The Japanese report

In ​April 2015 did four Japanese researchers, Tomohito Hamazaki, Harumi Okuyama, Yoichi Ogushi and Rokuro Hama, publish an analysis of several cholesterol studies in Japan and other countries. The 116 pages long report has a clear conclusion: Cholesterol is a friend, not an enemy.

The report is introduced with the following words:

• "High cholesterol levels are recognized as a major cause of atherosclerosis. However, for more than half a century some have challenged this notion. But which side is correct, and why can’t we come to a definitive conclusion after all this time and with more and more scientific data available?  We believe the answer is very simple: for the side defending this so-called cholesterol theory, the amount of money at stake is too much to lose the fight. The issue of cholesterol is one of the biggest issues in medicine where the law of economy governs."
• "The Japanese public ... generally accept everything physicians say; unfortunately, this is also complicated by the fact that physicians don’t have enough time to study the cholesterol issue by themselves, leaving them simply to accept the information provided by the pharmaceutical industry."
• "The relationship between all-cause mortality and serum cholesterol levels in Japan is a very interesting one: mortality actually goes down with higher total or low density lipoprotein (LDL) cholesterol levels, as reported by most Japanese epidemiological studies of the general population."
• "The theory that the lower the cholesterol levels are, the better is completely wrong in the case of Japan—in fact, the exact opposite is true. Because Japan is unique in terms of cholesterol-related phenomena, it is easy to find flaws in the cholesterol hypothesis."
• "Our purpose in writing this supplementary issue is to help everyone understand the issue of cholesterol better than before, and we hope that we lay out the case for why a paradigm shift in cholesterol treatment is needed, and sooner rather than later."

Heart disease

The Ibaraki Prefecture Health Study: A total of 30,802 men and 60,417 women were followed for 10.3 years. The figure shows relationship between LDL levels and the Hazard Ratio (HR) for all-cause mortality. Dark gray shading represents coronary heart disease (CHD) deaths. The width of each column is proportional to the number of participants in that group. The vertical lines represent 95% confidence intervals.

Comment: The study shows different patterns for men and women.

• Men: All-cause mortality is U-shaped, with lowest rate with LDL between 120 and 139 mg/dl. The highest mortality rate is when LDL is below 80 mg/dl. CHD (coronary heart disease) deaths increases slowly with LDL levels.
• Women: All-cause mortality for women decreases with increasing LDL levels. There's no correlation between CHD deaths and LDL levels.  

Meta-analysis of total cholesterol levels and all-cause mortality in Japan. This meta-analysis included five reports. The width of each column is proportional to the number of participants in that group. The vertical lines represent 95% confidence intervals. Total number of subjects: 173,539.

Hazard ratios (HRs) for all-cause mortality and ischemic heart disease (IHD) according to total cholesterol level in Norway: HUNT 2 Study. A total of 52,087 Norwegians aged 20–74 years were followed to calculate cause-specific mortality for 10 years. HRs were adjusted for age, smoking, and systolic blood pressure. The height of the black bar denoting IHD deaths is set according to the ratio between the numbers of IHD deaths and all-cause deaths within the same column. HRs for IHD mortality in cholesterol categories II to IV were not significantly different from those in cholesterol category I in men or women. The width of each column is proportional to the number of participants in that group.

Comment: Both studies show that there is a correlation between low cholesterol and high all-cause mortality. This contradicts the 2015 guidelines for cholesterol levels set by the  The other results are less conclusive when compared directly.

Aged people and heart disease

Leiden, Netherlands: A total of 724 residents aged 85 or above were followed for nearly 10 years. The group with the highest cholesterol level had the lowest all-case mortality.

Honolulu, USA: A total of 3,572 Japanese-American men aged 71–93 years were followed from baseline (1991–1993) to the end of 1996. There were 727 deaths (20% of the total) over the study period.

Fifteen-year follow up of 67,413 men in Vorarlberg, Austria: Vorarlberg Health Monitoring and Promotion Programme. The width of a column is proportional to the number of participants. The left column in each age group represents the lowest total cholesterol quartile, the middle column represents the reference group containing the second and third quartiles combined, and the right column represents the highest cholesterol quartile. Adjusted for age (Cox’s proportional hazards model).

Fifteen-year follow up of 82,237 women aged 20–95 years in Vorarlberg, Austria: Vorarlberg Health Monitoring and Promotion Programme.

