Why the Niacin Flush May Be Surprisingly Beneficial to Your Health – Sandy Shaw
Why the Niacin Flush May
Be Surprisingly
Beneficial to Your Health
Why We Think It May Reduce the Risk of ALZHEIMER’S
DISEASE and Other Inflammatory Diseases,
Including Atherosclerosis, Type 2 Diabetes,
Ulcerative Colitis, Even Male Pattern Baldness
Sandy Shaw
WITHOUT PRIOR REVIEW BY DURK
It has not been read or reviewed by Durk Pearson, due to considerations that include, importantly, that time has run out and Shaw insisted on getting the paper into print without waiting. Shaw and Pearson have extensively discussed the elements of the paper so that the word “we” can be understood as the result of these discussions, but not based upon a reading of the Shaw paper by Pearson. Other than this disclaimer, everything remains as it is in the paper, with an expectation that more will be written on this subject to appear in future Durk & Sandy newsletters.
Here we explain why we think that the niacin flush (see description of flush just below this paragraph) may be a key part of the cardioprotective effect of high dose immediate-release (flushing) niacin’s highly protective effects on lipid metabolism, such as potent reductions of LDL and VLDL and triglycerides, while increasing HDL and, moreover, why the niacin flush may play an important role in reducing the risk of Alzheimer’s disease, atherosclerosis, type 2 diabetes, and other inflammatory diseases.
NOTE TO OUR READERS: This paper has become both much longer and included much more complex data and mechanistic detail to evaluate than the author (Sandy Shaw) originally anticipated. As a result, we have included in this first section of the paper the basic elements of how we believe the protective mechanism works and data on some diseases that we believe supports that interpretation. The next issue of our newsletter and subsequent issues will include the remaining parts of our analysis, with data from other diseases that we believe also appear to have significant risk reduction by the same mechanism, including detailed analysis of atherosclerosis, type 2 diabetes, and other diseases that show up in our literature searches. Sandy conceived the idea of examining the literature on the subject, read the papers discussed, and analyzed the data presented in the papers. Durk has read the analysis in detail and is in agreement with it.
WHAT IS THE FLUSH? Though we have heard it described as a transient skin reddening (from increased blood flow) accompanied by a sensation of heat associated with itching, we have come to realize that not everybody is feeling the same thing when they say “niacin flush.” The reason is that while both of us find the niacin flush AS WE EXPERIENCE IT to be pleasant, many people find it intolerable. Thus we think that what people who hate the flush mean when they say “niacin flush” is not a pleasant hot with mild itch but a hot with very unpleasant biting and burning sensations (as if being bitten by an insect or stabbed with tiny knives). In studying the mechanisms involved in the niacin flush to the extent they are now understood, we have an idea why many people are having this unpleasant flush. After we have discussed how we understand some of what causes the niacin flush, we will explain what we think may be making it intolerable for many. See below in section on “What Is Intolerable About the Niacin Flush?”
We understand, then, that some people can’t tolerate the niacin flush (caused by acute release of the prostaglandin PGD2) and, as a result, won’t use high dose plain niacin. It may be possible to reduce the flush to a tolerable level and, if so, it might be a much more sensible strategy (so you would end up with a flush like what we experience) than eliminating the flush if you want to get niacin’s lipid-lowering benefits. What has happened to niacin research, however, as a result of this crash program by certain drug companies to get rid of the flush, is that data on plain niacin and the effects of the acute release of prostaglandin D2 (which induces the flush) have to a considerable extent disappeared as more and more research focuses on the “extended release” or other non-flushing versions of niacin, which are not the same as plain niacin. “Extended release” niacin is not the same as plain niacin, for which extensive literature exists showing its potent lipid benefits. You do not see very much in the literature on head to head comparisons of “non-flushing niacin” (ersatz niacin) to plain (flushing) niacin in humans to identify what it is that the flush is doing.
THE KEY TO UNDERSTANDING THE EFFECTS OF THE PGD2-CAUSED NIACIN FLUSH IS TO REALIZE THAT IN ITS ACUTE RELEASE, PGD2 ACTS IN MANY MODEL SYSTEMS AS AN ANTI-INFLAMMATORY. IN ITS CHRONIC RELEASE, PGD2 APPEARS TO BE USUALLY PRO-INFLAMMATORY, but depending on the rate at which PGD2 is released, the amount released, and the state of inflammation in the tissue where it is released, you can get a pro-inflammatory or an anti-inflammatory effect. As the old saying goes, the devil is in the details and it is the rush to avoid considering the details so as to rapidly develop a non-flushing niacin that is leading to the rash abandonment by some drug companies and health practitioners of flushing (immediate-release or plain) niacin.
It has long been known that the prostaglandins PGD2 and PGE2 are responsible for inflammation induction (Haworth, 2007). Later in the biosynthetic pathways of these prostaglandins, anti-inflammatory circuits are induced (Haworth, 2007). Here is where we believe is the source of a major misunderstanding concerning the pro-inflammatory or anti-inflammatory effects of PGD2, the prostaglandin that causes the niacin flush: A CHRONICALLY HIGH LEVEL OF A SIGNALING MOLECULE, SUCH AS PGD2 (generally pro-inflammatory when at a chronically high level) CAN INTERFERE WITH SIGNALS BY ACUTELY RELEASED (PULSATILE) AMOUNTS OF THAT SIGNALING MOLECULE (generally anti-inflammatory) BY THE ACUTE SIGNAL SIMPLY BEING “LOST” IN THE NOISE OF THE CHRONICALLY HIGH LEVEL OF THAT MOLECULE. Hence, we think that chronically high levels of PGD2 are likely to prevent or reduce the effect of acute signals of PGD2 that would otherwise be anti-inflammatory. See sections on Alzheimer’s disease (AD) below, where chronically high PGD2 signaling is thought to be a major cause of the neurodegenerative features characteristic of AD (Maesaka, 2013).
- Haworth and Buckley. Resolving the problem of persistence in the switch from acute to chronic inflammation. Proc Natl Acad Sci U S A. 104(52):20647-8 (2007).
- Maesaka, Sodam, Palaia, et al. Prostaglandin D2 synthase: apoptotic factor in Alzheimer plasma, inducer of reactive oxygen species, inflammatory cytokines, and dialysis dementia. J Nephropathol. 2(3):166-80 (2013).
