See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/343862920

The use of moist exposed burn ointment (MEBO) for the treatment of burn
wounds: a systematic review

Article  in  Journal of Plastic Surgery and Hand Surgery · August 2020

DOI: 10.1080/2000656X.2020.1813148

CITATION READS

1 424

3 authors, including:

Nigel Tapiwa Mabvuure Christopher Felix Brewer

Norfolk and Norwich University Hospitals NHS Foundation Trust Cambridge University Hospitals NHS Foundation Trust
63 PUBLICATIONS   657 CITATIONS    19 PUBLICATIONS   111 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

facial palsy View project

Facial Palsy View project

All content following this page was uploaded by Nigel Tapiwa Mabvuure on 16 March 2021.

The user has requested enhancement of the downloaded file.

IPHS#1813148, VOL 0, ISS 0

The use of moist exposed burn ointment (MEBO) for the treatment of
burn wounds: a systematic review
Nigel Tapiwa Mabvuure, Christopher Felix Brewer, Kevin Gervin, and Siobhan Duffy

QUERY SHEET

This page lists questions we have about your paper. The numbers displayed at left are hyperlinked to the location of the query in
your paper.

The title and author names are listed on this sheet as they will be published, both on your paper and on the Table of Contents. Please
review and ensure the information is correct and advise us if any changes need to be made. In addition, please review your paper as
a whole for typographical and essential corrections.

Your PDF proof has been enabled so that you can comment on the proof directly using Adobe Acrobat. For further information on
marking corrections using Acrobat, please visit http://journalauthors.tandf.co.uk/production/acrobat.asp; https://authorservices.
taylorandfrancis.com/how-to-correct-proofs-with-adobe/

The CrossRef database (www.crossref.org/) has been used to validate the references.

AUTHOR QUERIES

Q1 Please confirm the article type as set in the proof is accurate.

