Do debates actually harm medicine?

It was during the second year of DM , I was exposed to an interesting kind of academic activity in a conference -the medical debate. The issue in question was a thorny one – with conflicting data. The speakers were excellent and put forth brilliant points for and against the motion. As it so happened, they were both comparable not just in their credentials as doctors but also in their public speaking and debating skills.
As I attended more conferences, it soon became clear the debates were a fixture in many medical conferences and were eagerly awaited by the audience. The speakers in turn tried to do as much justice to the topic as possible. Unfortunately there were some problems

  • Some speakers were better than others – and their point of view was more likely to resonate with the audience
  • Even if the speakers were convinced of the other person’s argument to some extent, it soon became a contest in hyperbole

Most importantly it became a contest between individuals and not ideas. This is of course, entirely predictable to anyone who knows how the human mind works. We back people with ideas, not just ideas. Arjuna vs Karna, Tesla vs Edison, Steve Jobs vs Bill Gates – we revel in the great individual rivalries. You might scoff at this rather ridiculous over simplification. After all, the attendees of conferences are highly educated people, not given to emotional support of one speaker or the other.
Unfortunately, education doesn’t make us immune to biases and blind spots. The nature of the debate means that, the speakers have little incentive to expound on the gray areas and are forced to make emphatic statements in front of an audience baying for certainty. Medical science is rife with uncertainty and it is this uncertainty that make it interesting. To dissect the gray zone in which the debate occurs, the speakers and the audience need to know a lot of details. Some of these details like statistics are boring to the average listener ,but is at least available in public domain. Others are effectively hidden from the prying eyes of the public and are probably sleeping safe in the digital vaults of the companies that conduct the studies.
The need for certainty of the average audience member and the near universal reluctance to dig deeper into difficult topics makes it much harder to convey the nuances involved. At the end of the day, the audience are at risk of taking home the message they understand / like the best – setting the stage for eloquence based medicine. So I wondered, if the non-specialist audience could be influenced by the better speaker in the debate, would it make sense for the companies to ask for a particular speaker to speak for their product? Perhaps I’m being too cynical – I will leave it to your judgment. The moderator of the debate should be a person with impeccable credentials and should be able to rise above the need to simply give the audience some satisfying soundbites. Luckily in most debates this is the case – but not always.
More than posters, oral paper presentations and invited orations, debates spark enormous interest. With great popularity,comes great responsibility. Sometimes, I wonder if the moderator should make slide – instead of , or after the rebuttal. People remember what they see – so this should probably tilt the balance towards the truth (if one is known).

After every debate, I’m reminded of the final scene in the Kamal movie Nayagan, where the don’s grandson asks him a very simple question “Neenga nallavara,kettavara? (Are you a good man or bad man?)” – to which Kamal replies ‘I don’t now”. Some audience are like Kamal’s grandson – hoping for a simple answer to a very complex and nuanced question. I wish we had Kamals to tell them the truth.


The T-score tamasha

Elderly people fracture when they fall, more often than the youngsters. In order to know who is at high risk, so that we can intervene before the fracture, doctors measure how much bone a person has. This is called bone mineral density (we also need to measure bone quality, not just quantity – but that’s a post for another day)

Bone mineral density is measured by a machine called DEXA.(dual energy xray absorptiometry). What we are measuring is the content of the mineral per unit volume. However only have two dimensional measurement – so we express it as gm /cm2.
Now imagine this – you are checking blood glucose and want to know if it’s normal. You have a value and if it exceeds a particular number, you have diabetes. Unfortunately in bone density, we don’t compare the g/cm2 units . Instead we have two unwieldy scores called T scores and Z scores. The T score is (to put simply) as measure of how far away from the avg bone density of a person in the prime of youth. The Z score is a measure of how far away from your peer group (age). Both these change depending on who you take as reference.
In 1994, the WHO defined osteoporosis as T score of less than /equal to -2.5. This means that approx 1/100 people in their 30s have a bone density as low. Pretty simple.
Unlike bone density, no other parameter in modern medicine is compared to the values in the youth. We don’t say only 1 in 1/100 young women have higher waist circumference (thankfully), even though many biological parameters get worse with aging.

How did this happen?

