Advertisement

Prevalence of Severe (500 to 2,000 mg/dl) Hypertriglyceridemia in United States Adults

Published:January 20, 2011DOI:https://doi.org/10.1016/j.amjcard.2010.11.008
      A growing amount of evidence has supported an association between elevated triglyceride levels and cardiovascular disease. However, little information regarding co-morbidities, levels of other cholesterol types, or medication use among adults with severe hypertriglyceridemia (SHTG; (500 to 2,000 mg/dl) is available. We examined the data from 5,680 subjects, ≥20 years old, who had participated in the National Health and Nutrition Examination Survey from 2001 and 2006, to evaluate the epidemiology of adults with SHTG. Approximately 1.7% of the sample had SHTG, equating to roughly 3.4 million Americans. The participants with SHTG tended to be men (75.3%), non-Hispanic whites (70.1%), and aged 40 to 59 years (58.5%). More than 14% of those with SHTG reported having diabetes mellitus, and 31.3% reported having hypertension. Only 14% of the subjects with SHTG reported using statins, and 4.0% reported using fibrates. The factors significantly associated with having SHTG included high-density lipoprotein <40 mg/dl (odds ratio [OR) 11.45, 95% confidence interval [CI] 6.28 to 20.86), non–high-density lipoprotein 160 to 189 mg/dl (OR 9.74, 95% CI 1.68 to 56.40) or non–high-density lipoprotein ≥190 mg/dl (OR 24.99, 95% CI 3.90 to 160.31), diabetes mellitus (OR 3.04, 95% CI 1.45 to 6.37), and chronic renal disease (OR 7.32, 95% CI 1.45 to 36.94). In conclusion, SHTG is rare among adults in the United States and the use of pharmacologic intervention is low among those with SHTG.
      Little information is published regarding the epidemiology of adults with severe hypertriglyceridemia (SHTG; 500 to 2,000 mg/dl). Descriptive data on this condition are of public health importance, given that elevated serum triglycerides (TG) are associated with an increased risk of coronary heart disease and pancreatitis.
      Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and treatment of High Blood cholesterol in Adults (Adult Treatment Panel III) final report.
      Using data obtained from the National Health and Nutrition Examination Survey (NHANES) from 2001 to 2006, our objectives were to estimate the distribution of high TG levels in the United States adult population, assess the distribution of TG levels as they relate to the use of dyslipidemic agents, and evaluate the association between SHTG and personal characteristics and habits, levels of other cholesterol types, co-morbidities, and medication use.

      Methods

      The present study used data from the NHANES (2001 to 2006) surveys. Specific details about NHANES have been provided by the Centers for Disease Control (available at: http://www.cdc.gov/nchs/nhanes.htm). In brief, NHANES is a program of studies designed to assess the health and nutritional status of adults and children in the United States. The survey is unique in that it combines interviews and physical examinations. The data obtained from NHANES are our best resource for estimating the prevalence of SHTG in the United States adult population. The NHANES participants were initially interviewed in their homes and then were asked to visit a mobile examination center, where they completed additional questionnaires and underwent a physical examination and fasting blood draw.
      In the past, the NHANES surveys were conducted on a periodic basis, and the data were released as single, multiyear data sets. For example, NHANES III covered the 6 calendar years from 1988 to 1994 and has generally been analyzed as a single 6-year survey. Since 1999, NHANES has been planned and conducted as a continuous annual survey. For a variety of reasons, including disclosure issues, the continuous NHANES survey data have been released as public use data files in 2-year increments (e.g., NHANES 2001 to 2002, NHANES 2003 to 2004, NHANES 2005 to 2006). For the present study, NHANES 2001 to 2006 were combined to achieve an optimal sample size for the analyses.
      All adults who participated in the NHANES 2001 to 2006 were included in our study, except for those aged <20 years, pregnant and lactating women, and anyone for whom morning, fasting TG levels had not been taken.
      Approximately 50% of the NHANES participants provided a fasting blood sample during a morning examination. On collection, the blood samples were stored at −20°C until they were shipped to the Lipoprotein Analytical Laboratory at Johns Hopkins University (Bethesda, Maryland) for testing. At the laboratory, the TG levels were measured enzymatically through a series of reactions in which the TGs were hydrolyzed to produce glycerol. Using glycerol oxidase, the glycerol was oxidized and eventually converted into phenoazone. The absorbance was measured at 500 nm
      Centers for Disease Control
      NHANES 1999–2000 Data Documentation.
      . Quality control procedures, including the testing of blind quality control samples, were followed and can be found in the laboratory files for each survey at the NHANES Web site (available at: http://www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm).
