BMJ 2010; 341: 259 – 308 (7th August)

First off, calcium supplements (article on page 289, editorial pp 260 – 261). You remember how we’ve all been advised to start everyone on calcium supplements when we start them on bisphosphonates? Because all the studies on bisphosphonates were actually on using calcium plus bisphosphonates, so nobody was sure if bisphosphonates would really work the way they were supposed to without their trusty sidekick? Bad idea, it turns out. A meta-analysis on the subject of calcium supplements backs up what was initially shown by a study a couple of years back – they increase the chances of a heart attack. By around 30%. Since your typical patient on calcium supplements is likely to be elderly and frail, probably with co-morbidities, that 30% of existing risk is going to add up to a clinically significant risk.

So what, if anything, do we do about this? Do we stop putting patients on calcium? As the editorial points out, we don’t actually have much in the way of evidence that it’s doing any good, and we do now have solid evidence that it may be doing some people harm. This issue, as it happens, came up straight away for me; the manager at the nursing home I cover asked about the calcium supplement I was supposed to have prescribed for Mrs X and which never made it to the pharmacy, and I told her that, ooops, I must have forgotten to print the prescription off again (having just spent eight and a half years working with a computer system that defaults to printing all prescriptions at the end of a consultation and then moved two months ago to a computer system that defaults to not printing prescriptions at the end of a consultation, I have something of a habit of this), but, you know, funny she should bring it up, because, as it happened… So I explained about the study and she said, oh, yes, a couple of years ago Dr B. (the colleague of mine who covered the nursing home before I did) came in for his weekly round and crossed off the calcium supplements for every patient who was on them, and told me a new study had just come out showing they were harmful. (The initial study, I asume.) Then new patients (she continued) came in who were already on calcium supplements and she asked him whether he wanted them crossed off [the supplements, not the patients], and he said, no, turns out they’re not so harmful as initially thought. (The inevitable follow-up heated debate, I assume.) I told her that I’d wait on the supplements for the time being. At least until I have a chance to read all the follow-up letters and see how the general consensus of opinion shakes out at the end of it all.

Anyway, I digress. What do we do about putting patients on calcium supplements? What do we do about the patients who are already on it? What do we do about patients who need Vitamin D supplements, given that it now seems to be practically impossible to get them without calcium attached? And, on a deeper level, what lessons have we learned, boys and girls, about prescribing something on a might-as-well-because-it-might-help basis with no solid evidence that it actually is of help? Even something as innocent as a substance you find in your glass of milk can, it seems, be harmful, so maybe we should all think a tad bit more carefully before adding yet another drug to the list of official recommendations. Why do I get the feeling that that isn’t going to happen?

Another thought-provoking point comes from the Short Cuts page, from a summary of an article published in the Annals of Internal Medicine (page 277). This one looked at the advance directives homeless people came up with if asked, and found that they were much less likely than the general population to conclude that they wouldn’t want to be resuscitated if in a coma. Does this mean that they actually value their lives more highly than the general population? Or just that they’re more used to the idea of living, and finding some satisfaction in, a situation that most of us would assume to be intolerable? Either way, it’s the opposite of what I would have expected, so it makes me think, which is always good.

On page 288 is a study showing that white matter hyperintensities on brain MRIs are associated with an increased risk of stroke, cognitive decline, dementia, and death. I can’t offhand recall ever reading a report that a white matter hyperintensity had been spotted on the MRI of any of my patients, but now I know what to tell them if I do. Though it’s not as though there’s much you can do about it, so ignorance is probably bliss. On page 290, we learn that beta-HPVs are associated with increased risks of SCCs.

Finally, on pages 297 – 8, a really useful article about how to investigate mildly abnormal ALTs. Points in history-taking come under the general category of considering what patients might be putting into their bodies that might damage their livers – alcohol, prescribed or OTC medications, or viruses. (In the case of the latter, of course, what you would be asking about would be risk factors for Hep B or C.) Next, a couple of useful points about the actual levels – it seems that levels less than 5 times the upper limit of normal are considered mild and only higher levels are considered severe. Since the first thing that patients usually want to know when their LFTs come back abnormal is “Are they seriously high?”, this is useful information for reassuring them. Though it is, unfortunately, completely contradicted by the next point, which is that it’s possible for even severe chronic liver disease causing substantial liver damage to show up only as mild derangement of the LFTs. So don’t be falsely reassured. Any elevation in the ALT, even mild, that persists for three months or more should really be investigated further.