Back in the 1990s, when DEXA machines became commercial, only three companies were making the machines. They still do. Norland, Hologic and GE. Let’s take a detour and see how a DEXA machine works. It is surprisingly similar to an Xray. X rays are passed from the top and are received at the bottom. The patient lies down in the middle. If the patient’s bone block more X rays from reaching the receiver, we conclude she has more bone. Or bone mineral. Pretty simple right? Well not quite.
From this generalization, we need to convert the attenuation( the loss of xrays while passing through the patient) into an estimate of bone density. Since the patient isn’t a skeleton, the Xrays pass through other stuff like muscles, fat and organs, each of which cause some attenuation. This is also why we can calculate the fat and muscle mass with a DEXA machine. The algorithms which convert this attenuation into the bone mineral density estimate – in gm/cm2 are complex and proprietary.
Because of this proprietary algorithms, the same person’s bone mineral density when measured by a GE Lunar would be, say 6% higher than the same bone mineral density measured by a Hologic machine. Now that’s like saying you measure tomatoes by two balances and one shows 1 kg and the other shows a little more. The natural tendency is to ask which is right? Or perhaps work towards standardizing the measurement.
To standardize measurement, the companies had to work together on the innards of the DEXA machine. Alas – they didn’t. Instead, they decided to bring in a statistical definition based on T and Z scores.

Ever since, the doctor community adopted what the big corporate boys decided. Unfortunately not all doctors like numbers (I know some who absolutely detest math). This has caused a situation where just to interpret a bone density measurement, you need to deal with unwiedly concepts like mean, standard deviation, scores etc. As a side effect, it has made doctors learn these concepts well and think about how their population varies compared with the Caucasians. (a thinking that’s almost never seen when interpreting any other test).

When a change comes, it brings both good stuff and bad stuff along. The yin and yang, the dark and light always exist together. It’s upto us to focus on the good.

Walking beer factories !

Let’s start with a fundamental question – do you need to drink to get drunk? The surprising answer is no. Our body can make ethanol, thus making us walking beer factories. This phenomenon of endogenous ethanol production is called Auto-Brewery.

Auto-brewery is fascinating and very few doctors would think about this when encountering a patient who they think is drunk. Imagine the plight of a teetotaler trying to persuade the doctor that he is drunk,but not because he drank 🙂

How does this happen?

While very little is known about the auto-brewery syndrome, the following are considered necessary for its development

  1. High carbohydrate intake
  2. Prolonged stay of the food in the gut due to
    1. Gut dysmotility
    2. Surgical alteration , creating a ‘vat’
  3. Colonization by organizations which cause fermentation of carbohydrates
    1. Candida spp
    2. Sacharomyces cerevisiae – both due to prolonged antibiotic therapy


  • Medical
    • Chronic -The above 3 conditions tend to coexist in one of the commonest patient groups- obese diabetics. They may have autonomic dysfunction leading to intestinal dysmotility. Many of these patients also have Non Alcoholic Fatty liver disease -NAFLD. If a subgroup of these patients have endogenous ethanol production, their liver disease may well be ‘alcoholic’. There is very little data to support or refute this claim. In a study of blood alcohol levels, the diabetic patients were found to have 5 times as much BAC (blood alcohol concentration) as non diabetics (1). While the authors conclude that this is not enough to be picked up in routine breath analyser tests, the implications of such long standing increased BAC on liver are intriguing, to say the least.
    • Acute- Acute alcohol intoxication has been reported in a patient who hasn’t touched alcohol in 30 years (2) ! Some cases of metabolic encephalopathy in which no apparent cause is forthcoming may be due to ethanol intoxication from endogenous production. However we must be careful to rule out the usual suspects and should only entertain this diagnosis if BAC is high in a teetotaler or abstinent patient.
  • Medico legal
    • It is unlikely that endogenous ethanol production is severe enough to cause positive breath analyser test in cases of drunken drive (3). This line of argument also doesn’t hold much water in the courts if the published medico legal literature is anything to go by.

To conclude auto-brewery is an interesting medical oddity. As the mechanistic insights are still not very clear, we must be cautious in making this diagnosis. Whether auto-brewery is the culprit in a subset of diabetic patients with neuropathy with NAFLD is not known.

Further reading

  1. Hafez EM, Hamad MA, Fouad M, Abdel-Lateff A. Auto-brewery syndrome: Ethanol pseudo-toxicity in diabetic and hepatic patients. Hum Exp Toxicol. 2017 May;36(5):445–50.