      The following risk factors were included in our analysis and categorized as yes or no unless otherwise specified: gender (male or female); age (20 to 39, 40 to 59, and ≥60 years); race (non-Hispanic white, Mexican/Hispanic, non-Hispanic black, and other); smoking status (never, current, former); alcohol consumption; body mass index (under/normal weight <25, overweight 25 to <30, and obese ≥30 kg/m2); total cholesterol (<200, 200 to <240, and ≥240 mg/dl); high-density lipoprotein (HDL; ≥60, 40 to 59, and <40 mg/dl); non-HDL (<130, 130 to 159, 160 to 189, and ≥190 mg/dl); C-reactive protein quintiles (0.01 to 0.06, 0.07 to 0.14, 0.15 to 0.28, 0.29 to 0.56, and 0.57 to 25.40 mg/dl); diagnosed diabetes mellitus; and a history of myocardial infarction, heart failure, coronary heart disease, hypertension, stroke, angina, chronic renal disease, liver disease, and hypothyroidism. In addition, the use of TG-raising medications, including levothyroxine, estrogens, and β blockers, was assessed. Information on the status of each of these diseases, conditions, or use of medications was obtained by participant self-report, with the exception of several diseases (i.e., liver disease and chronic renal disease), for which the variables were combined to classify the participants as having or not having the disease. Liver disease was defined as a total bilirubin >2 mg/dl, alanine aminotransferase >100 U/L, or aspartate aminotransferase >100 U/L. Chronic renal disease was defined as albuminuria (urine albumin/urine creatinine 100≥30) and an abnormal glomerular filtration rate. Stage 1 was albuminuria with an estimated GFR >90 ml/min/1.73 m2; stage 2, albuminuria with an estimated GFR of 60 to 89 ml/min/1.73 m2; stage 3, a GFR of 30 to 59 ml/min/1.73 m2; stage 4, a GFR of 15 to 29 ml/min/1.73 m2; and stage 5, a GFR of <15 ml/min/1.73 m2. The use of the following dyslipidemic agents alone and combined was also evaluated: niacin by prescription only, over-the-counter niacin, if the study participant reported taking ≥250 mg/day, statin, fibric acids, bile acid sequestrants, ezetimibe, and omega-3 (Lovaza or omega-3 supplements at a dose of ≥250 mg/day). Lovaza was combined with omega-3 supplements in the NHANES 2005 to 2006 data. We used the 2002 National Cholesterol Education Program guidelines
      Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and treatment of High Blood cholesterol in Adults (Adult Treatment Panel III) final report.
      to categorize the TG levels as follows: normal, <150 mg/dl; borderline high, 150 to 200 mg/dl; high, 200 to <500 mg/dl; and severe, 500 to 2,000 mg/dl.
      • Durstine J.L.
      • Grandjean P.W.
      • Cox C.A.
      • Thompson P.D.
      Lipids, lipoproteins, and exercise.
      For a subset of our evaluation, the TG levels were dichotomized as severe (500 to 2,000 mg/dl) or not severe (<500 mg/dl).
      Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and treatment of High Blood cholesterol in Adults (Adult Treatment Panel III) final report.
      Descriptive statistics, including the mean values for continuous variables and percentages for categorical variables, were used to summarize the data. The 95% confidence interval (CI) was calculated around all percentages. Crude and adjusted multivariate logistic regression models were used to calculate the odds ratios (ORs) and 95% CIs for the association between the risk factors and having SHTG. The adjusted models included the following covariates: gender, age, ethnicity, total cholesterol, HDL, non-HDL, smoking, alcohol, C-reactive protein quintiles, physical activity, body mass index, diagnosed diabetes, myocardial infarction, heart failure, coronary heart disease, hypertension, stroke, angina, chronic renal disease, liver disease, hypothyroidism, levothyroxine, estrogens, and β blockers. Statistical analyses were conducted using Statistical Analysis Systems, version 9.1 (SAS Institute, Cary, North Carolina).

      Results

      Our sample included 5,680 adults ≥20 years old who had participated in NHANES 2001 to 2006 and who had had the fasting TG level measured as a component of the NHANES. The distribution of TG levels in the sample are listed in Table 1. More than 67% of the participants had normal TG levels, 14.2% had borderline high TG levels, and 16.3% had high TG levels. The prevalence of SHTG (500 to 2,000 mg/dl) was 1.7%, equating to roughly 3.4 million Americans. Three participants had TG levels >2,000 mg/dl. Given the small number, these patients were excluded from our analysis.
      Table 1Distribution of triglyceride (TG) levels in adults ≥20 years old: results from National Health and Nutrition Examination Survey (NHANES) 2001 to 2006
      Triglyceride Level (mg/dl)ATP-III Guidelines ClassificationSample Size (n = 5,680)Population Estimate (n = 197,088,927)Percentage of Population
      <150Normal3,812133,660,63467.8%
      150–200Borderline high83927,933,11414.2%
      200–<500High93932,056,08916.3%
      500–2,000SHTG873,357,2141.7%
      >2,000SHTG381,8770.0%
      Total5,680197,088,927100.0%
      ATP-III, Adult Treatment Panel III; SHTG, severe hypertriglyceridemia.