Q2 Please confirm the authors’ affiliations as set in the proof are accurate.
Q3 Please provide the complete physical address of the corresponding author.
Q4 Please provide an "Abbreviations" list for the article as it is mandated by the journal style.
Q5 Please provide the volume number and page range for reference [2] in the references list.
Q6 Please provide complete details for reference [15] in the references list.
Q7 Please provide the volume number and page range for reference [28] in the references list.
 JOURNAL OF PLASTIC SURGERY AND HAND SURGERY
1 64
https://doi.org/10.1080/2000656X.2020.1813148
2 65
3 66
4 REVIEW ARTICLE 67
5 68
6 The use of moist exposed burn ointment (MEBO) for the treatment of burn 69
7 wounds: a systematic review 70
8 71
9 72
10
Nigel Tapiwa Mabvuurea, Christopher Felix Brewera, Kevin Gervinb and Siobhan Duffyc 73
11 a
St. Andrews Centre for Plastic Surgery, Broomfield Hospital, Broomfield, UK; bBelfast Health and Social Care Trust, Belfast City Hospital, Belfast, 74
12 UK; cGreater Glasgow and Clyde NHS Health Board, Glasgow, UK 75
13 76
Q2
14 77
15 ABSTRACT ARTICLE HISTORY 78
16 Moist exposed burn ointment (MEBO) is an oil-based herbal paste, purported to be efficacious in manag- Received 1 April 2020 79
ing burn wounds and more commonly used in Asia and North Africa. A PRISMA-compliant systematic Revised 21 July 2020
17 80
review was performed to analyse the evidence for the use of MEBO on burn wounds. The wound healing Accepted 17 August 2020
18 81
19 rate was the primary outcome of interest. PubMed-listed randomised controlled trials (RCTs) comparing 82
KEYWORDS
the efficacy of MEBO with placebo, standard care or other therapies in the treatment of partial thickness
20 MEBO; MEBT; burn; moist 83
burns in adults and children were eligible for inclusion (November 2019). Six RCTs were eligible. The exposed burn ointment;
21 majority of trials comparing wound healing between MEBO and SSD favoured MEBO (two of three). There 84
moist exposed burn
22 may be improved healing in MEBO-treated wounds vs. those treated with povidone-iodine þ bepanthenol 85
23
Q4 therapy; review
86
cream. There was no difference between MEBO and Acquacel Ag, but Helix Aspersa had faster healing
24 rates than MEBO. However, all evidence was from moderately to poorly reported trials with a high risk of 87
25 bias. In conclusion, the evidence for MEBO in English-language literature was poor and inconsistent with 88
26 respect to wound healing rate and analgesis compared to 1% SSD, Acquacel Ag, Helix aspersa cream and 89
27 povidone-iodine þ bepanthenol cream. Blinded RCTs comparing MEBO to both placebo and other com- 90
28 mon topical treatments may further improve the confidence in concluding their analysis. There is good 91
29 evidence that MEBO is as safe as its comparators as shown by the low complication rate. 92
30 93
31 94
32 95
33 96
Introduction supporting the use of MEBO for treating a wide range of ailments
34 97
[15]. Ang et al. noted that some hospitals in China have reported
35 There are several nonsurgical treatment options for superficial 98
survival rates reaching 90% in patients receiving MEBO following
36 and partial thickness burns. They include inorganic and biosyn- 99
burn injuries covering 40–80% of their total body surface area
37 thetic dressings, topical treatments such as silver sulfadiazine 100
(TBSA) [16]. These claims, including those that MEBO prevents
38 (SSD) and alternative/complementary therapies such as honey. 101
shock in burned patients [17] if confirmed would make this herbal
39 The evidence for many of these treatments, however, remains 102
paste an important addition to burn management algorithms.
40 weak. A 2013 Cochrane review by Wasiak et al. assessed the effi- 103
Despite these claims, it has been noted there is a paucity of
41 cacy of a range of dressings for treating burn wounds could not 104
scientifically rigorous studies assessing MEBO’s efficacy for treating
42 draw firm conclusions due to the lack of adequately powered 105
burn wounds [16]. The Cochrane review by Wasiak et al. [1] did
43 high-quality trials [1]. Similarly, the 2013 Cochrane review by 106
not include an assessment of the efficacy of MEBO although one
44 Hoogewerf et al. also failed to draw conclusions on topical treat- 107
nonrandomised comparative study of MEBO vs. Aquacel Ag by
45 ments for facial burns owing to a paucity of high-quality evidence 108
Mabrouk et al. [18] was listed as awaiting assessment. Hoogewerf
46 [2]. The optimal treatment options for superficial and partial thick- 109
47 et al.’s Cochrane review included one study comparing MEBO 110
ness burns remain controversial and subject to an active with SSD for routine care of facial burns [19], however noting the
48 investigation. 111
49 small sample size and absence of intention to treat analysis as 112
Moist exposed burn ointment (MEBO), an oil-based herbal barriers to forming evidence-based conclusions [2]. Similarly, the
50 paste, was developed in 1989 in Beijing. It remains a popular top- 113
51 Cochrane review by Norman et al. [20] concluded there was only 114
ical treatment for burn wound treatment especially in Asia and low/very low-grade evidence relating to the effect of MEBO on
52 115
the Middle East. The main ingredient in MEBO is beta-sitosterol, a the incidence of infection and wound healing time due to study
53 116
plant-derived sterol with reportedly anti-inflammatory and anti- imprecision and reporting inconsistency in the included trials
54 117
pyretic properties [3–5]. The oily component of MEBO is thought [21,22]. No systematic reviews of studies assessing MEBO have
55 118
to improve wound moisture retention. Given the recognised been published to date.
56 119
importance of moisture in wound healing [6–8], the hypothesis For these reasons, this systematic review was performed to col- 120
57
58 that MEBO positively affects wound healing appears intuitive. late and analyse the evidence for the use of MEBO from English- 121
59 Numerous anecdotal reports [9,10], animal studies [11,12] and language literature. The clinical question, with reference to PICOS, 122
60 non-comparative reports [13,14], concluded that MEBO was effica- was as follows: 123
61 cious in treating burns and other wound types. Several more  Participants: Patients with burn wounds, 124
62 reports have been published in Chinese literature with authors  Intervention: Treatment of burn wounds with MEBO, 125
63 126
Q3 CONTACT Nigel Tapiwa Mabvuure n.mabvuure@doctors.org.uk St. Andrews Centre for Plastic Surgery, Broomfield Hospital, Broomfield CM1 7ET, UK
ß 2020 Acta Chirurgica Scandinavica Society
 2 N. T. MABVUURE ET AL.

