Nutrition Affects Chronic Wound Outcomes

Reference: Ohura T, Nakajo T, Okada S, Omura K, Adachi K. Evaluation of effects of nutrition intervention on healing of pressure ulcers and nutritional states (randomized controlled trial). Wound Repair Regen. 2011;19(3):330–336.

Rationale: Inadequate nutrition is believed to limit cell function and repair, and it is a vital factor to address before enrolling patients in a clinical trial. Guidelines recommend patients consume 30 kcal to 35 kcal and 1.25 g to 1.5 g of protein per kg of body weight for optimal PU prevention based on weak evidence.

Objective: Conduct a multicenter, open RCT with carefully controlled nursing care exploring effects of a standardized nutritional intervention on PU healing and patient nutritional states.

Methods: Consenting elderly, tube-fed inpatients with a National Pressure Ulcer Advisory Panel (NPUAP) Stage III or IV PU in the coccygeal, trochanteric, calcaneal, or sacral area; a serum albumin of 2.5 to 3.5; a Japanese Pressure Ulcer Score (OH) ≤ 8.5; and a Braden scale of 9 to 17 were enrolled in the study. Patients were excluded if they had severe renal or liver dysfunction, uncontrolled or unmanageable diabetes mellitus or other systemic conditions, a malignant tumor within the last 5 years, more than 1 PU, or if their PU was infected or contained > 20% necrotic tissue obscuring its surface or was undermined > 2 cm. Qualifying patients were randomly assigned to receive (n = 30) or not receive (n = 30) the nutritional intervention, stratifying patient assignment to minimize group differences on enrollment based on important PU risk factors for delayed healing. Control patients were assigned to receive the same caloric intake as prior to enrollment. Intervention patients received calories calculated using the Harris-Benedict equation with an active factor of 1.1 and a stress factor of 1.3 to 1.5. Both groups received their supplements, adding 4.38 g protein, 2.23 g fat, and 15.62 g carbohydrate per 100 mL of a commercially available feeding formula for patients hospitalized with extensive burns or after surgery. Patients were treated for 12 weeks after their diet reached specified supplement values. Wounds were classified according to NPUAP staging. Healing was documented using the Depth, Exudate, Size, Inflammation/Infection, Granulation, and Necrotic Tissue (DESIGN) Tool. Nursing treatment met Japanese Society of Pressure Ulcers³ standards, with all patients turned every 2 hours and using a standardized pressure-redistribution mattress and nursing activities for both groups. In addition to PU area, depth, DESIGN, OH, and Braden scale parameters, the following patient nutritional outcomes were recorded at enrollment, intervention day 0, and every 2 weeks through week 12: body weight; various skinfold thickness and limb circumference measurements; and laboratory measures of blood levels of albumin, prealbumin, cholinesterase, cholesterol, lymphocyte count, hemoglobin, copper, zinc, and iron. Appropriate analyses of variance, chi-square analyses, or Wilcoxon tests as well as Fisher’s exact test for safety effects compared groups at baseline and following interventions using P < .05 for treatment efficacy and safety effects. 

Results: Both groups were comparable on all parameters at enrollment. Due to withdrawals for protocol violations or discontinuance, 21 intervention patients and 29 control patients were analyzed for efficacy. Protein and calorie intake increased in the intervention group, accompanied by correspondingly greater increases over time for patients receiving the intervention in the physical parameters of weight (P < .001), waist circumference (P < .001), suprailiac skinfold thickness (P < .005), and thigh circumference (P < .05) as well as 2 laboratory parameters: pre-albumin, a marker for protein deficiency (P < .05), and copper (P < .001). During the 12-week trial, 4 (14%) subjects healed in the control group and 7 (33%) in the intervention group (significance not reported). The PU area and depth decreased more in the intervention group over time, with P < .05 after 8 or more weeks of treatment. There were no significant differences between groups over time in OH scores, any DESIGN parameter, or in PU risk as measured by Braden scale scores. 

Authors’ Conclusions: Using standardized mattresses and nursing treatment, a controlled nutritional intervention enhanced PU healing, despite groups being comparable for PU risk factors.

Reference: Armstrong DG, Hanft JR, Driver VR, Smith AP, Lazaro-Martinez JL, Reyzelman AM, Furst GJ, Vayser DJ, Cervantes HL, Snyder RJ, Moore MF, May PE, Nelson JL, Baggs GE, Voss AC; Diabetic Foot Nutrition Study Group. Effect of oral nutritional supplementation on wound healing in diabetic foot ulcers: a prospective randomized controlled trial [published online ahead of print June 19, 2014]. Diabet Med. 2014;31(9):1069–1077.

Rationale: Ten to 25% of those with diabetes mellitus will develop a DFU. Elderly individuals whose diets were supplemented with arginine, glutamine, and β-hydroxy β-methylbutyrate increased wound-related hydroxyproline, reflecting increased collagen deposition. These results suggest that these nutritional supplements, which decrease protein breakdown and preserve nitrogen balance, may help heal DFUs.

