JSUR Journal

Introduction

Currently, the flaps used to cover complex wounds are subject to failure in the event of occlusion of the vascular pedicle. Fortunately, the flap can be saved if revascularization only briefly occurs [1,2]. For this reason, postoperative monitoring has been highlighted as a key parameter in early identification of these cases. Among these, the parameters that always stood out in the literature were a) color, b) temperature, c) turgor, and d) bleeding scarified edges. However, the time interval between the occurrence of occlusion and clinical signs may delay the diagnosis. In this context, several authors have introduced the evaluation of patchwork complex measures with hand-specific work and costly measures such as Doppler ultrasound of the vascular pedicle [3-5], measurement of tissue oxygenation, intra-arterial or intravenous catheters, microdialysis [6], changes in metabolism [7,8], and evaluation with probes for thermal diffusion [9]. However, another measure that is also studied, and is less complex, is blood serum glucose, the point of interest in this study.

Decreases in glucose serum levels and glycogen storage occur early in the flaps with a good prognosis [10], returning to normal by the seventh day after the procedure. In pathological cases, where there is formation of clots or thrombi, a decrease in flap sugar levels is more pronounced, more frequently observed in congestion situations (venous involvement), and also in ischemia or both [11-13]. Even with the use of anticoagulants, clot formation can occur [14]. Following this line of reasoning, glycemic measures can be used to monitor flaps and decrease rate in blood glucose values would be indicative of occlusive changes in the pedicle [15]. In the literature, there have been statements about flap monitoring using an absolute glycemic measure, which can be hard to rely on because the established relationship values are influenced by systemic glucose concentration, which can vary among individuals. With continuous monitoring, the high cost becomes unfeasible in most of the healthcare centers worldwide. For these reasons, the aim of this study was to establish a confidence limit interval to determine the extent of flap distress, taking into account the glycemic measures by measurement with glucometers available in almost all hospitals, which reduces costs and facilitates the specificity and dissemination of this evaluation, which is important in grafts.

Materials and Methods

This experiment was conducted in the laboratory of Microsurgery of the Hand and Upper Limb Department of Orthopedics and Traumatology of a Surgery reference center in Brazil, approved by the internal ethics committee (972,081,013). For the development of the project we used male Wistar rats of the isogenic strain SHR (N = 20, age = 2.5 months, body weight = 280– 300 g), provided by the Experimental Models Development Center for Medicine and Biology of the Federal University São Paulo (CEDEME–UNIFESP). During the study, the experimental animals were kept in a vivarium, with light/dark cycle (12 h:12 h), temperature of 21 ± 2°C, receiving water and standard rat chow ad libitum.

Anesthesia

The animals were anesthetized by intra-peritoneal injection with xylazine anesthetic solution composed by 1 U/100g and ketamine 1 U/100g.

They were divided into two homogeneous groups. They underwent surgery to establish an inguinal flap where the systemic blood glucose levels were measured (via flow) and at the flap edge.

Groups:

• Exposed: 10 SHRs underwent surgery for inguinal flap dissection, followed by occlusion of the pedicle vein.
• Controls: 10 SHRs underwent surgery for inguinal flap dissection, leaving an intact pedicle.

Surgical Technique

Once anesthetized, the animals were submitted to trichotomy of the abdominal region at the level of the knee, with the group assignment randomly chosen. The antisepsis in incised locations was achieved with 70% alcohol. The rat was placed supine and the legs were fixed to the table plane, with a tape. Based on the femoral artery, an inguinal flap, measuring 3 cm long and 2 cm wide, was drawn parallel to the midline, including the femoral artery inside. Then, the skin was incised, followed by blunt dissection of the planes on the medial side, exposing the femoral artery and its branch to the inguinal flap. After exposure of the pedicle (measuring approximately 2 cm with an outside diameter of the vessel of at most 2 mm), the flap was dissected from the cranial portion to flow, isolating the vascular pedicle. In the group in which the venous occlusion was performed, the vein was dissected at the emergence of the pedicle vessels and connected with Prolene 7-0. The surgical procedure was performed under 25x magnified view through a microscope.

Blood Glucose Measurement

Blood glucose level was measured in a drop of blood, both from the caudal vein of the animal (Figure 2B; through venipuncture needle) and at the flap edge (Figure 2A; through the inguinal flap edge fragment), with the aid of a specific Local device (Accu-Chek active; Roche Pharmaceutical Chemicals S/a) composed of sensitive strips for biochemical determination of glucose (Accu-Chek active glucotrend).

Samples were collected at different times, and the data recorded in a spreadsheet and displayed in graphs, as described below:

1. 0 minutes–before connecting the vein;
2. 30, 60, 90, and 120 minutes–after connecting the vein.

