Nl-Fr

View abstract

This abstract is assigned to sessions
TitleHierarchical cluster analysis using blood test data in ocular sarcoidosis
Accept poster if oral is not possible ?Yes
PurposeSarcoidosis is the most frequent disease causing uveitis in Japan, and it is a disease with unknown etiology which forms non-caseating epithelioid cell granulomas in the whole body. The purpose of this study was to group the patient diagnosed with ocular sarcoidosis by cluster analysis using blood test data obtained from patients.
MethodsWe analyzed 41 patients diagnosed with ocular sarcoidosis (13 males and 28 females, average age 55.4 ± 18.3 years) at the Department of Ophthalmology, Tokyo Medical University Hospital between 2015 and 2019. Cluster analysis was performed using the values of peripheral blood leukocyte count, CH50, C-reactive protein (CRP), serum soluble interleukin-2 (sIL-2) receptor, and angiotensin I converting enzyme (ACE) from blood test data obtained at diagnosis.
ResultsThe hierarchical cluster analysis grouped the patient population into 5 clusters: 1) peripheral blood leukocyte count elevated group, 2) CH50 elevated group, 3) serum ACE elevated group, 4) normal range group (peripheral blood leukocyte count, CH50, CRP, serum sIL-2 receptor, serum ACE), 5) CH50, CRP, serum sIL-2 receptor and serum ACE elevated group. Elevations of serum sIL-2 receptor and ACE correlated with bilateral hilar lymphadenopathy (BHL) (p = 0.01). Cluster 5 showed significant BHL involvement compared to the other clusters.
ConclusionOcular sarcoidosis could be grouped by hierarchical cluster analysis based on blood test data. The results of analysis may predict the presence or absence of extraocular lesions of sarcoidosis.
Conflict of interestNo
Author 1
Last nameMATSUSHIMA
Initials of first name(s)R
DepartmentTokyo Medical Univesity
CityTokyo
CountryJapan
Author 2
Last nameUsui
Initials of first name(s)Y
DepartmentTokyo Medical Univesity
CityTokyo
CountryJapan
Author 3
Last nameShimizu
Initials of first name(s)H
DepartmentTokyo Medical Univesity
CityTokyo
CountryJapan
Author 4
Last nameGoto
Initials of first name(s)H
DepartmentTokyo Medical Univesity
CityTokyo
CountryJapan