Validity of Diagnostic Codes and Estimation of Prevalence of Diabetic Foot Ulcers using a Large

Electronic Medical Record Database

Avivit Cahn1,2, Talya Altaras2, Tal Agami2, Ori Liran2, Colette E. Touaty2, Michel Drahy2, Rena Pollack1,2,
Itamar Raz1, Gabriel Chodick3,4, Inbar Zucker4’5

1
Diabetes Unit, Dept. of Endocrinology and Metabolism, Hadassah Medical Center, Hebrew
University of Jerusalem, The Faculty of Medicine, Jerusalem, Israel, 2Maccabi Healthcare Services,
Israel, 3Maccabi Institute for Research and Innovation, Maccabi Healthcare Services, Tel-Aviv, Israel,
4
School of Public Health, Sackler Faculty of Medicine, Tel Aviv University, Tel-Aviv, Israel, 5The Israel
Center for Disease Control, Ministry of Health, Ramat Gan, Israel

Running title: Validity of coding and prevalence of DFU

Word count: 3156

References: 15

Tables: 4 Figures: 1

Keywords: diabetic foot, prevalence, ulcer, electronic medical records

Corresponding author:

Dr. Avivit Cahn

Diabetes Unit, Endocrinology and Metabolism Unit

Hadassah Hebrew University Hospital, PO Box 12000

Jerusalem, Israel 91120

Tel: 97226776498

Fax: 97226437940

Email: [email protected]

This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/dmrr.3094

This article is protected by copyright. All rights reserved.

Abstract

Aims: To assess the validity of the diagnostic codes relating to diabetic foot ulcer (DFU) in the
electronic medical records of a large integrated care provider and to assess the prevalence of DFU
among its members.

Materials and Methods: Data were obtained from the diabetes registry of Maccabi Healthcare
Services (MHS), a 2.1-million-member sick-fund in Israel, which included 125,665 patients in 2015.
We randomly selected and reviewed ~400 patient files from each of the following categories during
study period: 1) had a diagnostic code of DFU; 2) had a diagnostic code, or clinical condition
suggestive of DFU including: leg-ulcer, amputation, DFU in quartiles proximate to 2015 or
abnormality reported by nurse; 3) patients at high risk for DFU (age >35 and one of the following:
peripheral artery disease, neuropathy, DFU during 2011-2014, eGFR<30 ml/min/m2 or foot
deformity). The patients' charts were reviewed by study physicians and DFU was validated or
refuted.

Results: Relying upon diagnostic codes entered by physicians, the positive-predictive-value (PPV)
was 73.1% (95% CI 67.6-78.2) and the sensitivity was 48.2% (95% CI 45.8-50.7%). The PPV of the
diagnostic codes listed by podiatrists was significantly lower while that of codes listed by nurses was
higher but with lower sensitivity. The estimated annual prevalence of DFU in the diabetes registry of
MHS was 1.2% (95%CI 1.0-1.5%).

Conclusions: Diagnostic codes alone cannot be used reliably to create a DFU registry. Nevertheless,
the data collected provide an estimate of the prevalence of DFU among patients included in the MHS
diabetes registry.

This article is protected by copyright. All rights reserved.

Introduction

Diabetic foot is defined as an infection, ulceration or destruction of tissues of the foot associated
with neuropathy and/or peripheral artery disease in the lower extremity of people with diabetes.
Diabetic foot ulcer (DFU) is defined as a full thickness lesion of the skin of the foot. (1). DFU is one of
the most devastating complications in patients with diabetes, and it is estimated that approximately
20% of moderate to severely infected DFU may lead to some level of amputation (2). The lifetime
risk of a patient with diabetes to develop a foot ulcer may be as high as 34% and the global
prevalence of diabetic foot ulcers (DFU) is estimated at 6.3% (2,3).

