Emerging Treatments and Technologies

O R I G I N A L A R T I C L E

Maggot Therapy for Treating Diabetic

Foot Ulcers Unresponsive to
Conventional Therapy
RONALD A. SHERMAN, MD, MSC hospitals around the world for treating
bone and soft-tissue infections (14). With
the introduction of antibiotics and other
improvements in wound care, by the
1960s maggot therapy was used only as
OBJECTIVE To assess the efficacy of maggot therapy for treating foot and leg ulcers in
salvage therapy for the most serious
diabetic patients failing conventional therapy. wounds.
Over the past few years, there has
RESEARCH DESIGN AND METHODS Retrospective comparison of changes in been a resurgence in the use of maggot
necrotic and total surface area of chronic wounds treated with either maggot therapy or standard therapy (15), even though its optimal role
(control) surgical or nonsurgical therapy. has not been clearly defined. Large pro-
spective clinical trials have not been con-
RESULTS In this cohort of 18 patients with 20 nonhealing ulcers, six wounds were treated ducted for maggot therapy, and there are
with conventional therapy, six with maggot therapy, and eight with conventional therapy first, no commercial backers to support such
then maggot therapy. Repeated measures ANOVA indicated no significant change in necrotic studies. To assess the utility of maggot
tissue, except when factoring for treatment (F [1.7, 34] 5.27, P 0.013). During the first 14
days of conventional therapy, there was no significant debridement of necrotic tissue; during the
therapy, we analyzed the clinical course
same period with maggot therapy, necrotic tissue decreased by an average of 4.1 cm2 (P 0.02). and outcomes of a cohort of diabetic pa-
After 5 weeks of therapy, conventionally treated wounds were still covered with necrotic tissue tients whose foot and leg ulcers were
over 33% of their surface, whereas after only 4 weeks of therapy maggot-treated wounds were treated with conventional (control) or
completely debrided (P 0.001). Maggot therapy was also associated with hastened growth of maggot therapy.
granulation tissue and greater wound healing rates.

CONCLUSIONS Maggot therapy was more effective and efficient in debriding nonheal- RESEARCH DESIGN AND
ing foot and leg ulcers in male diabetic veterans than was continued conventional care. METHODS
Diabetes Care 26:446 451, 2003 Patient selection
Between 1990 and 1995, 143 patients
with 260 nonhealing wounds were re-
ferred to the maggot therapy service for

I
mpaired wound healing is a common jective has not been met (3), despite many
and costly problem for those with dia- advances in wound care. evaluation and found to be appropriate
betes. Nonhealing diabetic foot ulcers We evaluated the efficacy of an un- candidates for maggot therapy. Non-
account for 2550% of all diabetic hospi- conventional but simple treatment, long candidates were referred elsewhere, usu-
tal admissions, and most of the 60,000 acclaimed as an effective last resort: mag- ally for surgical resection of osteomyelitis
70,000 yearly amputations in the U.S. got therapy. For 70 years, maggot therapy or rapidly advancing soft-tissue infection.
(1,2). Almost 15% of all diabetic patients has been recognized as an aid in debride- After documenting informed consent,
will develop one or more foot ulcers, and ment and wound healing (4 6). Medici- maggot therapy candidates were followed
1525% of those ultimately will require nal maggots secrete digestive enzymes weekly, and their wounds were traced
amputation (2). It is no wonder that one that selectively dissolve necrotic tissue and photographed for at least 8 weeks or
of the disease prevention objectives out- (7), disinfect the wound (8 10), and until hospital discharge. Whenever possi-
lined in the Healthy People 2000 stimulate wound healing (1113). During ble, patients continued receiving the
project was a 40% reduction in the ampu- the 1930s, maggot debridement therapy treatments prescribed by their primary or
tation rate for diabetic patients. That ob- (MDT) was routinely used in hundreds of wound care team for the first 2 weeks of
observation. If the wound did not im-
prove, and if the patient and primary care
From the Veterans Affairs Medical Center, Long Beach, California and the Department of Medicine, Univer-
team consented, then maggot therapy was
sity of California, Irvine, California. initiated. Maggot therapy occasionally
Address correspondence and reprint requests to Dr. Ronald A. Sherman, Department of Pathology, was initiated without baseline observa-
Medical Sciences Bldg, Room D-440, University of California, Irvine, CA 92697-4800. E-mail: tion, especially with patients who refused
[email protected]. further attempts at standard wound care
Received for publication 14 January 2002 and accepted in revised form 1 November 2002.
Abbreviations: MDT, maggot debridement therapy. or patients who were already scheduled
A table elsewhere in this issue shows conventional and Systeme International (SI) units and conversion for amputation. In all, 20 nonhealing foot
factors for many substances. and leg wounds in 18 diabetic patients

