APPENDIX D: Summary Evidence Tablesfor

APPENDIX D: Summary Evidence Tablesfor

LMBPTMLipid Biomarkers and Risk of Cardiovascular Disease

NOTE: Scoring Information see: Christenson et al. 2011

For brevity, apolipoprotein was shortened to ‘Apo’ in the abstraction tables

Apolipoprotein-B

Bibliographic Information
- Author (s)
- Yr Published/Submitted
- Publication
- Author Affiliations
- Funding / Study
- Design
- Facility/Setting
- Time Period
- Population/Sample
- Comparator
- Study bias / Practice
- Description
- Duration
- Training
- Staff/Other Resources
- Cost / Outcome Measures
- Description (s)
- Recording method / Results/Findings
- Number of Participants/CVD events
- Findings/Effect Size
- Stat. Significance/Test(s)
- Results/Conclusion Bias
1. Author(s): Steffen, et.al.
- Year: 2014
- Publication: ArteriosclerThrombVascBiol
- Affiliations: Dept of Laboratory Medicine and Pathology and Divof Biostatistics, School of Public Health, Uni of Minnesota, Minneapolis;Dept of Lab Medicine; National Institutes of Health Molecular Disease Branch, National Heart, Lung, and Blood Institute, Bethesda, MD; Uni of Washington School of Medicine, Div of Cardiology, Uni of Washington Medical Center, Seattle; Cardiovascular Health Research Unit, School of Public Health and Dept of Biostatistics; Uni of Washington, Seattle; Dept of Preventive Medicine, Feinberg School of Medicine, NorthwesternUni, Chicago, IL; Dept of Medicine, Div of Cardiology, Johns Hopkins University
- Funding: This research was supported by the following contracts, N01-HC-95159 through N01-HC-95169 from the National Heart,Lung, and Blood Institute. / - Design: Prospective cohort study
- Facility/Setting: Multi-institutionaaal
- Time period: 8.5-year follow-up
-Population/Sample: 4679 Multi-Ethnic Study of Atherosclerosis (MESA) participants
- Comparator: Standard lipid panel to evaluate the risk of CHD.
- Study bias: None / -Description: Compared the standard lipid panel with nonstandard measurements ApoB and the ratio of ApoB/ApoA-I as well as NMR spectroscopy-derived measures of total LDL-P and the ratio of LDL-P to HDL-particles (HDLP) for evaluating CHD risk. We then determined whether these lipoprotein measures may impart risk independent of cholesterol measures.
- Duration: 8.5 years
- Training: N/A
- Staff/Other resources: NA
- AssociatedCost: Not provided / - Description: Used the AHA/ACC risk calculator score as a baseline prediction model and determined whether
individual additions of ApoB, ApoB/ApoA-I, LDL-P, or
LDL-P/HDL-P improved CHD risk prediction.
Recording Method:
Cox proportional hazards analyses were conducted, and adjustments were made for sex, hypertension medication, systolic blood pressure, age, diabetes mellitus, smoking, and race/ethnicity, with individual adjustmentsfor LDL-C, non-HDL-C, or TC/HDL-C, followed by a combination of TC, LDL-C, HDL-C, and triglycerides. / - Number of Participants/ CVD events: 4679/ 233
- Findings/Effect Size:
Risk of Incident CHD for Lipoprotein Apo-B Markers Adjusted for Non-lipid Variables (sex, HT meds, systolic BP, age, DM, smoking, and race/ethnicity)+Specified Lipid Measures (TC+LDL-C+HDL-C+TGs)
HR (95% CI): 1.23 (0.63 – 2.41)
p-value: 0.55
- Statistical Significance/Test(s): Risk of Incident CHD were calculated as Hazard Ratios with 95% confidence intervals and p- values. Adjustments were made for non- lipid measures,including sex, systolic blood pressure, hypertension medication use, age, and race/ethnicity.
- Results/conclusion biases: None
Quality Rating(10 point maximum): 9 (Good)
Effect Size Magnitude Rating: Moderate / Study (3 pts maximum): _3__ / Practice (2 pts maximum): 1_
<10-yr follow-up / Outcome measures (2 pts. maximum): _2_ / Results/findings (3 pts maximum): _3_
Bibliographic Information
- Author (s)
- Yr Published/Submitted
- Publication
- Author Affiliations
- Funding / Study
- Design
- Facility/Setting
- Time Period
- Population/Sample
- Comparator
- Study bias / Practice
- Description
- Duration
- Training
- Staff/Other Resources
- Cost / Outcome Measures
- Description (s)
- Recording method / Results/Findings
- Type of Findings
- Findings/Effect Size
- Stat. Significance/Test(s)
- Results/Conclusion Bias
2. Author(s): Kappelleet al.,
- Year: 2010
