Reduction in Foot Ulcer Incidence: Relation to compliance with a prophylactic foot care program

Diabetic foot lesions are one of the most serious causes of morbidity among diabetic people and often require a long hospital stay. Despite intensive therapy, many of these patients will require a lower-extremity amputation(LEA), with a high economic and social cost. The estimated cost due to LEAs in diabetic patients in Spain was 5,289 million pesetas ($39.5 million U.S.)(1). Area 7 in Madrid has the lowest LEA incidence of all Caucasian populations(2), with a progressive decline since 1989, reaching a ≥50% decrease in 1997. Preventive strategies are mainly focused on early peripheral neuropathy detection, diagnosed by a neuropathy disability score (NDS) ≥6. Diabetic people diagnosed with neuropathy are then included in a continuous prophylactic foot care program(FP). This program has been available in our area since 1993. Our study assessed the effectiveness of the FP in reducing the foot ulcer incidence in diabetic patients included in this program between 1993 and 1996 and in the follow-up completed in December 1999.

We designed a prospective clinic-based study in which all diabetic patients who attended the outpatient clinic of the endocrinology service and who were diagnosed as having peripheral neuropathy based on an NDS ≥6 were included in a screening and prophylactic FP. The NDS is included in the standards of medical care for diabetic people from area 7 in Madrid (population of 565,000). The number of people with diagnosed diabetes is estimated to be∼ 19,000.

The design of the screening program has been described previously elsewhere(3). In short, we recruited diabetic patients with an NDS ≥6 who were considered to be suffering from peripheral neuropathy according to standard criteria(4). All of these diabetic patients were tested for peripheral vascular disease (PVD) and morphological plantar deformities; for visual and motor capacity that enabled them to inspect their own feet; and for self-foot care that included the manner in which they walked barefoot, correct performance of foot hygiene, callus care,nail trimming, water temperature checking, the use of heating pads and other methods to warm up the feet, “bathroom surgery,” the use of products for foot care, the method used for inspection of feet and shoes, and the proper use of shoes, socks, and clothes. Patients with severe PVD (grade 2 ischemia or higher), which was defined as the presence of intermittent claudication or the absence of at least one foot pulse (dorsalis pedis or posterior tibial), were excluded because these patients are unlikely to be influenced by FPs. The other patients were included in a continuous FP that consisted of four 90- to 120-min sessions held during 1 week. The first session was individual, and personal characteristics of the foot care of each patient were noted. The main goal of this session was to make the patient realize his/her lack of normal sensitivity in the feet compared with the rest of the body and his/her loss of sensory perception of pain.

Patients should agree that because of the decrease or loss in their protective pain sensation, it should be carefully evaluated. First, visual and motor capacity to inspect their own feet was explored. Patients were considered to have adequate eyesight capacity if they were able to read a letter 0.4 cm in size from 50 cm away (with their usual glasses if necessary)and acceptable physical mobility if they could see the soles of their feet,using a mirror if necessary. The patients were evaluated for the proper use of shoes, socks, and clothes; the manner in which they walked barefoot; the correct performance of foot hygiene; callus care and nail trimming; water temperature checking; the use of heating pads, hot water bottles, and other methods to warm up the feet; bathroom surgery; the use of products for foot care; the time and method used for the inspection of feet and shoes; as well as the use of podiatry services.

These data were collected in a questionnaire(5) adapted from one previously published (6). To enhance their knowledge and to change unsuitable behaviors, patients were included in groups of three to six people and instructed during three consecutive sessions over the course of 1 week. If a patient was unable to inspect his/her feet, one of the patient's relatives was then instructed on how to perform the task or at least monthly access to a chiropody service was provided. A continuous foot care treatment program was performed, including regular monthly visits during the first 6 months to evaluate the changes in the patient's behavior and diabetes treatment. In addition, patients could contact the clinic whenever they felt it necessary to evaluate some finding in their feet. Thereafter,reviews took place at least every 6 months. At least once a year a full clinical review took place, including a thorough neurological examination that followed the American Diabetes Association (ADA) recommendations(7).

Since 1998, a neurothesiometer (Arnold Horwell, London) has been available in our department, and a vibration perception threshold (VPT) was included in the annual review.

Patients were considered as having compliance with the FP if they met the following requirements: completion of the educational program, a change in inadequate foot care behavior during the first 6 months, regular visits to the podiatrist, at least one foot review every 6 months, and an annual medical care diabetic treatment review.

