Research

                   Research Proposal by Jessica John Posko

Title: Is Manuka honey dressing more effective than conventional povidone iodine dressing in the healing of diabetic foot ulcers?

Description

In local context, the prevalence for diabetes in Singapore is 12.8%, which means that 1 in 8 adults are diabetic (IDF, 2015). An estimated 700 diabetics have lower limb amputations annually due to diabetic foot complications locally (Nather et al., 2010). Although various treatments are used to treat a diabetic foot in Singapore, most modern approach results in adverse reactions, uneconomical, time-consuming and requires a lot of effort. Consequently, this leads to low quality patient care due to longer length of stay in local hospitals and increasing healthcare costs. Hence, it is paramount to use effective diabetic wound treatment to improve overall patient quality of life.  

This research aims to study the therapeutic effect of Manuka honey on the duration of diabetic foot ulcer wound healing in comparison to conventional povidone iodine 10% dressing in an acute tertiary care hospital here in Singapore.

Background and Justification

                   The introduction of antibiotics caused decreased use of honey in wound healing but the widespread problem of multiple anti-biotic resistant bacteria resulted in a renewed interest in honey application for wound healing (Sharp, 2009). Few studies were done to compare different conventional dressings with honey dressing in treating wounds, specifically in treating diabetic foot ulcers wounds. In a latest randomized study by Gulati et al., (2014), gamma-radiated honey were used comparing to povidone iodine in treating chronic wounds which resulted in 31.82% in the honey group attaining complete wound healing while none in the povidone group. However, Moghazy et al., (2010) contended that processed honey might alter its effectiveness specifying the alteration in the ration of its components. 

                   This shows the uncertainty that still exists in using pure and raw honey or processed honey in wound treatment. Another study conducted by Parimala and Rani (2014) among 100 patients in a hospital in South India also favours the treatment of honey in contrast to povidone. But the use of purposive sampling in the study is biased as it influences the outcome. Besides that, the quality of life experienced by patients that are measured after the treatment is very subjective and influenced as patients are not blinded to the study and may feel obliged to make positive comments.

Using a combination of betadine and silver dressing in comparison to honey dressing, 4 out of 5 different wound categories studied showed significantly faster wound healing time with honey dressing compared to conventional dressing, specifically for diabetic foot wound which took 28.5 days with honey compared to 45 days with conventional dressing, p<0.05 (Kshirsagar et al., 2014).                         However, limited studies were conducted on the effect of a combination dressing with honey in the treatment of diabetic foot ulcers.

On the contrary, in a prospective randomized trial, no statistical significance were found in the wound healing time between honey and povidone iodine 10% dressed group in Wagner Grade- II diabetic foot ulcers, but patients treated with honey showed less pain than patients in control group. (Shukrimi, Sulaiman, Halim & Azril, 2008). Using the same Wagner diabetic foot classification system, Jan, Shah, Khan, Fayyaz and Ullah (2012) studied all four grades of foot ulcers healing time also showing no significant statistical change in healing rate with p<0.6658. However, the results seem unfair since the type of honey wasn’t mentioned.   

In the context of burn wounds, the duration of wound healing was shorter in honey group healing in 18.16 days compared to SSD of 32.68 days (Baghel, Shukla, Mathur & Randa, 2009). The general results, reflecting the nature of burn patients, are similar to the report by Gupta et al., (2011). However, both groups did not describe the concentration of honey used in their study.

                  Furthermore, the effectiveness of honey dressing in comparison to SSD dressing among patients with superficial and partial thickness burn covering less than 40% of total body surface area (TBSA) was studied by Malik, Malik and Aslam (2010), which showed a shorter wound healing rate of 21 days in honey group and 24 days in SSD. Based on similar types of study subjects researched in India on the other hand, Subrahmanyam (1998) proved 16% higher epithelisation rate and reparative rate in honey group than SSD, showing the efficacy of honey in early subsidence of acute inflammatory changes, faster wound healing and good infection control. Recruiting the same number of patients as Subrahmanyam (1998) but with 15% TBSA burn, Mashood, Khan and Sami (2006) found that an additional 2 weeks was needed for the 1% SSD cream treated wounds to heal compared to honey. 

