Primary prevention guidelines recommend calculation of lifetime cardiovascular disease (CVD) predicted risk in patients who may not meet criteria for high short-term (10-year) Adult Treatment Panel III risk for coronary heart disease (CHD). Extreme obesity and bariatric surgery are more common in women who often have low short-term predicted CHD risk. The distribution and correlates of lifetime CVD predicted risk, however, have not yet been evaluated in bariatric surgical candidates. Using established 10-year (Adult Treatment Panel III) CHD and lifetime CVD risk prediction algorithms and presurgery risk factors, participants from the Longitudinal Assessment of Bariatric Surgery-2 study without prevalent CVD (n = 2,070) were stratified into 3 groups: low 10-year (<10%)/low lifetime (<39%) predicted risk, low 10-year (<10%)/high lifetime (≥39%) predicted risk, and high 10-year (≥10%) predicted risk or diagnosed diabetes. Participants were predominantly white (86%) and women (80%) with a median age of 45 years and median body mass index of 45.6 kg/m 2 . High 10-year CHD predicted risk was common (36.5%) and associated with diabetes, male gender, and older age, but not with higher body mass index or high-sensitivity C-reactive protein. Most participants (76%) with low 10-year predicted risk had high lifetime CVD predicted risk, which was associated with dyslipidemia and hypertension but not with body mass index, waist circumference, high-density lipoprotein cholesterol, or high-sensitivity C-reactive protein. In conclusion, bariatric surgical candidates without diabetes or existing CVD are likely to have low short-term, but high lifetime CVD predicted risk. Current data support the need for long-term monitoring and treatment of increased CVD risk factors in bariatric surgical patients to maximize lifetime CVD risk decrease (clinical trial registration, Long-term Effects of Bariatric Surgery, indentifier NCT00465829 , available at: http://www.clinicaltrials.gov/ct2/results?term=NCT00465829 ).
Recently, long-term, or lifetime, cardiovascular disease (CVD) risk scores have been developed using long-term follow-up data to capture the long incubation period for development of CVD. Using 30-year follow-up data in Framingham participants, Lloyd-Jones et al calculated that at 50 years of age the lifetime predicted CVD risk (i.e., risk of CVD during remaining lifetime) is 1 in 2 for men and 1 in 3 for women. Lifetime CVD predicted risk has been stratified into high (≥39% risk of CVD event during lifetime follow-up) versus low (<39% risk) according to categories of traditional CVD risk factors. Applying this lifetime CVD risk stratification to National Health and Nutrition Examination Survey (NHANES) 2006 data showed that 2 in 3 United States adults with “low” 10-year predicted coronary heart disease (CHD) risk have high lifetime predicted CVD risk. Primary prevention guidelines have recommended the use of long-term, or lifetime, CVD predicted risk stratification in addition to short-term 10-year (Framingham/Adult Treatment Panel III) CHD risk estimation, particularly in women and younger patients who are less likely to meet criteria for high 10-year CHD risk. However, although extreme obesity and bariatric surgery are more common in women than men, the distribution and correlates of lifetime predicted CVD risk has not yet been evaluated in bariatric surgical candidates or others with extreme obesity. Therefore, the objective of the present report is to describe the presurgery distribution and correlates of lifetime CVD predicted risk as an adjunct to 10-year CHD predicted risk in adults enrolled in a large multicenter observational study of bariatric surgery.
Methods
As previously described in detail, the Longitudinal Assessment of Bariatric Surgery (LABS) is a series of National Institutes of Health–funded multicenter longitudinal observational cohort studies designed to rigorously assess risks and benefits of bariatric surgery. LABS includes several studies (LABS-1, LABS-2, and LABS-3) including progressively smaller cohorts with more detailed data collection. From February 1, 2006 through February 17, 2009, patients ≥18 years old without previous weight loss surgery seeking bariatric surgery by participating surgeons at 10 locations throughout the United States were identified for recruitment to LABS-2. The study is registered at http://www.clinicaltrials.gov , all centers had institutional review board approval, and all participants provided informed consent. By study enrollment closure (April 2009) 2,458 participants attended a baseline (presurgery) research visit, which occurred after the surgery approval process was complete and within 30 days before their scheduled bariatric surgical procedure. Of these, 186 were excluded from the present report because of missing key data points at baseline (i.e., CVD, total cholesterol, high-density lipoprotein [HDL] cholesterol, systolic or diastolic blood pressure, hypertension medication, or smoking) and 202 were excluded for prevalent CVD (defined as history of ischemic heart disease, heart failure, or stroke) because they are not part of the lifetime risk scoring algorithm, leaving a sample of 2,070.
