Revisiting the Role of Amniotic Membrane Dressing in Acute Large Traumatic Wounds: A Randomized Feasibility Study at a Level 1 Trauma Centre

Abstract

Introduction:

Acute large traumatic wounds require temporary dressing prior to the definitive soft tissue reconstruction, as the physiological derangement during the immediate postinjury period delays the definitive surgical intervention. Selecting an ideal dressing material from numerous available synthetic dressings and skin substitutes poses a challenge. Although amniotic membrane (AM) scaffold has a definitive role in promoting wound healing in burns and chronic wounds, however, its efficacy in acute large traumatic wound is lacking. The present trial aimed to evaluate the safety and efficacy of AM in wound bed preparation before the definitive soft-tissue reconstruction in acute large traumatic wounds.

Methods:

Sixty patients with acute large traumatic wounds (>10 cm × 10 cm) were divided into two groups (conventional dressing and AM dressing) using simple mixed block randomization. Wounds were assessed using the Bates Jensen Score at various timelines for the signs of early wound healing. The primary outcome was to evaluate the time taken for the wound bed preparation for definitive soft-tissue reconstruction. The secondary outcome was the pain assessment and complications, if any.

Results:

There was significant reduction in the wound exudate as well as peripheral tissue edema in the intervention group (P = 0.01). AM dressing was significantly less painful (P = 0.01). The incidence of wound infection and need for debridement was decreased in the intervention group. However, the time interval to definitive soft-tissue coverage was statistically insignificant and comparable in both the groups. No adverse reactions were seen in either group.

Conclusion:

AM dressings are safe and efficacious with significant reduction in wound exudates and peripheral edema. However, these dressings do not hasten the wound maturation as compared to conventional dressings. AM dressings can be used as a less painful alternative to conventional dressing in the management of large acute posttraumatic wounds.

INTRODUCTION

Posttraumatic soft-tissue injury is an important cause of morbidity. Majority of these wounds are complex in nature due to the involvement of different tissue planes with or without tissue loss. Management of these wounds requires regular wound dressing before the definitive cover either by a skin graft or a flap. Numerous advancements are being made in acute management of soft-tissue injury in the form of synthetic dressings and skin substitutes, but remain out of reach to the vast majority of populations, especially low-middle income countries (LMICs).[1,2] With the availability of a variety of options for dressing in today’s world, it becomes necessary to select an ideal option. The most important issue in large/deep wound is to find a temporary cover that is comfortable and decreases pain, avoids infection, and possibly promote wound healing. The best available option for the management of large traumatic wounds is an autologous skin graft/flap but is invasive and can be detrimental in the immediate postinjury period.

Amniotic membrane (AM) is used as a biological dressing for decades and is easily available.[3] Its regenerative properties such as angiogenic, anti-inflammatory, anti-fibrotic, anti-microbial, and nonimmunogenic augment the physiological process of wound healing.[4,5,6] Various studies have highlighted its efficacy in promoting the healing of small and superficial chronic wounds and burn wounds.[7,8,9] However, there is a paucity of literature on its use in acute large traumatic wounds.[10,11,12]

A randomized control pilot study was conducted at a level 1 Trauma Center of LMIC to evaluate its safety and efficacy as an alternative wound dressing in an acute large posttraumatic wound. We hypothesized that AM scaffold will shorten the time duration to the final definitive soft-tissue reconstruction of these wounds by promoting early wound bed maturation.

METHODS

Study design

This was an open-label, pilot, experimental, randomized control study carried out under the Division of Trauma Surgery and Critical Care, at a level 1 Trauma Center, from August 2019 to November 2021, after obtaining ethical clearance from the institute.

The present study included 60 patients of either sex aged between 18 and 60 years with traumatic soft-tissue injuries having a wound dimension of 10 cm × 10 cm or more, requiring in-patient care with regular wound dressings till the definitive soft-tissue reconstruction either by skin grafting or the flaps cover. The exclusion criteria were any injured patients with comorbidities, skin ailments, immunodeficiency disorders, on steroids, significant associated injuries, infected wounds, and sepsis.

