Lies, damned lies, and statistics

18. April 2009 William Davis (8)
In the last Heart Scan Blog post, I discussed the question of whether statin drugs provide incremental benefit when excellent lipid values are already achieved without drugs.

But I admit that I was guilty of oversimplification.

One peculiar phenomenon is that, when plaque-causing small LDL particles are reduced or eliminated and leave relatively benign large LDL particles in their place, conventional calculated LDL overestimates true LDL.

In other words, eliminate wheat from your diet, lose 25 lbs. Small LDL is reduced as a result, leaving large LDL. Now the LDL cholesterol from your doctor's office overestimates the true value.

Anne raised this issue in her comment on the discussion:

I eliminated wheat - and all grains - from my diet nearly three years ago (I eat low carb Paleo). My fish oils give me a total of 1680 mg EPA and DHA per day, and my vitamin D levels since last year have varied between 50 ng/ml and 80 ng/ml. However, my lipid profile is not like either John's or Sam's:

LDL cholesterol 154 mg/dl
HDL cholesterol 93 mg/dl
Triglycerides 36 mg/dl
Total cholesterol 255 mg/dl

My cardiologist and endocrinologist are happy with my profile because they say the ratios are good, no one is asking me to take a statin. My calcium score is 0.


However, if we were to measure LDL, not just calculate it from the miserably inaccurate Friedewald equation, we would likely discover that her true LDL is far lower, certainly <100 mg/dl. (My preferred method is the bull's eye accurate NMR LDL particle number; alternatives include apoprotein B, the main apoprotein on LDL.)

So Anne, don't despair. You are yet another victim of the misleading inaccuracy of standard LDL cholesterol determination, a number that I believe should no longer be used at all, but eliminated. Unfortunately, it would further confuse your poor primary care doctor or cardiologist, who--still believe in the sanctity of LDL cholesterol.

By the way, the so-called "ratios" (i.e., total cholesterol to HDL and the like) are absurd notions of risk. Take weak statistical predictors, manipulate them, and try to squeeze better predictive value out of them. This is no better than suggesting that, since you've installed new brakes on your car, you no longer are at risk for a car accident. It may reduce risk, but there are too many other variables that have nothing to do with your new brakes. Likewise cholesterol ratios.
Tags :

Comments (8) -

  • Jeff

    4/18/2009 12:02:00 PM |

    I am in a similar boat.  I eliminated wheat, added cod liver oil(1-1.5 TBS or so per day), and take vitamin D supplements(gel caps around 4-5K IU per day).  Here were my results:

    TOTAL: 272
    HDL: 76
    LDL(calculated): 184
    Triglycerides: 62

    I strongly suspect the LDL is BS due to large particle size.  I will only be able to tell with the NMR, which I will do sometime this year.  I am not worried so I am in no rush.

  • arnoud

    4/18/2009 1:26:00 PM |

    Dr. Davis, thank you for providing such eye-opening insights in the interpretation of lipid testing results, and for explaining the limits of the usefulness of these measurements.

    However, in the typical doctor's office, the high (calculated) LDL prompts the doctor to push for treatment (including statins).  My doctor tells me that I need not avoid statins, as he is taking those himself, and he wouldn't if he'd think they were a problem.

    I can take my NMR Liposcience LDL particle count result to my doctor, and tell him that my real LDL number is 1/10th of the LDL particle count.  This actual number looks great!  How can I show my doctor that this calculation is correct (LDL particle count divided by 10), and that the standard Framingham calculated LDL should be ignored.   Is there a reference paper I can show my doctor, which explains the science behind the "LDL particle count divided by 10" rule?

  • john elfrank

    4/18/2009 1:35:00 PM |

    I had a coronary calcium scan a  few years ago. My score was about 350 with most of the calcium in the LAD.

    My Manhattan cardiologist responded by putting me on the treadmill and doing an eco stress test. I passed it with flying colors.


    I went to my internist who said  I should be concerned about that calcium score. I said my cardio won't give me any other tests. He said to go back and tell him I have chest pains. I did, got the angiogram and a stent for the 80% blockage in my "widow-maker" LAD.