      The distribution of risk factors across the TG levels is listed in Table 2. The participants with SHTG tended to be men (75.3%), non-Hispanic whites (70.1%), and aged 40 to 59 years (58.5%). In addition, a large majority of those with SHTG reported participating in habits that might be risk factors for elevated TG levels, including a current or former smoker (70.1%), consuming alcohol (71.5%), and participating in <150 minutes/week of exercise (64.0%). The lipid profiles of the participants with SHTG were not unexpected. More than 70% of the participants with SHTG had HDL levels <40 mg/dl, 73.5% had non-HDL cholesterol levels ≥190 mg/dl, 66.7% had total cholesterol levels of ≥240 mg/dl. The mean TG level for the participants with SHTG was 784.5 mg/dl compared to 272.6 mg/dl for the participants with high TG levels, 172.1 mg/dl for the participants with borderline high TG levels, and 92.7 mg/dl for those with normal TG levels. More than 14% of the those with SHTG reported having diabetes, and 31.2% reported having hypertension.
      Table 2Risk factor distribution among adults ≥20 years old, United States National Health and Nutrition Examination Survey (NHANES) 2001 to 2006 population, by Adult Treatment Panel III classification of triglyceride (TG) levels
      Descriptive VariableSerum TG Level (mg/dl)
      <150 (n = 3,812) [n = 133,660,634]150–<200 (n = 839) [n = 27,933,114]200–<500 (n = 939) [n = 32,056,089]500–2,000 (n = 87) [n = 3,357,214]
      Gender
       Male46.2% (45–48)51.1% (48–54)59.2% (55–63)75.3% (62–89)
       Female53.8% (52–55)48.9% (46–52)40.8% (37–45)24.7% (11–38)
      Race/ethnicity
       Hispanic10.9% (8.7–13)12.2% (7.5–17)11.8% (8.0–16)14.9% (6.6–23)
       Non-Hispanic white70.8% (67–74)75.2% (69–81)77.6% (73–83)70.1% (60–81)
       Non-Hispanic black13.3% (11–16)6.3% (4.1–8.6)5.6% (3.8–7.4)5.7% (2.3–9.0)
       Other5.1% (4.0–6.1)6.2% (4.1–8.4)4.9% (2.6–7.2)9.3% (0.7–18)
      Age (years)
       20–3942.3% (40–45)28.1% (24–32)29.8% (25–34)31.1% (19–44)
       40–5937.8% (36–40)43.0% (39–47)41.9% (38–46)58.5% (46–71)
       ≥6019.9% (18–22)28.9% (26–32)28.3% (25–32)10.4% (4.0–17)
      Currently insured80.7% (78–83)82.0% (79–85)83.1% (80–86)67.7% (53–82)
      Body mass index (kg/m2)
       Underweight (<18.50)2.1% (1.5–2.7)0.2% (0.0–0.4)0.2% (0.0–0.5)
       Normal weight (>18.50–<25)38.7% (37–40)17.8% (15–21)14.7% (12–17)10.5% (0.2–21)
       Overweight (25–<30)31.1% (29–33)40.9% (37–45)36.6% (33–40)46.1% (36–57)
       Obese (≥30)26.2% (24–28)39.0% (35–43)46.4% (43–50)43.5% (33–54)
      Smoking status
       Current24.6% (23–27)23.4% (20–27)27.5% (24–31)39.9% (25–55)
       Former23.4% (21–26)28.9% (25–33)28.7% (25–32)30.2% (18–42)
       Never51.9% (49–55)47.6% (42–53)43.8% (40–47)29.9% (18–42)
      Alcohol consumption67.8% (65–71)65.7% (61–70)64.4% (59–69)71.5% (57–86)
      Physical activity (minutes/week)
       <15054.2% (52–57)59.3% (55–64)64.3% (60–69)64.0% (53–75)
       ≥15045.8% (43–48)40.7% (36–45)35.7% (31–40)36.0% (25–47)
      Abdominal obesity (cm)
       >102 (men), >88 (women)42.5% (40–45)64.3% (61–68)66.1% (62–70)63.5% (49–78)
       ≤102 (men), ≤88 (women)54.8% (53–57)32.7% (29–36)30.7% (27–35)31.4% (18–45)
      Carbohydrate (% total intake)
       <6086.4% (85–88)83.7% (79–88)87.2% (84–90)89.3% (81–97)
