Categories
Health

As discussed in the post “Does closure technique matter”, I rarely use sutures anymore. Tissue adhesives and/or steri-strips provide identical outcomes with much greater ease and less pain for patients. However, there are still a sizable minority of lacerations that seem to require sutures, which raises another question: should I be using absorbable or nonabsorbable sutures?
This is not a new or cutting edge topic. I have heard, in various contexts, that absorbable sutures are just as good (and therefore probably better from a patient’s perspective) since medical school. That being said, none of my residency supervisors actually used absorbable sutures, and so neither did I. There seemed to be a lingering concern about absorbable sutures and an increased risk of infection. So, what does the evidence actually say?
Let’s start with a meta-analysis from 2007 that looked for any RCT comparing absorbable to nonabsorbable suture, regardless of patient age or location (so it included both outpatient and operating room patients). (Al-Abdullah 2007) They found 7 studies, which ranged in size from 44 to 166 patients (so our conclusions are going to be uncertain whichever way they lie). The headline news was that there was no statistical difference in cosmetic outcomes, scars, infections, or dehiscence. However, a lack of statistical significance does translate to equivalence. Depending on the cosmetic measures used, point estimates were all over the place, with huge confidence intervals, and this data could easily be consistent with a big clinical difference between the two groups (in either direction). Similarly, dehiscence was not statistically different, but the OR (0.16) favoring (surprisingly) absorbable sutures was such that it certainly could be clinically significant. Infection was the only outcome with a reassuring odds ratio, at exactly 1.00 in the overall group, but when they looked at only the subgroup of traumatic wounds, absorbable sutures looked potentially better (OR 0.42 95% CI 0.007-2.51). I think the big take-away from this paper is that as of 2007, there was just not enough data available to make any firm conclusions. The headline from the abstract was no difference, but the numbers actually seem like they might favour absorbable sutures. 
The inclusion of surgical patients may be problematic for generalizability to emergency patients. Clean, perfectly straight incisions may heal differently or require different management than the dirty, traumatic lesions we see in the ED. Additionally, just comparing the broad categories of “absorbable” and “nonabsorbable” may not be fair. Each category contains numerous options, each of which might perform differently. Finally, these odds ratios are not all that helpful without some knowledge of the baseline risk or absolute numbers. A 50% increase in infection from 1% to 1.5% is very different than 20% to 30%. Looking at individual RCTs might help us understand some of these details.
Karounis (2004) is a single-center RCT from a pediatric ED where lacerations requiring repair were randomized to plain catgut or nylon. The exclusions were extensive, including any wound that could be repaired with tissue adhesive, any contamination, bites, prior keloid formation, areas of high tension, and any comorbidities that could cause problems with healing or infection. Unlike other studies, this study used a standardized care plan for most decisions. Unfortunately, that included the use of a dry dressing, when absorbable sutures theoretically require a moist environment to dissolve. (Most resources suggest either petroleum jelly or topical antibiotics when using absorbable sutures.) Both infection and dehiscence are somewhat subjective evaluations, and both were measured by an unblinded nurse (who also removed nonabsorbable sutures) about 1 week later. They include 95 patients, and the headline news was that there were no statistical differences between the groups, but the study is just too small to make any definitive claims. For the primary outcome of cosmesis at 4 months, 62% of the absorbable group had perfect outcomes as compared to 49% of the nonabsorbable (RR 0.73, 95% CI 0.45-1.17). A 13% absolute improvement or NNT of 7.5 seems like it would be clinically important. There are similar non-statistical but potentially clinically significant differences seen in both infection (0% vs 2% favoring absorbable) and dehiscence (2% vs 11% favoring absorbable). 