To the best of our knowledge, no data regarding the prevalence of DFU in Israel has been reported.
Increasingly, disease specific registries are being developed in regions and countries where
electronic medical records (EMR) are available and are an important means of monitoring
prevalence, incidence, trends and treatment outcomes (4). A national diabetes registry has been
created in Israel in 2014 by the Israel Center for Disease Control (ICDC) of the Ministry of Health (5).
One of the registry aims is to collect data regarding diabetes complications including diabetic foot
pathologies. A dedicated DFU registry may aid in understanding disease burden, monitoring the
quality of care and reducing inequities. However, ascertainment of DFU from the EMR may be
unreliable, since the positive predictive value (PPV) and sensitivity of diagnostic codes in use for DFU
are currently unknown.

The purpose of this study was to assess the validity of the diagnostic codes used to describe DFU in
the computerized database of Maccabi Healthcare Service (MHS), the second largest sick fund in
Israel which includes over 2 million individuals. Additionally, the data collected was used to form an
estimate of the prevalence of DFU among patients listed in the MHS diabetes registry.

Materials and Methods

Study population and selection of DFU diagnostic codes

Data were obtained from the diabetes registry of MHS, a 2.1-million-member sick fund in Israel,
which included 125,665 patients in 2015. We reviewed the MHS internal diagnostic codes (called ‘Y’
codes) entered during inpatient and outpatient encounters, aiming to define those which may be
used to establish a DFU registry. Key words including: diabetes, ulcer, foot, leg, peripheral arterial
disease, and neuropathy were used to identify the diagnostic codes relevant for the formation of the
registry. The Y-codes of “ulcer foot chronic”, “diabetic foot”, “ulcer diabetic lower limb” and
“diabetic neuropathic ulcers” were identified as frequently used relevant codes.

Patients who had a listing of any of the Y-codes as an active diagnosis during clinic visit by physician
or podiatrist during 2015 were considered as having a DFU during that year.

This article is protected by copyright. All rights reserved.

The International Classification of Diseases, Ninth Revision, with Clinical Modification (ICD-9) code of
250.7 is commonly used in hospital’s EMR to document peripheral arterial disease (PAD) as well as
DFU, and is therefore non-specific, and was not used to define DFU in our study.

The study was approved by the local Helsinki committee of MHS.

DFU validation

Patient files were reviewed by the study physicians who were general practitioners working within
MHS, familiar with its EMR system and trained in the study definitions of DFU. The principal
investigator of the trial was the referee in any case of diagnostic dilemma and randomly double-
checked the files for errors.

The positive predictive value (PPV) of the diagnostic codes of DFU was ascertained by a single
physician who reviewed the medical charts of a random sample of 400 patients in the diabetes
registry, for whom at least one of the DFU codes was noted during 2015 by a physician or a
podiatrist (figure 1).

The negative predictive value (NPV) of the absence of diagnostic codes was estimated by selecting
and reviewing the medical charts of 412 patients with diabetes with clinical conditions suggestive of
DFU or in whom codes suggestive of DFU were inputted such as leg ulcer, amputation, code 250.7,
DFU codes listed in the end of 2014 or beginning of 2016, or foot abnormality reported by a nurse.
Additionally, a random sample of 400 patients was chosen from a population at high risk for DFU.
Patients included in this category were of age over 35 years and had at least one additional risk
factor including advanced chronic renal failure (eGFR<30 mg/ml/m2), listing of diagnostic codes of
peripheral arterial disease, foot deformity or neuropathy during year 2015, or listing of DFU codes by
physicians in years 2011-2014.

Study physicians reviewed the entire patient chart including doctor, nurse, or podiatrist visits,
purchase of wound dressings and antibiotics. Inpatients encounters were reviewed as well. The
physician noted if the patient did or did not have DFU during year 2015. Leg ulcers (occurring above
the malleolus) during year 2015 were noted as well. If the diagnostic code of DFU was erroneously
used, the cause for the error was recorded.

For the comparison of the demographic, clinical, laboratory and outcomes data of DFU vs. leg ulcer
patients, data were extracted automatically from the EMR (and not manually by study physicians) –
to ensure uniformity.

This article is protected by copyright. All rights reserved.