In this section, we look at PGD2 release in model systems:
DATA ON ACUTE AND CHRONIC RELEASE OF PGD2
Prostaglandin D2 in the Resolution of Inflammation
Ulcerative Colitis
There is considerable difficulty in interpreting the huge amount of scientific literature on prostaglandins when you read that a certain tissue level is associated with a certain stage of an inflammatory disease. Most of the literature that we’ve seen is bogged down with difficulties in interpreting the role of the prostaglandin in the inflammatory process because of not knowing whether the tissue level measured is part of a pulsatile release or a chronic release.
A very interesting recent paper (Vong, 2010) was published by scientists who believe that the increased expression of prostaglandin D2 synthesis and its EP1 receptor that they detected in individuals in long-term remission from ulcerative colitis suggest that the release of PGD2 in response to acute releases of inflammatory stimuli could be important antiinflammatory protection to maintain colonic mucosal homeostasis.
To study this phenomenon in human patients, the scientists took rectal biopsies from patients with active ulcerative colitis, which have elevated levels of PGE2, PGI2, and PGF2alpha. “Several studies of experimental colitis suggest important roles for PGD2 in promoting the resolution of inflammation and long-term alterations in colonocyte and barrier function …” They examined PGD2 levels in biopsies for ulcerative colitis patients, comparing them with those of healthy individuals who had no prior history of UC or those from healthy individuals who had experienced a prior bout of UC but had been in remission without medication for >4 years. “We observed a pronounced elevation of PGD2 synthesis and DPI receptor expression only in healthy individuals with a prior history of UC. In these individuals, as has been observed in animal studies, the elevated mucosal PGD2 may contribute to the maintenance of colonic tissue homeostasis and possibly, also to an increased risk of colorectal cancer.”
One difficulty here is that the biopsy represents a snapshot of PGD2 levels at one point. Was the PGD2 being released as an acute pulse at that point or was it being measured at a chronic level? The authors are hot to track down the cause of this association (the apparent anti-inflammatory effect of PGD2 in maintaining remission in UC) and say, “we believe that PGD2 plays an important role in the initial maintenance of mucosal homeostasis.” We might as well add our own hypothesis to the mix. On the basis of other data on anti-inflammatory effects of pulsatile release of PGD2, we would expect the most protective effects of PGD2 release in this model to occur at an optimal level of a pulse of PGD2 released over a limited time period in response to pro-inflammatory stimuli, but not too little to prevent inflammation so as to maintain remission, or too much to increase PGD2 to levels that would potentiate inflammation and, perhaps, be part of an increased risk of colorectal cancer the scientists here mention as a possibility.
For example, the scientists note that their results are “consistent with studies of rodents in which prolonged elevations of PGD2 synthesis were observed after resolution of colitis.” This prolonged elevation contributed not only to resolution of inflammation but also to “long term alterations in epithelial function, some of which may have contributed to an increased susceptibility to colon cancer.” This suggests that the protective response of the immune system in some animals and humans of increasing PGD2 release in response to inflammation to modulate that inflammation may go too far and result in long-term adverse effects such as increased risk of colon cancer. It appears to us that pulsatile PGD2 release a few times a day with immediate release niacin supplementation may offer better protection against a variety of inflammatory diseases.
- Vong, Ferraz, Panaccione, et al. A pro-resolution mediator, prostaglandin D(2), is specifically up-regulated in individuals in long-term remission from ulcerative colitis. Proc Natl Acad Sci U S A.107(26):12023-7 (2010).
MALE PATTERN BALDNESS (Nieves, 2014)
The hair follicle in male pattern baldness balding areas (but not in normal hair of balding men) have chronically high levels of prostaglandin D2 accompanied by lower levels of prostaglandin E2. One way that minoxidil has been found to work is by increasing prostaglandin E2, which in this model “normalizes” the PGD2/PGE2 ratio. However, PGE2 is an inflammatory molecule, so you wouldn’t want to increase it very much, and that is undoubtedly the “secret” of minoxidil, to NORMALIZE the ratio of PGD2/PGE2 so as to eliminate a chronically high PGD2 level.
One of the signals of the catagen phase of hair growth, where hair growth ceases for a time and some hair follicles die, is the release of very large amounts (7 fold higher than baseline) of PGD2 (Nieves, 2014). For that reason, there is interest in blocking PGD2 as a “treatment” for balding. But once again, there is a risk that blocking PGD2’s unwanted effects will also block important beneficial effects of PGD2. This, not surprisingly, is a major problem in medicine, that the change you want in a certain tissue at a certain time and by a certain amount may cause harm elsewhere where you do not want that change.
- Nieves and Garza. Does prostaglandin D2 hold the cure to male pattern baldness? Exp Dermatol. 23(4):224-7 (2014).
ANTI-INFLAMMATORY ROLE OF PGD2 IN ACUTE
LUNG INFLAMMATION (Murata, 2012)
The authors of this paper (Murata, 2012) studied the role of PGD2 signaling in acute lung injury (ALI). Administering endotoxin (lipopolysaccharide (LPS), a potent bacterial inflammatory factor) increased edema and neutrophil infiltration into the wild type mouse lung, typical effects seen in inflammation. “Treatment with either an agonist to the PGD2 receptor, DP, or a degradation product of PGD2, 15-deoxy-delta12,14-PGJ2, exerted a therapeutic action against ALI.” The effect of LPS inhalation by the wild type mice peaked on day 1, hence this was an acute effect. The authors found, however, that whether PGD2 had an anti-inflammatory effect or a pro-inflammatory effect depended upon the stage at which the PGD2 was administered, with PGD2 at later stages of ALI being anti-inflammatory (reducing the invasiveness of neutrophils).
- Murata, Aritake, Tsubosaka, et al. Anti-inflammatory role of PGD2 in acute lung inflammation and therapeutic application of its signal enhancement. Proc Natl Acad Sci. 110(13):5205-10 (2013).
ALZHEIMER’S DISEASE: MICROGLIAL PGE2
(PROSTAGLANDIN E2) SIGNALING VIA EP4
RECEPTOR SUPPRESSES ALZHEIMER
ASSOCIATED INFLAMMATION
A signaling system is reported here (Woodling, 2014) that the authors found to regulate an important protective anti-inflammatory mechanism in the early stages of Alzheimer pathology that decreases significantly (along with its protective effect) as the disease progresses. This is a signal from the prostaglandin PGE2 to its EP4 receptor. See section below on the PGE2 receptor system (EP1, 2, 3, and 4) and new findings suggesting that it is a key to some of the antiinflammatory properties of DHA (docosahexaenoic acid, an omega 3 fatty acid found in fish oils) and possibly that of curcumin.