Statistical Analysis

Statistical analysis was performed using descriptive statistics such as mean, standard deviation, standard error (mean standard deviation), and correlation to the characterization of quantitative variables in the study population. In the tests for comparison of quantitative variables, we initially checked the normality by the Shapiro-Wilk test because the sample size was small. The Shapiro-Wilk test did not reject the hypothesis of normality (p > 0.05), indicating that the data (n = 20) are derived from a normal distribution, and therefore parametric tests can be used.

We used the paired t-test for comparisons between glucose measurements, and we used Student’s t test for comparisons between groups. A more refined analysis was performed using a linear models generalized test with repeated measures in order to check the influence over time, with a comparison of groups as well as checking group-time interactions. The results were expressed as mean, maximum, minimum, standard deviation, IC, risk factor, percentage, absolute values used for each test, and a p value < 0.05 was considered statistically significant.

Results

Experimental Data

Glycemic behavior of the groups was evaluated in three stages, first, as a function of time, comparing the groups (control vs. exposed) to measure the degree of flap distress, and a second analysis between glycemia collection sites (Local vs systemic) to observe the times at which they were correlated. In the final stages, the times that had a positive correlation were used to determine the range of glucose values of the Local test that might indicate the boundaries between healthy tissue and distress, expressed always as the percentage of the glycemic value of the commercial test in reference to the systemic blood glucose measurement.

Parametric Data Evaluation

To establish reliable indices in comparisons, we first established normality of the samples by using the Shapiro-Wilk normality test. Normality was not rejected for the variables investigated, with p > 0.05 (Table 1).

Glycemic Performance Evaluation of the Local Test

We observed that the glycemic index in the exposed group decreased over time, while in the control group it remained stable, characterized by a regular upward curve (Figure 1).

In the group of interest, this reduction was associated with disruption of blood flow as measured by the Local test, which caused a constant reduction in glycemic rates by exposing the tissue to hypoxia. Note that this variation was significant when comparing the exposed and control groups from 30 until 120 minutes (0, p = 0.985; 30 min, p = 0.01; 60 min, p = 0.002; 90 min, p <0.0001; 120 min, p = 0.001) (Table 2,3).

Evaluation of Systemic Glycemic Behavior

On comparing the groups for systemic measurement, there was no significant difference over time (0, p = 0.985; 30 min, p = 0.1; 60 min, p = 0.12; 90 min, p = 0.81; 120 min, p = 0.51) (Table 2,3 and Figure 2).

Correlation Between the Local and Systemic Tests

There was a positive correlation in the exposed group at three stages of evaluation, at time 30 minutes where a weak but positive correlation (r = 0.602) was observed, and at times 60 and 90 minutes, with a positive and strong correlation (r = 843 and r = 782, respectively) (Table 4,5) (Figures 3).

In the control group this correlation was positive only at 60 minutes (r = 664) and 90 minutes (r = 824) (Table 6 and 7) (Figures 3).

Local/Systemic Glucose Ratio

After determining the success of the experimental model, we evaluated the times when the glucose values of the Local test correlated with those of the systemic test, to establish the maximum rate, minimum, and average in groups according to the relationship of these two variables (Table 8, Figure 11). It was possible to determine, for this sample, below 50% (GR/GS×100) value on the retail test, indicating flap damage. Values above 60% indicated good quality of the flap, and those with values between 50 and 60% should be observed carefully.

Figure 1: Glucose flap per group versus time.

Figure 2: Systemic blood glucose level per group versus time.

Figure 3: Representation of the correlation between the Local and systemic test results in the exposed group and control, depending on the indicated time.

Figure 4: Flap length.

Figure 5: Flap width.

Figure 6. Flap and Pedicle dissection.    

Figure 7: Flap harvested isolated only for the pedicle.

Figure 8: Systemic blood glucose measument form the rat tail.

Figure 9: Flap blood glucose measument from the edge of the flap.

Figure 10: Flap pedicle interrupted.

Figure 11: Representation of minimum and maximum values of the groups, control and exposed, for the percentage of Local glycemic test results correlating with the systemic blood glucose test results.

Table 1: Representation of the normality test data.

Table 2: Comparison between groups (control and exposed) depending on the times studied.

Table 3: Representation of test values, confidence interval, and variations.

Table 4: Analysis of paired data between systemic blood glucose value and Local blood glucose test results of the exposed groups at the evaluated times.

Mean blood glucose values by time and group

Table 5: Analysis of the correlation between systemic value and Local test results of exposed groups at the evaluated times by Pearson index.

Table 6: Analysis of paired data between systemic blood glucose value and Local test results of the control group at the evaluated times.

Table 7: Correlation analysis by Pearson index between systemic and Local blood glucose of control groups at the evaluated times.

Table 8: Representative percentage of the glucose ratio of the Local test in relationship to the systemic test (GR/GS x100).