All patients with acute traumatic soft-tissue injuries were screened for eligibility for recruitment, based on the inclusion and exclusion criteria of the study. Prior consent for participation in the trial was taken from patients or their authorized legal representative. The eligible patients were randomized into two cohorts using simple mixed block randomization, to maintain the balance in the intervention and control group. Allocation concealment of the patients was done by sequentially numbered opaque-sealed envelopes. The envelopes were prepared by a person not associated with the conduct of the study. The envelopes were sequentially opened.

In Group I (intervention group), AM dressing was applied to the wound, which was held in place with a secondary dressing that was changed daily. In Group II (conventional dressing group) patients, conventional wound dressing using normal saline was done, which was changed daily. In both groups, the wound was assessed on each dressing by senior surgical residents using the Bates Jensen Score (BJS) wound assessment tool and was followed till the wound bed was ready for the definitive soft tissue reconstruction (skin grafting/flap).[13] The BJS wound assessment tool consider multiple wound parameters including wound size and depth. The wound size was measured along the longest and widest aspect of the wound surface using ruler in centimeters. The wound depth was assessed on the basis of extent of tissue damage, whether involving skin and/or additional tissue like muscle, tendon, nerve and joint capsule. The eligibility of the wound bed for the definitive cover was based on the clinical decision of the same treating surgeon in both the cohorts. During this assessment period, pain assessment was also done with the Visual Analog Scale (VAS).[14] The time frame for the outcome analysis was day 1 (initial baseline), subsequently on the 7^th, 14^th, and or 28^th day of the treatment period, whichever was earlier.

Any adverse reactions and signs of wound infection in either group such as purulent discharge with a foul odor were noted and a wound swab evaluation from the suspected area was done when required.

Placenta was obtained from patients/subjects undergoing an elective caesarean section after taking written consent. The donor blood serological testing was performed prior to surgery, the placenta was used to harvest only if the serological test was negative for HIV, Hepatitis B and C and venereal disease research laboratory at baseline. A repeat serological test was also done after 3 months.

All the eligible placenta was collected in a sterile storage container-containing 0.9% saline solution (NaCl) with 1% Antibiotic-Antimycotic (Gibco, USA) transport media (constituents of the media) and it was transported to Stem Cell Facility (DBT Centre of Excellence for Stem Cell Research) AIIMS. Placenta was processed for AM preparation in current Good Manufacturing Practices lab as per the standardized protocol. Placenta was transferred to a sterile tray under laminar flow hood and amnion was separated from the chorion by blunt dissection. Membrane was washed thoroughly with normal saline to remove the blood clots. Following this, membrane was placed on Sterile tray and spread with the stromal surface up. AM was de-epithelialized using 0.25% Good Manufacturing Practices-grade trypsin–ethylenediaminetetraacetic acid (T-EDTA) (Gibco, USA).[15] Briefly, AM was incubated with T-EDTA for 30 min at 37°C followed by gentle mechanical scraping and washing with Phosphate Buffered Saline thrice. Sterile Nitrocellulose paper was used to fold the membrane while cutting it into small pieces and this was transferred on the exposed stromal surface of the AM into the vial containing cryopreservation media 1:1 mixture of glycerol (glycerol 85%) and Dulbecco’s modified Eagle medium High Glucose (Gibco, USA). All AM used in the study were of wet type.[16] A small piece of AM was sent for bacterial and fungal sterility test. All the vials were labelled and sealed with paraffin and stored at − 80°C in deep freezer for further use [Figures 1 and 2].

Figure 1.

Process of amniotic membrane preparation

Figure 2.

Depicting amniotic membrane harvesting, processing, storage and application as dressing scaffold on traumatic wound

For transport, vials were shifted in cold boxes. AM was immediately processed in a sterile condition for application over the wound. AM was washed with normal saline in a sterile tray. The obtained AM was applied on a large posttraumatic wound with the epidermal layer in apposition and the stromal layer on the outside [Figure 3].[17]

Figure 3.

