IJCS | Volume 33, Nº1, January / February 2019

88 Albano et al. PET and SPECT in endocarditis Int J Cardiovasc Sci. 2020;33(1):87-93 Review Article Search strategy A comprehensive literature search of the PubMed/ MEDLINE, Scopus, Embase and Cochrane library databases was conducted to find relevant published articles about the diagnostic accuracy of WBC SPECT and 18F-FDG PET/CT in patients affected by IE. We used a search algorithm based on a combination of the terms: a) ‘‘SPECT” OR “Single-photon emission computed tomography” OR “WBC” OR “radiolabeled leukocytes” OR “PET” OR “positron emission tomography” AND b) “endocarditis” OR “heart infection”. No beginning date limit was used; the search was updated until August 31, 2019. Only articles in the English language were selected; pre-clinical or not in vivo studies, review, letters, editorials and conference proceedings were excluded. To expand our search, references of the retrieved articles were also screened for additional studies. Studies considering cardiovascular implantable electronic device infections were excluded by this review. All literature studies collected were managed using EndNote Web 3.3. Study selection All articles reporting patients with IE evaluated by WBC SPECT and 18F-FDG PET/CT in clinical setting were eligible for inclusion. Two researchers (DA and FB) independently reviewed the titles and abstracts of the retrieved articles. The same two researchers then independently reviewed the full-text version of the remaining articles to determine their eligibility for inclusion. Disagreements were resolved by a third opinion (RG). Moreover, in case of studies that included the same population, the report with the highest number of enrolled patients was considered for the analysis. Data abstraction For each included study, the following data were extracted – authors’ names, year of publication, type of study, number of patients, diagnostic test, diagnostic criteria, reference standard, diagnostic performance. The main findings of the articles included in the review are reported in the Results section. Results Literature search The comprehensive computer literature search revealed 665 articles (Figure 1). On reviewing the titles and abstracts, 645 articles were excluded because the data reported data were not within the field of interest of this review. Twenty articles were selected and retrieved in full-text version 7-25 ; no additional study was found when screening the references of these articles. In total, 20 articles were included in the systematic review, four about WBC SPECT 7-10 and 16 about 18F-FDG PET/CT 11-25 Qualitative analysis Characteristics of the studies are detailed in Tables 1 and 2. The IE group included 16 [18F] FDG PET/CT (overall 988 patients) and four SPECT/CT studies (overall 166 patients). Among the PET/CT studies, seven analyzed only prosthetic valve endocarditis (PVE), 9,13,15,16,19-21 two only native valve endocarditis (NVE), 11,23 and the remaining seven analyzed a mixed population or the type of endocarditis was not reported. 12,14,17,18,22,24,25 Among SPECT studies, two included only PVE, 8,9 and the remaining two papers included both NVE and PVE. 7,10 In only one paper, 9 both SPECT and PET/CT techniques were used to study IE. The pooled sensitivity of 18F-FDG PET/CT was 68% (95% CI 55–87), with a high heterogeneity (I2 = 94%, p < 0.001), whereas pooled sensitivity of WBC SPECT was 80% (95% CI 67–94) with a lower heterogeneity (I2=75%, p=0.017). Thepooledspecificityof 18F-FDGPET/ CT was 86.8% (95% CI 82–95) with a high heterogeneity (I2 = 86%, p < 0.001), whereas WBC SPECT showed a pooled specificity of 98% (95% CI 94–100) with no heterogeneity (I2 = 0%, p = 0.625). In a sub-analysis, pooled sensitivity of 18F-FDGPET/CT andWBC SPECT for NVE was 71% (95% CI 49–93) with a high heterogeneity (I2 = 95%, p < 0.001), while pooled sensitivity for PVE was 81% (95% CI 78–93) with a significant heterogeneity (I2 = 67%, p < 0.001). Pooled specificity of 18F-FDG PET/ CT and WBC SPECT for NVE was 96% (95% CI 93-100) with a low heterogeneity (I2 = 52%, p = 0.016), while pooled specificity for PVE was 92% (95% CI 86-96) with a significant heterogeneity (I2 = 79%, p < 0.001). Of 17 manuscripts considering the diagnostic performance of 18F-FDG PET/CT, 11 showed specific preparation before PET/CT scan [9,13,15-18,21-25] and five did not. 11,12,14,19,20 In six studies, 9,13,21-24 participants underwent dietary preparation to promote myocardial suppression (high-fat, low-carbohydrate diet), without heparin injection; in two studied only heparin injection was suggested; 16,18 in the remaining three works, 15,17,25 both myocardial suppression and heparin injection were done. Despite this, there was strong heterogeneity in