Interobserver and Intraobserver Reproducibility of Plain X-Rays in the Diagnosis of Diabetic Foot Osteomyelitis

Prévia do material em texto

http://ijl.sagepub.com/
Wounds
The International Journal of Lower Extremity
 http://ijl.sagepub.com/content/12/1/12
The online version of this article can be found at:
 
DOI: 10.1177/1534734612474304
 2013 12: 12 originally published online 1 February 2013International Journal of Lower Extremity Wounds
Cecilia-Matilla and Juan V. Beneit-Montesinos
Francisco J. Álvaro-Afonso, Jose L. Lázaro-Martínez, Javier Aragón-Sánchez, Esther García-Morales, Almudena
Osteomyelitis
Interobserver and Intraobserver Reproducibility of Plain X-Rays in the Diagnosis of Diabetic Foot
 
 
Published by:
 http://www.sagepublications.com
 can be found at:The International Journal of Lower Extremity WoundsAdditional services and information for 
 
 
 
 
 http://ijl.sagepub.com/cgi/alertsEmail Alerts: 
 
 http://ijl.sagepub.com/subscriptionsSubscriptions: 
 http://www.sagepub.com/journalsReprints.navReprints: 
 
 http://www.sagepub.com/journalsPermissions.navPermissions: 
 
 What is This?
 