- Publication: Journal of Internal Medicine
- Affiliations: [1] Department of Endocrinology, [2] Nephrology, [3] Cardiology; University Medical Center Groningen and University of Groningen, Groningen, The Netherlands
- Funding: Grants from the Netherlands Heart Foundation and the Jan Kornelis de Cock Foundation / - Design: Prospective case-cohort study
- Facility/Setting: Population study; residents of Groningen, Netherlands
- Time period: median follow-up 7.9 yr (7.5-8.1 yrs range)
-Population/Sample: Participants of the PREVEND study; 6948 subjects aged 28-75 years at baseline without previous cardiovascular disease and who did not use lipid powering drugs initially; 47.8% men in entire cohort; 253 first cardiac events (69.9%): 185 men (72.5%) and 68 (63.6%) women
- Comparator: multivariate models accounting for CHD risk factors (hypertension, diabetes, obesity, smoking, age, sex, triglyceride, TC/HDL-C)
- Study bias: None noted / -Description: multivariate models accounting for CHD risk factors with addition of ApoB/ApoA-I ratio
- Duration:median follow-up 7.9 yr (7.5-8.1 yrs range)
- Training: Not provided
- Staff/Other resources: Not provided
- Cost: Not provided / - Description: First MACE (combined endpoint of incident cardiovascular morbidity and mortality) included fatal/nonfatal MI, ischemic heart disease, coronary artery bypass grafting and percutaneous transluminal coronary angioplasty: (1) Hazard ratio when ApoB/A-I and TC/HDL-C included in the same model
- Recording Method: PRISMANT, the Dutch national registry of hospital discharge diagnoses; municipal register; death certificates / - Number of Participants/ CVD events: 6948/ 362
- Findings/Effect Size:
Hazard ratios (HRs) of Apo-B and their ratios for major adverse CVD events (Adjusted for Non-lipid Variables (sex, age)+Specified Lipid Measures (TGs)
HR* Apo-B(95% CI): 1.13(1.01-1.26); p<0.001
*HRsare givenper 1-SDincrease for all variables.
- Statistical Significance/Test(s): Cox proportional hazard analyses for association of MACEs and apolipoprotein ratios; hazard ratios per SD change; pair-wise models used to directly compare differences in HRs by means of Wald statistics; two-sided p<0.05 significant
- Results/conclusion biases: statin treatment after baseline was not tracked; if statin treatment was started during follow-up, it could underestimate effectiveness of prediction of ApoB/A-I ratio
Quality Rating: 7 (Fair)
(10 point maximum)
Effect Size Magnitude Rating: Moderate / Study (3 pts maximum): _2__
Lack of generalizability and potential bias / Practice (2 pts maximum): 1_
<10 yr follow-up / Outcome measures (2 pts. maximum): _2_ / Results/findings (3 pts maximum): _2_
Potential bias
Bibliographic Information
- Author (s)
- Yr Published/Submitted
- Publication
- Author Affiliations
- Funding / Study
- Design
- Facility/Setting
- Time Period
- Population/Sample
- Comparator
- Study bias / Practice
- Description
- Duration
- Training
- Staff/Other Resources
- Cost / Outcome Measures
- Description (s)
- Recording method / Results/Findings
- Number of Participants/ CVD events
- Findings/Effect Size
- Stat. Significance/Test(s)
- Results/Conclusion Bias
3. Author(s): St-Pierre et.al.
- Year: 2006
- Publication: American Journal of Cardiology
- Affiliations: [1] The Institute on Nutraceuticals and Functional Foods, [2] The Lipid Research Center and CHUL Research Center, and the [3] Québec Heart Institute, Laval Hospital, Laval University, Ste-Foy, Québec, Canada.
- Funding: Operating grant from the Canadian Institute for Health Research; training fellowship from Heart and Stroke Foundation of Canada / - Design: Prospective cohort study
- Facility/Setting: Population study; not provided
- Time period: 13-year follow-up; 1985 to 1990-1998
-Population/Sample: 2,072 CHD-free men from the Quebec Cardiovascular Study; 35-64 years old at start; 173 nonfatal MI, 57 fatal coronary events
- Comparator: multivariate model accounting for conventional risk factors (age, BMI, systolic blood pressure, cigarette smoking, diabetes, medication use, HDL-C, log-transformed triiglycerides) for CHD risk
- Study bias: Only men / -Description: multi-variable model accounting for conventional risk factors for CHD risk with addition of ApoB
