- Email: [email protected]
Silver sulphadiazine and the healing of partial thickness burns: a prospective clinical trial
Silver sulphadiazine and the healing of partial thickness burns: a prospective clinical trial H. S. STERN Burns Unit, Repatriation
General Hospital, Concord, Sydney, Australia
Summary-A clinical trial was devised to determine whether the healing of partial thickness burns was retarded by the use of silver sulphadiazine cream (SSD) compared with simple, non-bacteriocidal dressings. Biopsy-confirmed partial thickness burns of at least 2% confluent area were dressed according to a strict protocol to compare the rate of epithelialisation of the control dressing, tulle gras, with that obtained with the use of silver sulphadiazine. The same comparative dressing regimen was carried out on a series of split thickness graft donor sites. Twenty such donor sites and fifteen burn areas were admitted to the trial. The mean time to healing of SSD-treated burn areas was longer than that for tulle gras; the difference when analysed by paired t-test was statistically significant (~~0.05). There was very little difference in the compared healing rates of the donor sites, which implies that SSD does not retard epithelialisation of dermal depth injuries but rather that the delayed healing is a unique response of the partial thickness burn wound to the presence of SSD. A review of the relevant literature is included in the discussion.
thickness burns with individual confluent areas >29& It did not exclude patients with full thickness burns in other areas, or with complicating factors such as inhalational injury or myoglobinuria. Having been admitted to the trial, the patient gave signed consent after the protocol was explained. Each separate burn area was considered in two subsections, to be dressed with tulle or SSD respectively, and each subsection was biopsied with a 4 mm diameter skin punch (Stiefel Co.) prior to the commencement of dressings. (ii) Dressings. The materials used were Silvazine”’ (Smith & Nephew: silver sulphadiazine 1% w/ w, chlorhexidine digluconate 0.26 w/w) and Jelonet’” (Smith & Nephew No. 7477) with no antibiotic impregnation. The section designated for SSD was cleansed daily and dressed with gauze, heavily coated with the cream, plus padding. The tulle section was dressed with the paraffin gauze, cotton gauze, padding and bandage. The tulle was left intact for 5 days and only the outer dressing changed as required f-or the exudate. The tulle was changed if it fell off. By convention within the trial, the distal subsection of a limb or trunk burn was dressed
The clinical impression had been gained that the healing of partial thickness burns was retarded by the use of silver sulphadiazine (SSD) cream, if compared with simple non-bacteriocidal dressing materials. A prospective trial was devised to confirm or deny that impression using simple openweave paraffin gauze, tulle gras, as the control dressing on a series of patients with partial thickness burns. A second group of patients was investigated by carrying out the same dressing regimen on a number of split thickness skin graft donor sites. The intention was to identify whether any difference in healing rates noted in the burns for the two dressing materials was reproducible in a defect of similar histological depth but different aetiology, which would suggest a specific effect of SSD on epithelialisation. Materials and methods Group I-Burn patients
(i)
selection. This group comprised those patients at the Burns Unit, Repatriation General Hospital. Concord, who displayed, at the time of initial assessment, recent (<24hour-old), untreated, macroscopically partial Patient
581
582
BRITISH JOURNAL
with SSD and the proximal part with tulle, in an attempt to minimise gravity-induced contamination of the tulle area with cream. Trunk burns that crossed the midline cyere divided in the sagittal plane. The regimen continued until healed, or to 21 days, at which time management decisions regarding residual areas were made. (iii) Histology. All punch biopsies were placed in formalin, accurately labelled and sent for histopathological review. The slides were stained with haematoxylin and eosin. The pathologist examined the slides at magnifications of x 2.5, x 10 and x 40 and commented on remaining viable basal epidermal cells, the presence of viable or necrotic papillary dermis and surviving healthy epithelial adnexae. Thus subsections of a particular area could be confirmed as comparable depth injuries for treatment by the two dressing regimens. If biopsy showed full thickness burn, the area was removed from the trial and treated appropriately. (iv) Healing was said to have occurred when a subsection appeared clinically fully epithelialised. The number of days since initial injury was recorded. Group II-Donor
site patients
(i) Patient selection. This Table 1
group
Group I-Burn
comprised
a
OF PLASTIC SURGERY
random selection of patients with apparently uniform thickness split skin graft donor sites, mostly following the use of an electric dermatome. These patients required skin graft for either burns, general trauma, skin malignancy or other reconstructive problems. The donor sites were not biopsied. (ii) Dressings. The regimen was essentially as that for Group I except that the tulle gras was left intact for 8 days. (iii) Healing. As for Group I. The results paired t-test.