446 DIABETES CARE, VOLUME 26, NUMBER 2, FEBRUARY 2003

 Sherman

Figure 1Flow diagram of patients referred to and followed by the maggot therapy service, 18 of whom had wounds and data that qualified for this
analysis.

monitored by our service for at least 2 of porous dacron chiffon or a nylon stock- ventional surgical or nonsurgical therapy
weeks were found to have contours ing was secured to the hydrocolloid ring selected by their primary care staff or the
that could be measured by planimetry, with glue and tape (16,17). This cage-like hospitals wound care team.
making them eligible for this study (Fig. dressing was then topped with a light
1). gauze pad to absorb the necrotic drainage. Wound evaluations
The top layer of gauze was replaced every Ulcer length, width, circumference, and
Administration of therapy 4 6 h, but the cage dressing and maggots surface area were calculated from digi-
Maggot therapy was administered by ap- were left in place for cycles of about 48 h. tized photographic images (Mocha; Jan-
plying disinfected fly larvae to the wound, Maggots were removed by pealing back dell Scientific, San Rafael, CA). Primary
within a cage-like dressing, as previously the dressing with one hand while wiping outcome measures included 1) change in
described (16,17). Larvae of Phaenicia up the larvae with a wet gauze pad held in relative and absolute amounts of necrotic
(Lucilia) sericata, reared and disinfected the other hand. One or two cycles were tissue (defined as nonperfused, nonviable
in our insectary (18), were placed within applied each week; saline- or 0.125% so- soft tissue); 2) change in relative amounts
the wound (five to eight larvae per square dium hypochloritemoistened gauze of granulation tissue (defined as viable,
centimeter) with loose gauze. A ring of dressings were applied during the period well-vascularized, undifferentiated tis-
hydrocolloid (cut from Duoderm; Conva- between MDT cycles and after maggot sue); 3) change in wound surface area
tec, Princeton, NJ) was placed onto the treatment was complete. Patients not re- over time; and 4) the length of time until
skin surrounding the wound. A covering ceiving maggot therapy received the con- complete wound healing. The wound

DIABETES CARE, VOLUME 26, NUMBER 2, FEBRUARY 2003 447

Maggot debridement therapy

Table 1Characteristics of 18 patients and 20 wounds undergoing treatment with standard care or maggot therapy

Conventional therapy MDT

n wounds monitored 14 14
Duration (weeks) 40 (13.5;4312) 44 (21;4318)
Size (cm2) 6.3 (6.6;0.515.5) 13.3 (7.8;0.942)
Circumference (cm) 9.4 (10.3;2.516.6) 13.5 (10.7;3.327.7)
Depth to periosteum or into bone (%) 14 21
Necrotic tissue (% total surface) 44 (35;0100) 38 (38;090)
Granulation tissue (% total surface) 18 (0;090) 19 (1;0100)
Pathologic precursors (26)
Ischemic (%) 7 7
Neuropathic (%) 86 64
Mixed or undefined (%) 7 29
Prior treatment (n [%])
Dry gauze, saline, petrolatum, aloe, other gel 3 (21.4) 3 (21.4)
Topical antimicrobial 1 (7.1) 1 (7.1)
Chemical debriding agent 1 (7.1) 0
Sharp debridement, incision and drainage, other surgical procedure 5 (35.7) 8 (57.1)
Three or more different nonsurgical methods 4 (28.6) 2 (14.3)
Patient age (years) 68 (72;5382) 63 (62;5374)
Underlying illnesses
Peripheral venous or arterial disease (%) 64 93
Cigarette smoker (%) 23 14
Mean ideal body weight (%) 114 129
Mean albumin (g/dl) 3.7 3.7
Mean hemoglobin (g/dl) 12.4 13.2
Receiving systemic antibiotics (n [%]) 2 (14) 3 (21)
Data are means (median and range), unless otherwise specified. Six wounds received conventional therapy only, six received maggot therapy only, and eight received
conventional therapy followed by maggot therapy.

healing rate, based on studies by Gilman (except when less than five cases were ex- apy, and one patient was taken for a
(19) and Margolis et al. (20), was defined pected, thereby invoking Fishers exact below-knee amputation before maggot
as the change in surface area divided by test). Changes in tissue quality and sur- therapy was initiated.