Data used in this study were collected at the baseline (presurgery) research visit. Age, gender race, ethnicity, medications, diabetes and cigarette smoking were self-reported by standardized questionnaire, completed ≤30 days of surgery, after presurgical medical workup. Current smoking was defined as smoking within the previous year to avoid bias caused by recent quitting. Height and weight were measured according to standard protocol and body mass index was calculated as kilograms per meter squared. Waist circumference was measured according to a standard protocol using a Gulick II Tape Measure (model 67020; Gays Mill, Wisconsin) by measuring around the abdomen horizontally at the midpoint between the highest point of the iliac crest (hip bone) and the lowest part of the costal margin (ribs). A single measurement of systolic and diastolic blood pressures was obtained by a trained and certified researcher using a Welch Allyn Spot Vital Signs monitor (model 4200B, Welch Allyn, Skaneateles Falls, New York). Sleep apnea and hypertension were determined by clinical researchers based on chart review, physical assessment, or self-report (e.g., sleep study or self-reported continuous positive airway pressure use, blood pressure measurements, and medication).
Personnel at clinical sites drew fasting blood specimens. All assays described in this report were performed by the Northwest Lipid Metabolism and Diabetes Research Laboratories (Seattle, Washington). Using plasma samples, total cholesterol and triglycerides were quantified using a Roche Modular-P AutoAnalyzer (Roche Diagnostics, Mannheim, Germany) using methods standardized to Centers for Disease Control and Prevention reference methods. HDL cholesterol was determined using precipitation procedures with polyethylene glycol (Immuno, Vienna, Austria). Low-density lipoprotein (LDL) cholesterol was calculated using the Friedewald equation except for participants whose triglycerides were ≥400 mg/dl, for whom LDL cholesterol was measured directly by β-quantification. Non-HDL cholesterol was calculated as total cholesterol minus HDL cholesterol. High-sensitivity C-reactive protein (hs-CRP) was measured using the Behring Nephelometer II (Dade Behring, Deerfield, Illinois). Lipid levels were reported as continuous variables and using the following cutoffs for nonoptimal lipids: total cholesterol ≥200 mg/dl, LDL cholesterol ≥100 mg/dl, non-HDL cholesterol ≥130 mg/dl, triglycerides ≥150 mg/dl, and HDL cholesterol <40 mg/dl for men and <50 mg/dl for women.
Participants (without baseline CVD) were classified into 3 risk strata according to 10-year Adult Treatment Panel III CHD predicted risk and lifetime CVD predicted risk : low 10-year/low lifetime risk, low 10-year/high lifetime risk, and high 10-year predicted risk as has been reported for NHANES 2006 data. Specifically, short-term (10-year) predicted risk for hard CHD (myocardial infarction or coronary death) risk was calculated using Framingham point scores from the Adult Treatment Panel III risk assessment tool. Points are assigned for gender, age, total cholesterol, HDL cholesterol, systolic blood pressure, treatment for hypertension, diabetes, and cigarette smoking. “High” 10-year predicted CHD risk was defined as 10-year hard CHD risk ≥10% or diabetes and “low” 10-year predicted risk as 10-year hard CHD risk <10% and no diabetes.
For participants with low 10-year CHD predicted risk, lifetime predicted risk for CVD (all atherosclerotic CVDs) was calculated using the existing lifetime CVD risk algorithm, with cutpoints for systolic and diastolic blood pressures, total cholesterol, diabetes, and smoking as previously described. High lifetime CVD predicted risk was defined as ≥1 increased or major risk factor, i.e., systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or treated hypertension, or total cholesterol ≥200 mg/dl or treated hypercholesterolemia, or diabetes, or current cigarette smoking. Low lifetime CVD predicted risk was defined as meeting all these criteria: systolic blood pressure <140 mm Hg, diastolic blood pressure <90 mm Hg, total cholesterol <200 mg/dl, no diabetes, not currently smoking cigarettes, and no antihypertensive or lipid-lowering medications. High and low lifetime CVD predicted risk have been previously defined based on lifetime risk estimates from Framingham showing that high had ≥39% lifetime risk and low had <39% lifetime risk of CVD. This lifetime risk stratification has also been shown to differentiate higher from lower levels of subclinical atherosclerosis in younger adults in multiethnic cohorts. In a secondary analysis, lifetime CVD risk strata were reclassified by adding low HDL cholesterol (<40 mg/dl for men, <50 mg/dl for women) to the definition of high lifetime risk as was previously reported using NHANES 2006 data.
