Background

Atherothrombotic Disease (ATD) is the most common cause of morbidity and mortality in the western world.  Many attempts have been made to attempt to predict the population at risk of ATD, with the aim of treating at-risk patients and thereby in theory preventing the onset of clinical ATD.  The current predictive tools either do not predict well, or they predict well in the short term, but not in the long term.  The purpose of this paper is to present a means of ATD risk prediction based on the risk factor profile of patients who developed some form of clinical ATD during the Bowling Green Study (BGS) time frame of 4 November 1974 to 1 January 2023.  This approach will treat the various risk factors as interdependent factors as advocated by the Framingham Heart Study.

Methods and Materials

The findings in this paper are based on a chart review of all patients known by the author to have developed some form of clinical ATD in the BGS timeframe.  The risk factors to be examined are cigarette smoking, dyslipidemia, and hypertension. (Other risk factor data is available but not needed to define the populations at risk of ATD.)  The BGS involves patient of both genders and all ages. Full risk factor data is available for 907 BGS patients, with an additional 47 patients whose lipids were treated prior to entry to the BGS. The lipid data is comprised of low-density lipoprotein cholesterol (LDL-c) and high-density lipoprotein cholesterol (HDL-c), combined together in a ratio.  Cigarette smoking status is determined by direct questioning.  Blood pressure is determined by trained office staff.

Results

Examination of the BGS database reveals that ATD risk factors are rarely present as single entities in patients with ATD.  The BGS predictive tool is therefore a graph with the Cholesterol Retention Fraction (CRF, defined as [LDL-c minus HDL-c]/LDL-c) on the ordinant and systolic blood pressure (SBP) on the abscissa.  The BGS has developed a threshold line with CRF-SBP co-ordinates of (0.74, 100) and (0.49, 140).  A CRF-SBP plot above this threshold line provides a first order approximation of ATD risk.  This graph may in turn be subdivided into quadrants with a horizontal line at CRF = 0.70 and a vertical line at SBP = 140.  Using this approach one can include patients whose lipids and/or SBP have been treated.  The average age of ATD onset of the patients in the various quadrants can then be calculated to determine a second order approximation of ATD risk.  A third order of ATD risk cn be gained by subdividing the BGS graph into sextiles of CRF versus SBP.  The lower the average age of ATD onset, the higher the near-term risk of ATD. Reverting back to the general population, one can create a table of all of the CRF  data, stratified by age and gender, and stratified by presence/absence of clinical ATD. One can then generate revere Kaplan-Meier curves (time to event) to determine when to begin treatment, depending upon the ATD risk level one wishes to accept prior to initiating therapy. In the general population, serial lipid testing can inform the treating physician of any changes in ATD risk based on changes from baseline values. Finally, the author has in his physical possession the databases of eight angiographic regression trials. Examining the end-of-trial lipids in terms of angiographic outcome aids in determining the amount of lipid lowering that must be obtained to maximize plaque stabilization/regression.

Conclusions

The population at risk of ATD can be readily identified on the basis of the data in the BGS.  Early and repeated lipid testing, combined with cessation of cigarette smoking (most important) and control of SBP offers the opportunity to prevent ATD, at least until very late in life.

Atherothrombotic Disease (ATD) is the leading cause of morbidity and mortality in the western world. [1,2]  ATD refers to atherosclerotic disease, with emphasis on the thrombosis that so often causes the acute clinical event, such as acute myocardial infarction or acute cerebral infarction.   Based on the concept of “An ounce of prevention is worth a pound of cure,” and over 50 years of personal experience in the field of preventive cardiology and interventional lipidology (Harvey Hecht, MD, used with permission), it is this author’s opinion the best means of preventing ATD is the determination of who is at risk of early and middle-aged ATD. It is the common experience of this author and others, there is “no safe place” to totally prevent ATD. In other words, if one lives long enough and does not die of something else, he/she will eventually experience an ATD event.  Thus the goal should be to identify those people at risk of ATD in the early and middle-aged years of life, treating them to prevent plaque formation, and to identify older at-risk patients (aged 75 years or older), treating them in the hope of delaying their ATD events till very late in the lifespan. Unless the physician wishes to treat everyone (herd approach), the author accepts as true that a physician can not protect his/her patient unless that physician can predict which patients are at risk of ATD and moreover, that the better that a physician can predict the population at risk of ATD, the better  he/she can protect that patient from ATD.  Such an attempt requires accurate tools with which to make the prediction, but it also requires that testing be done early in life and be repeated at intervals to determine if the prediction has changed over time.  This has been referred to as primordial prevention. [3] Prediction of the population at risk of ATD has been attempted by various groups over the last 50+ years, beginning with the Framingham Heart Study (FHS).  The FHS has published numerous papers on this topic and has identified three main ATD risk factors (cigarette smoking, dyslipidemia, and hypertension) with some contribution by the very high blood sugar levels of uncontrolled diabetes. [4-8]. The FHS has specifically created a Framingham Risk Score (FRS) to predict those patients at risk of ATD. [9] In the FHS, the lipid predictor that best predicts risk is total cholesterol (CT) divided by high-density lipoprotein cholesterol    (HDL-c) [10-11]; however in the FRS, because of the promise of greater statistical power, the CT:HDL-c  ratio has been discarded and CT and HDL-c are utilized as independent factors. (Peter Wilson, MD, personal communication).  Other studies have built on the FHS findings, specifically the Atherosclerosis Risk in the Community and the Multi-Ethnic Study of Atherosclerosis, and some other  studies have attempted to predict the population at risk of ATD. These latter predictors include the Pooled Cohort Equations (PCE) [12], SCORE2 [13], QRISK [14], Interheart [15], and the Sheffield Tables [16].  Most recently, the American Heart Association has introduced the Predicting Risk of cardiovascular disease EVENTs (PREVENT) [17] to try to improve ATD risk prediction. There are many other predictors, but they are not used in the daily practice of medicine.  Of the cited studies, the FRS, PCE, and PREVENT use CT and HDL-c as their lipid predictors and as independent risk factors.  QRISK and the Sheffield Tables use the CT:HDL-c ratio as their lipid predictors.  SCORE2 uses (CT  minus HDL-c, termed non-HDL-c cholesterol) and Interheart uses a ratio between lipoproteins apo B and apo A-1).  Reports have been published concerning ATD events in patients without the standard ATD risk factors. [18-20]  And of course, in 2017, the president of the American Heart Association sustained an acute myocardial infarction while delivering  lecture at the annual scientific assembly.  Concurrently, recent studies have shown that the above noted predictors are indeed faulty.  Sud found that neither the FRS nor the PCE predicted ATD events very well over the long term, though they were fairly accurate very the short term. [21] Saleh has suggested that coronary artery calcium scoring be added to the PCE to improve its predictive ability. [22] And finally Nurmohamed has proposed adding coronary artery CAT scans to the predictive tool, again to enhance its accuracy. [23]  The problem with these latter two approaches that they identify people who already have disease, perhaps even significant ATD. In light of the above, there is a need to enhance the prediction of the population at risk of ATD.  It will be the purpose of this paper to present a novel predictive tool, based almost entirely on the precepts of the FHS and to present a program based on this risk assessment that may well lead to the prevention of clinical ATD, at least until very late in life.

