Introduction

Pathergy test (PAT) is used in the majority of diagnosis criteria sets for Behget's disease (BD). They are Dilsen criteria (D)1, Japan revised criteria (J)2-3, International criteria (I)4, Iran traditional format criteria (IR)5, and the Classification Tree (CT)6.

Many factors influence the result of the PAT: Needle gauge (from 25 to 21), sharpness ofthe tip (sharp tip of disposable needles or manually blunted tip), direction of the needle penetration (vertical or diagonal, intra-dermal or subcutaneous), in situ injection (serum saline, sodium urate, streptococcal antigens) or just a skin puncture with no injection, the number of skin punctures (from one to several), the delay to read the result (24 to 48 hours), the description of a positive reaction (papule or pustule, reactive erythema surrounding the papule or pustule), and finally where to draw the line between a positive and a negative result. All these factors may influence the result of the PAT, and none of them is standardized. Therefore the results differ from one center to another, introducing a great bias to the accuracy of different diagnosis criteria.

The aim of this study was to analyze the sensitivity, the specificity, and the accuracy of these criteria with the PAT as a criterion being included, and excluded.

2. MATERIALS AND METHODS

2.1 Inclusion criteria

The diagnosis was clinical (it was not based on a particular diagnosis criterion).

2.2 Method

All Behcet's patients of the Behcet's Disease Registry (4717 on January 20, 2002) were included in the study. Consecutive control patients (2407) were those who were referred to the BD Unit with a possible diagnosis of BD, but proving the contrary after evaluation.

2.3 Statistical analysis

The sensitivity, the specificity, and the accuracy of the diagnosis criteria were calculated first by taking into account the result of the PAT, and second by excluding the result of the PAT.

A confidence interval (CI) at 95% was calculated for each percentage.

3. RESULTS

3.1 General data

There was a significant drop in the sensitivity and the accuracy of all diagnosis criteria, while the specificity improved slightly.

3.2 Dilsen criteria

The sensitivity dropped from 86% (CI: 1) to 68% (CI: 1.3). The specificity improved from 91% (CI: 1.1) to 99% (CI: 0.3) whereas the accuracy dropped from 88% (CI: 0.7) to 79% (CI: 0.9). The differences in all the parameters were statistically significant.

3.3 Japan revised criteria

The sensitivity dropped from 88% (CI: 0.9) to 80% (CI: 1.1). The specificity did not change. It was 91% (CI: 0.7) with and without the PAT.

The accuracy dropped from 91% (CI: 0.7) to 86% (CI: 0.8). The difference in the sensitivity and the accuracy was statistically significant.

3.4 International criteria

The sensitivity dropped from 82% (CI: 1.1) to 66% (CI: 1.4). The specificity improved from 98% (CI: 0.5) to 99% (CI: 0.9) whereas the accuracy dropped from 88% (CI: 0.7) to 77% (CI: 0.9). The difference in the sensitivity and the accuracy was statistically significant.

3.5 Iran criteria (traditional format)

The sensitivity dropped from 93% (CI: 0.7) to 79% (CI: 1.2). The specificity improved from 96% (CI: 0.8) to 97% (CI: 0.6) whereas the accuracy dropped from 97% (CI: 0.4) to 85% (CI: 0.8). The difference in the sensitivity and the accuracy was statistically significant.

3.6 Classification tree criteria

The sensitivity dropped from 97% (CI: 0.5) to 89% (CI: 0.9). The specificity improved from 97% (CI: 0.7) to 98% (CI: 0.5) whereas the accuracy dropped from 97% (CI: 0.4) to 92% (CI: 0.6). The difference in the sensitivity and the accuracy was statistically significant.

Table 1. Sensitivity, specificity, and accuracy of diagnosis criteria_

PATHERCY INCLUDED PATHERGY EXCLUDED

Sensitivity Specificity Accuracy Sensitivity Specificity Accuracy

Table 1. Sensitivity, specificity, and accuracy of diagnosis criteria_

PATHERCY INCLUDED PATHERGY EXCLUDED

Sensitivity Specificity Accuracy Sensitivity Specificity Accuracy

CRITERIA

%

CI

%

CI

%

a

%

CI

%

CI

%

CI

Dil sen

86

1.0

91

1.1

88

0.7

68

1.3

99

0.3

79

0.9

Japan

88

0.9

97

0.7

91

0.7

80

1.1

97

0.7

86

0.8

International

82

1.1

98

0.5

88

0.8

66

1.4

99

0,9

77

1.0

Iran

93

0.7

96

0.8

97

0.4

79

1.2

97

0.6

85

0.8

Classification

91

0.5

97

0.7

97

0.4

89

0.9

98

0.5

92

0.6

Tree

The pathergy test is an important criterion for the modern diagnosis and classification criteria of BD. As stated in the introduction part, many factors influence the result of the pathergy test. As a result, many scientists hesitate to employ this tool, especially those not being very familiar with BD.

The sensitivity of all the criteria using the pathergy test dropped when its results were disregarded. The Dilsen criteria was the big looser, it lost 18% of its sensitivity. It was followed by the International criteria (16%) while the least affected were the Classification Tree, and the Japan revised criteria with 8% each.

The absence of the pathergy test promotes greatly the specificity of the Dilsen criteria by improving it by 8%. For the other criteria the improvement was just 1%, or absent (Japan revised criteria).

The accuracy also dropped in the absence of the pathergy test. Most affected were the Iran (traditional format) and the International criteria, while the least affected was the Classification Tree criteria.

The highest accuracy remained for the Classification Tree criteria at 92%.

Was this article helpful?

0 0
How To Deal With Rosacea and Eczema

How To Deal With Rosacea and Eczema

Rosacea and Eczema are two skin conditions that are fairly commonly found throughout the world. Each of them is characterized by different features, and can be both discomfiting as well as result in undesirable appearance features. In a nutshell, theyre problems that many would want to deal with.

Get My Free Ebook


Post a comment