Metabolic health in the time of COVID-19 – an update.
By George Henderson and Grant Schofield
Topics covered here
- Metabolic health and COVID-19 risk
- Possible roles of HDL in lung disease
- The microbiome and viral resistance
- Selenium and COVID-19
- Vitamins D and K2, and COVID-19
Bottom line for all of this:
Eating whole unprocessed food, especially that low in sugar, refined and processed carbohydrates, is likely to have benefits for your immune system and viral resistance generally.
If you are pre-diabetic, diabetic, overweight, or insulin resistant then this effect may be even more important.
COVID-19 numbers are falling sharply in NZ and we are on course to “eliminate” the virus, or probably more accurately restrict it so it’s easily managed. Vaccines are being trialed around the world already; but there are significant barriers to making a safe and effective vaccine for this disease, so we are not holding our breath. You certainly wouldn’t bet the country on a timely, safe, and highly effective vaccine.
There are also encouraging signs that some anti-viral drug combinations developed for hepatitis C treatment may be effective enough against COVID-19 (they need to be highly specific against HCV, but because COVID-19 is not usually a persistent virus, a drug that reduces the viral load by a log factor or two will probably be good enough, the immune system can do the rest).
Both these approaches take time (and money) and it may be years before a vaccine can reach everyone. And even then, it is unlikely to give the lasting herd immunity that some vaccines do. We have flu vaccines, but flu still goes round, and this is only partly due to unpredicted strains appearing that weren’t in the vaccine – when healthy elderly people have a flu vaccine, the rates of seroconversion or seroprotection (two different ways of measuring whether there are enough antibodies to prevent infection) are low, and the immunity doesn’t last as long. Immunologists worry that the same might be true of COVID-19 – not everyone infected has seroprotection going forward.
Metabolic disease and COVID-19 risk.
It is beyond obvious that metabolic disease greatly increases the risk of death in people who become infected with COVID-19.
New research from Glytec, the main supplier of diabetes software in the US, suggests that in-hospital mortality is more than quadrupled in COVID-19 patients with diabetes and hyperglycemia.
“People with diabetes who have not yet been infected with the SARS-CoV-2 virus [which causes COVID-19] should intensify their metabolic control as needed as means of primary prevention of COVID-19 disease,” according to expert panel guidance in the Lancet Diabetes & Endocrinology.
Diabetes and hyperglycaemia, as well as the metabolic syndrome and most cases of hypertension, can have their control greatly improved by carbohydrate restriction, calorie restriction, and/or fasting. The fewer carbohydrates are eaten, the fewer drugs are required to maintain glycemic control, and the more reliable (or “intense”) it is, because the usual, unavoidable errors in estimating diet and dose have less impact when numbers (of carbs, of drug units) are in a lower range. None of this is new or controversial knowledge, the main thing that is required from the practitioner is the ability to ignore the low-quality evidence from epidemiology (saturated fat, whole grains, low carbohydrate “diet score”) that is used by some people to confuse the issue. This evidence, such as it is, has zero to do with the utility of carbohydrate restriction for the control of metabolic disease. (But saturated fat and whole grains will get their time in court when we discuss immunity).
While there has been some focus on obesity and much on type 2 diabetes and the metabolic syndrome in the literature, an emerging risk factor seems to be HDL cholesterol which has causal links to respiratory fitness.
In this pre-print study of cases in 3 hospitals
Fasting hypoglycaemia was found in 21.4% of patients in mild 1 (14-day recovery) group with no case of fasting hyperglycaemia. In the mild 2 (30 day recovery) group, 34.1% of the patients had fasting hypoglycaemia, and 2.3% had fasting hyperglycaemia. Compared with mild COVID-19 patients, we found that 24% of severe COVID-19 patients had fasting hyperglycaemia and 4% had fasting hypoglycaemia.
Patients in the severe group had a lower level of serum total protein (59[58–63] vs. 65[63–70]), serum albumin (36[34–39] vs. 41[37–43]), total cholesterol (3.6[3.3–4.0] vs. 3.8[3.5–4.4]), and HDL-C (0.88[0.81–1.10] vs. 1.05[0.93–1.50]) compared with the mild 1 group (Table 2).
The difference in HDL cholesterol is more significant than that in total cholesterol, and is more likely to be causal. Why? Because HDL cholesterol is a marker of respiratory fitness, and ApoA1 is causal in lung health; it is involved in both the antioxidant protection of lung cells, and the lung’s immune function, which relies on “reverse cholesterol transport”-type movement of cholesterol.