Materials and Methods

The Bowling Green Study (BGS) of the Primary and Secondary Prevention of Atherothrombotic Disease was conceived in 1968 while the author was in medical school at Ohio State University and doing a cardiology rotation.  This provided the initial background for the BGS and the first tentative steps in the investigation that led to the BGS predictive tool were undertaken while the author served on active duty in the United States Air Force (1971-1974) at RAF Lakenheath Hospital in England.  The major input to the BGS began in November of 1974 when the author set up his practice of family medicine in Bowling Green, the county seat of Wood County in northwest Ohio.  The major portion of Wood County is rural in nature, the population of Wood County at the last census was 132,248 people, the vast majority being of northern European descent.  The chief minority is people of Latin American descent.  The population of Bowling Green at last census was 28,500, with 66?ing students at Bowling Green State University, which is the chief employer in Wood County.  Multiple small industries exist. When the author set his practice in 1974, he provided medical care for people of both genders and of all ages, thus giving him a typical patient cohort for Wood County.  The author’s main purpose, besides providing good medical care for his patients, was to attempt to predict the population at risk of ATD.  In this he followed the precepts of the FHS.  The effort to collect enough data to make such a prediction was not a randomized controlled clinical trial.  There were no placebo groups.  Everyone who met the available criteria for lipid management, according to the FHS, was offered treatment.  Some accepted therapy, but many did not.  In the mid-1970’s, the existing medical community was not interested in preventive medicine, as least as regards dyslipidemia, and even treatment for hypertension was not universal.  Cigarette smoking was widespread, at least in the author’s male patients. Additionally, in the mid-1970’s, the medications used to treat either dyslipidemia or hypertension (or diabetes, for that matter) were not very efficacious.  AS a result, many of the author’s patients sustained ATD events.  Meanwhile, the author continued to collect ATD risk factor data. ATD risk factor data collected consisted initially of the patient’s blood pressure and height/weight (converted into Body Mass Index).  This was collected on each and every patient presenting to the author’s office.  At times, the author was also able to collect fasting lipid data: CT and triglycerides (TG) until 1 January 1978, when HDL-c data became available and was incorporated into the lipid panel, along with a calculation of the low-density lipoprotein cholesterol (LDL-c) following the Friedewald formula.  The Friedewald formula is as follows: LDL-c = CT minus HDL-c minus TG/5. (24)  HDL-c was measured by the precipitation method. The measurement of a lipid panel was done on the basis of family history of dyslipidemia and/or ATD, and a personal history of dyslipidemia and/or ATD, as well as the presence of any known ATD risk factors such as hypertension or diabetes—and indeed for any reason at all. Moreover, a determination of blood glucose performed precisely two hours after a 100 gm carbohydrate meal as suggested by the Wood County Hospital dietician at the author’s request.  (A copy of the protocol  is presented in the Appendix I.)  Finally, in the summer of 1981, when the author recognized the paramount importance of cigarette in the pathogenesis of ATD, the author began actively accumulating tobacco use data for everyone presenting to his office aged 15 years or older.  And after reading surveys in 1991 that cigarette smoking was occurring in adolescents aged 10-14 years, smoking surveys were done on everyone aged 10 years or older.

All of this data was collected into a general population (Gen Pop) database.  As noted previously, ATD risk factor analysis, at the time of the BGS initiation, was not precise and medications to treat dyslipidemia, hypertension, and diabetes were not very efficacious.  Moreover, cigarette smoking was rampart in Wood County and many people did not wish to pursue a course of preventive medications.  Additionally, the author received no support from the established medical community.  Likely as a result, many of the author’s patients developed some form of clinical ATD. By 1981, the author was able to separate out those people who had developed some form of clinical ATD from the Gen Pop and place them into an ATD data base (ATD Pop). Inspection of the ATD Pop database revealed that while interpreting cigarette smoking, blood pressure and blood sugar data was straight forward, the interpretation of the lipid data was less clear.  Some people developed ATD with high levels of LDL-c, but others did so with low levels of LDL-c.  Moreover, some people with low levels of HDL-c developed ATD, but others did so with high levels of HDL-c.  But then in 1981 and article was published asking the question:”Is the LDL:HDL Ratio the Best Lipid Predictor.” (The author regrets that this article is lost to him and hence he can not give proper attribution.)  The author subsequently inspected the ATD Pop database and found that when people with low levels of LDL-c developed ATD, their associated HDL-c  tended to be very low.  Furthermore, when people with high HDL-c  levels developed ATD, their associated LDL-c tended to be very high.  When these were not the cases, the younger patients were virtually always cigarette smokers, whereas older ATD patients usually had hypertension with/without diabetes.  Consequently, in 1981, the BGS began to use the LDL-c:HDL-c ratio as its preferred lipid predictor. In 1983, it occurred to the author that what he really wanted to know was the amount of cholesterol accumulating within the artery wall. This, he reasoned, was best estimated by the amount of cholesterol entering the artery wall (LDL-c) minus the amount of cholesterol exiting the artery wall via reverse cholesterol transport (HDL-c), that difference being divided by LDL-c, to give an estimate of the percentage of the LDL-c being retained within the arterial wall.  This fraction the author termed the Cholesterol Retention Fraction (CRF, defined as [LDL-c minus HDL-c]/LDL-c).  The CRF is abnormal at 0.70 or higher. After a conversation with Roger Illingworth, MD, in the early 1980’s, in which Illingworth pointed out that reverse cholesterol transport could not compensate for very large influxes of LDL-c, the author returned to the ATD Pop database and noted that once LDL-c exceeded 169 mg/dl (4.4 mmoles/L), ATD events could occur despite an optimal level of CRF, although the lower the CRF, the later in life did the ATD event occur.  The author therefore considers a state of dyslipidemia to exist whenever the CRF is 0.70 or higher and/or LDL-c is 170 mg/dl (4.4 mmoles/L) or higher.  This scenario is termed the Cholesterol Threshold (CThr).

Caveat

All scientific hypotheses must be based on factual observations.  The observations so far discussed and those yet to come were all based on the measurement of HDL-c and calculation of LDL-c by the Friedewald formula, as described earlier in this paper.  Until the spring of 1999, HDL-c was measured by the precipitation method (Precip) in which fasting serum is infused with either heparin manganese or phosphotungstic acid, both of which precipitate out all of the lipoproteins except HDL. (Other precipitants have been used. Different methodologies give differing HDL-c measurements, though not usually much different.)  In the spring of 1999, at least in the author’s area of practice, the methodology of HDL-c measurement was changed to the enzymatic method (Enz),with no advanced warning to the medical community.  These two differing methodologies (Precip versus Enz) give differing results, but this time on the order of 10 mg/dl (0.25 mmoles/L). The Enz method gives the higher value and the Precip method gives the lower value.  Consequently, the LDL-c level calculated by the Friedewald formula will be on the order of 10 mg/dl (0.25 mmoles/L) lower for the Enz method than for the Precip method.  A lipid ratio, such as used in the BGS, will, as a result, be significantly altered, dependent upon the methodology by which HDL-c is measured. This is finding is not inconsequential.  In 2008, the author reported the case of one of his patients who sustained an acute myocardial infarction at age 53 years. [25]  The author had provided medical care for him over the years, almost always relating to employment issues.  Neither the patient nor his wife had  ever smoked cigarettes. He was not hypertensive or known to be diabetic.  He was slim, not obese.  He had no personal or family history of dyslipidemia, diabetes, or ATD. Hence the author had never tested his lipids.  One day in 2008, the patient sustained an acute myocardial infarction while in Toledo, Ohio.  He went to a local hospital to which he was admitted.  A lipid panel was performed on admission and was mildly abnormal.  The Enz method of HDL-c measurement was utilized at that hospital.  When he returned to the author’s care, the author was able to obtain a copy of the lipid panel results.  Using the formula provided by Wood County Hospital, the lipid panel was converted from the Enz methodology to its Precip equivalent and now the lipid panel was much more abnormal--especially the CRF—and the patient sustained his myocardial infarction when it would have been predicted by the Precip method.  The formula for conversion of Enz HDL-c to Pecip HDL-c  (using the auto-analyzer at Wood County Hospital) is as follows:  HDL-c (Precip) = [HDL-c (Enz) minus 12]/0.93. Since the vast majority of the author’s lipid panels utilized the Precip methodology and since the angiographic regression studies, to be discussed later in this paper, were performed using the Precip methodology, the author has elected to continue using the Precip data, converting the Enz based lipids to their Precip based equivalents.  The Precip based lipid equivalents can be roughly approximated by subtracting 10 mg/dl (0.25 mmoles/L) from HDL-c (Enz), adding 10 mg/dl (0.25 mmoles/L) to LDL-c (Enz), and subtracting 0.1 from the CRF.

