Table 2. Metabolic disease.

	Author, year	Study design	Age	% male	HIV Virological Control	NCD Examined	Main findings
1	Timmons 2014, USA	1. Sub-study of RCT (ACTG 5152s) study (n = 82) PWH 2. Cross-sectional analysis of Indiana University Cohort study, PWH not on ART (n = 47), 3. IU study PWH on ART (n = 49)	Years (mean, SD) 5152s cohort: 1. 35.6¬±8.3 2. Not on ART: 37.0¬±11.5 3. On ART: 40.3¬±8.1	1. 92% 2. 81% 3. 65%	1. 85113.80 ± 123468.52 2. 89.13 ± 151.86 3. 707.95 ± 1173.68 4. 16595.87 ± 38213.40 (mean, SD)	Lipids, lipoproteins, glucose, insulin, Insulin sensitivity was estimated by HOMA-IR. DXA scans at baseline and week 48 in ACTG 5142.	LPS correlated positively with HOMA-IR and triglycerides, particularly in ART naïve. A negative correlation was seen between sCD14 and HDL cholesterol, particularly ART naïve. A negative correlation was seen between sCD14 and trunk fat, limb fat and lean mass by DXA.
2	Villanueva-Millan 2019, Spain	Cross sectional study and a nested case control study of 1.PWH with MS (n = 11) and2. without MS (n = 40)	Years (mean, SD) 1. 48.38±‚ 0.89 2. 52.30¬±1.1	66%	100% undetectable/ suppressed viral load	MS as classified by the NCEP-ATP III Fasting plasma levels of glucose, triglycerides, total cholesterol, LDL, HDL, AST, and ALT, insulin, IL-6, PAI-1 and MCP-1. HOMA-IR for insulin resistance	No differences were found in α-diversity, LBP or sCD14 between MS- and MS+ patients. In MS group, found reduction of several species including Eubacterium eligens, Faecalibacterium prausnitzii, Roseburia intestinalis, Roseburia inulinivorans, Ruminococcus flavefaciens, Subdoligranulum sp, Sutterella wadsworthensis and the Bifidobacterium genus of the Actinobacteria phylum. HIV+MS+ showed higher PAI-1, triglycerides-to-HDL-ratio compared to HIV+MS- The abundance of R. flavefaciens was negatively correlated with PAI-1 and LDL.
3	Armstrong 2021, USA	Cross sectional study of PWH and HIV negative, 1. HIV + MSM untreated- (n = 14) 2. HIV+ MSM, treated (n = 20) 3. HIV+ MSM, treated, with Lipodystrophy- (n = 25) 4. HIV- MSW (n = 22) 5. HIV—MSM- (n = 32)	Years (median, IQR) 1. 34 (26.5–40.3) 2. 46 (42.8–50.5) 3. 60 (54–64) 4. 33 (27.3–38.5) 5. 34 (29.8–44.5)	100%	1.101,400 (20,300–292,514) 2.20 (0–20) 3.0 (0–20) copies/mL, 4.NA 5.NA (median, IQR)	Untargeted metabolomics; Serum analysis of fasting triglycerides, glucose, insulin, LDL, HDL, leptin, and adiponectin and inflammatory cytokines. Diet History Questionnaire II	VSURF selected immune markers that were positively correlated with the metabolic disease score included LBP, ICAM-1, IL-16, IL-12, and GM-CSF. LBP in turn correlated with other markers of systemic inflammation, a loss of beneficial microbes such as butyrate-producing bacteria, and a higher BMI
4	Amador-Lara 2022, Mexico	Cross sectional study of PWH (n = 60)	Years (mean, SD) MS +42.8 ± 10.1, MS—39.4 ± 11.4	76%	92.8% undetectable/ suppressed viral load	Caloric intake measurement Serum analysis of fasting triglycerides, glucose, insulin, LDL, HDL, leptin, HbA1c and adiponectin and inflammatory cytokines. Triglycerides/HDL index, > 3 = predictor of cardiovascular risk HOMA-IR, VAI, FLI, ASCVD FINDRSIC, calculated	Decreased α- diversity in MS+ with a predominance of Prevotella, higher abundance of phylum Bacteroidetes, Proteobacteria and Fusobacteria and a significant decrease in SCFA‐producing bacteria. Higher serum levels of hsCRP in MS+. Higher scores in the HOMA-IR, VAI, FLI, TG/HDL ratio and FINDRISC scale in MS group compared to MS-. Genes related with insulin signalling pathway were significantly increased in MS-.
