or prolonged steroiduse. Patients were excluded if they did not agree to thestudy or were unable to participate in the follow-uprequired by the study. The study was conducted in theemergency department of the Ottawa General Hospital, anadult teaching hospital at the University of Ottawa. The studyprotocol was approved by the Research Ethics Board of theOttawa General Hospital.

All eligible patients were prospectively enrolled in thestudy, and informed consent was acquired. Lacerationswere stratified by means of block randomization into facialand nonfacial wounds to assure homogeneity betweengroups.

All lacerations were first assessed to determine whetherdeep sutures or debridement would be needed. Woundsrequiring debridement or deep sutures were treated inaccordance with the standard practice of the treatingphysician: 1% lidocaine, with or without epinephrine foranesthesia, with or without irrigation with normal salinesolution. Only at the time of skin closure were the lacera-tions randomized to treatment consisting of skin closure with5-0 or 6-0 monofilament suture or octylcyanoacrylate tissueadhesive. Patients with wounds deemed to require no deep-

Registered eligible lacerations (n=179)

Not randomized (n=43)(35 refused, 8 missed)

Randomized (n=136)

Allocated to sutures (n=68) Allocated to octylcyanoacrylate (n=68)

Early follow-up (n=53) Early follow-up (n=53)

3-Month follow-up (n=50) 3-Month follow-up (n=55)

1-Year follow-up (n=40) 1-Year follow-up (n=37)

Table 1. Clinical and demographic characteristics of the study groupscompleting 1-year follow-up.

Parameters Octylcyanoacrylate (n=37) Sutures (n=40)

Mean age (y) [mean±SD] 34.7±12.4 39.6±18.3Male sex (%) 88.5 75.0Wound length (cm) [mean±SD] 2.2±1.4 2.4±1.2Deep sutures (%) 10.8 20.5Nonfacial injury (%) 16.2 17.5

P was nonsignificant for all characteristics.

Figure 1.Profile of the randomized trial.

→→

→→

W O U N D R E P A I RQuinn et al

D E C E M B E R 1 9 9 8 3 2 : 6 A N N A L S O F E M E R G E N C Y M E D I C I N E 6 4 7

clinical variables: absence of step-off, contour irregularities,wound-margin separation greater than 2 mm, edge inver-sion, excessive distortion, and overall cosmetic appearance.Each of these categories was graded on a 0-or-1-point scale.A total cosmetic score was derived by the addition of thescores of the 6 categorical variables. A score of 6 is optimal, ascore of 5 or less suboptimal. We compared the percent-ages of wounds from each group demonstrating optimalcosmesis.

At the 3-month and 1-year follow up visits, photographsof the patients’ healing wounds were taken by the researchassistant in a standard fashion involving a 1:3 macro settingwith a ring flash on 100 Ektachrome slide film (EastmanKodak, Rochester, NY).12 These photographs were rated forcosmesis on a previously validated visual analog cosmesisscale that has been demonstrated to be a reliable and validoutcome measure of cosmesis.13The visual analog cosme-sis scale is a 100-mmlinewith “worst scar” at the left end of theline, for a score of0, and “best scar” at the right end, scored as“100” (Figure 2). The photographs were rated by a cosmeticsurgeon blinded to the wound-closure method.

We computed descriptive statistics of all outcomes usingSPSS 7.5 for Windows software (Chicago, IL). For the pri-mary outcome (visual analog cosmesis scores) and othercontinuous outcomes, the treatment groups were comparedwith the use of the Student’s t test for independent samples.We used χ2 analysis to compare wound-evaluation scoresand other dichotomous outcomes. The intraclass correla-tion coefficient and κ-coefficient were used to measureagreement for continuous and categoric data, respectively.We calculated the power of the sample size using 90%power and an .05 α-error rate.

wound care were immediately assigned to skin closure withmonofilament sutures or with octylcyanoacrylate.

Sutured wounds were anesthetized and cleansed at thetreating physician’s discretion and the skin closed with a5-0 or 6-0 monofilament suture under sterile techniques.A dressing was applied for at least 48 hours.