  2. Welch BT, Coelho Prabhu N, Walkoff L, Trenkner SW. Auto-brewery Syndrome in the Setting of Long-standing Crohn’s Disease: A Case Report and Review of the Literature. J Crohns Colitis. 2016 Dec;10(12):1448–50.

  3. Logan BK, Jones AW. Endogenous ethanol “auto-brewery syndrome” as a drunk-driving defence challenge. Med Sci Law. 2000 Jul;40(3):206–15.

Drugs: Do we get what we pay for?

Most doctors are very particular about the insulin dosing : for a good reason. Small changes in insulin dosing can result in undesirable changes in blood glucose. Sometimes we give higher concentration of insulin when patients require a larger dose. One assumption is that the patient has some sort of insulin resistance/secretory defect and that the increased dose will take care of the problem. Endocrinologists also enquire about the insulin storage – and we have all met the occasional patient who has stored insulin in the freezer or made some egregious mistake in the insulin injection technique.

Hitherto, we have assumed that , if you buy a vial of insulin containing 300 Units of insulin, it contains 300 IU of insulin. After all, what could be more natural than this?.

Today I came across a study that challenges one of our most natural beliefs. You can read the paper here.

Study in brief

Carter and Heinemann purchased insulin vials from different pharmacies in the United States. They quantitatively tested the insulin concentration in 18 10-ml vials from two major manufacturers (if you are into diabetes management, you won’t get any brownie points for guessing). They used a standard analytic method – Quadrapole Time-of-Flight (QTOF) mass spectrometry to quantify insulin.

Unfortunately, not even a single sample contained the amount of insulin that it should contain. The concentrations ranged from 13.9 to 94.2 IU/ml ( for an insulin that should contain 95 IU/ml). The lossis slightly better with NPH insulin, as shown below. However NPH throws a different problem – high variability between vials.So a patient might get only 1/6th of the regular insulin that he pays for.Even the lot to lot difference is more – in essence, while we might prescribe some dose of insulin, only God knows what’s going in. ( call it luck if you are an atheist)

What makes this issue even more troublesome is that neither the patient nor the doctor can do anything about it. Even the pharmacist can’t do much about it, because the problem ( as postulated by the authors) is more proximal – perhaps a break in the cold chain.

This study is the first to show such huge loss of insulin at the last link in the cold chain (pharmacy). The study was not done in some remote African village- but in the USA. In a hot country like India, with poor knowledge about storage of insulin, I wonder what the situation will be. There are no similar data for analogue insulins.

We have had similar problems with thyroxine (especially the government supplied ones). Similar issue was noted with Vitamin D preparations in a study conducted by AIIMS.You can access it here.



  1. If the blood glucose control changes from month to month, it may have something to do with the lot to lot variability
  2. Endocrinology is more than just numbers.
  3. Much of our certainty may be ill founded. The numbers we encounter are at best, estimates and approximations.
  4. We need demand better drug quality control – after all our patients are paying for the drug.

Painless pricks

Problem: Your diabetic patient is not checking blood glucose frequently. She is sick of pain in the prick sites and would rather not check blood sugar

Solution: Make the pricks painless and hope the patient will check blood glucose regularly

Some people might scoff at the solution. How can you make drawing the blood from your patient’s finger painless? It doesn’t make sense? You might perhaps think that making the needle thinner should help. Unfortunately that wouldn’t be enough. Because even thinner needles also have to prick deep enough to draw blood. Besides one might need a thinner needle that goes in just the right amount to draw blood. Often the patient squeezes the finger – which leads to more pain.

One option is to use Lancing pens – they are significantly better than just pricking with any old needle. However they still continue to be painful. This is a problem especially for children with diabetes – there is ample data to show that frequent glucose monitoring and corrective steps are necessary to achieve optimal glycemic control.

It is in this backdrop that a device has been introduced. It’s called Genteel – and it promises pain free blood glucose testing. How does it make a prick  pain free?

  1. It uses vacuum to draw the blood vessels
  2. It plunges only deep enough to hit the blood vessel ( for most people), and avoids the nerve endings*
  3. It has a vibration mechanism that distracts the patient from feeling the minimal pain. The end result is an almost painless prick. The device isn’t available in India at the moment, but it can be ordered online for a price of $ 119(after 10$ discount). (shipping costs extra).