5	Gelpi 2020, Denmark & Norway	Case control study of 1. PWH MSM (n = 281), 2. PWH non-MSM (n = 124), 3. HIV- MSM (n = 43), 4. HIV- (n = 68)	Years (mean, SD) 1. 53.2 (11.3), 2. 53.4 (10.4), 3. 40.1 (9.6) 4. 60.6 (9.1),	1. 100%, 2. 48.4%, 3. 100% 4. 66.2%,	95.2% undetectable/ suppressed viral load	Non-fasting serum HDL- C, triglycerides, and glucose. CT measures of SAT and VAT.	Identified HMI consisting of lower diversity and increased relative abundance of Gammaproteobacteria and Desulfovibrionaceae and decrease in several Clostridia species. The HMI was positively correlated with sCD14 and LBP and associated with excess risk of MS, hypertriglyceridemia, diabetes, and hypertension A 1-unit increase in α-diversity was associated with a 28% reduced risk of MS. A history of severe immunodeficiency was associated with a 5-fold excess risk between HMI and MS. It also modified the association between the HMI and a 30-cm2 larger VAT area.
6	Gelpi 2022, Denmark	Cross sectional study of (n = 383) PWH	Years (median, IQR) 52 (46.1‚ 61.0)	84.10%	95.3% undetectable/ suppressed viral load	Demographics and lifestyle questionnaires. CT measures of SAT and VAT.	The HMI was associated with higher KTR, QKR and lower KA and Trp concentrations, driven by increase in the family Desulfovibrionaceae and genus Eisenbergiella and reduction in the Lachnospiraceae CAG 56 and Coprococcus 2. KTR was associated with higher VAT-to-SAT ratio and larger VAT area. LBP levels were associated with higher levels of KTR and QKR and associated with larger VAT and SAT areas.
7	Gogokhia 2020, USA	Cross sectional study of (n = 35) PWH	Years (mean, SD) 1. visceral obesity- 51.0 (5.9) 2. general obesity- 45.2 (6.6) 3. lean- 45.7 (7.9	1. 86.7% 2. 0% 3. 90.9%	1. 93.3%, 2. 66.7% 3. 100% undetectable/ suppressed viral load	Serum measurement of hsCRP, adiponectin, leptin, IL-6, sCD14 and MCP-1. CT measures of SAT and VAT.	Higher α-diversity index in participants with lean body type, compared with visceral and general obesity. α-diversity was negatively correlated with VAT area, waist/hip ratio and sCD14. Increased abundance of family Prevotellaceae, genus Erysipelatoclostridium and Bacteroides in the lean group. Increased abundance of family Bacteroidaceae and reduction in Flavobacteriaceae in obese groups. Increased abundance of Dialister and reduced Anaerostipes, Coprococcus, Lactobacilli, Oscillibacter, Atopobium and Alloprevotella genus in viscerally obese groups. Increased abundance of Parabacteroides and Lactobacillus genus and reduction in Solobacterium and Alloprevotella genus in the general obese group. Positive correlation seen between sCD14 and VAT, and between Leptin and SAT area. Obese groups had higher hsCRP, leptin, sCD14 and IL-6 compared to lean groups.
8	Gelpi 2019, Denmark	Case control study of 1.PWH (n = 864) 2 HIV—(n = 75)	Years (median, IQR) 1.52 (47–60.4), 2. 59.8 (51.6–67.1)	83%	95.1% undetectable/ suppressed viral load	BMI, hip, and waist measurements; Serum measurement of hsCRP, total cholesterol, LDL, HDL and triglycerides.	PWH had higher KTR and kynurenine concentrations. QKR was associated with higher hs-CRP and neopterin concentrations while KA was associated with lower hs-CRP and neopterin concentrations. In PWH, increased waist-to-hip ratio was associated with increased KTR and QKR and decreased KA concentrations.
9	Dirajlal-Fargo 2019, USA	Randomised controlled trial of (n = 231) ART naïve PWH	Years, median 36	90%	36307.81 copies/mL (median)	Serum measurement of hsCRP, D-dimer, sCD14, sCD163, IL-6. DXA at baseline and week 96. CT measures of SAT and VAT. Insulin sensitivity was estimated by HOMA-IR.	Initially BDG significantly decreased after 4 weeks of ART initiation, but then increased at week 24 and maintained through week 96. A 2-fold higher BDG level at week 96 was associated with an 8% increase in trunk fat and a 7% increase in total fat.
