A Randomized, Controlled Trial to Assess the Effect of Topical Insulin Versus Normal Saline in Pressure Ulcer Healing

Login toDownload PDF version
Ostomy Wound Management 2016;62(6):16–23
Shine Stephen, MSc; Meenakshi Agnihotri, MSc; and Sukhpal Kaur, MSc, PhD


Insulin has been used in wound healing to increase wound collagen, granulation tissue, wound tensile strength, and local production of insulin-like growth factors by fibroblasts. Saline is a widely used irrigating and wound dressing solution. Patients admitted to an acute care facility who had a Grade 2 or Grade 3 pressure ulcer were recruited to participate in a randomized, controlled trial to compare the effect of normal saline-impregnated gauze and insulin dressing in pressure ulcer healing. Persons with immunodeficiency, diabetes mellitus, pregnancy, osteomyelitis, and peripheral vascular illness were not eligible for the study.

Study participants were randomized to receive either normal saline dressing gauze or insulin dressing twice daily for 7 days. At baseline, patient demographic data and ulcer history were recorded. Baseline and follow-up ulcer assessments (days 4 and day 7) included ulcer measurement (length and width) and completion of the Pressure Ulcer Scale for Healing (PUSH version 3.0) tool. Patients in the control group received dressings of sterile gauze soaked with normal saline; patients in the intervention group received topical insulin (1 U/cm2 wound area). The insulin was sprayed over the wound surface with an insulin syringe, allowed to dry for 15 minutes, and then covered with sterile gauze. To ascertain the safety of study participants, blood glucose levels were measured with a glucometer 10 minutes before and 1 hour after the topical insulin application in the intervention group. Treatment efficacy was determined by assessing the reduction in wound area and PUSH scores at follow-up. Statistical analysis was performed; data are expressed as mean ± SD and percentage for continuous and categorical variables respectively. The differences in PUSH score and ulcer sizes between the 2 groups were analyzed using independent t-test, and within-group differences were analyzed using ANOVA with repeated measures; Greenhouse-Geisser correction was applied for the 3 consecutive measurements (day 1, day 4 and day 7). Fifty (50) patients (40 men, 10 women), ages 42.46 ± 15.47 years, with 50 ulcers, 25 in each treatment group, were enrolled. At baseline, demographic variables and wound characteristics were comparable between the 2 groups. By day 7, mean wound area had decreased from 11.79 ± 8.97 cm2 (day 1) to 11.43 ± 9.06 cm2 in the saline group (P = 0.566) and from 9.61 ± 6.39 cm2 (day 1) to 6.24 ± 4.33 cm2 (P <0.01) in the insulin group. Mean PUSH scores decreased from 10.52 ± 2.37 at baseline to 10.36 ± 2.40 on day 7 in the saline group (P = 0.475), and from 10.28 ± 1.10 to 8.52 ± 1.58 on day 7 (P <0.01) in the insulin group. No significant decrease in blood glucose level before and after insulin application (P >0.05) was observed. Treatment with topical insulin was found be safe and effective in reducing pressure ulcer size as compared to normal saline-soaked gauze. Future studies utilizing larger sample sizes, longer follow-up times, and different types of chronic wounds and control treatments are warranted.  


A pressure ulcer is a wound caused by incessant pressure or repeated friction that damages the skin and its underlying architecture.1 A 2009, cross-sectional descriptive study2 conducted in Northern California revealed a pressure ulcer incidence between 0% and 5.4% and prevalence between 12% and 19.7%. The major risk factors were serum albumin level <3 (74%), fecal and/or urinary incontinence (73%), fragile skin (67%), and bedbound (63%). Increased life expectancy of the population, admission in long-term-care facilities, increasing health care costs, and advances in treatment modulate the management focus of these wounds.1 A systematic literature review3 performed to consolidate the Cochrane Collaboration guidelines for systematic reviews in Medline, Cumulative Index to Nursing and Allied Health Literature (CINAHL), and Web of Science among the articles published from 2001 to 2013 revealed the cost of pressure ulcer prevention per patient per day varied from 2.65 ? to 87.57 ? across all settings, and the cost of pressure ulcer treatment per patient per day varied from 1.71 ? to 470.49 ? across different settings. A randomized, clinical trial4 performed in 2004 reported the cost of materials needed to treat a pressure ulcer ranged from $112 to more than $6000. A bottom-up costing study5 conducted to estimate the annual cost of treating pressure ulcers in the UK revealed a varied treating cost of £1064 (Grade 1) to £10,551 (Grade 4). 

