Patrick Holford, Shark Liver Oil and Walnuts

Patrick Holford on ITV Lunchtime 16 April 2008
Former Visiting Professor Patrick Holford is Head of Science and Education at Biocare. From time to time Holford has nothing but harsh words for randomised controlled trials and the perceived iniquity of systematic reviews or meta-analyses. Unless they confirm a point of view that he already holds, of course, or that he can adapt to the self-aggrandisement of his opinions. And so it is with some delight and no obvious trace of irony that Holford welcomes the release of a systematic review and meta-analysis that evaluates the impact of incorporating walnuts into the diet and outcomes for blood lipids as a proxy for a risk factor for cardiovascular disease. Or, as Holford so pithily phrases it, Go (wal) nuts this summer – walnuts lower your cholesterol (as ever, you need to go to the home page to read the first paragraph but Holford is handily re-cycling his blog posts as email newsletters which must be value for money).

Holford’s account of the paper largely follows the pattern and quotations of the better-written Food Navigator account: Walnuts can cut cholesterol, say Harvard researchers. We will concentrate on Holford’s own, special insights.

I’ve long been an advocate of eating lots of nuts and seeds which are very good sources of both [sic] antioxidants, minerals, protein and essential fats…

I recommend eating a small handful of raw nuts or seeds every day. After all, if a tree is going to grow from it it’s got to be a concentrated source of nutrients, including protein. Whenever you eat fruit be sure to have a few nuts or seeds – the carbohydrate/protein combination helps stabilise your blood sugar and hence makes you feel fuller for longer.

I imagine that many people’s grandmothers have made remarks about the healthfulness of nuts and the good old days when dinner started with soup and ended with a nut-cracking session.

However, Holford advocates “eating lots of nuts and seeds” which somehow equates to “a small handful of raw nuts or seeds every day” and the meta-analysis only included studies that involved comparatively high quantities of walnuts – between 10-24% of total calories. The nutritional status and impact of nuts is a promising area of research. However, the fact that something is true of walnuts does not necessarily make it true of other nuts and seeds and the quantities are very different. Plus, these studies were all short-term, so although the results look interesting, the authors conclude:

Overall, high-walnut-enriched diets significantly decreased total and LDL cholesterol for the duration of the short-term trials. Larger and longer-term trials are needed to address the effects of walnut consumption on cardiovascular risk and body weight.

Holford’s remark about the tree is embarrassing and we have been here before with the well-known botanist and nutritionist, Gillian McKeith, who may well share a blithe indifference to water and photosynthesis with Holford. As Andy Lewis of Quackometer expresses it:

there are certain woo-like beliefs that seeds and sprouts are ‘bursting’ with all the ‘energy’ that a plant will need for its life. Utter rot. Plants obtain their energy from photosynthesis and nutrients and water from soil. A seed’s job is to produce a leaf or two and a small root so that it can start extracting the stuff from the environment that it will need to grow. In that sense, a seed is no more special than any other plant matter. [Luckily] seeds do not contain all that energy the nutriquacks talk about. Imagine the energy in an acorn required to make an oak tree. One wrong tap and it would go off like a nuclear bomb. Dangerous walking in Autumn.

As for eating nuts or seeds with fruit: i) that would surely depend upon the fruit (not all of them have a high glycaemic index or load, eg, plums); ii) not everybody has unstable blood sugar of a form that is clinically relevant; iii) one might imagine that the fat content of nuts and seeds would retard emptying of the stomach and contribute to any sustained feelings of fullness.

From time to time, one does wonder what the University of Teesside would have done if Holford had ever attempted to teach his students such nutritional science or nutritionism.

In a comprehensive example of the saying, “the parts that were good weren’t original and the parts that were original weren’t good”, Holford derails his post about the cholesterol-lowering benefits of walnuts to digress onto shark liver and almonds by way of an irrelevant comparison of the ORAC value of walnuts relative to tomatoes and a side-swipe at dietitians.

Only 7 walnuts gives you 2,000 ORACs – that’s the antioxidant power of half a dozen tomatoes…Often slated by dieticians due to their high calorie and fat content all nuts are good sources of phytosterols which lower cholesterol. They also contain vitamin E (alpha tocopherol) and other tocopherols. Almonds are a particularly good source of calcium and magnesium. They also contain something called squalene which is cancer preventive, great for your skin, immune boosting and helps lower cholesterol. The richest source of squalene is shark liver oil.