- Duration: 13 years follow-up
- Training: Not provided
- Staff/Other resources: Not provided
- Cost: Not provided / - Description: Occurrence of first CHD event (coronary death, MI) (1) Effect of apolipoprotein B on CHD risk associated with high Framingham risk score
- Recording Method: Questionnaire data was confirmed by review of hospital records and death certificates / - Number of Participants/ CVD events: 2,072/230
- Findings/Effect Size:
Relative risk (RR) of Apo-B and their ratios for CHD (Adjusted for Non-lipid Variables (sex, BMI, type 2 DM, Systolic BP, smoking)+Specified Lipid Measures (HDL, TGs)
RR, 95% CI = 1.89 (1.31, 2.73)
- Statistical Significance/Test(s): Student’s t-test; Wilcoxon’s test; Cox’s proportional hazard models (to estimate rates of CHD events); Lifetest procedure; ROC curves; likelihood ratio test
- Results/conclusion biases: Translation of additive ApoB value unclear
Quality Rating: 8 (Good)
(10 point maximum)
Effect Size Magnitude Rating: Moderate / Study (3 pts maximum): _3__ / Practice (2 pts maximum): 1_
Insufficient details provided for model based on Framingham risk score / Outcome measures (2 pts. maximum): _2_ / Results/findings (3 pts maximum): _2_
Data do not permit effect size calculation; potential bias
Bibliographic Information
- Author (s)
- Yr Published/Submitted
- Publication
- Author Affiliations
- Funding / Study
- Design
- Facility/Setting
- Time Period
- Population/Sample
- Comparator
- Study bias / Practice
- Description
- Duration
- Training
- Staff/Other Resources
- Cost / Outcome Measures
- Description (s)
- Recording method / Results/Findings
- Number of Participants/ CVD events
- Findings/Effect Size
- Stat. Significance/Test(s)
- Results/Conclusion Bias
4. Author(s): Lamarche B et. al.
- Year: 1996
- Publication: Circulation
- Affiliations: [1] Lipid Research Center, Laval University Hospital Research Center, Ste-Foy [2]Dept of Social and Preventative Medicine, Laval University, and [3] Dept of Medicine, University of Montreal, Quebec, Canada
- Funding: National health Research Development Program of Health and Welfare Canada; Heart and Stroke Foundation of Canada; Medical Research Council of Canada; Fournier Pharma/Jouveinal / - Design: Prospective cohort study
- Facility/Setting: Population study; not provided
- Time period: 5-year follow-up
-Population/Sample: Quebec Cardiovascular Study; 2,155 men free of CHD; 45-76 years with a mean of 56.5 yrs; 99% of French Canadian descent; 50 cases MI, 51 cases effort angina, 15 deaths to IHD-related causes
- Comparator: multivariate model accounting for conventional risk factors (age, systolic blood pressure, total/DHL cholesterol ratio, diabetes mellitus, smoking, and medication use) for CHD risk
- Study bias: only men of French-Canadian descent / -Description: multi-variable model accounting for conventional risk factors for CHD risk with addition of ApoB ; total/HDL ratio used
- Duration: 5 years
- Training: Not provided
- Staff/Other resources: Not provided
- Cost: Not provided / - Description: First onset ischemic heart disease (angina, coronary insufficiency, nonfatal MI, coronary death) (1) relative risk of IHD for an increase of 1 SD in each lipid (2) relative rate of IHD for synergistic effects of low and high ApoB and Total/HDL-C
- Recording Method: Questionnaires; direct measurements; medical records; death certificate / - Number of Participants/ CVD events:2155/ 116
- Findings/Effect Size:
Five-Year Relative Rates (RR) of Ischemic Heart Disease According to Apolipoprotein B and A-I Levels After Control for nonlipid (age, systolic blood pressure, diabetes mellitus, smoking habits, and medication) and lipid (LDL, HDL, Total/HDL) variables
RR (95% CI) Apo-B: 1.41 (1.29 – 1.54)
- Statistical Significance/Test(s): Proportional hazard models; hazard ratios by use of coefficients from proportional hazard models
- Results/conclusion biases: Five year follow up may not be sufficient to determine full risk profile
Quality Rating: 8 (Good)
(10 point maximum)
Effect Size Magnitude Rating: Moderate / Study (3 pts maximum): _2__
Population not generalizable / Practice (2 pts maximum): 1_
<10-yr follow-up / Outcome measures (2 pts. maximum): _2_ / Results/findings (3 pts maximum): _3_