Statistical analyses

Categorical variables were described by frequencies and percentages. Continuous and count
variables were described using mean (± standard deviation [SD]).

Clinical characteristics between patients with DFU vs. patients with leg ulcers were compared by chi-
square test or- t-test as appropriate All analyses were conducted using IBM-SPSS version 24. Validity
calculations were conducted using epi.R package in R (R Foundation for Statistical Computing,
Vienna, Austria).

Results

PPV of DFU

Within the diabetes registry of 2015 we identified 1562 patients who had diagnostic codes of DFU
listed during the year. The number of appearances of each code in all patients during the year was:
“ulcer foot chronic” – 257; “diabetic foot” – 4689; “ulcer diabetic lower limb” – 155; and “diabetic
neuropathic ulcers” – 31. A random sample of 400 patients was chosen from this group. The sample
did not differ from the group with respect to age, gender, region, diabetes duration, insulin use or
HbA1c (all p>0.05, data not shown). The medical files of these 400 patients were reviewed by the
study physician for explicit documentation of DFU, or for an alternative explanation for its listing. An
active DFU (true positive) was noted in the files of 237/400 = 59.3% (95% CI: 54.3%-64.1%) of
patients. False positive cases of DFU were most often due to use of the codes to describe a foot at-
risk (such as callus removal, nail cutting, foot deformity etc.), leg ulcers or a history of DFU which
was erroneously listed as an active diagnosis.

False positive rates were higher in those patients in which the diagnostic codes were listed by
podiatrists alone, since they were more likely to input DFU to describe the routine care of a patient
with a foot at-risk (table 1). In the 283 cases where DFU code was listed by a physician at least once
during 2015, DFU was validated in 207 yielding a PPV of 207/283 = 73.1% (95% CI: 67.7%-78.2%),
whereas the PPV of DFU codes listed by podiatrists was low - 30/117 = 25.6% (95% CI: 18.0%-34.5%).

According to our local guidelines, patients with diabetes should undergo annual or bi-annual foot
testing by the nursing staff. The nurses document foot abnormalities in the charts in coded fields,
and there are explicit fields for DFU. We studied whether including these listings may improve our
prediction. In our sample of 400 patients, nurses listed a DFU in 103 patients of whom 85 indeed had
an active DFU, leading to a PPV of 85/103=82.5% (95% CI: 73.8%-89.3%). However, complementing
the DFU codes with the nurses’ listings significantly lowered the sensitivity in our sample. Of the 237
patients who had a validated DFU, the nurses documented a DFU in only 85/237 = 35.9% (95% CI:
29.8%-42.3%).

We additionally studied whether available data regarding the purchase of advanced wound care
products may assist in identification of patients with DFU. In our sample of 400 patients, advanced
wound care products were purchased by 190 patients, and among the 237 patients who had a

This article is protected by copyright. All rights reserved.

validated DFU, only 148/237 = 62.4% (95% CI: 55.9%-68.6%) purchased advanced wound care
products. Therefore, the inclusion of this variable in the definition of a DFU led to a reduction in the
sensitivity of the diagnosis of DFU in our sample. Still, 148/190 = 77.9% (95% CI: 71.3%-83.6%) of the
patients who purchased advanced wound care products had a DFU and an additional 29/190 =
15.2% (95% CI: 10.5%-21.2%) had a leg ulcer, indicating a high PPV of this variable.

False negative rate of DFU

The rates of DFU in the different categories of high-risk patients who did not have DFU codes listed
during 2015 are shown in table 2. The diagnostic codes of leg ulcers were frequently used to
describe DFU, whereby 41% of the patients with diagnostic codes of leg ulcers, and no codes of DFU,
actually suffered of DFU (table 2). Additional high risk categories included patients in whom nurses
listed a DFU in their specific coded fields; patients in whom a physician or podiatrist had listed a
coding of DFU during the quartiles proximate to 2015 (i.e. last quartile of 2014 or first quartile of
2016); patients who had the coding of 250.7 (an ICD-9 code used for peripheral arterial disease and
diabetic foot) or the coding of amputation listed in 2015. Of note, the two latter categories included
very few patients, since most patients who had these codes, also had DFU codes listed during that
year, and had been included in the assessment of the true positive rates. The prevalence of DFU in
these categories was low ranging from 5-28%.