SAt2 1 (surface area at time t2) (surface area at time t1)

(t2 1) ) t2 1
circumferencet 2 1 [(circumference at time t 1) (circumference at time t2)]/2

the mean circumference over time: face area over time were evaluated using Differences between patients receiv-
where t1 initial time of observation, t2 repeated measures ANOVA. The hypoth- ing conventional or maggot treatments
final time of observation, and t2 1 esis of equality of means was discarded are described in Table 1. Maggot-treated
the period of observation (t2 t1), in when the probability (P) of a type I error wounds tended to be larger and contain
weeks. Wound healing rates were calcu- was 5%. Analyses were performed with more necrotic tissue than wounds treated
lated for t2 1 4 weeks, t2 1 8 weeks, SPSS statistical software (SPSS, Chicago, with conventional therapy, but these dif-
and t2 1 duration of treatment. IL). ferences were not statistically significant.
The average and median number of mag-
Statistical analysis RESULTS Of the 20 ischemic and got treatments per wound was 15 and 10,
Normally distributed ordinal and interval neuropathic ulcers in these 18 patients, respectively. Conventional treatments
data were analyzed using Students t test six wounds were treated only with con- prescribed by the patients primary pro-
or logistic regression when variance was ventional therapy, six with MDT, and viders or wound care team were consis-
equal, and Welchs t test when variance eight with conventional therapy first, then tent with the standard of wound care
was not equal. Ordinal and interval data MDT. For the six wounds (in five pa- practiced in our facility: nonmedicated
not normally distributed were evaluated tients) not treated with maggot therapy, dry dressings or saline-moistened wet-
using the Mann-Whitney U test. Nominal one patient did not consent to MDT, three to-dry gauze changed every 8 h (four
data were analyzed using Pearsons 2 test patients spouses did not consent to ther- wounds); topical antimicrobials adminis-

448 DIABETES CARE, VOLUME 26, NUMBER 2, FEBRUARY 2003

 Sherman

Table 2Results of 28 treatments with conventional wound care or maggot therapy

Conventional therapy MDT

n wounds monitored 14 14
Average duration of therapy (weeks) 5.3 (4.16.4) 4.7 (3.56.0)
Debridement
Initial necrotic tissue (% total area) 44 (2267) 38 (2255)
Initial surface area of necrotic tissue (cm2) 2.7 (1.24.4) 5.0 (2.08.0)
Necrotic tissue (% total surface area) at week 2 39 (1958)* 7 (113)*
Percent change in surface area of necrotic tissue 8% (8 to 25)* 59 (125 to 7)*
Weekly change in surface area of necrotic tissue (cm2) 0.4 (0.4 to 1.2) 3.9 (8.4 to 0.6)
Wound size and healing
Initial surface area (cm2) 6.3 (4.28.5) 13.3 (6.420.2)
Surface area at 4 weeks (cm2) 10.9 (6.115.7) 11.8 (4.818.7)
Change in surface area (cm2) 5.0 (0.199.8)* 3.8 (7.0 to 0.6)*
Weekly change in surface area (cm2) 1.15 (0.242.1)* 0.78 (1.6 to 0.1)*
Weekly % change in surface area (final/initial) 27 (4.150) 2 (22 to 18)
Healing rate at 4 weeks 0.08 (0.15 to 0.0002)* 0.08 (0.200.14)*
Healing rate at 8 weeks 0.02 (0.08 to 0.04)* 0.07 (0.040.11)*
Wounds completely closed (%) 21 (044) 36 (765)
Average time until wound closure (weeks) 18 (828) 15 (326)
Quality of wound base: preparation for graft or surgical closure
Initial granulation tissue (% total area) 18 (233) 19 (435)
Granulation tissue at 4 weeks (% total area) 15 (130)* 56 (3477)*
Data are means (median and range), unless otherwise specified. Six wounds received conventional therapy only, six received maggot therapy only, and eight received
conventional therapy followed by maggot therapy. *Differences in mean values where the probability (P) of a type I error is 0.05.