II. Apolipoprotein A-I

Bibliographic Information
- Author (s)
- Yr Published/Submitted
- Publication
- Author Affiliations
- Funding / Study
- Design
- Facility/Setting
- Time Period
- Population/Sample
- Comparator
- Study bias / Practice
- Description
- Duration
- Training
- Staff/Other Resources
- Cost / Outcome Measures
- Description (s)
- Recording method / Results/Findings
- Number of Participants/ CVD events
- Findings/Effect Size
- Stat. Significance/Test(s)
- Results/Conclusion Bias
1. Author(s): Kappelleet al.,
- Year: 2010
- Publication: Journal of Internal Medicine
- Affiliations: [1] Department of Endocrinology, [2] Nephrology, [3] Cardiology; University Medical Center Groningen and University of Groningen, Groningen, The Netherlands
- Funding: Grants from the Netherlands Heart Foundation and the Jan Kornelis de Cock Foundation / - Design: Prospective case-cohort study
- Facility/Setting: Population study; residents of Groningen, Netherlands
- Time period: median follow-up 7.9 yr (7.5-8.1 yrs range)
-Population/Sample: Participants of the PREVEND study; 6948 subjects aged 28-75 years at baseline without previous cardiovascular disease and who did not use lipid powering drugs initially; 47.8% men in entire cohort; 253 first cardiac events (69.9%): 185 men (72.5%) and 68 (63.6%) women
- Comparator: multivariate models accounting for CHD risk factors (hypertension, diabetes, obesity, smoking, age, sex, triglyceride, TC/HDL-C)
- Study bias: None noted / -Description: multivariate models accounting for CHD risk factors with addition of ApoB/ApoA-I ratio
- Duration:median follow-up 7.9 yr (7.5-8.1 yrs range)
- Training: Not provided
- Staff/Other resources: Not provided
- Cost: Not provided / - Description: First MACE (combined endpoint of incident cardiovascular morbidity and mortality) included fatal/nonfatal MI, ischemic heart disease, coronary artery bypass grafting and percutaneous transluminal coronary angioplasty: (1) Hazard ratio when ApoB/A-I and TC/HDL-C included in the same model
- Recording Method: PRISMANT, the Dutch national registry of hospital discharge diagnoses; municipal register; death certificates / - Number of Participants/ CVD events: 6948/ 362
- Findings/Effect Size:
Hazard ratios (HRs) of Apo-B and their ratios for major adverse CVD events (Adjusted for Non-lipid Variables (sex, age)+Specified Lipid Measures (TGs)
HR* Apo-AI(95% CI): 0.82(0.73-0.92)p<0.001;
*HRsare givenper 1-SDincrease for all variables.
- Statistical Significance/Test(s): Cox proportional hazard analyses for association of MACEs and apolipoprotein ratios; HRs per SD change; pair-wise models used to directly compare differences in HRs by means of Wald statistics; two-sided p<0.05 significant
- Results/conclusion biases: statin treatment after baseline was not tracked; if statin treatment was started during follow-up, it could underestimate effectiveness of prediction of ApoB/A-I ratio
Quality Rating: 7 (Fair)
(10 point maximum)
Effect Size Magnitude Rating: Moderate / Study (3 pts maximum): _2__
Lack of generalizability and potential bias / Practice (2 pts maximum): 1_
