Povidone-iodine solution in wound treatment.

The second factor complicating decisions about using

povidone-iodine is that it may be used in a variety of

ways. Wounds may be irrigated or soaked once or

repeatedly with povidone-iodine solution.

Povidone-iodine solution also can be applied for longer periods as

part of the dressing. There are no studies comparing the

effects of these methods. Povidone-iodine solution also

may be used full strength (10%) or diluted to any

desired concentration prior to use. Research results

should be interpreted based on the specifics of the

application used.

Recent positions taken by two federal agencies -- the

Food and Drug Administration (FDA) and the Agency

for Health Care Policy and Research (AHCPR) -- have

implications for the use of povidone-iodine solution in

wound treatment. The FDA has approved povidone-iodine

for use in nonprescription first-aid antiseptic products.[2]

Use of the term "first aid" implies that povidone-iodine can

be used for short-term treatment (approximately 1 week)

and on relatively superficial and acute wounds. In assessing

the evidence regarding use of povidone-iodine, the FDA

report states:

Controlled studies on wound healing were conducted in

animals and humans and involved various types of dermal 

wounds.... Both superficial and deeper wounds were studied

with a contralateral control.... Results showed that there

were no statistically significant differences in mean healing

times between any of the treatment groups and their

controls. In addition, microscopic analysis showed no

differences in wound healing in the groups studied. These

pathological and histological studies did not indicate any

deleterious effect of povidone-iodine on wound healing.

However, there was also no evidence demonstrating that

povidone-iodine might aid wound healing.[2]

The FDA has issued no

position statement on

povidone-iodine use

for prolonged periods

or in treating chronic

wounds.

The AHCPR has

published guidelines for

treatment of pressure ulcers.[3] These guidelines state,

"Do not clean ulcer wounds with skin cleansers or

antiseptic agents (eg, povidone-iodine, iodophor,

sodium hypochlorite solution [Dakin's solution],

hydrogen peroxide, acetic acid)."[[3].sup.(p50)] Due to the stature of

the AHCPR, as well as the strong proscriptive wording,

the guidelines may be used in future liability actions

where pressure ulcers were treated with

povidone-iodine. Murphy argues that clinicians will be held

accountable to the guidelines as "the most effective

and appropriate standard of care based on current

and exhaustive scientific research and available

evidence."[[4].sup.(p31)] Murphy believes that if treatment is not

in accordance with the guidelines, clinicians must

document why. The implication here is that use of

povidone-iodine, as well as other antiseptic agents, can no longer

be considered a customary treatment for pressure ulcers.

Although the guidelines are directly applicable only to

pressure ulcers, the evidence on which they are based

appears to be applicable to all wound treatments

involving povidone-iodine. This update will examine the

research evidence regarding the safety and efficacy of

povidone-iodine solutions in preventing infections and

in promoting wound healing.

Safety

Wound healing is a complex process involving many

physiological events. Immunological resources are

recruited to fight infection and debride damaged tissue.[5]

Blood supply in the healing area is reestablished

(angiogenesis).[6] Regeneration of tissue (cell proliferation,

fibroplasia)[6] follows, replacing damaged or destroyed

tissue. The area to be healed is decreased via wound

contraction.[6] Closure of the wound is achieved through

epithelial cell migration.[6] Finally, remodeling of scar

tissue occurs to approximate prior appearance and

function.[6] A safe treatment should promote, or at least

not impair, this process.

In Vitro Studies

Researchers have examined the effect of

povidone-iodine on several of the cellular components of the

wound healing mechanism. Van Den Broek and

co-workers[7] found povidone-iodine solution at concentrations

greater than 0.05% to be toxic to granulocytes;

monocytes showed some effects of toxicity at concentrations

above 0.05% and complete toxicity at

concentrations above 1%. Tatnall and colleagues[8] found

concentrations of povidone-iodine greater than 0.004% to

be 100% toxic to keratinocytes. Lineaweaver et al[9]

identified 0.05% as a safe concentration of povidone-iodine

for fibroblasts; higher concentrations (including the

10% concentration that is commonly used in clinical

practice) were 100% cytotoxic. These studies show that

in vitro povidone-iodine, unless it is diluted to a far lower

concentration than that commonly used in clinical

settings, is toxic to all of the cell types that are essential to

the healing process.