"Survival rate is definitely better in elderly people with high total or LDL cholesterol levels than in those with low levels. The proportion of people with FH or similar conditions among the elderly population is much smaller than that among younger populations, which explains why an inverse correlation between total cholesterol (or LDL cholesterol) and all-cause mortality becomes prominent with age in all countries.  High LDL cholesterol levels might also be related to better cognitive function. The memory function of 193 functionally independent and community-dwelling elderly participants aged ≥80 years was cross-sectionally examined in the Key to Optimal Cognitive Aging ( KOCOA) Project, a prospective study undertaken in Okinawa, Japan. High LDL cholesterol levels and low triglyceride/HDL cholesterol ratios were associated with high Scenery Picture Memory Test scores after adjustment for many confounding factors. When viewed together with the data presented above on cholesterol and longevity, it seems clear that high cholesterol levels should not be considered unhealthy especially in elderly people." - Hamazaki, Okuyama, Ogushi and Hama

Stroke

Hazard ratio for stroke mortality in both men and women according to total cholesterol level. A total of 65,594 men and women in Japan aged 40– 89 years were followed for a mean 10.1 years. The width of each column is proportional to the number of participants. HR = Hazard ratio; CI = confidence interval.

​Cancer

​One of the biggest contributors to the inverse correlation between the lowest cholesterol levels and the highest all-cause mortality is cancer mortality.

Association between cancer incidence and serum cholesterol level in men: JPHC Study [26]. A total of 33,368 Japanese men and women aged 40–69 years were followed for more than 10 years. The width of each column is proportional to the person years of that group.

​Infectious disease

LDL and the other lipoproteins are the nonspecific frontline against a wide variety of infectious agents. Lipoproteins have long been known to bind to and inactivate bacteria, bacterial fragments (lipopolysaccharides, LPS), and viruses.  The body can handle the entire phase of infection with a safety margin if lipoproteins neutralize a considerable proportion of toxic agents in blood. Without the buffer of lipoproteins, however, the immune system needs to deal with all the infection-related materials directly.

In a multiethnic cohort of 55,300 men and 65,271 women that was followed for 15 years (1979–1993), the association between total cholesterol and risk of infections (other than respiratory and HIV) was examined. Cholesterol was found to be inversely related to various infections, including all infections, in both sexes. The reduction of risk for all infections was 8% in both sexes.

Liver disease

Hazard ratios (HRs) for liver cancer and cirrhosis mortality according to low density lipoprotein (LDL) cholesterol level: Ibaraki Prefectural Health Study. A total of 16,217 participants aged 40–79 years (men: 34%) were followed for 14.1 years. The HRs for liver cancer and cirrhosis mortality were calculated using a multivariable Cox proportional hazards model. Covariates were age, sex, alanine transaminase, body mass index, alcohol intake, and smoking status. The height of the black bar denoting liver cirrhosis mortality is set according to the ratio between the numbers of cirrhosis deaths and total cancer + cirrhosis deaths. The width of each column is proportional to the number of participants. Note that zero mortality from liver cirrhosis was found in the group with LDL cholesterol ≥120 mg/dl (≥3.09 mmol/l).

Hazard ratio (HR) for liver cancer incidence according to total cholesterol level in Korea: NHIC study in Korea. A total of 1,189,719 Korean adults enrolled in the National Health Insurance Corporation who underwent biennial medical examinations from 1992 through 1995 were observed for 14 years until cancer diagnosis or death. Cox’s proportional hazards models with attained age as the underlying time metric were used to calculate HRs with adjustment for smoking, alcohol consumption, body mass index, physical activity, hypertension, and fasting serum glucose. The width of each column is proportional to the number of participants of that group; however, because of a limited number of women, the width of women’s columns is multiplied by 4. The trends in HRs for both sexes did not loose significance even after excluding cases during the first 10 years of observation.

Conclusion

• "In this supplementary issue, using data in large part from Japan where the mean life expectancy has been the longest in the world for decades, we have tried to show that cholesterol is not an enemy but a friend."
• "The general Japanese population with high total cholesterol levels—or with high levels of low density lipoprotein (LDL) cholesterol— have very often been shown in cohort studies to have low all-cause mortality."
• "What about the most relevant disease with respect to cholesterol, coronary heart disease (CHD)? Some epidemiological studies have found an association between high cholesterol levels and CHD mortality in Japanese men. In Japanese women, however, this association has been found in just one study, the NIPPON DATA80 study with a 17.3-year follow up. In fact, other studies have shown that CHD mortality in Japanese women is not related to cholesterol levels at all or even has an inverse association with cholesterol levels. Closer examination of the studies reveals that the positive association between cholesterol levels and CHD mortality in men is largely explained by the presence of participants with familial hypercholesterolemia (FH)."
  
• "Cholesterol levels also have some association with cancer, infection, and liver disease: subjects with high cholesterol levels have lower incidence and mortality rates from these diseases. With regard to liver disease specifically, if cholesterol levels are high enough, serious liver disease does not develop."
  

Dr. Tomohito Hamazaki

---

References
​ - Hamazaki, Okuyama, Ogushi and Hama: "Towards a paradigm shift in cholesterol treatment. A re-examination of the cholesterol issue in Japan". 
​ - Journal of Clinical Lipidology: "National Lipid Association recommendations for patient-centered management of dyslipidemia " (2015).
 - Nord-Trøndelag Heart Study (HUNT): Homepage.
​