Intra-operative application of amniotic membrane, (a) Normal saline wash, (b) Spreading of amniotic membrane, (c) Application over raw area, (d) View after application

Statistical analysis

Descriptive statistics was used to analyse the data. All the data were analysed with the help of STATA 16.0 (StataCorp, College Station, TX, USA) software. Quantitative variables were presented as in percentage and in absolute numbers. Quantitative data were tested for normality and presented in mean ± standard deviation if normality fulfilled otherwise presented in median (interquartile range [IQR]). Kolmogorov–Smirnov test was used to test the normality of quantitative variables. A Student’s t-test was used to compare/observe the differences in the quantitative variable between the groups if normal otherwise the Mann–Whitney U-test was performed to compare the nonnormal variables. Chi-square test/Fisher’s exact test was used to establish the association between groups and the variables. A value of P < 0.05 was considered to represent the statistical significance of the study.

RESULTS

A total of 138 patients with acute large traumatic wounds (>10 cm × 10 cm) were assessed for eligibility during the study period. Sixty patients who met the inclusion criteria were randomized into two equal cohorts. Sixty-four patients refused participation while the rest 14 patients were excluded due to associated comorbidity or concomitant head injuries. All the recruited patients were analyzed for the outcomes till the definitive cover of the wound with no loss to follow-up [Figure 4].

Figure 4.

CONSORT flow diagram of trial participants. AM: Amniotic membrane

Majority of the patients were young males (95%). The median age of the patients in both groups was comparable (P = 0.32). The median injury severity scores (ISS) were also comparable between the groups with a median ISS of 13 (IQR 10–17.25) in the intervention group and 15.5 (IQR 10.75–23.25) in the control group (P = 0.77). Road-traffic injury was the most common mode of injury noted in 83.3% followed by fall from height (6.6%) and machine/glass cut injuries (6.6%) [Table 1].

Table 1.

Demographics of study population

Variables	Intervention (AM) group (n=30), n (%)	Control (conventional dressing) group (n=30), n (%)	P
Age (years), median (IQR)	30 (25−45)	35 (24.75−41.25)	0.32
ISS, median (IQR)	13 (10−17.25)	15.5 (10.75−23.25)	0.77
Sex (males)	27	30	-
Mechanism of injury
RTI	25 (83.3)	25 (83.3)	-
FFH	1 (3.33)	3 (10)
Railway associated injury	2 (6.66)	0
Machine/glass cut injury	2 (6.66)	2 (6.66)

AM: Amniotic membrane, IQR: Interquartile range, RTI: Road traffic injury, FFH: Fall from height

The wound assessment on day 1 of the recruitment, the median BJS was comparable between the intervention group (35.5 IQR 32.75–40.75) and (37 IQR 35.75–40) in the control. In both, groups, reductions in BJS were noted across the predefined timeframes of the analysis (i.e., day 1, 7, 14, and or on the day of definitive cover whichever was earlier) and were comparable (P = 0.35, 0.18, 0.37, and 0.12, respectively). The median BJS was comparable between groups (28 [IQR 26–30] in the intervention and 26 [25–30] in the control group, P = 0.12) at the time of the definitive coverage of the wound [Table 2].

Table 2.

Timeline assessment of the large acute post traumatic wounds

	Intervention (AM) group	Control (conventional dressing) group	P
Wound assessment-day 1 BJS, median (IQR)	35.5 (32.75−40.75)	37 (35.75−40)	0.35
Wound assessment-day 7 BJS, median (IQR)	30 (27−37.25)	31 (25.75−34.25)	0.18
Wound assessment-day 14 BJS, median (IQR)	28 (27−32)	30 (26−33)	0.37
Time interval to definitive cover in days-median (IQR)	10 (7−16.5)	11.5 (8.75−15)	0.76
BJS at definite cover, median (IQR)	28 (26−30)	26 (25−30)	0.12
Total patients who underwent definitive cover by day 7, n (%)	9 (30)	6 (20)	0.37
Total patients who underwent definitive cover within day 14, n (%)	21 (70)	24 (80)	0.37
Total patients who underwent definitive cover beyond day 14, n (%)	9 (30)	6 (20)	0.37

BJS: Bates Jensen score, IQR: Interquartile range, AM: Amniotic membrane

The median time to the definitive cover in the intervention group was 10 days (IQR of 7–16.5 days), whereas in the conventional group, it was 11.5 days (IQR of 8.75–15 days) and was comparable (P = 0.76). However, the time to the definitive cover was reduced by 1½ days in the intervention group. On the 7^th day of assessment, 30% (n = 9) of patients in the intervention group underwent definitive cover as compared to only 20% in the control group (n = 6). However, on the 14^th day or beyond, 10% more cases in the control group underwent the definite cover and were statistically insignificant (P = 0.37).