were analysed
statistically
a
Results Group I Fifteen burn areas of at least 2% in 8 patients were admitted to the trial. Details of nature of burn, site and healing rate are presented in Table 1. The subsections of three burns (20%) failed to heal spontaneously and proceeded to excision and graft. All three burns were hot water scalds in two elderly female patients (KP and ME), which were deep dermal burns with viable adnexal elements on original biopsy and on repeat biopsy prior to grafting. These burns were omitted from statistical analysis. The mean healing rate of the SSD-treated burns was 19.25 days. The mean healing rate of the
patients Subsection healing iday)
Patient
Age/xx
Type of’ burn
Site
Six
VG
67
M
Flame
DR
24
M
Flame/flash
GM KP
47 70
M F
Hot oil Hot water scald
ME PC DA
87 24 22
F M M
Hot water scald Flame/flash Flame
DT
24
M
Flame/flash
Abdomen L Thigh Abdomen Chest L Leg R Arm L Arm * Abdomen * Thigh * Buttocks L Arm L Arm L Leg Chest L Arm
5 4 4 5 8 2 2 2 3 5 3 2 3 4 2
Mean healing time
t= 2.509 with 11 degrees of freedom p < 0.05 i.e. significant difference. * Burn area excised and grafted; results omitted from analysis. Paired t-test statistic:
using
(“J
SSD
Tulle
15 26 5 4 17 25 15 19 19 21 35 19 30 24 16
8 9 5 5 10 11 13 19 19 21 23 17 35 24 13
19.25
14.41
SILVER SULPHADIAZINE
AND HEALING
OF PARTIAL
THICKNESS
tulle-treated burns was 14.41 days. Statistical analysis showed that the difference in healing rates was significant (t = 2.509. p < 0.05). Group II Twenty donor sites in 18 patients were admitted to the trial (Table 2). The mean healing rate of the SSD-treated donor sites was 10.85 days. The mean healing rate of the tulle-treated donor sites was 10.5 days, Statistical analysis showed that the difference in healing rates was not significant (t =0.941, p>O.O5). Discussion The aims of the study were : or deny the clinical impression that SSD retards the healing of partial thickness burns: (ii) to identify any effect of SSD on the healing rate of split skin graft donor sites.
(9
to confirm
Table 2
Group II-donor
site patients Subsection healing i days i Donor site
SSD
Tulle IO
RT
49
M
Burn
LThigh RThigh
9 11
AF
28
F
Burn
LThigh
9
9
AH
67
M
Neoplasm
LArm
10
IO
RJ
63
M
Neoplasm
LThigh
10
IO
DH
15
M
Burn
RThigh LThigh
8 8
8 8
AD
73
M
Neoplasm
L Thigh
18
13
M
Neoplasm
L Thigh
12
I’
M
Neoplasm
L Thigh
14
II
64
F
Burn
RThigh
11
I1
RM
65
M
Neoplasm
R Thigh
14
II
DS
66
M
Neoplasm
RThigh
13
14
FD
67
M
Neoplasm
RThigh
10
IO
PW
28
M
Burn
LThigh
10
11 10
JR WC zo
78 68
II
MB
15
M
Burn
LThigh
9
sz
37
M
Burn
LThigh
9
9
HL
38
M
Burn
RThigh
7
9
SS
80
F
Burn
Abdomen
14
13
DJ
74
M
Neoplasm
RThigh
I1
I1
10.85
10.5
Mean healing time Paired r-test statistic:
r=0.941 with 19 degrees of freedom p>>O.O5 i.e. no significant difference
BURNS: A PROSPECTIVE
CLINICAL
TRIAL
583
The rationale was that if burns did heal more slowly with SSD, the donor site behaviour would indicate if this was a uniform effect of SSD on epithelialisation of wounds of this depth, or a response unique to the burn wound. It should be noted that under normal circumstances many of the deep dermal burns admitted to the trial would have been treated by early excision and grafting (Janzekovic, 1970) by the Burns Unit at Repatriation General Hospital, Concord. Clinical experience would dictate that certain burns, such as the three hot water scalds in elderly females that ultimately proceeded to grafting, would not heal well spontaneously despite the initial appearance and biopsy confirmation of their partial thickness. Since SSD became widely available as a bacteriocidal dressing cream for burns (Fox, 1968), it has come to be regarded by some as the standard topical agent in burns therapy (Hermans, 1984). This popularity is largely due to its confirmed ability to reduce bacterial colonisation of burns and thus reduce morbidity and mortality from burn wound sepsis. Fox (1968) reported its success in treating Pseudomonas infected burns in mice and man and others have since verified its bacteriocidal effectiveness (Ballin, 1974; Lowbury et al., 1976; Pegg et al., 1979; Hermans and Hermans. 1984; Waymack et al., 1986). Little interest seems to have been focused on the rate of epithelialisation in partial thickness burns dressed with SSD. In an uncontrolled comparison of healing and infection rates in burns of all depths before and after the introduction to general use of the topical antimicrobial agents SSD and sulphamylon, Pegg et al. (1979) found no difference in the average number of days in hospital. Hermans (1984) stated simply that SSD “. . . does not hamper epithelial outgrowth” with no specific evidence presented to support this assertion. No significant difference in healing rates of partial thickness burns was found by Waymack et al. (1986) when they compared the use of SSD and aquaphor gauze. Kaufman et al. (1985) questioned whether SSD interfered with mitosis or migration of epithelial cells after finding slower wound healing in experimental burns in guinea-pigs dressed with SSD, or combinations including SSD. Tulle gras has been used previously as a comparison dressing to Inadine (Wilson et al., 1986), polyvinyl alcohol-backed tulle (Hart and Lawrence, 1984) and SSD-impregnated tulle (Lawrence, 1977). Lawrence (1977) found no difference in healing rates of burns treated with simple tulle gras and