In Vivo Studies

Lineaweaver et al[9] irrigated surgically induced wounds in

rats with several solutions, including saline and 1%

povidone-iodine, three times daily. At 4 days postsurgery,

tensile strength in the wounds treated with

povidone-iodine was only 21% that of the wounds treated with

saline. There were no differences at 8, 12, or 16 days

postsurgery, despite continued irrigation.

Epithelialization was found to be delayed in the wounds treated with

povidone-iodine at 4 and 8 days postsurgery, but not

thereafter.

Br[Angstrom]nemark et all[10] used a microwound in the cheek

pouch of a hamster as a model to investigate the effects

of povidone-iodine on microcirculation. Exposure for 60

minutes to 1% povidone-iodine solution resulted in

cessation of blood flow in surface capillaries. Circulation

did not resume within 1 hour. No alteration in blood

flow was found with exposure to 1% povidone-iodine

solution for 5 minutes, nor was any alteration of blood

flow in capillaries covered with epithelial tissue found

with exposure to 1% povidone-iodine solution for 60

minutes. No circulatory changes were seen in the cheek

pouches of control hamsters in which the wounds were

treated with saline. Brennan and Leaper[11] created

wounds in rabbit ears. The wounds were enclosed in a

plastic chamber. When the wounds were fully

granulated, they were "flooded" with saline or one of several

antiseptic solutions, including 5% and 1%

povidone-iodine. Brennan and Leaper did not clearly state how

long the solution remained in contact with the wound.

They inspected the wounds microscopically to examine

the effects of the test solutions on the microcirculation

in the granulation tissue. Saline had "little or no effect"

on the blood flow. The 1% povidone-iodine solution

gave similar results. With the 5% povidone-iodine

solution, however, there was cessation of blood flow in the

granulation tissue, which did not fully recover for 72

hours. Br[Angstrom]nemark and colleagues[12] examined the effects

of disinfectants on microcirculation in wounds to

connective, synovial, and nerve tissue in mice, hamsters, and

rabbits as well as dermal tissue in humans. The findings,

based on qualitative evaluation using vital micrography,

electron micrography, and vital angiography showed "a

very slight reaction" in wound microcirculation when

exposed to a 1% solution of povidone-iodine.

Br[Angstrom]nemark and colleagues commented that other

disinfectants produced a much greater reaction.

Hughes-Papsidero and Levine[13] exposed the carotid 

artery and vagus nerve in wound beds of rabbits. Wounds

were kept open until the arteries were 100% covered

with granulation tissue. One group of rabbits received

daily topical application of saline, and a second group

of rabbits received daily topical application of 10%

povidone-iodine. The wounds in both groups were

treated identically except for the topical agent that was

used. No differences in rate of healing were found

between the groups, nor was any damage to arterial

tissue identified.

Kjolseth and coworkers[14] compared the effects of

bacitracin (500 U/g), silver nitrate (0.5%), silver

sulfadiazine (1%), mafenide acetate (8.5%), and

povidone-iodine (10%) on two measures of healing in

full-thickness wounds in mouse ears. The test substances

were applied once daily to the wounds and covered with

a dressing. Wounds treated with povidone-iodine required

more time to epithelialize (11.8 [+ or -] 0.55 days) than did

controls (7.2 [+ or -] 0.7 days) or wounds treated with silver

sulfadiazine (7.1 [+ or -] 0.3 days) or mafenide acetate (7.3 [+ or -] 0.3

days). Interestingly, the wounds treated with povidone-iodine

showed complete neovascularization in less time

(15.0 [+ or -] 0.4 days) than the wounds treated with the other

topical agents (range: 15.3 [+ or -] 0.7 to 18.4 [+ or -] 0.56 days).

Gruber et al[15] compared times to complete wound epithelialization

in both partial-thickness and full-thickness

wounds in rats. Wounds were treated by applying hydrogen

peroxide (3%), povidone-iodine, acetic acid (0.25%), or

saline (control) to the surface of the wound four times

daily. Healing times for the wounds treated with

povidone-iodine or saline were not different. In an additional

experiment, split-thickness skin graft donor sites in humans were

treated with the same antiseptics or saline by application

with cotton blotter four times daily. Again, no differences

in the time to epithelialization (pink surface free of scabs)

was found between wounds treated with saline or

povidone-iodine. Gruber et al did not specify the concentration of

povidone-iodine used in their study; it is identified as

"Betadine" and therefore is probably the commercially

available 10% stock solution.