    Now my lipid profile (I have dyslipidemia) is LDL 23, HDL 23, triglycerides 350 (1000 w/o meds). I had thyroidectomy in 1991 and take synthroid 200 mg.

    My combo thereapy is:
    2000 Niaspan, 40 Simvastatin, 200 Co-enzyme Q10, 1200 fish oil. It'a about as aggressive as my body can stand. Tricor and other fibrates interfere with synthroid absorption (I bet you didn't know that).

    My questions are:
    1. Would it be better for me to take the new combo Simvastatin/Niaspan drug rather than take them separately?

    2. Just passed a nuclear stress test. Should I insist on another angiogram soon?

    3. Would another calcium scan be useful?

    Thanks,

    John

  • sk

    4/18/2009 3:38:00 PM |

    this is absolutely spot on!  My numbers prior to NMR showed a total cholesterol of 150, HDL of 41, and TRG of 53.  Because of family history, my internist had me take NMR study and results showed that my particle number for LDL was 1795 and all small particles.  Since eliminating wheat and being on a statin my particle number is down to 1305,but still all small. Not sure that size can be changed, probably genetic.  
    Sadly, many out there think they have a fine profile from indirect measurement, and reality is that many probably do not.

  • Kiwi

    4/18/2009 10:05:00 PM |

    What is the recommended range for the ApoB test?
    My lab gives this:

    Male reference range 0.52 - 1.09 g/L
    Female reference range 0.49 - 1.03 g/L

    Using the Immunoturbidimetric method.
    VAP and NMR tests not available here.

    http://www.labnet.co.nz/testmanager/index.php?fuseaction=main.DisplayTest&testid=292

  • Dr. B G

    4/20/2009 9:42:00 PM |

    Anne,

    Those are FANTASTIC, phenomenal labs !!!

    You go GIRL!

    -G

  • Dr. B G

    4/20/2009 9:42:00 PM |

    Anne,

    Those are FANTASTIC, phenomenal labs !!!

    You go GIRL!

    -G

  • Ravi

    4/23/2009 9:42:00 AM |

    I strongly suspect the LDL is BS due to large particle size. I will only be able to tell with the NMR, which I will do sometime this year. I am not worried so I am in no rush.

Add comment

Loading
The happy homeotherm

The happy homeotherm

24. November 2010 William Davis (14)
If you were a "cold blooded" poikilotherm unable to regulate internal body temperature, you would have to sun yourself on rocks to raise your body temperature, just like turtles and snakes. When it got cold, your metabolic rate would slow and you might burrow into the mud to hide.

You and I, however, are homeotherms, terrestrial animals able to regulate our own internal body temperature. Principal responsibility for keeping your body temperature regulated falls with the thyroid gland, your very own thermoregulatory "thermostat."

But internal body temperature, even in a homeotherm, varies with circadian rhythm: Highest temperature occurs in the early evening around 8 p.m.; the low temperature nadir occurs at around 4 a.m.

The notion that normal human temperature is 98.6 degrees Fahrenheit is a widely-held fiction, a legacy of the extraordinary experience of 19th century German physician, Carl Reinhold August Wunderlich, who claims to have measured temperatures of one million people using his crude, uncalibrated thermometer to obtain axillary (armpit) body temperatures.

Dr. Broda Barnes was a 20th century American proponent of using the nadir body temperature to gauge thyroid function. Like Wunderlich, Barnes also used axillary temperatures.

Modern temperature assessments have employed radiotransmitting thermistors that are swallowed, with temperatures tracked as the thermistor travels through the stomach, duodenum, small intestine, large intestine, rectum, then peek-a-boos back out. Such internal "core temperature" assessments have shown that:

--Axillary temperatures do not track with internal core temperatures very well, often veering off course due to external factors.
--Axillary temperatures are subject to ambient temperatures, such as room temperature, and are affected by clothing.
--Axillary temperatures are more susceptible to physical activity, e.g., increased with exercise or physical work.

Even right vs. left axillary temperatures have been shown to vary up to 2 degrees Fahrenheit.