We have previously mentioned that it is unlikely that most people regard ORACs and what are commonly understood as antioxidants (eg, vitamins A, C and E) as equivalent. Plus, it seems a little odd to compare a dense food with a small water content with a foodstuff like tomato that is largely comprised of water.

Registered Dietitians regularly praise the Mediterranean Diet and similar diets that include 1-3 servings a day of legumes and nuts (we should remember, of course, that people with allergies to nuts or seeds need practical alternatives). The BDA’s Fat Facts information leaflet (last updated in 2004) explicitly praises nuts.

In smaller amounts [omega 3 fats] can be obtained from plant foods such as rapeseed, walnuts, soya, flax and linseed oils. We really need to increase our intake of these healthy fats. Not only can they can keep our hearts healthy, there is growing evidence that they offer many other health benefits as well.

Similarly, the BDA’s Cholesterol Facts information leaflet:

Make sure your daily diet contains …wholegrains, at least five portions of fruit and vegetables, low-fat dairy foods and lean cuts of meat and fish as well as eggs, nuts and pulses.

I have a small concern about Holford’s implicit handwave assurance that “eating lots of nuts and seeds” is desirable for the general population but have opted to put it in a footnote.[a]

As for almonds, it’s a little difficult to see what they have to contribute to a discussion about walnuts but suffice it to say that a PubMed search for squalene cancer almonds did not return any papers although there was a single result for squalene cancer nuts. So, at best, it seems as if Holford’s claims for almonds and squalene might be a little premature.

As ever, HolfordWatch is not sponsored for any of its writing and we are not lobbied by anyone so, with hand on heart, a clear conscience and innocent of undue influence by shark lobbyists, we report that there is a comparatively tiny amount of research concerning shark-derived squalene and cancer and that most of it is at the stage of petri dishes or small-scale animal studies. Not the sort of evidence that would lead most of us to promote shark liver squalene as “cancer preventive”.

As for the other claims relating to skin care etc. -leaving aside the state of the evidence for those -we must report that deep sea sharks are harvested for squalene for cosmetics products and their population is declining sufficiently that some manufacturers have been persuaded to remove it from their brands.

There is a distinct lack of a respectable cohort of high-quality evidence to support Holford’s assertions, so, please, cancel the trip on the factory fishing-trawler or the order for shark liver oil capsules from your favoured purveyor of supplements and leave the sharks alone. If you are interested in squalene, it is present in amaranth, olive oil, rice bran and wheat germ.

In summary, in the space of a short blog post cum email, Holford over-extrapolates the evidence from a systematic review and meta-analysis to support his own, inconsistently-expressed views on nuts and seeds; neglects to mention appropriate caveats; is currently premature in promoting squalene as a “cancer preventive”; makes inappropriate comparisons of the ORAC values of substances that have very different water percentages (let’s overlook the issue of the relevance of ORACs for now); and perpetuates the familiar, embarrassing nutritionism assertion concerning the awesome nutrient power of nuts and seeds.

We would remind you that this nutritional knowledge is disseminated by a former Professor at Teesside University. BANT is supporting an attempt by nutritionistas trained in similar beliefs and with similar standards of evidence to infiltrate the NHS, specifically in association with a vascular check programme. If this is allowed to happen, somewhere soon, in a clinic near you, somebody will be solemnly told all about the remarkable nutrient content of nuts/seeds that re-interprets scientific understanding and, doubtless, the equally awesome power of supplements. Shark liver oil capsules, anyone?


[a] The authors of the systematic review and meta-analysis acknowledge that there are several confounders and limitations that arise from the included trials but these do not appear in Holford’s account.

Although certain positive results were seen with these trials, there are some limitations. Primarily, the studies had relatively small sample sizes and short durations of follow-up. For the 3 parallel design studies, the small sample sizes could have led to ineffective randomization and potential confounding (20, 25, 26). Although authors adjusted for some covariates, it might not have been enough to eliminate bias. However, when we excluded these studies from the subgroup analysis of higher quality trials, it was apparent that they did not appreciably affect the results. The small sample sizes also decrease statistical power to detect minor, but clinically significant, changes. Long-term effects are clinically important for lipid profiles and other CVD risk factors. The longest follow-up time was 6 mo so presumed health benefits cannot be extrapolated beyond the duration of these studies. Furthermore, because lipid profiles change soon after switching diets, walnut consumption would need to be maintained indefinitely to maintain lower lipid concentrations. Compliance with the diets was acceptable according to authors, but long-term adherence is often a concern with dietary interventions. Finally, the amount of walnuts consumed in these trials was relatively large, representing 5–25% of total calories (30–108 g/d). This level of consumption might be difficult to maintain in a nonresearch setting.

