Abstract: Objective. This study examined the impact of diabetes on inflammatory responses among individuals with infected chronic wounds. The findings provide evidence of the extent to which infection may go unrecognized among persons with diabetes. Methods. This study was a secondary analysis of another study. Chronic wounds were assessed for signs of infection and classified as infected or noninfected based on quantitative cultures of wound tissue. Infected wounds were those with greater than 105 organisms per gram of wound tissue or containing b-hemolytic Streptococcus. Diabetes status was determined through medical record review. The classical signs of inflammation (ie, pain, erythema, edema, heat, and purulent exudate) were identified using the Clinical Signs and Symptoms Checklist (CSSC). Clinical assessments were blinded to the infection status. Results. Seventeen subjects with infected, nondiabetic foot ulcer wounds were included in the sample. Seven (41%) had type 2 diabetes; none had type 1 diabetes. Diabetes was not associated with wound type, wound size, wound duration, or wound oxygen levels. The frequencies with which each of the 5 signs of inflammation was expressed did not differ by diabetes status with the exception of erythema. Three (43%) of the 7 persons with diabetes expressed erythema in their infected chronic wound compared to 8 (80%) of the 10 persons without diabetes. Conclusions. Because erythema is due to local vasodilation, the inactivation of nitric oxide by advanced glycosylation endproducts may be a key pathophysiological mechanism that underlies its diminished expression in persons with diabetes.

Acknowledgment: This study was funded in part by HSR&D, Nursing Research Initiative, Department of Veteran’s Affairs (NRI 1-005-01), Washington, DC, and National Institutes of Health, National Institute of Nursing Research (NINR RO1 NR07721), Bethesda, Md.

I

t is commonly asserted that diabetes causes diminished inflammatory responses,¹ although this assertion has never been empirically verified. The purpose of this study was to examine the impact of diabetes on localized inflammatory responses among individuals with infected chronic wounds. The specific aim was to describe differences in the expression of pain, erythema, edema, heat, and purulence in infected chronic wounds comparatively between individuals with diabetes and those without diabetes. The findings provide evidence of the extent to which infection may go unrecognized among persons with diabetes.

Methods

       This study was a secondary analysis of data collected for another study² to examine clinical signs of localized chronic wound infection. Chronic wounds were assessed for signs and symptoms of infection and classified as infected or noninfected based on quantitative cultures of viable wound tissue. Clinical assessments were blinded to the infection status of the wound.

Setting and Sample

       A 100-bed veteran’s facility in the Midwest served as the primary study setting. The University of Iowa Institutional Review Board approved human subject procedures, and informed consent was obtained from all participants.
       Subjects were enrolled using the following criteria: 1) presence of a full-thickness, nonarterial chronic wound, 2) white blood cell count (WBC) greater than 1,500 cells/mm³, 3) platelet count greater than 125,000/mm, 4) no coagulopathies, and 5) no anticoagulation therapy. These criteria were employed to decrease the risks associated with wound biopsy. For this study, a subset of subjects with infected, nondiabetic foot ulcer wounds was selected. Wound etiology was determined based on clinical history and physical and diagnostic examinations (eg, ankle-brachial indices, pulse volume recordings, Doppler studies, angiography, or air plethysmography). Diabetic foot ulcers were excluded because of their direct association with diabetes and were defined as ulcers on the plantar or dorsal foot surface in patients with diabetes. Infected wounds were defined as those with greater than 105 organisms per gram of wound tissue or containing β-hemolytic Streptococcus at any level. This definition of wound infection is widely accepted.^3–6 Laboratory procedures used to culture wound specimens have been reported elsewhere.²

Study Variables

       Study variables included diabetes status and clinical signs and symptoms of infection associated with inflammation. Diabetes status was determined through medical record review and subject report. Level of diabetes control was measured using HbA1c values. The Clinical Signs and Symptoms Checklist (CSSC) was developed to standardize the assessment of signs and symptoms associated with localized infection and contains 5 items specific to the classical signs of inflammation (ie, pain, erythema, edema, heat, and purulent exudate). Development and testing of the CSSC was reported elsewhere.⁷ Secondary variables were measured in order to more fully describe the sample including demographics, albumin level, wound type, size, and duration, and oxygen level (T_CPO₂).

Study Procedures

       The research team collected the study data. The T_CPO₂ sensor was secured on the skin just proximal to the study wound and equilibrated for 20 minutes prior to recording values. Following cleansing with saline gauze, the wound margin was outlined on transparent film and the surface area determined using a Lasico Planimeter/Digitizer (Lasico, Los Angeles, Calif). The depth of the wound was measured using a cotton-tipped swab placed in the deepest portion of the wound. A dry gauze dressing was placed in the wound cavity for 1 hour to standardize conditions for assessing wound exudate. Blood samples were collected. After removing the dry gauze dressing, the wound was assessed using the CSSC. A second observer repeated this step independently. Wound tissue biopsy was performed according to the procedures outlined by Stotts.⁸ Tissue specimens were transported to the laboratory for quantitative cultures.

Results

       Fifty-two nondiabetic foot ulcer wounds were assessed for signs of infection with the CSSC and for infection status using quantitative tissue cultures. Of these, 17 (33%) were infected based on the aforementioned criteria. Diabetes was not significantly associated with having an infected wound.
       The mean (SD) age of the 17 subjects with infected chronic wounds was 67.7 (14.01) years; 13 (77%) were male, and 16 (94%) were white. Mean albumin level was 3.5 (0.71) gm/dL, and mean WBC was 8821.4 (3896.19) K/uL. Seven (41%) had type 2 diabetes; none had type 1 diabetes. Diabetes was not significantly associated with age, gender, albumin level, or WBC. Mean HbA1c among subjects with diabetes was 7.0 (1.27).
       Of the 17 infected wounds, 7 (41%) were venous ulcers, 5 (29%) were pressure ulcers, 3 (18%) were traumatic wounds, and 2 (12%) were secondary incisions. Seven (41%) were located on the lower leg, 4 (24%) on the malleolus, 2 (12%) on the sacrum, 2 (12%) on the heel, and 2 (12%) on the foot. The 2 study ulcers located on the foot were on subjects without diabetes. The mean surface area was 11.3 (19.50) cm2; mean depth was 1.0 (1.46) cm. The mean TCPO2 level was 44.2 (18.33) mmHg, and the mean duration was 147.4 (337.19) weeks. Diabetes was not associated with wound type, size, duration, or T_CPO₂ levels. Table 1 presents the frequencies for each of the 5 signs of inflammation by diabetes status. Diabetes was not significantly associated with the expression of any of these signs or symptoms. However, it is clear that fewer persons with diabetes express erythema in their infected chronic wounds compared to persons without diabetes.

Table 1

Discussion

       Although diabetes was not associated with having an infected chronic wound or other parameters, such as wound oxygen levels, the findings of this study are noteworthy with respect to expression of inflammatory signs of infection. Individuals with diabetes with infected chronic wounds expressed erythema less often than individuals without diabetes with infected chronic wounds. Since erythema is caused by local vasodilation, the inactivation of nitric oxide by advanced glycosylation endproducts⁹ may be a key pathophysiological mechanism that underlies this observation.

Conclusion

       This preliminary study needs to be replicated with a larger sample in order to more fully identify differences in inflammatory responses between individuals with diabetes and those without diabetes. However, these findings suggest that at least some inflammatory responses are markedly diminished with diabetes.