- Feb 1, 2013OnlineFirst Version of Record 
 
- Mar 15, 2013Version of Record >> 
 by guest on November 4, 2014ijl.sagepub.comDownloaded from by guest on November 4, 2014ijl.sagepub.comDownloaded from 
The International Journal of Lower 
Extremity Wounds
12(1) 12 –15
© The Author(s) 2013
Reprints and permission: 
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1534734612474304
ijl.sagepub.com
Osteomyelitis is an inflammatory process of infectious 
nature that affects the bone marrow and the surrounding 
cortical bone and is one of the most common complications 
in diabetic foot ulcers.1 Unlike hematogenous osteomyeli-
tis, diabetic foot osteomyelitis often occurs contiguously as 
a result of deep penetration of an infection of soft tissue 
nearby.2 The prevalence of osteomyelitis in diabetic foot 
ulcers ranges from 20% to 70.4% depending on the popula-
tion studied, the techniques used for diagnosis, and the cen-
ter’s level of expertise in treating the diabetic foot.3-7
The Infectious Diseases Society of America (IDSA) 
clinical practice guideline for the diagnosis and treatment of 
diabetic foot infections8 includes standardized tests such as 
the probe-to-bone test and plain X-rays. Both tests are 
noninvasive, are not very expensive, and are accepted 
from the point of view of early diagnosis used in special-
ized units. According to a recent study,7 the combination 
of these 2 diagnostic tests has a sensitivity and specificity 
similar to other more expensive diagnostic tests such as 
magnetic resonance. Another study reports that bone changes 
seen on radiographs have no prognostic value when diabetic 
foot osteomyelitis is treated surgically and that they are more 
related to the prognosis of the soft tissue affected.9
It was recently shown that the probe-to-bone test is an 
operator-dependent test in the diagnosis of diabetic foot osteo-
myelitis,10 but there is currently no study on the diagnostic vari-
ability of plain radiography among clinicians with different or 
equal experience in the management of the diabetic foot.
The aims of this study were to analyze the interobserver 
and intraobserver variability of plain radiography in the 
diagnosis of diabetic foot osteomyelitis depending on the 
training of the professional involved.
Materials and Methods
We conducted an observational study between October 2009 
and July 2011, which included 123 patients with diabetic 
1Complutense University of Madrid, Madrid, Spain
2La Paloma Hospital, Las Palmas de Gran Canarias, Spain
Corresponding Author:
Francisco J. Álvaro-Afonso, Unidad de Pie Diabético, Clínica 
Universitaria de Podología, Edificio Facultad de Medicina, Pabellón I, 
Universidad Complutense de Madrid, Avda Complutense s/n, 28040 
Madrid, Spain.
Email: fraalv@hotmail.com
Interobserver and Intraobserver 
Reproducibility of Plain X-Rays in the 
Diagnosis of Diabetic Foot Osteomyelitis
Francisco J. Álvaro-Afonso, DPM, PhD1, Jose L. Lázaro-Martínez, DPM, PhD1, 
Javier Aragón-Sánchez, MD, PhD2, Esther García-Morales, DPM, PhD1, 
Almudena Cecilia-Matilla, DPM, PhD1, and Juan V. Beneit-Montesinos MD, PhD1
Abstract
The purpose of this study was to analyze the interobserver and intraobserver variability in plain radiography in the diagnosis 
of diabetic foot osteomyelitis. A prospective observational study was performed from October 1, 2009, to July 31, 2011, 
on patients with diabetic foot ulcers, with clinically suspected osteomyelitis who were admitted to the Diabetic Foot Unit 
of the Complutense University of Madrid. Two professional groups examined 123 plain X-rays, each group comprising 3 
different levels of clinical experience. To analyze intraobserver variability, 2 months later plain X-rays were reanalyzed by 
one of the clinical groups. When using only plain radiography for the diagnosis of osteomyelitis in the diabetic foot, low 
concordance rates were observed for clinicians with a similar level of experience: experienced clinicians (K
11AB
 = .35, 
P < .001), moderately experienced clinicians (K
22AB
 = .39, P < .001), and inexperienced clinicians (K
33AB
 = .40, P < .001). 
Intraobserver agreement was highest in experienced clinicians (K
11A
 = .75, P < .001), followed by moderately experienced 
clinicians (K
22A
 = .61, P < .001) and inexperienced clinicians (K
33A
 = .57, P < .001). Plain radiography for the diagnosis of 
diabetic foot osteomyelitis is operator dependent and shows low association strength, even among experienced clinicians, 
when interpreted in isolation without knowing the clinical characteristics of the lesion.
Keywords
diabetic foot, diabetic foot infections, foot ulcer, diabetic foot osteomyelitis, plain X-ray
474304 IJL12110.1177/1534734612474304The International Journal of Lower Extremity WoundsÁlvaro-Afonso et al
© The Author(s) 2013
Reprints and permission: http://www.
sagepub.com/journalsPermissions.nav
Translational Research
 by guest on November 4, 2014ijl.sagepub.comDownloaded from 
Álvaro-Afonso et al 13
foot ulcers, with clinical suspicion of osteomyelitis based 
on the fulfillment of at least one of the following criteria: 
presence of 2 or more signs of inflammation such as pain, 
redness, heat, loss of function, foul odor, lymphangitis or 
crepitus, besides the presence of suppuration or even bone 
fragments11; swollen, erythematous toe with loss of normal 
contours or “sausage toe”12; and ulcers with discharge that 
did not improve with appropriate treatment for a period of 
at least 6 weeks.13 All patients with diabetes who underwent 
foot surgery in the 3 preceding months and patients with 
diabetic foot ulcers associated with Charcot foot were 