As noted previously, the author inspected his ATD database in the late 1970’s and early 1980’s, once enough patients had sustained enough ATD events.  (See previous comments about the suboptimal risk predictors and medications to control the various risk factors—and the distinct lack of interest in preventive cardiology and interventional lipidology.)  It was at this point that the paramount importance of cigarette smoking became obvious—until that time cigarette smoking was just another ATD risk factor.  Almost as important as that finding was the observation that the various ATD risk factors rarely acted alone.  The various ATD risk factors virtually always acted in concert with one another.   Therefore the author began to look for various risk factor combinations that might be predictive of ATD.  Various risk factor combinations were examined, but the only combination that proved predictive was the      LDL-c  : HDL-c ratio paired with systolic blood pressure (SBP)—and then only when stratified by cigarette smoking.  The author, a few years later, replaced the LDL-c  : HDL-c ratio with the CRF, due to the latter’s somewhat better predictive ability.  (Feeman, unpublished data)  (See Figure I.)  The initial CRF-SBP plots amounted to a scattergram, but once the CRF-SBP plots were stratified by cigarette smoking, a clear band of those plots was present at CRF levels at or above 0.70.  There were a few outliers, representing elderly patients with isolated hypertension, but by far and away the CRF-SBP plots of ATD patients lay at or above the CRF=0.70 threshold.  The author terms this main group the “main sequence.”  In light of this, in 1989 (when enough patients had sustained ATD events) the author elected to draw a sloping line of demarcation, based on the principle of the fewest false negatives.  (In other words, if the author told the patient that his ATD risk was low, the author did not want to have given false reassurance.)  The co-ordinates of this line are CRF-SBP plots (0.79, 100) and (0.54, 140).  This line was termed the Atherosclerotic Disease Threshold Line.  In the last decade of the twentieth century, the author contacted the principle investigators of eight angiographic regression trials and asked for copies of their databases.  Actually, more than eight principal investigators were contacted, but only eight responded in the affirmative.  In any event, the end of trial average CRF-SBP plot of one of those trials (the Lipoprotein and Coronary Atherosclerosis Study, or LCAS) lay just below the threshold line.    LCAS was a secondary prevention trial. [26]  The author then began to consider that since most of the patients in his practice had never had a lipid panel tested, those with an elevated C Thr threshold might well have subclinical atherosclerosis and therefore might  be better  served by lowering the threshold line to (0.74,100) and (0.49/140).  These new co-ordinates, in the year 2000, created a new threshold line, now termed the angiographic stabilization/regression line (ASRL). [27] The threshold line co-ordinates have not been changed since that time.  This graph, and its ASRL, are now termed the BGS Graph.  (See Figure I.) Table I gives the outcomes of BGS ATD patients from 1978 till the end of 2003.  At the end of 2003 major insurance changes caused the author to lose about 25% of his patients, who were taken over by Wood County Hospital.  As a result the author was unable to determine the outcomes of the “lost” patients, especially those who developed multi-system ATD (coronary, cerebral arteries, and peripheral  arteries, in various combinations ).  The BGS Graph (1978-2003) does give outcomes data and reveals that people who smoke cigarettes and have a CRF-SBP plot above the ASRL have the youngest age of ATD onset, age at multi-system disease and age at death.  Those with CRF-SBP plots below the ASRL and who have never smoked cigarettes have the oldest ages.  Indeed the average age of ATD onset in these latter patients is so old that this plot position, in people who have never smoked cigarettes, implies virtual immunity to ATD.  The results for those who smoked cigarettes in the past are not far behind.  Incidentally the average age of ATD onset is earlier for those people whose CRF-SBP plot lies below the ASRL but who smoke cigarettes than that of people who have a CRF-SBP plot above the ASRL but who never smoked cigarettes.  This finding illustrates the prime importance of cigarette smoking.  Table II gives the CRF-SBP plot position for the (1978-2003) ATD patients and those added to the BGS after 2003, with respect to the average age of ATD onset. The findings are similar to those of Table I. Simply using the BGS graph with its ASRL provides a rough estimate of ATD risk. A CRF-SBP plot below the ASRL, in the absence of any history of cigarette smoking, has such low risk of ATD that this scenario implies virtual immunity to ATD, at least until very old age.  A CRF-SBP plot above the ASRL implies significant ATD risk. To enhance ATD risk prediction, the author has placed two lines of demarcation on the BGS Graph.  The horizontal line was placed at CRF = 0.70 and the vertical line at SBP = 140 mm Hg.  This divides the graph into quadrants.  The left upper quadrant (zone I)represents those with pure dyslipidemia; the right upper quadrant (zone II), those with dyslipidemia and hypertension; the right lower quadrant (zone III), those with pure hypertension; and the left lower quadrant, those with neither dyslipidemia nor hypertension (zone IV).  (See Figure II.)   Tables III A-C give the average ages of ATD onset in each of the areas in Figure II.  Table III-A gives the average age of ATD onset for all patients while Table III-B shows the same data, but for males, and Table III-C shows the data for females.  The data in these tables is stratified by cigarette smoking status.  Patients with CRF-SBP plots in the let upper quadrant are said to be in Zone I, with patients having CRF-SBP plots lying in the area above the ASRL being in Zone I-A and those below the ASRL being in Zone I-B.  The right upper quadrant is termed Zone II, and all CRF-SBP plots in Zone II are above the ASRL.  The right lower quadrant is termed Zone III, with patients above the ASRL being in Zone III-A and those below the ASRL being in Zone III-B.  The left lower quadrant is termed Zone IV, with the area lying above the ASRL being termed Zone IV-A and the area below the ASRL being termed Zone IV-B. Each of these zones can be subdivided by cigarette smoking status.   Table III-A contains the data for the one patient with full lipid and blood pressure data, but whose cigarette smoking status is not known.  Tables A-C can be color coded for the average age of ATD onset.  Those patients who presented to the BGS with ATD events occurring younger than 64 years of age and younger are placed in the red zone; those with presenting ages of 65-74 years, in the yellow zone; and those with presenting ages of 75 years or older, in the green zone.  It will be clear that red zone patients lie mainly in the current smokers category, whereas green zone patients lie mainly in the never smokers category.  (The never smokers category does include people who have never smoked cigarettes, but may have smoked pipes or cigars, or may have chewed tobacco.)    Zone I-A had the youngest ATD patients (average age of ATD onset at 59 years) and adding in Zone I-B does not much change the average age of ATD onset (60 years).  Zone II patients are second in the age category (average age of 65 years).  Zone III-A patients are older still (average of 69 years) and  Zone III-B are even older (average age of 71 years).  The data for Zone IV-A and IV-B follows a similar pattern. In Table III-A, virtually all of the “red zone” patients are found in current cigarette smoking cohorts and “red zone” ex-smokers/never smokers mainly in Zones I-A and I-B.  Conversely all of the “green zone” patients fall into the never smoker category.  Patients in Zone II have their ATD events on the average of 6 years later than do Zones I-A and I-B.  Zone III patients lie mostly (78% of patients) in Zone III-A and have their ATD events a few years earlier than do Zone III-B patients. Tables III-B and III-C reveal that males and females have similar patterns. The advantage of this approach is that it better defines ATD risk in terms of lipids and blood pressure.  The left lower quadrant can contain those with CThr and/or treated hypertension.   Using this approach, those patients with an abnormal CThr can now be moved to the left upper quadrant if the  SBP is below 140 mmHg or to the right upper quadrant if SBP is elevated, and treated dyslipidemic patients can be added to either the left upper quadrant or the right upper quadrant, again depending on blood pressure. Those with treated hypertension, but no dyslipidemia (treated or not) can be shifted from the left lower quadrant to the right lower quadrant.  This permits better definition of ATD risk and it examines that risk in terms of risk factors combinations and now analyzes the data of 954 patients instead of 907 patients.  However, this modification makes little difference to the average age of ATD onset.  