       ≥6013.6% (12–15)16.3% (12–21)12.8% (9.8–16)10.7% (2.6–19)
      High-density lipoprotein (mg/dl)
       <409.2% (7.9–10)22.8% (19–27)39.4% (36–43)70.4% (57–84)
       40–<6052.4% (50–54)60.4% (56–65)52.2% (48–56)28.7% (15–42)
       ≥6038.4% (36–41)16.8% (14–20)8.4% (5.4–11)0.9% (0.0–2.8)
      Non-HDL cholesterol (mg/dl)
       <13047.9% (46–50)20.8% (18–23)12.6% (10–15)2.7% (0.0–6.8)
       130–<16030.2% (28–32)32.1% (28–36)24.8% (22–28)6.3% (0.7–12)
       160–<19015.2% (14–17)28.3% (24–32)30.1% (27–34)17.6% (6.4–29)
       ≥1906.7% (5.5–7.9)18.9% (16–22)32.5% (29–36)73.5% (61–86)
      Total cholesterol (mg/dl)
       <20061.5% (59–64)43.6% (39–49)35.5% (33–38)11.0% (3.9–18)
       200–<24028.7% (27–31)35.3% (31–40)37.3% (33–41)22.3% (12–33)
       ≥2409.8% (8.4–11)21.1% (18–24)27.2% (24–30)66.7% (55–79)
      Mean triglycerides92.7% (91.3–94.0)172.1% (170.6–173.6)272.6% (266.2–279.0)784.5% (698.3–870.7)
      LDL levels (mg/dl)
       <10034.4% (32–37)26.1% (23–30)31.4% (28–35)64.0% (53–75)
       100–<13034.5% (32–37)31.0% (27–35)30.8% (27–34)19.1% (8.9–29)
       130–<16021.4% (20–23)26.8% (22–31)22.8% (20–26)1.6% (0.0–4.3)
       160–<1907.4% (6.2–8.6)11.7% (8.8–15)9.6% (7.5–12)10.5% (2.0–19)
       ≥1902.2% (1.6–2.9)4.5% (2.5–6.5)5.4% (3.6–7.1)4.8% (0.0–10)
      C-reactive protein quintiles (mg/dl)
       1 (0.01–0.06)16.0% (14–18)5.3% (3.7–6.9)3.8% (2.0–5.6)7.0% (0.4–14)
       2 (>0.06–0.14)23.6% (21–26)17.7% (14–21)17.9% (15–21)14.3% (3.4–25)
       3 (>0.14–0.28)21.0% (19–23)21.2% (18–25)20.9% (18–24)29.8% (17–42)
       4 (>0.28–0.56)19.6% (18–21)28.0% (23–33)28.3% (25–31)28.7% (17–40)
       5 (>0.56–25.40)19.8% (18–21)27.7% (24–32)29.1% (26–32)20.3% (10–31)
      Hemoglobin A1c (%)
       <797.3% (97–98)95.0% (93–97)92.2% (90–94)82.%5 (77–88)
       7–<81.5% (1.1–1.8)2.1% (0.8–3.4)3.4% (2.0–4.8)4.3% (0.0–8.9)
       8–<9.50.6% (0.3–0.9)1.4% (0.4–2.4)2.0% (0.8–3.2)0.6% (0.0–1.9)
       ≥9.50.5% (0.2–0.8)1.4% (0.5–2.4)2.0% (1.0–3.0)11.8% (6.7–17)
      Co-morbidities (n)
       057.9% (56–60)45.5% (41–50)39.6% (35–45)41.4% (29–54)
       125.0% (23–27)27.2% (22–32)31.6% (28–35)32.5% (18–47)
       29.4% (8.3–11)14.6% (12–18)15.1% (13–18)13.6% (2.9–24)
       ≥37.7% (6.6–8.7)12.7% (10–15)13.7% (11–16)12.5% (5.2–20)
      Diabetes mellitus5.2% (4.6–5.9)10.0% (7.4–13)12.5% (9.3–16)14.6% (5.8–23)
      Myocardial infarction2.8% (2.2–3.5)3.5% (2.1–4.9)6.1% (4.5–7.8)1.7% (0.0–4.0)
      Heart failure1.9% (1.5–2.4)2.5% (1.2–3.9)3.0% (1.7–4.2)1.7% (0.0–4.0)
      Coronary heart disease3.1% (2.5–3.7)4.2% (2.8–5.7)5.0% (3.4–6.6)5.3% (0.1–10)
      Hypertension25.5% (24–27)35.7% (31–40)41.0% (37–46)31.2% (20–42)
      Stroke2.4% (1.8–3.0)3.2% (1.8–4.6)3.9% (2.5–5.3)3.0% (0.0–6.5)
      Angina pectoris2.3% (1.7–2.8)3.0% (1.7–4.3)4.7% (3.0–6.3)1.6% (0.0–4.8)
      Liver disease1.1% (0.8–1.5)2.0% (0.6–3.5)1.7% (0.6–2.7)2.8% (0.0–8.5)
      Hypothyroidism9.2% (8.0–10)10.9% (7.9–14)11.6% (8.9–14)10.2% (1.8–19)
      Cancer5.6% (4.6–6.5)7.4% (5.0–9.7)5.8% (3.9–7.8)9.5% (0.0–19)
      Chronic kidney disease
       Stage 12.8% (2.2–3.4)2.6% (1.4–3.7)3.6% (2.6–4.7)9.9% (0.5–19)
       Stage 23.5% (2.8–4.2)4.8% (3.1–6.4)7.0% (4.7–9.2)6.6% (1.2–12)
       Stage 36.7% (5.6–7.7)11.7% (9.0–14)9.5% (7.2–12)6.6% (0.4–13)
       Stage 40.2% (0.1–0.4)0.4% (0.1–0.8)0.4% (0.0–0.7)