             Although less pain was reported in honey group, no tools were mentioned about how it was measured. In a study carried out in Pakistan by Sami et al., (2011) four years ago from now, similar dressing material of honey and 1% SSD cream were used among the same number or patients of 50 presenting with corresponding results but with an additional emphasis on the cost where honey dressing amounted less than half (48.5%) of SSD.

              In the streamline of burn wound treatment, precision was shown in the study where honey dressing was compared to mafenide acetate dressing in Iranian patients with superficial thermal burns (Maghsoudi et al., 2011) resulting the same as the study by Subrahmanyam (1998) with specification to the amount of honey used as 16-30ml depending on wound size. However, a wide age range of patients from 3 to 70 years old used here may influence the outcome as the wound healing rate is different at certain age. On the contrary, Hashemi, Bayat, Kazemei and Azarpira (2011) tested mafenide acetate against honey dressing on rabbits with auricular burns in rabbits revealing the ineffectiveness of honey in deep wounds given its high healing property. Although relevant, this study revolves around animals and is not patient-oriented to be used as reference in clinical practice.

            Given that the therapeutic effect of honey versus conventional dressing on various wound healing researches were carried out in a number of countries, there are none done locally here in Singapore.

                                    

Research Objective

To compare the effectiveness of honey dressing with conventional povidone iodine 10% dressing in achieving a shorter duration of wound healing among diabetic patients with foot ulcers aged between 35 to 55 years old in a local acute tertiary care hospital in Singapore.

PICOS:

Patient/Problem:  Diabetic patients with foot ulcers

Intervention: Manuka honey dressing

Comparison: Povidone iodine 10% dressing

Outcome: Duration of diabetic ulcer wound healing

Setting: Diabetes Outpatient Centre in National University of Singapore (NUHS)

Research Question

Is Manuka honey dressing more effective than conventional povidone iodine 10% dressing in achieving a shorter duration of wound healing among diabetic patients with foot ulcers aged between 35 to 55 years old at the Diabetes Outpatient Centre in National University Hospital in Singapore (NUHS)?

Hypothesis

Manuka honey dressing is more effective than povidone iodine 10% dressing in achieving a shorter duration of wound healing among diabetic patients with foot ulcers aged between 35 to 55 years old at the Diabetes Outpatient Centre in NUHS

Null Hypothesis

There is no difference between Manuka honey dressing and povidone iodine 10% dressing in achieving a shorter duration of wound healing among diabetic patients with foot ulcers aged between 35 to 55 years old at the Diabetes Outpatient Centre in NUHS.

Ethical Considerations

Prior to carrying out this research study, a voluntary informed consent from the research participants is obtained ensuring their legal capacity to give consent without applying any elements of force. Adequate explanation on the informed consent will be given as detailed in Appendix A to ensure participants understand about the research to be conducted without any intentions of fraud. The confidentiality of participants is ensured by storing participant’s data in the office computer for analysis purposes encrypted with strong password that is only accessible with an authorized staff pass. The anonymity of the research respondents will be maintained by using study codes.

     Besides that, an ethics approval for this study is obtained with a structured letter as described in Appendix B from the Singhealth Centralized Institutional Review Board, the Director of Nursing of National University Hospital, National Health Group Domain Specific Review Board and Singapore Health Sciences Authority to ensure all relevant policies and procedure were abided and provide all research participants with assurance that the research is of high quality, ethical and safe and to show that the results of this research can be relied upon and used as evidence in future researches.

Sampling size

A sample size of 110 diabetic participants is included. The probability of type I error (>) of 0.05 (2-sided) and powers of 0.80 is used for the sample size estimation. The larger the sample size, the lower the sampling error (Endacott & Botti, 2005).     

Target population: All adults with diabetic foot ulcer  

Accessible population:  All adults with diabetic foot ulcer at the Diabetes Outpatient Clinic in NUHS

Sampling Method

A total of 110 participants will be recruited by convenience sampling method. This is the most common type of non-random sample, whereby; the subjects are whoever that’s convenient to the researcher (Panacek & Thompson, 2007). Randomized numbers labelled on opaque sealed envelopes containing either honey dressing or alginate dressing treatment will be distributed to the first 110 participants who met the sampling criteria at the Diabetes Outpatient Centre in National University Hospital of Singapore by a diabetic wound nurse to select samples. Subjects are allocated into two study arms, the honey treatment group and the control group dressed with povidone iodine 10%.                      Approximately two is required to reach a desired amount of 120 sample size. However, biasness may exist in the method of subject selection into the sample (Panacek & Thompson, 2007).          