Wounds closed with octylcyanoacrylate were prepared ina standard manner. Wounds were cleansed with a chlorhexi-dine solution, and hemostasis was achieved with pressureor topical 1:1,000 epinephrine as appropriate. Wound edgeswere manually approximated with the use of fingers or for-ceps. The physician conducting the repair topically paintedthe octylcyanoacrylate over the manually apposed woundedges using the applicator tip, careful not to apply the adhe-sive between the wound edges. The wound was held for30 seconds to allow complete polymerization.

All wounds in the study were treated by staff physicians,and all study physicians were trained in the use of tissueadhesives by the principal investigator (JQ). To ensure thatall eligible patients were enrolled, the ED logbook waschecked daily.

Patients were asked to return for early suture removal,wound assessment, or both—3 to 5 days after repair forfacial lacerations and 10 days after repair for extremity andtorso lacerations. Patients returned or contacted the inves-tigators immediately if they had any questions or concernsabout their wounds. Wound dehiscence was treated bymeans of primary closure with the previously randomizedmethod of closure. Infections were treated with antibioticsat the treating physician’s discretion. Follow-up was con-ducted 3 months and 1 year after wound repair (Figure 1).

All wounds were evaluated with the use of a previouslyvalidated wound-evaluation scale 5 to 10 days, 3 months,and 1 year after wound repair.11 The score was assignedby a research nurse aware of the method of closure andvalidated by a second nurse blinded to the method of closure(sutures and glue removed). The wound score addresses 6 Figure 2.

One-year visual analog cosmesis scores.

69±19.6 mm

69±14.3 mm

OctylcyanoacrylateSutures

0 100Worst scar Visual analog cosmesis scale Best scar

Table 2. Correlations among early, 3-month, and 1-year outcomes forwound scores.

Time κ-Value (95% CI)

Early–3-month .34 (.10 to .58)3-Month–1-year .71 (.52 to .90)

Strength of κ-coefficients18: .2–.4=fair agreement; .4–.6=moderate agreement; .6–.8=substan-tial agreement; >8=almost perfect.

Agreement between research nurses in rating thewounds was excellent (κ=.84; 95% CI, .53 to 1.0).

D I S C U S S I O N

This study is the first to document with sufficient power thatno difference exists 1 year after treatment in the cosmeticoutcomes of wounds repaired with sutures and those closedwith octylcyanoacrylate tissue adhesive.

It is not surprising that no differences are found 1 year afterrepair as well. We previously found no difference betweenwounds at 3 months, and histologically there appear to beno differences in wounds repaired with tissue adhesivesand those closed with sutures after the sutures have beenremoved and the adhesive has sloughed off.5,14

Problematic wounds are traditionally monitored untilfull maturity at 1 year before any attempts at surgical scarrevision are made.10 However, some have also suggestedthat 1 year should be the standard for cosmetic evaluation.These recommendations are not based on any scientificevidence and seem unwarranted on the basis of our find-ings and the biology of wound healing.

At 1 year, collagen has fully matured and a wound isclose to 100% of the skin’s original tensile strength becausemost collagen remodeling has taken place, thus the rationalefor delaying cosmetic assessment. However, at 3 months awound has 80% of the original tensile strength, so it isunlikely that significant changes in wound remodeling orscar widening would occur between 3 months and 1 yearthat would change cosmetic outcome. Wounds appear tochange significantly between 10 days (when a wound hasonly 10% of the undamaged skin’s original strength) and3 months. During this time dynamic changes in collagenmaturation and inflammatory responses occur that canaffect the eventual cosmetic outcomes.15-17 Our work hasalso demonstrated this clinically, with a poor association ofwound scores between 10 days and 3 months and a strongassociation of scores assigned at 3 months and at 1 year.18,19

W O U N D R E P A I RQuinn et al

6 4 8 A N N A L S O F E M E R G E N C Y M E D I C I N E 3 2 : 6 D E C E M B E R 1 9 9 8

R E S U L T S

We enrolled 135 patients with 136 lacerations during a 5-month study period beginning October 15, 1995, and end-ing March 15, 1996. Follow-up on 98 lacerations was com-pleted at 3 months (reported previously).5 We attemptedto contact all patients by telephone, and 77 of the originallacerations were available for 1-year follow-up.

