Metabolic Risk for Cardiovascular Disease

Here’s what I wrote on goodreads:

“The book is well sourced and actually does a good job of covering much of the material. But the editor has done a poor job, and as a result the book seems very sloppy compared to similar scientific publications. There are multiple spelling errors and typos along the way, and it frankly seems as if the book was ‘published too fast’, before all the errors could be corrected. At first I punished this severely when I rated it by only giving the book 2 stars, but I realized this was too harsh. There’s a lot of interesting stuff included in the book.”

Here’s the kind of thing I’m talking about:

“Numerous cardiovascular abnormalities may be encountered in obese subjects (Table 6.4) it is not written properly in the PDF files that I have but this version seems correct. Health service usage and medical costs associated with obesity …”

That comment was one of a kind (fortunately), but there are a lot of errors and typos. At one point they talk about a marginally insignificant finding with an associated P-value of 0.52. This kind of stuff makes you look sloppy. The book is a Wiley-Blackwell publication and you kind of expect a bit more from books like these.

I’ve dealt with many of the topics covered in the book before (e.g. here, here and here, Khan Academy, etc.). I got the book in part to have a book in which I knew I could easily find a reference if/when I needed one, so that I wouldn’t have to look around a lot, and I think it’ll serve that purpose reasonably well. I gave the book 3 stars on goodreads. The book deals with many of the things you’d expect a book like this to cover; lipid and lipoprotein metabolism, insulin resistance and its role in cardiovascular disease, the obesity epidemic, hypertension, type 2 diabetes and the metabolic syndrome, tobacco use and cardiovascular disease and the role of physical exercise and nutrition, among other things. There was some interesting stuff in the book, but not a lot which was all that surprising. I really liked parts of chapter 11 on diabetes management and cardiovascular risk reduction; the chapter went over some reviews and a few major studies well known to people who’re interested in these things (ACCORD, ADVANCE), and the interpretation of the data by the author was somewhat different from interpretations I’ve seen in the past. One main point in the chapter is that lowering of Hba1c may be more effective in preventing cardiovascular events/disease progression among patients without overt cardiovascular disease; the argument being that lowering of blood glucose may protect vessels from getting damaged, but once they’re damaged lowing of Hba1c may not do much difference because it’s basically too late (in part because glycemic control may play a greater relative role in the early course of the disease process, compared to other factors, than it does in the later stages, where other mechanisms may conceivably take over to a greater extent – he doesn’t spell this out explicitly but I’d be surprised if he has not been thinking along those lines). In terms of previous trials looking at the link between glycemic control and cardiovascular disease (CVD), researchers have usually looked disproportionately at diabetics with manifest CVD; this is understandable as these patients are high risk. But such applied selection mechanisms in the past may mean (among other things) that these studies may have been underpowered to find the effects they were looking for. This is an interesting line of argument I have not seen before. If you’re wondering why this is important, it’s important because whereas the link between small-vessel disease and glycemic control is incontrovertible and has been for a long time, the link between macrovascular complications (CVD, etc.) and glycemic control has long been questionable, with a lot of mixed findings. Study selection designs and similar mechanisms may help partially explain why previous studies have not been able to establish a clear relationship. There are of course other complicating factors as well. As I think I’ve said before, until it’s perfectly clear to me that glycemic control and macrovascular disease are unrelated (or at least until we know in more detail how they are related), I’ll pretend that better glycemic control may have a protective effect on both small and large blood vessels. Note that the reason why this is important is also that diabetics make up a huge proportion of all heart disease patients; in Denmark the Danish Endocrine Society noted in a report published a few years ago (I can no longer find it online, unfortunately) that roughly half of all Danish patients with chronic ischaemic heart disease, AMI or heart failure have diabetes (of course a lot of them didn’t know that they did, but that’s a different discussion).

I’ve added some observations from the book below as well as a few comments:

…

“a general rule is that CVD risk approximately doubles for each 20mmHg increment of systolic BP and 10mmHg increment of diastolic BP above 115/75mmHg […] a substantial excess risk of stroke death among those who are overweight or obese may be largely accounted for by a higher blood pressure [31].”