127 Secondary outcomes of interest were post-dressing pain reduc- 190

128 tion, complications and wound infections. 191
129 192
130 Search strategy 193
131 194
132 The EMBASE and MEDLINE databases were searched from incep- 195
133 tion to November 2019 using the term: (MEBO OR MEBT OR 196
134 ‘moist exposed burn therapy’ OR ‘moist exposed burn treatment’ 197
135 OR ‘moist exposed burn ointment’).ti,ab. The search was dupli- 198
136 cate-filtered and limited to human studies reported in English. 199
137 200
138 Study selection 201
139 202
140 Two authors independently assessed titles and abstracts for rele- 203
141 vance and verified by a third. 204
142 205
143 Data extraction 206
144 207
145 Data extraction was performed by one author and independently 208
146 verified by two others. Data extracted from each study included 209
147 bibliometric indices (authorship, year of publication, the country 210
148 in which study was conducted and type of study), anatomical 211
149 area, the extent of burn injury (TBSA), population characteristics 212
150 and outcomes. 213
151 214
152 215
Assessment of the risk of bias in included studies
153 216
154 The risk of selection, performance, detection, attrition, reporting 217
155 and other biases in each study was assessed using the Cochrane 218
156 Collaboration’s risk of bias (RoB) assessment tool (RevMan version 219
157 5.2.11, Cochrane Collaboration) [23]. 220
158 Descriptive statistics were used to synthesise data. 221
159 222
Figure 1. PRISMA flow chart.
160 223
Results
161 224
162  Comparisons: Comparison to either standard of care Search results 225
163 or placebo, 226
 Outcomes: Effect on wound healing measures. The search results are summarised in Figure 1. Eight articles
164 227
 Study design: Comparative studies. remained after applying the exclusion criteria [16,17,19,22,24–27].
165 228
However, since three of these articles all reported data from a sin-
166 229
gle RCT by Ang et al. [16,19,27], their data were pooled and were
167 230
Methods treated as one article. A similar approach was taken by Dat et al.
168 231
in a Cochrane review of studies of Aloe vera for treating acute
169 A systematic review was performed following PRISMA guidelines. 232
and chronic wounds [28]. Therefore, the eight eligible articles rep-
170 The review protocol was not published or registered prospectively 233
resent data from six RCTs.
171 on a registry. 234
172 235
173 Characteristics of included studies 236
174 Inclusion and exclusion criteria 237
All studies provided level 3 evidence on the Oxford scale. The
175 238
Randomised controlled trials (RCTs) comparing the efficacy of included RCTs (578 patients) were undertaken in Greece [22,26],
176 239
Germany [17], Singapore [16,19,27] and Egypt [24,25] between
177 MEBO with placebo, standard wound care or other therapies in 240
2002 and 2011 (Table 1). Tsoutsos [26], Hindy [24] and their
178 treating superficial and partial-thickness burn wounds in adults 241
respective teams included only patients with facial burns whilst
179 and children were eligible. Non-comparative studies, retrospective 242
Allam et al. [25] included only patients with hand burns. The
180 studies, reviews, animal studies, expert opinion articles and pre- 243
remaining studies did not specify the affected anatomical region
181 liminary reports were excluded. Economic analyses were only 244
[16,17,19,22,27]. Only two trials included patients with non-ther-
182 included if they also investigated clinical outcomes. 245
mal burns although the aetiology remains unclear [17,25].
183 246
The depths of burns in the patients included in the trials were
184 247
deep partial thickness (DPT) [26], superficial partial thickness (SPT)
185 Outcome measures 248
[24] and two studies included patients with both SPT and DPT 249
186
The primary outcome of interest was the effect of MEBO on burns [22,25]. Two trials included patients with partial-thickness
187 250
wound healing. Therefore, time to wound healing, wound healing burns but did not specify whether these were SPT or DPT
188 251
rate, transepidermal water loss (TEWL) and reduction in wound [16,17,19,27]. Only Ang et al. [16,19,27] detailed the method used
189 252
surface area were considered primary outcome measures. to determine burn depth.
 315
314
313
312
311
310
309
308
307
306
305
304
303
302
301
300
299
298
297
296
295
294
293
292
291
290
289
288
287
286
285
284
283
282
281
280
279
278
277
276
275
274
273
272
271
270
269
268
267
266
265
264
263
262
261
260
259
258
257
256
255
254
253

Table 1. Summary of eligible studies.

