The inclusion and exclusion criteria

The inclusion criteria were: (1) type 2 DM with Wagner grade 1–3 foot ulcers, (2) age 30–80 years, (3) Chinese Han from Hubei province. The exclusion criteria were: (1) patients with Wagner grade 4 or 5 DFUs, (2) bleeding diatheses, (3) malignant hypertension, (4) hepatic or renal failure, (5) mental illness, (6) cancer, (7) current pregnancy, (8) diabetic ketoacidosis or diabetic lactic acidosis, and (9) neuropathic ulcer. Diagnosis criteria: (1) DM was defined in 1999 WHO standard, (2) the DFU was classified by Wagner scale,15 (3) cardiovascular events included coronary heart disease events (non-fatal myocardial infarction, angina), total stroke, cardiovascular death, coronary and carotid revascularization,16 17 and (4) cardiovascular death was deemed when a fatal event occurred after myocardial infarction, ventricular fibrillation, sudden death, or heart failure. All other deaths were classified as non-cardiovascular.18 19 All myocardial infarctions, strokes, and deaths—that is, all cardiovascular disease events and all other deaths were adjudicated by an end point assessment committee unaware of treatment allocation, (5) cigarette smoker was defined as subjects who had smoked at least one cigarette daily for 1 year,19 (6) malignant hypertension was defined as per the American College of Cardiology/American Heart Association guidelines, (7) albuminuria was defined as per the American Diabetes Association criteria,20 and (8) major amputation was defined as any above-the-ankle amputation.18

Study subjects and design

A total of 195 patients with DFU were randomly divided into either continuous intrafemoral thrombolysis group (98 patients) or conventional therapy group (97 patients) between May 2010 and December 2015. Characteristics of the patients are shown in table 1.

This trial was registered in ClinicalTrial.gov (NCT01108120), and was a randomized, parallel group trial of investigations with or without continuous intrafemoral thrombolysis. The continuous intrafemoral thrombolysis group received continuous intrafemoral urokinase injection for 7 days, and conventional therapy group just received wound debridement and dressing change. Then, a follow-up of average 6.5 years was performed (figure 1). The actual time of completion of the study was January 2020.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Trial profile.

Study objective

The primary end point of this study was to compare the percentage of completed ulcers closure between two groups at the first month and the terminal of the follow-up study. The secondary end point was the time of ulcer healing, incidence of ulcer recurrence, incidence of amputation, the cardiovascular event and the cardiovascular death during the follow-up period.

Study procedures

Eligible patients were numbered according to their admission order. The numbers were arranged in odd or even groups, with the odd number as continuous intrafemoral thrombolysis group and the even number as conventional therapy group. After diabetic dietary advices, all patients received insulin therapy to control blood glucose (BG) within a range of 5–10 mmol/L during first month intervention period as much as possible. Then the patients received conventional care for their ulcers such as wound debridement including remove extensive callus and necrotic tissue, and broad spectrum antibiotics when ulcers showed clinical signs of infection. Adjustments to the treatment are performed when indicated on the basis of microbiologic cultures and sensitivity testing. All participants received a regular dressing change with sterile gauze 2–3 times per week to the end of the study if needed.

Patients were examined weekly after the terminal of the intrafemoral urokinase injection assay. Then the patients were followed monthly for the first year and followed trimonthly to January 2020. At each study visit, ulcers were assessed for area via wound tracing, ulcer closure, or adequate granulation tissue formation. The ulcer area was determined in square millimeters by multiplying the largest width and length of the ulcer. The largest ulcer was considered as the study ulcer when more than one ulcer was present. The measurement was performed after revision of the ulcer.21 During the follow-up period, the treated target of hemoglobin A1c (HbA1c) was <7.0%, and the therapy medicines and strategy were determined by the responsible physicians based on Chinese Diabetes Society guidelines 2010.

The continuous intrafemoral urokinase injection approach

Based on our previous report in Chinese in 2007 (Guangda Xiang, Chinese Journal of Diabetes, 2007, 15(11): 649–650), a continuous intrafemoral urokinase injection approach was performed for patients in continuous intrafemoral thrombolysis group. In brief, an ultrasound Doppler examination of vessels including artery and venous of lower limbs was performed. To avoid pulmonary infarction, a filtrator was placed in the inferior vena cava before the thrombolysis process if ultrasound results showed venous thrombosis. Next, a percutaneous artery canal was inserted from femoral at the base of thigh into the distal of popliteal artery in affected lower limb as far as possible from femoral artery. After finishing this process, the outside part of this artery catheter is fixed at thigh (online supplemental figure 1), and the patients must keep in supine position in the bed for 7 days. When the above operation was completed, 200 000–400 000 units urokinase was injected via the catheter to diseased foot. Afterward, the continuous infusion urokinase (100 mL 0.9% sodium chloride+1 000 000 unit urokinase at a rate of 4 mL/hour) was administered with an artery pump via femoral artery for 7 days. Considering that catheter-associated infection risk increased after 7 days,22 as well as the effective role of the duration of catheterization was 5–9 days in our previous study (Guangda Xiang, Chinese Journal of Diabetes, 2007, 15(11): 649–650), thus, we chose 7 days as the course of treatment. In addition, the safety outcomes such as infections and bleeding or oozing blood for continuous intrafemoral urokinase injection approach were observed.

Definition of ulcer outcome

Three professional physicians assessed ulcer healing and they were blinded to the intervention. Ulcer outcome was defined as: (1) complete ulcer closure was defined as skin closure (100% re-epithelialization) without drainage or dressing requirements, (2) improved ulcer was defined as 50% reduction of the ulcer area but not complete ulcer closure, (3) unchanged ulcer was defined as decreased or increased ulcer area <50%, (4) impaired ulcer area was defined as increased ulcer area ≥50% or amputation above or below the ankle.

