PET/CT in Extrapulmonary TB in HIV Patients: a Clinical Trial

Materials and Methods Eighteen HIV-positive patients, 18 years old or more, with extra pulmonary TB were enrolled after providing written informed consent. Baseline FDG-PET/CT was performed before or as soon as possible after initiating TB treatment, followed by FDG-PET/CT after 2 months of TB treatment and at the end of TB treatment according to guidelines (6 months/9-12 months if bone or CNS TB). However, real duration of treatment was left at the discretion of physician, blinded to the results of second and third FDG-PET/CT.

Patients Collected data were: demographics, mode of HIV infection, nadir CD4 cell count, history/date of cART start, mode/date of TB diagnosis, HIV immunovirological status at the time of TB diagnosis, date of TB treatment, culture and DST (Drug Susceptibility Test) results for M. tuberculosis, histopathology results, IRIS (immune reconstitution inflammatory syndrome, documented by treating physician), corticosteroids administration and TB relapse during a follow-up of at least 36 months after the end of TB treatment. TB was classified as proven if culture, molecular biology (PCR) was positive for M. tuberculosis/ histopathology showed caseous granulomatosis, or probable if the patient improved with TB treatment without other diagnosis and an additional criterium (i.e. CSF analysis compatible with TB meningitis, biopsy showing granulomas). Clinical and biological follow up was performed every 4 months during at least 36 months.

FDG-PET/CT analysis Two experienced nuclear medicine physicians, blinded to clinical, bacteriological and histopathological findings, interpreted FDG-PET/CT images. They followed a predefined reading frame, looking first at 9 lymph node sites (cervical, axillary, mediastinal, hilar, retroperitoneal, portal-hepatic, mesenteric, iliac, inguinal) then at other FDG uptake including organ. A site was designated as abnormal if FDG activity was increased relative to that of adjacent normal soft-tissue. The number of abnormal sites (lymph node sites and organs) was counted. Qualitative assessment was made by Visual score for each site: score 1 was defined as less than 5 visible active lymph nodes, score 2 as 5 to 10 visible active lymph nodes and score 3 as too much lymph nodes to count, or massive confluence of lymph nodes at a site. The 3 greatest lymph node sites per patient in terms of visual score were described in detail in our analysis.

FDG-PET were quantitatively analyzed using SUVmax. SUVmax corresponds to the maximum Standardized Uptake Value (SUV) in a voxel within a circular region of interest drawn manually for each site showing FDG uptake. SUVmax was measured on the most active lymph node in each lymph node site. Concerning organ FDG uptake, only SUVmax of the most active organ was described. Percentage change in SUVmax between serial FDG-PET/CT (or DeltaSUVmax) was calculated as follows: %DSUVmax = (SUVmaxT0 - SUVmaxT2) / SUVmaxT0 x 100 (with T0 the baseline FDG-PET/CT and T2 either PET2 or PET3). DeltaSUVmax was considered significant when ≥ 20%. In our analysis, the PET2 performed after 2 months of TB treatment was compared to the baseline FDG-PET/CT (PET1). PET3, performed at the end of TB treatment, was compared to PET1.

Inclusion criteria

• HIV-infected patient
• aged at least 18 years old
• with high clinical suspicion of extrapulmonary tuberculosis

Exclusion criteria

• predominant neurological symptoms
• pregnancy
• positive smear sputum for tuberculosis
• MDR or XDR tuberculosis