584
those treated with tulle gras medicated with silver sulphadiazine, both in the clinical and the experimental setting. No comparison was performed to SSD cream dressings. Previous reports of the healing of burns have depended on the clinical appearance of the burn initially to diagnose its depth (Jackson, 1953; Hansbrough et al.. 1984; Hart and Lawrence, 1984; Keswani and Patil, 1985; Wilson et al., 1986), Jackson (1953) having described in detail the criteria that most clinicians would use in this evaluation. Other means of estimation of depth have included knowledge of the burning agent, tissue staining with fluorescein or patent blue, thermography, radioisotope scans (Lynch ef al., 1964; Lynch, 1979) and ultimately “the retrospectoscope”-seeing how it heals. By biopsying the burn areas in this trial, the pathologist could confirm or deny the clinician’s initial impression, and justify the inclusion of the patient in the trial. Reports were based on Lever’s classification in Histopathology of the Skin (Lever and Lever, 1983) which details the depth of burn loss, viable dermis and adnexae when present. The limitations that arise are two:
(9 One cannot have guaranteed
uniformity of depth of injury across a 2% area, nor is it justifiable to take multiple punch biopsies of each subsection. Thus one biopsy only was taken from each subsection, ensuring reasonable similarity of depth of injury in compared sections, and the distal part of the burn always dressed with SSD to ensure than any slight variation in depth was randomised and this potential source of bias eliminated. (ii) Confirmation of depth does not specify a healing rate no matter what dressing material is used. Other factors such as nutrition, vascularity, contamination or infection, desiccation, systemic illness or system failure will influence the rate of healing. An histologically deeper burn in a fit young patient may heal more rapidly than a more superficial burn in a debilitated older patient. Within this trial an attempt was made to mitigate the effects of many of these biological variables by having each burn wound act as its own control. Analysing the results, several interesting issues emerge. The mean healing rate of the partial thickness burns dressed with SSD cream was slower than that of the tulle-dressed subsections, and the difference was statistically significant. The impli-
BRITISH JOURNAL
OF PLASTIC SURGERY
cations of this delayed healing are far-reaching if each extra day to achieve epithelialisation is thought of as another day off work for the outpatient or another day of hospital expenses for the inpatient. The almost identical healing rates of the tulleand SSD-dressed subsections of the donor wounds shows that SSD does not inhibit epithelialisation of other dermal depth injuries. Thus the delay in Group I would appear to be a unique response by the partial thickness burn wound to the presence of SSD. A theoretical explanation for this is discussed later. In considering the variability of absolute healing times for those subsections of Group I dressed with SSD, or those dressed with tulle, the very broad spectrum of injury covered by the term “partial thickness burn” is highlighted. It should be noted that such a grouping includes the very superficial flash burn of patient DR which healed in 5 days, and the much more substantial deep dermal burn of patient PC which only healed after 35 days. Three burn areas displayed a great difference in the healing rates with SSD or tulle (patients VG, KP, PC). These remain essentially unexplained. All burnt limbs were elevated, so dependent oedema was not a factor. Histology of the biopsied subsections was similar and nothing in the history of the injury (e.g. unprotected area, last item of flaming clothing removed) shed any light on the disparity. It may be that SSD retards the healing of partial thickness burns significantly by delaying the separation of the burn eschar/coagulum. SSD reduces bacterial colonisation (Ballin, 1974; Pegg et al., 1979) and thus should reduce the tissue levels of bacterial enzymes which enhance eschar separation. Microbiological swabs would verify the previously documented bacteriocidal effects of SSD. However, spectrophotometry is the biochemical approach to assaying tissue collagenase and elastase levels. To eliminate potential bias in this technique, the tissue biopsies must contain equivalent weights of dermis and fat from the compared subsections. Further, a method must be found to eliminate the yellowish tissue discolouration and coating caused by SSD as this will interfere with the spectrophotometric enzyme analysis. This trial will continue, to verify or disprove this theory. Acknowledgements The author wishes to thank Mr P. Haertsch. Consultant Plastic Surgeon at the RGHC, under whose care all of the reported
SILVER SULPHADIAZINE
AND HEALING
OF PARTIAL
THICKNESS
patients were admitted, the nursing staff of the Burns Unit and Ward 630 of the RGHC for their attentiveness to the prescribed dressing regimens, and Mr P. Kennedy for his assistance within the trial.