Apolipoprotein A-I (apoA-I) and high-density lipoproteins (HDL) mediate reverse cholesterol transport out of cells. Furthermore, HDL has additional protective functions, which include anti-oxidative, anti-inflammatory, anti-apoptotic, and vasoprotective effects. In contrast, HDL can become dysfunctional with a reduction in both cholesterol eﬄux and anti-inflammatory properties in the setting of disease or the acute phase response. These paradigms are increasingly being recognized to be active in the pulmonary system, where apoA-I and HDL have protective effects in normal lung health, as well as in a variety of disease states, including acute lung injury (ALI), asthma, chronic obstructive pulmonary disease, lung cancer, pulmonary arterial hypertension, pulmonary fibrosis, and viral pneumonia. Similar to observations in cardiovascular disease, however, HDL may become dysfunctional and contribute to disease pathogenesis in respiratory disorders.
HDL also protects populations of anti-inflammatory regulatory T-cells, T-Reg.
HDL could be worth looking at further in lung disease
To sketch in the possibility of this, we’re going to explore this American paradox – when the 397 residents of a Boston homeless shelter were tested for COVID-19, 146 people tested positive. As we might expect when 397 people are living together with little opportunity for distancing. However they reported the same low rate of cold and flu symptoms as the people testing negative.
Cough (7.5%), shortness of breath (1.4%), and fever (0.7%) were all uncommon among COVID-positive individuals.
This is another preprint, so results may change, but we wanted to see if we could explain this: what are the metabolic features associated with homelessness, if any?
We found a study of a homeless population in Estonia. N=51, 90% men. Is this is a suitable proxy for a US population? The result is striking enough that it may well be. These people had a wide range of metabolic variation across lipid, glycemic and inflammatory markers, but not one of them had HDL of <1 mmol/L or below the normal range (which has no upper limit). Usually, around 40% of a population will have low HDL. ApoA1 was normal or high, no-one was low.
More than half of the investigated patients had values of measured markers (hsCRP, TChol, LDL-Chol, TG, HbA1c, ApoA1, ApoB, Lp(a), Gluc) within normal range. Surprisingly, 100% of subjects had HDL-Chol within endemic norm.
In this peripatetic population, it is likely that a high activity level contributes to the HDL level. Of course our juxtaposition of these two studies proves nothing, and there are many reasons why homeless populations might be at higher risk after all, but it is an interesting finding. At any rate, COVID-19 is not picking off the highly fit, and the effect of HDL and ApoA1 on lung function helps explain this.
Besides physical activity, a low carb diet (which also improves VO2max) also raises HDL levels. As do saturated fats of the dairy and coconut type, which may have other benefits, as we will see in the next section. Saturated fat foods tend to raise LDL cholesterol, so is LDL cholesterol also relevant? This review suggests it is.
It is now increasingly recognized that lipids and lipoproteins play an important role in host defense as part of the innate immune system. For example, lipoproteins including HDL, chylomicrons, VLDL, and LDL can bind and neutralize LPS, lipoteichoic acid, and viruses.
Numerous studies have shown that animals that have elevations in serum lipid/lipoprotein levels are protected from the toxicity of LPS, whereas animals with low circulating lipid/lipoprotein levels are more sensitive to the toxic effects of LPS. Studies have shown that HDL may inhibit the ability of certain viruses to penetrate cells.
In other words, this is not a time to worry about the effect of your diet on cholesterol, if that diet helps you reach other goals.
Immunity and the microbiome.
As we mentioned earlier, flu vaccine responses in health elderly people are pretty disappointing, and this has been our major effort to reduce infections that cause pneumonia deaths in this population. The Cochrane review points out that RCT evidence is slim.
These results indicate that 30 people would need to be vaccinated to prevent one person experiencing influenza, and 42 would need to be vaccinated to prevent one person having an influenza-like illness.
Given how common influenza-like illnesses are, this, though better than no protection, is not the coverage expected from a really effective vaccine.
When we looked into the evidence for various supplements “boosting immunity”, some of the best evidence was for probiotics boosting vaccine responses, especially for flu vaccines in the elderly.
This research allows us to look closer at the data – probiotics and prebiotics double the rates of seroconversion and seroprotection (2 similar measures of immunity) in adults after vaccination for common flu strains. The effect is even stronger if the trails are restricted to only those looking at healthy elderly people (those not in hospital). There is a dose-response effect from longer duration of exposure to probiotics before vaccination.
Now, if you can double or triple antiviral immunity after a vaccine by giving a probiotic, then rates of immunity without the probiotic have to be pretty low. And if you dig down into these studies, they are around 30%. If you can double (or triple) that rate you’d have herd immunity.