Treatment recommendations guide evidence-based care for patients with existing pressure ulcers. The treatment recommendations should apply to all individuals with pressure ulcers regardless of the setting. Dressings used for pressure ulcers include hydrocolloid, transparent film, hydrogel, alginate, normal saline, foam, polymeric membrane, silver-impregnated, honey-impregnated insulin, cadexomer iodine, gauze, silicone dressings, collagen matrix, and composite dressings.6

Insulin and wound healing. Wound healing is a complex biologic process that affects gene and protein levels; clot formation, inflammation, granulation tissue development, and remodeling also influence wound healing. One important process initiated during the early stages of healing is reepithelialization; this involves the proliferation, migration, and differentiation of keratinocytes from the wound margins.7,8 In addition to maintaining the growth and development of different cell types, numerous experimental studies suggest insulin has the ability to enhance these processes, particularly proliferation, migration,9 and extracellular matrix secretion by keratinocytes, endothelial cells, and fibroblasts.10

Saline and wound healing. Clinical practice suggests a form of mechanical debridement occurs with frequent wet-to-dry dressing changes that aids in tissue rebuilding. In a quasi-experimental study11 conducted to assess the effectiveness of normal saline in wound healing, it was hypothesized that normal saline dressings have an osmotic effect: the dressing becomes hypertonic with evaporation of water and this provides an osmotic gradient for absorption of wound fluid, which aids in wound healing. Physiological saline is a widely used irrigating and wound dressing solution known to be compatible with human tissue. It causes no damage to new tissue and does not affect the functions of fibroblast and keratinocytes in healing wounds.11

The current study was conducted to evaluate the efficacy of topical insulin in pressure ulcer healing as compared to using saline-soaked dressings.

Materials and Methods

The study was conducted in neurosurgical intensive care units and neurology wards, male and female surgical wards, and trauma wards of the Advanced Trauma Centre of Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh, India over a period of 6 weeks (July 2011 to August 2011). PGIMER is a referral center for several North Indian states.

A patient survey was conducted to select study participants according to the inclusion criteria, which stipulated patients with Grade 2 or Grade 3 pressure ulcers according to European Pressure Ulcer Advisory Panel1 were eligible (see Table 1). Patients with immunodeficiency, diabetes mellitus, pregnancy, osteomyelitis, or peripheral vascular illness were excluded. owm_0616_stephen_table1

The study was approved by the Institutional Ethics Committee. Permission for the study also was obtained from the respective head of the departments. Written informed consent was obtained from all the patients/caregivers.

Study design and procedures. A randomized, controlled trial research design was employed. After baseline assessment of inclusion criteria and grade of pressure ulcer, patients were randomly assigned to the control group (normal saline dressing) or intervention group (insulin dressing), using the nonreplacement lottery method. Protocols for normal saline dressing and insulin dressing use were developed after an extensive literature search, which stipulated twice daily dressing. Researchers’ personal experience and suggestions from coinvestigators were considered for protocol development. Developed protocol was reviewed by experts from nursing and the medical field, all postgraduates in their area of expertise. 

Patients in the control group received a dressing that comprised sterile gauze soaked with normal saline (0.9%). Patients in the intervention group received regular insulin (human Actrapid, Novo Nordisk, Bangalore, India) dressing, 1 U/cm2 wound area. Insulin was cautiously and slowly sprayed over the wound surface with an insulin syringe and needle to avoid dripping out of the wound bed and allowed to dry for 15 minutes before the wound was covered with sterile gauze. The designated intervention was performed by one of the authors twice daily (morning and evening) for 7 continuous days in both groups. To ascertain the safety of study participants, blood glucose levels were measured with a glucometer 10 minutes before and 1 hour after topical insulin application in the intervention group. Normal saline (0.9%) dressing was changed twice daily after soaking sterile gauze with normal saline. The dressing then was secured with sterile gauze and adhesive binding.

Ulcer healing was assessed using the Pressure Ulcer Scale for Healing (PUSH version 3.0).12 The PUSH tool was developed by the National Pressure Ulcer Advisory Panel as a quick and reliable instrument to monitor changes in pressure ulcer status over time. The tool comprises 3 wound characteristics (ie, ulcer size [length and width], exudate amount, and tissue type) that are scored and summed to give a PUSH total score of 0 to 17. A score of 0 indicates complete healing and 17 indicates greatest severity of pressure ulcer. A comparison of total scores measured over time provides an indication of improvement or deterioration in pressure ulcer healing. 

PUSH score and ulcer size (area) were calculated on day 1 before starting the intervention and on days 4 and 7. A sterile transparent paper was placed on the wound to mark the wound borders. The 2 largest perpendicular diameters were measured (in cm) using a ruler. Wound area was calculated by multiplying the 2 diameters to obtain ulcer area in cm2. Tissue type (ie, the tissue present in the wound [ulcer] bed) also was observed and recorded. The estimate of the exudate amount (drainage) present after the removal of the dressing and before applying the new dressing was noted and recorded; exudate amount was scored on a 0 to 3 scale as none, light, moderate, or heavy, respectively. The investigator measured the wound using strict aseptic precautions. Demographic and clinical information was compiled on data collection sheets for the study and then entered into a database.