Despite the cytotoxicity documented in in vitro studies,

the results of in vivo studies seem to suggest that

povidone-iodine does not interfere with healing,

especially if it is used at concentrations of 1% or lower.

Povidone-iodine may temporarily decrease blood flow in

the wound bed at higher concentrations, as shown by

Brennan and Leaper,[11] but concentrations of 1% or less

do not appear to have this effect.[10-12] The effects of

repeated use of povidone-iodine solution on

microcirculation have not been investigated. All of the above

studies[7-15] used healthy animal or human subjects with

acute, surgically induced wounds.

Systemic Toxicity

Patients have developed systemic iodine toxicity as a

result of iodine absorption from wounds dressed with

gauze soaked in povidone-iodine or when

povidone-iodine solution was used as a wound irrigant.[16-18]

Patients developed decreased renal function or renal

failure following 10 hours of continuous irrigation of

their wounds with povidone-iodine or 17 days to 5 weeks

of wound dressing with gauze soaked in

povidone-iodine. The four patients described in these reports,[16-18]

aged 50 to 83 years, had multiple health problems,

including preexisting renal insufficiency (n = 3), diabetes

(n = 2), and congestive heart failure (n = 1). The

wounds involved were pressure ulcers or debrided septic

hip wounds. Systemic toxicity does not appear to be a

common occurrence. No toxicity has been reported with

povidone-iodine used as a brief rinse or soak.

Efficacy

In Vitro Studies

In the laboratory, povidone-iodine has been

demonstrated to be effective at killing a broad range of the

pathogens generally associated with wound infection.[19]

Berkelman et al[20] found that povidone-iodine solutions

diluted to concentrations of 0.1% to 5% were more

effective in killing common wound contaminants than

was the 10% stock solution. Even the 10% solution was

completely effective within 4 minutes of exposure.[20] Van

Den Broek and coworkers[7] found povidone-iodine to be

effective against Staphylococcus aureus at concentrations of

0.005% or higher, but they questioned whether this

effectiveness would be true in vivo. Lineaweaver and

associates9 found povidone-iodine to be an effective

bactericide at a concentration of 0.001%.

Povidone-iodine is not always effective at killing

common bacteria. Anderson[21] discussed two reports of

povidone-iodine stock solution (10%) contaminated

with Pseudomonas sp. The contamination apparently

occurred during production of the povidone-iodine

solution. The bacteria remained viable for several weeks

and were eventually involved in patient infections. Why

povidone-iodine failed to kill these bacteria is not

known. These isolated incidents are inconsistent with

other in vitro findings[7,9,19,20] and cannot be explained by

the concentration of povidone-iodine solution involved.

In Vivo Studies

There are a small number of studies in which the ability

of povidone-iodine to control infection in dermal

wounds was examined. Dire and Welsh[22] studied wounds

treated in a hospital emergency department. No

differences were found in infection rates between wounds

irrigated with povidone-iodine and those irrigated with

normal saline.

Rodeheaver et al[23] inoculated experimental wounds in

guinea pigs with 102 to 107 organisms of S aureus. Ten

minutes later, the wounds were irrigated with either

povidone-iodine solution (10%) or normal saline. Four

days after treatment, the authors found no difference

between the two groups in the number of viable bacteria

present in the wounds or in the number of wounds with

visible purulent exudate. When the same experiment

was conducted using povidone-iodine surgical scrub, the

wounds treated with povidone-iodine had higher rates of

infection than those treated with saline. Wounds

contaminated with 103 organisms showed 60% infection

when treated with povidone-iodine versus 0% with

saline. Inoculation with 104 organisms produced 90%

infection when treated with povidone-iodine versus 0%

with saline. With 105 organisms, wounds treated with

povidone-iodine were 100% infected versus 15% for

saline controls.

Edlich and coworkers[24] also created wounds in guinea

pigs, which they inoculated with S aureus. Five minutes

later, the wounds were irrigated with either a

povidone-iodine solution (10%) or saline. After 4 days, wound

infection as shown by visible purulent exudate was lower

for the wounds treated with povidone-iodine than for

the saline-treated wounds. There were no differences in

percentage of positive cultures and area of induration.

