In our previous study of subjects attending a lipid disorder clinic, we found a strong association between plasma homocysteine and betaine excretion 
, and the urinary betaine excretion was the strongest determinant of plasma homocysteine in the male subjects 
. We were therefore surprised when similar associations were not seen in several subsequent studies, some briefly described elsewhere 
and some unpublished 
. We have shown here that our previous results 
were serendipitous consequences of the population we sampled. One cohort that did not show the expected correlation was the ACS study group, who were stabilized subjects recruited at a clinic visit approximately four months after hospital admission following a cardiovascular event. These differed from the lipid disorders clinic subjects in that almost all were receiving standard cardiovascular medications including lipid lowering agents. Few of them had elevated plasma lipids when studied, whereas in contrast, the subjects recruited from the lipid clinic required specialist outpatient management for disordered lipid profiles. The differences were large; the bottom quartile of non-HDL cholesterol in the lipid clinic subjects was above the top quartile in the ACS cohort. We found a positive correlation between an elevated betaine excretion and plasma homocysteine in sub-groups of the ACS cohort with elevated plasma lipids, with non-HDL cholesterol a strong marker of this (with the correlation increasing with increasing non-HDL cholesterol). In addition, in the total ACS population, there was a confounding trend for low betaine excretion to negatively correlate with plasma homocysteine; this trend was weak or absent in the lipid disorders cohort, and was stronger in the subsets of ACS subjects treated with drugs such as diuretics and proton pump inhibitors that were more commonly prescribed to ACS subjects. The median plasma betaine concentrations in both sexes of the ACS population were close to normal 
and were much higher than those in the lipid disorders cohort. Plasma betaine is not independent of plasma lipids and an inverse relationship that has been previously described 
was present in the ACS population in this study 
. The other major difference between the two study populations is age, which is an independent determinant of plasma homocysteine and does not explain the difference; in contrast, although the non-HDL cholesterol concentration efficiently defines sub-groups in which betaine excretion determines plasma homocysteine, the lipid concentration itself did not correlate with plasma homocysteine.
When we examined the cohort of subjects without diabetes who were attending a lipid disorders clinic, we expected that this group would have a high proportion of subjects with pre-diabetic conditions, and we found in this group that betaine (both plasma concentrations and urinary excretion) was a major determinant of plasma homocysteine concentrations 
, and in male subjects the urinary betaine excretion was the strongest determinant of homocysteine, stronger than folate 
. In this mainly folate-replete population we concluded that homocysteine was a marker of betaine deficiency, and found that betaine loss appears to be associated with vascular risk 
. Homocysteine is a well-known risk factor for cardiovascular disease 
. However, studies in which folate supplementation decreased the mean plasma total homocysteine concentrations have not shown a corresponding decrease in subsequent vascular events 
. There has been considerable recent discussion of this paradox and a number of suggested resolutions put forward, which is not surprising given the strong association between homocysteine and both cardiovascular disease and heart failure 
. However, it is usually overlooked that betaine may be the major determinant of plasma homocysteine 
which is largely controlled by the liver enzyme betaine-homocysteine methyltransferase 
, and most discussion is focussed on the role of folate and vitamin B12
mediated homocysteine lowering. There is now a growing awareness of importance of betaine in human health 
. It has dual roles as a key osmolyte and as a source of methyl groups, and tissue concentrations are typically an order of magnitude higher than circulating blood concentrations 
. More than 20% of patients with diabetes mellitus excrete abnormal amounts of betaine 
. Since betaine is normally metabolized, and this metabolism inevitably involves the conversion of homocysteine to methionine 
, it seemed plausible that betaine loss in the urine could be both a major determinant of plasma homocysteine concentrations and be itself pathogenic, and our initial observation supported this model.
Here we ask why this association between betaine excretion and homocysteine was not subsequently observed in the ACS study. This population, and the lipid disorders clinic cohort, were quite different and neither is typical of the general population. The contrast in control of homocysteine between the cohorts is most strongly associated with the large difference in the plasma lipid profiles of the two populations, and another factor is the difference in the degree of medication of the subjects. The correlation between urine betaine excretion and plasma homocysteine increases in subsets of the present study with higher lipids, but the number of subjects becomes too small to detect significant correlations at lipid concentrations well below those in the earlier study. A confounding factor is the presence of subjects with abnormally 
low betaine excretions, as seen in previous studies 
: these also appear to be high risk subjects (unpublished data). Plasma non-HDL cholesterol appears to be a marker of subjects in whom betaine excretion is related to plasma homocysteine, but itself is not statistically a determinant of homocysteine. These results confirm observations made in a smaller study of ambulant elderly subjects 
. It is possible that drugs affect betaine and homocysteine metabolism, but the differences we observed may reflect differences between the sub-populations that are prescribed these drugs; the cross-sectional design means that we cannot distinguish these alternatives, but both are plausible. In the case of fibrates it is more likely to be an effect of the drug 
, and the same might be true of the angiotensin II antagonists reported here, but in many cases it is likely that the results reflect the selection of patients who are treated with that drug. The correlations seen with low betaine excretions is obviously associated with subjects being treated with diuretics or proton pump inhibitors, but it does not necessarily follow that the drug treatment is responsible for the effect, and since most subjects are on multiple drugs, the effects suggested by are not independent. In the metabolic syndrome where high blood pressure is common, subjects are likely to be treated with one or more of the drugs listed in and . Subjects with the metabolic syndrome are also likely to have elevated lipids, and we have previously suggested that these are subjects who are likely to become betaine deficient 
. Age itself may not be a critical factor, but further investigation would be justified into the diseases associated with age, and the effects of medication used to treat these. We acknowledge that in our analyses we have made ad-hoc multiple comparisons within subgroups, and therefore type I errors are not excluded. However, the consistency of the patterns of correlations suggests robustness in the results, but a cautious interpretation is required.
In conclusion, the results of the two studies can be reconciled, but the apparent contradictions are a warning about generalizations based on a selected patient population. The reconciliation confirms the association between betaine and lipid metabolism, and is consistent with a role for betaine in the metabolic syndrome. These cross-sectional studies suggest a number of hypotheses to be tested. The results are consistent with our previous suggestions 
that betaine plays an important role in the metabolic syndrome, and this could be tested by longitudinal studies of supplementation. The cross-sectional results suggest that betaine deficiency is most apparent in subjects with elevated plasma lipids, and the hypothesis that there is a causal connection needs to be tested; it cannot be excluded that it is the elevated lipid that affects betaine, since in rats a high lipid diet appears to cause an elevation in betaine excretion 
. As for mechanisms, the osmolyte role of betaine needs to be considered as well as its role as a methyl donor, and this is also consistent with the possibility that betaine metabolism is affected by drugs that modify the renin-angiotensin system. Studies to identify such connections may suggest improved therapeutic strategies that could benefit those subjects with the metabolic syndrome who have elevated lipids and increased vascular risk.