tered three times daily (three wounds); the between-subject factor, and initial dicated no significant change in necrotic
acemannan hydrogel applied 13 times surface area of necrotic tissue as a covari- tissue, except when factoring for treat-
daily (one wound); hydrocolloid pads ate in the analysis. The sphericity assump- ment (F [1.7, 34] 5.27, P 0.013).
12 times weekly (one wound); multiple tion was not met, so the Huyn-Feldt Repeated measures ANOVA for each
nonsurgical modalities (two wounds); or correction was applied. The ANOVA in- treatment arm alone demonstrated that
bedside surgical debridement up to three
times weekly (three wounds).
Maggot therapy was associated with
faster debridement and wound healing
than conventional therapy (Table 2, Figs.
2, 3). MDT-treated wounds saw a 50%
reduction in necrotic surface area (half-
debrided) in 9 days, whereas conven-
tionally treated wounds did not reach that
stage until day 29 (P 0.001). Within 2
weeks, maggot-treated wounds were cov-
ered by only 7% necrotic tissue (0.9 cm2),
compared with 39% necrotic tissue (3.1
cm2) for conventionally treated wounds
(P 0.009). Within 4 weeks, maggot-
treated wounds were completely de-
brided, whereas wounds treated with
conventional therapy for an average of 5
weeks were still covered with necrotic tis-
sue over 33% of their surface (P 0.001).
The efficacy of MDT was further eval-
uated using repeated measures ANOVA, Figure 2Surface area of necrotic tissue over time, during treatment with maggots (F; n 14)
with necrotic tissue surface area as a or standard therapy (E; n 14). Six wounds received conventional therapy only, six received
within-subject factor, treatment group maggot therapy only, and eight received conventional therapy followed by maggot therapy. Error
(MDT versus conventional therapy) as bars indicate standard error. *P 0.05.