excluded.
All patients underwent plain radiography of the foot (dor-
sal-plantar, lateral, medial oblique, lateral oblique, or both) 
based on the location of the ulcer under study. The presence 
of at least one of the following radiographic findings was 
considered radiographic evidence of osteomyelitis14: abnor-
mal cortical bone, periosteal elevation, affected bone mar-
row, osteolysis, and sequestrations (segments of necrotic 
bone separated from healthy bone by granulation tissue).
The radiographs were interpreted in isolation and individu-
ally by 2 groups of professionals (group A and group B), each 
composed of 3 clinicians with different levels of experience:
 • Inexperienced: Professional without experience in 
the treatment of the diabetic foot
 • Moderately experienced: Professional whose 
experience ranges from 6 to 12 months in the treat-
ment of the diabetic foot
 • Very experienced: Professional with several years 
of experience in the treatment of the diabetic foot
Selection of clinicians with different expertise levels was 
made by alternate allocation within agroup of candidates 
available in the Diabetic Foot Unit.
The clinicians were aware of the location of the lesion but 
not of its clinical characteristics. Radiographs were interpreted 
in 10 series consisting of 12 radiographs each, except for series 
number 10, which consisted of 15 radiographs. Each clinician 
interpreted a total of 123 radiographs, each radiograph belong-
ing to different patients. An interpretation regimen was estab-
lished that consisted of 1 series per clinician per day.
To analyze intraobserver variability, 2 months later the 
radiographs were interpreted again by each of the 3 clini-
cians of group A, following the same protocol (Figure 1).
Statistical Analysis
Data were collected confidentially by the principal investi-
gator and analyzed using the software package SPSS v15.0 
for Windows. Qualitative variables were described using 
frequency distributions and percentages, and quantitative 
Figure 1. Sequence of clinicians who interpreted the plain X-rays.
 by guest on November 4, 2014ijl.sagepub.comDownloaded from 
14 The International Journal of Lower Extremity Wounds 12(1)
variables were described by their mean, standard deviation, 
maximum, and minimum. Analysis of agreement was per-
formed by Kappa statistics, using the Landis and Koch 
criteria15 to study the strength of association. Differences 
were assumed significant at P < .05 for a confidence inter-
val of 95%.
We conducted the study in accordance with the 
Declaration of Helsinki (2000 revision).16 This prospective 
study did not require approval by the ethics committee, since 
the procedures did not alter or exceed the scope of our stan-
dard medical care.
Results
A total of 123 patients were included in this study, of whom 
88 (72%) were men and 35 (28%) were women. The mean 
age of the patients was 65 ± 13.3 years, with a mean duration 
of diabetes of 16.0 ± 12.2 years. In our study population, 14 
patients (11%) had type 1 and 109 patients (89%) had type 2 
diabetes mellitus. Mean glycated hemoglobin was 52 ± 3 
mmol/mol (6.9 ± 1.9%), and the mean glucose level was 141 
± 46.3 mg/dL. Mean wound duration was 35.4 ± 95.3 weeks.
Table 1 summarizes the agreement indices obtained for 
the 2 groups of clinicians.
Discussion
Diagnosis of osteomyelitis of the diabetic foot is one of the 
most difficult and controversial diagnoses in the management 
of diabetic foot infections. The IDSA guideline8 recommends 
performing plain radiography in suspected osteomyelitis of 
the diabetic foot, given that it is a widely available and 
inexpensive imaging test. One of the limitations of this test is 
its low sensitivity in the early stages of infection, since in 
some cases radiographic bone changes are not apparent until 
10 to 14 days after development of bone infection; for this 
reason, it is recommended to take serial radiographs after 2 or 
4 weeks.8,17,18
Recent studies6,7 demonstrate the importance of the cor-
relation between clinical information and diagnostic inter-
pretation to increase the sensitivity and specificity of this 
imaging test. It should be noted that plain radiography per-
formed in isolation has low sensitivity in diagnosing dia-
betic foot osteomyelitis (54%).19
In certain health care settings, diagnosis of osteomyelitis 
in the diabetic foot is done by a professional using diagnos-
tic tests in isolation without knowing the clinical character-
istics of the lesion. Sometimes the professional who makes 
the diagnosis and the professional who explored the ulcer 
are not one and the same person. Our study demonstrates 
that the strength of agreement even among experienced cli-
nicians is low when using plain radiographs in isolation 
without knowing the clinical characteristics of the lesion 
(K
11AB
 = .35, P < .001). In our study, clinicians who inter-
preted the radiographs only knew the location of the lesion 
and did not know the clinical characteristics of the lesion or 
the result of palpation of bone in the ulcers. If the explored 
ulcer presents a fistulous tract, we believe radiographic 
interpretation will be easier if the clinician explored the 
ulcer previously or at least received clinical information 
about it, as this will make the final diagnosis more reliable.
Experienced clinicians coincided more in their radio-
graphic diagnoses when seeing the same radiograph twice, 
suggesting that the diagnosis given by an experienced clini-
cian is more reliable when making a second interpretation 
of a radiograph (K
11A
 = .75, P < .001).
The main finding of the present article is the determina-