As reported in a 2012 paper (Ruan, 2012), the EP1 receptor for PGE2 appears to be the key target for DHA and fish oils. There they showed that, in cultured stromal cells, the IC50 for fish oil (that is, the amount that inhibited 50% of the PGE2 activity) was 18 mg/L or 54 μM. The authors calculated that, for a 150 pound human containing 4-5 liters of blood, “consuming 100 mg. fish oil should yield IC50 results.” (This depends, of course, on how the DHA partitions in the blood and tissues, but the calculation provides a crude estimate.) The authors then indicate that they would recommend taking 500-1000 mg fish oil daily on the basis of their findings.
It is interesting to note the opposing effects of PGD2 (the prostaglandin that induces the niacin flush) and PGE2 in the balding model (above), where chronically high PGD2 resulted in suppression of PGE2. A pulsatile release of PGD2 (an ACUTE release) as in the niacin flush would be anti-inflammatory, not pro-inflammatory as with chronically high PGD2. Hence, you could see an INCREASE in PGE2 by suppressing chronically high PGD2. The balding model, in fact, shows hair growth and the cessation of hair follicle death resulting from slight modulation in the ratio of PGD2/PGE2, in which PGE2 is increased, while chronically high PGD2 levels are reduced to “normalize” the ratio. The niacin flush causes pulsatile, not chronic, release of PGD2. It is relevant to note that another paper (see just below) describes CHRONICALLY high PGD2 signalling in full-blown Alzheimer’s. We predict, in fact, that high dose niacin in the immediate-release flushable form will REDUCE the risk of Alzheimer’s, and that getting rid of the flush would probably eliminate this protective effect. If you could get rid of the flush and still retain all the protective benefits of the flush, then fine, go ahead and get rid of it. But so far, the focus seems to be on suppressing the flush without adequately understanding what the flush has to do with the protective effects of immediate release niacin.
Also, note in the urate crystal inflammation model (below) that a 5.2 fold pulsatile (acute) increase in PGD2 was anti-inflammatory, decreasing inflammatory signaling by PGE2. The opposing effects of certain dose and time-dependent releases of PGD2 and PGE2 would appear to be a system to examine closely in relation to Alzheimer’s.
- Woodling, Wang, Priyam, et al. Suppression of Alzheimer-associated inflammation by microglial prostaglandin-E2 EP4 receptor signaling. J Neurosci.34(17):5882-94 (2014).
- Ruan and So. Screening and identification of dietary oils and unsaturated fatty acids in inhibiting inflammatory prostaglandin E. BMC Complement Altern Med.12:143 (2012).
ALZHEIMER’S DISEASE:
INCREASED LEVELS OF A FORM OF
PGD2 SYNTHASE
In a very complex analysis (Maesaka, 2013), researchers found that a form of PGD2 synthase, L-PGDS, can in a chain of biochemical reactions, convert arachidonic acid to 15deoxyPGdelta12,14 J2(15dPGJ2), the primary ligand for peroxisome proliferator activator receptor gamma (PPARgamma), and that 15dPGJ2 has been reported to induce apoptosis in human astrocytes and cortical neurons, which could be prevented by inhibitors of L-PGDS, such as IGF, insulin, and erythropoeiten as well as PGE1, PGE2, and COX2 and caspase inhibitors. The authors identified L-PGDS “as a dominant inducer of apoptosis in AD plasma,” presumably by increasing PGD2 signalling to a chronically high level.
- Maesaka, Sodam, Palaia, et al. Prostaglandin D2 synthase: apoptotic factor in Alzheimer plasma, inducer of reactive oxygen species, inflammatory cytokines, and dialysis dementia. J Nephropathol. 2(3):166-180 (2013).
Interestingly, another paper (Ryan, 2008) reports that 15dPGJ2 (that as noted just above induces apoptosis in brain astrocytes and cortical neurons) impairs phosphatidylcholine synthesis by promoting cysteine cross-linking in the enzyme cytidyltransferase alpha. This cross-linking could be reduced by N-acetylcysteine (Ryan, 2008).
The sleep impairments observed in Alzheimer patients may be, at least to some extent, linked to chronically high PGD2 signaling, as PGD2 is an important sleep-inducing molecule (Urade 1999). The signal of a transient pulse of a substance is lost in the noise of a continuously high background level of that substance.
- Ryan, Chen, Vennalaganti, et al. 15-deoxy-delta12,14-prostaglandin J2 impairs phosphatidylcholine synthesis and induces nuclear accumulation of thiol-modified cytidyltransferase. J Biol Chem.283(36):24628-40 (2008).
- Urade and Hayaishi. Prostaglandin D2 and sleep regulation. Biochim Biophys Acta. 1436:606-15 (1999).
REDUCTION OF URATE CRYSTAL INFLAMMATION
BY ACUTELY ELEVATED PGD2
After struggling through the analysis of prostaglandin D2 synthase’s link to apoptosis in Alzheimer’s disease (just above this paragraph), if you did, you may be hoping for something a little simpler. This one is.
Here, researchers found (Jung, 2007) that in mice fed root extracts of traditional oriental medicinal plants,* inflammation elicited by injecting 2 mg of monosodium urate crystals into the pouch resulted in a rapid and dramatic decrease in the measured inflammatory parameters, including neutrophil density, IL-6 and TNFalpha mRNA. Leukocyte count, IL-6, prostaglandin E2, along with prostaglandin D2 were examined in the pouch exudate. Remarkably, the concentration of the potentially anti-inflammatory Prostaglandin D2 rose 5.2 fold. The authors of this 2007 paper were very excited about these results and thought this could point to a novel way to treat inflammation, the cause of the intense pain of uric acid crystals in gout. They seem to have been right, but nothing appears to have come of this.
* Dried roots of Acanthopanax senticosus, Angelica sinensis, and Scuttelaria baicalensis.
- Jung, Schumacher, Kim, et al. Reduction of urate crystal-induced inflammation by root extracts from traditional oriental medicinal plants: elevation of prostaglandin D2 levels. Arthritis Res Ther.9:R64 (2007).