I discussed the RCT by Holger (2004) in the discussion of skin adhesives. The quick summary was that there was no difference in cosmetic outcomes, with moderately good outcomes (80/100), at 9-12 months in facial lacerations when comparing absorbable sutures (rapid gut) to nonabsorbable (nylon) to skin adhesive (Dermabond) in this small RCT with multiple limitations. They don’t report at all on infection or other complications. 
The next RCT was published after our original meta-analysis. It is an unblinded RCT from 2 hospitals (one pediatric only) looking at linear lacerations less than 8 cm in length, and randomizing to either Vicryl Rapide of Prolene. (Tejani 2014) They excluded facial lacerations, wounds with visible contamination, delayed presentation (>12 hours), areas of tension, and significant comorbidities. (I would have preferred a more real-world study, because I don’t expect there to be a difference in perfectly selected patients. I want to see how absorbable sutures perform in the real world, with a mix of somewhat contaminated wounds or patients with comorbidities.) It was a convenience sample of daytime patients, and there is a very high risk of selection bias (they only included 113 patients out of 4600 lacerations that presented). The primary outcome of cosmesis was judged by 2 blinded plastic surgeons at 3 months, and there was absolutely no difference, although neither result was great (54/100 vs 55/100). There were no dehiscences in either group. There was not a statistically significant difference in infection rate, but the study is way underpowered for this outcome. There were 4 infections with Vicryl Rapide (11%) as compared to 1 with Prolene (3%), an odds ratio of 0.2 (95% CI 0.1-1.9), and a difference that definitely could be clinically significant. (The direction of this finding is the opposite of that in Karounis 2004). An 8% absolute increase would be important, but also doesn’t make any sense to me in this carefully selected low risk group. I don’t see infection rates anywhere close to 11%, and I use a lot of absorbable sutures even in contaminated wounds. Their definition of infection included erythema or tenderness at the wound at 10 days, which may be irrelevant or could even represent normal healing or inflammatory response to sutures. The majority of these complications were documented through telephone follow-up, and neither the clinician nor the patient were blinded, so this is a very low reliability finding. Although the overall methodology is lower quality, this study does deserve kudos for being the only one registered with clinicaltrials.gov (NCT00933829) and for actually following the original protocol. 
There is one more single centre pediatric ED study focused on linear facial lacerations, and randomizing to fast-absorbing gut or nylon. (Luck 2008) They had similar exclusions to the other studies. They enrolled 88 patients, but only 47 completed the study. Their primary outcome was cosmesis, and was excellent and identical in this study (92 vs 93/100). There were no infections at all. There were 2 dehiscences with gut and none with nylon (although if it doesn’t impact long term cosmesis, it might not matter much.) What this study adds is a parental survey, which suggests that absorbable sutures were found to be more convenient (91% vs 75%) and were more likely to be recommended in the future (96% vs 79%). However, all 3 parents who perceived there to be a complication were in the absorbable suture group (the suture unraveled prematurely in their eyes). It is possible this ‘complication’ could be minimized with good discharge instructions. 
There are many many problems with these studies, aside from the obvious fact that they are way too small to comment on important complications. For the researchers out there, this should be a very easy RCT to get done. If doing so, please focus on making the results broadly applicable. If you exclude all the patients I see, your results will be useless to me.
The many potential sources of heterogeneity in this data make it hard to summarize. It probably isn’t fair to simply consider all absorbable sutures as a single category. Different sutures absorb at very different rates. Likewise, performance could theoretically vary among the variety of nonabsorbable sutures available to us. You could imagine a scenario where one absorbable suture is much better, but another is much worse, and when you combine all the data they all just come out looking the same. 