“Despite the fact that obesity has been shown to be an independent risk factor for CVD, many studies have reported that obese patients with established CVD have a better prognosis than do patients with ideal bodyweight; the socalled “obesity paradox.” […] The improved survival of obese individuals is paradoxical principally because of the assumption that excessive weight is always and invariably injurious. As a matter of fact, among patients with congestive heart failure, subjects with higher BMI are at decreased risk for death and hospitalization compared with patients with a “healthy” BMI [2]. Further, obesity was associated, in a prospective cohort study, with lower all-cause and cardiovascular mortality after unstable angina/non-ST-segment elevation myocardial infarction treated with early revascularization [89]. The obesity paradox may reflect the lack of discriminatory power of BMI to adequately reflect body fat distribution [20,87,90]. Since BMI measures total body mass, i.e. both fat and lean mass, it may better represent the protective effect of lean body mass on mortality. This negative confounding may have been under-appreciated in prior studies that did not adjust for measures of abdominal obesity. It is possible that the favorable prognosis implications associated with mildly elevated BMI might actually reflect intrinsic limitations of BMI to differentiate adipose tissue from lean mass. The lack of specificity of BMI could dilute the adverse effects of excess fat with the beneficial effects of preserved or increased lean mass [91]. […] Another issue to consider is that normal-weight patients may have a significantly higher percentage of high-risk coronary anatomy compared with obese patients [97]. […] Another limitation in most studies reporting an obesity paradox in patients with CVD is that non-intentional weight loss, which would be associated with a poor prognosis, is not assessed as BMI is measured only at the beginning of the study. Patients who have decompensated heart failure may lose weight because of extensive caloric demands associated with the increased work of breathing […] the excess health risk associated with a higher BMI declines with increasing age. An explanation for the lack of a positive association between BMI and mortality at older ages is that, in older persons, higher BMI is a poor measure of body fat and may simply represent a measure of increased physical activity with preserved lean mass. Sarcopenic obesity, which is defined as excess fat with loss of lean body mass, is a highly prevalent problem in the older individual. […] in view of the importance of body fat distribution, one could argue that, instead of targeting bodyweight per se, one should pay more attention to the WC [waist circumference] and conservation of lean mass as a critical goal in intervention programs [23].”

“Self-reported diabetes mellitus is often used in studies, but that approach underestimates the true prevalence of diabetes mellitus, and may misclassify a sizable fraction of the participants. […] it has been estimated that the lifetime risk of T2DM for persons born in the USA in 2000 is approximately 33% for men and 39% for women [9].”

“Summary analyses have reported that about 65% of deaths among diabetic patients are from vascular or heart disease, 13% are from diabetes itself, 13% are from neoplasms, and the rest are from other causes [14]. Most data concerning diabetes and death in adults are concerned with T2DM, and the limited data on mortality associated with type 1 diabetes mellitus have suggested that approximately one-third are from diabetes itself, one-third are from kidney disease, and one-third are from cardiovascular disease [15,16].” [I should note that some of these numbers sound wrong to me, but for now I’ll just report the numbers. I may have a closer look at the studies later. Note that ‘deaths from diabetes’ is a variable which is incredibly hard to get right in general; everybody dies, but diabetics die faster – deaths incontrovertibly ‘directly attributable’ to diabetes like DKA or hypoglycemic coma don’t make up all the ‘excess deaths’.] Researchers have investigated the effect of diabetes on life expectancy. An Iowa study showed that estimated life expectancy was 59.7 years at birth for diabetic men and 69.8 years in diabetic women, and it was estimated that diabetes reduced the lifespan by 9.1 years in diabetic men and 6.7 years in diabetic women [17]. From US national survey data it has been estimated that men known to have diabetes at age 40 years will lose 11.6 life-years and similarly affected women will lose 14.3 life-years [9].” [Again, for now I’ll just report the numbers…]

“The Centers for Disease Control reported that there were 8 million diabetic American adults with CVD in 1997 and the number increased to more than 11 million in 2007 […] reports suggest that diabetic patients continue to experience CVD at a high rate and are surviving, which has resulted in an increased prevalence of diabetic patients with CVD [21]. […] Fewer diabetes complications such as mortality, renal failure, and neuropathy have been observed for adult T1DM patients in the Pittsburgh Epidemiology of Diabetes Complications Study over recent years. On the other hand, risk of proliferative retinopathy, overt nephropathy, and clinical CAD have not declined over the long-term follow-up interval of 30 years [97]. […] Overall 1-, 2-, and 5-year survival after myocardial infarction in a population-based Swedish cohort was 94%, 92%, and 82%, respectively, in non-diabetic patients and 82%, 78%, and 58%, respectively, in diabetic patients.” [I.e., the proportion of diabetics who can expect to survive one year after an MI corresponds to the proportion of non-diabetics who can expect to survive five years.]

