best to check nonfasting lipids

A recent article confirms several other studies suggesting that checking nonfasting lipid levels was as at least as good as getting fasting ones; this study compared results of fasting vs nonfasting from the same individual, looking at clinical outcomes (see lipids nonfasting ASCOT jamaintmed2019 in dropbox, or doi:10.1001/jamainternmed.2019.0392).

Details:

-- posthoc perspective follow-up of 8270 participants in the Anglo-Scandinavian Cardiac Outcomes Trial-Lipid Lowering Arm (ASCOT-LLA), comparing nonfasting and fasting lipid levels measured 4 weeks apart in each individual

-- Includes both those with cardiovascular disease and those with no prior history

-- 82% male, mean age 63, current smoker 35%, BMI 28, blood pressure 165/95, heart rate 71, fasting glucose 115, creatinine 1.1

-- prior stroke 12%, previous peripheral vascular disease 6%, diabetes 30%, family history of premature CHD 23%, aspirin 20%

-- median follow-up 3.3 years: ASCOT-LLA trial terminated prematurely because of finding a highly significant relative risk reduction for the primary coronary endpoint achieved by atorvastatin 10 mg/d vs placebo

-- data collected from 1998-2003, analyzed in 2016-2019

-- primary outcome: major coronary events (nonfatal MI and fatal CHD), 212 events

-- secondary analyses: ASCVD events (including MI, stroke, and ASCVD deaths), 351 events

-- lipid profiles were slightly different between fasting and nonfasting, with modestly higher triglyceride levels but negligible differences in LDL, HDL, and total cholesterol levels: total cholesterol was about 3 mg/dL higher nonfasting, LDL about 8 mg/dL higher fasting, HDL about 2 mg/dL higher fasting, triglycerides 35 mg/dL higher nonfasting. Ratio of total to HDL cholesterol about 0.05 higher nonfasting [ie, insignificant differences in LDL and HDL and ratio of chol/HDL, and modestly increased triglycerides nonfasting]

-- those who had a coronary event were older (66 vs 63 years) more likely to have a history of stroke or TIA (14% vs 10%) or diabetes (36% vs 26%), and had more cardiovascular risk factors (median of 4 vs 3)

Results:

-- overall: extremely similar numbers of clinical events comparing fasting and nonfasting lipids, with more coronary events tracking with LDL cholesterol, non-HDL cholesterol, and the ratio total to HDL cholesterol. Triglycerides did not matter at all. And there was a strong trend for about a 35% protective effect for HDL (fasting: HR 0.59 (0.34-1.01), nonfasting 0.69 (0.40-1.18))

    -- also very similar results in those with no prior history of vascular disease (primary prevention group)

-- per 40 mg/dL change of LDL:

    -- nonfasting lipids: adjusted HR 1.32 (1.08-1.61), p=0.007, adjusted for age, sex, race, smoking, diabetes, BMI, and what treatments were done in ASCOT trial for hypertension (atenolol or amlodipine) or cholesterol (atrovastatin or placebo)

    -- fasting lipids: aHR 1.28 (1.07-1.55), p=0.008

-- in primary prevention group, per 40 mg/dL change of LDL

    -- nonfasting lipids: aHR 1.42 (1.13-1.78), p=0.003

    -- fasting lipids: aHR 1.37 (1.11-1.69), p=0.003

-- results for major coronary events were consistent when the patient was randomized to atorvastatin 10 mg/d or placebo, and similar for ASCVD events

-- concordance of fasting and nonfasting lipids for classifying participants into appropriate ASCVD risk categories was 94.8% as calculated by the 2013 ACC/AHA risk calculator, 98.6% per the QRISK2 algorithm

    -- and, no differences in reclassification from high risk to low risk categories between fasting and nonfasting lipids for the ACC/AHA guidelines (1.41% vs 1.40%) or QRISK2 (0.42% vs 0.45%)

Commentary:

-- remarkably similar clinical reductions utilizing fasting vs nonfasting lipids overall, including those in the the exclusively primary prevention group

-- one large advantage of the study is that it used individual level data from the same patient. It is always a bit of a trick to interpret/apply community data from large cohorts and to assume they apply to an individual patient (so-called “ecological fallacy”). So, this study mitigates this issue by comparing results of nonfasting and fasting clinical endpoints for the same patients, and is therefore more directly applicable to clinical practice