       Stage 50.2% (0.0–0.3)0.1% (0.0–0.2)0.2% (0.0–0.4)
      Cortical steroid therapy1.3% (0.8–1.7)1.9% (0.7–3.0)1.3% (0.5–2.1)
      Estrogen use8.5% (7.4–9.6)8.6% (5.8–11)9.1% (6.7–12)1.9% (0.0–4.6)
      β-Adrenergic blocker6.8% (5.8–7.8)13.8% (11–17)15.7% (13–19)12.7% (4.3–21)
      Levothyroxine5.6% (4.6–6.6)7.1% (4.9–9.2)8.8% (6.4–11)5.4% (0.0–12)
      Data are presented as % (95% CI).
      Missing values not listed.
      Almost 85% of our sample reported they were not using any dyslipidemic agent (data not shown). The participants who refused to answer or did not know whether they had taken prescription medications in the past month (n = 5) were classified as not using any dyslipidemic agent. Statins were the most commonly reported dyslipidemic agent, with 11.8% of the sample using this medication (12.1% men and 11.5% women). A larger proportion of the study participants reported using monotherapy instead of polytherapy (13.0% vs 2.3%, respectively). Overall, a low percentage of subjects with SHTG reported using any dyslipidemic medications. This was especially true for the women. As shown in Figure 1, almost 14% of subjects with SHTG reported using statins and 4.0% reported using fibrates. The least commonly reported dyslipidemic agent used among subjects with SHTG was prescription niacin, with <1% reporting the use of this medication.
      Figure thumbnail gr1
      Figure 1Use of specific dyslipidemic agents in adults with SHTG.
      The association between the risk factors and having SHTG compared with not having SHTG (<500 mg/dl) are listed in Table 3. After adjusting for the risk factors listed in Table 3, the factors significantly associated with having SHTG included HDL <40 mg/dl (OR 11.45, 95% CI 6.28 to 20.86), non-HDL 160 to 189 mg/dl (OR 9.74, 95% CI 1.68 to 56.40) or non-HDL ≥190 mg/dl (OR 24.99, 95% CI 3.90 to 160.31), diabetes (OR 3.04, 95% CI 1.45 to 6.37), and chronic renal disease (OR 7.32, 95% CI 1.45 to 36.94).
      Table 3Odds ratios (ORs) and 95% confidence intervals (CIs) of specific risk factors and their association with severe hypertriglyceridemia (SHTG): National Health and Nutrition Examination Survey (NHANES) 2001 to 2006
      Risk FactorsCrude OR (95% CI)Adjusted
      Adjusted ORs included full model: gender, age, ethnicity, total cholesterol, HDL, non-HDL, smoking status, alcohol consumption, C-reactive protein quintiles, physical activity, body mass index, diagnosed diabetes mellitus, myocardial infarction, heart failure, coronary heart disease, hypertension, stroke, angina, chronic renal disease, liver disease, hypothyroidism, levothyroxine, estrogens, and β blockers.
      OR (95% CI)
      Gender
       Female1.01.0
       Male3.26 (1.60–6.64)1.44 (0.74–2.80)
      Age (years)
       20–391.01.0
       40–591.90 (1.06–3.42)1.24 (0.49–3.14)
       ≥600.57 (0.26–1.26)0.50 (0.150–1.68)
      Race
       Non-Hispanic white1.01.0
       Mexican/Hispanic1.34 (0.69–2.57)1.20 (0.53–2.71)
       Non-Hispanic black0.51 (0.28–0.94)0.75 (0.35–1.61)
       Other1.78 (0.67–4.78)1.65 (0.43–6.25)
      Smoking status
       Never1.01.0
       Current2.57 (1.30–5.09)1.55 (0.70–3.44)
       Former2.00 (1.13–3.55)1.74 (0.93–3.24)
      Alcohol consumption
      Referent group was no disease or no treatment.