PGD2 IN THE SKIN
A recent paper (Shimura, 2010) reports on the role of PDG2 in allergic responses in the skin, focusing on mast cells expressing the hematopoeitic PGD synthase found in dendritic cells. They discussed rapid excretion of PDG2 in response to various allergens, including an irritant compound. “A possible anti-pruritic [anti-itch] potential of PGD2 in the scratching behavior of mice was recently proposed.” [It was AFTER Sandy performed the experiment on her itchy skin described just below that she read about this finding. Serendipity!] When released rapidly in response to allergens, PGD2 can act as an anti-inflammatory, while when released in excess quantities it exacerbates the allergic response (Shimura, 2010).
- Shimura, Satoh, Igawa, et al. Dendritic cells express hematopoietic prostaglandin D synthase and function as a source of prostaglandin D2 in the skin. Am J Pathol.176:227-37 (2010).
SANDY’S ITCHY SKIN—DISCOVERING THE EFFECT OF
ELIMINATING THE NIACIN FLUSH THE HARD WAY
Sandy had developed an itchy skin condition (probably a result of severe hypothyroidism) and had, as a result, discontinued high dose niacin about a month ago because of increased itchiness and of stabbing sensations (see description of the niacin flush above) during the flush. The itchiness got worse until it became such a serious problem that she had to take two prescription drugs to keep the itchiness under control. Hydrocortisone cream didn’t help at all, which is consistent with the reported effect of chronically elevated PGD2 in blunting the antiinflammatory effect of corticosteroids (Barnes, 2009). This doesn’t PROVE that it was chronically high PGD2 in her skin that made the hydrocortisone salve ineffective, but is consistent with data showing that effect. As a matter of fact, Sandy has a mild case of COPD, which is reported to exhibit resistance to the antiinflammatory effects of corticosteroids, suggesting the possibility that she has chronically elevated PGD2 in her lungs. The fact that it hasn’t gotten progressively worse over the years, as COPD typically does, MAY be due to her ingestion of high dose immediate-release niacin which could be reducing the inflammatory activity by discharging the release of PGD2 via pulses, thereby preventing chronic PGD2 release as a sort of constant dribble rather than as pulses. This is our hypothesis. There might be another way to explain all this, and we certainly can’t prove there isn’t (given that it is impossible to disprove a negative), but we think our explanation is quite plausible and consistent with all the data we’ve seen.
It is interesting to note that curcumin restores corticosteroid sensitivity in monocytes exposed to oxidants by maintaining HDAC-2 (histone deacetylase 2) levels (Gonzalez, 2012); HDAC-2 levels are known to be reduced in COPD. Could this be another example of a natural substance that reduces chronically high levels of PGD2? It is certainly consistent with a large amount of the literature on COPD, PGD2, and HDAC-2.
- Role of HDAC2 in the pathophysiology of COPD. Annu Rev Physiol.71:451-64 (2009).
- Gonzalez, Ballester, Lopez-Posadas, et al. Effects of flavonoids and other polyphenols on inflammation. Crit Rev Food Sci Nutr. 51(4):331-62 (2011).
EXPERIMENT: When I (Sandy) realized that it might be the loss of the flush that was causing my monster skin itch, I took 400 mg of niacin on an empty stomach to induce the flush. After that, the flush ensued and went on for the normal period of time it usually does after I take niacin, during which the itching was intensified, the itching then subsiding to a very low level. Plus, a little bonus, my painful knee osteoarthritis had become much less painful when I went out to the supermarket a few minutes later. It had become worse during the month I was off high-dose niacin. NOTE: I still took Personal Radical Shield during this period and so was getting between 250 and 500 mg/day of niacin, but my regular dose of niacin was around 2 grams a day in addition to that.
For a relationship between osteoarthritis and prostaglandin D2, see:
- Zayed, Li, Chabane, et al. Increased expression of lipocalin-type prostaglandin D2 synthase in osteoarthritic cartilage. Arthritis Res Ther. 2008, 10(6):R146 doi:10.1186/ar2581 (2008).
Mitigation of Inflammation with Foods
A 2012 paper (Wu, 2012) reported that polyphenols and other compounds found in foods such as fruits, berries, vegetables, nuts, whole grains, and foods of marine origin contain components can “play an important role in attenuating and mitigating chronic pro-inflammatory processes associated with chronic diseases,” such as atherosclerosis, ischemic heart disease, cancer, obesity, inflammatory bowel disease, Crohn’s disease, diabetes and autoimmnune diseases. Surprisingly few papers on the mitigation of inflammation discuss prostaglandins explicitly, revealing that there is an immense area here where increased knowledge could promote improvements in controlling the chronic inflammatory diseases associated with aging.
When you control a biochemical pathway to regulate processes taking place far downstream, it is usually best to think of ways to regulate closer to the downstream site that is a problem because regulation at the upper end of the chemical pathway will affect many processes that have nothing to do with your problem and may produce unwanted off-target effects. THAT, in fact, is one of the major problems with statins … that their effects are taking place far upstream in the process of synthesizing cholesterol (Cederberg, 2015) and there are frequent undesired effects such as myopathy and an alarmingly high increased risk of developing type 2 diabetes.
- Wu and Schauss. Mitigation of inflammation with foods.J Agric Food Chem.60:6703-17 (2012). Cederberg, Stancakova, Yaluri, et al. Increased risk of diabetes with statin treatment is associated with impaired insulin sensitivity and insulin secretion: a 6 year follow-up study of the METSIM cohort. 58:1109-17 (2015).
STATINS IN COMBINATION WITH NIACIN
As we note in the paragraph above, statins have been found to reduce insulin sensitivity, and this is associated with a large increase in the risk of developing type 2 diabetes. The combination of statins with niacin also have a significant effect in reducing the beneficial effects of increased HDL cholesterol that is seen with immediate release niacin (Keene, 2014). The mechanism that causes statins to increase the risk of type 2 diabetes are, we think, a likely place to look for what causes this adverse effect of statins on the protective effect of niacin on reduced risk of non-fatal heart attacks, the most common kind of heart attack.