We found no differences in clinical or demographiccharacteristics between glue and suture repair in the 77patients who completed 1-year follow up (Table 1). Neitherdid we find differences between groups in the early woundscores and the 3-month visual analog scores of those whodid and did not complete 1-year follow-up (86% versus91% and 67 versus 70 mm for octylcyanoacrylate andsutures, respectively).

Likewise, we found no differences in the primary out-come, visual analog cosmesis score (69 versus 69 mm,P=.95) (Figure 2), or in the percentage of optimal woundsscores (73% versus 68%, P=.60) (Figure 3) at 1 year foroctylcyanoacrylate and sutures, respectively. This studyhad sufficient power (90% with.05 α-error) to determinea 13-mm difference on the visual analog cosmesis scale.Previous work has shown the minimal clinically impor-tant difference on the visual analog cosmesis scale to be15 mm.9

We noted poor agreement between early and 3-monthwound scores (κ=.34; 95% confidence interval [CI], .10 to.58) but very strong agreement for the wound scores at 3months and 1 year (κ=.71; 95% CI, .52 to .90). A modestcorrelation was found between the visual analog cosmesisscores at 3 months and 1 year (.48; 95% CI, .30 to .63)(Table 2).

Octylcyanoacrylate Sutures(n=37) (n=40)

100

80

60

40

20

0

73 68

% Optimal Wound Scores

Figure 3.One-year wound evaluation scores.

Table 3. Indications for octylcyanoacrylate.

Anatomic Location Indication

Face Most cutaneous closuresLips and mucosa Not recommendedExtremities and torso Cutaneous closures, deep sutures recommended; not

over jointsHands and feet Minor lacerations only; generally not recommended

W O U N D R E P A I RQuinn et al

D E C E M B E R 1 9 9 8 3 2 : 6 A N N A L S O F E M E R G E N C Y M E D I C I N E 6 4 9

R E F E R E N C E S1. Stussman BJ: National Hospital Ambulatory Medical Care Survey: 1994 EmergencyDepartment Summary: Advance Data from Vital Health Statistics, No. 275. Hyattsville, MD:National Center for Health Statistics, 1996.

2. Quinn JV, Drzewiecki A, Li MM, et al: A randomized, controlled trial comparing a tissueadhesive with suturing in the repair of pediatric facial lacerations. Ann Emerg Med1993;22:1130-1135.

3. Bruns TB, Simon HK, McLario DJ, et al: Laceration repair using a tissue adhesive in a chil-dren’s emergency department. Pediatrics 1996;98:673-675.

4. Keng TM, Bucknall TE: A clinical trial of tissue adhesive in skin closure of groin wounds.Med J Malaysia 1989;44:122-128.

5. Quinn JV, Wells GA, Sutcliffe T, et al: A randomized trial comparing octylcyanoacrylate tis-sue adhesive and sutures in the management of lacerations. JAMA 1997;277:1527-1530.

6. Maw JL, Quinn JV, Wells GA, et al: A prospective comparison of octylcyanoacrylate tissueadhesive and sutures for the closure of head and neck incisions. J Otolaryngol 1997;26:26-30.

7. Singer AJ, Hollander JE, Valentine SM, et al: Prospective randomized controlled trial of anew tissue adhesive (2-octylcyanoacrylate) versus standard wound closure techniques for lacer-ation repair. Acad Emerg Med 1998;5:94-104.

8. Quinn JV: Clinical wound evaluation. Acad Emerg Med 1996;3:298-299.

9. Quinn JV, Wells G: Outcome measures in clinical wound studies: The reliability of cosmesisscales with a determination of a minimal clinically important difference. Acad Emerg Med1998;5:583-586.