Number of
participants Result
First author, year (Male:Female) Anatomical Mean age in Study
(Country of origin) (Extent of burn) Burn thickness Burn aetiology region years (range) design Comparison(s) Outcome (measure) MEBO Comparison
Hirsh, 40 adults (M25:F15) Partial thickness Thermal Not specified 41.15 (20–65) RCT MEBO vs. 1% SSD Day 12 Wound healing (mean TEWL): 11 7
2008 (Germany) (<20% TBSA) (n ¼ 20 each) Day 0 Inflammation (WCC and CRP): NR NR
Day 8 inflammation (WCC and CRP): NR NR
Pain (mean VAS score D1): 5 5
Pain (mean VAS score D12): 3.8 3.5
Adverse effects: 0 0
Hindy, 2009 (Egypt) 60 patients (<25% Superficial Thermal Facial NR RCT Aquacel Ag (I) vs. MEBO Mean time to complete healing (II vs I 10.35 ± 2.8 10.05 ± 2.3,
TBSA in >12- partial thickness (II) vs. saline (III) vs III): 12.05 ± 2.4¥
years-old, <15% (n ¼ 20 each) Pain (mean VAS score D1 & D2): 3.1 ± 1.9 NR
TBSA in <12- Itching (proportion itch-free) (%): 65 25, 10¥
years-old) Adverse effects: NR NR
Ang, 115 patients Partial thickness Thermal Not specified 22 RCT MEBO (n ¼ 57) vs. 1% Wound healing rate (days to 17 20
2002 (Singapore) (<40% TBSA) SSD (n ¼ 58). 75% healing):
Pain (mean week 1 post-dressing 2.9 3.5
verbal numerical rating scale):
MRSA infection rate (%): 37.4 38.5
Adverse effects NR NR
Allam, 2007 (Egypt) 106 patients Partial thickness Thermal Hand NR RCT MEBO vs. 1% SSD cream Mean days to healing (superficial 10.48 ± 2.66 14.53 ± 3.83
(<25% TBSA) and a polyethylene partial thickness)
bags (n ¼ 53 each) Mean days to healing (deep 30.50 ± 5.10 36.60 ± 5.08
partial thickness)
Pain Score 0 (%): 35.85 33.96
Pain Score 1 (%): 47.17 41.51
Pain Score 2 (%): 16.98 24.53
Pain Score 3 (%) 0 0
Adverse effects NR NR
Carayanni, 211 adults Partial thickness Thermal Not specified 42.68 (18–75) RCT MEBO (n ¼ 104:50 deep Time to wound healing (days to 50% 8.7 10.75
2011 (Greece) (<15% TBSA) partial and 54 reduction in TEWL for superficial
superficial partial) vs. partial thickness burns)
Povidone Reduction in admission for deep 3.02 2.79
iodine þ bepanthenol partial thickness burns (days)
cream (n ¼ 107: 52 Pain (mean morning VAS on day 2) 3 4.2
deep partial and 55
superficial partial) Pain (mean evening VAS on day 2) 3.8 4.7
Adverse effects (allergic reactions, (%) 10.6 7.5
Wound infection rate (%) 5.8 7.5
Tsoutsos, 46 adults (extent Deep partial Not specified Facial (20–70) RCT Helix aspersa extract Days to full epithelialisation 15 ± 3 11 ± 2
2009 (Greece) not reported) thickness (Elicina cream) (27) vs. (photographic assessment)
MEBO (n ¼ 16) Days to eschar detachment 11 ± 2 9 ± 2
D4 pre-intervention pain score (VAS) 6.50 ± 0.89 6.22 ± 1.25
D4 30 mins post-intervention pain 4.50 ± 0.52 3.52 ± 0.80
score (VAS)
Adverse effects (allergic reactions) 0 0
Abbreviations. CRP: C-reactive protein; MEBO: moist exposed burn ointment; SSD: silver sulfadiazine; TBSA: total body surface area; TEWL: transepidermal water loss ;VAS: visual analog scale; WCC: white cell count.

378
377
376
375
374
373
372
371
370
369
368
367
366
365
364
363
362
361
360
359
358
357
356
355
354
353
352
351
350
349
348
347
346
345
344
343
342
341
340
339
338
337
336
335
334
333
332
331
330
329
328
327
326
325
324
323
322
321
320
319
318
317
316
 4 N. T. MABVUURE ET AL.

379 442
380 443
381 444
382 445
383 446
384 447
385 448
386 449
387 450
388 Figure 3. Risk of bias graph: review authors’ judgements about each risk of bias 451
389 item presented as percentages across all included studies. 452
390 453
391 concealed allocation. Across most trials, the risk of selection bias 454
392 is unclear (Figure 3). 455
393 None of the trials had a low risk of performance bias although 456
394 there was insufficient information in three of the trials [24–26] to 457
395 allow an assessment. The remaining three trials were at high risk 458
396 of performance bias due to the non-blinding of patients and 459
397 some study personnel [16,17,19,22,27]. 460
398 Three trials [16,19,22,26,27] were at low at risk of detection 461