Research Design and Methodology

In this study, a non-equivalent, single- blinded comparison quasi experimental design will be used. Quasi-experiment differs from true experiment design because participants aren’t randomly picked from a significant population and aren’t randomly allocated into group (Martella, Nelson, Morgan & Marchand-Martella, 2013). Since this study attempts to focus on the concentration of participants specifically from the Diabetes Outpatient Centre in National University Hospital of Singapore by using a non-probability convenience method sampling, a randomized controlled trial (RCT) cannot be carried out. Consequently, majority of the subjects will not be randomly allocated to treatment groups.  

          Quasi-experimental design provides significant experimental control based on natural setting because it compares the performance of the experimental group with the control group (Martella et al., 2013). This quantitative experimental design is suitable in this case study as subjects are divided into two the experimental honey dressing group and the control povidone group. A quasi-experimental design uses comparison groups that have a close resemblance with the treatment group in relation to baseline properties (White & Sabarwal, 2014). Participants are selected as similar to each other as possible so that an equal and fair comparison can be done with the experimental and control group. However, it is unsure whether or not the groups are comparable. Given for an instance the subjects are randomly selected in this study, it is very unlikely that the groups are similar to each other because the groups are often non-equivalent.    

Inclusion criteria

Participants deemed suitable based on the University of Texas diabetic foot ulcer classification system, aged between 35 to 55 years old, male or female, has unilateral or bilateral diabetic foot ulcer, cognitively stable to give voluntary informed consent, has a follow up with the Diabetes Outpatient Centre in NUHS and are available for 2 weeks.

Exclusion criteria

Participants with known allergy to honey, povidone iodine and cotton, those with burn wounds and skin grafts donor sites, unwilling to attend follow-up sessions, pregnant females, those with multiple medical illnesses besides diabetes as results may be influenced by confounding factors, infected diabetic foot ulcers requiring antibiotics, ulcers with bone inflammation (osteomyelitis), those on steroid therapy and with arterio-venous insufficiencies.

Independent variable

The types of wound dressing application

Dependent variable

The duration for the diabetic foot ulcer wound to heal

Extraneous Variable

The social status, occupation, education level, diet, age, marital status and the social status of participants.

Recruitment process

Registered nurses at the Diabetes Outpatient Centre at NUHS collect the data. The diabetes wound specialist nurses and doctors at the centre will receive a written and verbal instruction about the process stages of recruitment, data accumulation and treatment procedure. This involves informing eligible participants who met the criteria regarding the main intention of the study, explaining the data collection and treatment procedure, including the randomization of honey and povidone iodine dressing treatment during their visit to gain consent.

              Randomized numbers labelled from 1 to 120 on opaque sealed envelopes containing either honey or povidone dressing treatment and a diabetic foot questionnaire form to be filled is distributed by the diabetes wound specialist nurse at the centre to the first 120 participants eligible. The participants, clinical and research staff are blinded to the contents of the envelopes. Two diabetes wound nurse and one consultant will review the wound to ensure there are no contradictions to the treatment to be applied. 120 qualified participants will then be divided into 60 subjects in the experimental honey treatment group and 60 in control povidone iodine group who will receive the treatment at the Diabetes Outpatient Centre at NUHS from 9am to 5pm on weekdays and 9am to 1pm on Saturday.

Data collection  

Participants will have their parameters and clinical history taken and will fill a diabetic foot questionnaire form structured as the diabetes foot care questionnaire by the Diabetes Care Program of Nova Scotia because it holistically identifies participant’s foot care knowledge, promoting foot care as described in Appendix C. According to Polit and Beck (2010), questionnaire is inexpensive, preserves anonymity, has no interviewer biasness and saves energy and time. Duplex ultrasound is done to assess any lower limb arterio-venous insufficiencies and X—ray is done to rule out any bone inflammations like osteomyelitis. One consultant and two diabetic wound nurse will inspect the foot ulcers at the centre using geometrical rules in square centimetres to measure the total surface area, depth, size and width of the wound on the first visit and consequently every 2nd week interval for 24 weeks of treatment period. Wounds in both treatment groups is first debrided by a resident with at least 2 years of clinical experience and cleansed with warm normal saline prior to dressing application during follow up visits. Raw unprocessed Manuka honey soaked gauze is applied over the wound in the honey treatment group. Similarly, povidone iodine 10% soaked gauze is applied in the control treatment group. Both dressing are covered with sterile gauze, cotton pad and bandaged.