SAS 9.2 (SAS Institute, Cary, North Carolina) was used for all analyses. Results are reported as median (interquartile range) or percentage (SE). Between-group differences were tested using Pearson chi-square test for differences in proportions or Wilcoxon-rank sum test. Logistic regression was used to test whether differences in lipid categories by gender or risk strata persisted when adjusted for lipid-lowering medication.
Results
Most study participants were white women, with a median age of 45 years and a median body mass index of 45.8 kg/m 2 ( Table 1 ). Diabetes, hypertension, and sleep apnea were common and median hs-CRP levels were high, but prevalence of smoking was low. Median lipid levels were not high, but nonoptimal lipid levels were common. Compared to study participants, LABS-2 participants excluded from this report because of existing CVD before surgery (n = 202) had similar median body mass index and prevalence of smoking but were more likely to be men, older and have a higher prevalence of diabetes, hypertension, sleep apnea, and use of antihypertensive or lipid-lowering medications but lower median levels of hs-CRP and most lipids ( Table 1 ). In study participants, compared to women, men had higher median body mass index and triglycerides levels and were more likely to have diabetes, hypertension, and sleep apnea or to use lipid-lowering or antihypertensive medication, but had a lower prevalence of nonoptimal lipids (except for HDL cholesterol and triglycerides) and lower median hs-CRP. In logistic regression models adjusted for lipid-lowering medications, men remained less likely to have nonoptimal total cholesterol, LDL cholesterol, and non-HDL cholesterol (odds ratio 0.59 to 0.76, p <0.05 for all comparisons), whereas the odds ratio for triglycerides ≥150 mg/dl was 1.23 (95% confidence interval 0.99 to 1.54, p = 0.06) for men versus women.
| Variable | Study Sample (no prevalent CVD) | Excluded Participants (with prevalent CVD) | ||||
|---|---|---|---|---|---|---|
| Total | Women | Men | p Value † | Total | p Value ‡ | |
| (n = 2,070) | (n = 1,665) | (n = 405) | (n = 202) | |||
| White | 85.8 (0.8%) | 84.3 (0.9%) | 92.0 (1.4%) | <0.001 | 88.1 (2.3%) | 0.65 |
| Women | 80.4 (0.9%) | — | — | 59.9 (3.5%) | <0.001 | |
| Age (years) | 45 (36–53) | 44 (36–53) | 46 (38–55) | <0.01 | 56 (47–62) | <0.001 |
| Body mass index (kg/m 2 ) | 45.8 (41.7–51.3) | 45.6 (41.6–50.8) | 47.2 (42.6–53.1) | <0.001 | 45.0 (41.1–51.1) | 0.16 |
| Waist circumference (cm) | 130.5 (120.8–142.3) | 127.5 (118.5–137.5) | 146.9 (136.0–158.6) | <0.001 | 133.6 (124.3–146.5) | <0.01 |
| Diabetes mellitus | 32.7 (1.0%) | 30.5 (1.1%) | 41.5 (2.5%) | <0.001 | 58.9 (3.5%) | <0.001 |
| Smoker | 13.0 (0.7%) | 13.6 (0.8%) | 10.4 (1.5%) | 0.08 | 11.9 (2.3%) | 0.65 |
| Hypertension | 57.5 (1.1%) | 54.9 (1.2%) | 68.4 (2.3%) | <0.001 | 82.7 (2.7%) | <0.001 |
| Systolic blood pressure (mm Hg) | 129 (120–139) | 128 (120–138) | 132 (122–143) | <0.001 | 130 (118–140) | 0.78 |
| Diastolic blood pressure (mm Hg) | 80 (73–86) | 80 (73–85) | 80 (73–87) | 0.10 | 74.5 (69–83) | <0.001 |
| Sleep apnea | 50.8 (1.1%) | 45.5 (1.2%) | 72.4 (2.2%) | <0.001 | 67.3 (3.3%) | <0.001 |
| High-sensitivity C-reactive protein (mg/L) | 7.2 (4.1–13.5) | 7.7 (4.5–14.4) | 5.8 (2.9–9.1) | <0.001 | 6.1 (3.1–12.1) | 0.02 |
| Antihypertensive medications | 50.6 (1.1%) | 48.5 (1.2%) | 59.5 (2.4%) | <0.001 | 77.2 (3.0%) | <0.001 |
| Lipid-lowering medications | 25.4 (1.0%) | 23.1 (1.0%) | 34.8 (2.4%) | <0.001 | 61.9 (3.4%) | <0.001 |