The prebiotic studies are not that great, there are fewer of them, and the one that involved feeding isolated bean fibre to elderly people sounds unnecessarily harsh, but they do demonstrate the principle that just as you are what you eat, so are your microbes. As far as we can tell, there are two types of food that put the microbiome out of balance – one is refined carbohydrates (sugar and flour), which are missing the fibres that the better microbes like. So if you do eat grains, because you have no metabolic reason not to, be sure to eat whole grains, and if you do eat sugar, try to keep it in the fruit as much as possible. (but remember that gluten does weaken tight junctions in the gut, which may be undesirable for reasons that will appear later)
The other thing that good microbes don’t like a lot of is unsaturated fat. The beneficial gram positive bacteria that keep the gut in shape metabolise saturated fats – these are prebiotic. They have a limited capacity to tolerate unsaturated and especially polyunsaturated fatty acids. This may be why the “good shepherd” bacteria lactobacillus reuterii is declining in the US population, where it used to be very common. If you make a full-fat yoghurt, you need to use either dairy, or coconut, and these are the animal food and the plant food highest in saturated fat. There are yoghurts made with soy milk, but this is low in fat (1.61 g/L). The bacteria are fermenting sugars in the milk (and are able to make their own saturated fats) and to make a higher-fat yoghurt with soy it would probably be necessary to hydrogenate the soy oil.
After the NZ level 3 lockdown rush to buy fast food, there were probably a few people nursing a gut-ache. Refined carbohydrates and deep-fried seed oils can cause changes in the microbiome by killing off desirable species and boosting the population of more inflammatory ones. This is probably not great for immunity in the context of the ageing immune system.
COVID-19 sometimes appears as a gut infection, with symptoms of gastrointestinal distress and diarrhoea. Evidence from the US is that mild cases of COVID-19 have a longer course if gut symptoms appear first (30 days vs 14 days). Evidence from China is that gut symptoms early in the course of the disease predict more severe lung pathology.
Why would this be the case? If COVID-19 strongly infects the gut, it may interact with gut bacteria. This has been observed with polio, where specific bacteria enhance infection. The risk is that as viral infection of the lung proceeds, it will then be accompanied by LPS and other PAMPS from gut bacteria, and it is possible that signals from some bacteria are more inflammatory and predispose more to autoimmunity and sepsis than signals from, for example, lactobacillus species.
Of course it goes without saying that age and the comorbidities associated with poor COVID-19 prognosis will be associated with GERD, IBS, SIBO and the other clinical signs of dysbiosis. Anticholinergics (used to treat IBS) and PPI’s (used to treat GERD) are both associated with an increased risk of pneumonia.
How relevant this is is unclear, but it seems unlikely to be irrelevant.
Selenium and COVID-19 cure rates.
In our original factsheet we highlighted the likely role of selenium in viral outbreaks. Selenium is critical to immune cell function, especially in T-cells and macrophages where it provides an antioxidant defence against the ROS these cells generate, and also has critical functions in endothelial cells. Many RNA viruses, including HIV, HCV, ebola, and flu and coronaviruses, encode for large numbers of selenocysteine residues. This means that they have a high demand for selenium, and that infection can cause selenium deficiency if levels are low (the virus is sequestering selenium, which is unavailable to the body and lost when virus particles shed). Selenium helps the viral genome stay stable, and it is more likely to mutate in a selenium-deficient host, so it’s actually in our interests to give a virus like COVID-19 all the selenium it needs – getting enough selenium for both virus and host is a win-win situation.
A new study from China correlates COVID-19 cure rates with soil selenium status. While this is not individual-level data, the correlations are both large and highly specific and are unlikely to be due to different diagnostics or other random variations between regions.
Examining data from provinces and municipalities with more than 200 cases and cities with more than 40 cases, researchers found that areas with high levels of selenium were more likely to recover from the virus. For example, in the city of Enshi in Hubei Province, which has the highest selenium intake in China, the cure rate (percentage of COVID-19 patients declared ‘cured’) was almost three-times higher than the average for all the other cities in Hubei Province. By contrast, in Heilongjiang Province, where selenium intake is among the lowest in the world, the death rate from COVID-19 was almost five-times as high as the average of all the other provinces outside of Hubei.
Most convincingly, the researchers found that the COVID-19 cure rate was significantly associated with selenium status, as measured by the amount of selenium in hair, in 17 cities outside of Hubei.
New Zealand soil is relatively low in selenium; we recommend, at least for the duration of the pandemic, eating either 2-3 Brazil nuts per day or supplementing 50-100mcg selenium. Clinicians brand sodium selenite drops provides a cheap option with lower toxicity than other supplement forms, but many multivitamins made for the NZ market supply useful amounts of selenium. Selenium from supplements should not exceed 200mcg per day.
Vitamin D and vitamin K2
Vitamin D status is also highly correlated (at an individual level) with the risk of dying from COVID-19 in Indonesia. It may be that some of the low vitamin D is due to an effect of infection, but this does not completely explain the protective effect of higher levels, which is consistent with animal experiments in lung disease models. We were lucky in NZ and Australia that COVID-19 hit at the end of summer, when vitamin D levels are highest, but many people will need to supplement in winter to maintain adequate levels.