There is still a large differential in ATD risk in the various subdivisions noted above.  This can be further investigated by eliminating the ASRL and dividing the CRF and the SBP into sextiles and then determining the average age of ATD onset for each CRF-SBP cohort.  Doing this means that the CThr patients and those on lipid modifying therapy must of necessity be excluded from the analysis.  Determining the CRF-SBP cohorts for the 906 patients for whom the CRF, SBP and cigarettes smoking data is known—excluding the one patient with CRF and SBP data, but no cigarette smoking data—means that there are 36 cohorts to match with the BGS ATD patients.  (See Tables IV A-D.) Tables IV A-D give the distribution of the various CRF-SBP cohorts in the BGS ATD population for those patients who were not receiving medications for either hypertension or dyslipidemia at their presentation to the BGS.  This encompasses 79% (716/907) of the BGS ATD population for whom full lipid data is available. With the loss of 21% of the BGS ATD database, stratifying the CRF-SBP cohorts by cigarette smoking status  is problematical due to fewer patient numbers.  To reiterate, the red zone includes all of the patients whose average age of ATD onset is 64 years or earlier.  The yellow zone includes all of the patients whose average age of ATD onset is between 65 and 74 years of age.  Finally, the green zone is comprised all of the patients whose average age of ATD onset is 75 years or older.  (This is equivalent to early onset ATD, middle-aged onset of ATD, and old age onset of ATD.)  Men and women are grouped together in these Tables.   Inspection of Table IV-A reveals that 75% (316/422) of the red zone patients lie in the CRF= 0.70 or higher portion of the Table, whereas 22% (62/277) of the yellow zone patients occupy this area (CRF = 0.70 or higher) and green zone patients are absent from this area.  In the area occupied by those with CRF values of 0.60-0.69 lie the CRF-SBP cohorts of most of the other red zone patients (14%, or 57/422) and the CRF-SBP cohorts of yellow zone patients (38%, or 104/277). But again no green zone patients are present.  Finally in the area occupied by patients with CRF values of 0.59 or less, lie the CRF-SBP plots of the remaining red zone patients (12%, or 49/422) and yellow zone patients (43%, or 111/277).  All green zone patients have CRF-SBP plots in this area.  The author interprets all of this to mean that the risk of early onset ATD is mainly limited to people with a CRF of 0.70 or higher.  However, Table IV-A takes into account all comers who are not taking anti-hypertensive/dyslipidemic medications and is not stratified by cigarette smoking status. Table IV-B provides the same analysis, but limited to current cigarette smokers.  It will be clear that 99% (267/269) of patients are associated with red zone risk, regardless of CRF-SBP cohort status.  There is one yellow zone patient (in the CRF = 0.65-0.69 section) and one green zone patient (in the CRF= 0.60-0.64 section). Table IV-C provides the same analysis for people who have quit smoking cigarettes for at least 6 months and have not resumed smoking.  91% (43/47) of red zone patients have CRF values of 0.70 or higher, as do 43% (63/146) of yellow zone patients and 29% (4/14) of green zone patients.  Of the patients with CRF values of 0.60-0.69, 9% (4/47) of red zone CRF-SBP have cohorts in this area, as do 21% (30/146) yellow zone patients and 71% (10/14) green zone patients.  Finally the area of Table IV-C for patients with CRF values of 0.59 or less contains no red or green zone patients, but does contain the cohorts of 36% (53/146) yellow zone patients.  It will be clear that once current cigarette smokers are removed from consideration, there is a marked shift away from red zone (early onset) age at ATD onset, favoring yellow and green zones. Table IV-D gives the data for patients who have never smoked cigarettes, though they may have smoked cigars or pipes or chewed tobacco.  For patients who have CRF values of 0.70 or higher, red zone patients congregate in this area of the Table with 78% (28/36) being located there, as are yellow zone patients ( 48% or 71/147) and 38% ( 21/56 )green zone patients.  For those with CRF values of 0.60-0.69,  22% (8/36) of red zone patients have CRF-SBP cohorts in this area, as compared  to 15% (22/147) yellow zone patients and 43% (24/56) green zone patients.  For those patients with CRF values of 0.59 or less, no red zone patients are present, whereas 37% (54/147) of yellow zone patients and 20% (11/56) of green zone patients are present.  

In summary, Tables IV B-D reveal the marked shift away from early onset ATD as one moves from current to past to never cigarette smoking.  Moreover, most of the red zone (early onset ATD) patients lie in the CRF of 0.70 or higher area.  Additionally, as the CRF falls, the average age of ATD onset appears to  grow older. Once the average age of ATD onset for a patient’s CRF-SBP cohort is determined, the next question to be asked is when treatment of a lipid disorder should be commenced.  The answer to this question is that it depends upon what risk of ATD one is willing to run before beginning treatment.  This author has  privately contacted a number of his collegues, none of whom would go on the record on this matter.  One, speaking on the condition of anonymity, suggested that a lifetime risk of 15% would be a suitable threshold for initiating treatment.  Recent investigations of ATD risk have indicated fault with current guidelines, with the main fault being with long term projections. (21-23) Hence this author decided to utilize that recommendation as it came from an internationally known investigator and this author’s data fits better with lifetime ATD risk.  The topic of when to begin dyslipidemic therapy can be initiated by revisiting the Gen Pop and creating a table with all of the CRF determinations, stratified by age and gender—and then by the presence/absence of ATD patients in each of the cohorts thus created.  (See Tables V  A-C.)  The ATD risk in each cohort is determined by the number of ATD events and number of patients within each cohort.  Table V-A is comprised of all-comers; Table V-B, for males; and Table V-C, for females.   If one examines all of the CRF values, arranged by CRF sextile, patient age/gender and number of ATD patients per CRF cohort, then one can generate reverse Kaplan-Meier curves—that is time to event rather than time of survival.  This is seen in Table V A-C and Figures III A-F.  For purposes of comparison, in Figures III A-F, the Kaplan-Meier curves for men and women are placed on the same graph. The curves for men and women are quite similar.  Table V-B is used for men and Table V-C for women. By way of illustration, and for example, the male cohort with a cumulative age of 49 years or less and a CRF of 0.80 or higher has an ATD risk of  25% (48/190) prevalence of ATD.  The ATD prevalence can be calculated for each age-CRF sextile.  For each CRF sextile, a curve can be created giving the ATD prevalence for each age group.  When this is done, a series of six Kaplan-Meier curves is created.  The curve for the CRF of 0.80 or higher rises more rapidly than does the curve for a CRF=O.75-0.79, which in turn rises more rapidly than the curve for CRF=0.70-0.74, which in turn rises more rapidly than the curve for CRF=0.65-0.69, which in turn rises more rapidly than the curve for CRF=0.60-0.64, which, finally, rises more rapidly than the curve for CRF=0.59 or lower. These curves are presented in terms of both males (brown color) and females (orange color).  It will be apparent that the curves for males and for emales are very close to be overlapping.  This implies that the various trials that have been done to evaluate dyslipidemia therapy (mainly involving men) will also be applicable to women.  This finding is supported by noting that the ATD risk curves are virtually identical for men and women, whether those curves lie above the ASRL or below the ASRL.  (Data available, not shown)  Two caveats are note for this analysis. First, that follow-up in incomplete and it is not impossible that some of the patients who were lost to the author’s practice may have had values that may have influenced the prevalence of ATD or CRF-age cohort.  The author had no control over the loss of those patients.  When Wood County Hospital took over the insurance coverage of those patients, it did so not knowing their lipid values.  Thus there should be no selection bias involved.  If anything, the “lost” patients who sustained ATD events unknown to the author should shift the curves to the left.  And secondly, that the lipid values presented in this study are, as noted previously, upon the precipitation method of HDL-c measurement.  The same approach could be used for lipids based on the enzymatic methodology of HDL-c measurement, using the corrections described earlier.  To utilize this approach, using the 15% lifetime risk parameter, one simply determines one’s CRF in order to select the appropriate Kaplan-Meier curve and then moves up the prevalence (vertical) arm of the Table until one gets to the 15% level.  One then crosses over to the appropriate point on the Kaplan-Meier curve and drops down to the age group (horizontal) axis to determine the age group at which the 15% risk is associated.  