| Total cholesterol (mg/dl) | 185 (161–211) | 188 (163–214) | 173 (151–199) | <0.001 | 169 (142–201) | <0.001 |
| Total cholesterol ≥200 mg/dl | 34.7 (1.1%) | 37.1 (1.2%) | 24.9 (2.2%) | <0.001 | 26.7 (3.1%) | 0.02 |
| Low-density lipoprotein cholesterol (mg/dl) | 109 (87–132) | 111 (89–133) | 101 (79–124) | <0.001 | 95 (73–120) | <0.001 |
| Low-density lipoprotein cholesterol ≥100 mg/dl | 61.3 (1.1%) | 63.7 (1.2%) | 51.6 (2.5%) | <0.001 | 42.1 (3.5%) | <0.001 |
| High-density lipoprotein cholesterol (mg/dl) | 43 (37–51) | 45 (38–53) | 37 (32–42) | <0.001 | 40 (33–48) | <0.001 |
| Low high-density lipoprotein cholesterol ⁎ | 66.1 (1.0%) | 66.7 (1.2%) | 63.7 (2.4%) | 0.26 | 69.8 (3.2%) | 0.29 |
| Nonhigh-density lipoprotein cholesterol (mg/dl) | 140 (116–165) | 141 (117–166) | 134 (111–161) | 0.01 | 127 (103–158) | <0.001 |
| Nonhigh density lipoprotein cholesterol ≥130 mg/dl | 60.6 (1.1%) | 62.2 (1.2%) | 54.1 (2.5%) | <0.01 | 45.5 (3.5%) | <0.001 |
| Triglycerides (mg/dl) | 138 (100–192) | 136 (98–190) | 146 (109–200) | <0.001 | 162 (109–228) | <0.001 |
| Triglycerides ≥150 mg/dl | 42.8 (1.1%) | 41.4 (1.2%) | 48.4 (2.5%) | 0.01 | 54.5 (3.5%) | <0.01 |
⁎ Low high-density lipoprotein cholesterol defined as <40 mg/dl in men and <50 mg/dl in women.
More than 1/3 of participants (36.5%) without CVD had high short-term (10-year) predicted risk, 48.0% had low 10-year but high lifetime predicted risk, and only 15.5% had low 10-year and low lifetime predicted risk ( Table 2 ). Therefore, 75.6% (993 of 1,314) of those with low 10-year risk had high lifetime risk. Addition of low HDL cholesterol to the definition of high lifetime risk ( Table 2 ) reclassified 210 participants as high lifetime risk, such that 91.3% (1,200 of 1,314) of those with low 10-year CHD predicted risk had high lifetime CVD predicted risk. Low 10-year/high lifetime risk was more common for women versus men, for younger versus older men ( Figure 1 ). For women in each age group, and for men <50 years old; ≥50% of those with low 10-year risk had high lifetime risk, whereas high 10-year CHD risk predominated for men >50 years old.
| Lifetime Predicted CVD Risk | Low 10-Year CHD Risk/Low Lifetime CVD Risk | Low 10-Year CHD Risk/High Lifetime CVD Risk | High 10-Year CHD Risk |
|---|---|---|---|
| Standard algorithm (high-density lipoprotein cholesterol not included) | 321 (15.5%) | 993 (48.0%) | 756 (36.5%) |
| With low high-density lipoprotein cholesterol ⁎ as major risk factor | 114 (5.5%) | 1,200 (58.0%) | 756 (36.5%) |
⁎ Low high-density lipoprotein cholesterol defined as <40 mg/dl for men and <50 mg/dl for women.
Characteristics of participants in each risk category are listed in Table 3 . Diabetes status was the greatest contributor to 10-year risk classification; only 10.6% in the high 10-year CHD risk category did not have diabetes. Most participants (92.3%) with diabetes had other increased or major risk factors (data not shown). Compared to overall low 10-year risk, high 10-year risk was also characterized by male gender, older age, lower HDL cholesterol, systolic blood pressure, hypertension, and greater use of antihypertensive medications, by definition. High 10-year CHD risk was also characterized by a larger waist circumference, higher triglycerides, more nonoptimal triglycerides, sleep apnea, and lipid-lowering medication use (not included in risk algorithm). High 10-year risk was not characterized by a higher body mass index and was associated with lower median diastolic blood pressure, hs-CRP, total cholesterol, LDL cholesterol, and non-HDL cholesterol.
