Journal of the American College of Cardiology Vol. 44, No. 5, 2004
Published by Elsevier Inc.

EDITORIAL COMMENT

Low and Lowered
Cholesterol and Total Mortality*

Michael H. Criqui, MD, MPH,
Beatrice A. Golomb, MD, PHD=C2=A7
San Diego, California

The paper by Strandberg et al. (1) in this issue of the Journal
confirms that in relatively young, healthy men from higher
socioeconomic strata, a (naturally) low totalcholesterol (TC) is
associated with increased longevity (2); on average, such men will have
lower levels of low-density lipoprotein (LDL) cholesterol. Thus, if you
are fortunate to have a low TC as a young man, typically a product of a
lot of nature and a little nurture, you can expect to do a little
better than your peers not so favored.

Can one extrapolate these mortality benefits of naturally
low TC in advantaged middle-aged men to middle-aged
men of lower socioeconomic status (SES)? Can they be
extrapolated to the very low end of the lipid range, or to TC
lowered by treatment, or to women or to elderly persons? In
each case, the available evidence suggests perhaps not.

Lowest levels. Several prospective population studies have
suggested that the mortality benefit associated with lower
TC plateaus at levels around 180 mg/dl (3). Indeed, total
mortality actually appears to increase somewhat at levels
below 180 mg/dl, a finding partially but not completely
explained by illness-related reductions in TC and confounding
(3). This study does not contradict those observations.
The cohort of Finnish men in this study had generally high
cholesterols by current standards; only 8% of the population
had a TC
194 mg/dl (5.0 mmol/l), and when the authors
split this group in half, the lower half, with TCs
182
mg/dl (4.7 mmol/l), actually had a slightly higher mortality
(27.2%) than the upper half (24.5%). Although the numbers
were small and this difference was not significant, the result
is concordant with a summary analysis of population studies
of both men and women (3).

Lowered by treatment. Lowered TC, either by behavioral
intervention ormuch more potently by drug therapy, is
a question distinct from naturally low TC. The evidence
that pharmacologic therapy of dyslipidemia in higher risk
middle-aged men results in a total mortality benefit is
extensive and convincing (4). However, from the limited
clinical trial evidence available, lowering TC or LDL
cholesterol alone produces no independent benefit for clinical
events after the ratio of TC to high-density lipoprotein
(HDL) cholesterol has been considered (5,6). In only two
clinical trials has the issue of the most relevant lipid measure change
been addressed, and change in either the TC/HDL
ratio (5) or the apolipoprotein (apo) B/apoA1 ratio (6)
accounted for all the benefit from drug therapy. After
consideration of changes in these ratios, changes in TC and
LDL added no predictive value. Conversely, changes in
these ratios remained strong predictors even after consideration of
changes in TC and LDL. The benefits of statins and other dyslipidemic
drugs appear proportional to the degree of improvement in the TC/HDL
ratio, and by far
the largest reductions in event rates reported to date
occurred with combination statin and niacin therapy, with
the attendant profound lowering of the TC/HDL ratio (7).
Emerging data on the dramatic HDL-increasing effects of
cholesterol ester transfer protein inhibitors may prove important for
future therapy (8).

Increasingly, evidence shows that the level of TC or LDL
alone is a poor guide to whether to begin therapy. A recent
trial of high-risk patients showed proportional benefit at
every level of baseline TC and LDL (9), making use of such
criteria for guiding therapy in high-risk patients (as contrasted with
assessing risk) of questionable value.

Women. What about extrapolations to women? To date,
there is no evidence for a total mortality benefit in women
from dyslipidemia therapy. Two of three trials that have
released data on gender-specific total mortality with lipid
therapy (6,10) have actually shown increases in the treatment
group of 57% and 12%, respectively, whereas the third
reported a 1% decrease (11). None of these differences was
individually statistically significant. The authors of this
editorial have not been successful in obtaining genderspecific
mortality data from other trials to further evaluate
this question, even from trials completed more than a
decade ago. The Cholesterol Treatment Trialists Collaboration
is reportedly addressing such questions. However, a
PubMed search revealed only two publications to date from
this group: both were methods papers, and both were nearly
a decade old (12,13). We believe these questions have some
urgency. (Since the initial submission of this editorial, a
meta-analysis of drug treatment of hyperlipidemia in women
has come to a similar conclusion, and the authors have expressed a
similar frustration at their inability to obtain gender-specific data
from many studies [14]).

SES. What of socioeconomic status? In this study all were
men from the highest social class, a fact that is extenuated
by noting, in the present study we examined within-group
differences, which are probably less sensitive to the selective nature
of the cohort. However, a meta-analysis by Law and Thompson (15)
showed that the relation of low cholesterol to cancer on long-term
follow-up was not evident in a meta-analysis of high-SES samples but
was evident and
significant in meta-analyses of populations of low SES.
Indeed, there appeared to be a graded relationship from
high-SES populations, to mixed SES, to low SES. This
finding cannot be glibly dismissed because there are plausible
effect modifiers linked to SES, such as dietary and
environmental differences.

Elderly. Can we extrapolate the favorable observations
concerning low TC in younger persons to older persons?
The best evidence isnot really. This article does not
address the prognostic implications of low TC once old age
is attained. Some evidence suggests that elevated TC in old
age is protective (16), and other evidence indicates that in
the same person, a TC measured at an older age is much less
predictive than a TC measured at a younger age, and a TC
measured more distant in time is more predictive than a
recent one, even after adjusting each TC measure for the
other (17). Only consideration of HDL cholesterol allows
much predictive power at middle or older ages (18).