Most Heart Attacks Are Non-Fatal, So the Reported
Highly Protective Effect of Niacin Against the
Incidence of Non-Fatal Heart Attacks Is Important
In a paper (Keene, 2014) providing a meta-analysis of 117,411 patients, very interesting differences between the effects of niacin taken by patients NOT RECEIVING STATINS (BEFORE THE STATIN ERA) and those who, later, were taking niacin and statins emerged. Very statistically significant results showed that in those patients NOT taking statins, niacin was associated with a significant reduction in non-fatal heart attacks (myocardial infarction) (odds ratio was 0.69, 0.56 to 0.85, p=0.0004). However, in studies where statins were already being taken, niacin showed no significant effect on the incidence of non-fatal heart attacks. The researchers say, “In the current era of widespread use of statins in dyslipidaemia, substantial trials of these three agents [niacin, fibrates, or CETP inhibitors, all of which increase HDL levels] do not support this concept [that increasing HDL would generally reduce cardiovascular events].” Statins interfere with an important protective effect of niacin. It is important to note that NON-FATAL MYOCARDIAL INFARCTIONS are the most COMMON type of heart attack, so reduction of these heart attacks is not at all unimportant, though the word FATAL may be somewhat distracting from the significance of these results.
Studies with statins HAVE repeatedly shown that reductions in LDL cholesterol with statins “has repeatedly been found to reduce cardiac events and all cause mortality in the setting of both secondary and primary prevention.” However, the increased protection against cardiovascular events expected by increased HDL has not been seen. See paragraph above. Hence, something in the interaction of statins with niacin and fibrates, where niacin and fibrates increase HDL cholesterol and which (when taken WITHOUT CONCURRENT INGESTION OF STATINS), reduces the risk of non-fatal myocardial infarctions, results in a loss of that protective effect of niacin or fibrates.
The studies with niacin were confounded by the fact that some of the patients were given aspirin or laropriprant to inhibit flushing. Laroopriprant is thought to interfere with prostaglandin pathways which, as you will see below in our discussion of prostaglandins, could be very important, and the authors of this paper (Keene, 2014) suggest that this effect “could confound the effect of niacin.”
- Keene, Price, Shun-Shin, and Francis. Effect on cardiovascular risk of high density lipoprotein targeted drug treatments niacin, fibrates, and CETP inhibitors: meta-analysis of randomised controlled trials including 117,411 patients. 349:g4379 (2014) doi: 10.1136/bmj.g4379 (18 July 2014).
2012 REVIEW PAPER ON PGD SYNTHASE AND PGD2
IN IMMUNE RESPONSE (Joo, 2012)
A good recent review of PGD synthase and PGD2 described a variety of model systems in which PGD2 exhibited a proinflammatory or antiinflammatory effect. For example, a paper is cited in which a human model of an acute inflammatory response induced by the administration of LPS provides data “strongly” supporting anti-inflammatory effects of PGD2. PGD2 and its cyclopentenone prostaglandin derivatives are reported by the review’s authors to have antiinflammatory properties with functions in the resolution of inflammation, and “there is considerable interest in the importance of PGD2 and its distal products in the mediation and resolution of inflammation.” (Joo, 2012). Other models show pro-inflammatory effects. The devil is in the details of the amounts released, the time course over which the release takes place, and the inflammatory milieu of the tissue involved (as we have noted above, an acute signal of a molecule released into an environment with a high chronic level of that molecule may not convey the acute signal very well or at all).
- Joo and Sadikot. PGD synthase and PGD2 in immune response. Mediators Inflamm. 2012:503128. doi: 10.1155/2012/503128 (2012).
Interestingly, in sleeping sickness, PGD2 is increased and the microorganism responsible for the disease has been shown to induce the production of PGD2 in culture. PGD2 is a well-known sleep promoting prostaglandin. The review authors (Joo, 2012) published a paper earlier in which they showed that PGD2 inhibits a “key proinflammatory immunoglobulin cell surface receptor TREM-1 in vitro in macrophages.” In another of the authors’ papers, they showed that the administration of PGD2 in a mouse model of P. aeruginosa lung infection resulted in enhanced clearance of P. aeruginosa from the lungs. Moreover, their study showed that mice that had COX-2 inhibited (via knockout) had enhanced clearance of the bacterium and that this effect was related to a decrease in PGE2. There we see the interaction and apparent opposing effects between PGD2 and PGE2 that has appeared in other studies.
This paper is useful for presenting an analysis of several models of inflammation and the effects of various forms of PGD2-synthase and PGD2.
PGE2 RECEPTOR SYSTEM
The inducible and inflammatory COX-2 pathway synthesizes the release of inflammatory amounts of PGE2 (Ruan, 2012). There are three different PGE2 synthases, enzymes that manufacture PGE2 from arachidonic acid. The PGE2 produced is then received by one of four PGE2 receptors, EP1, 2, 3, and 4), which is associated with pain, vascular diseases, and cancer cell growth (Ruan, 2012). DHA’s anti-inflammatory effects are mediated, at least in part, by its action at the EP1 receptor of PGE2 (Ruan, 2012) while the antiinflammatory action of curcumin is reported to be via the reduction of IL-1 beta-stimulated microsomal PGES. PGES enzymes convert the prostaglandin PGH2 to PGE2 (Kats, 2013).
- Ruan and So. Screening and identification of dietary oils and unsaturated fatty acids in inhibiting inflammatory prostaglandin E. BMC Complement Altern Med.12:143 (2012).
- Kats, Bage, Georgsson, et al. Inhibition of microsomal prostaglandin E synthase-1 by aminothiazoles decreases prostaglandin E2 synthesis in vitro and ameliorates experimental periodontitis in vivo. FASEB J.27:2328-41 (2013).
YIN-YANG BALANCE BETWEEN
PROINFLAMMATORY AND ANTI-INFLAMMATORY
PROSTAGLANDINS
This is what some researchers refer to as a “yin yang” system, where you have (for example) prostaglandins PGE2 and PGD2 working a balance of pro-inflammatory and anti-inflammatory effects depending on how much is released, the time course of the release, and the inflammatory milieu of the tissue where they are released. Acute inflammation onset and resolution have also been identified in a paper (Rajakariar, 2007) as being regulated by the balance of PGD2 and 15d-PGJ2.
- Rajakariar, Hilliar, Lawrence, et al. Hematopoietic prostaglandin D2 synthase controls the onset and resolution of acute inflammation through PGD2 and 15-deoxy-delta12,14 PGJ2. Proc Natl Acad Sci U S A. 104(52):20979-84 (2007).
“The devil may be in the details,
But so are the angels.”
—Sandy Shaw
It would appear that all of the most important chronic diseases of man, including cardiovascular disease, cancer, and Alzheimer’s disease are critically regulated by this sort of balancing act. The devil may be in the details, but so are the angels. Another way of putting it is that you shouldn’t be in too great a rush to throw out the devils until you understand whether, by doing so, you are throwing out some or all of the angels.
