Anemia Prevalence Among Patients With Diabetic Foot Ulcers Necessitating Surgery on Admission: A Preliminary, Retrospective Comparative Study

Anemia is more common among patients with diabetes compared with the general population, even after adjusting for renal function.^1-3 Based on the WHO sex-specific definitions of anemia as Hb less than 12 g/dL in women and less than 13 g/dL in men, up to 23% of patients with diabetes could be classified as having anemia.^2,4 Variables such as HbA_1c and C-reactive protein, as well as risk factors such as albuminuria and diabetes duration, are associated with low levels of Hb.^5,6 In the presence of altered microcirculation, anemia is considered to be an aggravating factor for wound healing, ulcer infection, lower extremity amputation, and death.^6,7

A recent meta-analysis demonstrated a clear association between the presence of anemia and DFUs; anemia was significantly more prevalent in patients with diabetes and foot complications compared to those with diabetes but without foot complications (P =.01).⁸ The same authors concluded that the severity of anemia was also associated with the severity of DFD.⁸ To the current authors’ knowledge, only 2 studies have reported such an association among patients who required amputation owing to DFD.^9,10 No studies were identified that included patients who had undergone conservative surgery as the initial management of foot complications. Furthermore, in the Middle East region, late presentation to a health care provider is common; a recently published study noted that patients had advanced DFUs at the first consultation, which often necessitated admission for surgery.¹¹

This retrospective pilot study is an attempt to determine a possible relationship between DFD severity and anemia among a homogenous series of patients who underwent surgical treatment on their first admission. Therefore, the primary purpose of this preliminary report is to look for a potential association between Hb level and the severity of DFD using comparative subgroups of infected vs. noninfected DFUs.

Sample selection

The records of patients who underwent surgical treatment for DFD between January 2019 and December 2019 were retrospectively examined. The ethics committee of the authors’ institution approved the research project prior to commencement of the study. The series was continuous, and data from all patients who met the inclusion criteria were included in the analysis. Only patients experiencing their first episode of DFD were included. In cases of multiple admissions, the charts of the first admission were extracted for inclusion.

Inclusion and exclusion criteria

For this study, the term diabetic foot disease (DFD) encompasses DFU, diabetic foot infection, and diabetic Charcot deformity. Toenail infections and callosities were excluded. Conservative foot surgeries (ie, all types of nonamputation surgery) and primary minor toe amputations were accepted for inclusion. Standalone vascular or endovascular procedures were excluded, as were midfoot, ankle, and below- and above-knee amputations. Patients with a previous history of DFU or amputation were excluded. The severity of DFD was based on the presence of infection. The 2 groups were defined as the noninfected group (University of Texas stage A or Wagner grade 0 to 2) and the infected group (University of Texas stage B and D or Wagner grade 3 and 4). Patients with a history of gastrointestinal bleeding were excluded.

Data extraction and analysis

Demographic data, preoperative Hb level, comorbidities, and relevant variables were recorded in an Excel spreadsheet (Microsoft Corporation). Data analysis was performed using StatsDirect software (version 3, UK Cambridge). Means were compared using the t-test. Logistic regression was used to find correlation between continuous and categorical variables. Correlation between 2 continuous variables was computed via the Pearson correlation test.

Characteristics of the study sample

During the study period, 69 patients were admitted for diabetic foot surgery. After applying the inclusion and exclusion criteria, 37 patients who underwent nonvascular surgery for DFD were selected for analysis. Twenty-one patients had a noninfected DFU, and 16 patients had either an infected DFU or osteomyelitis. The mean age of the overall sample was 72 ± 12.1 years. Demographic data, comorbidities, and test results are summarized in the Table.

Prevalence of anemia and Hb level

Twenty-six patients (70.3%) had low levels of serum Hb. Of these, 13 (32.4% of the total sample) had severe anemia. The overall mean Hb level was 11.55 g/dL ± 2.3. Anemia (per the WHO sex-based definition) was less pronounced among males than in females (mean, 11.7 g/dL and 11.3 g/dL, respectively [r = 0.1, P =.46]). The mean Hb levels for the noninfected and infected groups were 11.7 g/dL ± 2.4 and 10.3 g/dL ± 2.1, respectively (t = 0.8, P =.033).