399 bias because outcome-assessors were either blinded or non-blind- 462
400 ing was unlikely to have affected outcome measurement. Two tri- 463
401 als [24,25] had an unclear risk of bias whilst one had a high risk 464
402 of detection bias [17]. 465
403 Four trials had low risk [17,22,24,26], one was at high risk 466
404 [16,19,27] and the remaining trial had an unclear risk of attrition 467
405 bias [25]. 468
406 Only two trials had a low risk of reporting bias [16,19,22,27]. 469
407 Three trials had a high risk of reporting bias [17,24,25] and the 470
408 risk in the remaining trial was unclear [26]. 471
409 472
410 473
411 Effects of interventions 474
412 Meta-analysis was precluded by the heterogeneous reporting of 475
413 outcomes; poor definition of the study population (i.e. not sepa- 476
Figure 2. Risk of bias summary: review authors’ judgements about each risk of
414 bias item for each included study. rating SPT and DPT burns); missing data and poor reporting 477
415 (Table 1). A narrative synthesis was performed. 478
416 479
Three studies compared MEBO with 1% SSD cream
417 Measures of wound healing (primary outcome) 480
[16,17,19,25,27]. Others compared with sodium carboxymethylcel-
418 Both Ang [16,19,27], Allam [25] et al. observed faster healing in 481
lulose silver (Aquacel Ag) [24], povidene iodine plus bepanthenol
419 patients treated with MEBO compared to SSD. The mean number 482
cream [22] and Helix aspersa extract (Elicina cream) [26]. Only
420 of days to 75% wound healing recorded by Ang et al. for the 483
Hindy [24] included a negative control group (saline-soaked
421 MEBO and SSD groups were 17 and 20 respectively [16,19,27]. In 484
gauze dressing).
422 Allam et al’s study [25], patients with SPT burns healed faster 485
423 compared to when treated with MEBO compared to SSD 486
424 Characteristics of studies excluded after reading full-texts (10.5 ± 2.7 versus 4.5 ± 3.8 days) (p < 0.05). Similarly, patients with 487
425 DPT burns treated with MEBO healed faster (30.5 ± 5.1 vs. 488
Two studies were excluded after reading full-texts: an economic
426 36.6 ± 5.1 days) (p < 0.05). Hirsch et al. measured day 12 TEWL and 489
analysis that did not assess efficacy [29] and a nonrandomised
427 markers of inflammation such as CRP and leukocyte count as 490
study [18].
428 491
markers of wound healing [17] comparing MEBO with SSD. The
429 492
Risk of bias in included studies finding that patients treated with MEBO had greater loss of water
430 493
Figures 2 and 3 summarise and illustrate the authors’ risk of bias transepidermally (11 vs. 7) suggests that these wounds healed
431 494
assessment. The trials by Hirsch [17], Hindy [24], Allam [25] and slower than those treated with SSD. However, the levels of other
432 495
their respective colleagues had the highest or unknown risk of markers were not reported.
433 496
bias across all domains (Figure 2). The studies by Ang [16,19,27] Patients treated with Helix aspersa cream by Tsoutsos et al.
434 497
and Tsoutsos’ [26] teams had a lower risk of bias whilst Carayanni healed significantly faster compared to MEBO, as assessed on
435 498
et al. [22] was judged least likely to be biased. photographs by blinded assessors (11 ± 2 vs. 15.3 days) [26]. This
436 499
Three trials had a low risk of selection bias owing to detailed planimetric finding was corroborated by another that the eschar
437 500
random sequence generation [16,19,22,26,27]. Although the detachment was faster in patients treated with Helix aspersa
438 501
remaining three trials all randomised patients into study arms, cream [26].
439 502
their procedures were unclear [17,24,25]. Only Ang et al. Hindy found no difference in the healing times of MEBO and
440 503
441 [16,19,27] detailed allocation concealment making it unclear Aquacel Ag (10.35 ± 2.8 vs. 10.05 ± 2.3 days) [24]. Compared to 504
whether the remaining five trials [17,22,24–26] adequately povidone-iodine þ bepanthenol cream, patients treated with
 JOURNAL OF PLASTIC SURGERY AND HAND SURGERY 5