References Ballin, J. C. ( 1974). Evaluation of a new topical agent for burn therapy~~~silversulphadiazine. Joumalqfthr Anlerican Medical Associa~bn, 230, I 184. Fox, C. L. (1968). Silver sulphadiazine: a new topical therapy for Pseudomonas in burns. Archires ofSurgery, %, 184. Hansbrougb, J. F., Carroll, W. B., Zapata-Sivent, R., Dominic, W. J., Wang, X-W. and Wakimoto, A. (1984). Clinical experience with biobrane biosynthetic dressing in the treatment of partial thickness burns. Burns, 10.415. Hart, N. B. and Lawrence, J. C. (1984). Tulle gras dressings. Burns, Il. 26. Hermans, N. T. H. E. and Hermans, R. P. (1984). Preliminary report on the use of a new hydrocolloid dressing in the treatment of burns. Burns, 11. IX. Hermans, R. P. (1984). Topical treatment of serious infections with special reference to the use of a mixture of silver sulphadiazine and cerium nitrate: 2 clinical studies. Burns. II. 59. Jackson, D. McG. (1953). The diagnosis of the depth of burning. British JournalqfSurgery. 40. 588. Janzekovic, A. (1970). A new concept in the early excision and immediate grafting of burns. Journalqt’Traum~. 10, 1103. Kaufman. T.. Nathan. P.. Levin, M.. Hebda. P. A.. Eichenlaub. E. H. and Korol. B. (1985). drug-loaded synthetic dressings: effect on contraction. epithelialization. and collagen synthesis of deep second-degree experimental burns. Annuls of’Plasric Surger?. 14.410. Keswani, M. H. and Patil, A. R. (1985).Boiled potato peel as a burn wound dressing. Burns. 11. 220. Lawrence, J. C. ( 1977). An experimental and clinical evaluation
BURNS: A PROSPECTIVE
CLINICAL
TRIAL
585
of a tulle gras dressing medicated with silver sulphadiazine. Burns. 3, 186. Lever, W. F. and Lever, G. S. (1983). Thermal burns. In Hisroporhology qfthe Skin. Philadelphia: J. B. Lippincott. Lowbury, E. J. L., Babb, J. R., Bridges, K. and Jackson, D. M. (1976). Topical chemoprophylaxis with silver sulphadiazine and silver nitrate chlorhexidine creams: emergence of sulphonamide-resistant gram-negative bacilli. British Medical Journal, 1,493. Lynch, J. B. (1979). Thermal burns. In Grabb, W. C. and Smith, J. W. (Eds) Plastic Surgery. 3rd Edition. Boston: Little. Brown and Co. Lynch, J. B., Bray, J. P., Lewis, S. R. and Blocker, T. G. Jr. (1964). Studies with radioisotope labelled albumin in experimental burn wounds. JournalqfSurgical Research, 4.226. Pegg, S. P., Ramsay, K., Meldrum, L. and Laundy, M. ( 1979). Clinical comparison of maphenide and silver sulphadiazine. Scandinavian Journal qf’Plastic and Reconstructire Surgery, 13. 95. Waymack, J. P., Nathan, P., Robb, E. C., Plessinger, R., Rapien. J., Krummel, R., Jenkins, M. and MacMillan, B. G. (I 986). An evaluation of aquaphor gauze dressing in burned children. Burns. 12. 443. Wilson, G. R., Fowler, C.. Ledger, J. and Thorley, M. ( 1986). Evaluation of a new antiseptic dressing in minor burns. Burn.,. 12,518
The Author Harvey S. Stern, BSc(Med), FRCSEd, Registrar. Department of Plastic Surgery and Burns Unit, Repatriation General Hospital, Concord. Sydney. Requests for reprints to: Mr Harvey Bondi Junction. NSW 1022. Australia.