Studies such as this demonstrate that normal oral temperature upon arising is around 97.2-97.3 degrees Fahrenheit. While we lack data correlating thyroid function with circadian temperature variation, the a.m. nadir does indeed, as Dr. Barnes originally suggested, seem to track thyroid status quite well: lower with hypothyroidism, higher with normal or hyperthyroidism.

I have been using 97.3 degrees F orally as the cutoff for confirming or uncovering thyroid dysfunction, particularly when symptoms or blood tests (TSH, free T3, free T4) are equivocal, a value that has held up well in the majority of cases. I find it helpful when, for instance, someone complains of cold hands and feet and has normal TSH (1.5 mIU/L or less in my view) but low free T3. An a.m. oral temperature of, say, 95.7 degrees F, suggests that there will be a favorable response to T3 supplementation. And it nearly always plays out that way.

Wouldn't it be interesting to know if there was insight into thyroid status provided by also examining the circadian behavior of temperature (e.g., height or timing of the peak)?
Tags :

Comments (14) -

  • Anonymous

    11/24/2010 9:08:56 PM |

    perhaps a disturbed circadian rhythm results in excess prolactin and that to hypothyroidism... thanks to our modern lighting keeping us up till late at night.

  • Anand Srivastava

    11/24/2010 9:38:59 PM |

    The temperature over the day actually tracks the cortisol level. The T3 is present in the blood, but is taken up by the cells only when cortisol is present. In the morning till 12:00 the cortisol is highest. Before sleeping it dips to its lowest waking value. It will dip further in the night. 2 hours before sleeping it will start to rise. I guess the temperature tracks the cortisol with a delay of a couple of hours. Providing the lowest temperature immediately at waking up.

    There is a good temperature graph method by Dr.Rind to monitor thyroid function and also Adrenal function.
    http://www.drrind.com/therapies/metabolic-temperature-graph

    It seems that thyroid is responsible for defining the body temperature, while adrenal is responsible for varying the temperature as per requirement.

    The test requires taking the temperature 3 times a day using an analogue meter (like the mercury thermometer). The temperature is taken at 3, 6, and 9 hours after waking, and then averaged. Remember to keep the thermometer in the mouth for around 10 minutes and don't do anything. Less will not be sufficient, and movement changes the temperature.

    This test is done for at least 5 days. The average temp will define the thyroid function. If it is too low, it indicates thyroid problem. If the variation day to day is too much then it indicates a cortisol problem.

    I am not sure what the average temperature should be but I think it should be at least 98.2F. The temperature variation has been recommended to be below .3F.
    Mine varies 1F. I have a lot of adrenal insufficiency symptoms.

  • LeonRover

    11/24/2010 9:47:14 PM |

    I am constantly astonished that population studies of common markers such as temperature and blood pressure have such low samples as in the hundreds.

    My mother would take my temperature when I was young and might call our GP if it was over 100 F.

    I had assumed (until recently) that decent statistically large databases, covering 000's of people, divided by gender, age, time of day, diseases etc., from which one might get some interesting were available.

    Alas, there is only Broda Barnes, or studies with 190 in sample.

    It is disturbing.

  • Daniel A. Clinton, RN, BSN

    11/24/2010 11:03:09 PM |

    Dr. Davis,
    I'm a big fan. It's refreshing for me to find people with a deeper understanding than me. Would you treat the hypothetical patient in your post with the normal TSH but low T3 with iodine supplementation before prescribing T3, or jump right to med therapy?
    It boggles my mind that a good chunk of hypothyroidism is simple iodine deficiency, but that truth is hidden from all but those who vigilantly try to seek it out.

  • Lori Miller

    11/25/2010 12:55:41 AM |

    Something that affects temperature in women: the time of the month.

  • Anonymous

    11/25/2010 2:58:00 AM |

    alot of low t3 is just leptin resistance.  Fix the leptin resistance with no sugars and starches, plenty of green veges, fish and a bit of lean meat, and the thyroid starts converting t4 back into t3 and not reverse t3.  I actually believe the t3 supplementation is like giving insulin to a T2DM- it just kills them quicker!