So how should we investigate further? Well, first off, of course, you repeat the test. Along with it, you’re supposed to check gamma-GT and AST. Raised values of gamma-GT are associated with alcoholic liver disease. (I’m sure I remember reading that they’re associated with a stack of other liver pathologies as well, so I’m not terribly convinced that this is a useful follow-up investigation in someone with a raised ALT.) AST is for looking at AST:ALT ratios. Ratios >2 are suggestive of alcoholic liver disease, whereas ratios <1 typically suggests hepatic steatosis or chronic viral hepatitis. What ratios of 1 to 2 suggest is not specified – probably that the test hasn’t given you a clearcut answer one way or the other. Again, I’m not convinced I’d bother doing it, regardless of what the authors say. (EDITED TO ADD: According to the learning.bmj module, it’s worth doing to look for alcohol as the cause.) Prothrombin time, on the other hand, can be useful in showing up problems with liver function (also look at what the albumin shows). FBC may show a raised MCV (suggestive of alcoholic damage) and/or a low platelet count (hypersplenism and portal hypertension resulting from cirrhosis).


So, let’s say you’ve done all those and the raised ALT is persisting. What are the likely culprits? Alcohol and obesity/Type 2 DM are the two most likely causes. Next lot of recommended tests are:

Hep B surface antigen and Hep C antibody (actually, I’ll just find the box to tick for hepatitis screening, which will be much easier)
Ferritin/iron studes (to exclude haemochromatosis)
Fasting glucose (to exclude diabetes, obviously).

The next lot after that should be:

Antinuclear/smooth muscle/liver-kidney microsomal-1/mitochondrial antibodies (you know what? I’ll just tick the ‘auto-immune’ box). I would have thought these would be in the first round of follow-up tests, but apparently not.
Alpha-1 antitrypsin levels
Copper/caeruloplasmin levels

Somewhere in there, it may of course also be helpful to do an ultrasound, although it isn’t clear where this should place in the hierarchy of usefulness. Personally, I’d put in for it when I start asking for diagnostic blood tests.

EDITED NOV 2012: According to the BMJ learning module, viral screening, ferritin and autoantibodies should actually be in the first round of retesting, as some liver diseases fluctuate and can show up with mild alterations in transaminases on one occasion and normal levels on follow-up. Patients should abstain from alcohol while awaiting retesting. For second-line if all that’s normal, consider caeruloplasmin, tissue transglutaminase antibodies, and alpha-1 antitrypsin.

If Hep C antibody is positive, the patient should be tested for active viral replication, and a negative test should be repeated three months later to avoid false negatives.

Haemochromatosis is strongly suggested by a transferrin saturation >45%. Negative genetic testing doesn’t rule it out – liver biopsy should be done in those cases. First-degree relatives should be screened by genetic testing.

(back to original article)

Finally, I was never quite clear on what non-alcoholic fatty liver disease actually was. It is, it seems, the hepatic manifestation of the metabolic syndrome. Basically, you’re supposed to advise patients on lifestyle changes and look at all their various cardiovascular risk factors – weight, blood pressure, blood sugar, insulin resistance.

EDITED FURTHER, Feb 2014: Just doing the Pulse Online Learning module on fatty liver, and it seems I finally have an answer to what it means if the AST:ALT ratio is between 1 and 2. In patients with established NAFLD, it can be a worrying sign of progression, as the ALT falls as fibrosis progresses. (In fact, so does the amount of fat in the liver, so it becomes harder to pick up on USS.)

One 2010 study in Gut recommended the cut-off for referral should be an AST:ALT ratio of >0.8. Other useful signs of progression are:

  • Raised bilirubin
  • Low albumin
  • Low platelets
  • Abnormal clotting

For scoring of severity, try using the NAFLD fibrosis score (recommended) or the Fibrosis 4 score (uses fewer indicators).

Referral seems to be largely for purposes of monitoring, as treatment is still with lifestyle management even in cases of more advanced FLD. (There is some early evidence for a possible role of glitazones.)

Also, to clear up what was on my part some terminology confusion, stages of fatty liver are:

  • Fatty liver
  • NASH – non-alcoholic steatohepatitis – inflammation as well as the fat.
  • Fibrosis
  • Cirrhosis (develops in about 1 – 2% of patients with NAFLD)

About Dr Sarah

I'm a GP with a husband and two young children.
This entry was posted in BMJ, BMJ 2010, BMJ 341, Impact. Bookmark the permalink.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s