      1.21 (0.66–2.23)1.24 (0.57–2.71)
      Body mass index (kg/m2)
       Under/normal weight (<25)1.01.0
       Overweight (25–<30)4.49 (1.54–13.10)1.92 (0.63–5.84)
       Obese (≥30)4.53 (1.51–13.65)1.80 (0.56–5.80)
      Total cholesterol (mg/dl)
       <2001.01.0
       200–<2403.56 (1.52–8.32)0.68 (0.21–2.23)
       ≥24023.92 (11.25–50.89)2.90 (0.63–13.36)
      High-density lipoprotein (mg/dl)
       ≥401.01.0
       <4012.06 (6.78–21.44)11.45 (6.28–20.86)
      Non–high-density lipoprotein (mg/dl)
       <1301.01.0
       130–1593.05 (0.50–18.38)2.12 (0.31–14.33)
       160–18912.80 (2.35–69.74)9.74 (1.68–56.40)
       ≥19084.82 (18.47–389.53)24.99 (3.90–160.31)
      C-reactive protein quintile (mg/dl)
       1 (0.01–0.06)1.01.0
       2 (0.07–0.14)1.17 (0.33–4.14)0.51 (0.14–1.87)
       3 (0.15–0.28)2.63 (0.89–7.77)0.90 (0.25–3.23)
       4 (0.29–0.56)2.31 (0.74–7.21)0.51 (0.14–1.89)
       5 (0.57–25.40)1.72 (0.58–5.14)0.33 (0.09–1.15)
      Physical activity (minutes/week)
       <1501.01.0
       ≥1500.74 (0.47–1.17)0.64 (0.40–1.03)
      Diabetes mellitus
      Referent group was no disease or no treatment.
      2.18 (1.09–4.35)3.04 (1.45–6.37)
      Myocardial infarction
      Referent group was no disease or no treatment.
      0.48 (0.13–1.83)0.22 (0.05–0.89)
      Heart failure
      Referent group was no disease or no treatment.
      0.77 (0.20–2.98)0.71 (0.15–3.22)
      Coronary heart disease
      Referent group was no disease or no treatment.
      1.45 (0.52–4.05)3.03 (0.59–15.41)
      Hypertension
      Referent group was no disease or no treatment.
      1.04 (0.63–1.72)0.92 (0.49–1.74)
      Stroke
      Referent group was no disease or no treatment.
      1.06 (0.30–3.68)0.56 (0.12–2.72)
      Angina pectoris
      Referent group was no disease or no treatment.
      0.56 (0.08–3.91)0.35 (0.05–2.61)
      Chronic renal disease
      Referent group was no disease or no treatment.
      4.68 (1.65–13.26)7.32 (1.45–36.94)
      Liver disease
      Referent group was no disease or no treatment.
      2.04 (0.28–15.07)0.57 (0.07–4.60)
      Hypothyroidism
      Referent group was no disease or no treatment.
      1.02 (0.41–2.53)1.84 (0.82–4.13)
      Levothyroxine
      Referent group was no disease or no treatment.
      0.82 (0.24–2.79)0.57 (0.12–2.70)
      Estrogen use
      Referent group was no disease or no treatment.
      0.20 (0.05–0.81)0.79 (0.14–4.48)
      β Blocker use
      Referent group was no disease or no treatment.
      1.39 (0.66–2.89)1.76 (0.71–4.37)
      low asterisk Adjusted ORs included full model: gender, age, ethnicity, total cholesterol, HDL, non-HDL, smoking status, alcohol consumption, C-reactive protein quintiles, physical activity, body mass index, diagnosed diabetes mellitus, myocardial infarction, heart failure, coronary heart disease, hypertension, stroke, angina, chronic renal disease, liver disease, hypothyroidism, levothyroxine, estrogens, and β blockers.
      Referent group was no disease or no treatment.

      Discussion

      Elevated TG levels have been shown to be associated with an increased risk of coronary heart disease and pancreatitis.
      Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and treatment of High Blood cholesterol in Adults (Adult Treatment Panel III) final report.
      Our results have indicated that approximately 3.4 million Americans have TG levels of 500 to 2,000 mg/dl, representing 1.7% of the total population. Moreover, we found that most subjects with SHTG in our sample reported participating in <150 minutes/week of exercise, were overweight or obese, and had a large waist circumference. Adipose tissue is known to be a dynamic endocrine organ, activating a series of immune responses associated with a chronic pro-inflammatory state, as well as insulin resistance and disturbed lipid profiles.
      • John B.J.
      • Irukulla S.
      • Abulafi A.M.
      • Kumar D.
      • Mendall M.A.
      Systematic review: adipose tissue, obesity and gastrointestinal diseases.