Stern,
Paper received 5 October 1988. Accepted IO February 1989 after revision
33 Bourke
Street.
General Hospital, Concord, Sydney, Australia
Summary-A clinical trial was devised to determine whether the healing of partial thickness burns was retarded by the use of silver sulphadiazine cream (SSD) compared with simple, non-bacteriocidal dressings. Biopsy-confirmed partial thickness burns of at least 2% confluent area were dressed according to a strict protocol to compare the rate of epithelialisation of the control dressing, tulle gras, with that obtained with the use of silver sulphadiazine. The same comparative dressing regimen was carried out on a series of split thickness graft donor sites. Twenty such donor sites and fifteen burn areas were admitted to the trial. The mean time to healing of SSD-treated burn areas was longer than that for tulle gras; the difference when analysed by paired t-test was statistically significant (~~0.05). There was very little difference in the compared healing rates of the donor sites, which implies that SSD does not retard epithelialisation of dermal depth injuries but rather that the delayed healing is a unique response of the partial thickness burn wound to the presence of SSD. A review of the relevant literature is included in the discussion.
thickness burns with individual confluent areas >29& It did not exclude patients with full thickness burns in other areas, or with complicating factors such as inhalational injury or myoglobinuria. Having been admitted to the trial, the patient gave signed consent after the protocol was explained. Each separate burn area was considered in two subsections, to be dressed with tulle or SSD respectively, and each subsection was biopsied with a 4 mm diameter skin punch (Stiefel Co.) prior to the commencement of dressings. (ii) Dressings. The materials used were Silvazine”’ (Smith & Nephew: silver sulphadiazine 1% w/ w, chlorhexidine digluconate 0.26 w/w) and Jelonet’” (Smith & Nephew No. 7477) with no antibiotic impregnation. The section designated for SSD was cleansed daily and dressed with gauze, heavily coated with the cream, plus padding. The tulle section was dressed with the paraffin gauze, cotton gauze, padding and bandage. The tulle was left intact for 5 days and only the outer dressing changed as required f-or the exudate. The tulle was changed if it fell off. By convention within the trial, the distal subsection of a limb or trunk burn was dressed
The clinical impression had been gained that the healing of partial thickness burns was retarded by the use of silver sulphadiazine (SSD) cream, if compared with simple non-bacteriocidal dressing materials. A prospective trial was devised to confirm or deny that impression using simple openweave paraffin gauze, tulle gras, as the control dressing on a series of patients with partial thickness burns. A second group of patients was investigated by carrying out the same dressing regimen on a number of split thickness skin graft donor sites. The intention was to identify whether any difference in healing rates noted in the burns for the two dressing materials was reproducible in a defect of similar histological depth but different aetiology, which would suggest a specific effect of SSD on epithelialisation. Materials and methods Group I-Burn patients
(i)
selection. This group comprised those patients at the Burns Unit, Repatriation General Hospital. Concord, who displayed, at the time of initial assessment, recent (<24hour-old), untreated, macroscopically partial Patient
581
582
BRITISH JOURNAL
with SSD and the proximal part with tulle, in an attempt to minimise gravity-induced contamination of the tulle area with cream. Trunk burns that crossed the midline cyere divided in the sagittal plane. The regimen continued until healed, or to 21 days, at which time management decisions regarding residual areas were made. (iii) Histology. All punch biopsies were placed in formalin, accurately labelled and sent for histopathological review. The slides were stained with haematoxylin and eosin. The pathologist examined the slides at magnifications of x 2.5, x 10 and x 40 and commented on remaining viable basal epidermal cells, the presence of viable or necrotic papillary dermis and surviving healthy epithelial adnexae. Thus subsections of a particular area could be confirmed as comparable depth injuries for treatment by the two dressing regimens. If biopsy showed full thickness burn, the area was removed from the trial and treated appropriately. (iv) Healing was said to have occurred when a subsection appeared clinically fully epithelialised. The number of days since initial injury was recorded. Group II-Donor
site patients
(i) Patient selection. This Table 1
group
Group I-Burn
comprised
a
OF PLASTIC SURGERY
random selection of patients with apparently uniform thickness split skin graft donor sites, mostly following the use of an electric dermatome. These patients required skin graft for either burns, general trauma, skin malignancy or other reconstructive problems. The donor sites were not biopsied. (ii) Dressings. The regimen was essentially as that for Group I except that the tulle gras was left intact for 8 days. (iii) Healing. As for Group I. The results paired t-test.