10. Edlich RF: Tissue adhesives: Revisited. Ann Emerg Med 1998;31:106-107.

11. Hollander JE, Singer AJ, Valentine S, et al: Wound registry: Development and validation.Ann Emerg Med 1995;25:675-685.

12. Storrow AB, Stack LB, Peterson P: An approach to emergency department photography.Acad Emerg Med 1994;1:454-462.

13. Quinn JV, Drzewiecki AE, Stiell IG, et al: Appearance scales to measure cosmetic outcomesof healed lacerations. Am J Emerg Med 1995;13:229-231.

14. Howell JM, Newsome J, Bresnahan K: Histologic effect of butyl-2-cyanoacrylate on skin lac-erations. Acad Emerg Med 1996;3:426-427.

15. Hunt TK, Van Winkle W Jr: Normal repair, in Fundamentals of Wound Management. NewYork: Appleton-Century-Crofts,1979.

16. Trott A: Wounds and Lacerations: Emergency Care and Closure. St. Louis, MO: Mosby,1991:18.

17. Carrico TJ, Mehrof AI, Cohen IK: Biology of wound healing. Surg Clin North Am1984;64:721-733.

18. Landis JR, Koch GG: The measurement of observer agreement for categorical data.Biometrics 1977;March:159-174.

19. Streiner DL, Norman GR: Health Measurement Scales: A Practical Guide To TheirDevelopment and Use, ed 1. New York: Oxford University Press, 1991:8.

20. Hollander JE, Blasko B, Singer AJ, et al: Poor correlation of short and long term cosmeticappearance of lacerations. Acad Emerg Med 1995;2:983-987.

21. International Standards Organization, Association for Advancement of MedicalInstrumentation: Biological Evaluation of Medical Devices, vol 4. Arlington, VA, 1995.

22. Quinn JV. Tissue Adhesives in Wound Care. Hamilton, Ontario, BC: Decker, 1998.

Previous investigation has also demonstrated a similar poorcorrelation on comparison of early and long-term (6- to9-month) cosmetic outcomes.20 However, ours is the firststudy to demonstrate that minimal changes occur between3 and 12 months.

A potential limitation of this study is that fewer than60% of those lacerations originally enrolled were availablefor follow-up at 1 year. However, we are confident in ourconclusions, for several reasons. We initially overestimatedthe sample size to account for this loss to follow-up, andour sample size at 1 year still afforded us the power todetect an important clinical difference, had one existed.Furthermore, the dropout rate was similar between thestudy groups, and we found no differences in clinical ordemographic characteristics between patients in the 2study groups at 1 year. Finally, the early wound scores, 3-month visual analog scores, and clinical and demographiccharacteristics were the same in both groups for thosepatients who were not available for 1-year follow-up.

Octylcyanoacrylate (Dermabond) is the first medical-grade tissue adhesive (compared with other cyanoacrylateadhesives, which to date have been industrial-grade adhe-sives packaged for medical use). It has better tensile strengththan the available butylcyanoacrylates and is the onlycyanoacrylate adhesive that currently has met the stan-dards for nontoxic medical devices, hence its “medical-grade” designation.21,22 Caregivers should understandthe proper indications and application of octylcyanoacry-late before starting to use it. Table 3 outlines appropriateareas for its use, as based on our experience, and physi-cians should understand that proper wound care shouldnot be sacrificed.22 Wounds still may need to be anes-thetized for proper exploration and cleansing. The useof topical anesthetic agents helps achieve hemostasis andpermits proper cleansing while allowing a needleless clo-sure. Most important, tissue adhesives should always beapplied topically, never between wound edges. Althoughproficiency comes with experience, we have shown that,with minimal training, adhesives may be incorporatedinto practice and provide cosmetic outcomes as good asthose found in wounds sutured by certified emergencyphysicians.

In conclusion, there are no discernible differences inthe cosmetic outcomes of traumatic lacerations treatedwith octylcyanoacrylate tissue adhesive and sutures at 1year. Assessment of wounds 3 months after wound closureprovides a good measure of long-term cosmetic outcome.