505 MEBO healed faster as shown by faster 50% TEWL reduction (8.7 [16,19,27]. Similarly, no statistically significant difference was seen 568
506 vs. 10.75 days) [22]. in the incidence of Staphylococcus and Pseudomonas infections 569
507 In summary, more trials comparing wound healing between in groups receiving either MEBO or povidone-iodi- 570
508 MEBO and SSD favoured MEBO. There may be improved healing ne þ bepanthenol creams (5.8 vs. 7.5%, respectively) [22]. 571
509 in MEBO-treated wounds vs. those treated with povidone-iodi- In summary, the incidences of adverse reactions and wound 572
510 ne þ bepanthenol cream. There was no difference between MEBO infections were low and no statistically significant differences 573
511 and Acquacel Ag, but Helix Aspersa had faster healing rates were noted between MEBO and any of the comparators. 574
512 than MEBO. 575
513 576
Discussion
514 577
Pain and itch
515 There has been a suggestion that there are “double standards” in 578
516 Two studies found no difference in the analgesic effects of MEBO assessing complementary and alternative therapies in medicine 579
517 and SSD. Hirsch et al. found no statistically significant difference [30]. It is therefore crucial that these treatments are subjected to 580
518 in mean VAS scores between the MEBO and SSD groups on both the same scientifically rigorous analysis as used for ‘traditional’ 581
519 days 1 and 12 (both mean 5 on day 1 then 3.8 and 3.5, respect- treatments. As such, this systematic review was performed with 582
520 ively on day 12) [17]. Similarly, Allam et al. found no statistically the aim of pooling data relating to the efficacy of MEBO for the 583
521 significant difference in pain scores between MEBO and SSD [25]. treatment of burn wounds. Such a synthesis for the first time 584
522 One study found a statistically significant difference in pain pro- allows both surgeons and patients to appraise collated and syn- 585
523 files between MEBO and SSD [16,19,27]. MEBO patients rated their thesised evidence of MEBO and several comparators and is there- 586
524 pain as less than that of the SSD group after one week (2.9 vs 3.5 fore crucial to cost-effectiveness calculations by these groups. 587
525 mean post-dressing verbal numerical pain rating (VNPR) score). Data from six RCTs, all level 3 evidence, mostly poorly reported, 588
526 However, the MEBO group had a higher mean VNPR on admission 589
were eligible for inclusion. The heterogeneity of study methods,
527 (5.09 vs. 4.72) (p-value unreported) which may partially explain the 590
comparators and outcome measures precluded meta-analysis.
528 pain score differences. Furthermore, by the third week, there were 591
Even the three studies comparing MEBO to 1% SSD were suffi-
529 no longer any differences in analgesic effect. 592
ciently heterogeneous to preclude meta-analysis of only
530 In Hindy’s study, the MEBO group rated their pain as signifi- 593
these studies.
531 cantly less than that of the Aquacel Ag and saline control groups 594
The results of this review should be interpreted with the fol-
532 during the first 48 h [24]. However, although the mean VAS score 595
533 lowing caveats in mind. There were varying anatomical locations 596
for the MEBO group was 3.1 ± 1.9, those of the Aquacel Ag and
534 and not all papers specified. The appropriateness of 1% SSD as a 597
saline control groups were not reported. MEBO also had a greater
535 comparator or standard treatment is debatable since SSD has 598
ichthyotic effect than Acquacel Ag as shown greater proportions
536 been shown to be consistently associated with poorer healing 599
itch-free patients in the MEBO group compared to the Aquacel
537 outcomes compared to treatments such as skin substitutes, silver- 600
Ag and saline control groups (65 vs. 25% and 10%, respect-
538 containing dressings and silicon-coated dressings [1]. Other fac- 601
ively) [24]
539 There was no difference in the analgesic properties of MEBO
tors precluding meta-analysis included variability in outcome 602
540 and povidone-iodine þ bepanthenol cream (mean morning VAS
measurement. Surrogates included time to complete wound heal- 603
541 scores: 3.0 vs. 4.2, respectively, mean evening VAS scores: 3.8 vs. ing [24–26], TEWL [17], time to 75% healing [16,19,27] and 50% 604
542 4.7 respectively) [22]. reduction in TEWL [26]. 605
543 Tsoutsos et al. found that pain scores were significantly 606
544 improved with Helix Aspersa compared to MEBO (4.50 ± 0.52 vs. 607
MEBO effect on wound healing
545 3.52 ± 0.80) [26]. In this trial, mean VAS scores were similar 608
546 between the MEBO and Helix aspersa groups before dressings
The results do not consistently favour MEBO or any of its compa- 609
547 were applied (6.50 ± 0.89 vs. 6.22 ± 1.25). rators. Of the three studies comparing the wound healing proper- 610
548 In summary, the majority of trials comparing wound healing ties of SSD with MEBO, two favoured MEBO [16,25] whilst one 611
549 between MEBO and SSD favoured MEBO. There may be improved favoured SSD [17]. One of the favourable studies was poorly 612
550 healing in MEBO-treated wounds vs. those treated with povidone- reported, exposing it to significant biases [25]. The small improve- 613
551 ment in the other favourable study was not statistically significant 614
iodine þ bepanthenol cream. There was no difference between
552 [16,19,27]. This small and statistically insignificant benefit should 615
MEBO and Acquacel Ag, but Helix Aspersa had faster healing rates
553 be interpreted in the context of moderate risk of bias due to 616
than MEBO.
554 issues with blinding and failure to analyse on an intention to treat 617
555 basis. 618
Incidence of adverse effects
556 Wound healing was also reportedly improved in MEBO com- 619
The incidence of adverse effects was very low in all studies and
557 pared to Acquacel Ag [24] and povidone-iodine þ bepanthenol 620
for all interventions. One MEBO vs. SSD RCT reported no adverse
558 cream groups [22]. The results in the Acquacel Ag trial were at 621
outcomes for either intervention but the other two RCTs did not
559 risk of bias due to poor reporting [24]. The highest quality study 622
report their incidence of adverse effects [16,19,25,27]. There was
560 used an indirect surrogate measure, that is, reduction in TEWL to 623
no difference in allergy rates between MEBO and either Helix
561 624
aspersa [26] or povidone-iodine þ bepanthenol creams [22]. The suggest faster healing in SPT but not DPT burns [22]. This trial by
562 625
respective rates of adverse effects of MEBO and Aquacel Ag were Carayanni et al. was at risk of intrinsic bias as it was funded by a
563 626
not reported in the sole trial comparing the two [24] manufacturer of MEBO [31].
564 627
Only two trials reported the incidence of wound infections Patients receiving Helix aspersa cream healed faster than those
565 628
[16,19,22,27]. There was no difference in the incidence of methicil- receiving MEBO [26]. This study by Tsoutsos et al. [26] was at
566 629
lin-resistant Staphylococcus aureus infections between the MEBO moderate risk of bias and also compared MEBO to a treatment
567 630
and SSD groups in one trial (37.4 vs. 38.5%, respectively) that is not standard of care. Due to these quality issues, these
 6 N. T. MABVUURE ET AL.