         Blood glucose level is measured using glucometer upon every visit to maintain optimum level. Photograph is taken on the first visit of the participant for baseline comparison and consequently on every follow- up visits. Bacterial swab is taken for culture and sensitivity testing prior to treatment and at the end of the research duration. The presence of granulation, epithelialization, size, depth, width and signs of infection is assessed by the physician on the first visit. Consequently on every 2nd week interval, the outcome is assessed by a consultant blinded by the group allocation and dressings were changed before the consultant enters the room. If infection develops during the course of study, antibiotics is prescribed with regular bacterial swabs.

          The percentage of wound healing is recorded as: [(area of first day- area of second time/area at first time)] x 100 (Malik et al., 2010). The University of Texas Diabetic Wound Classification (UT) system is used to grade the ulcers (refer to Appendix D). The UT system is more descriptive including the risks for amputation and prediction of ulcer healing combining grade and stage proved better than the Wegner system (Oyibo et al., 2001). Data collected during each follow-up is entered in Mircosoft Excel spread sheet by the registered nurses.    

Data Analysis

At the final 6 months, a statistician blinded to the treatment type received by both groups will analyse the data using Social Package for Social Science (SPSS) program version 23.0. The gender, dressing type, age, time of dressing application, wound size and duration of wound healing of participants is collated and summarised in percentage and frequency distribution. To study the mean distinction between both study arms on the duration of wound healing, data will be statistically analysed using parametric t-test and analysis of variance (ANOVA). Level of significance is set as p<0.05. To study the wound variables and baseline demographics, independent samples t-test is used to compare the treatment and control group.

Validity

Type 1 error is kept <0.05. As the data generated in Mircosoft Excel spreadsheet keeps changing, data was copied onto multiple sheets so that it’s not lost for future analysis and to prevent internal validity threats. Selection bias was reduced by distributing sealed envelopes which are randomly numbered.

Reliability

The University of Texas Diabetic Wound Classification system is extensively used in clinical trials by health practitioners (Oyibo et al., 2001). Thus, the chances for correlation are high when evaluating the test for stability over time by test-retest.  

Conclusion

The rate of diabetic foot ulcers is increasing over time especially in Singapore. The usage of honey as an effective dressing material is inexpensive making it affordable for patients and reducing healthcare costs. It is a safe alternative without any known adverse complications reported. With faster wound healing, patients’ visits to healthcare facilities will be reduced consequently minimizing unnecessary admissions. With shorter length of stay in hospitals, healthcare professionals will have more time to provide better quality patient care. Hence, Manuka honey could be the next effective treatment implemented in clinical practices with extensive research.     

        

        

                    

References

Baghel, P., Shukla, S., Mathur, R., & Randa, R. (2009). A comparative study to       evaluate the effect of honey dressing and silver sulfadiazene dressing on wound healing in burn patients. Indian Journal Of Plastic Surgery, 42(2), 183. doi: 10.4103/0970-0358.59276.

Endacott, R., & Botti, M. (2005). Clinical research 3: Sample selection. Intensive          and Critical Care Nursing, 21(1), 51-55. doi: 10.1016/j.iccn.2004.11.001.

Gulati, S., Qureshi, A., Srivastava, A., Kataria, K., Kumar, P., & Ji, A. B. (2014). A prospective randomized study to compare the effectiveness of honey dressing vs. povidone iodine dressing in chronic wound healing. Indian Journal of Surgery, 76(3), 193-198. doi: 10.1007/s12262-012-0682-6.

Hashemi, B., Bayat, A., Kazemei, T., & Azarpira, N. (2011). Comparison between topical honey and mafenide acetate in treatment of auricular burn. American Journal Of Otolaryngology, 32(1), 28-31. doi: 10.1016/j.amjoto.2009.07.005.

International Diabetes Federation (2015), IDF Diabetes Atlas 6th edition. Retrieved September 9 2015, from http://www.idf.org/sites/default/files/attachments/WP%20factsheet.pd.