“In the mid-1990s there was considerable interest in the potential benefit of antioxidant nutrients and CVD risk reduction [100–103]. Since that time a series of randomized controlled intervention trials have failed to demonstrate a benefit of vitamin E or other antioxidant vitamin supplementation on CVD risk [104, 105]. The most recent work focusing on vitamins C and E confirm these earlier trials [106]. At this time the data do not support a recommendation to use antioxidant vitamins for the prevention or management of CVD. […] The three major dietary omega-3 polyunsaturated fatty acids (PUFAs) are alphalinolenic acid (ALA, 18:3n-3), eicosapentaenoic acid (EPA, 20:5n-3), and docosahexaenoic acid (DHA,22:6n-3). The later two fatty acids are sometimes referred to as very-long-chain n-3 fatty acids. […] a number of studies have reported an inverse association between dietary n-3 fatty acids, CVD and stroke risk [41]. Intervention data have demonstrated that EPA and DHA, but not ALA, benefit cardiovascular outcomes in primarily and secondary prevention studies [42] […] Of note, the relationship between arrhythmea and EPA and DHA has recently been questioned [45]. The major source of ALA in the diet is soybean and canola oils, whereas the major source of EPA and DHA is marine oils found in fish.”

“The lipoproteins are defined by their density, for example, very low density (VLDL), low-density (LDL), and high-density (HDL). In this instance, “density” is mostly related to the triglyceride and cholesterol content; the more lipids in a lipoprotein the lower its density, as measured by how readily it floats toward the top of a tube during ultracentrifugation. TG-rich lipoproteins transport an energy source, triglyceride, to muscle and adipose tissue for use and storage. TG-rich lipoproteins also contain cholesterol, and can deliver the cholesterol to peripheral tissues and the arterial wall. LDL is a transporter of primarily cholesterol from the liver to peripheral tissues. HDL also functions to transport cholesterol but in the reverse direction as VLDL and LDL, from peripheral tissues to the liver. Lipoproteins also are required to transport fat-soluble vitamins.”

“Relatively consistent evidence indicates that increasing the carbohydrate content of the diet at the expense of fat results in dyslipidemia [7–9]. The majority of the evidence suggests that carbohydrate-induced hypertriglyceridemia results from an increased rate of hepatic fatty acid synthesis [10,11] and subsequent production of hepatic triglyceride-rich particles, very-low-density lipoprotein (VLDL) […] Within the context of a stable bodyweight, replacement of dietary fat with carbohydrate results in higher triglyceride and VLDL cholesterol concentrations, lower HDL cholesterol concentrations and a higher (less favorable) total cholesterol to HDL cholesterol ratio [16–21]. […] Sedentary individuals characterized by visceral adiposity are at particularly high risk for carbohydrate-induced hypertrygliceridemia [9]. […] Studies performed in the mid 1960s demonstrated that changes in dietary fatty acid profiles altered plasma total cholesterol concentrations in most individuals […] Many studies have since confirmed these early observations using a variety of different experimental designs [29]. When carbohydrate is displaced by saturated fatty acids, LDL cholesterol concentrations increase, whereas when carbohydrate is displaced by unsaturated fatty acids LDL cholesterol concentrations decrease, with the effect of polyunsaturated fatty acids greater than monounsaturated fatty acids […] When carbohydrate is displaced by saturated, monounsaturated or polyunsaturated fatty acids, HDL cholesterol concentrations are increased, with saturated fatty acids having the greatest effect and polyunsaturated fatty acids having the least effect.”

“Some agents affect HDL and TG in the same direction. Drinking alcoholic beverages and postmenopausal estrogen treatment raise HDL and TG. Testosterone lowers HDL and TG. Since we do not have a way as yet to evaluate the function of HDL in reverse cholesterol transport [one of the chapters spends a significant amount of time on that one – there’s a lot more to be said about that stuff than what’s in the wiki article], we cannot be confident that these or any changes in HDL concentration affect atherosclerosis in the direction expected from the relation of HDL concentrations and CHD risk [59,65]. There is also no clear relation between genetic variants in enzymes or transporters in HDL metabolism that cause either very low or high HDL cholesterol concentrations and CHD [74].” [HDL is usually termed ‘good cholesterol’, but in reality it’s much more complicated than that. We are very sure by now that high ‘anything which is not HDL’ is bad for you, though – in fact:] “The combination of VLDL cholesterol and LDL cholesterol, named “non-HDL cholesterol” [2], or perhaps better “atherogenic cholesterol,” is a measurement that generally predicts CVD better than LDL-C [LDL-Cholesterol].”