Statins and CAC: What is This Paper, Really?

Edition: 2019-06-02

This forum article is about this JACC paper published 2018-09-25:
Impact of Statins on Cardiovascular Outcomes Following Coronary Artery Calcium Scoring
As published, the paper is pay-walled at Science Direct.

If the pay-wall was intentional (which would have the effect of discouraging public debate), it wasn’t completely successful, because Dr. Jeffery Dach has made the PDF available on his site.

Jeffrey also blogged about it at:
Calcium Score Determines Who to Treat With Statin Drug
He remains a statin skeptic after reading: “No doubt this new methodology will improve the NNT (Number Needed to treat) for the statin drug. However, one must consider the adverse effects of long term statin use, and ask if the trade off is worth the reduction in quality of life? We await further research to answer this question.“

A Google search also turned up the Appendix. I have not yet turned up the “Extras (1) Online Data” shown on the SD page. And yes, a dive that deep is necessary.

Someone with at least partial access to JACC also shared with me some of the letters to JACC (but not the author responses to the criticisms).

TL;DR? My Bullets : My Perspective : My Conjectures

I’ll “quote, with emphasis” from the paper as I go, adding remarks. Although only a 10-page PDF, the published periodical pages numbers run from p3233 to 3242.

p3233 ABSTRACT: “However, the relative impact of statins on ASCVD outcomes stratified by CAC scores is unknown.”
And unless the Extras show up and say something, it remains unknown, no thanks to this paper.

p3233 footnote: “Dr. Villines has received an honorarium from Boehringer Ingelheim.”
They make a statin. After reading the paper, I think I see another honorarium in his future.

p3234: “The authors identified consecutive subjects without pre-existing ASCVD or malignancy who underwent CAC scoring from 2002 to 2009 at Walter Reed Army Medical Center.”
So these were “active duty military, retirees, and other Department of Defense health care beneficiaries and their dependents.”, on standard diets for that population, easily the disastrous Solder Systems Natick diets (which obeys USDA-DGA). We can at least say that the data are valid for people subject to that (if not subject to the exclusion criteria).

p3234: “Coronary artery calcium (CAC) scoring, a non-invasive measure of coronary artery atherosclerotic plaque burden, improves the accuracy of contemporary risk scores for predicting ASCVD outcomes, and has been suggested as a means to optimize patient selection for statin therapy.”
That might be an agenda indicator there.

p3234: “Randomized controlled trials assessing CAC guided prevention in a broad screening population have not been performed, likely due to concerns over trial size, costs, and the inherent difficulty establishing equipoise to withhold statins from patients at high risk for cardiovascular events due to a significantly elevated CAC score.”
Translation: The RCT we’d like to see will never be run. IRB would consider a non-statin arm to be unethical today. In some saner future, the reverse would be the case.

The First Elephant in the Room (a Dead One)

p3235: “Death data, including cause of death, were extracted for all patients from the MDR and National Death Index and cross-referenced to the Veterans Affairs Beneficiary Identification Records Locator Subsystem as well as the Social Security Death Index.”
But other than one number, death data was not included, not reported even by arm, nor commented on.
Why?

If you contemplate the screening criteria, and Table 1, you can see the challenges they faced. There’s a fair amount of detail on how they attempted to account for this tsunami of confounding factors and compliance considerations.

Might they have crossed the border from de-confounding into obfuscation? Another statin skeptic, Angela A. Stanton PhD, apparently wrote the JACC a [published] letter, but it’s behind a pay-wall too: Confusing Data Analysis, and there’s nothing on her web site. It ended with “At the end, the authors concluded the opposite of what their numbers actually show.” Unfortunately, just what data Stanton was relying on was as just as opaque as the paper under discussion.

p3236: “After applying exclusion criteria, there were 13,644 consecutive patients…”
p3237: “There were 209 deaths (1.5%) from any cause.”
And that’s all we hear about ACM in this paper, or the Appendix.

So the absolute difference in ACM between the arms cannot exceed 1.53%. If all the deaths happened in the untreated arm, I suspect we’d have heard. This leaves us wondering how it did play out, and why it wasn’t reported. Usually there is some mention, at least “no statistical difference.” Here, just silence.

For ACM, cause of death would also matter. It would be useful to see comparative CoD outcomes for things like cancer, diabetes complications, cardiomyopathies, accidental trauma (esp. falls), suicide. This data was available, but not included or reported (more below).

There are also non-fatal outcomes to consider, such as dementia, depression, myalgia, and Quality of Life scores. This data was likewise probably available, but not included or reported. Nor can we infer ACM from the 1.53%, because subjects entered and left the study population over time.

The Next Elephant in the Room

The charts and graphs focus on aSHR (adjusted sub-hazard ratio), which I presume is necessary to turn this paltry percentage into various visible differences, much as the pernicious RR (relative risk) statistical gambit does.