Objective: Conduct a RCT exploring the efficacy of arginine, glutamine, and β-hydroxy β-methylbutyrate nutritional supplements on healing of University of Texas Grade 1-A noninfected, nonischemic, superficial DFUs.

Methods: After institutional board approval, a 16-week, multicenter, prospective, double-blind RCT assigned consenting, community-dwelling individuals with controlled type 1 or type 2 diabetes mellitus to drink a nutritional supplement dissolved in 237 mL of water twice daily of either a low-glycemic control powder (control) or an identically packaged powder (commercially available experimental supplement) similar in caloric content with 7 g L-arginine, 7 g L-glutamine, and 1.5 g calcium β-hydroxy β-methylbutyrate (providing 1.2 g β-hydroxy β-methylbutyrate dihydroxy-β-methylbutyrate) added.  Subjects were assigned to groups using stratified randomization by area (1–3 cm² and > 3–10 cm²), duration (1–6 months and > 6–12 months), and glycemic control (hemoglobin A_1c [HbA_1c] < 9% or 9%–12%) to assure similar risk of nonhealing in control and experimental groups. Subjects were excluded if they had a DFU > 10 cm² in area, longer than 12 months’ duration, Charcot deformity that could not be adequately offloaded, or HbA_1c > 12%. Subjects reported quality of life (QoL) using the EuroQol-5D and Diabetic Foot Ulcer Scale Short Form and had blood drawn for routine testing at enrollment and on a weekly basis. Coprimary outcomes were percent of patients with a healed study DFU at 16-week study end and time to complete healing. Secondary outcomes were percent of subjects experiencing ≥ 15% wound area reduction during the first week of study or ≥ 50% wound area reduction during the first 4 weeks of study, change in study DFU area or patient-reported QoL, and incidence of complications including infections, new DFU appearance, DFU recurrence, or amputation. Sample size was set at 134 per group based on an expected 20-subject drop-out rate and a chi-square test with a power of 80% to detect a 20 percentage-point difference between groups as statistically significant using one-sided test with alpha = 0.025. This exceeded the sample size (n = 98) required for log rank differences in percentage of coprimary outcome events based on the same assumptions. 

Results: Of 1052 subjects screened, 130 experimental and 141 control subjects were enrolled from 38 US, European, and Taiwanese centers between July 2008 to August 2010. One experimental subject never received the product and did not qualify for analysis. The 2 groups were comparable on all baseline characteristics and did not differ significantly on any primary or secondary efficacy or safety outcome. However, a post hoc analysis of 127 subjects with a baseline serum albumin ≤ 40 g/L showed that there was a significantly greater proportion of subjects with total wound healing at 16 weeks in the experimental (31/61; 50.8%) group than the control (23/66; 34.9%) (P = .03). Similarly, a greater proportion of experimental (60.3%) than control (39.3%) subjects healed by 16 weeks in the subset of 119 subjects with an ankle-to-brachial index (ABI) < 1.0 (P = .008). A similar difference was found between experimental and control proportions healed for subjects with combined ABI < 1.0 and serum albumin ≤ 40 g/L (P = .0042). No clinically significant differences were observed in any safety or blood coagulation variable. 

Authors’ Conclusions: Although no significant improvement was observed in overall DFU healing as a result of the experimental compared with the control supplement, patients with risk of poor limb perfusion and/or protein malnutrition may benefit from receiving the experimental as compared with the control supplement. Further studies using the experimental supplementation in these high-risk subgroups are needed to confirm or refute these findings.

Clinicians are only beginning to learn how to study the effects of nutritional supplements on healing. Though neither of these studies reported consistent healing benefits of nutritional supplements on overall complete PU or DFU healing, both studies illustrate important opportunities to improve study designs and outcome measurements in wound care. A protein supplement administered for at least 8 weeks within a good standard of PU care reduced PU area and depth significantly more than a control supplement.¹ Although this finding was based on a small study with more than 25% dropouts in the intervention group, it highlights the merit of future RCT studies exploring potential effects of diet for improving physical parameters and reducing clinically important wound characteristics like depth and area⁴ that are strongly associated with healing time.⁵ As suggested following the DFU study,² it may be more fruitful to focus nutritional supplements on patient-centered risk factors, such as malnutrition and/or compromised circulation, than on all patients with a DFU. Perhaps clinicians can learn from oncology and other fields regarding the power of patient-centered care: focusing interventions on precisely what each patient needs rather than assuming one therapy fits all. It makes more sense to provide nutritional supplements to those deficient in the nourishment provided than to test efficacy of nutrient supplements on all subjects regardless of their deficiencies or needs. Both studies point out the need for further RCTs, each shedding light on how to perform key studies teaching how to give the right patient the right therapy at the right time to derive the best results. 

Laura Bolton, PhD
Adjunct Associate Professor
Department of Surgery
Rutgers Robert Wood Johnson
Medical School
New Brunswick, NJ