Between November 1993 and December 1996, 3,254 diabetic patients (17% of estimated diabetic population) visited our outpatient setting office. A total of 318 patients (9.2%) with neither past nor recent history of foot lesions were recruited from the screening and included in the study. Of these patients, 223 showed compliance with the FP (group A), whereas 95 patients did not complete the requirements and were placed in group B. Causes for noncompliance with the FP were as follows: 41 did not complete the educational program, 52 did not change their foot care behaviors, and 2 did not visit the podiatrist regularly. Groups A and B were comparable in terms of age, sex,duration of diabetes, and HbA_lc levels(Table 1). Because substantial reductions in morbidity can be obtained by instituting an adequate screening as simple as an NDS and enrolling patients in a basic FP, the existence of a control group lacking these requirements was considered ethically unacceptable. This letter compares patients with and without compliance to an FP, bearing in mind that the characteristics of both groups of patients are comparable. In addition, patients from group B received identical screening and an educational program at baseline and were followed in the same office setting, so we believe that this group received a better than conventional treatment. Patients were followed between 3 and 6 years (mean 4.6). Aχ² test was used to evaluate the differences in foot ulcer incidence within both groups.

During the follow-up, 13 patients died (8 men and 5 women, 69 ± 9 years of age [means ± SD]). Of these, 6 died as a result of heart ischemic disease, 3 died because of neoplasm disease (breast, colon, and prostate, respectively), and 3 died of unknown causes. The final patient, who died because of renal failure, had previously suffered an LEA.

In total, 37 foot ulcers were detected in 32 diabetic patients (68.6± 17 years of age), with 7 ulcers in 6 patients from group A and 30 ulcers in 26 patients from group B. Of these ulcers, 20 progressed to LEAs(73.2 ± 7.6 years of age), with 19 from group B and 1 from group A. The incidence of first ulcers and the annual distribution are shown in Table 1. The cumulative total foot ulcer rate was 11.6%, with 3.1% from group A and 31.6% from group B(P < 0.001). The number of first ulcers per 100 diabetic patients per year (95% CI) for the whole study population was 2.1 (1.9-2.3), with 0.5(0.4-0.6) from group A and 6.8 (5.4-8.2) from group B (P < 0.01). With regards to the VPT, the number of first ulcers per year of diabetic patients with VPT scores >25 V was 4.3% in group B, compared with 0.05% in group A (relative risk 8). The figures for diabetic patients with VPT scores<25 V were 4.4 vs. 0.2% for group B versus group A, respectively (relative risk 22).

Following international recommendations(7,8),educational programs must be offered to diabetic people experiencing a decrease in their protective pain sensation, even though the efficacy may be based on poor evidence(9,10). Our data show that diabetic people who complied with the FP had a 13-fold lower risk of suffering first foot ulcers than people who did not follow the program. This reduction may be higher in diabetic patients with less severe polyneuropathy (VPT <25 V). This finding shows, in a very indicative way,the importance of early intervention. Moreover, the program was designed to be widely applied in all of the ordinary medical office settings, even with limited financial resources. We compared the results from both group A and group B, considering group B as the control group. Thus, the control group(group B) had better than usual medical therapy, but without a continuous educational program. In addition, both groups were matched in relation to other risk factors, so that the difference in foot ulcer incidence might depend on compliance with the FP.

Our study clearly supports the hypothesis that when an FP is present, many patients change their inappropriate footcare behavior, with at least a 13-fold decrease in foot ulcer presentation. Some peculiar facts in our FP could explain our results. First, following ADA(7) recommendations,intervention was limited to high-risk people with previous neuropathy diagnosed by NDS. Second, continuous education was maintained until the changes in the patient's behavior were achieved, in contrast to other education programs with a rigidly scheduled session or a maximum of 3 months of surveillance. This FP integrates preventive educational and therapeutic strategies in a continuous way. Finally, free podiatry care was provided to all patients unable to perform their foot self-care. These three aspects,altogether, are critical in the program's efficacy.

There are several ways to diagnose a diabetic patient as having neuropathy(4). The NDS had been proven to be useful in epidemiological trials, using simple equipment that is available in any clinic(11,12). In addition, a recent study conducted in the primary care setting in the U.K.(13)—designed to search for risk factors in a large population sample—identified diabetes duration, amputation, and an NDS ≥6 as high-risk categories for foot ulceration, after stepwise multiple regression. Our study confirms that simple clinical measures of neuropathy, particularly NDS, are the best predictors of diabetic foot ulceration and could be applied in any medical office setting.

Moreover, patients with neuropathy that is diagnosed with these methods could have different risk categories. When the foot ulceration rate in people who complied with the program is compared with the rate in people who were noncompliant (according to VPT scores higher or lower than 25 V), diabetic people compliant with the FP and having a VPT score >25 V showed an eightfold lower risk for foot ulceration. This risk decreases even more(22-fold) in diabetic patients with less severe neuropathy (VPT scores <25 V). These results support the hypothesis that program efficacy could be greater when it is applied early, especially in people with less severe neuropathy.

In conclusion, our data suggest that a screening program based on the NDS is useful in detecting people with a high risk of foot ulcer development. Intervention programs based on continuous and well-organized education,supported by regular podiatry assistance, have demonstrated to be very effective in decreasing the incidence of foot ulceration. Considering how simple the program is, and that prevalence of clinical neuropathy in Spain is 22.7% (12), this program could be extended to any primary care clinic. Based on these findings, an FP in the primary setting is ongoing in Spain.