Statistical analysis. Data from all study participants were tabulated and expressed as mean ± SD and percentage for continuous and categorical variables, respectively. The chi-square test was employed for demographic and wound history variables. The differences in PUSH score and ulcer sizes between the 2 groups were analyzed using independent t-test, and within-group differences were analyzed using ANOVA with repeated measures with Greenhouse-Geisser correction for 3 consecutive measurements. Statistical evaluation of the data was performed using the Statistical Package for Social Sciences (SPSS) version 16.0, Chicago, IL. A P value <0.05 was considered statistically significant.


Baseline characteristics of patients. Of the 78 patients assessed for eligibility, 11 did not meet inclusion criteria and 7 declined to participate. Ultimately, because 5 study participants in each group were lost to follow-up, analysis was performed for 50 persons who completed the 7-day intervention (see Figure 1). owm_0616_stephen_figure1

The mean age of participants in the control group was 41.56 ± 16.53 years (17 Grade 2 and 8 Grade 3 pressure ulcers) and 43.36 ± 14.41 years in the intervention group (20 Grade 2 and 5 Grade 3 pressure ulcers). Nineteen (19) patients in the control group and 21 in the intervention group were male; 6 in the control group and 4 in the intervention group were female. The mean baseline wound area was 11.78 ± 8.97 cm2 and 9.61 ± 6.38 cm2 in the normal saline and insulin groups, respectively. Mean PUSH 3.0 score was 10.52 ± 2.37 in the saline group and 10.28 ± 1.10 in the insulin group. The groups were comparable regarding pressure ulcer grade and patient characteristics (age, gender), as well as clinical characteristics (duration of hospital stay, number of days in hospital before the start of the study, duration of pressure ulcer before starting intervention, and frequency of position change), mean wound area, and PUSH score at baseline (P >0.05) (see Table 2). owm_0616_stephen_table2

Assessment of wound healing as per wound area. The decrease in wound area on day 4 and day 7 was significant in the intervention group (P <0.05). Mean wound area did not differ significantly between 3 days of observation (day 1, day 4, day 7) in the control group (F = 0.382, P = 0.566); whereas, the decrease was significant in the insulin group (F = 7.661, P = 0.009). Overall, mean wound area decreased from 11.79 ± 8.97 cm2 on day 1 to 11.43 ± 9.06 cm2 on day 7 in the control group (P = 0.566) and from 9.61 ± 6.39 cm2 to 6.24 ± 4.33 cm2 in the intervention group (P <0.01) (see Table 3). owm_0616_stephen_table3

Assessment of wound healing as per PUSH score. At day 7, the decrease in the mean PUSH score in the intervention group from 10.28 ± 1.10 to 8.52 ± 1.58 was significant (P <0.01). Between day 4 and day 7, a significant difference was noted between groups in PUSH scores (insulin: F = 14.273, P <0.01; saline: F = 0.609, P = 0.475) (see Table 4). owm_0616_stephen_table4

Pre- and post-intervention blood glucose level in the intervention group. A total of 350 blood glucose level measurements were obtained before and after insulin application. Mean blood glucose levels (mg/dL) before and after intervention were 130.62 ± 24.94 and 128.82 ± 25.14, respectively. Paired t-test revealed the slight reduction in blood glucose was not significant (P >0.05) (see Table 5). owm_0616_stephen_table5


Wound healing is a complex and dynamic process; the person’s health status influences the wound healing environment. The phases of normal wound healing involve hemostasis, inflammation, granulation, and maturation.13 Cell types key to the healing process include macrophages and fibroblasts. Macrophages engulf and destroy bacteria and clean the wound site of debris. Fibroblasts synthesize collagen, the principle component of connective tissue.

Several clinical trials reported the beneficial effect of insulin compared to other agents in wound healing.14 Human growth hormone receptors are present throughout the skin; insulin acts on these receptors and increases reepithelialization as well as collagen content, granulation tissue, wound tensile strength, and local production of insulin-like growth factors by fibroblasts.15-17 Insulin also stimulates proliferation and migration of human keratinocytes, which stimulates cell growth and enhances wound healing.18

In this randomized, controlled trial, normal saline was used as the control. It is compatible with human tissues (ie, does not damage new tissues)19 and does not affect the function of fibroblasts and keratinocytes in healing wounds.7

The results of the present study suggest topical insulin significantly increases the rate of pressure ulcer healing compared with normal saline. A randomized, double-blind, placebo-controlled study14,20 with insulin and zinc also reported wounds treated with insulin healed faster. The dose of insulin (1 U/cm2 wound area) used in the present study also was found to be safe and effective for pressure ulcer management — none of the study participants in the insulin group developed hypoglycemia, and blood glucose levels before and after insulin application did not change significantly (P >0.05).