Despite what may be aSHR grade inflation, the data they did chart and graph are quite clear that even restricted to MACE, statins have:
=0: zero MACE benefit at CAC=0,
and what may be
≈0: nil MACE benefit at CAC 1…100.

p3237: “In the 10-year NNT analysis, there was no significant effect of statins among patients without any CAC. Patients with a CAC of 1 to 100 had a trend toward benefit”
Only people who [can, and actually do] read this paper are going to see that key admission, which is in there more than once. People seeing only the released charts might figure it out (I did), but your doctor probably won’t.
Why wasn’t it the subtitle?

p3238: “There was no observed benefit for statins in patients with elevated CAC (101+) and high ASCVD risk (n=185), but the relatively low number of patients in the non-statin group limited the analysis (23 not on statins, 162 on statins). On the basis of the overall trend of the data, it does appear that this group is likely an outlier.”
Translation: this data doesn’t fit our bias. Must be mistaken.

p3239: “Our study helps provide valuable information on the effect of statin therapy in a real-world population without known ASCVD who underwent CAC scoring.”
No kidding, and the statin pushers who actually read it, and understand it, are not going to be terribly happy about that, even with the spin in the paper. So they spun it as best they could.

p3240: “STUDY LIMITATIONS. Given the retrospective design, patients were not pre-assigned to statin therapy.”
Indeed. Given the propensity for consensus MDs to push statins (and for .mil folks to follow orders), just how did these individuals escape that, or are they "statin deniers", and if so, what else are they doing differently in life? Goes to confounding, your honor.

p3240: “Because patients were included in the statin treatment group if they received a prescription in the initial interval following CAC screening, their post-baseline assignment introduces some artifact into the cumulative incidence function curve. Reassuringly, results were consistent in our sensitivity analysis using 2 years instead of 5 years as the cutoff for statin assignment.”
Reassuring. That’s the sort of sentiment that I’d expect to hear from a team looking for, rather than looking at.

p3240: “Although all deaths and their causes were ascertained using the National Death Index, it is possible that some deaths may have been misclassified (cardiovascular vs. noncardiovascular).”
Yep. We had the outcome data. We looked at it. Nope, we aren’t sharing.

p3240: “As in any prevention study, we cannot rule out a healthy user bias (40), whereby a patient that is more likely to receive preventive therapy may also be more likely to engage in other healthy activities that reduce their chance for MACE, such as exercise or a healthy diet.”
That sword has two edges. Note they are assuming that the statin arm is the users with health bias.

The Third Elephant in the Room

A JACC editorial regarding this article, by Khurram Nasir MD, MPH, MSc, leads off with a 2015 quote from Harlan Krumholz: This Test Might Tell You If You Don’t Need A Statin

The editorial is also pay-walled, but party visible here:
Message for 2018 Cholesterol Management Guidelines Update
Time to Accept the Power of Zero
This was another blissfully ignored subtext topic in the paper. Khurram points out that the study results strongly support reversing an AHA reclassification of CAC test from IIa to IIb, an error they made in 2013.
Class IIa: Weight of evidence/opinion is in favor of usefulness/efficacy.
Class IIb: Usefulness/efficacy is less well established by evidence/opinion.

It would appear that the ACC is not interested in actually improving cardiac health, or in having public discussion of their failed policies. Statin-pushing continues apace, however.

Main Critique Points

• You’ve been MACEd
  Dead silent on all-cause mortality, much less other adverse outcomes. Why?
• Statins useless even for MACE at low CAC
  This should have been the paper subtitle, and wasn’t.
• CAC useful for mis-treatment avoidance
  Get a CAC scan. If you have a low or zero score, don’t even consider having statins (or PCSK9 inhibitors) inflicted upon you, for all you will get is side effects.
• Stealth {possibly pseudo} science strikes again
  Where consensus medicine and nutrition are incorrect, real data, and public debate thereof, are not their friends. One of their clumsy tools for delaying the truth is hiding their junk science behind pay-walls. Is that what’s going on here?

My Perspective

If you are a .mil, active or retired, or a dependent, and not on a dissident diet, and are more concerned with MACE than ACM, this paper’s data might give you a tiny advantage.

But for someone following the Undoctored / 2014+ Wheat Belly program, and having a CAC score under 100, get even more skeptical about statins (my posture on those).

If you have a score over 100, there’s no way to discover just what (if anything) the data say. This paper doesn’t even include any theory of operation for the claimed benefit, making it impossible to assess whether there are alternatives that are even more effective (and my bet is that there are).

My Conjectures

People are getting very skeptical about statins. There are tens of billions of dollars at stake on that, and there is also potential liability lurking. From the perspective of Big Pharma, and the Hoodwinked Medical Guilds, the house of cards must be made to stand for as long as possible, by almost any means necessary. The moneyed interests have an unlimited budget for hiring lab-coated shills for this holding action.

The merit of the CAC scan as a risk score, over wild guesses like LDL-C and FRS, is evident to anyone who even glances at it. The medical guilds have only recently partially relented on their multi-decade resistance to CAC. People are getting CAC scans on their own. Consensus med has no (zero) effective advice for preventing score growth, much less slowing, arresting and reversing it. They desperately need a story for patients losing patience with the unpromising prognosis under Standard of Care today.

If Big Pharma™ could get Big Med™ to conjure up some fake news about statins being beneficial in CAC scores: hey, double prizes. If the full data set doesn’t really support the sales pitch, it would of course be important to not allow the actual facts to be too closely examined outside the guild (and guild members can be counted on to not read their own journals). And do try to keep a lid on the actual agenda when authoring the novella.
___________
Bob Niland [disclosures] [topics] [abbreviations]