10	Cheru 2018, USA	Cross sectional study of 1. PWH (on ART n = 149), 2. HIV elite controllers (n = 10) and 3. HIV negative (n = 69)	Years (mean, SD) 1. HIV 47 ± 7, 2. 53.4 ± 8, H3. IV—46 ± 7	63%	92% undeteactable viral load	Body Composition and Dietary Assessment; Serum measurements MCP-1, CXCL10, IL-6, and sCD14, HbA1c, total cholesterol, LDL, HDL, triglycerides,	I-FABP was significantly higher in PWH on ART and correlated with intake of sugar and saturated SCFA. Higher levels of serum I-FABP were inversely associated with weight, BMI, SAT, and VAT in PWH. Serum I-FABP also correlated negatively with waist and hip circumference. I-FABP correlated with MCP-1, CXCL10, sCD163, and LPS among all participants
11	Moon 2018, USA	Cross sectional study of WIHS cohort (n = 48) of PWH 1. with and 2. without DM and HIV negative 3. with and 4. without DM	Years (median, IQR) 1. 52 (51‚ 55), 2. 51 (42‚ 60), 3. 53 (52‚ 57), 4. 49 (42‚ 51)	0	1. 62.5% 2. 91.7% undetectable/ suppressed viral load 3. NA 4. NA	Targeted metabolomics of 130 metabolites: 20 amino acids, 6 biogenic amines, 12 acylcarnitines, 78 glycerophospholipids, and 14 sphingolipids	No difference in α-or β diversity by diabetes status or HIV serostatus. Four genera (Finegoldia, Anaerococcus, Sneathia, and Adlercreutzia) were less abundant in HIV+ DM+ and showed positive correlations with diabetes-associated glycerophospholipids. In HIV+ DM+, plasma levels of KTR, branched-chain amino acid and proline metabolism pathways were higher, while glycerophospholipids were lower
12	Hoel 2018, Denmark & Norway	Cross sectional study (n = 84) of PWH 1. with and 2. without DM and HIV negative 3. with and 4. without DM	Years (median, IQR) 1.54 (51‚ 58) 2.57 (54‚ 60), 3.53 (52‚ 57) 4.49 (42‚ 51)	1. 86%, 2. 90%, 3. 92%, 4. 69%	100% undetectable/ suppressed viral load	Serum glucose, HbA1c, lipid parameters. To assess endothelial dysfunction, L-arginine and ADMA were measured. Plasma levels of neopterin, kynurenine and tryptophan and the KTR was determined	The lowest α-diversity was found in PWH with DM, followed by DM alone, HIV alone, and healthy controls. MSM, Physical activity, metformin use and HDL cholesterol which were associated with higher α-diversity measures, and smoking and Framingham risk score being associated with lower α-diversity. MSM status was associated with lower predicted abundance of tryptophan metabolizing microbes. HIV+ DM+ had higher plasma KTR and higher neopterin compared to HIV- DM+ KTR and neopterin were positively correlated with ADMA and negatively correlated with L-arginine/ADMA-ratio, particularly in HIV+ DM+ patients.
13	Hove-Skovsgaard 2017, Denmark	Cross sectional study of 1. PWH with DM (n = 25), 2.PWH without DM (n = 25), 3. HIV- with DM (n = 22) and 4.HIV- without DM (n = 28)	Years (median, IQR) 1. 58 (55–61) 2. 55 (50–58), 3. 57 (55–60), 4.57 (53–60)	1. 92% 2. 96%, 3. 89%, 4. 72%	1. 26 (16–38) 2. 27 (15–40), copies/mL, (median, IQR) 3. NA 4. NA	Framingham risk score, Serum measurements of hsCRP ADMA and L-arginine	A higher ADMA and lower L-arginine/ADMA ratio was found in HIV + DM+. 50% of HIV + DM+ had hsCRP above 3 mg/L. A positive correlation between ADMA and TMAO was seen in HIV + DM+.
14	Jayanama 2022, Thailand	Cross sectional study of 1. PWH, (n = 20) with prediabetes, and 2. PWH (n = 20) with normoglycaemia	Years (mean, SD) 1.50.9¬±5.6, 2. 51.8±6.6	65%	100% undetectable/ suppressed viral load	75-g OGTT and HbA1c measurement	Lower α—diversity in the prediabetes group and a significant difference in β-diversity was observed between prediabetes and normoglycemia groups. Increased abundances of two genera in Firmicutes (Streptococcus and Anaerostignum) in the prediabetes group. Increased abundances of 13 genera (Akkermansia, Gastranaerophilales, Desulfovibrio, Butyricimonas, Colidextribacter, Christensenellaceae R 7 group, Victivallis, Uncultured Bacteroidota, Uncultured phylum Firmicutes, Holdemanella, UCG-005, Eubacterium ruminantium group, and family Oscillospiraceae-associated group in normoglycaemia group.

Abbreviations: ADMA, asymmetric dimethylarginine; ASCVD, Atherosclerotic Cardiovascular Disease, AST, aspartate aminotransferase; CT, computed tomography; DM, diabetes milletus; DXA, dual-energy X-ray absorptiometry; FLI, fatty Liver Index; FINDRISC, Finnish Diabetes Risk Scale; HDL, HbA1c, hemoglobin A1c; HDL, high density lipoprotein, HOMA-IR, homeostasis model assessment-insulin resistance; HMI, HIV-related microbiota index; I-FABP, fatty acid binding protein; KA, kynurenic acid; KTR, Kynurenine to Tryptophan Ratio; LBP, lipopolysaccharide binding protein;LDL, low-density lipoprotein; MCP-1, monocyte chemotactic protein-1; MS, metabolic syndrome; MSM, men who have sex with men; NCEP-ATP III, National Cholesterol Education Program Adult Treatment Program III; OGTT, oral glucose tolerance test; QKR, quinolinic-to-kynurenic acid ratio SAT, subcutaneous adipose tissue; SCFA, short chain fatty acids; VSURF, Variable Selection Using Random Forest; VAI, visceral Adiposity Index; VAT, Visceral adipose tissue