The authors make some useful observations and caveats about their own paper. However, it might have been useful if they had explicitly mentioned that small sample sizes mean that it is difficult to observe any adverse effects such as for people with phytosterolaemia. Phytosterolaemia is a particularly interesting disorder because few cases have been identified but there is some concern that people with phytosterolaemia are being misdiagnosed with hyperlipidaemia (high cholesterol or a form of familial hypercholesterolemia). Essentially, people with this condition exhibit a mutation in the relevant genes that both allows phytosterols (including stanols) to pass into their circulation more easily and also blocks their excretion, resulting in 10-25 times greater levels of phytosterols in the circulation than in non-affected individuals.

This is not a case where because some phytosterols are good, more must be better. People with phytosterolaemia are at greater risk for: premature atherosclerosis; premature death; coronary heart disease and associated chronic disorders; accumulations and deposits in the tendons that provoke symptoms that resemble arthritis; and liver damage.

The BDA acknowledges that general advice on cholesterol-lowering might not suit everyone:[b]

Reducing blood cholesterol is only one modifiable risk factor in reducing heart disease and needs to be considered
in the context of other risk factors. Specialist dietary advice should be given to those with inherited disorders.

The BDA also stresses nuts should preferably replace other sources of fat in the diet rather than be used in addition to them.

Plausibly, although someone with phytosterolaemia might be able to handle a modest amount of phytosterols from nuts or seeds (I can’t quantify modest because there isn’t sufficient research), this might not be true if they are also using functional foods that are enriched with phytosterols as the quantities in such products can be high. However, people with phytosterolaemia or who have chronically high cholesterol levels despite compliance with lifestyle recommendations and/or statin therapy should be guided by their health adviser on an appropriate diet that addresses their special needs.
[b] The BDA might need to consider updating some of its information leaflets in light of the recent Clinical Guidelines and Evidence Review for Lipid Modification: cardiovascular risk assessment and the primary and secondary prevention of cardiovascular disease. The authors of those guidelines report that there is “insufficient evidence to recommend [the use of plant sterols or stanols]” (pg. 119).

People should not routinely be recommended to take plant sterols and stanols for the primary prevention of CVD. [pg. 104]


Filed under dietitian, nutrition, patrick holford

6 responses to “Patrick Holford, Shark Liver Oil and Walnuts

  1. Pingback: Twitted by jdc325

  2. If you replace 25% of your calories with walnuts, isn’t that 25% less calories to be provided by “bad”, cholesterol-raising foods?

    In other words, how do we know that the benefit of substituting 25% of your calories with walnuts wouldn’t also be provided by substituting 25% of your calories with anything else vaguely “healthy”?

    Did any of these studies have a proper control group?

    • Short and long answer. For some of the studies in the meta-analysis and systematic review the nuts were an add-on to the diet, in others they were a substitute for other sources of calories and there were adjustments elsewhere.

      Overall, the 13 studies selected for inclusion represented 365 participants, with diets lasting between four and 24 weeks (6 weeks on average) and walnuts providing between 10 and 24 per cent of total calories.

      One study allowed 15 g of walnut oil in addition to the 37 g of walnuts (15). Another study incorporated the walnuts into meat products to facilitate participant blinding (16). It was apparent from the remaining studies that patients were given or recommended specific daily amounts of whole walnuts. The control diets were diverse, as outlined in Table 1. Four studies used low-fat diets, with a cutoff of ,30% energy from fat in 3 studies and ,20% energy from fat in the fourth study. A Mediterranean-style diet was used in 3 studies. Three other studies used the average diets for the country: 2 average American and 1 traditional Japanese. One study reported that the control meals were cholesterol lowering, and another specified that control meals were provided but did not elaborate on the composition.

      The studies were selected to be controlled trials – the control diets included Standard American, Japanese and Mediterranean. Some of the studies provided the meals.

      Studies were accepted for the meta-analysis if they were a controlled trial that evaluated a walnut-enriched diet compared with a control diet.

      I have to agree that substituting 25% of calories with something approximately more healthy than the usual choice might be A Good Thing.

  3. seapea

    I hope these replies are moderated.
    I agree with everything you have said, but you need an editor. Your thoughts are well organized but not clearly expressed.

  4. seapea

    I just left a comma splice in a comment criticizing your writing!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s