Jan, W. A., Shah, H., Khan, M., Fayaz, M., & Ullah, N. (2012). Comparison of conventional pyodine dressing with honey dressing for the treatment of diabetic foot ulcers. Journal of Postgraduate Medical Institute, 26(4), 402-407.

Kshirsagar, A., Mahna, A., Pednekar, A., Vekariya, M., Chaudhari, R., & Gupta, V. et al. (2014). To compare manuka honey dressing with conventional dressing in surgical wound. Journal of Evidence Based Medicine And Healthcare, 1(11), 1418-1424. doi: 10.18410/jebmh/2014/211.

Maghsoudi, H., Salehi, F., Khosrowshahi, M. K., Baghaei, M., Nasirzadeh, M., & Shams, R. (2011). Comparison between topical honey and mafenide acetate in treatment of burn wounds. Annals of Burns and Fire Disasters, 24(3), 132-137.

Malik, K. I., Malik, M. A. N., & Aslam, A. (2010). Honey compared with silver sulphadiazine in the treatment of superficial partial-thickness burns. International Wound Journal, 7, 413–417.

Martella, R. C., Nelson, J. R., Morgan, R. L., & Marchand-Martella N. E. (2013). Understanding and interpreting educational research. New York, NY: Guilford Press.

Mashood, A. A., Khan, T. A., & Sami, A. N. (2006). Honey compared with 1% silver sulfadiazine cream in the treatment of superficial and partial thickness burns. Journal of Pakistan Association of Dermatologists, 16, 14-19.  

Moghazy, A., Shams, M., Adly, O., Abbas, A., El-Badawy, M., & Elsakka, D. et al. (2010). The clinical and cost effectiveness of bee honey dressing in the treatment of diabetic foot ulcers. Diabetes Research And Clinical Practice, 89(3), 276-281. doi: 10.1016/j.diabres.2010.05.021.

Nather, A., Chionh, S. B., Tay, P. L., Aziz, Z., Teng, J. W., Rajeswari, K., … Nambiar, A. (2010). Foot Screening for Diabetes. ANNALS Academy of Medicine Singapore, 39(6), 472-475. Retrieved from http://www.annals.edu.sg/pdf/39VolNo6Jun2010/V39N6p472.pdf.               

Oyibo, S. O., Jude, E. B., Tarawneh, I., Nguyen, H. C., Harkless, L. B., & Boulton, A. J. M. (2001). A comparison of two diabetic foot ulcer classification systems: The Wagner and the University of Texas wound classification systems. Diabetic Care, 24, 184-188.

Panacek, E., & Thompson, C. (2007). Sampling methods: Selecting your subjects. Air Medical Journal, 26(2), 75-78. doi: 10.1016/j.amj.2007.01.001.

Parimala, P. J. & Rani, S. R. (2014). A Comparative Study on the Effect of Honey and Povidone Iodine Ointment on Pain, Wound Healing and Quality of Life of Patients with Diabetic Ulcer. International Journal of Health Sciences and Research (IJHSR), 4(7), 179-185.

Polit, D. F., & Beck, C. T. (2010). Essential of nursing research: Appraising evidence for nursing practice. Walnut Street, Philadelphia: Lippincott Williams & Wilkins.

Sami, A. N., Mehmood, N., Qureshi, M. A., Zeeshan, H. K., M, I. A., & Khan, M. I. (2011). Honey compared with silver sulfadiazine as burn wound dressing. Annals of Pakistan Institute of Medical Sciences, 7(1), 22-25.

Sharp, A. (2009). Beneficial effects of honey dressings in wound management. Nursing Standard, 24(7), 66-74. doi: 10.7748/ns2009.10.24.7.66.c7331.

Shukrimi, A., Sulaiman, A. R., Halim, A. Y., & Azril, A. (2008). A comparative study between honey and povidone iodine as dressing solution for Wagner type II diabetic foot ulcers. Med J Malaysia, 63(1), 44-46

Subrahmanyam, M. (1998). A prospective randomised clinical and histological study of superficial burn wound healing with honey and silver sulfadiazine. Burns, 24(2), 157-161. doi: 10.1016/s0305-4179(97)00113-7.

White, H., & Sabarwal, S. (2014). Quasi-experimental design and methods. Methodological Briefs: Impact Evaluation 8, UNICEF Office of Research, Florence.