I couldn’t find evidence that this device reduces pain/ improves the frequency of glucose testing /improves glycemic control. There are various testimonials by children who have used it and found it to significantly reduce the pain. Since they probably don’t have any commercial bias, I decided to take their testimonial at face value. However that doesn’t negate the need for some solid scientific evidence.

Realistically we have three choices
1. Wait for the evidence to accrue and avoid experimenting
2. Have a demo pen in the clinic and have the kids try it out. If they like it they can buy it.
3. Do a trial of the device in Indian setting – preferably including the soft end points such as pain and frequency of glucose testing and harder ones like HbA1c.

I favour option 2&3. The reason is simple – it isn’t too costly, there’s a 4 month trial period during which the device can be returned if found useless, and unlike non invasive glucose monitoring systems this one isn’t some black box approach to diabetes.

There are other approaches to relatively painless blood glucose testing. One option is Abbot Freestyle LibrePro. Unlike Abbot Freestyle LibrePro, Genteel is an one time investment. And it can be easily shared among family members*.(the lancets have to be changed)The advantages are obvious. In case you know some type 1 diabetic child who could benefit, please share it with your patient.

Disclosure: I have no conflicts of interest to declare.


The Plain Language Movement & Law

The plain language movement started in both sides of the Atlantic in the 1970s to make law easy to understand. The legal documents were plagued by legalese and were thus inaccessible to the commoner. This problem can be traced back to almost a 1000 years when William, the Duke of Normandy defeated the Anglo-Saxon King Harold in the Battle of Hastings in 1066. As William and his followers spoke a dialect of French, English became the language of the common and lowly folk.
The courts and lawyers soon followed suit. Within a few decades the Legal system had became inscrutable to the common man. With the ascendancy of English came the urge to rid the system of the French and Latin terms and replace them with crisp Anglo Saxon words. The push to make common sense in common language fashionable had a reasonable amount of success.
The legal system and the people benefited a lot from making things simple. Unfortunately, the Plain Language movement only focused on the law, not medicine.

Saving Medicine From Medicalese

Flip(or click) through the pages of any medical journal and you will see how hard our language has become for anyone outside our profession to make sense of. Even among doctors, each discipline has its own jargon and stylistic idiosyncrasies making it harder for others to understand. We live in a time when obfuscation is celebrated as a skill and straight talk is scoffed at.
To give an example, I was reading a top endocrinology journal yesterday and was dismayed to find that the pages have been hijacked by genes, genes and more genes or molecules,molecules and more molecules. It felt like the journal had written in 100 size font in invisible ink – look, this is for the experts. No one else is welcome.
I am not arguing that the top journals should dumb down their content or ask authors to keep click baity titles. However I’m certain that the scientific community will be better served by a Cochrane style plain language summary for every scientific article. In fact developing a written version of the elevator pitch is likely to narrow our focus on what matters. However, most journals don’t have the space/ inclination for such summaries. We need a plain language movement for medicine.

What can we do in the meantime?

Kudos. It is a free online service to explain about your research in plain English. Each paper gets these four pieces of information – Title, What is about, Why is it important and the Perspectives of the author. Kudos also provides shareable links and can automatically post to Facebook, Twitter and LinkedIn. It can even track the response your article is generating! (It’s like having your own Altmetric dashboard)
Here’s a plain language summary of one of our papers – Tumor(s) Induced Osteomalacia- A curious case of double Trouble
If you are an academic, check out Kudos. It’s free and the experience can help you focus on what matters.

The Insulin plant

Yeah..you read that right. There’s actually a plant called insulin plant and it’s supposed to reduce blood glucose levels(no surprise there).You have probably heard of several natural remedies for diabetes and are rolling your eyes now.. The diabetes armamentarium is brimming with antidiabetic agents which are effective and proven. Some like GLP1 analogues and SGLT2 inhibitors have even proven to have cardiovascular benefits.

So why bother about a plant?