In addition, we reviewed a sample of 400 patients randomly selected from those who were at high
risk of DFU, yet not included in any of the previous categories. The prevalence of DFU in this high
risk population was low (1.25%). This indicates that the vast majority of patients with DFU had been
included in the previous categories – i.e. had diagnostic codes of DFU or codes suggestive of DFU.

Estimating the prevalence of DFU

To estimate the prevalence of DFU among patients who were listed in MHS diabetes registry during
year 2015 we multiplied the number of patients in each risk category by the prevalence of DFU
found in the sample. We included an additional 0.1% of the remaining individuals in the diabetes
registry (table 3). This resulted in an estimated number of 1,555 patients with DFU which comprised
1.2% (95% CI 1.0-1.5) of the patients listed in the diabetes registry at 2015.

Creating a registry of DFU based upon the diagnostic codes

Aiming to capture the majority of DFU patients (maximal sensitivity) with an acceptable compromise
to specificity, DFU was defined according to the DFU codes when listed by a physician. This yielded a
PPV of 754/1032=73.1% (95% CI 67.6-78.2%), and a sensitivity of 754/1555=48.2% (95% CI 45.8-
50.7%) (table 3). Thus, only about half of the patients with DFU could be identified by these criteria –
a sensitivity much lower than acceptable for a registry.

This article is protected by copyright. All rights reserved.

Clinical characteristics of patients with DFU vs. leg ulcers

Due to the frequent confusion observed in our study in the use of the diagnostic codes of leg ulcers
and foot ulcers, we compared the demographic and clinical parameters of patients with leg ulcers
and no DFU vs. patients with DFU and no leg ulcers – diagnoses established by review of clinical
notes (table 4). Patients with leg ulcers were more likely to have venous insufficiency and to be
taking anticoagulants, and less likely to have longstanding (>10 years) diabetes, require dialysis, take
insulin, or suffer from PAD, neuropathy or albuminuria (table 4).

Discussion

In our study we assessed the feasibility of creating a DFU registry based on the diagnostic codes of
DFU in the EMR of MHS. Basing the registry on the codes entered by physicians and verification of
the diagnosis by abstraction of clinical notes revealed low sensitivity (48.2%) and moderate PPV
(73.1%). Therefore, currently, the diagnostic codes cannot be reliably used alone to create a DFU
registry. Nevertheless, the data collected provides an estimate of the annual prevalence of DFU
among the patients listed in the diabetes registry of MHS.

Complementing the diagnostic codes with the nurses’ listings or the purchase of advanced wound
care products improved the PPV, yet at the cost of further reduction in the sensitivity. One of the
most common coding errors, performed by all health care professionals including physicians, nurses
and podiatrists was the erroneous listing of leg ulcers as DFU and vice versa. Leg ulcers and foot
ulcers, although often leading to the similar untoward outcome of amputation, are distinct clinical
entities with different epidemiology, pathophysiology and treatment, as was demonstrated in our
database and shown by others (9). DFU is a pathology stemming from neuropathy, often combined
with PAD, and is treated by multidisciplinary care including local wound care, offloading, metabolic
control, antibiotics and revascularization if needed (6,7,8). Leg ulcers are more likely to be secondary
to venous insufficiency and the mainstay of their treatment other than local wound care is
compression stockings (9). Although chronic leg ulcers are not uncommon in patients with diabetes,
due to their tendency to poor wound healing, and their increased rates of heart failure and
subsequent leg edema, they are a distinct entity and should not be included in a DFU registry.

