NEW YORK - A coffee drinker's gender, as well as the brewing technique, may predict whether the invigorating drink will be associated with an increase in cholesterol, a new study suggests.

An analysis of data from more than 21,000 adults aged 40 and older revealed that consuming coffee brewed in a French press was associated with the highest increase in cholesterol, followed by espresso. Filtered coffee, on the other hand, appeared to have no impact on men and only a small risk increase was seen in women, researchers report in Open Heart.

"In the 1980s, The Tromsoe Study was the first to show that unfiltered boiled coffee was associated with higher cholesterol levels than filtered coffee," said first author Aasne Lirhus Svatun, a medical student at UiT The Arctic University of Norway in Tromsoe.

"Less has been known about espresso," she told Reuters Health by phones. "This study found that espresso coffee was associated with higher cholesterol than filtered coffee, but lower cholesterol than boiled and plunger coffee. The relationship between espresso coffee and elevated cholesterol was stronger for men than for women. The take-home message is therefore that brewing method and sex influence how coffee affects cholesterol levels. The more unfiltered the brew is, the more it will affect the cholesterol."

Coffee beans contain diterpenes, cafestol and kahweol, which all raise cholesterol levels in humans. "In filtered coffee, these compounds mainly get stopped by the paper filter, and the finished brew contains low concentrations of diterpenes. For unfiltered coffee like plunger coffee, the coffee beans and water are in contact for a long time, and there is no filter," Svatun explained.

To take a closer look at whether the gender of the drinker and the type of brewing was associated with increased cholesterol, she and her colleagues turned to data from the Tromsoe Study, one of Norway's most comprehensive population studies through the last 48 years, which has been repeated every sixth to seventh year between 1974 and 2016. Participants in that study filled out a questionnaire, had physical exams and simultaneous non-fasting venous blood samples collected by trained personnel. The questionnaire asked about amount of coffee consumed and how it was brewed.

Consumption of three to five cups of espresso daily was significantly associated with increases in serum total cholesterol (S-TC) of 0.09 mmol/L for women and 0.16 mmol/L for men, compared with not drinking espresso.

Moreover, consumption of six or more cups of French press coffee per day was also associated with increases in S-TC of 0.30 mmol/L for women and 0.23 mmol/L for men, compared with not drinking French press coffee.

Meanwhile, consumption of six or more cups of filtered coffee daily was associated with 0.11 mmol/L higher S-TC levels (95% CI, 0.03 to 0.19) for women and a 0.06 mmol/L increase (95% CI, -0.01 to 0.14) for men. The researchers noted that there were significant sex differences for all coffee types except French press coffee.

So, should coffee lovers all switch to the filtered version?

"Norwegians love coffee, and we are the second-highest coffee consumers worldwide," Svatun said. "For the Tromsoe Study, it is essential that this information goes out to the general public, so that they can make their own informed healthy choices. People who drink many cups of unfiltered coffee per day, or have elevated cholesterol, could benefit from changing to filtered coffee. This is more important for those who drink" French press coffee.

The new study is "interesting," but flawed in many ways, said Dr. Howard Weintraub, clinical director of NYU Langone's Center for the Prevention of Cardiovascular Disease.

First, Dr. Weintraub told Reuters Health by phone, the study does not account for participants' use of milk, cream and sugar. The researchers also did not comment on the rest of participants' lipid profile.

"You would want to know whether HDL cholesterol went up or down, whether triglycerides went up or down," Dr. Weintraub said.

Beyond that, Dr. Weintraub said, the elevations in cholesterol reported in the study were small, "and the only reason the numbers approach significance is because there were 11,000 participants. In a study with huge numbers a small change can become statistically significant."