Finally, if TC is a poor predictor in the elderly, or perhaps
even a protective factor, should we lower TC in elderly
persons? Certainly they are at high cardiovascular risk and,
from this standpoint, should benefit from intervention if
comparable relative risk reductions obtain and benefits are
not undercut by unsuspected hazards. To date only one
trialProspective Study of Pravastatin in the Elderly at
Risk (PROSPER)has specifically randomized a large
number of elderly patients (age 70 to 82 years, mean 75
years) to drug therapy, and all were also at elevated vascular
risk for reasons in addition to age (19). The results were not
reassuring. The overall benefit for vascular events was a
modest 15%, there was no benefit for vascular events in
women, there was no benefit for stroke, and there was no
benefit for total mortality in both genders combined (readers
were not provided gender-specific mortality rates).

Moreover, there was a disturbing significant excess of incident cancer
in the treatment group, reminding us that elderly persons are at
increased risk for cancer as well as cardiovascular disease.

In conclusion, low TC is not uniformly a good thing, and
many unanswered questions remain. Appropriate access to
relevant data would help clarify some of these questions for
researchers and clinicians alike. There is a natural tendency
by authors to highlight the most positive findings in their
study, particularly where commercial sponsorship is involved.
Perhaps medical journals, as a prerequisite for
publication, should require investigators and sponsors to
make study data accessible after a suitable period to preclude, or at
least attenuate, bias in publication.

Reprint requests and correspondence: Dr. Michael H. Criqui,
UCSD School of Medicine, Division of Epidemiology, Department
of Family and Preventive Medicine, San Diego, California
92093-0607. E-mail: mcriqui@ucsd.edu.

REFERENCES
1. Strandberg TE, Strandberg A, Rantanen K, Saloma VV, Pitkl K,
Miettinen TA. Low cholesterol, mortality, and quality of life in old
age during a 39-year follow-up. J Am Coll Cardiol 2004;44:
1002 8.
2. Klag MJ, Ford DE, Mead LA, et al. Serum cholesterol in young
men and subsequent cardiovascular disease. N Engl J Med 1993;
328:313 8.
3. Jacobs D, Blackburn H, Higgins M, et al. Report of the Conference on
Low Blood Cholesterol: Mortality Associations. Circulation 1992;86:
104660.
4. Gould AL, Rossouw JE, Santanello NC, Heyse JF, Furberg CD.
Cholesterol reduction yields clinical benefit: impact of statin trials.
Circulation 1998;97:94652.
5. Natarajan S, Glick H, Criqui M, Horowitz D, Lipsitz SR, Kinosian B.
Cholesterol measures to identify and treat individuals at risk for
coronary heart disease. Am J Prev Med 2003;25:507.
6. Gotto AM Jr., Whitney E, Stein EA, et al. Relation between baseline
and on-treatment lipid parameters and first acute major coronary
events in the Air Force/Texas Coronary Atherosclerosis Prevention
Study (AFCAPS/TexCAPS). Circulation 2000;101:47784.
7. Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin,
antioxidant vitamins, or the combination for the prevention of coronary
disease. N Engl J Med 2001;345:158392.
8. Brousseau ME, Shaefer EJ, Wolfe ML, et al. Effects of an inhibitor
of
cholesterol ester transfer protein on HDL cholesterol. N Engl J Med
2004;350:150515.
9. Heart Protection Study Collaborative Group. MRC/BHF Heart
Protection Study of cholesterol lowering with simvastatin in 20,536
high-risk individuals: a randomised placebo-controlled trial. Lancet
2002;360:722.
10. Scandinavian Simvastatin Survival Study Group. Randomised trial of
cholesterol lowering in 4,444 patients with coronary heart disease: the
Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:
13839.
11. Hague W, Forder P, Simes J, Hunt D, Tonkin A, LIPID Investigators.
Effect of pravastatin on cardiovascular events and mortality in
1,516 women with coronary heart disease: results from the Long-term
Intervention with Pravastatin in Ischemic Disease (LIPID) study. Am
Heart J 2003;145:64351.
12. Simes RJ. Prospective meta-analysis of cholesterol-lowering
studies:
the Prospective Pravastatin Pooling (PPP) project and the Cholesterol
Treatment Trialists (CTT) collaboration. Am J Cardiol 1995;76:
122C6C.
13. Cholesterol Treatment Trialists (CTT) collaboration. Protocol
for a
prospective collaborative overview of all current and planned
randomized
trials of cholesterol treatment regimens. Am J Cardiol 1995;75:
11304.
14. Walsh ME, Pignone M. Drug treatment of hyperlipidemia in women.
JAMA 2004;291:224352.
15. Law MR, Thompson SG. Low serum cholesterol and the risk of
cancer: an analysis of the published prospective studies. Cancer Causes
Control 1991;2:25361.
16. Weverling-Rignsburger A, Blauw G, Lagaay A, Knook D, Meinders
A, Westendorp R. Total cholesterol and risk of mortality in the oldest
old. Lancet 1997;350:111923.
17. Shipley MJ, Pocock SJ, Marmot MG. Does plasma cholesterol
concentration predict mortality from coronary heart disease in elderly
people? 18 year follow up in Whitehall study. BMJ 1991;303:8992.
18. Kronmal RA, Cazin KC, Ye Z, Omenn GS. Total serum cholesterol
levels and mortality risk as a function of age. Arch Intern Med
1993;153:106573.
19. Shepherd J, Blauw GJ, Murphy MB, et al. Pravastatin in elderly
individuals at risk of vascular disease (PROSPER): a randomised
controlled trial. Lancet 2002;360:162330.
1010 Criqui and Golomb JACC Vol. 44, No. 5, 2004
Editorial Comment September 1, 2004:100910