-- this study suggest several reasonable conclusions:

    -- nonfasting lipid assessment is at least as good as fasting lipids in evaluating cardiovascular risk, through 2 validated risk assessment instruments, and with essentially no individual risk reclassification changes dictated by clinical outcomes by using either nonfasting or fasting samples

    -- this study confirms that individual risk assessment leading to statin therapy initiation is appropriate using non-fasting samples

    -- risk assessment was robust in both the total group as well as in the exclusively primary prevention cohort

-- There were 2 studies finding that nonfasting triglycerides were a strong predictor of cardiovascular events, as noted in http://gmodestmedblogs.blogspot.com/2016/05/fasting-lipids-not-so-fast.html  . And other studies have found that nonfasting lipids may have a stronger association with cardiovascular outcomes, especially the nonfasting triglycerides levels

-- there are several disadvantages to fasting lipid measurements:

    -- it may require patients to come back to the laboratory at a different time, likely decreasing the likelihood of getting the blood test done and determining an important clinical assessment

    -- this is particularly an issue for older patients, where the trip to the health center when fasting may be more difficult, and they may experience weakness as a result of not eating

    -- and there was a study in diabetics finding that 1 in 4 had symptoms of hypoglycemia en-route to getting fasting blood tests

    -- the more rapid results from getting the blood test at the time of the visit could conceivably have cardiovascular benefit, given that several studies have shown statin benefit within 6 months of initiation

-- even before the study, many medical societies have supported nonfasting lipid assessments: the 2016 and 2018 European Atherosclerosis Society and the European Federation of Laboratory Medicine Consensus Statements recommended nonfasting lipids for routine vascular care, the Canadian guidelines of 2016 similarly suggested nonfasting lipid assessments, and more recently the 2018 ACC/AHA guidelines did so as well

-- limitations of the ASCOT-LLA study include the fact that few nonwhite individuals were assessed, and the role of lipids in the development of cardiovascular disease does have some ethnic variability

--but, i should add that some authors suggest that fasting triglycerides should be checked when concerned about triglyceride-induced pancreatitis, but I could not find studies to justify this conclusion: it seems to me to be more physiologically likely that pancreatitis is associated with disturbed lipid metabolism, so that a high lipid meal would lead to more prolonged elevation of the triglyceride level (larger area under the curve) throughout the day, though one concern is that there is no standardized meal to assess nonfasting triglyceride levels. but this is all still pretty controversial (European guidelines use nonfasting, the Am Heart Assn suggests getting repeat fasting TG if the nonfasting is >400 (was 200, but most recently 400). though if really high nonfasting triglycerides (eg >1000), no need to repeat with fasting: see http://gmodestmedblogs.blogspot.com/2018/11/new-cholesterol-guidelines-2018.html, or https://www.heart.org/-/media/files/health-topics/cholesterol/chlstrmngmntgd_181110.pdf].  However, there was a somewhat intriguing article looking at nonfasting triglycerides in the Copenhagen City Heart Study, following 117K people and finding an increased association of nonfasting triglycerides even at 167-265 mg/dL being associated with >2-fold risk of pancreatitis, increasing to >8-fold in those with nonfasting triglycerides >443 mg/dL, higher risk than for MI (see triglycerides pancreatitis nonfasting low levels jamaintmed2016 in dropbox, or doi:10.1001/jamainternmed.2016.6875)

so, another nail in the fasting lipid coffin...  this study adds much better evidence that nonfasting lipid assessments are reasonable and appropriate for us to use, based on actual clinical outcomes in individual patients comparing their own fasting vs nonfasting lipid values.  to me, there is no reason to get fasting lipids for cardiovascular risk assessment/treatment, assuming that the lab can get direct LDL measurements if the triglycerides are too high to calculate the LDL

see http://gmodestmedblogs.blogspot.com/2016/05/fasting-lipids-not-so-fast.html  which reviews 3 prior blogs on fasting lipids, including the European Atherosclerosis Society consensus statement suggesting routine nonfasting lipids. This also includes a 2008 article by the same lead author as the current one, also suggesting at least as good a cardiac risk factor assessment by using nonfasting lipids

geoff

If you would like to be on the regular email list for upcoming blogs, please contact me at  gmodest@uphams.org

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/
 to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list