631 results should, therefore, be interpreted with caution. At present, [2] Hoogewerf CJ, Van Baar ME, Hop MJ, et al. Topical treat- 694
632 poor evidence shows no difference in wound healing properties. ment for facial burns. Cochrane Database Syst Rev. 2013. Q5 695
633 [3] Nirmal SA, Pal Subodh C, Subhash CM, et al. Analgesic and 696
634 anti-inflammatory activity of b-sitosterol isolated from 697
MEBO effect on secondary outcomes
635 Nyctanthes arbortristis leaves. Inflammopharmacology. 2012; 698
636 Similar caution should be exercised in interpreting the effects of 20(4):219–224. 699
637 the interventions on secondary outcomes. In summary, there was [4] Loizou S, Lekakis I, Chrousos GP, et al. Beta-sitosterol exhib- 700
638 no difference in pain profiles in two [17,25] of three trials compar- its anti-inflammatory activity in human aortic endothelial 701
639 ing MEBO with SSD and in one trial comparing MEBO with povi- cells. Mol Nutr Food Res. 2010;54(4):551–558. 702
640 done-iodine þ bepanthenol cream [22]. One 3-week trial found [5] Gupta MB, Nath R, Srivastava N, et al. Anti-inflammatory 703
641 that MEBO had better analgesis than SSD but only in the first and antipyretic activities of beta-sitosterol. Planta Med. 704
642 week [16,19,27]. Helix aspersa cream had a greater analgesic effect 1980;39(2):157–163. 705
643 than MEBO [26]. [6] Dyson M, Young S, Pendle CL, et al. Comparison of the 706
644 The claim that MEBO was more analgesic than Acquacel Ag effects of moist and dry conditions on dermal repair. J 707
645 could not be verified since pre-treatment VAS scores were not Invest Dermatol. 1988;91(5):434–439. 708
646 reported [24]. Furthermore, the Acquacel Ag trial was either inad- [7] Eriksson E, Perez N, Slama J, et al. Treatment of chronic, 709
647 equately randomised, poorly reported or both. The identity of nonhealing abdominal wound in a liquid environment. Ann 710
648 outcome assessors and blinding protocols were also not specified. Plast Surg. 1996;36(1):80–83. 711
649 Wound infection and adverse effects rates were similarly low [8] Breuing K, Eriksson E, Liu P, et al. Healing of partial thick- 712
650 in both MEBO and comparator groups in all studies. However, ness porcine skin wounds in a liquid environment. J Surg 713
651 none of the studies reported adverse events according to item 19 714
Res. 1992;52(1):50–58.
652 of the CONSORT scale which advises recording adverse events 715
[9] Alexander G, Rajacic N, Ebrahim MK, et al. MEBO and hot
653 with reference to standardised criteria. 716
bitumen burns. Burns. 2003;29(8):863–865.
654 717
[10] Atiyeh BS, Dham R, Yassin MF, et al. Treatment of toxic epi-
655 718
dermal necrolysis with moisture-retentive ointment: a case
656 Limitations 719
report and review of the literature. Dermatol Surg. 2003;
657 This systematic review was conducted following PRISMA guide- 720
29(2):185–188.
658 721
lines. However, some limitations remain. The methodological flaws [11] Jewo PI, Fadeyibi IO, Babalola OS, et al. A comparative
659 722
of the eligible studies and their effect on this review’s conclusions study of the wound healing properties of moist exposed
660 723
have been extensively described above. This abundance of pre- burn ointment (MEBO) and silver sulphadiazine. Ann Burns
661 724
clinical evidence of efficacy would suggest a likely positive clinical Fire Disasters. 2009;22(2):79–82.
662 725
efficacy. However, the evidence was not high quality enough to [12] Zhang HQ, Yip TP, Hui I, et al. Efficacy of moist exposed
663 726
664 definitively answer this question. Furthermore, there was a selec- burn ointment on burns. J Burn Care Rehabil. 2005;26(3): 727
665 tion bias towards trials indexed on MEDLINE and EMBASE, argu- 247–251. 728
666 ably the two preeminent general healthcare databases. Several [13] Ioannovich JD, Magliacani G, Costagliola M, et al. Moist 729
667 case reports written in unscientific formats and listed on the man- exposed therapy of partial-thickness burn wounds. A multi- 730
668 ufacturer’s website were excluded as they were largely anecdotal center study. Eur J Plast Surg. 2003;26(7):338–345. 731
669 [15]. Papers were limited to English-language publications. [14] Atiyeh BS, Dham R, Costagliola M, et al. Moist exposed 732
670 Attempts were made to contact all authors for clarification of therapy: an effective and valid alternative to occlusive 733
671 missing data required but only one author responded. dressings for postlaser resurfacing wound care. Dermatol 734
672 Surg. 2004;30(1):18–25. 735
673 Conclusions [15] MEBO GROUP n.d. http://en.mebo.com/DaShiJi/index/id/ 736
674 290.html. (accessed November 23, 2019). Q6 737
675 The evidence for MEBO in English-language literature was poor [16] Ang ES, Lee ST, Gan CS, et al. Evaluating the role of alter- 738
676 and inconsistent with respect to wound healing rate and analge- native therapy in burn wound management: randomized 739
677 sis compared to 1% SSD, Acquacel Ag, Helix aspersa cream and trial comparing moist exposed burn ointment with conven- 740
678 povidone-iodine þ bepanthenol cream. Blinded RCTs comparing tional methods in the management of patients with 741
679 MEBO to both placebo and other common topical treatments second-degree burns. MedGenMed. 2001;3(2):3. 742
680 such as paraffin wax, which may also provide a moist and [17] Hirsch T, Ashkar W, Schumacher O, et al. Moist exposed 743
681 exposed environment, may further improve the confidence in burn ointment (MEBO) in partial thickness burns - a 744
682 concluding their analysis. There is good evidence that MEBO is as randomized, comparative open mono-center study on the 745
683 safe as its comparators as shown by the low complication rate. efficacy of dermaheal (MEBO) ointment on thermal 2nd 746
684 degree burns compared to conventional therapy. Eur J 747
685 Med Res. 2008;13:505–510. 748
686 Disclosure statement 749
[18] Mabrouk A, Boughdadi NS, Helal HA, et al. Moist occlusive
687 No potential conflict of interest was reported by the author(s). R V 750
dressing (Aquacel Ag) versus moist open dressing
688 R V 751
(MEBO ) in the management of partial-thickness facial
689 752
burns: a comparative study in Ain Shams University. Burns.
690 References 753
2012;38(3):396–403.
691 754
[1] Wasiak J, Cleland H, Campbell F, et al. Dressings for superfi- [19] Ang ES, Lee ST, Gan CS, et al. The role of alternative ther-
692 755
cial and partial thickness burns. Cochrane Database Syst apy in the management of partial thickness burns of the
693 756
Rev. 2008;4:CD002106. face - experience with the use of moist exposed burn
 JOURNAL OF PLASTIC SURGERY AND HAND SURGERY 7