The determination of toe brachial index and ankle-brachial index

All subjects were asked to refrain from coffee and smoking for at least 8 hours before the examinations and room temperature was kept between 22℃ and 24℃. Participants were kept in a supine position with their feet at heart level after 20 min of acclimatization. The blood pressure was determined sphygmomanometrically in the right arm. Then, the cuff was placed around the ankle immediately above the malleolus and the ankle pressure was measured with a Doppler signal followed by release sphygmomanometrically on the dorsal right foot.23 The ankle-brachial index (ABI) was calculated by the ratio of the ankle systolic pressure to the brachial systolic pressure.24

The measurement of toe pressure was followed, a 25 mm digital cuff was placed around the proximal phalanx of the hallux and the Doppler signal was measured from the distal pad of the great toe.25 The cuff was inflated to a maximum of 200 mm Hg and then slowly deflated. The point at which the arterial waveform reappeared was defined as the toe systolic blood pressure. The toe brachial index (TBI) was calculated as the toe systolic pressure divided by the brachial systolic pressure.24

The determination of peripheral neuropathy

Peripheral neuropathy was based on loss of sensation in peripheral neuropathy including vibration, proprioception, temperature, and pin-prick sensations.26 All patients were evaluated by vibration sensation using a 128 Hz tuning fork27 and pin-prick sensation using Neuro tip and temperature sensation using warm and cool rods and Achilles tendon reflex using a tendon hammer. Then, we used the Neuropathy Deficit Score (NDS) to evaluate peripheral neuropathy. The perceptions were scored 0 if present and normal, and 1 if absent, reduced, or uncertain. On either side, the ankle reflex was scored 0 if present and normal, and 2 if absent. The maximum NDS score was 10. The severity of neuropathy was graded as follows: mild (NDS scores: 3–5), moderate (NDS scores: 6–8), and severe (NDS scores: 9–10). We defined peripheral neuropathy as at least moderate NDS scores with or without symptoms (NDS ≥6).

The measurements of transcutaneous oxygen tension

Local skin oxygenation was evaluated by measurement of transcutaneous oxygen tension (TcPO2).28 TcPO2 is a non-invasive method reflecting local arterial skin blood flow and oxygenation, and it can be used as a predictor of ulcer heal.29 TcPO2 was measured by an electrochemical transducer and the measurements of TcPO2 were taken at the dorsum of the foot in the first intermetatarsal space (PF5001, Pehimed, Sweden). A calibration period of method was 10–15 min, the TcPO2 signal was continuously recorded, and has good reproducibility.30

Biochemical measurements

Blood samples were obtained from all patients after a 12-hour fast. Serum total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), aspartate transaminase (AST) and alanine transaminase (ALT) were measured enzymatically. HbA1c was measured by liquid chromatography (G8-90SL, Tosoh, Japan). Fasting blood glucose (FBG) and postprandial 2-hour glucose (after 75 g glucose loading, 2-hour BG) was measured by a glucose oxidase procedure (SYNCHRON Biochemical Analysis System, Suzhou, China). Creatinine was measured enzymatically. Total plasma fibrinogen was analyzed with a syneresis method,31 which measures the amount of clottable protein incorporated into the clot, after addition of thrombin to the plasma sample. As our previously described,32 urinary albumin was examined with nephelometry (N antiserum to human albumin kit; BN prospec, Siemens, Germany) and urinary creatinine (Creatinine Plus V.2 kit; Cobas c701, Roche, Germany) according to the manufacturer’s instructions. Coefficients of variation for these assays were 1%–2% (HbA1c, TC, HDL-C, creatinine), 2%–3% (BG, TG, LDL-C, UACR), and 3%–6% (fibrinogen, AST, ALT).

Health-related quality of life

At the end of follow-up study, we randomly selected 20 participants in both groups to complete a validated health-related quality of life (HRQoL) questionnaire that is specific to patients with DFU, the DFS-SF.33 The DFS-SF tool contained 29 items consisting of 6 subscales that were related to leisure, physical health, dependence or daily life, negative emotions, being worried about ulcers or feet, and being bothered by ulcer care. Each domain had a score ranging from 0 to 100, with higher scores implying better quality of life. Data generated were used to evaluate the effect of the interventions on HRQoL.

Sample size calculation and statistical analysis

We were conducting a pilot study with no prior data of the effects of continuous intrafemoral thrombolysis on DFU. Based on the previous study, the ulcer healing rate was 45% with a regular dressing change. In this study, we want to achieve a target of 70% ulcer healing rate. Therefore, we performed a power calculation based on the ulcer healing rate. Based on an expected difference in ulcer healing rate of 25%, with a power of 0.90 and an α of 0.05 (two-sided), we calculated that 80 subjects would be needed. Considering that patients drop out in long-term follow-up, we enrolled 100 participants in each group.

Continuous normally distributed data are reported as mean±SD. The Student’s t-test was used to evaluate differences in continuous variables between groups. Contingency tables were analyzed using the χ² test. Kaplan-Meier methodology with log-rank test was used to evaluate the time to complete ulcer closure and cumulative incidence of cardiovascular events and cardiovascular mortality between groups.

HRs and 95% CIs were determined by univariate and multivariate Cox proportional hazards regression analyses for end point of ulcer closure and cardiovascular events. Initial univariate analyses identified demographic and clinical variables that predicted ulcer closure and cardiovascular events. Significant variables were entered into subsequent multivariate Cox regression with the variables included in the models using forward stepwise selection, with a probability value of <0.05 required for entry into the model. A probability value of 0.05 or less was considered significant. All statistical analyses were performed using SPSS statistics V.18 software.