  • Anonymous

    11/25/2010 7:13:17 AM |

    How do you suggest to measure the oral basal body temperature?  30 years ago the suggestion was to use the axillary temp.  Now with the ear digital thermometers that would seem to be easier and more accurate.   We used 97.2 axillary, you suggest 97.3 orally, no problem but how do you suggest to measure

  • qualia

    11/25/2010 12:10:07 PM |

    totally agree with @anonymous - supplementing T3 is an extremely crude and potentially harmful (long term) measure. leptin and physical activity plays a huge role in the conversion of T4 to T3. also iodine, selenium and zinc status. listen to the last podcast by byron richards covering all of these correlations in his "thyroid health class" episode (mp3): http://www.wellnessresources.com/audio/podcast_112410.mp3

  • Dr. William Davis

    11/25/2010 2:20:43 PM |

    Thanks, Daniel. It is always worth trying iodine first; crudely estimated, about 30% of people will respond just to iodine. If that doesn't work, then a T3 preparation or a combination T4/T3 preparation is worth considering.


    Hi, Lori--Yes, indeed. A big effect. I advocate measuring it during the first 7 days after menstrual bleeding starts, the time when temperature is lowest.

    Anonymous--I actually have plenty of slender people (e.g., BMI <23), therefore presumably with normal leptin levels, who still display the low-T3 effect. I believe there is more to this issue.

  • steve

    11/25/2010 4:46:31 PM |

    what actually constitutes a low T3?
    Is it below the range the lab considers normal, with in the range, but at bottom 10% or the 40 to 50 % level of the range?

  • Anonymous

    11/25/2010 8:01:57 PM |

    Hi Dr Davis
    I have read several articles and studies that link lower T3 levels to calori restriction diets and to slower aging.
    Wouldn't T3 supplementation accelerate aging process?

  • Anonymous

    11/26/2010 8:17:45 PM |

    Hi Dr Davis
    sigh!  You are right on most things including about butter being insulinogenic (but I still eat it anyway) but you have not grasped the basic idea yet.

    BMI is such a crude tool.  It does not account for how much is fat and how much is muscle.  Very skinny people are highly leptin resistant (think the osteoporotic grandmother who also has IHD - most from the generation above us are skinny).  That is why BMI in critical care units is inversely related to mortality.  Lean and very muscular people are unlikely to be leptin resistant, though they may have a BMI of 26.  People with low BMI, or marathon runners with stress fractures are more likely to be leptin resistant.  People with BMIs of over 26 are usually  quite leptin resistant though there may be some exceptions to that rule.  We are all leptin/insulin resistant because this is how we age, but most of us are ageing too quickly.  
    Byron Richards, despite being a dreaded naturopath (and recommending wheat!), understands this connection very well:
      
    http://www.byronrichards.com/index.php/thyroid_leptin/entry/obesity_causes_thyroid_problems/

    PS  my child is now off her t4! - we followed a modified rosedale.  Thyroid supplementation, testosterone supplementation, growth hormone treatment and insulin for  T2 do not treat the underlying cause, and will accelerate the ageing process and cause sarcopenia, whether you are fat or skinny.

  • Anonymous

    2/16/2011 5:23:35 PM |

    I have found this website helpful regarding thyroid www.thyroidbook.com

  • Paul

    5/10/2011 8:55:16 PM |

    I am hypothyroid and on T4 T3 combination therapy.  In 2010, I followed Michael Pollan's "Food Rules" less the whole grains: I was weight stable.  However, starting this year (and still following the organic, whole food approach) I also embarked on strict carb restriction.  I lost about 16 pounds and then I hit a plateau.  I also got a tooth abscess and malaise.  My annual thyroid tests showed a TSH of <1, but a T4 at the bottom of the range and a hypothyroid T3.  My T4 T3 have been revised upwards and I am staying with carbohydrate restriction.

    Perhaps this suggests that hypothyroid patients need to be cautioned when embarking on carbohydrate restriction. Or, perhaps I am a one-off.  Also, are (under-treated) hypothyroid people susceptible to carb diets as such diets may effectively reduce the extent of their thyroid deficiency?

Add comment

Loading