The purpose of creating a DFU registry is to enable monitoring trends in DFU with respect to
different regions and levels of care. A registry may enable following the clinical course of persons
with DFU from diagnosis to outcome, aiming to improve quality of care and reduce inequities.
Furthermore, a validated registry may enable studying the impact of introducing new technologies,
or of reorganization of foot care services on complication rates. However, a registry with a low
sensitivity presents a significant barrier when seeking to meet these objectives since if the majority
of patients are not identified it is not possible to truly monitor the quality of care at a national level.
Due to the high rates of recurrence of DFU, it may be difficult to attain high specificity of the registry

This article is protected by copyright. All rights reserved.

since healed ulcers are often listed as prevalent cases. Nevertheless, the aim should be to strive for
the highest possible sensitivity, yet with the least compromise to specificity.

The results of this study indicate that existing clinical codes of DFU cannot be used to create a
registry without supporting evidence, and alternative means should be sought. Natural language
processing (NLP) of clinical notes was used to identify critical limb ischemia in the EMR of Mayo
Clinic (10). NLP demonstrated a significantly higher PPV, and a similarly high sensitivity vs. using ICD-
9 billing codes alone. NLP of our database may be considered as a means to develop a DFU registry,
and may assist in differentiating leg ulcers vs. foot ulcers and a history of DFU vs. prevalent DFU.
Alternatively, a prospective registry may be considered. A prospective registry will require education
of the health care personnel caring for patients with DFU, and possibly the inclusion of new codes in
the EMR which clearly designate DFU. The code “diabetic foot” has erroneously been used to
describe a “foot at-risk” and therefore a code which explicitly includes the word “ulcer” such as
“diabetic foot ulcer” may be a better choice. As we are in advanced stages of improving the infra-
structure of diabetic foot care in our country, education regarding the listing of DFU may be an
integral part of this reorganization.

The estimated prevalence of DFU in MHS is 1.2% (95% CI 1.0-1.5). This estimate is lower than
previously published in other studies due to the large denominator which is based on a
comprehensive diabetes registry, unlike other studies which included patients within diabetes clinics
or referral centers (3,8). Of note, the average age of persons included in our registry is 65.0±12.1
years, which is much higher than the age of persons with diabetes but no DFU in the other studies.
Elderly patients with diabetes are a heterogeneous group composed of patients with longstanding
diabetes as well as those with incident diabetes – the latter with better glycemic control and lower
complication rate (11). As a group, the elderly are characterized by better glycemic control, and
lower rates of insulin use, compared to younger patients (12). In Israel in 2016, 6.5% of patients age
≥65 had an HbA1c >9.0% compared to 15.0% of patients age <65 (13). We therefore hypothesize
that the inclusion of a large population of elderly, who are relatively well controlled, reduced the
prevalence of DFU. Furthermore, the MHS population is of higher socioeconomic status (14), and
better glycemic control than that observed countrywide (13). In our registry of year 2015, 8.1% of
the patients had a HbA1c >9.0% compared to the national rate of 11.0% (13). Therefore, the national
rates of DFU may be higher than our estimate for MHS. Finally, our study assessed the prevalence of
DFU and not of diabetic foot syndrome, as collected in another study (15), and thus did not include
patients with Charcot foot, or other foot pathologies without ulceration.

In conclusion, we are currently unable to rely upon the diagnostic codes listed in the EMR for
creating a DFU registry. However, the study provides a preliminary estimate of the prevalence of
DFU in MHS, which can be used for the assessment of disease burden and resource allocation.

Acknowledgements

Funding for this trial was provided by the D-CURE foundation (OTZMA initiative) represented by the
Medical Research and Development Fund for Health Services Jerusalem

This article is protected by copyright. All rights reserved.