DIABETES CARE, VOLUME 26, NUMBER 2, FEBRUARY 2003 449

Maggot debridement therapy

therapy) did not reach statistical signifi-

cance.
These findings support the benefits of
maggot therapy claimed by earlier au-
thors. Mumcuoglu et al. (24) reported
effective debridement for 24 of 27 non-
healing wounds in 22 diabetic patients
treated with an average of six maggot
treatments over the course of 2 weeks; 12
wounds were debrided within just 1
week. Rayman et al. (25) and Fleisch-
mann et al. (26) similarly found maggot
therapy to be a valuable treatment for de-
briding diabetic foot wounds. However,
reported outcomes were subjective, there
were no control groups, and the effects on
wound closure were not evaluated. Only
two prior studies of maggot therapy in-
corporated control groups. Sherman et al.
(22) demonstrated that maggot therapy
Figure 3Percent of wound base covered by granulation tissue, over time, during treatment with enhanced the closure rate of pressure ul-
maggots (F; n 14) or standard therapy (E; n 14). Six wounds received conventional therapy cers in spinal cord injury patients, and
only, six received maggot therapy only, and eight received conventional therapy followed by Wayman et al. (23) demonstrated that
maggot therapy. Error bars indicate standard error. *P 0.05. maggot therapy was associated with more
rapid debridement and reduced cost
when compared to hydrogel for the treat-
MDT was associated with a significant de- got therapy completely debrided all but ment of venous stasis ulcers. Neither of
crease in necrotic tissue (mean debride- 17% of the wound base, whereas 1 week these prospective studies evaluated dia-
ment of 4.1 cm2 within 2 weeks; P of standard therapy failed to debride ne- betic foot ulcers.
0.02), whereas conventional therapy was crotic tissue from 39% of the wound base The present study demonstrated that
not associated with any decrease in ne- (P 0.012). the benefits of maggot debridement ther-
crotic tissue over the same period. Two maggot-treated patients com- apy reported for pressure and venous sta-
Maggot therapy was associated with plained of pain during therapy, but the sis ulcers can also be realized by diabetic
hastened growth of granulation tissue and pain was not severe enough to cause them patients with chronic ischemic and neu-
greater wound healing rates. Within 4 to abandon maggot debridement. The ropathic wounds. The results of this study
weeks, maggot-treated wounds were not same two patients complained of pain should not be misinterpreted as suggest-
only debrided, but were covered with during conventional dressing changes ing that conventional therapy is generally
healthy granulation tissue over about also. ineffective. The overall efficacy of conven-
56% of their wound base. In contrast, tional therapy was not assessed in this
granulation tissue covered only 15% of CONCLUSIONS The current en- study, and it is likely that conventional
the base of those wounds treated conven- thusiasm for maggot debridement has treatments are very effective in treating
tionally (P 0.016). been fueled more by anecdotal reports most wounds. This study simply evalu-
Debridement and healing rates of the and personal experience than by scientific ated wounds that were not responding to
eight wounds treated first with conven- studies. Until now, there has not been a conventional care, and demonstrated that
tional therapy, then with maggot therapy, study comparing maggot therapy to con- MDT was far more effective in treating
were compared by paired t tests. Surface ventional treatments for diabetic foot these chronic wounds than was a trial of
areas at the beginning of conventional wounds. The present analysis demon- another standard therapy. The findings
therapy (3.8 cm2) and maggot therapy strated that maggot therapy is more effec- suggest that we should not consider mag-
(9.7 cm2) were not significantly different, tive and efficient in debriding nonhealing got therapy only as a last resort (an alter-
but their weekly change was significant foot and leg ulcers in diabetic male veter- native to amputation); rather, we should
(increase of 1.0 cm2/week in surface area ans than the typical conventional treat- consider using MDT earlier during the
of wounds during conventional therapy ment currently prescribed. Maggot course of therapy, as a second- or third-
versus decrease of 0.9 cm2/week in sur- therapy was also associated with a more line option.
face area of the same wounds during mag- rapid decrease in wound size and an in- Many questions remain unanswered,
got therapy; P 0.018). Similarly, the crease in granulation tissue, making the and a large prospective evaluation is war-
average portion of wound base covered wounds ready for surgical closure. The ranted. Although MDT debrided wounds,
with necrotic tissue was equivalent at the higher number of patients actually decreased their size, and prepared them
start of conventional therapy (43%) and achieving complete wound closure within for closure more rapidly than did conven-
5.6 weeks later when maggot therapy the 8-week study period (14% with mag- tional therapy, the rate of wound closure
was initiated (37%). Just 1 week of mag- got therapy vs. 0% with conventional was not significantly higher than that as-