were analysed
statistically
a
Results Group I Fifteen burn areas of at least 2% in 8 patients were admitted to the trial. Details of nature of burn, site and healing rate are presented in Table 1. The subsections of three burns (20%) failed to heal spontaneously and proceeded to excision and graft. All three burns were hot water scalds in two elderly female patients (KP and ME), which were deep dermal burns with viable adnexal elements on original biopsy and on repeat biopsy prior to grafting. These burns were omitted from statistical analysis. The mean healing rate of the SSD-treated burns was 19.25 days. The mean healing rate of the
patients Subsection healing iday)
Patient
Age/xx
Type of’ burn
Site
Six
VG
67
M
Flame
DR
24
M
Flame/flash
GM KP
47 70
M F
Hot oil Hot water scald
ME PC DA
87 24 22
F M M
Hot water scald Flame/flash Flame
DT
24
M
Flame/flash
Abdomen L Thigh Abdomen Chest L Leg R Arm L Arm * Abdomen * Thigh * Buttocks L Arm L Arm L Leg Chest L Arm
5 4 4 5 8 2 2 2 3 5 3 2 3 4 2
Mean healing time
t= 2.509 with 11 degrees of freedom p < 0.05 i.e. significant difference. * Burn area excised and grafted; results omitted from analysis. Paired t-test statistic:
using
(“J
SSD
Tulle
15 26 5 4 17 25 15 19 19 21 35 19 30 24 16
8 9 5 5 10 11 13 19 19 21 23 17 35 24 13
19.25
14.41
SILVER SULPHADIAZINE
AND HEALING
OF PARTIAL
THICKNESS
tulle-treated burns was 14.41 days. Statistical analysis showed that the difference in healing rates was significant (t = 2.509. p < 0.05). Group II Twenty donor sites in 18 patients were admitted to the trial (Table 2). The mean healing rate of the SSD-treated donor sites was 10.85 days. The mean healing rate of the tulle-treated donor sites was 10.5 days, Statistical analysis showed that the difference in healing rates was not significant (t =0.941, p>O.O5). Discussion The aims of the study were : or deny the clinical impression that SSD retards the healing of partial thickness burns: (ii) to identify any effect of SSD on the healing rate of split skin graft donor sites.
(9
to confirm
Table 2
Group II-donor
site patients Subsection healing i days i Donor site
SSD
Tulle IO
RT
49
M
Burn
LThigh RThigh
9 11
AF
28
F
Burn
LThigh
9
9
AH
67
M
Neoplasm
LArm
10
IO
RJ
63
M
Neoplasm
LThigh
10
IO
DH
15
M
Burn
RThigh LThigh
8 8
8 8
AD
73
M
Neoplasm
L Thigh
18
13
M
Neoplasm
L Thigh
12
I’
M
Neoplasm
L Thigh
14
II
64
F
Burn
RThigh
11
I1
RM
65
M
Neoplasm
R Thigh
14
II
DS
66
M
Neoplasm
RThigh
13
14
FD
67
M
Neoplasm
RThigh
10
IO
PW
28
M
Burn
LThigh
10
11 10
JR WC zo
78 68
II
MB
15
M
Burn
LThigh
9
sz
37
M
Burn
LThigh
9
9
HL
38
M
Burn
RThigh
7
9
SS
80
F
Burn
Abdomen
14
13
DJ
74
M
Neoplasm
RThigh
I1
I1
10.85
10.5
Mean healing time Paired r-test statistic:
r=0.941 with 19 degrees of freedom p>>O.O5 i.e. no significant difference
BURNS: A PROSPECTIVE
CLINICAL
TRIAL
583
The rationale was that if burns did heal more slowly with SSD, the donor site behaviour would indicate if this was a uniform effect of SSD on epithelialisation of wounds of this depth, or a response unique to the burn wound. It should be noted that under normal circumstances many of the deep dermal burns admitted to the trial would have been treated by early excision and grafting (Janzekovic, 1970) by the Burns Unit at Repatriation General Hospital, Concord. Clinical experience would dictate that certain burns, such as the three hot water scalds in elderly females that ultimately proceeded to grafting, would not heal well spontaneously despite the initial appearance and biopsy confirmation of their partial thickness. Since SSD became widely available as a bacteriocidal dressing cream for burns (Fox, 1968), it has come to be regarded by some as the standard topical agent in burns therapy (Hermans, 1984). This popularity is largely due to its confirmed ability to reduce bacterial colonisation of burns and thus reduce morbidity and mortality from burn wound sepsis. Fox (1968) reported its success in treating Pseudomonas infected burns in mice and man and others have since verified its bacteriocidal effectiveness (Ballin, 1974; Lowbury et al., 1976; Pegg et al., 1979; Hermans and Hermans. 1984; Waymack et al., 1986). Little interest seems to have been focused on the rate of epithelialisation in partial thickness burns dressed with SSD. In an uncontrolled comparison of healing and infection rates in burns of all depths before and after the introduction to general use of the topical antimicrobial agents SSD and sulphamylon, Pegg et al. (1979) found no difference in the average number of days in hospital. Hermans (1984) stated simply that SSD “. . . does not hamper epithelial outgrowth” with no specific evidence presented to support this assertion. No significant difference in healing rates of partial thickness burns was found by Waymack et al. (1986) when they compared the use of SSD and aquaphor gauze. Kaufman et al. (1985) questioned whether SSD interfered with mitosis or migration of epithelial cells after finding slower wound healing in experimental burns in guinea-pigs dressed with SSD, or combinations including SSD. Tulle gras has been used previously as a comparison dressing to Inadine (Wilson et al., 1986), polyvinyl alcohol-backed tulle (Hart and Lawrence, 1984) and SSD-impregnated tulle (Lawrence, 1977). Lawrence (1977) found no difference in healing rates of burns treated with simple tulle gras and
