Why ova counts are worthless for measuring helminth population

Egg counts were once commonly used to estimate helminth populations in infected humans, and more commonly in animals. They were used to determine whether or not a treatment to kill the helminths, a process called helminth therapy, was appropriate. Before modern anti-helminthics in particular treatment was very unpleasant, and quite dangerous, so treatment was far more dangerous in the case of light infections than to leave them to die of old age.

This policy of only treating large infections prevails, only subjects with high populations of hookworm for instance are supposed to be given anti helminthic drugs (helminth therapy) according to CDC policy (see graphic from CDC image web site below). It was this customary use of the term helminth therapy that lead me at the very beginning to adopt the use of helminthic therapy, the two phrases meaning exactly the opposite.

I do not claim the term’s invention, I read it in an early paper since lost speculating about the possible use of helminths via deliberate infection to treat diseases like Crohn’s.

The arguments against egg counts as an indicator of helminth number

  • Parasitology texts note that each day a female hookworm or whipworm produce from 2,000 to 20,000 ova. For this reason alone any estimate based on an ova count must have a variance of x10. That is the answer can only be expressed as a range, with the higher end of the range being ten times the low end. So a typical answer would be “from 20-200 hookworm”. Utterly useless if you are trying to detect the loss of one or two, or five or twenty, hookworm. On this basis alone the test is too crude for the purposes most with deliberate helminth infections are trying to use it for. 
  • Nor are the numbers, 2000 at the low end, 20,000 at the top so commonly quoted as simple facts accurate or agreed on. Nor do we know why this variance is stated in the texts on the subject, many of which record very different minimums and maximums for daily ova production.* Ova production of each species is not a settled, precise and agreed upon number. The cause for this may be any of a number of reasons. Speculating one can easily think that amongst which are that the ranges reflect the weakness of the original study used to determine them, and the author’s acknowledgement of that uncertainty in the results. Or that they indicate ignorance and were originally stated as an estimate. It could as well reflect a long forgotten disagreement about what they are, and so on. We do not know, all we know is that different texts quote vastly different numbers for the upper and lower limits, based on long forgotten original sources. So as vague as an answer as “from 20-200 ” is, it is not vague enough.
  • Volume of stool throughput: An egg count is a measure of density, so it can be affected by the how much stool the day’s output of ova, whatever that is, ends up in. It is easy to see that both the volume of stool produced in a given day, as well as how fast that stool moves past the jumping off point, can affect ova in stool density measurements. Any factor affecting those two things can affect the results of an egg count. Different foods and drugs affect the speed of material through the intestines, as well as the volume of stool produced by the amount and type of food eaten in a given period.  Things like the speed of material through the intestines (constipation or diarrhoea at the extremes), amount eaten, amount of fluid drunk, the nature of the food, fibre content of food, etc. If you think of the extremes one can see this is going to have an enormous impact on density of ova per gram of faeces. On Monday you may produce an estimate of 10-100, two days later 20-200. Has your population magically increased, doubled, in the interim? Or did you simply get over the MSG you ate on Sunday night?
  • Ova output: Ova production in all helminths is affected by various drugs, some known, most inevitably unstudied and therefore unknown, for an effect few care about. So, antibiotic use so profoundly reduces ova production that parasitology texts recommend not doing ova tests for two weeks after the subject stops taking the antibiotics. Other drugs almost certainly have various effects as well. What are you taking, and what are you eating, or drinking, etc? Any idea of their impact on ova production?
  • Methods, precision and consistency: Any count depends on extreme precision and replication of methods and precision from test-to-test. Only an experienced lab technician, with a gram scale accurate to a hundredth of a gram, and with lab moves as precise, is capable of accurately counting ova in stool. I forgot that they need good control of their gag reflex, too. Stains are no aid, there are none. Identifying each and every ova in a gram of faeces is difficult work, look at how useless western labs are for even detecting ova in infected individuals. In someone producing 100 ova per gram .1 grams will contain 10 ova, so the slightest variation in weight can have a big effect on the number observed, that effect then being amplified by the multipliers used to derive worm population from egg counts. Is the person conducting the test fastidiously scraping and washing ever particle of faeces off the gathering and weighing devices to examine it. Have you ever dealt with a gram of faeces? It may not seem much, but it quickly becomes so when you have to examine ever particle of it looking for an object with the same colour as the background material, that is about 40 microns wide. And then propose to do it a few days a week until… when you stop.
  • Ova production varies tremendously with time. It is well documented for instance that hookworm ova production falls by an estimated 50% about one year after infection. But there is little information on whether this is invariably 50% or exactly when it happens or how fast. Or how fast the drop occurs, etc. Nor is the observation that this drop occurs any guarantee that it is the only drop, what about an increase, during a hookworm’s lifetime. In fact the addition of even a few juveniles to an adult population, before they even reach sexual maturity and start producing eggs, somehow stimulates the already resident adults to increase ova production. But no one has documented for how long.
  • Variations in ova producing due to genetic differences in helminth populations:* Within a species infesting somewhere in the vicinity of 700 million people worldwide, from Australia and Polynesia, so Zambia and Cameroon, on to Mexico and Columbia, there has to be significant genetic variants within it, N. Americanus. Given the small number of studies, and that they were all completed no later than the early 20th Century when tools were not as sophisticated as now, it is likely that there will be significant differences in ova production from one infestation to another based on genetic variations between local hookworm populations. Do the useless numbers for average ova output per day on which this entire test is based even pertain to the hookworms you host?
  • Variations in ova production due to differences in host genetics: Helminths because of the nature of their relationship to their host, and in particular for the species used by Autoimmune Therapies in helminthic therapy, have to play tricks on host immune response in order to feed, which is why we use them therapeutically of course. If stressed by a robust immune response, as in someone with Crohn’s or Colitis, which I believe are disease that arise in part out of a genetic adaptation to high worm burden environments, hookworm or whipworm inhabiting such a person will produce fewer ova per adult female because those helminths are subjected to a stronger host immune response directed at them, affecting their vitality and ability to feed. Furthermore this is likely to be particularly true of hookworm or whipworm. There is further evidence that these diseases, in their most common forms, do arise out of adaptation by humans to high hookworm and whipworm specifically burden environments. Which is that these diseases typically manifest as chronic inflammation affecting the tissues exactly where hookworm (the terminal ileum) and whipworm (colon) attach to feed. So an individual with Crohn’s affecting the terminal ileum as one example is very likely to produce fewer ova per adult female hookworm than someone with those genes. Has anyone gotten around to determining and publishing a table of differentials based on host genotype for helminth ova production, similar to the differing coefficients of thermal expansion for different elements and alloys? Did the study subjects have the Crohn’s genes but not the disease, did they not but you do, etc?
  • McMaster egg counts have been abandoned by doctors of veterinary medicine as being useless as a practical tool for estimating worm numbers. They would know.