Conflict of interests: The authors have no relevant conflicts to report

Prior Presentation: Parts of this study were presented at 78th Scientific Sessions of the American
Diabetes Association, Orlando, FL, 22-26 June 2018

References

1. Definitions and criteria, 2015. Available at: http://iwgdf.org/guidelines/definitions-criteria-

2015/ Last accessed 20.9.2018
2. Armstrong DG, Boulton AJM, Bus SA. Diabetic Foot Ulcers and Their Recurrence. N Engl J
Med 2017;376(24):2367–75.
3. Zhang P, Lu J, Jing Y, Tang S, Zhu D, Bi Y. Global epidemiology of diabetic foot ulceration: a
systematic review and meta-analysis. Ann Med. 2017;49(2):106–16.
4. Shalev V, Chodick G, Goren I, Silber H, Kokia E, Heymann AD. The use of an automated
patient registry to manage and monitor cardiovascular conditions and related outcomes in a
large health organization. Int J Cardiol. 2011 Nov 3;152(3):345-9
5. https://www.health.gov.il/UnitsOffice/ICDC/disease_Registries/Pages/diabetes_reg.aspx
6. Schaper NC, Van Netten JJ, Apelqvist J, Lipsky BA, Bakker K; International Working Group on
the Diabetic Foot (IWGDF). Prevention and management of foot problems in diabetes: A
Summary Guidance for Daily Practice 2015, based on the IWGDF guidance documents.
Diabetes Res Clin Pract. 2017 Feb;124:84-92.
7. Jeffcoate WJ, Vileikyte L, Boyko EJ, Armstrong DG, Boulton AJM. Current Challenges and
Opportunities in the Prevention and Management of Diabetic Foot Ulcers. Diabetes Care.
2018 Apr;41(4):645-652
8. Cahn A, Elishuv O, Olshtain-Pops K. Establishing a multidisciplinary diabetic foot team in a
large tertiary hospital: a workshop. Diabetes Metab Res Rev. 2014 Jul;30(5):350-3
9. Heyer K, Protz K, Glaeske G, Augustin M. Epidemiology and use of compression treatment in
venous leg ulcers: nationwide claims data analysis in Germany. Int Wound J. 2017
Apr;14(2):338-343
10. Afzal N, Mallipeddi VP, Sohn S, Liu H, Chaudhry R, Scott CG, Kullo IJ, Arruda-Olson AM.
Natural language processing of clinical notes for identification of critical limb ischemia. Int J
Med Inform. 2018 Mar;111:83-89.
11. Lee BK, Kim SW, Choi D, Cho EH. Comparison of Age of Onset and Frequency of Diabetic
Complications in the Very Elderly Patients with Type 2 Diabetes. Endocrinol Metab (Seoul).
2016 Sep;31(3):416-423
12. American Diabetes Association. Economic Costs of Diabetes in the U.S. in 2017. Diabetes
Care. 2018 May;41(5):917-928.
13. National program for quality indicators in community healthcare, 2014-2016 report.
Available on: http://israelhpr.org.il/. Accessed on July 28, 2018
14. National insurance survey on membership in health plans 2016, published on August 2017.
Available on:
https://www.btl.gov.il/Publications/survey/Documents/seker289/seker_289.pdf, table 31
page 64. Accessed on July 28, 2018

This article is protected by copyright. All rights reserved.

15. Bohn B, Grünerbel A, Altmeier M, Giesche C, Pfeifer M, Wagner C, Heise N, Best F, Fasching
P, Holl RW. Diabetic foot syndrome in patients with diabetes. A multicenter
German/Austrian DPV analysis on 33 870 patients. Diabetes Metab Res Rev. 2018 May
3:e3020.

This article is protected by copyright. All rights reserved.

Table 1 – DFU rates in the sample

Code inputted Code inputted Total

by physician by podiatrist (N=400)
(N=283) (N=117)
Foot ulcer 207 (73.1%) 30 (25.6%) 237 (59.2%)
Clinical diagnosis
Leg ulcer and no foot ulcer 27 (9.5%) 7 (6.0%) 34 (8.5%)
per chart review
No leg or foot ulcer 49 (17.3%) 80 (68.4%) 129 (32.3%)
Rates of DFU in a sample of 400 patient files. The clinical diagnoses were validated or refuted by
abstraction of the clinical notes by study physicians.