One begins treatment when that age group is reached, though it is the author’s preference to initiate treatment in the decade prior to reaching the age group specified by the Kaplan-Meier curves in order to give treatment a “head start.”  Of course, any desired risk can be used, depending upon the investigator’s preferences. It is common knowledge that lipids change with the length of life lived.  In the author’s long experience, such changes may occur rapidly (yearly) or slowly (over decades).  This has been termed “trajectory jumping.”  This is an important concept because it is clear that ATD events do not happen the moment that one achieves a certain CRF, but rather it takes time for the CRF to form plaque within the artery wall.  Knowing the history of CRF trajectory jumping—that is, how long one has had a given CRF—permits one to determine the need for therapy.  Also, it may well be useful to know the rate of change of the “trajectory jumping,” since if a patient begins with a low CRF trajectory (say 0.59 or lower), but his/her CRF is rapidly changing then the goal of treatment is to prevent long term ATD, even though that patient is of no immediate ATD risk:  the low short term risk but high long term risk scenario.  After all, it would seem to be easier to prevent ATD earlier in life than to try to stabilize/regress more advanced ATD later in life.  This concept is the basis for lipid testing in childhood with follow up testing depending on the CRF trajectory and trajectory jumping.  The author considers this concept to be vital in the fight against ATD. The above evaluation and treatment of dyslipidemia represents the ideal practice of epidemiology, though the United States Preventive Task Force Service (USPTFS) has opined against it. [28] The  USPTFS is simply wrong on this topic, for the reasons outlined above.  In Wood County and the city of Bowling Green, at least in the author’s long experience, lipid panels are rarely ordered by the physicians who attended the patient prior to that patient being seen by the author.  Hence the author has no idea how long the dyslipidemia has been present and hence the state of any ATD that may be present (subclinical ATD).  As a result, the author must assume that the dyslipidemia may have been present for a long enough period of time to cause plaque formation and to begin treatment immediately—especially when the patient’s CRF value is on the more rapidly progressive trajectory, or in the older patients when the CRF is on the moderately progressive trajectory.  The minimally progressive trajectory (CRF of 0.59 or lower) is rarely treated and if the patient remains on this trajectory, he will not reach the 15% threshold until the latter part of the eighth decade of life and she will not reach that threshold until the ninth decade of life if ever.   The author has published an article detailing ATD risk factors in patients aged 19 years or less in his family practice. [29]  Cigarette smoking was present in 10% of these patients; elevated CRF (0.70 or higher), in 15%; elevated LDL-c (125 mg/dl [3.2 mmoles/L] or higher), in 23%; low HDL-c (39 mg/dl [1.0 mmoles/L] or lower) in 28%; and elevated CT (200 mg/dl [5.2 mmoles/L] or higher), in 14%.  TG, 2 hr postprandial glucose levels , SBP, and body mass index were not elevated in the vast majority of cases.   Males tended to have lower HDL-c, whole females tended to have higher LDL-c.  This counteracts the idea that children do not have ATD risk factors and supports early and repeated testing. The next question is how far to lower LDL-c.  The author has in his physical possession the actual data bases of eight published angiographic regression trials.   These trials are as follows: 1) Saint Thomas Atherosclerosis Regression trial (STARS)  [30], 2] the Program on the Surgical Control of the Hyperlipidemias (POSCH) [31], 3] the National  Heart, lung, and Blood Institute Type II Coronary Intervention Study (NHLBI) [32], 4] Lipoprotein and Coronary Atherosclerosis Study (LCAS) (26), 4)  Familial Atherosclerosis Treatment Study (FATS) [33], 5] Pravastatin Limitation of Atherosclerosis in the Coronary Arteries (PLAC-1) [34], Lopid Coronary Artery Trial (LOCAT) [35], and the Heidelberg Study [36].  POSCH is a surgical (partial ileal bypass) trial; STARS, FATS, and NHLBI, resin trials, though FATS had a statin arm and a niacin arm to go with the resin : LCAS and PLAC-1 are statin trials, though in LCAS resins might be used to achieve lipid goals if those goals were not achieved by fluvastatin; LOCAT is a fibrate trial; and Heidelberg is a diet and exercise trial.  Angiographic regression trials are known to predict clinical outcomes, with plaque regression being associated with fewer subsequent ATD events and plaque progression associated with more subsequent ATD events. [37,38] Table VI-A shows the rates of plaque progression at the end-of-trail  LDL-c levels of each trial.  Since the goal of dyslipidemia therapy is and always must be the stabilization/regression of plaque, it is clear that the lower the end-of-trial LDL-c, the lower the rate of associated plaque progression.  The best results were obtained with POSCH and NHLBI, whereas the least good results were obtained in LOCAT and the Heidelberg trial—at the same end-of-trial LDL-c levels.  This result is at odds with the current mantra that it does not make a difference how LDL-c is lowered—only that the goal is achieved. Table VI-B show the same findings as Table VI-A, but for end-of-trial LDL-c levels below 99 mg/dl (2.5 mmoles/L).  There is basically no difference if plaque progression rate rates at any end-of-trial LDL-c level below 80 mg/dl (2.0 mmoles/L)  This result is at odds with the current mantra that “Lower is better” with respect to LDL-c . The author has advocated the use of the CRF and CThr and yet he has used the end-of-trial LDL-c to measure the success of dyslipidemia therapy.  The author favored LDL-c in this analysis because it was marginally better at predicting plaque progression.  The question is why this is so.  The answer may lie in the fact that when choline and carnitine are delivered to the terminal ileum, they are metabolized to trimethyl alanine (TMA).  TMA is then absorbed into to intestinal blood vesels and taken to the liver where TMA is oxidized to YMA-oxide (TMAO).  TMAO is thrombogenic and also blocks reverse cholesterol transport. [39] This hypothesis is suggested by the remarkable decreases in plaque progression when choline and carnitine are shunted away from the terminal ileum (POSCH) or are bound to resins (NHLBI).  The intermediate results (plaque progression) in FATS and LCAS) when resins may/may not have been given at maximally tolerated doses would further support this hypothesis.  (FATS used colestipol as the baseline therapy, with one arm receiving lovastatin and the other arm receiving niacin.  In LCAS resins were given only when fluvastatin failed to achieve goals of lipid therapy.)  PLAC-1, LOCAT, and Heidelberg did not involve the use of resins and had the worst results. This is pertinent because in POSCH, using the CRF as the measure of dyslipidemia treatment results, a rise of the CRF at one year after initiation of therapy—even minutely—always resulted in plaque progression at the three-year angiogram.  Conversely, a fall in the CRF at one year after initiation of therapy—even minutely-- always resulted in plaque non-progression.  No exceptions. [40] Additionally, the author has reported the case of a post-myocardial infarction female patient who reached her 104th birthday, still taking her statin, some 25 years post infarction, and being followed by CRF.  Her LDL-c levels never achieved 79 mg./dl (2.0 mmoles/L). [41]  The author has also reported the case of a man with ATD risk factors, but free of clinical ATD, who reached his `100th birthday, still taking his statin, before passing away.  Because his HDL-c was very low, he was followed by LDL-c level and his LDL-c was basically kept below 99 mg/dl ( Precip but 89 mg/dl if Enz) (2.6 mmoles/L and 2.3 mmoles/L respectively) until the more potent statins, whereupon his LDL-c was lowered even farther to below 80 mg/dl (Precip, but below 70 mg/dl if Enz) ( 2.1 mmoles/L and 1.8 mmoles/L respectively ).  His total time of therapy was 29 years. [42] Finally, for any treatment proposal, the “proof of the pudding is in the eating.”  To this end the author has examined his database for patients with AMI/ACI, both male and female.  This involves inspection of records dating back 30-40 years and such data can be difficult to extract.  Even so, there were 133 male and 70 female patients with the diagnosis of AMI/ACI,  Of those patients only 5 male and 9 female patients had one of these events within the last 10 years and only 1 male and 3 female patients within the last 5 years.  All but one of the 5 males and 5 of the nine females were current cigarette smokers, which means that at best the author could only delay the onset of their AMI/ACI events—or as the author tells his patients, ”I can save you from many conditions, but not from the effects of cigarette smoking. You have to quit smoking.” There were no strokes.  The author suggests that in addition to the power of statins to prevent ATD, the guidelines offered in this manuscript may have played a major role in these results.