450 DIABETES CARE, VOLUME 26, NUMBER 2, FEBRUARY 2003

 Sherman

sociated with standard therapy. A larger mansviable antiseptic in chronic osteo- therapy for treating venous stasis ulcers.
study, preferably with subjects whose dis- myelitis. Proceedings of the International Arch Dermatol 132:254 256, 1996
ease is not as advanced, might better dem- Assembly of the Inter-state Postgraduate 17. Sherman RA: A new dressing design for
onstrate the impact of maggot therapy on Medical Association of North America 371: treating pressure ulcers with maggot ther-
365372, 1929 apy. Plast Reconstr Surg 100:451 456,
complete wound closure. In addition to 5. Pechter EA, Sherman RA: Maggot ther- 1997
issues of efficacy and safety, future studies apy: the medical metamorphosis. Plast Re- 18. Sherman RA, Wyle FA: Low-cost, low-
also must address the cost-effectiveness of constr Surg 72: 567570, 1983 maintenance rearing of maggots in hospi-
MDT and conditions in which maggot 6. Mumcuoglu KY: Clinical applications for tals, clinics, and schools. Am J Trop Med
therapy is likely to be futile (e.g., at what maggots in wound care. Am J Clin Derma- Hyg 54:38 41, 1996
measurable level of hypoperfusion is an tol 2:219 227, 2001 19. Gilman TH: Parameter for measurement
extremity wound unlikely to respond to 7. Vistnes LM, Lee R, Ksander GA: Proteo- of wound closure. Wounds 3:95101,
maggot therapy). lytic activity of blowfly larvae secretions in 1990
experimental burns. Surgery 90:835841, 20. Margolis DJ, Gross EA, Wood CR, Lazarus
1986 GS: Planimetric rate of healing in venous
8. Robinson W, Norwood VH: Destruction ulcers of the leg treated with pressure
Acknowledgments This work was sup- of pyogenic bacteria in the alimentary bandage and hydrocolloid dressing. J Am
ported, in part, by grants from the Spinal Cord tract of surgical maggots implanted in in- Acad Dermatol 28:418 421, 1993
Research Foundation of the Paralyzed Veter- fected wounds. J Lab Clin Med 19:581 21. Levin ME: The diabetic foot: pathophysi-
ans of America (1990), the California Para- 586, 1934 ology, evaluation, and treatment. In The
lyzed Veterans of America (1991), and the 9. Simmons SW: A bactericidal principle in Diabetic Foot. 4th ed. Levin ME, ONeal
Andrus Foundation of the American Associa- excretions of surgical maggots which de- LW, Eds. St. Louis, C.V. Mosby, 1988, p.
tion of Retired Persons (19921995). stroys important etiological agents of pyo- 150
Toni Espinoza Ferrel, MPH, provided criti- genic infections. J Bacteriol 30:253267, 22. Sherman RA, Wyle FA, Vulpe M: Maggot
cal statistical assistance. Many individuals 1935 debridement therapy for treating pressure
contributed to patient care, wound assess- 10. Pavillard ER, Wright EA: An antibiotic ulcers in spinal cord injury patients. J Spi-
ment, data-entry, chart reviews, or administra- from maggots. Nature 180:916 917, 1957 nal Cord Med 18:7174, 1995
tive support, including, Julie Sherman; 11. Baer WS: The treatment of chronic osteo- 23. Wayman J, Nirojogi V, Walker A, Sowin-
Randall Sullivan, RN; Rodney Wishnow, MD; myelitis with the maggot (larva of the ski A, Walker MA: The cost effectiveness
and Frederic Wyle, MD. blow fly). J Bone Joint Surg Am 13:438 of larval therapy in venous ulcers. J Tissue
475, 1931 Viability 10: 9196, 2000
12. Robinson W: The role of surgical maggots 24. Mumcuoglu KY, Ingber A, Gilead L, Stess-
References in the disinfection of osteomyelitis and man J, Friedmann R, Schulman H, Bi-
1. Moss SE, Klein R, Klein BEK: The preva- other infected wounds. J Bone Joint Surg chucher H, Ioffe-Uspensky I, Miller J,
lence and incidence of lower extremity Am 15:409 412, 1933 Galun R, Raz I: Maggot therapy for the
amputation in a diabetic population. Arch 13. Prete PE: Growth effects of Phaenicia seri- treatment of diabetic foot ulcers (Letter).
Intern Med 152:610 616, 1992 cata larval extracts on fibroblasts: mecha- Diabetes Care 21:2030 2031, 1998
2. Frykberg RG: Epidemiology of the dia- nism for wound healing by maggot 25. Rayman A, Stansfield G, Woollard T,
betic foot: ulcerations and amputations. therapy. Life Sci 60:505510, 1997 Mackie A, Rayman G: Use of larvae in the
Adv Wound Care 12:139 141, 1999 14. Sherman RA: Maggot therapy in modern treatment of the diabetic necrotic foot.
3. National Center for Health Statistics: medicine. Infect Med 15:651 656, 1998 The Diabetic Foot 1:713, 1998
Healthy People 2000 Final Review. 15. Sherman RA: Maggot therapy: the last five 26. Fleischmann W, Russ M, Moch D, Mar-
Hyattsville, MD, Public Health Service, years. Eur Tiss Repair Soc Bull 7:9798, quardt C: [Biosurgery: Maggots, are they
2001 (DHHS publ. no. 01-0256) 2000 really the better surgeons?]. Chirurg 70:
4. Baer W: Sacro-iliac jointArthritis defor- 16. Sherman RA, Tran J, Sullivan R: Maggot 1340 1346, 1999

DIABETES CARE, VOLUME 26, NUMBER 2, FEBRUARY 2003 451

The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.