The findings of this study regarding visible purulent

exudate appear to be contradictory to the results of the

study by Rodeheaver et al.[23]

Kucan and associates[25] examined infection rates in

pressure ulcers under various treatment conditions. The

threshold for infection was defined as 105 bacteria per

gram of tissue on biopsy. Wounds were treated with

gauze dressing saturated with either saline or

povidone-iodine (10%). Treatment was given for 3 weeks. At the

end of that time, 78.6% of the saline-treated wounds had

bacteria counts below the infection threshold, compared

to with 63.6% of the wounds treated with

povidone-iodine. No statistical analysis of this difference was

reported. The saline dressings were changed every 4

hours, whereas the povidone-iodine dressings were

changed every 6 hours. No explanation was given for this

difference in procedure.

Other researchers have investigated the antibacterial

properties of povidone-iodine for use in surgery. Amstey

and Jones[26] found povidone-iodine to be no more

effective than normal saline for preventing infection

when used as a vaginal irrigant before vaginal

hysterectomy. Sindelar and Mason[27] found that superficial

infections of surgical wounds irrigated with 10%

povidone-iodine solution prior to closure had an overall infection

rate of 2.9%. Wounds irrigated with saline had a 15.1%

infection rate.

Viljanto[28] found that irrigation of appendectomy 

wounds with 1% povidone-iodine before closure

resulted in a reduction in wound infection when

infection before surgery was isolated to the appendix (2.6%

infection rate versus 8.6% for saline controls). If

infection had spread beyond the appendix before surgery,

however, there was no difference in infection rates

between wounds irrigated with povidone-iodine and

saline-irrigated controls.

In several of the studies discussed, wounds treated with

povidone-iodine solution were compared with wounds

treated with saline to assess impairment of healing. If

povidone-iodine is helpful in promoting wound healing

by decreasing infection rates, healing rates would be

expected to be faster in wounds treated with

povidone-iodine. Some researchers,[9,13,15] however, found no

difference in healing rates between wounds treated with

povidone-iodine and saline-treated controls. This finding

suggests that the antibacterial effects of the

povidone-iodine either did not promote healing or were offset by

some other effect such as cytotoxicity.

Although povidone-iodine clearly is an efficient

bactericide in vitro, the benefits in treating actual wounds

appear to be inconsistent at best. Several studies

demonstrated no difference in infection rates between

wounds treated with povidone-iodine solution and

wounds treated with other topical agents. In the only

study involving chronic wounds, povidone-iodine

treatment seemed to be inferior to treatment with saline.[25]

Summary and Clinical Considerations

Most in vivo research on the use of povidone-iodine was

conducted on experimentally or surgically induced

acute wounds. The applicability of these findings to the

chronic wounds typically seen by physical therapists has

not been demonstrated. Apparently, some people

assume that chronic wounds, especially those containing

necrotic tissue, have a greater risk of infection and

therefore a greater need for treatment to decrease the

number of surface bacteria. Moreover, many patients

with chronic wounds have generally compromised

health status and may be less able to produce an effective

immune response to bacterial invasion. These patients,

therefore, may need more assistance to prevent infection

than patients with acute wounds and fewer systemic

complications.

Conversely, patients with compromised health status

may be more susceptible to the cytotoxic character of

povidone-iodine. In light of the findings of Berkelman et

al[20] that povidone-iodine was more effective when

diluted to a concentration of 0.1% to 5%, use of a less

concentrated solution than the 10% stock solution may

be prudent if povidone-iodine is the treatment of choice.

Limited evidence[23] suggests that povidone-iodine

surgical scrub may increase infection rates when used in open

wounds.

Edlich and associates[29] examined the decrease in

effectiveness of antibiotic therapy when the amount of time

that a wound was left open before closure increased.

They found that bacteria became coated with a

"fibrinous coagulum" derived from the wound drainage, which

served to protect them from antibiotic action. Howell et

al[30] suggest that the effectiveness of povidone-iodine in

wounds might also be decreased through this

mechanism. This suggestion may explain the finding by Kucan

and colleagues[25] that povidone-iodine did not reduce

bacteria counts in pressure ulcers. Further investigation

with chronic wounds is necessary to establish the benefit,

if any, of using povidone-iodine as an adjunct to wound

treatment.