757 ointment (MEBO) compared with silver sulphadiazine. Ann ointment or silver sulphadiazine cream both combined 820
758 Acad Med Singap. 2000;29(1):7–10. with a polyethylene bag. Ann Burns Fire Disasters. 2007; 821
759 [20] Norman G, Christie J, Liu Z, et al. Antiseptics for burns. 20(3):144–148. 822
760 Cochrane Database Syst Rev. 2017;7:CD011821. [26] Tsoutsos D, Kakagia D, Tamparopoulos K. The efficacy of 823
761 [21] Li Y, Wang N, Zhou C. A comparison between “moist Helix aspersa M€ uller extract in the healing of partial thick- 824
762 ointment” and 0.25% iodophor, silver sulfadiazine paste ness burns: a novel treatment for open burn management 825
763 and 0.1% rivanol in the treatment of deep II degree burn protocols. J Dermatolog Treat. 2009;20(4):219–222. 826
764 wounds. Zhonghua Zheng Xing Shao Shang Wai Ke Za Zhi. [27] Ang E, Lee ST, Gan CSG, et al. Pain control in a random- 827
765 1994;10(5):342–345. ized, controlled, clinical trial comparing moist exposed 828
766 [22] Carayanni VJ, Tsati EG, Spyropoulou GCH, et al. Comparing burn ointment and conventional methods in patients with 829
767 oil based ointment versus standard practice for the treat- partial-thickness burns. J Burn Care Rehabil. 2003;24(5): 830
768 ment of moderate burns in Greece: a trial based cost 289–296. 831
769 effectiveness evaluation. BMC Complement Altern Med. [28] Dat AD, Poon F, Pham KB, et al. Aloe vera for treating acute 832
770 2011;11:122. and chronic wounds. Cochrane Database Syst Rev. 2012; Q7 833
771 [23] Higgins JPT, Altman DG, Gøtzsche PC, et al. The Cochrane [29] Atiyeh BS, Dham R, Kadry M, et al. Benefit-cost analysis of 834
772 Collaboration’s tool for assessing risk of bias in randomised moist exposed burn ointment. Burns. 2002;28(7):659–663. 835
773 trials. BMJ. 2011;343:d5928. [30] Bower H. Double standards exist in judging traditional and 836
774 [24] Hindy A. Comparative study between sodium carboxy- alternative medicine. BMJ. 1998;316(7146):1694. 837
775 methyl-cellulose silver, moist exposed burn ointment, and [31] Lundh A, Lexchin J, Mintzes B, et al. Industry sponsorship 838
776 saline-soaked dressing for treatment of facial burns. Ann and research outcome. Cochrane Database Syst Rev. 2017; 839
777 Burns Fire Disasters. 2009;22:131–137. 2:MR000033. 840
778 [25] Allam AM, Mostafa W, Zayed E, et al. Management of the 841
779 acute partial-thickness burned hand; moist exposed burn 842
780 843
781 844
782 845
783 846
784 847
785 848
786 849
787 850
788 851
789 852
790 853
791 854
792 855
793 856
794 857
795 858
796 859
797 860
798 861
799 862
800 863
801 864
802 865
803 866
804 867
805 868
806 869
807 870
808 871
809 872
810 873
811 874
812 875
813 876
814 877
815 878
816 879
817 880
818 881
819 882

View publication stats