Of the 13 elements of the BJS, statistically significant changes across the timelines were noted in two variables. The exudative amount was significantly less in AM group as compared to the conventional dressing group on day 14 (P = 0.01). Peripheral tissue edema was also significantly less in the AM group on day 7 (P = 0.01). Skin color surrounding the wound also showed a reduction in the AM group but was not statistically significant (P = 0.07). Other elements of the BJS did not show any significant variation across timelines among the two study groups [Table 3].

Table 3.

Timeline analysis of the parameters of Bates Jensen score for wound assessment

Parameters	BJS on day 1			BJS on day 7			BJS on day 14
	AM	CG	P	AM	CG	P	AM	CG	P
Size	3.5	3.65	0.31	3.43	3.65	0.19	3.6	3.44	0.27
Depth	3.12	3.2	0.31	3.08	3	0.90	2.9	3	0.20
Edges	2.87	2.66	0.17	2.37	2.37	1	2.09	2.1	0.48
Undermining	2.54	2.33	0.28	2.16	2.04	0.33	1.81	2.1	0.24
Necrotic tissue type	2.08	2.04	0.44	1.83	1.7	0.3	1.72	1.7	0.46
Necrotic tissue amount	2.04	2.04	1	1.83	1.5	0.11	1.54	1.5	0.42
Exudate type	2.83	2.79	0.44	2.6	2.2	0.11	2.63	2.7	0.43
Exudate amount	2.83	3.16	0.06	2.91	2.25	0.41	2.09	2.3	0.01
Skin colour surrounding wound	1.95	2.37	0.07	1.25	1.3	0.31	1	1.2	0.07
Peripheral tissue oedema	6.4	6.06	0.29	3.6	3	0.01	4.21	3	0.08
Peripheral tissue induration	2.56	2.43	0.9	1.76	1.8	0.9	1.5	1.61	0.9
Granulation tissue	4.36	4.33	0.9	2.96	3.03	0.9	2.5	2.53	0.9
Epithelialization	4.6	4.73	0.9	4.13	4.4	0.9	4	4	1

BJS: Bates Jensen score, AM: Amniotic membrane, CG: Current good

The pain assessment using the VAS score was significantly less in the intervention group on day 7 (median VAS 3 with IQR 2–5) (P = 0.01). The rate of wound infections was 26.7% (n = 8) in the intervention group while in the control group it was 40% (n = 12) (P = 0.27). The need for wound debridement was 13.30% and 20% in the intervention and control groups respectively and was comparable (P = 0.48). No adverse effects were noted.

All patients underwent definitive soft tissue reconstruction primarily by skin grafting in both the cohorts, except one patient in the intervention cohort, who underwent a regional flap [Figure 5].

Figure 5.

Spectrum of wounds after amniotic membrane application

DISCUSSION

In the present study, we analysed the efficacy of AM dressing in acute large traumatic wounds that required definitive cover either by skin graft/flap as only a few case reports have analysed its healing potential in acute traumatic wounds.[10,11,12]

A statistically significant reduction in the amount of wound exudate and peripheral tissue edema was observed in the intervention group in the present study (P = 0.01). Erythema surrounding the wound margin also showed a reduction however, this finding could not achieve statistical significance [Table 3]. In concurrence with our results, Dehghani et al., in their randomized controlled trial (RCT) on the efficacy of AM dressing in pressure sores have reported a significant reduction in wound discharge and faster wound healing following AM application.[18] Similarly, Loeffelbein et al. in their prospective clinical trial on AM dressing on the split thickness skin graft (STSG) donor site also reported a significant reduction in wound exudates with fewer dressing changes.[19] Except for these 3 elements of BJS, no other elements showed any significant change as in our study as we analysed only the early wound healing [Table 3]. Majority of the studies on the efficacy of AM dressing have been done on various chronic wounds of different etiologies where they have assessed complete wound healing.[7,8] BJS was used to assess wound healing in our study as it has a wider representation of variables for the sequential wound assessment.[13,20]