What Is Intolerable About the Niacin Flush?
The niacin flush appears to be the devil in the details that most disturbs people who would use niacin but can’t stand the flush. There are many interacting biochemical pathways regulating pro-inflammatory and anti-inflammatory effects in the skin, but it would be helpful to isolate some particularly important players in the niacin flush. First, of course, is PGD2, the acute release of which is closely associated with the strength and timing of the flush in relation to when you took the niacin. Preventing the acute release of PGD2 comes close to getting rid of the flush, but doesn’t eliminate it entirely, pointing to other regulatory factors being involved (Kamanna, 2009).
But there are a number of different prostaglandins that have interactions with each other and can counter-regulate each other, and so on. Importantly, as we describe above, PGE2 seems to be in a balancing act with PGD2 in some models of inflammation, so that the nastiness of the niacin flush MAY be linked to the release of PGE2 (our supposition). A prediction of this hypothesis would be that inhibitors of the release of PGE2 would reduce how much of the nasty burning and sensations of insects biting would occur during the flush. A recent paper (Gonzalez, 2012) reported that a study of the inhibitory effects of flavonoids on the release of PGE2 in peritoneal macrophages found that some flavonoids were as effective as aspirin in this inhibitory activity, including quercetin, resveratrol, apigenin, genistein, or kaempferol, but the authors were surprised that luteolin, fisetin, or morin did NOT inhibit the PGE2 release in the peritoneal macrophages.
Here, again, we have another remarkable little clue on the psychodynamics of the niacin flush: a paper (Papaliodis, 2008) that reports that luteolin suppresses the niacin flush!! Another piece for the puzzle of what the niacin flush is all about. Moreover, to add a little additional spice to this, it has been reported (Ren, 2009) that some flavonoids produce a little skin flush of their own. The paper (Gonzalez, 2012) also mentioned that flavonols appear to exert the highest activity (in papers on anti-inflammatory effects via inhibition of lipoxygenases). Cocoa is an excellent source for flavonols. We are currently taking a flavonol-enriched supplement (expensive, however) called Cocoa-Via® as a source of these flavonols.
- Kamanna, Ganji, Kashyap. The mechanism and mitigation of niacin-induced flushing. Int J Clin Pract.63(9):1369-77 (2009).
- González, Ballester, López-Posadas, et al. Effects of flavonoids and other polyphenols on inflammation. Crit Rev Food Sci Nutr. 51(4):331-62 (2011).
- Papaliodis, Boucher, Kempuraj, and Theoharides. The flavonoid luteolin inhibits niacin-induced flush. Br J Pharmacol. 153:1382-7 (2008).
- Ren, Kaplan, Hernandez, et al. Phenolic acids suppress adipocyte lipolysis via activation of the nicotinic acid receptor GPR109A (HM74a/PUMA-G). J Lipid Res. 50:908-14 (2009).
REDUCING THE NIACIN FLUSH
A recent paper (Jacobson, 2010) that discussed in amazing detail how to reduce the niacin flush begins by noting that “[n]iacin is the most effective lipid modifying agent for raising high density lipoprotein [HDL] cholesterol …” The author explains that in clinical trials, over 60% of niacin users experienced mild or moderate flushing, with 5% to 20% of patients discontinuing niacin therapy because of the flush.
The author (Jacobson, 2010) goes on to explain various ways to reduce the flush, including taking niacin with meals or at bedtime with a low-fat snack and avoiding exacerbating factors such as alcohol or hot beverages. He also indicates that taking 325 mg of aspirin along with niacin suppresses the flush. However, aspirin has many effects on prostaglandin chemistry and as prostaglandins appear to play an important role in the beneficial cardioprotective effects of niacin, we do not recommend taking aspirin along with niacin to reduce the flush.
- A “hot” topic in dyslipidemia management—“How to Beat a Flush:” optimizing niacin tolerability to promote long-term treatment adherence and coronary disease prevention. Mayo Clinic Proc. 85(4):365-79 (2010).
IN THE BEGINNING,
THERE WAS IMMEDIATE RELEASE NIACIN
To get some feel for the data on clinical effects of immediate release niacin over its first fifty years of use, a paper published in 2005 in the Journal of Internal Medicine (Carlson, 2005) provides an informative review. At this point, the switchover to the non- flushing niacin had not occurred to a great extent, as Niaspan had just become available. Hence, the review focused largely on immediate release niacin that caused a strong flush. Indeed, the author of this review (Carlson, 2005) had been doing pioneering work on niacin for over 40 years.
It was discovered early (some of the early data were published by the author of this review with a coauthor (see Carlson, 1962) that nicotinic acid (niacin) lowers free fatty acids by inhibiting the mobilization of free fatty acids from adipose tissue, a process called lipolysis; the antilipolytic effect of niacin is now considered a major source of the benefits of niacin. The early researchers found that the inhibition of mobilization of free fatty acids by niacin did not change the overall energy metabolism in the heart but “switched its oxidative metabolism from lipids to carbohydrates.” This is a major effect of niacin in its protective cardiovascular role. The inhibitory effect of niacin on the rise of free fatty acids and triglycerides that occurs during emotional stress was reported in a 1962 paper (Carlson, 1962) on experiments in humans.
As of the date of this review, the author stated, “[i]t is now generally accepted that nicotinic acid is the most powerful drug for raising the concentration of HDL, in particular the subspecies HDL2.” He cites data from a study that found HDL cholesterol to rise by 50% in hyperlipidemic patients, but the subfraction of HDL2, the large HDL particles, increased by almost 100%.
Interestingly, the author referred to data showing that at that time researchers had already identified the niacin flush as being due to the release of prostaglandins (by experiments using the prostaglandin synthesis [cyclooxygenase] inhibitor indomethacin) and found that indomethacin markedly reduced the flush. Other studies published before the Carlson 1962 review suggested that PGD2 might be the prostaglandin responsible for the flush.
Carlson in that review also mentioned the discovery that niacin had fibrinolytic (clot busting) effects, having been shown to decrease the plasma levels of fibrinogen by 15% by inhibiting its synthesis by plasminogen activator inhibitor 1. The increased expression of PAI-1 was discussed by Carlson as being closely associated with hypertriglyceridemia (Carlson, 1962). A recent paper (Song, 2012) proposed that PGD2, which the authors found to be synthesized by COX-1 in platelets in both mice and humans, may “function as a homeostatic response to thrombogenic and hypertensive stimuli and may have particular relevance as a constraint on platelets during [flushing] niacin therapy.”