584
those treated with tulle gras medicated with silver sulphadiazine, both in the clinical and the experimental setting. No comparison was performed to SSD cream dressings. Previous reports of the healing of burns have depended on the clinical appearance of the burn initially to diagnose its depth (Jackson, 1953; Hansbrough et al.. 1984; Hart and Lawrence, 1984; Keswani and Patil, 1985; Wilson et al., 1986), Jackson (1953) having described in detail the criteria that most clinicians would use in this evaluation. Other means of estimation of depth have included knowledge of the burning agent, tissue staining with fluorescein or patent blue, thermography, radioisotope scans (Lynch ef al., 1964; Lynch, 1979) and ultimately “the retrospectoscope”-seeing how it heals. By biopsying the burn areas in this trial, the pathologist could confirm or deny the clinician’s initial impression, and justify the inclusion of the patient in the trial. Reports were based on Lever’s classification in Histopathology of the Skin (Lever and Lever, 1983) which details the depth of burn loss, viable dermis and adnexae when present. The limitations that arise are two:
(9 One cannot have guaranteed
uniformity of depth of injury across a 2% area, nor is it justifiable to take multiple punch biopsies of each subsection. Thus one biopsy only was taken from each subsection, ensuring reasonable similarity of depth of injury in compared sections, and the distal part of the burn always dressed with SSD to ensure than any slight variation in depth was randomised and this potential source of bias eliminated. (ii) Confirmation of depth does not specify a healing rate no matter what dressing material is used. Other factors such as nutrition, vascularity, contamination or infection, desiccation, systemic illness or system failure will influence the rate of healing. An histologically deeper burn in a fit young patient may heal more rapidly than a more superficial burn in a debilitated older patient. Within this trial an attempt was made to mitigate the effects of many of these biological variables by having each burn wound act as its own control. Analysing the results, several interesting issues emerge. The mean healing rate of the partial thickness burns dressed with SSD cream was slower than that of the tulle-dressed subsections, and the difference was statistically significant. The impli-
BRITISH JOURNAL
OF PLASTIC SURGERY
cations of this delayed healing are far-reaching if each extra day to achieve epithelialisation is thought of as another day off work for the outpatient or another day of hospital expenses for the inpatient. The almost identical healing rates of the tulleand SSD-dressed subsections of the donor wounds shows that SSD does not inhibit epithelialisation of other dermal depth injuries. Thus the delay in Group I would appear to be a unique response by the partial thickness burn wound to the presence of SSD. A theoretical explanation for this is discussed later. In considering the variability of absolute healing times for those subsections of Group I dressed with SSD, or those dressed with tulle, the very broad spectrum of injury covered by the term “partial thickness burn” is highlighted. It should be noted that such a grouping includes the very superficial flash burn of patient DR which healed in 5 days, and the much more substantial deep dermal burn of patient PC which only healed after 35 days. Three burn areas displayed a great difference in the healing rates with SSD or tulle (patients VG, KP, PC). These remain essentially unexplained. All burnt limbs were elevated, so dependent oedema was not a factor. Histology of the biopsied subsections was similar and nothing in the history of the injury (e.g. unprotected area, last item of flaming clothing removed) shed any light on the disparity. It may be that SSD retards the healing of partial thickness burns significantly by delaying the separation of the burn eschar/coagulum. SSD reduces bacterial colonisation (Ballin, 1974; Pegg et al., 1979) and thus should reduce the tissue levels of bacterial enzymes which enhance eschar separation. Microbiological swabs would verify the previously documented bacteriocidal effects of SSD. However, spectrophotometry is the biochemical approach to assaying tissue collagenase and elastase levels. To eliminate potential bias in this technique, the tissue biopsies must contain equivalent weights of dermis and fat from the compared subsections. Further, a method must be found to eliminate the yellowish tissue discolouration and coating caused by SSD as this will interfere with the spectrophotometric enzyme analysis. This trial will continue, to verify or disprove this theory. Acknowledgements The author wishes to thank Mr P. Haertsch. Consultant Plastic Surgeon at the RGHC, under whose care all of the reported
SILVER SULPHADIAZINE
AND HEALING
OF PARTIAL
THICKNESS
patients were admitted, the nursing staff of the Burns Unit and Ward 630 of the RGHC for their attentiveness to the prescribed dressing regimens, and Mr P. Kennedy for his assistance within the trial.