Table 2: Prevalence of DFU in high risk patients

No. of Sample Prevalence Estimated number of

patients size* of DFU in patients with DFU in
fulfilling sample population (95% CI)**
criteria during
2015
Diagnostic code of leg ulcer 532 100 41 (41%) 218 (170-270)
DFU per nurses chart 595 100 28 (28%) 167 (118-227)
DFU codes in Q4-2014 or Q1- 337 100 5 (5%) 17 (6-40)
2016
Diagnostic code of 250.7 52 52 14 (28%) 14
Diagnostic code of 118 60 12 (20%) 24 (13-38)
amputation in 2015
High risk patients† 8918 400 5 (1.25%) 111 (45-258)
All categories are mutually exclusive, whereby each row does not include patients in the previous
category. *Sample size – number of files reviewed by study physician – randomly selected from each
category. ** Number of patients with DFU in each category is calculated by multiplying prevalence in
sample by size of group. †As described in text.

This article is protected by copyright. All rights reserved.

Table 3 – Estimating the number of patients with DFU in 2015

Number of patients in group Estimated number with DF (95%

CI)
DFU code by physician 1,032 754 (698-807)
DFU code by podiatrist 530 136 (95-183)
High risk patients by categories 10,552 551 (352-833)
(per table 2)
Remaining patients in diabetes 113,551 114
registry (estimated 0.1% risk)
Total 125,665 1,555 (1,250-1,940)
An estimation of the number of patients with DFU calculated by multiplying the prevalence of DFU in
each category by the number of persons in that category in year 2015.

This article is protected by copyright. All rights reserved.

Table 4: Characteristics of patients with DFU vs. leg ulcer

DFU Leg ulcer P

N 315 92
Demographics Age (years ± SD) 68.2 ± 12.6 70.2 ± 10.6 0.15
Male gender, n (%) 198 (62.9%) 60 (65.2%) 0.68
Diabetes duration>10 years, n(%) 228 (72.4%) 53 (57.6%) <0.01
Medications Insulin Yes, n(%) 186 (59.0%) 42 (45.7%) 0.02
Anticoagulants Yes, n(%) 39 (12.4%) 30 (32.6%) <0.01
Laboratory HbA1c (%± SD) 8.1 ± 1.9 7.8 ±1.7 0.20
value GFR >60 196 (62.2%) 54 (58.7%) 0.16
(ml/min/1.73m2) 30-60 68 (21.6) 27 (29.3%)
<30 36 (11.4%) 6 (6.5%)
No data 15 (4.8%) 5 (5.4%)
ACR >300 54 (17.1%) 12 (13%) 0.04
30-300 78 (24.8%) 18 (19.6%)
<30 99 (31.4%) 44 (47.8%)
No data 84 (26.7%) 18 (19.6%)
Listed in IHD, n (%) 139 (44.1%) 33 (35.9%) 0.16
Registry CHF, n (%) 66 (21.0%) 19 (20.7%) 0.98
Dialysis, n (%) 18 (5.7%) 0 (0%) <0.01
Comorbidities Retinopathy 112 (35.6%) 26 (28.3%) 0.19
Neuropathy 132 (41.9%) 28 (30.4%) 0.05
Peripheral arterial disease 173 (54.9%) 35 (38.0%) <0.01
Venous insufficiency 94 (29.8%) 53 (57.6%) <0.01
Amputation before 2015 44 (14.0%) 9 (9.8%) 0.29
Outcomes Amputation during 1/2015-1/2018 78 (24.8%) 15 (16.3%) 0.09
Death till 1/2018 90 (28.6%) 22 (23.9%) 0.38
Diagnoses were verified by file review. Patients having both DFU and leg ulcer in 2015 were
excluded. P-values did not include the “no data” category. ACR – Albumin creatinine ratio; CHF –
Congestive heart failure; DFU – Diabetic foot ulcer; IHD – Ischemic heart disease; GFR – Glomerular
filtration rate.

This article is protected by copyright. All rights reserved.

Figure 1: Study design

This article is protected by copyright. All rights reserved.