- Nicotinic acid: the broad-spectrum lipid drug: a 50th anniversary review.J Intern Med. 258:94-114 (2005).
- Carlson and Oro. The effect of nicotinic acid on the plasma free fatty acids; emonstration of a metabolic type of sympathicolysis. Acta Med Scand. 172:641-5 (1962).
- Song, Stubbe, Ricciotti, et al. Niacin and biosynthesis of PGD2 by platelet COX-1 in mice and humans. J Clin Invest. 122(4):1459-68 (2012).
CONCLUSION:
In conclusion, there is a rush to develop a way to evade the niacin flush that causes an acute release of PGD2 in the use of niacin therapy to treat hyperlipidemia. This incredible rush, that has resulted in nearly a discontinuation of research on immediate release niacin is being done almost entirely for the purpose of making a non-flushing niacin available for commerce because it could make somebody a lot of money. Because of the entwined interests of government regulatory authorities at the FDA (users’ fees are a very important part of the FDA’s budget) and certain pharmaceutical companies which could benefit by the availability of such a non-flushing niacin product, this work is not (in our opinion) properly scientifically evaluating what it means to eliminate the niacin flush and what it might cost patients using the new forms of niacin (taking into account that many such patients would never take the flushing niacin in the first place) by actually comparing the two different forms of niacin. No such evaluation seems to be taking place. Just thought we ought to mention it …
The correct question that should be asked is not whether extended-release niacin has equivalent benefits to regular niacin—it doesn’t —but what benefits does it provide to which patients and what are its risks, questions that should be asked of any medicine. The almost desperate pursuit of “equivalence” between the two resembles a morbid fear of admitting that somebody might be giving up something of value by taking extended release niacin rather than immediate release niacin and—horror of horrors—finding out exactly what that something of value really is.
HUMAN STUDIES WITH “EXTENDED RELEASE” OR
“PROLONGED RELEASE” NIACIN LEAVE LITTLE
ROOM FOR DOUBT ABOUT THE LACK OF
EQUIVALENCE WITH THE BENEFITS OF
IMMEDIATE RELEASE NIACIN
- A human study of extended-release niacin on lipoprotein particle size, distribution and inflammatory markers in patients with coronary artery disease (Kuvin, 2006) found that compared with subjects who received placebo, 3 months of ER niacin resulted in significantly increased though “relatively small” increases in HDL and no significant change in total LDL levels. Regular niacin provides a very significant and clinically meaningful reduction in LDL. The patients participating in this study already had well-controlled LDL (that is, by prior treatment not including any form of niacin), so this study really did not explore the differences between ER niacin and immediate release niacin on the regulation of lipids.
- Kuvin, Dave, Sliney, et al. Effects of extended-release niacin on lipoprotein particle size, distribution, and inflammatory markers in patients with coronary artery disease. Am J Cardiol. 98:743-5 (2006).
- In another human study (Plaisance, 2008), the effects of aerobic exercise and ER niacin were examined in 15 men with the metabolic syndrome. ER niacin lowered fasting but had no effect on the postprandial triglyceride AUC (amount under the curve), while it did decrease the insulin AUC. Immediate release niacin, however, reduces fasting triglycerides by 20-50%.
- Plaisance, Mestek, Mahurin, et al. Postprandial triglyceride responses to aerobic exercise and extended-release niacin. Am J Clin Nutr. 88:30-7 (2008).
- In a third human trial (Jafri, 2009), ER niacin reduced LDL particle number and increased the number of HDL particles without changing total LDL cholesterol in patients with stable coronary artery disease. But the very significant lowering of LDL is considered a major protective feature of immediate release niacin.
- Jafri, Alsheikh-Ali, Mooney, et al. Extended-release niacin reduces LDL particle number without changing total LDL cholesterol in patients with stable CAD [coronary artery disease]. J Clin Lipidol. 3:45-50 (2009).
- In another human trial (Westphal, 2006) (randomized, placebo-controlled double blind, 30 men with metabolic syndrome), a short term (6 week) treatment with ER niacin increased adiponectin by 56% and leptin by 26.8% but there was no change in the levels of the proinflammatory factors TNFalpha, IL-6 and C-reactive protein, no improvement in endothelial function (as measured by FMD), and an actual deterioration in glucose and insulin parameters. Despite increased levels of adiponectin, the authors note that this “fails to improve atheroprotective functions attributed to adiponectin, such as insulin sensitivity, anti-inflammation, and endothelial function.”
- Westphal, Borucki, Taneva, et al. Extended-release niacin raises adiponectin and leptin. 193:361-5 (2007).
A review paper (Vosper, 2009) looking at niacin’s effects on prostaglandin chemistry, came to the conclusion that “recent advances in understanding of the contribution of prostaglandin metabolism to vascular wall health suggest that some of the beneficial effects of niacin may well result from activation of the same pathways responsible for the adverse [the flush] reactions. The purpose of this review is to emphasize that the search for agonists that show higher tolerability must take into account all aspects of signaling [by prostaglandin D2] through this [the DP1] receptor.”
- Niacin: a re-emerging pharmaceutical for the treatment of dyslipidaemia.Br J Pharmacol. 158:429-41 (2009).
In another review paper (Song, 2013), the authors conclude that while the pursuit of a more tolerable form of niacin with some benefits that might add to statins might be attractive to a commercial sponsor, “Such pragmatic reasoning may drive the progressive abandonment of niacin, a drug that has long been a mainstay of cardiovascular therapy, while we still poorly understand its many potentially relevant mechanisms of action and have an incomplete picture of its clinical utility.” (Hear, hear!)
- Song and FitzGerald. Niacin, an old drug with a new twist. J Lipid Res.54(10):2586-94 (2013).
A further review of niacin in its various forms (McCay, 2012) notes that “[w]hether other compounds that are converted to NA [nicotinic acid] or that contains NA, nicotinamide (NM) or their releasable moieties should be referred to as ‘niacin’ depends on the biological effects that are attributed to the compound, the interpretation of the evidence for the rates of uptake and metabolism, and/or the release of the chemical components (apparent bioavailability) that produce biological effects similar to the primary forms of niacin.”
- MacKay, Hathcock, Guarneri. Niacin: chemical forms, bioavailability, and health effects. Nutr Rev.70(6):357-66 (2012).