References Ballin, J. C. ( 1974). Evaluation of a new topical agent for burn therapy~~~silversulphadiazine. Joumalqfthr Anlerican Medical Associa~bn, 230, I 184. Fox, C. L. (1968). Silver sulphadiazine: a new topical therapy for Pseudomonas in burns. Archires ofSurgery, %, 184. Hansbrougb, J. F., Carroll, W. B., Zapata-Sivent, R., Dominic, W. J., Wang, X-W. and Wakimoto, A. (1984). Clinical experience with biobrane biosynthetic dressing in the treatment of partial thickness burns. Burns, 10.415. Hart, N. B. and Lawrence, J. C. (1984). Tulle gras dressings. Burns, Il. 26. Hermans, N. T. H. E. and Hermans, R. P. (1984). Preliminary report on the use of a new hydrocolloid dressing in the treatment of burns. Burns, 11. IX. Hermans, R. P. (1984). Topical treatment of serious infections with special reference to the use of a mixture of silver sulphadiazine and cerium nitrate: 2 clinical studies. Burns. II. 59. Jackson, D. McG. (1953). The diagnosis of the depth of burning. British JournalqfSurgery. 40. 588. Janzekovic, A. (1970). A new concept in the early excision and immediate grafting of burns. Journalqt’Traum~. 10, 1103. Kaufman. T.. Nathan. P.. Levin, M.. Hebda. P. A.. Eichenlaub. E. H. and Korol. B. (1985). drug-loaded synthetic dressings: effect on contraction. epithelialization. and collagen synthesis of deep second-degree experimental burns. Annuls of’Plasric Surger?. 14.410. Keswani, M. H. and Patil, A. R. (1985).Boiled potato peel as a burn wound dressing. Burns. 11. 220. Lawrence, J. C. ( 1977). An experimental and clinical evaluation
BURNS: A PROSPECTIVE
CLINICAL
TRIAL
585
of a tulle gras dressing medicated with silver sulphadiazine. Burns. 3, 186. Lever, W. F. and Lever, G. S. (1983). Thermal burns. In Hisroporhology qfthe Skin. Philadelphia: J. B. Lippincott. Lowbury, E. J. L., Babb, J. R., Bridges, K. and Jackson, D. M. (1976). Topical chemoprophylaxis with silver sulphadiazine and silver nitrate chlorhexidine creams: emergence of sulphonamide-resistant gram-negative bacilli. British Medical Journal, 1,493. Lynch, J. B. (1979). Thermal burns. In Grabb, W. C. and Smith, J. W. (Eds) Plastic Surgery. 3rd Edition. Boston: Little. Brown and Co. Lynch, J. B., Bray, J. P., Lewis, S. R. and Blocker, T. G. Jr. (1964). Studies with radioisotope labelled albumin in experimental burn wounds. JournalqfSurgical Research, 4.226. Pegg, S. P., Ramsay, K., Meldrum, L. and Laundy, M. ( 1979). Clinical comparison of maphenide and silver sulphadiazine. Scandinavian Journal qf’Plastic and Reconstructire Surgery, 13. 95. Waymack, J. P., Nathan, P., Robb, E. C., Plessinger, R., Rapien. J., Krummel, R., Jenkins, M. and MacMillan, B. G. (I 986). An evaluation of aquaphor gauze dressing in burned children. Burns. 12. 443. Wilson, G. R., Fowler, C.. Ledger, J. and Thorley, M. ( 1986). Evaluation of a new antiseptic dressing in minor burns. Burn.,. 12,518
The Author Harvey S. Stern, BSc(Med), FRCSEd, Registrar. Department of Plastic Surgery and Burns Unit, Repatriation General Hospital, Concord. Sydney. Requests for reprints to: Mr Harvey Bondi Junction. NSW 1022. Australia.
Stern,
Paper received 5 October 1988. Accepted IO February 1989 after revision
33 Bourke
Street.