Study limitations include a small sample size, short (7-day) study duration, and the fact the study was confined to a single setting. Also, a simple method (ruler) was used for wound measurement, and the PUSH score also contains the wound size variable. Due to the small sample size, outcomes were not controlled by ulcer grade or location. Recommendations for future studies include blinding of control and intervention groups, patient follow-up until complete healing occurs, comparison of insulin to moisture-retentive dressing treatment, and evaluation of patient quality of life and cost-effectiveness of the treatment. These studies also should be conducted in populations with other types of chronic ulcers.


The results of a randomized, controlled trial indicate insulin is a safe and effective topical agent that facilitates pressure ulcer healing when compared to normal saline. During the 7-day study, statistically significant differences in PUSH score and wound size change were observed between saline and insulin gauze dressings. Because the study duration was short (7 days), the long-term effect of topical insulin on pressure ulcer healing and other chronic wounds remains to be examined and compared to moisture-retentive dressing regimens. n


1. Jenkins ML, O’Neal E. Pressure ulcer prevalence and incidence in acute care. Adv Skin Wound Care. 2010;23(12):556–559.

2. Dealey C. Review of advances in pressure ulcer management since 1992. Br J Nurs. 2002;(11):486–490.

3. Demarré L, Van Lancker A, Van Hecke A, et al. The cost of prevention and treatment of pressure ulcers: a systematic review. Int J Nurs Stud. 2015;52(11):1754–1774.

4. Kerstein MD. Unexpected economics of ulcer care protocols. South Med J. 2004;(97):135–136.

5. Gallagher SM. Outcomes in clinical practice: pressure ulcer prevalence and incidence studies. Ostomy Wound Manage. 1997;43(1):28–38.

6. Aditya S, Mark SG, Nancy LT. Wound dressings and comparative effectiveness data. Adv Wound Care (New Rochelle). 2014;3(8):511–529.

7. Schilling JA. Wound healing. Surg Clin North Am. 1976;56(4):859–874.

8. Coulombe PA. Wound epithelialization: accelerating the pace of discovery. J Invest Dermatol. 2003;121(2):219–230.

9. Benoliel AM, Kahn-Perles B, Imbert J, Verrando P. Insulin stimulates haptotactic migration of human epidermal keratinocytes through activation of NF-kappa B transcription factor. J Cell Sci. 1997;110(17):2089–2097.

10. Madibally SV, Solomon V, Mitchell RN, Van de Water L, Yarmush ML, Toner M. Influence of insulin therapy on burn wound healing in rats. J Surg Res. 2003;109(2):92–100.

11. Lim, JK,Saliba L, Smith MJ, McTavish J, Raine C, Curtin P. Normal saline wound dressing —is it really normal? Br J Plast Surg. 2000;53(1):42–45.

12. National Pressure Ulcer Advisory Panel. National Pressure Ulcer Advisory Panel PUSH Tool Version 3.0. Available at: www.npuap.org. Accessed February 10, 2011.

13. Keast-Butler J. Honey for necrotic malignant breast ulcers. Lancet. 1980;2(8198):809.

14. Zhang XJ, Wu X, Wolf SE, Hawkins HK, Chinkes DL, Wolfe RR. How local insulin- zinc injection accelerates skin donor site wound healing? J Surg Res. 2007;14(2):90–96.

15. Pierre E, Perez-Polo J, Mitchell A. Insulin-like growth factor-1 liposomal gene transfer and systemic growth hormone stimulate wound healing. J Burn Care Rehabil.1997;18(4):287–289.

16. Biolo G, Fleming R, Wolfe R. Physiological hyperinsulinemia stimulates protein synthesis and enhances transport of selected amino acids in human skeletal muscle. J Clin  Invest. 1995;9(5):811–817.

17. Zhang X, Chinkes D, Irtun O, Wolfe R. Anabolic action of insulin on skin wound protein is augmented by exogenesis amino acids. Am J Physiol Endocrinol Metab.2002;28(2):308–315.

18. Lait M, Smith L. Wound management: a literature review. J Clinical Nurs. 1998;7(1):11–17.

19. Salami AA. A comparison of the effect of chlorhexidine, tap water and normal saline on healing wounds. Int J Morphol. 2006;24(4):673–676.

20. Greenway SE, Filler LE, Greenway FL. The role of topical insulin in wound healing. J Wound Care.1999;8(10):526–528.



Mr. Stephen is a staff nurse, ESIC Medical College Hospital, Parippally, Kollam, Kerala, India. Ms. Agnihotri and Dr. Kaur are faculty, National Institute of Nursing Education, Post Graduate Institute of Medical Education and Research, Chandigarh, India. Please address correspondence to Shine Stephen at: shinestephentn@gmail.com.