<10 yr follow-up / Outcome measures (2 pts. maximum): _2_ / Results/findings (3 pts maximum): _2_
Potential bias
Bibliographic Information
- Author (s)
- Yr Published/Submitted
- Publication
- Author Affiliations
- Funding / Study
- Design
- Facility/Setting
- Time Period
- Population/Sample
- Comparator
- Study bias / Practice
- Description
- Duration
- Training
- Staff/Other Resources
- Cost / Outcome Measures
- Description (s)
- Recording method / Results/Findings
- Number of Participants/ CVD events
- Findings/Effect Size
- Stat. Significance/Test(s)
- Results/Conclusion Bias
2. Author(s): Lamarche B et. al.
- Year: 1996
- Publication: Circulation
- Affiliations: [1] Lipid Research Center, Laval University Hospital Research Center, Ste-Foy [2]Dept of Social and Preventative Medicine, Laval University, and [3] Dept of Medicine, University of Montreal, Quebec, Canada
- Funding: National health Research Development Program of Health and Welfare Canada; Heart and Stroke Foundation of Canada; Medical Research Council of Canada; Fournier Pharma/Jouveinal / - Design: Prospective cohort study
- Facility/Setting: Population study; not provided
- Time period: 5-year follow-up
-Population/Sample: Quebec Cardiovascular Study; 2,155 men free of CHD; 45-76 years with a mean of 56.5 yrs; 99% of French Canadian descent; 50 cases MI, 51 cases effort angina, 15 deaths to IHD-related causes
- Comparator: multivariate model accounting for conventional risk factors (age, systolic blood pressure, total/DHL cholesterol ratio, diabetes mellitus, smoking, and medication use) for CHD risk
- Study bias: only men of French-Canadian descent / -Description: multi-variable model accounting for conventional risk factors for CHD risk with addition of ApoB ; total/HDL ratio used
- Duration: 5 years
- Training: Not provided
- Staff/Other resources: Not provided
- Cost: Not provided / - Description: First onset ischemic heart disease (angina, coronary insufficiency, nonfatal MI, coronary death) (1) relative risk of IHD for an increase of 1 SD in each lipid (2) relative rate of IHD for synergistic effects of low and high ApoB and Total/HDL-C
- Recording Method: Questionnaires; direct measurements; medical records; death certificate / - Number of Participants/ CVD events: 2155/ 116
- Findings/Effect Size:
Five-Year Relative Rates (RR) of Ischemic Heart Disease According to Apolipoprotein B and A-I Levels After Control for nonlipid (age, systolic blood pressure, diabetes mellitus, smoking habits, and medication) and lipid (LDL, HDL, Total/HDL) variables
RR (95% CI) Apo-A-I: 0.88 (0.81 – 0.97)
- Statistical Significance/Test(s): Proportional hazard models; hazard ratios by use of coefficients from proportional hazard models
- Results/conclusion biases: Five year follow up may not be sufficient to determine full risk profile
Quality Rating: 8 (Good)
(10 point maximum)
Effect Size Magnitude Rating: Moderate / Study (3 pts maximum): _2__
Population not generalizable / Practice (2 pts maximum): 1_
<10-yr follow-up / Outcome measures (2 pts. maximum): _2_ / Results/findings (3 pts maximum): _3_

III: Apolipoprotein B/A-I ratio