Figure: I

Figure: II

Figure: III-A

Figure: III-B

Figure: III-C

Figure: III-D

Figure: III-E

Figure III-F

Above ASR Line Below ASR Line

Sex Average Age of  + Past - + Past -

Male

ATD Onset Total Patients 126 130 86 20 14 8

 Total Patient Years 6659 8536 5913 1174 1041 623

 Ave. Age of ATD Onset 53 66 69 59 74 78

MSD Onset Total Patients 38 41 32 6 5 1

 Total Patient Years 2363 2983 2522 382 402 78

 Ave. Age of MSD Onset 62 73 79 64 80 78

Death Onset Total Patients 49 64 47 12 11 4

 Total Patient Years 3153 4780 3805 815 879 374

 Ave. Age of Death Onset 64 75 81 68 80 94

Female

ATD Onset Total Patients 65 56 137 18 15 34

 Total Patient Years 3852 3908 9955 1145 1003 2543

 Ave. Age of ATD Onset 59 70 73 64 67 75

MSD Onset Total Patients 22 24 49 6 7 16

 Total Patient Years 1534 1800 3931 440 532 1283

 Ave. Age of MSD Onset 70 75 80 73 76 80

ATD Onset Total Patients 26 23 79 9 7 23

 Total Patient Years 1830 1824 6542 650 533 1941

 Ave. Age of Death Onset 70 79 83 72 76 84

Table: I

Sex  + Past -

Male

Above ASR Line Total Patients 166 154 113

 Total Patient Years 8715 10215 7575

 Ave. Age of ATD Onset 53 66 67

Below ASR Line Total Patients 25 23 14

 Total Patient Years 1455 1650 1035

 Ave. Age of ATD Onset 58 72 74

Female

Above ASR Line Total Patients 87 67 158

 Total Patient Years 5050 4601 11595

 Ave. Age of ATD Onset 58 69 73

Below ASR Line Total Patients 29 22 48

 Total Patient Years 1741 1498 3437

 Ave. Age of ATD Onset 60 68 72

Table: II

Ave Age of ATD Onset

Σ ATD. Pop

Cigarettes

Zone + Past OT NT  Σ

IA 102 61 7 64  234

5362 3890 461 4193  13906

53 64 66 66  59

IB 2 1  2  5

97 38  174  309

49 38  87  62

II 83 89 7 102  281

4670 5888 481 7185  18224

56 66 69 70  65

IIIA 39 49 7 64  159

2172 3527 531 4817  11047

56 72 76 75  69

IIIB 11 17 1 17  46

680 1261 86 1242  3269

62 74 86 73  71

IVA 29 21 1 20  71

1561 1452 79 1502  4594

54 69 79 75  65

IVB 42 27  41 1 111

2474 1853  2891 75 7293

59 69  71 75 66

Σ 308 265 23 310 1 907

17016 17909 1638 22004 75 58642

55 68 71 71 75 65

Table: III-A

Ave Age of ATD Onset

Σ Male ATD. Pop

Cigarettes

Zone + Past OT NT  Σ

IA 65 44 7 35  151

3345 2786 461 2198  8790

51 63 66 63  58

IB 1   1  2

50   85  135

50   85  68

II 58 64 7 35  164

3105 4208 481 2394  10188

54 66 69 68  62

IIIA 25 30 7 15  77

1382 2114 531 993  5020

55 70 76 66  65

IIIB 4 12 1 2  19

208 880 86 120  1294

52 73 86 60  68

IVA 18 16 1 7  42

883 1107 79 516  2585

49 69 79 74  62

IVB 20 11  9 1 41

1197 770  667 75 2709

60 69  74 75 66

Σ 191 177 23 104 1 496

10170 11865 1638 6973 75 30721

53 67 71 67 75 62

Table: III-B

Ave Age of ATD Onset

Σ Female ATD. Pop

Cigarettes

Zone + Past OT NT  Σ

IA 37 17  29  83

2017 1104  1994  5115

55 65  69  62

IB 1 1  1  3

47 38  89  174

47 38  89  58

II 25 25  67  117

1565 1680  4791  8036

54 67  72  69

IIIA 14 19  49  82

790 1413  3824  6027

56 74  78  74

IIIB 7 5  15  27

472 381  1122  1975

67 76  75  73

IVA 11 5  13  29

678 345  986  2009

62 69  76  69

IVB 22 16  32  70

1277 1083  2224  4584

58 69  70  65

Σ 117 88  206  411

6846 6044  15030  27920

59 69  73  68

Table: III-C

CRF vs SBP: Σ Cigarettes

ΣATD. Pop

SBP: No Rx

SBP ≥0.80 0.75-0.79 0.70-0.74 0.65-0.69 0.60-0.64 ≤0.59 Σ

≥180 8 14 8 10 5 10 55

451 976 555 632 337 672 3623

56 70 69 63 67 67 66

160-178 9 18 16 11 10 17 81

522 1149 1158 767 684 1291 5571

58 64 72 70 68 76 69

140-158 32 32 24 18 15 41 162

1878 1930 1616 1234 932 2796 10386

54 60 67 69 62 68 64

130-138 31 21 22 11 17 29 131

1800 1307 1377 793 1163 1934 8374

58 62 63 72 68 67 64

120-128 29 32 24 20 13 31 149

1478 1899 1489 1174 884 2128 9052

51 59 62 59 68 69 61

≤118 21 17 20 19 12 49 138

1298 997 1121 1235 737 3106 8494

62 59 56 65 61 63 62

Σ 130 134 114 89 72 177 716

7427 8258 7316 5835 4737 11927 45500

57 62 64 66 66 67 64

Table: IV-A

CRF vs SBP

ΣATD. Pop

+ Cigarettes; SBP: No Rx

SBP ≥0.80 0.75-0.79 0.70-0.74 0.65-0.69 0.60-0.64 ≤0.59 Σ

≥180 3 4 1 5 1 4 18

145 233 56 244 76 236 990

48 58 56 49 76 59 55

160-178 4 6 5 3 3 5 26

228 342 308 167 167 317 1529

57 57 62 56 56 63 59

140-158 15 9 5 4 4 13 50

768 470 265 204 222 741 2670

54 52 53 51 56 57 53

130-138 14 9 11 1 7 6 48

726 462 600 66 448 369 2671

52 51 55 66 64 62 56

120-128 16 12 9 8 9 9 63

724 696 452 375 566 515 3328

45 58 50 47 63 57 53

≤118 8 6 11 9 9 21 64

375 298 620 507 474 1224 3498

47 50 56 56 53 58 55

Σ 60 46 42 30 33 58 269

2966 2501 2301 1563 1953 3402 14686

49 54 55 52 59 59 55

Table: IV-B

CRF vs SBP

ΣATD. Pop

Past Cigarettes; SBP: No Rx

SBP ≥0.80 0.75-0.79 0.70-0.74 0.65-0.69 0.60-0.64 ≤0.59 Σ

≥180 2 4 1 3 2 3 15

113 298 63 246 134 198 1052

57 75 63 82 67 66 70

160-178 5 6 3 3 2 4 23

294 394 221 240 111 290 1550

59 66 74 80 56 73 67

140-158 10 14 12 7 2 16 61

658 890 828 484 124 1164 4148

54 64 69 69 62 73 68

130-138 8 5 6 6 4 12 41

446 344 426 422 271 800 2709

56 69 71 70 68 67 66

120-128 8 9 5 6 3 7 38

481 581 335 414 230 501 2542

60 65 67 69 77 72 67

≤118 7 1 4 5 1 11 29

461 41 162 344 87 785 1880

66 41 41 69 87 71 65

Σ 40 39 31 30 14 53 207

2453 2548 2035 2150 957 3738 13881

61 65 66 72 68 71 67

Table: IV-C

CRF vs SBP

ΣATD. Pop

Σ- Cigarettes; SBP: No Rx

SBP ≥0.80 0.75-0.79 0.70-0.74 0.65-0.69 0.60-0.64 ≤0.59 Σ

≥180 3 6 6 2 2 3 22

193 445 436 142 127 241 1584

64 74 73 71 64 80 72

160-178  6 8 5 5 8 32

 413 629 360 406 684 2492

 69 79 72 81 86 78

140-158 7 9 7 7 9 12 51

452 570 523 546 586 891 3568

65 63 75 78 65 74 70

130-138 9 7 5 4 6 11 42

628 501 351 305 444 765 2994

70 72 70 76 74 70 71

120-128 5 11 10 6 1 14 47

273 622 702 385 88 1037 3107

55 57 70 64 88 74 66

≤118 6 10 5 5 2 17 45

462 658 339 384 176 1097 3116

77 66 68 77 88 65 69

Σ 30 49 41 29 25 65 239

2008 3209 2980 2122 1827 4715 16861

67 65 73 73 73 73 71

Table: IV-D

Cumulative ATD Risk

Σ Σ Gen Pop

Σ Σ Cigarettes

Age Group ≥0.80 0.75-0.79 0.70-0.74 0.65-0.69 0.60-0.64 ≤0.59 Σ

≤29 5 8 7 2 6 11 39

62 96 139 189 157 711 1354

8% 8% 5% 1% 4% 2% 3%

≤39 19 25 20 14 10 24 112

147 198 236 277 219 933 2010

13% 8% 5% 5% 3% 6%

≤49 62 58 43 27 20 55 265

255 311 332 342 276 1089 2605

24% 8% 7% 5%

≤59 101 86 70 43 39 97 436

337 374 408 393 316 1214 3042

30#% 8%

≤69 129 125 103 75 54 97 583

337 436 457 451 347 1278 3306

38)#% 8%

≤79 142 142 126 97 69 162 738

393 459 486 481 365 1328 3512

361& !%

Σ 146 157 139 109 78 191 820

398 466 501 495 377 1363 3600

374("!#%

Table: V-A

Cumulative ATD Risk

Σ Male Gen Pop

Σ Σ Cigarettes

Age Group ≥0.80 0.75-0.79 0.70-0.74 0.65-0.69 0.60-0.64 ≤0.59 Σ

≤29 1 4 2 2 3 3 15

45 50 81 98 78 289 641

2% 8% 2% 2% 4% 1% 2%

≤39 12 15 11 9 5 9 61

108 128 136 151 105 352 980

11% 8% 6% 5% 3% 6%

≤49 48 36 28 19 10 23 164

190 200 196 189 138 398 1311

25% 7% 6%

≤59 75 59 44 28 20 45 271

245 244 239 218 156 446 1548

31$%

≤69 91 77 58 39 24 58 347

268 276 262 242 170 472 1690

34("!%

≤79 101 88 71 52 29 72 413

280 291 276 259 175 490 1771

360& #%

Σ 104 91 77 56 32 79 439

284 295 284 264 179 500 1806

371'!$%

Table: V-B

Cumulative ATD Risk

Σ Female Gen Pop

Σ Σ Cigarettes

Age Group ≥0.80 0.75-0.79 0.70-0.74 0.65-0.69 0.60-0.64 ≤0.59 Σ

≤29 4 4 5 0 3 8 24

17 46 58 91 79 422 713

24% 9% 9% 0% 4% 2% 3%

≤39 7 10 9 5 5 15 51

39 70 100 126 114 581 1030

18% 9% 4% 4% 3% 5%

≤49 14 22 15 8 10 32 101

65 111 136 153 138 691 1294

22 % 5% 7% 5% 8%

≤59 26 27 26 15 19 52 165

92 130 169 175 160 768 1494

28!% 9% 7%

≤69 38 48 45 36 30 70 267

109 160 195 209 177 806 1656

350#% 9%

≤79 41 54 55 45 40 90 325

113 168 210 222 190 838 1741

362& !%

Σ 42 56 62 53 46 112 371

114 171 217 231 198 863 1794

373)##!%

Table: V-C

Trial Angiographic Outcome > 200 175-199 150-174 125-149 100-124 < 99>

POSCH Progression

Non Progression

∑

% Progression 26

15

41

63E

27

72

63Q

82

133

38'

83

110

25

94

106

11% 2

267

269

1%

NHLBI Progression

Non Progression

∑

% Progression 16

37

53

30% 1

10

11

9% 4

9

13

31% 0

4

4

0% 0

7

7

0% 0

2

2

0%

FATS Progression

Non Progression

∑

% Progression 5

7

12

42% 5

10

15

33% 7

8

15

47% 7

12

19

37% 9

18

27

33% 5

27

32

16%

LCAS Progression

Non Progression

∑

% Progression 1

2

3

33% 6

12

18

33

25

39

362

54

86

379

72

111

35!