There are many other aspects that vary from patient to patient and trial to trial that could all impact wound healing. For this specific question, one aspect that stands out is the varying wound care instructions provided by the trials. Absorbable sutures dissolve in moist environments, not the air, so wound treatment could have a big effect. 
Although they might seem objective on paper, complications like dehiscence and infection are quite subjective. That is problematic, in that treatment and short term follow-up for all these studies was done in an unblinded fashion.
I try to keep my practice relatively simple, but different hospitals stock different sutures, so I find that this gets complicated. In my simple world, I use plain gut sutures for everything with the exception of the face, where I want something faster absorbing, and so I use a fast absorbing gut. However, sometimes I don’t have either of those options. Vicryl rapide will absorb in about the same time as plain gut, so I sometimes use that as an alternative. The other sutures that are commonly stocked (vicryl or chromic gut) take a lot longer to absorb, and so don’t really have much of a role in my practice. 
Overall, the data is too weak to draw any firm conclusions. The best guess seems to be that cosmesis is likely to be similar when comparing absorbable and nonabsorbable sutures. The data is strongly conflicting when it comes to both infection and dehiscence, with lots of bias. If anything, the current data seems like it might favour absorbable sutures. However, the primary reason we care about both of those complications is long term cosmesis, which seems to be unaffected, so the minor complications may be inconsequential. We clearly need better data, but for now either option seems fine. Ultimately, there is no one size fits all answer, and a combination of clinical judgment and shared decision making will be required. My guess is that if you engage in shared decision making, you will almost always end up using absorbable sutures.
CanadiEM: Nice threads: a guide to suture choice in the ED
PEMBlog: Absorbable versus nonabsorbable sutures
EM Lit of Note: You Can Use Absorbable Sutures Anywhere
Preparation 
Topic 1: How late is too late for suturing lacerations?
Topic 2: Irrigation
Topic 3: Should I grab sterile gloves?
Laceration Repair
Topic 4: Does eversion matter?
Topic 5: Sutures, staples, glue, steri-strips… how do I choose?
Topic 6: What’s the deal with absorbable sutures?
Topic 7: How close should I place sutures?
Aftercare
Topic 8: Can wounds get wet?
Topic 9: Dressings
Topic 10: Topical antibiotics
Topic 11: Prophylactic antibiotics for animal bites
Topic 12: Other medications/treatments for wounds
Al-Abdullah T, Plint AC, Fergusson D. Absorbable versus nonabsorbable sutures in the management of traumatic lacerations and surgical wounds: a meta-analysis. Pediatr Emerg Care. 2007 May;23(5):339-44. doi: 10.1097/01.pec.0000270167.70615.5a. PMID: 17505281
Holger JS, Wandersee SC, Hale DB. Cosmetic outcomes of facial lacerations repaired with tissue-adhesive, absorbable, and nonabsorbable sutures. Am J Emerg Med. 2004 Jul;22(4):254-7. doi: 10.1016/j.ajem.2004.02.009. PMID: 15258862
Karounis H, Gouin S, Eisman H, Chalut D, Pelletier H, Williams B. A randomized, controlled trial comparing long-term cosmetic outcomes of traumatic pediatric lacerations repaired with absorbable plain gut versus nonabsorbable nylon sutures. Acad Emerg Med. 2004 Jul;11(7):730-5. doi: 10.1197/j.aem.2003.12.029. PMID: 15231459
Luck RP, Flood R, Eyal D, Saludades J, Hayes C, Gaughan J. Cosmetic outcomes of absorbable versus nonabsorbable sutures in pediatric facial lacerations. Pediatr Emerg Care. 2008 Mar;24(3):137-42. doi: 10.1097/PEC.0b013e3181666f87. PMID: 18347489
Tejani C, Sivitz AB, Rosen MD, Nakanishi AK, Flood RG, Clott MA, Saccone PG, Luck RP. A comparison of cosmetic outcomes of lacerations on the extremities and trunk using absorbable versus nonabsorbable sutures. Acad Emerg Med. 2014 Jun;21(6):637-43. doi: 10.1111/acem.12387. PMID: 25039547
In the first edition of the research roundup for 2019 we are going to cover pediatric head injury, oral antibiotics for complicated […]
A key scientific lesson that I have discussed numerous times on this blog is that the interpretation of any new study must […]
October 2020. A month that many have declared the worst of all time, but can it really be that bad if you […]
Another randomized control trial of “medical expulsive therapy”


© 2023 First10EM All rights reserved
Designed with Customizr Pro

source

Leave a Reply

Your email address will not be published. Required fields are marked *