      The TG levels can often be managed through lifestyle modification and improvements in body size, diet, and exercise. TG levels have been shown to decrease by approximately 6 mg/dl for every 10 lb (4.5 kg) of weight lost
      • Dattilo A.M.
      • Kris-Etherton P.M.
      Effects of weight reduction on blood lipids and lipoproteins: a meta-analysis.
      and by 5 to 38 mg/dl for every 1,200 to 2,200 kcal expended through brisk walking or jogging (approximately 15 to 20 miles/week).
      • Durstine J.L.
      • Grandjean P.W.
      • Cox C.A.
      • Thompson P.D.
      Lipids, lipoproteins, and exercise.
      The causes of hypertriglyceridemia include obesity, metabolic syndrome or type 2 diabetes, consuming a diet with a positive energy-intake balance, reduced physical activity, renal disease, autoimmune disorders, such as lupus, and the consumption of specific medications (i.e., corticosteroids, some estrogens, tamoxifen, some antihypertensive agents, isotretinoin, bile acid-binding resins, cyclophosphamide, antiretroviral agents, and some psychotropic agents).
      • Yuan G.
      • Al-Shali K.Z.
      • Hegele R.A.
      Hypertriglyceridemia: its etiology, effects and treatment.
      As reported by Ford et al,
      • Ford E.S.
      • Li C.
      • Zhao G.
      • Pearson W.S.
      • Mokdad A.H.
      Hypertriglyceridemia and its pharmacologic treatment among US adults.
      the factors significantly associated statistically with TG levels ≥200 mg/dl in the NHANES 1999 to 2004 included increased age, male gender, white race, high school graduation without additional education, current smoker, alcohol consumption in the previous 12 months, overweight or obese, and diabetes. Using identical definitions, we observed a significantly elevated risk of SHTG among men, those aged 40 to 59 years, current or former smokers, overweight or obese subjects, and those with low HDL, high non-HDL, high total cholesterol, diabetes, chronic renal disease, and liver disease in the univariate logistic regression models. However, the number of factors significantly associated with having SHTG in the multivariate models were reduced to a HDL <40 mg/dl, non-HDL ≥160 to 189 mg/dl, non-HDL ≥190 mg/dl, diabetes, and chronic renal disease. It was difficult to determine whether a biologic or pathophysiologic explanation exists for the reason the risk of SHTG was attenuated for many of the personal characteristics and habits in the multivariate models. The TG levels are statistically correlated and biologically interconnected with many factors, including total cholesterol levels, low-density lipoprotein, and HDL, as well as obesity, diabetes, and hypertension.
      Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and treatment of High Blood cholesterol in Adults (Adult Treatment Panel III) final report.
      This adds to the challenge of identifying independent associations between the risk factors and cardiovascular events. However, it might have resulted from the smaller sample size of patients with SHTG. Nevertheless, our results provide evidence that the well-described and understood relation between elevated TG levels and both diabetes and renal disease remains substantial among subjects with SHTG.
      Only 15.3% of our total sample and 20.0% of the subjects with SHTG reported they were taking ≥1 dyslipidemic agent. Even though a large majority of the subjects included in our analysis reported having health insurance, we suspected that many of the subjects learned of their SHTG for the first time through their participation in NHANES and consequently had not been appropriately medicated for their condition at enrollment. Ford et al
      • Ford E.S.
      • Li C.
      • Zhao G.
      • Pearson W.S.
      • Mokdad A.H.
      Hypertriglyceridemia and its pharmacologic treatment among US adults.
      reported that <4% of the NHANES participants in their sample who had TG levels ≥200 mg/dl were using niacin or a fibric acid (specifically, fenofibrate or gemfibrozil) and >15% were using statins or ezetimibe. Our findings were similar among the subjects with SHTG. The 3 most commonly used dyslipidemic agents reported among subjects with SHTG in our sample of NHANES participants were statins (13.8%), fibric acids (4.0%), and over-the-counter niacin products (2.4%).
      We found that the prevalence of SHTG was considerably lower among women than men. Specifically, we found that <25% of the subjects with SHTG in our sample were women. Similar findings have been reported in other studies.
      • Donahue R.P.
      • Jacobs Jr, D.R.
      • Sidney S.
      • Wagenknecht L.E.
      • Albers J.J.
      • Hulley S.B.
      Distribution of lipoproteins and apolipoproteins in young adults: the CARDIA study.
      • Heiss G.
      • Schonfeld G.
      • Johnson J.L.
      • Heyden S.
      • Hames C.G.
      • Tyroler H.A.
      Black-white differences in plasma levels of apolipoproteins: the Evans County Heart Study.
      • Tyroler H.A.
      • Glueck C.J.
      • Christensen B.
      • Kwiterovich Jr, P.O.