The optimal method of application of povidone-iodine

has not been clearly established. Brief contact, such as

wound irrigation, especially if followed by a saline rinse,

or use of diluted solutions might minimize the risk of

cytotoxicity. Prolonged contact such as packing the

wound with gauze saturated with povidone-iodine,

however, might enhance the bactericidal effects. The safety

and efficacy of these alternatives have not been

compared in either acute or chronic wounds.

The use of povidone-iodine for wound packing requires

particular scrutiny. This treatment option precludes the

use of occlusive or semiocclusive dressings. A thorough

discussion of the benefits of these "moist environment"

dressings is beyond the scope of this update; however,

these dressings have been shown to decrease wound

infection rates.[31] Saydak[32] reported the results of a pilot

study comparing wound healing rates in pressure ulcers

dressed with either an unsaturated, amorphous hydrogel 

absorption dressing (Hydra-Gran [dagger]) (moist healing

environment) or povidone-iodine solution cleansing

followed by normal saline rinse and dry gauze dressing.

Wounds treated with the absorption dressing healed

more quickly. The percentage of decrease in the length

of the longest axis of the wound was more than double,

and the percentage of decrease in depth was more than

nine times that of wounds treated with povidone-iodine

and dry gauze. No data were reported regarding

infection rates. Although this study did not isolate the effect

of povidone-iodine from that of dry dressings, it suggests

that moist environment dressings may be a viable,

possibly safer, alternative to povidone-iodine.

Wound packing with gauze soaked with povidone-iodine

also presents a small danger of systemic toxicity.

Andrews[33] discusses these concerns in patients receiving

povidone-iodine dressings for prolonged periods. The

author recommends that such patients be observed for

symptoms of iodine toxicity, which include

"hypercalcemic metabolic acidosis, cardiovascular instability

(bradycardia, hypertension), elevation of hepatic enzymes,

central nervous dysfunction, and progressive renal

insufficiency." In addition, these patients should be

monitored every 3 days for elevated serum urea, nitrogen,

sodium, potassium, and osmolarity.[33]

Povidone-iodine solution appears to be a relatively safe

treatment for small acute wounds. Its safety for

treatment of patients with extensive or chronic wounds has

not been adequately investigated. The evidence

regarding efficacy of povidone-iodine solution in treating

patients with acute wounds is inconclusive. There is

insufficient evidence to demonstrate effectiveness in

treating chronic wounds. Better alternative treatments

may be available, including the use of moist

environment dressings. Clinically and legally adequate reasons

to use povidone-iodine solution for managing wounds

do not currently exist.

References

[1] Mayer DA, Tsapogas MJ. Povidone-iodine and wound healing: a

critical review. Wounds. 1993;5:14-23.

[2] 56 Federal Register 33644 at 33662.

[3] Clinical Practice Guideline No. 15: Treatment of Pressure Ulcers. Rockville,

Md: Agency for Health Care Policy and Research, Public Health

Service, US Department of Health and Human Services; 1994.

[4] Murphy RN. Legal and practical impact of clinical practice

guidelines on nursing and medical practice. Advances in Wound Care.

September/October 1996;9:31-34.

[5] Kloth LC, McCulloch JM. The inflammatory response to wounding.

In: McCulloch JM, Kloth LC, Feedar JA, eds. Wound Healing: Alternatives

in Management. Philadelphia, Pa: FA Davis Co; 1995:1-15.

[6] Daly TJ. Contraction and re-epithelialization. In: McCulloch JM,

Kloth LC, Feedar JA, eds. Wound Healing. Alternatives in Management.

Philadelphia, Pa: FA Davis Co; 1995:32-46.

[7] Van Den Broek PJ, Buys LMF, Van Furth R. Interaction of

povidone-iodine compounds, phagocytic cells, and microorganisms. Antimicrob

Agents Chemother. 1982;22:593-597.

[8] Tatnall FM, Leigh IM, Gibson JR. Comparative toxicity of antimicrobial

agents on transformed keratinocytes. J Invest Dermatol 1987;89:

316-317. Abstract.

[9] Lineaweaver W, Howard R, Soucy D, et al. Topical antimicrobial

toxicity. Arch Surg. 1985;120:267-270.

[10] Br[Armstrog]nemark PI, Albrektsson B, Lindstrom J, Lundborg G. Local

tissue effects of wound disinfectants. Acta Chir Scand. 1966;357(suppl):

166-176.