tion that simple radiography is operator dependent when 
interpreted in isolation, even among experienced clinicians. 
The lack of agreement among professionals with the same or 
different experience can lead to different diagnostic 
approaches and therapies that may sometimes be inadequate, 
by giving too much clinical relevance to such studies. This 
suggests that specialization programs and programs aimed at 
associating diagnostic techniques for osteomyelitis in the 
diabetic foot are necessary, as a delay in diagnosis increases 
the likelihood of complications such as amputations, with 
the psychosocial impact this generates.
Declaration of Conflicting Interests
The authors declared no conflicts of interest with respect to the 
authorship and/or publication of this article.
Funding
The authors received no financial support for the research and/or 
authorship of this article.
Table 1. Agreement Indices Between the 2 Groups of Clinicians.
Group Clinician
Kappa 
Coefficient P Value
Strength of 
Agreementa
A A
1
 and A
2
.46 <.001 Moderate
A A
2
 and A
3
.49 <.001 Moderate
A A
1
 and A
3
.33 <.001 Low
B B
1
 and B
2
.18 <.001 Not significant
B B
2
 and B
3
.32 <.001 Low
B B
1
 and B
3
.27 <.001 Low
A and B A
1
 and B
1
.35 <.001 Low
A and B A
2
 and B
2
.39 <.001 Low
A and B A
3
 and B
3
.40 <.001 Low
Intraobserver A
1
 and A
1
.75 <.001 Good
Intraobserver A
2
 and A
2
.61 <.001 Good
Intraobserver A
3
 and A
3
.57 <.001 Moderate
Abbreviations: A1, clinically very experienced member of group A; 
A2, clinically medium-experienced member of group A; A3, clinically 
inexperienced member of group A; B1, clinically very experienced 
member of group B; B2, clinically medium-experienced member of group 
B; B3, clinically inexperienced member of group B.
aLandis and Koch criteria.15
 by guest on November 4, 2014ijl.sagepub.comDownloaded from 
Álvaro-Afonso et al 15
References
 1. Aragon-Sanchez FJ, Cabrera-Galvan JJ, & Quintana-Marrero 
Y, et al. Outcomes of surgical treatment of diabetic foot osteo-
myelitis: a series of 185 patients with histopathological confir-
mation of bone involvement. Diabetologia. 2008;5:1962-1970.
 2. Aragon-Sanchez J. Evidences and controversies about recur-
rence of diabetic foot osteomyelitis: a personal view and an 
illustrated guide for understanding. Int J Low Extrem Wounds. 
2012;11:88-106.
 3. Grayson ML, Gibbons GW, Balogh K, Levin E, & 
Karchmer AW. Probing to bone in infected pedal ulcers. A 
clinical sign of underlying osteomyelitis in diabetic patients. 
JAMA. 1995;273:721-723.
 4. Shone A, Burnside J, Chipchase S, Game F, & Jeffcoate W. 
Probing the validity of the probe-to-bone test in the diagno-
sis of osteomyelitis of the foot in diabetes. Diabetes Care. 
2006;29:945.
 5. Lavery LA, Armstrong DG, Peters EJ, & Lipsky BA. Probe-
to-bone test for diagnosing diabetic foot osteomyelitis: reli-
able or relic? Diabetes Care. 2007;30:270-274.
 6. Morales Lozano R, Gonzalez Fernandez ML, Martinez 
Hernandez D, Beneit Montesinos JV, Guisado Jimenez S, & 
Gonzalez Jurado MA. Validating the probe-to-bone test and 
other tests for diagnosingchronic osteomyelitis in the diabetic 
foot. Diabetes Care. 2010;33:2140-2145.
 7. Aragon-Sanchez J, Lipsky BA, & Lazaro-Martinez JL. Diag-
nosing diabetic foot osteomyelitis: is the combination of 
probe-to-bone test and plain radiography sufficient for high-
risk inpatients? Diabet Med. 2011;28:191-194.
 8. Lipsky BA, Berendt AR, & Cornia PB, et al. 2012 Infectious 
Diseases Society of America clinical practice guideline for the 
diagnosis and treatment of diabetic foot infections. Clin Infect 
Dis. 2012;54:132-173.
 9. Aragón-Sánchez J, Lázaro-Martínez JL, Campillo-Vilorio N, 
Quintana-Marrero Y, & Hernández-Herrero MJ. Controversies 
regarding radiological changes and variables predicting ampu-
tation in a surgical series of diabetic foot osteomyelitis. Foot 
Ankle Surg. 2012;18:233-236.
10. Garcia Morales E, Lazaro-Martinez JL, Aragon-Sanchez 
FJ, Cecilia-Matilla A, Beneit-Montesinos JV, & Gonzalez 
Jurado MA. Inter-observer reproducibility of probing to bone 
in the diagnosis of diabetic foot osteomyelitis. Diabet Med. 
2011;28:1238-1240.
11. Shank CF, & Feibel JB. Osteomyelitis in the diabetic foot: 
diagnosis and management. Foot Ankle Clin. 2006;11: 
775-789.
12. Rajbhandari SM, Sutton M, Davies C, Tesfaye S, & Ward JD. 
“Sausage toe”: a reliable sign of underlying osteomyelitis. 
Diabet Med. 2000;17:74-77.
13. Lipsky BA, Berendt AR, & Deery HG, et al. Diagnosis and 
treatment of diabetic foot infections. Plast Reconstr Surg. 
2006;117(suppl 7):S212-S238.
14. Lipsky BA, Peters EJ, & Senneville E, et al. Expert opinion 
on the management of infections in the diabetic foot. Diabetes 
Metab Res Rev. 2012;28(suppl 1):S163-S178.
15. Landis JR, & Koch GG. The measurement of observer agree-
ment for categorical data. Biometrics. 1977;33:159-174.
16. World Medical Association Declaration of Helsinki: ethical 
principles for medical research involving human subjects. 
JAMA. 2000;284:3043-3045.
17. Teh J, Berendt T, & Lipsky BA. Rational imaging. Investi-
gating suspected bone infection in the diabetic foot. BMJ. 
2009;339:b4690.
18. Berendt AR, Peters EJ, & Bakker K, et al. Diabetic foot 
osteomyelitis: a progress report on diagnosis and a system-
atic review of treatment. Diabetes Metab Res Rev. 2008;24 
(suppl 1):S145-S161.
19. Dinh MT, Abad CL, & Safdar N. Diagnostic accuracy of 
the physical examination and imaging tests for osteomyeli-
tis underlying diabetic foot ulcers: meta-analysis. Clin Infect 
Dis. 2008;47:519-527.
 by guest on November 4, 2014ijl.sagepub.comDownloaded from