Vitamin K status has also correlated with survival from COVID-19, but again this may be in part an effect of infection (vitamin K levels were not tested – a marker of calcium transfer was used, which shows the function of vitamin K2 in the body, and is known to decline in kidney disease). Vitamin D3 with vitamin K2 supplements are available – from Clinicians again – no they don’t pay us, they just have a good evidence-based policy when it comes to formulating supplements! Vitamin K1 is found in green leafy plants, vitamin K2 (a small amount of which is made from K1 in the body) is sourced from some fatty animal foods, and some fermented animal and plant foods. The best dietary sources of D3 are salmon, eggs, and fatty pork.
 Nie, Shuke & Zhao, Xueqing & Zhao, Kang & Zhang, Zhaohui & Zhang, Zhentao & Zhang, Zhan. (2020). Metabolic disturbances and inflammatory dysfunction predict severity of coronavirus disease 2019 (COVID-19): a retrospective study. 10.1101/2020.03.24.20042283.
 Gordon EM, Figueroa DM, Barochia AV, Yao X, Levine SJ. High-density Lipoproteins and Apolipoprotein A-I: Potential New Players in the Prevention and Treatment of Lung Disease. Front Pharmacol. 2016;7:323. Published 2016 Sep 21. doi:10.3389/fphar.2016.00323
 Rueda CM, Rodríguez-Perea AL, Moreno-Fernandez M, et al. High density lipoproteins selectively promote the survival of human regulatory T cells. J Lipid Res. 2017;58(8):1514–1523. doi:10.1194/jlr.M072835
 COVID-19 outbreak at a large homeless shelter in Boston: Implications for universal testing. Travis P. Baggett, Harrison Keyes, Nora Sporn, Jessie M. Gaeta. medRxiv 2020.04.12.20059618; doi: https://doi.org/10.1101/2020.04.12.20059618
 Kaldmäe M, Zilmer M, Viigimaa M, et al. Cardiovascular disease risk factors in homeless people. Ups J Med Sci. 2011;116(3):200‐207. doi:10.3109/03009734.2011.586737
 Feingold KR, Grunfeld C. Lipids: a key player in the battle between the host and microorganisms. J Lipid Res. 2012;53(12):2487‐2489. doi:10.1194/jlr.E033407
 Demicheli V, Jefferson T, Di Pietrantonj C, Ferroni E, Thorning S, Thomas RE, Rivetti A. Vaccines for preventing influenza in the elderly. Cochrane Database of Systematic Reviews 2018, Issue 2. Art. No.: CD004876. DOI: 10.1002/14651858.CD004876.pub4.
 Lei WT, Shih PC, Liu SJ, Lin CY, Yeh TL. Effect of Probiotics and Prebiotics on Immune Response to Influenza Vaccination in Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Nutrients. 2017;9(11):1175. Published 2017 Oct 27. doi:10.3390/nu9111175
Han C, Duan C, Zhang S, et al. Digestive Symptoms in COVID-19 Patients With Mild Disease Severity: Clinical Presentation, Stool Viral RNA Testing, and Outcomes [published online ahead of print, 2020 Apr 15]. Am J Gastroenterol. 2020;10.14309/ajg.0000000000000664. doi:10.14309/ajg.0000000000000664
 Jinsong Zhang, Ethan Will Taylor, Kate Bennett, Ramy Saad, Margaret P Rayman, Association between regional selenium status and reported outcome of COVID-19 cases in China, The American Journal of Clinical Nutrition, , nqaa095, https://doi.org/10.1093/ajcn/nqaa095
 Thomson CD, Chisholm A, McLachlan SK, Campbell JM. Brazil nuts: an effective way to improve selenium status. Am J Clin Nutr. 2008;87(2):379‐384. doi:10.1093/ajcn/87.2.379
 Raharusun, Prabowo and Priambada, Sadiah and Budiarti, Cahni and Agung, Erdie and Budi, Cipta, Patterns of COVID-19 Mortality and Vitamin D: An Indonesian Study (April 26, 2020). Available at SSRN: https://ssrn.com/abstract=3585561
 Dofferhoff, A.S.; Piscaer, I.; Schurgers, L.J.; Walk, J.; van den Ouweland, J.M.; Hackeng, T.M.; Lux, P.; Maassen, C.; Karssemeijer, E.G.; Wouters, E.F.; Janssen, R. Reduced Vitamin K Status as A Potentially Modifiable Prognostic Risk Factor in COVID-19. Preprints 2020, 2020040457 (doi: 10.20944/preprints202004.0457.v1).