So save yourself the time, and the not inconsiderable inconvenience of working with your own faeces, as well as the expense. McMaster egg counts are worthless for population size estimates.

The only method of any utility is impractical, full endoscopy to survey populations in situ.

There is one other method.

Which is the elimination and necroscopy and count of dead hookworm, itself prone to error since one has to sieve the bowel movements of the subject for at least 24 hours after a dose of albendazole sufficient to kill them all. Adult hookworm are about 1.4 centimetres long and half a millimetre wide, so be careful they do not get crushed or missed in the process.

Or, much more fun, the same process for four days starting from the first bowel movement following ingestion of the first dose of mebendazole for whipworm. Most of which are a threadlike structure as thick as a human hair.

Notes:

The Public Health Image Library does not use static links for their images, I think they do this to prevent other sites linking to images there thereby putting the hosting and bandwidth charges on to the government. But if you want to find the original to verify it you can go here and search for “hookworm” or perhaps by “5245”, the image number.

 

Click to enlarge, image obtained from the Public Health Image Library USA, from http://phil.cdc.gov/phil/quicksearch.asp Search term “Hookworm” Image number 5245

Notes on this expanded version of my original post

At the time I wrote this a lot of people were using McMaster egg counts to try and track their helminth populations, despite my arguments, repeated arguments, that it was not fit for purpose, and unnecessary. Worse, the practice was being promoted, and the practitioners were using their results to draw erroneous conclusions, which to some had the air of validity. Perhaps because a microscope was involved?

Rather than have to restate my reasoning again I decided to write this post to link to. At the time I was fairly exasperated, and was not enjoying the online experience at all. Many of those who disagreed with me were quite shrill, and it got personal. As a result did not bother to repeated again all my reasons, the first alone seemed adequate to me. I have since expanded the post to include all my reasons.

But it was a serious situation, they sometimes used these erroneous results as a basis for demanding more hookworm, sometimes in my estimation the numbers involved could have caused anaemia.

Naturally this caused a lot of friction and upset, someone with a microscope KNEW they needed worms, the texts said it was a valid method for estimating the adult population, so I was just being stubborn and unreasonable. Because of the fact that textbooks persist, to be referred to and brandished, and are not updated to reflect changes in practice the arguments persisted. Even the, very happy for me, interjection of a veterinarian who validated my stance by saying McMaster egg counts had been abandoned years before, for the very reasons I gave, did not dissuade the zealots.

It was this event with discussion groups, before Facebook became popular, that for me was the turning point, after which I have not participated in online discussions much at all.

This is another reason to read science critically and never to blindly accept something just because someone managed to publish a book, or because they have a few letters after their name. While that may seem counter to my statements above, it is not. In this case reason, and many different reasons derived from it, was supported by the stance of the expert. Always think critically, and at the same time be flexible and humble enough to admit your errors and move on.

 

One thought on “Why ova counts are worthless for measuring helminth population

  1. Someone said to me after I had dug this out and updated it that it was no longer valid, that those who do this type of thing with helminths were using “fecalizers”. These have been around for years, I considered buying some, they come in bulk, to concentrated stool samples, or ova derived from them when we used to do stool tests to prove infection. It seems the person pointing this out to me is under the misapprehension, and indicated it was widespread, that improving the number of ova you actually detect in a given quantity of stool magically somehow makes the above points null and void. Think about it, ensuring you actually detect most or all of the ova in a gram of stool makes no difference to the other profound problems with egg counts, and I really did not address the issue of operator error or skill above, I did not want to inflame anyone needlessly. That is the only factor affected by using an ova concentrator like the fecalizer. Great name though, and great for it’s intended purpose: to concentrate ova efficiently, and to ensure even light infections with low egg counts are detected.

Comments are closed.