Diagnostic Value of Ultra-low Dose Thoracic Scanner for the Pulmonary Arteriovenous Malformation Detection in HHT Patient (ULD OSLER)

Diagnostic Value of Ultra-low Dose Thoracic Scanner for the Pulmonary Arteriovenous Malformation Detection in HHT Patient

Diagnostic Value of Ultra-low Dose Thoracic Scanner for the Pulmonary Arteriovenous Malformation Detection in HHT Patient

Hereditary hemorrhagic telangiectasia (HHT) is linked to a dysregulation of angiogenesis leading to the formation of arteriovenous malformations (AVM): cutaneo-mucous telangiectasia and visceral shunts. The diagnosis is clinical and based on Curaçao criteria: recurrent epistaxis, cutaneo-mucous telangiectasia, hereditary signs and presence of visceral AVM. Pulmonary AVMs (PAVM) expose patients to many potentially life-threatening complications, such as strokes or brain abscesses due to the right-left shunt created and the lack of filtration barrier of the pulmonary capillary within the AVM. These patients should therefore have regular monitoring throughout their life by a chest CT scanner every 5 to 10 years in the absence of PAVM at the initial scan or more often if PAVMs are present. The management of PAVMs is based on their early detection and embolization in interventional radiology during which is set up within the afferent artery of the PAVM an embolizing agent, the coil. However, the risk of cumulative irradiation exposure from thoracic scanners and repeated thoracic embolizations over time could be reduces by a decrease of X-rays dose. A new thoracic CT imaging protocol validated in the United States in the primary screening of lung cancer, the ultra-low dose protocol, is a CT scanner acquired at an irradiation dose equivalent to that of a frontal chest x-ray and in profile. The dose reduction is of 40 times the usual dose of a chest CT scanner. The lung parenchyma has a high natural contrast on thoracic CT images and there are few adjacent attenuating structures allowing a drastic reduction of dose. However, from this dose, the image quality is degraded with an increase of the image noise. The diagnostic performances have to be confirmed with qualitative and quantitative measurements. Thus, the objective of this study is to compare the sensitivity and the specificity of the current scanner and the ultra-low dose scanner to reduce the exposure to X-rays.

Hereditary Hemorrhagic Telangiectasia (HHT), diagnosis,, arteriovenous malformation, interventional radiology, pulmonary arteriovenous fistulas, embolization

CT scan acquisition will be performed at 140 kVp (kilovolt peak), 10 mAs(milliampere), CTDI 1.3 mGy with an irradiation dose 0.51 mSv (milliSievert). The CT image will be reconstructed and analyzed in parenchyma window. The CT images will be blinded and re-read by two independent senior radiologists 3 to 6 weeks apart between each re-reading.

CT acquisition will be performed at 80 kVp for patient with a body mass index<30, and 100 kVp with a body mass index>30, 10 mAs, CTDI 0.3 and 0.6 mGy with an irradiation dose 0.15-0.30 mSv. The CT images will be reconstructed and analyzed in mediastinal window. The CT images will be blinded and re-read by two independent senior radiologists 3 to 6 weeks apart between each re-reading.

Comparison of diagnostic performances (specificity and sensitivity) between ultra-low dose and low dose CT scanner

calculation of the sensibility, specificity and the predictive values using the Pearson's method after judgment of the presence or absence of pulmonary AVM to treat on each acquisition (low dose and ultra-low dose).

The effective dose was estimated from the product of the dose length product (DLP) using a conversion coefficient of 0.017 mSv ( milliSievert ) mGy (milligray )-1 cm-1 reported in the European Guidelines for chest examinations.

a grading score is used from "insufficient" to "good" to assess the quality of images produced by standard, iDose, and MRI reconstruction protocol

A confidence score is used (1 = no diagnostic confidence, 2 = average diagnostic confidence, 3 = good diagnostic confidence, 4 = excellent diagnostic confidence)

Inclusion Criteria: Patient monitored for a clinically confirmed Rendu-Osler disease and / or confirmed by molecular biology, seen for an initial assessment or for a reassessment of PAVM (after treatment or without treatment) Patient who agreed to participate in the study and signed the written informed consent Patient affiliated to a social security scheme or similar Exclusion Criteria: Women who are pregnant, nursing (lactating) or at risk of pregnancy (verified by a urine test before performing the scanner) Patient who is protected adults under the terms of the law (French Public Health Code). Patient who has not or poorly understand French