| Low 10-Year CHD Predicted Risk (total) | Low 10-Year CHD/Low Lifetime CVD Predicted Risk | Low 10-Year CHD/High Lifetime CVD Predicted Risk | High 10-Year CHD Predicted Risk | p Value § | p Value ∥ | p Value ¶ | |
|---|---|---|---|---|---|---|---|
| (n = 1,314) | (n = 321) | (n = 993) | (n = 756) | ||||
| White | 86.2 (1.0%) | 88.4 (1.8%) | 85.4 (1.1%) | 85.1 (1.3%) | 0.80 | 0.98 | 0.40 |
| Women | 86.3 (1.0%) | 87.2 (1.9%) | 86.0 (1.1%) | 70.2 (1.7%) | <0.001 | <0.001 | 0.58 |
| Age (years) | 41 (34–50) | 37 (31–44) | 43 (35–52) | 50 (42–57) | <0.001 | <0.001 | <0.001 |
| Body mass index (kg/m 2 ) | 45.6 (42.0–50.7) | 45.4 (42.0–50.6) | 45.6 (42.0–50.8) | 46.4 (41.5–52.4) | 0.18 | 0.22 | 0.93 |
| Waist circumference (cm) ⁎ | 127.9 (119.0–138.1) | 127.4 (118.8–138.1) | 128.0 (119.0–138.1) | 136.1 (125.0–147.0) | <0.001 | <0.001 | 0.55 |
| Diabetes mellitus | — | — | — | 89.4 (1.1%) | — | — | — |
| Smoker | 13.2 (0.9%) | — | 17.4 (1.2%) | 12.7 (1.2%) | 0.76 | <0.01 | <0.001 |
| Hypertension | 45.8 (1.4%) | 11.8 (1.8%) | 56.8 (1.6%) | 77.9 (1.5%) | <0.001 | <0.001 | <0.001 |
| Systolic blood pressure (mm Hg) | 128 (120–138) | 123 (116–129) | 130 (121–141) | 131 (122–142) | <0.001 | 0.24 | <0.001 |
| Diastolic blood pressure (mm Hg) | 80 (74–86) | 78 (73–83) | 81 (75–88) | 78 (71–85) | <0.001 | <0.001 | <0.001 |
| Sleep apnea | 44.3 (1.4%) | 39.9 (2.7%) | 45.7 (1.6%) | 62.0 (1.8%) | <0.001 | <0.001 | 0.07 |
| High-sensitivity C-reactive protein (mg/L) † | 7.5 (4.2–13.7) | 8.0 (4.3–14.1) | 7.4 (4.2–13.6) | 7.0 (3.7–13.1) | <0.01 | 0.02 | 0.50 |
| Antihypertensive medications | 37.4 (1.3%) | — | 49.5 (1.6%) | 73.7 (1.6%) | <0.001 | <0.001 | <0.001 |
| Lipid-lowering medications | 12.7 (0.9%) | — | 16.8 (1.2%) | 47.4 (1.8%) | <0.001 | <0.001 | <0.001 |
| Total cholesterol (mg/dl) | 188 (164–212) | 170 (153–185) | 198 (170–220) | 181 (153.5–210) | <0.001 | <0.001 | <0.001 |
| Total cholesterol ≥200 mg/dl | 36.5 (1.3%) | 0.0 (0.0%) | 48.3 (1.6%) | 31.5 (1.7%) | 0.02 | <0.001 | <0.001 |
| Low-density lipoprotein cholesterol (mg/dl) | 113.5 (93–134) | 99 (84–114) | 120 (97–141) | 102 (78–127) | <0.001 | <0.001 | <0.001 |
| Low-density lipoprotein cholesterol ≥100 mg/dl | 66.2 (1.3%) | 49.5 (2.8%) | 71.6 (1.4%) | 52.8 (1.8%) | <0.001 | <0.001 | <0.001 |
| High-density lipoprotein cholesterol (mg/dl) | 44 (38–53) | 43 (38–51) | 45 (38–53) | 41 (35–48) | <0.001 | <0.001 | 0.40 |
| Low high-density lipoprotein cholesterol (mg/dl) ‡ | 62.3 (1.3%) | 64.5 (2.7%) | 61.6 (1.5%) | 72.6 (1.6%) | <0.001 | <0.001 | 0.36 |
| Nonhigh-density lipoprotein cholesterol (mg/dl) | 141 (119–164) | 124 (106–139) | 150 (124–172) | 137.5 (110–166) | 0.03 | <0.001 | <0.001 |
| Nonhigh-density lipoprotein cholesterol ≥130 mg/dl | 62.6 (1.3%) | 40.8 (2.7%) | 69.6 (1.5%) | 57.1 (1.8%) | 0.02 | <0.001 | <0.001 |
| Triglycerides (mg/dl) | 129 (94–174) | 102 (80–142) | 135 (100–182) | 162 (116–227) | <0.001 | <0.001 | <0.001 |
| Triglycerides ≥150 mg/dl | 35.2 (1.3%) | 21.2 (2.3%) | 39.8 (1.6%) | 56.0 (1.8%) | <0.001 | <0.001 | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