[11] Brennan SS, Leaper DJ. The effect of antiseptics on the healing

wound: a study using the rabbit ear chamber. Br J Surg. 1985;72:

780-782.

[12] Br[Armstrong]nemark PI, Ekholm R, Albrektsson B, et al. Tissue injury

caused by wound disinfectants. J Bone Joint Surg Am. 1967;49:48-62.

[13] Hughes-Papsidero J, Levine H. Effects of benzoyl peroxide on the

stimulation of granulation tissue over major neck vessels. Otolaryngol

Head Neck Surg. 1984;92:362-365.

[14] Kjolseth D, Frank JM, Barker JH, et al. Comparison of the effects of

commonly used wound agents on epithelialization and

neovascularization. J Am Coll Surg. 1994; 179:305-312.

[15] Gruber RP, Vistnes L, Pardoe R. The effect of commonly used

antiseptics on wound healing. Plast Reconstr Surg. 1975;55:472-476.

[16] Aronoff GR, Friedman SJ, Doedens DJ, Lavelle KJ. Increased serum

iodide concentration from iodine absorption through wounds treated

topically with povidone-iodine. Am J Med Sci. 1980;279:173-176.

[17] Dela Cruz F, Brown DH, Leikin JB, et al. Iodine absorption after

topical administration. West J Med. 1987;146:43-45.

[18] D'Auria J, Lipson S, Garfield M. Fatal iodine toxicity following

surgical debridement of a hip wound: case report. J Trauma. 1990;30:

353-355.

[19] LaRocca R, LaRocca MAK, Ansell JM. Microbiology of

povidone-iodine: an overview. In: Digenis GA, Ansell JM, eds. Proceedings of the

International Symposium on Povidone. Lexington, Ky: College of

Pharmacy, University of Kentucky; 1983:101-119.

[20] Berkelman RL, Holland BW, Anderson RL. Increased bactericidal

activity of dilute preparations of povidone-iodine solutions. J Clin

Microbiol. 1982;15:635-639.

[21] Anderson RL. Iodophor antiseptics: intrinsic microbial 

contamination with resident bacteria. Infect Control Hosp Epidemiol. 1989:10;

443-446.

[22] Dire DJ, Welsh AP. A comparison of wound irrigation solutions used

in the emergency department. Ann Emerg Med. 1991;19:704-708.

[23] Rodeheaver GT, Bellamy W, Kody M, et al. Bactericidal activity and

toxicity of iodine-containing solutions in wounds. Arch Surg. 1982; 117:

181-186.

[24] Edlich RF, Custer J, Madden J, et al. Studies in management of the

contaminated wound, III: assessment of the effectiveness of irrigation

with antiseptic agents. Am J Surg. 1969;118:21-30.

[25] Kucan JO, Robson MC, Heggers JP, Ko F. Comparison of silver

sulfadiazine, povidone-iodine, and physiologic saline in the treatment

of chronic pressure ulcers. J Am Geriatr Soc. 1981;29:232-235.

[26] Amstey MS, Jones AP. Preparation of the vagina for surgery. JAMA.

1981;245.839-841.

[27] Sindelar WF, Mason GR. Irrigation of subcutaneous tissue with

povidone-iodine solution for prevention of surgical wound infections.

Surg Gynecol Obstet. 1979;148:227-231.

[28] Viljanto J. Disinfection of surgical wounds without inhibition of

normal wound healing. Arch Surg. 1980; 115:253-256.

[29] Edlich RF, Smith QT, Edgerton MT. Resistance of the surgical

wound to antimicrobial prophylaxis and its mechanisms of

development. Am J Surg. 1973;126:583-591.

[30] Howell JM, Dhindsa HS, Stair TO, Edwards BA. Effect of scrubbing

and irrigation on staphylococcal and streptococcal counts in

contaminated lacerations. Antimicrob Agents Chemother. 1993;37:2754-2755.

[31] Hutchinson JJ, Lawrence JA. Wound infection under occlusive

dressings. J Hosp Infect. 1991;17:83-94.

[32] Saydak SJ. A pilot test of two methods for the treatment of pressure

ulcers. Journal of Enterostomal Therapy. 1990; 17:139 -142.

[33] Andrews LW. The perils of povidone-iodine use. Ostomy/Wound

Management. 1994;40 (1):68, 70, 72, 73.