55

76

27%

PLAC-1 Progression

Non Progression

∑

% Progression 7

9

16

44!

15

36

588

32

70

54)

55

84

351

53

84

37

17

30

43%

Heidelburg Progression

Non Progression

∑

% Progression 6

6

12

50% 7

16

23

30% 8

16

24

33% 6

10

16

38% 2

9

11

18% 2

1

3

67%

LOCAT Progression

Non Progression

∑

% Progression 10

1

11

91"

8

30

73c

29

92

68w

50

127

618

35

73

52

21

38

45%

Table: VI-A

End of Trial LDL-C

POSCH Means Program on the Surgical Control of the Hyperlipidemias

NHLBI Means National Heart Lung and Blood Institute

FATS Means Familial atherosclerosis Treatment Study

LCAS Means Lipoprotein and Coronary Atherosclerosis Study

PLAC-1 Means Pravastatin Limitation of Atherosclerosis in the Coronary Arteries

Heidelburg Means Study on The Effects of Regular Physical Exercise and Low-Fat Diet on the Progression of Coronary Artery Disease

LOCAT Means Lopid Coronary Angiography Trial

Trial Angiographic Outcome 90-99 80-89 70-79 60-69 50-59 40-49 < 39>

POSCH Progression

Non-Progression

∑

% Progression 0

73

73

0% 0

69

69

0% 0

60

60

0% 0

31

31

0% 1

16

17

6% 0

4

4

0% 1

14

15

7%

NHLBI Progression

Non-Progression

∑

% Progression 0

1

1

0%  

0

1

1

0%    

FATS Progression

Non Progression

∑

% Progression 1

8

9

11% 3

9

12

25% 1

6

7

14% 0

2

2

0% 0

2

2

0%  

LCAS Progression

Non Progression

∑

% Progression 9

20

29

31% 9

20

29

31% 1

10

11

9% 2

4

6

33% 0

1

1

0%  

PLAC-1 Progression

Non Progression

∑

% Progression 8

10

18

44% 2

5

7

29% 2

1

3

67% 0

1

1

0% 1

0

1

100%  

Heidelburg Progression

Non Progression

∑

% Progression 2

1

3

67%      

LOCAT Progression

Non Progression

∑

% Progression 9

7

16

56% 5

6

11

45% 2

3

5

40% 1

2

3

33% 0

3

3

0%  

Table: VI-B

End of Trial LDL-C

POSCH Means Program on the Surgical Control of the Hyperlipidemias

NHLBI Means National Heart Lung and Blood Institute

FATS Means Familial atherosclerosis Treatment Study

LCAS Means Lipoprotein and Coronary Atherosclerosis Study

PLAC-1 Means Pravastatin Limitation of Atherosclerosis in the Coronary Arteries

Heidelburg Means Study on The Effects of Regular Physical Exercise and Low-Fat Diet on the Progression of Coronary Artery Disease

Appendix I

Atherosclerosis Screening Protocol

* A FAST IS REQUIRED. This means that you are not allowed to eat or drink anything but water, beginning at 7 p.m. the night before. It is o.k. to take daily medications unless the doctor tells you otherwise.

* Come to the hospital on the day of the test. You will have a blood test on arrival at the hospital laboratory. Then you are to go home or to the hospital cafeteria and eat a special breakfast. This consists of:

* 4 ounce glass of orange juice

* Bowl of cereal (3/4 cup) with milk and sugar

* 3 pieces of toast with butter and jelly

This breakfast must be eaten within 15 minutes, start to finish. (You will not need to take the whole fifteen minutes, but you must not exceed 15 minutes.) Look at your watch when you start and when you finish. Thereafter, do not eat or drink anything but water.

* Exactly 2 hours after finishing your breakfast come back to the hospital laboratory for your second blood test. Get there a bit early so that the technician can arrange their schedule to draw your blood on time.

* This completes your blood test. The results will be available in the office in 2-3 days and you will need to make an appointment to discuss the results. If you have any questions, please contact the office at 419-352-4665. Thank you.

*Sample schedule

* You do not need to follow these times; it is only an example to show time relationships. Remember you have to fast for 10-12 hours and have another blood test two hours after breakfast.

* 7 p.m. to 9 a.m.—FAST

* 9 a.m.--BLOOD TEST

* 9:15 a.m. to 9:30 a.m.--EAT BREAKFAST

*9:30 a.m. to 11:30 a.m.—FAST

* 11:30 a.m.—SECOND BLOOD TEST

Appendix II                                                        

Category                                                         Number of Patients                    

                                                                                           Male                              Female

ATHD

AMI…………………………………………………………….120                           64

ACI……………………………………………………………...14                               6

AP………………………………………………………………..83                             86

CHF……………………………………………………………..50                              74

+ test………………………………………………………….31                               12

Hx…………………………………………………………….259                             216

ATBD

TCVA…………………………………………………………44                                51

TIA…………………………………………………………….25                                28

ACS……………………………………………………………43                               22

SVID………………………………………………………….13                                 8

HCVA…………………………………………………………8                                  5

Hx…………………………………………………………127                              136

ATPVD

AAA………………………………………………………..35                                 13

ATPVD…………………………………………………….49                                 53

TAA………………………………………………………….8                                    7

Hx………………………………………………………….40                                  35

Glossary

ATD.............................................Atherothrombotic Disease  

ATHD……………………………. Atherothrombotic Heart Disease

AMI..............................................Acute Myocardial Infarction

ACI….………………….…………Acute Coronary Insufficiency

AP……………………………………………………...Angina Pectoris                                                                   …………                   CHF…………………………………………. ,,,,,,,,,,Congestive Heart Failure

+ ATHD Test…………………………………….... Positive Test for ATHD

Hx ATHD….…………………………………………. History of ATHD Event      

ATBD……………………………………………………Atherothrombotic Brain Disease

TCVA……………………………………………………Thrombotic Cerebrovascular Accident

 TIA………………………………………………………. Transient Ischemic Attack

 ACS………………………………………………………Asymptomatic Carotid Stenosis

 SVID……………………………………………………Small Vessel Ischemic Disease

Hx ATBD………………………………………………History of ATBD Event

ATPVD………………………………………………..Atherothrombotic Peripheral Vascular Disease AAA…………………………………………………..Abdominal Aortic Aneurysm

TAA………………………………………………….Thoracic Aortic Aneurysm

ATPVD……………………………………………….Atherothrombotic Peripheral Vascular Disease

Hx ATPVD………………………………………..History of ATPVD Event

This has been a long manuscript, but the author wanted to put all of his work, old and new, into one site, readily accessible to all.  As far as the virtual elimination of early/middle aged ATD is concerned, one must be able to predict the population at risk of ATD with high accuracy to assure that no future ATD patient is mischaracterized as being at little risk of an ATD event, only to have such an event occur.  It is not possible to predict everyone who will sustain an ATD event under age 75-80 years, but the vast majority of these patients are predictable and hence their ATD event preventable, so long as  the protocols described in this paper are followed. However, this ideal scenario is not always at hand.  Some patients will not be seen in the physician’s office until shortly before the ATD event occurs and the physician will not have the time needed to stabilize/regress any extant ATD.  Such cases do not represent a failure of ATD prevention but rather a failure to take advantage of the protocol. The following dicta are fundamental to ATD eradication and must be understood by the patient as well as by the treating physician. These dicta are taken from the findings of the FHS, with some supplementation by the author:

1. The protocol must be simple, so that the patient and physician alike can understand its guiding principles.  If the protocol is not easily understandable, the physician will not utilize it and neither will the patient.
2. As William B. Kannel, MD, the first director of the FHS, wrote back in the mists of time of preventive cardiology and interventional lipidology:  Heart attacks do not come as a bolt out of the blue, but rather are caused by certain conditions originally termed factors of risk [4-8], but now called risk factors. These risk factors are cigarette smoking, dyslipidemia, and hypertension, with some contribution by the very high blood sugar levels of uncontrolled diabetes mellitus. This author’s rank order. [43])
3. These risk factors are not present in an all-or-none manner.  The risk factors are present in a spectrum, shades of gray as it were.  Each spectrum flows from a more severe portion and a less severe portion.  The more severe end of the spectrum is associated with ATD; the less severe end of the spectrum, is not.  The more severe the risk factor, the earlier is the age of onset of ATD, and the less severe the risk factor, the later is the onset of ATD.
4. There is no risk factor unique to the ATD population. All ATD risk factors are present in the non-ATD population as well as in the ATD population, though the more severe risk factors will tend to be in the ATD Pop.
5. ATD risk factors do not act in a vacuum, but rather interplay with the other ATD risk factors in a risk factor milieu. Thus multiple moderate ATD risk factors may interact to produce clinical ATD events at the same age as a single severe risk factor.  Indeed the interaction of multiple moderate risk factors is more common in the ATD population than is the influence of a single severe risk factor.
6. Correction of any risk factor must reduce the incidence of ATD, usually by stabilizing/regressing plaque or other ATD processes such as clot formation (thrombosis).
7. There is no “safe place.”  If one lives long enough and does not die of something else, he/she will sustain an ATD event.  This is a natural consequence of aging.  The aim of preventive cardiology/interventional lipidology is to postpone that ATD event until very very late in the lifespan.
8. Kannel noted that the occurrence of an ATD event should be viewed as a failure of the physician’s professional competence. [44] Kannel may have been a bit premature, but his comment is fully realizable in today’s medicine.

To these precepts, this author would add the following corollaries:

1. The physician can not protect his/her patient from ATD unless he/she can accurately predict the population at risk of ATD. Or unless he/she wishes to put all of his/her patients on statins, but even then the goal of lipid therapy must be achieved or some treated patients may develop clinical ATD regardless of treatment.  So why treat everyone, even those whose lipids are naturally non-atherogenic?
2. The better the physician can predict the population at risk of ATD, the better he/she can protect his/her patient from the ravages of ATD.
3. It is easier to prevent plaque formation than to reverse an established plaque.  Moreover, the fibrous cap overlying a plaque can create turbulent blood flow, leading to additional build-up of a non-lipid plaque.  This is the rationale behind early and repeated lipid testing.  Patients can jump the various lipid trajectories.  Some do so rapidly, while others do so more slowly.  The scenario in play can only be determined by repetitive testing.  See Tables V A-C.
4. Treat the stronger ATD risk factors first. [44] This means that discontinuation of cigarette smoking is the first step in the treatment program.  The validity of this proposition is borne out by Table I, which shows that the average age of ATD onset is later for those patients who had quit smoking prior to their initial ATD event, but whose CRF and SBP remained above the ASRL, as opposed to those patients who corrected their dyslipidemia and/or hypertension, but continued to smoke cigarettes.  Tables III A-C further support this  finding, in that virtually all patients who smoke cigarettes have their ATD events in the youngest (aged 64 years or less) age group, regardless of CRF-SBP cohort.
5. Treat all ATD risk factors to plaque stabilization/regression levels, as otherwise the patient will continue to progress any extant plaque.  In the vernacular, “Go big or go home.”  The rationale for this precept is that if the high CRF has been present long enough, the patient may already have subclinical ATD.  The pre-clinical “incubation period” of ATD will be shorter or longer, depending upon the severity of the operative risk factors.  Risk factor severity determines the intensity of treatment, but all treatments must achieve plaque stabilization/regression levels.

When all of these precepts are taken into account, the protection of patients from ATD—and hence the virtual elimination of ATD, at least until very late in life—is straight forward. 

First and foremost is the determination of the patient’s ATD risk factors.  This must include a full lipid panel (CT, LDL-c, HDL-c, and TG, since LDL-c will likely be calculated by the Friedewald formula as noted earlier).  Knowing the patient’s cigarette smoking status is fundamental to determining risk of clinical ATD.  And of course knowing the patient’s SBP and 2 hour postprandial blood sugar (see Appendix for the protocol) is also important. The first step in treatment, as noted above, is the cessation of cigarette smoking.  If this is not accomplished, the plaques will continue to evolve, though the lower the CRF, the later in life the ATD event will occur. The next step is to calculate one’s CRF.  Remember that this author’s work was done when the Precip method of HDL-c measurement was the gold standard.  If the treating physician must use the Enz method of HDL-c measurement, he/she will have to make the adjustments the author noted earlier, which means that the CRF goals will need to be lowered by 0.1 and the LDL-c goals by 10 mg/dl (0.25 mmoles/L).  Then combine the CRF with the SBP and find the patient’s position on the BGS graph. If that CRF-SBP plot is located below the ASRL and the patient has never smoked cigarettes, the physician need go no further.  There is no “safe” place in interventional lipidology/preventive cardiology, but the patient is as close to a “safe” place as he/she can get.  The physician must also remember that lipids change over time, and with that change comes a change in ATD risk, so periodic follow-up is necessary.  A CRF that remains at levels of 0.59 or lower rarely requires treatment—unless of course the patient’s LDL-c is 170 mg/dl (4.4 mmols/L), in which case treatment is necessary.  A CRF-SBP plot above the ASRL, on the other hand, entails a large variation in ATD risk. The initial phase of follow-up for those with CRF-SBP plots above the ASRL is to decide the broad classification of the patient’s ATD risk: pure dyslipidemia,  dyslpidemic hypertension, or pure hypertension.  The ATD risk is highest for pure dyslipidemia; dyslipidemic hypertension has the next highest risk; and pure hypertension has the lowest risk of these three categories.  Given this finding, the physician should be careful to not use anti-hypertensive medications that elevate the CRF.  People who have neither dyslipidemia nor hypertension, and who have never smoked cigarettes, are at low risk.(45) The next step is to determine when to begin therapy.  Such a decision is based on the patient’s perceived risk of ATD.  This can be done using the method described in this manuscript. First, one simply locates the patient’s plot on the appropriate CRF graph as portrayed in Figures III A-F in light of the patient’s current age and begin therapy about a decade prior to the average age of ATD onset.  The presence of cigarette smoking implies an urgent need to begin therapy; the absence of cigarette smoking allows a more measured determination of when to begin therapy.  This decision can be enhanced by determining the patient’s ATD risk using the reverse Kaplan-Meier curves located in Tables V A-F.  The author suggests using a lifetime risk of 15%, but the threshold of ATD risk is up to the treating physician.  Treatment should be begun about a decade prior to the age indicated on the Figure.  Treatment should be begun about a decade prior to the age group indicated by the graph to allow time to achieve the desired goal, lest the patient “run out of time” and the ATD event occur prior to the patient achieving the desired goal of therapy.  The goal of therapy is determined by the angiographic studies as noted in Tables VI A-B. For those patients free of clinical ATD, a LDL-c goal of 99 mg/dl (Precip, but 89 mg/dl if Enz; 2.5 mmoles/L and 2.3 mmoles/L respectively). For people with clinical ATD a goal of 79 mg/dl or lower (Precip, but 69 mg/dl if Enz; 2.0 mmoloes/L and 1.8 mmoles/L respectively). ).  This latter goal could also be considered for those with very high baseline CRF or LDL-c levels, who may have subclinical ATD.  If the patient is undergoing an acute ATD event, an argument could be made to lower LDL-c to 55 mg/dl (1.4 mmoles/L) or lower during the acute inflammatory phase, but once the patient is stable, say at 6 months, then the LDL-c goal could be reduced to the 79 mg/dl (Precip, but 69 mg/dl if Enz) (2.0 and 1.8 mmoles/L respectively). Appendix II contains the breakdown of the various ATD events in the author’s ATD database, in terms of ATD of the coronary circulation, cerebral circulation, and the peripheral vascular circulatin.

Using the approach outlined in this manuscript, the author has had a remarkable reduction in ATD events in his practice of family medicine.  The approach is simple and easily implemented into medical practice.  It works for both genders and all ages.  Physicians can understand this methodology and more importantly, patients can understand it as well, and it is well known that a patient who understands why he/she is being treated is more likely to continue with that treatment.