Correlations between anemia and DFU severity and other comorbidities

A correlation was found between Hb level and the severity of the DFU (noninfected vs. infected). Hemoglobin values were significantly lower in the infected vs noninfected groups (P =.03).

A high correlation was found between Hb and serum creatinine levels (P =.025). No correlation was found with hypertension, coronary artery disease, or peripheral artery disease (P <.05).

Previous studies have demonstrated that anemia is more prevalent among patients with diabetes than in healthy populations, and that anemia is even more common in patients with diabetic foot complications. In this latter subset of patients, higher rates of adverse outcomes such as nonhealing ulcers, amputation, and mortality were reported when anemia was present.^9,12,13 The present study investigated the relationship between anemia and DFD severity in the previously unexamined subset of patients undergoing surgical treatment for DFD. Patients in this surgical population presented on first admission with an advanced DFU for which debridement and standard care alone would have been insufficient to achieve an optimal outcome. The preliminary findings in the present study support previous study findings in that anemia was more prevalent and more pronounced in this high-risk population, and it was linked to the severity of the foot complication.

The results of this study were in line with the majority of reports on the treatment of both anemia and DFU. The prevalence of anemia in the surgical population in this study was similar to the weighted rate of 69.7% reported in the meta-analysis by Yammine et al.⁸ The mean Hb value in the current retrospective comparative study (11.55 g/dL ± 2.27) was slightly higher than that reported previously⁸ (11.00 g/dL ± 1.13). Contrary to Chuan et al,¹² the current study reported no significant differences between men and women.

Khanbhai et al¹³ reported an inverse correlation between Hb level and foot disease progression from neuropathy to ulceration (P <.0001). Gezawa et al¹⁴ reported a significant association between anemia and the presence of osteomyelitis (P =.006). Similarly, the present study documented an increased prevalence of anemia among patients with advanced foot disease, whether infected DFU or osteomyelitis. Although the mean Hb level was lower in the infected group, it did not reach statistical significance. However, a correlation was found between advanced foot disease and both variables, infection and anemia.

Many comorbidities have been associated with worse outcomes for patients with DFD, including diabetes duration, advanced diabetic neuropathy, chronic kidney disease, peripheral artery disease, and coronary artery disease. In this preliminary series, Hb level was correlated with creatinine level but not with other comorbidities. This supports the meta-analysis conducted by Yammine et al,⁸ in which an association between anemia and creatinine levels was reported. Whether anemia is a predictor of, or an independent risk factor of, DFU outcome is still a matter of debate, however.

Anemia can impede tissue perfusion in individuals with compromised peripheral vascular status, such as patients with diabetes. These patients have impaired microcirculation,⁶ reduced erythrocyte deformability,¹⁵ chronic kidney disease,¹⁶ and malnutrition¹⁷; all such factors could decrease limb perfusion and aggravate limb ischemia.

The results of the current study may encourage health care providers to consider the Hb level at the outset of treatment and attempt to increase the Hb level in patients with diabetes early in the treatment course, particularly in such patients with foot complications that may require surgical management. It may also be wise to assess anemia, as well as manage it, in patients with diabetes and impaired renal function. In this study, creatinine level was the only other variable that was correlated to Hb level.

In addition, it may be useful to conduct randomized studies with large sample sizes to test for anemia in this patient population and evaluate the effect of transfusions on clinical outcomes.

This study has limitations, mainly with respect to sample size and study design. The authors acknowledge these disadvantages and, with respect to the article title, clearly indicate that this research study is a preliminary retrospective report of a subset of patients with diabetes who were admitted for a form of conservative surgical treatment of DFD. The findings concur with previous closely related studies showing an association between DFD severity and anemia. Although these results are in line with previous reports, the current findings should be tested in larger controlled trials involving patients undergoing surgical care.

This preliminary retrospective comparative study showed that patients who undergo surgical treatment of diabetic foot complications on their first admission may likely have anemia. The preliminary results also demonstrated lower Hb levels in the patients in the infected DFU group compared to those in the noninfected DFU group. Because a correlation was also found between Hb and creatinine levels, it may be of interest to further study this correlation and suggest early transfusions in patients who present with infected ulcers and/or impaired renal function.