In the present study, the incidence of wound infection, and the requirement of wound debridement were reduced by 16.3% and 6.3% in the intervention group respectively but were statically insignificant (P = 0.27 and 0.48 respectively). However, literature has highlighted the antibacterial potential of AM due to the inhibitory effects on the diverse panel of bacteria.[5,6] Ghalambor et al. have in their prospective study reported a significant reduction in the incidence of wound infection following AM application (P < 0.001).[21] The reduction in the exudate amount and peripheral tissue edema also highlight the anti-inflammatory potential of the AM dressing as reported in the literature.[4]

Sawhney in their study on AM as a biological dressing on burn wounds, have reported significantly faster mean healing time with amnion coverage than controls.[22] Dehghani et al., in his RCT too have reported faster partial as well as complete healing (P < 0.03 and P < 0.001 respectively) with the use of AM.[18] Branski et al., in their study on pediatric patients also reported a faster healing time in the amnion dressing group.[23] In the present study, the overall time to the definitive cover in the intervention group was decreased by 1½ days but was statistically insignificant (P = 0.76). Thirty percentage patients in the intervention group while only 20% of patients in the conventional group underwent early definite cover by day 7 with a comparable median BJS (P = 0.37). We hypothesize this observation to be clinically encouraging. However, a larger study is required to validate this observation in acute wounds [Table 2].

Kesting et al., in their review article, highlighted the analgesic potential of AM dressing. The AM scaffold is proposed to act as a protective physical barrier over the exposed nerve endings thereby reducing pain.[9] In addition, they also reduce the concentration of inflammatory cytokines in the raw area. Adly et al. in their study on patients with second-degree burn found AM dressing to be significantly less painful than polyurethane membrane dressing.[24] In the present study, a significant reduction in pain score was observed in the intervention group on the 7^th day of evaluation (P = 0.01). While on the 14^th day of assessment, the pain score was also less but was not statistically significant (P = 0.08). This observation highlights the analgesic potential of AM dressing in acute wounds.

Although considered safe, complications with the use of AM have been reported, like the risk of transmission of bacterial, viral, or fungal infections to the recipient if the donors are not adequately screened or due to improper processing and storage of AM.[17,25,26] No adverse reactions were noted with the use of AM in the present study.

The age of the patient and the injury severity among the study population were analysed as they have prognostic significance in clinical management.[27] The median age and the ISS were comparable between the group with a P = 0.77 and P = 0.32 respectively.

Limitations

To our best knowledge, this is the largest RCT analysing the role of AM in acute large traumatic wounds. The preliminary results suggest that AM dressing is safe and efficacious in large acute wounds. We do want to acknowledge the limitations of the study. This was a single centre, nonblinded study with a limited sample size. The results of the study are although promising but cannot be generalized. At the same time, being a pilot study, economic implication was not assessed; however cost can be estimated in further study done on the basis of encouraging preliminary data obtained in the current study.

CONCLUSION

AM application for the dressing of acute large traumatic wounds decreased the exudate amount and peripheral tissue oedema along with a good analgesic effect as compared to conventional dressing. These dressings do not hasten the maturation of wound bed prior to definitive soft tissue reconstruction. However, being a pilot study, further validation of its role in reducing the time to the definitive cover, the incidence of wound infection, and the rate of wound debridement needs a larger study.

Research quality and ethics statement

This study was approved by the Institutional Review Board AIIMS New Delhi vide Number (IEC 533/05.10.2018 RP-8/2018). The authors followed applicable EQUATOR Network (https://www.equator-network.org/) guidelines during the conduct of this research project.

Financial support and sponsorship

This study was financially supported by AIIMS intramural grant.

Conflicts of interest

There are no conflicts of interest.