LIVER TOXICITY
This review (McCay, Hathcock, Guarneri, 2012) noted that in the matter of LIVER TOXICITY, “[m]any severe reactions to NA, especially liver toxicity, have involved ill-advised or uninformed switching from NA preparations to ER-NA formulations without adjusting the dose.” That is, the same dose of niacin exhibits a greater risk of liver toxicity in the ER form. We suspect that this is caused by the loss of the PGD2 acutely released by plain niacin which in the ER form is more of a chronic release of PGD2, hence reducing the flushing effect but with the differences we have described in the body of this article above from an anti-inflammatory effect of acutely released PGD2 to a pro-inflammatory effect of chronically elevated levels of PGD2. John Hathcock, a co-author of this review, was formerly a prominent extensively-published scientist at the FDA who is an expert on matters of toxicity, writing frequently on toxicity issues involving nutrients. Dr. Hathcock left the FDA to become a scientist and analyst at the Council for Responsible Nutrition.
- (McCay, Hathcock, Guarneri 2012 reference in paragraph above)
MORE ON STUDIES OF “EXTENDED RELEASE”
AND “PROLONGED RELEASE” VS. IMMEDIATE
RELEASE NIACIN
It was a surprise to us that in their review discussed just above (McCay, Hathcock, Guarneri, 2012), the authors stated their belief that “the beneficial lipid lowering effects of both NA [nicotinic acid] and ER-NA [extended-release nicotinic acid] are well established with data showing reduction of total triglyceride levels by 20-50%, reduction of LDL-C levels by 10-25%, increases of HDL-C by 10-30% and reduction of lipoprotein a levels by 10-30% which includes preferential reduction of the more atherogenic, small dense LDL-C” as the data we have seen do not tend to support an equivalent effect on lipid levels between NA and ER-NA. As to the declaration of interest for this review, the authors simply note that McCay and Hathcock are “employed by The Council for Responsible Nutrition, a trade association representing dietary supplement manufacturers and ingredient suppliers.” Hence, we do not know why they came to this conclusion, telling us only that they did an extensive search of the literature.
In our own search, we found a number of papers reporting discrepancies between immediate or extended-release niacin with that of immediate-release niacin on lipid lowering that we found to be convincing that there was no such equivalence. For example, a study (Usman, 2014) showed that extended-release niacin could suppress post-meal triglyceride levels but unlike immediate release niacin, it had to be administered just before the fat-containing meal. NON-FATAL MYOCARDIAL INFARCTIONS, as the authors of that study (Usman, 2014) explained “[t]his disparity is relevant because extended-release niacin dominates clinical use, even though only immediate-release niacin prevented hard cardiovascular outcomes.” The authors went on to describe another study (Plaisance, 2008) in which the researchers found that bedtime dosing of <1500 mg extended-release niacin for six weeks failed to suppress postprandial (after meal) triglycerides the next day, unlike immediate-release niacin. The Plaisance et al study, however, involved repeated (that is, chronic) use of niacin over six weeks, whereas the Usman et al study was just for a single dose and the pattern of suppression of triglycerides, the Usman group suggested, depended on post-dose pharmacodynamics, referring to “disappointing cardiovascular effects of bedtime extended-release niacin …”
The paper by other authors (Vogt, 2007) showed that prolonged-release niacin (in this paper they used Niaspan) did increase HDL but not as effectively as immediate-release niacin and that only immediate-release niacin had been shown to reduce cardiovascular event rates.
- Usman, Qamar, Gadi, et al. Extended-release niacin acutely suppresses postprandial triglyceridemia. Am J Med.125(10):1026-35 (2012).
- Plaisance, Mestek, Mahurin, et al. Postprandial triglyceride responses to aerobic exercise and extended-release niacin. Am J Clin Nutr.88(1):30-7 (2008).
- Vogt, Kassner, Hostalek, et al. Prolonged-release nicotinic acid for the management of dyslipidemia: an update including results from the NAUTILUS study. Vasc Health Risk Manag.3(4):467-79 (2007).
SCHIZOPHRENIA AND THE NIACIN FLUSH
It is curious to note that a blunted flushing response to niacin has been observed in most schizophrenics, and even in a significant portion of first-degree relatives of schizophrenics (Shah, 1999), “suggest[ing] that niacin subsensitivity is a genetic trait …” (Messamore, 2003). A later paper by the same lead author (Messamore 2010) on niacin sensitivity and the arachidonic acid pathway in schizophrenia reported that, in a study of 20 schizophrenic adults compared to 20 controls, “[t]he schizophrenia-associated niacin response abnormality involves both diminished sensitivity and reduced efficacy.” This supports the possibility that the niacin flush plays a significant role in the clinical effects of niacin.
- Shah, Ramchand, Peet. The niacin skin flush test: first-degree relatives show responses intermediate between patients and controls. Schizophr Res.38:314 (1999).
- Relationship between the niacin skin flush and essential fatty acids in schizophrenia. Prostaglandins Leukot Essent Fatty Acids. 69:413-9 (2003).
- Messamore, Hoffman, and Yao. Niacin sensitivity and the arachidonic acid pathway in schizophrenia. Schizophr Res.122(1-3):248-56 (2010).
A HUMAN STUDY OF DIETARY NIACIN AND THE
RISK OF INCIDENT ALZHEIMER’S (Morris, 2003)
Here, it was reported that, in a group of 3718 participants of 65 years and older residents of a Chicago community, evaluated through the use of dietary data and at least two cognitive assessments to detect cognitive changes over a median 5.5 years, higher food intake of niacin was associated with a slower annual rate of cognitive decline. The authors of this study reported in two case control studies conducted by others that there were lower blood levels of a nicotinic acid metabolite among demented patients than among age and sex matched controls. The detection of such a difference in cognitive changes in individuals ingesting such small amounts of dietary niacin is really surprising, as the participants in the highest fifth of intake (with a median of 22.4 mg/day) had an 80% statistically significant reduction in risk compared with the lowest quintile (12.6 mg/day). These authors attempted to determine whether there was a difference in participants with an apoE4 allele, but could not detect any. At these levels of niacin, it is hard to imagine it would be possible to measure such a small difference.
- Morris, Evans, Bienias, et al. Dietary niacin and the risk of incident Alzheimer’s disease and of cognitive decline. J Neurol Neurosurg Psychiatry. 75:1093-9 (2004).
Sandy Shaw Copyright 2015