      Plasma high-density lipoprotein cholesterol comparisons in black and white populations: the Lipid Research Clinics Program Prevalence Study.
      Despite this low prevalence of SHTG, no evidence has suggested that women are protected from cardiovascular events. One meta-analysis of 17 prospective studies reported a statistically significant 76% increase in risk of cardiovascular disease for a 1-mmol/L increase in TG levels for women but only a 32% increase for the same increase for men.
      • Hokanson J.E.
      • Austin M.A.
      Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: a meta-analysis of population-based prospective studies.
      Given the cross-sectional design of the present study, the temporal association between each risk factor and the TG levels could not be ascertained. A limitation inherent in NHANES is that certain key measurements are based on self-report and therefore subject to misclassification. Self-reported measurements are subject to a variety of biases, including recall bias.
      • Natarajan S.
      • Lipsitz S.R.
      • Nietert P.J.
      Self-report of high cholesterol: determinants of validity in U.S. adults.
      However, as reported by the Centers for Disease Control, the prevalence estimates for several chronic conditions and health behaviors related to dyslipidemia, such as heart disease, stroke, hypertension, diabetes, and smoking history, were quite comparable between the NHANES data and the national averages (available at: http://www.cdc.gov/DataStatistics/). NHANES has been a valuable resource for answering questions regarding the factors that affect dyslipidemia management in adults. Despite these limitations, our report has included results for subgroups of subjects, such as minorities and the elderly, for whom data on SHTG are not readily available from other sources. An additional limitation of our study was that only 87 participants had SHTG, which resulted in wide CIs in the logistic regression models.

      Acknowledgment

      We have all met the criteria for authorship set forth by the International Committee for Medical Journal Editors. We wish to acknowledge Susan Johnson, MD, and Amy Meadowcroft, PharmD (both employees of GlaxoSmithKline), for their critical review of the overall analysis and presentation of final results during the development of our report.

      References

      1. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and treatment of High Blood cholesterol in Adults (Adult Treatment Panel III) final report.
        Circulation. 2002; 106: 3143-3421
        • Centers for Disease Control
        NHANES 1999–2000 Data Documentation.
        ((Revised 2007: Lab 13AM - Triglycerides and LDL Cholesterol)in: CDC, Atlanta2007: 1-5
        • John B.J.
        • Irukulla S.
        • Abulafi A.M.
        • Kumar D.
        • Mendall M.A.
        Systematic review: adipose tissue, obesity and gastrointestinal diseases.
        Aliment Pharmacol Ther. 2006; 23: 1511-1523
        • Dattilo A.M.
        • Kris-Etherton P.M.
        Effects of weight reduction on blood lipids and lipoproteins: a meta-analysis.
        Am J Clin Nutr. 1992; 56: 320-328
        • Durstine J.L.
        • Grandjean P.W.
        • Cox C.A.
        • Thompson P.D.
        Lipids, lipoproteins, and exercise.
        J Cardiopulm Rehabil. 2002; 22: 385-398
        • Yuan G.
        • Al-Shali K.Z.
        • Hegele R.A.
        Hypertriglyceridemia: its etiology, effects and treatment.
        Can Med Assoc J. 2007; 176: 1113-1120
        • Ford E.S.
        • Li C.
        • Zhao G.
        • Pearson W.S.
        • Mokdad A.H.
        Hypertriglyceridemia and its pharmacologic treatment among US adults.
        Arch Intern Med. 2009; 169: 572-578
        • Donahue R.P.
        • Jacobs Jr, D.R.
        • Sidney S.
        • Wagenknecht L.E.
        • Albers J.J.
        • Hulley S.B.
        Distribution of lipoproteins and apolipoproteins in young adults: the CARDIA study.
        Arteriosclerosis. 1989; 9: 656-664
        • Heiss G.
        • Schonfeld G.
        • Johnson J.L.
        • Heyden S.
        • Hames C.G.
        • Tyroler H.A.
        Black-white differences in plasma levels of apolipoproteins: the Evans County Heart Study.
        Am Heart J. 1984; 108: 807-814
        • Tyroler H.A.
        • Glueck C.J.
        • Christensen B.
        • Kwiterovich Jr, P.O.
        Plasma high-density lipoprotein cholesterol comparisons in black and white populations: the Lipid Research Clinics Program Prevalence Study.
        Circulation. 1980; 62: IV99-IV107
        • Hokanson J.E.
        • Austin M.A.
        Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: a meta-analysis of population-based prospective studies.
        J Cardiovasc Risk. 1996; 3: 213-219
        • Natarajan S.
        • Lipsitz S.R.
        • Nietert P.J.
        Self-report of high cholesterol: determinants of validity in U.S. adults.
        Am J Prev Med. 2002; 23: 13-21