For a couple of reasons

  • Research : There are many plants which have potential antidiabetic diabetic activity – infact there are at least 111 plants which are known to reduce blood glucose. (1). However the Indian patent laws do not allow patenting plants and more importantly medicines derived from natural products. I have always had trouble understanding the second clause – even if you do some fancy chemical extraction and make a useful substance that was essentially hidden underground for millennia, you wouldn’t get a patent in India. Consequently the incentive to exploit the “natural remedies” for commercial gain is very limited. Thus,most of these plants/plant based substances may never reach the market as a tablet. Does that mean we can’t study them or learn from them? Not really – one can essentially mimic a natural substance, tweak it, call it bioinspiration and pretend that the molecular structure was an epiphany during a coffee break ! Or at least apply for an AYUSH grant to do some research – I’m a novice here, but I guess there can’t a better time to apply for AYUSH grants than now. Even if we aren’t involved in the business of making drugs, if the natural form is safe enough, we can consume them. Even if the effect is modest.
  • Clinical: Apart from the research aspect, there is a huge public craze for cost effective natural remedies or drugs derived from plants. The runaway success of products like BGR34 is a testimony to this.

Now you might wonder, if this plant stuff is good, it should have a good scientific backing. Indeed there’s a good body of research behind this. But let’s be frank – research is often locked behind paywalls. Even when it is ‘accessible’ it isn’t truly accessible to those outside the profession – most people are turned off by graphs,tables and statistics. The idea of this post is to simply strip the complexity off the published scientific literature and bring the reader upto speed on this quirky plant.

Here’s a brief bio of the insulin plant in Q&A format

What exactly is the insulin plant?

This plant belongs to the Costaceae family – two species are common , the Costus igneus and the Costus pictus. The leaves of this plant are sometimes taken as supplements for reducing blood sugar. Known as the Spiral flag( insulin chedi in Tamil and Malayalam), the plant can grow upto 2 feet and has colorful flowers.


What does the plant contain?

It contains triterpenoids such as α and β amyrin,lupeol, stigmatsterol.,Diosgenin etc. That’s a lot of active principles- but mostly we are yet to understand how these substances interact with one another and whether isolating them is more useful than the natural mixture in which they are found.

How do I get this plant?

The insulin plant can be obtained from a nursery or someone who is already using it. Care should be taken to avoid mistaking some other plant for this. For the purposes of research, the identity of the plant needs to be confirmed by the Botanical Survey of India,Coimbatore. They give an authentication certificate with a number and date.

Is it safe for human consumption?

Published Toxicity studies in animals show no major toxic effects in the short term (2). Anecdotal human evidence seems to support this. However one should remember that with plants/plant products, there are a lot of variables one must account for – subspecies,soil,part of the plant, extract or whole leaves, growth in shade vs sunlight etc. Since there are no published long term human studies, we are essentially on our own when consuming this. Consequently, those at risk of hypoglycemia (elderly, recurrent hypos, comorbid illness, kidney diseases) and pregnant women should strictly avoid experimenting on themselves.

Is it effective in reducing blood sugar?

Much of the published research on this plant is from animal studies. These animal studies generally show a reduction in blood glucose. You can get a gist of the published research in the form of table by clicking here

Homogeneity is hard to obtain in these studies. Only limited human data is available. The absence of data doesn’t mean absence of useful effect though.

Does it have any other uses?

These days plenty of drugs reduce glucose. It is only natural to expect more !. Plant products tend to have pleiotropic effects and may well have off target effects which we don’t want. There are some of the effects of the insulin plant.

  1. Hypolipidemic effect
  2. Antioxidant effect
  3. Diuretic effect
  4. Anticancer effect
  5. Reduces TSH (3)

What does the current research mean?

Very little is known about the insulin plant – especially the human use of it. However, with the public clamor for natural remedies, there may be a future for this plant/its products. Because of its pleiotropic effects, it might have a role in conditions such as prediabetes,subclinical hypothyroidism apart from diabetes.

To conclude, the insulin plant is a potential plant therapy for diabetes. However at present we don’t know much about its human use and thus must proceed with caution.It opens up several research areas. If found useful in raw form, it may become one of the cheapest ways of treating diabetes.

Further Reading

1. Eddouks M, Bidi A, El Bouhali B, Hajji L, Zeggwagh NA. Antidiabetic plants improving insulin sensitivity. J Pharm Pharmacol. 2014 Sep;66(9):1197–214.

2. Hegde PK, Rao HA, Rao PN. A review on Insulin plant (Costus igneus Nak). Pharmacogn Rev. 2014 Jan;8(15):67–72.

3. Ashwini S, Bobby Z, Sridhar MG, Cleetus CC. Insulin Plant (Costus pictus) Extract Restores Thyroid Hormone Levels in Experimental Hypothyroidism. Pharmacognosy Res. 2017 Mar;9(1):51–9.